Methods and Protocols

in Food Science

Marco Gobbetti
Carlo Giuseppe Rizzello Editors

Basic Methods
and Protocols
on Sourdough
METHODS AND PROTOCOLS IN FOOD SCIENCE

Series Editor
Anderson S. Sant’Ana
University of Campinas
Campinas, Brazil

For further volumes:

[Link]
Methods and Protocols in Food Science series is devoted to the publication of research
protocols and methodologies in all fields of food science.
Volumes and chapters will be organized by field and presented in such way that the
readers will be able to reproduce the experiments in a step-by-step style. Each protocol will
be characterized by a brief introductory section, followed by a short aims section, in which
the precise purpose of the protocol will be clarified.
 Basic Methods and Protocols
on Sourdough

Edited by

Marco Gobbetti
Faculty of Science and Technology, Free University of Bozen-Bolzano, BOLZANO, Bolzano, Italy

Carlo Giuseppe Rizzello

Department of Environmental Biology, Sapienza University of Rome, Rome, Italy
Editors
Marco Gobbetti Carlo Giuseppe Rizzello
Faculty of Science and Technology Department of Environmental Biology
Free University of Bozen-Bolzano Sapienza University of Rome
BOLZANO, Bolzano, Italy Rome, Italy

ISSN 2662-950X ISSN 2662-9518 (electronic)

Methods and Protocols in Food Science
ISBN 978-1-0716-3705-0 ISBN 978-1-0716-3706-7 (eBook)
[Link]
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC,
part of Springer Nature 2024
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and
retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter
developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been
made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer
Nature.
The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.

Paper in this product is recyclable.

Preface to the Series

Methods and Protocols in Food Science series is devoted to the publication of research
protocols and methodologies in all fields of food science. The series is unique as it includes
protocols developed, validated, and used by food and related scientists as well as theoretical
basis are provided for each protocol. Aspects related to improvements in the protocols,
adaptations, and further developments in the protocols may also be approached.
Methods and Protocols in Food Science series aims to bring the most recent developments
in research protocols in the field as well as very well-established methods. As such, the series
targets undergraduate, graduate, and researchers in the field of food science and correlated
areas. The protocols documented in the series will be highly useful for scientific inquiries in
the field of food sciences, presented in such way that the readers will be able to reproduce the
experiments in a step-by-step style.
Each protocol will be characterized by a brief introductory section, followed by a short
aims section, in which the precise purpose of the protocol is clarified. Then, an in-depth list
of materials and reagents required for employing the protocol is presented, followed by a
comprehensive and step-by-step procedures on how to perform that experiment. The next
section brings the dos and don’ts when carrying out the protocol, followed by the main
pitfalls faced and how to troubleshoot them. Finally, template results will be presented and
their meaning/conclusions addressed.
The Methods and Protocols in Food Science series will fill an important gap, addressing a
common complain of food scientists, regarding the difficulties in repeating experiments
detailed in scientific papers. With this, the series has a potential to become a reference
material in food science laboratories of research centers and universities throughout the
world.

Campinas, Brazil Anderson S. Sant’Ana

v
Preface

Sourdough is one of the oldest examples of natural starters for food processing. During the
last decades, it has been rediscovered and largely used, all over the world, for making
leavened baked goods as the most suitable alternative to baker’s yeast and chemical
leavening.
A widely accepted technical definition describes the traditional sourdough as a mixture
of flour and water, spontaneously fermented by lactic acid bacteria and yeasts, and having
acidification and leavening capacities.
Each sourdough has a unique and complex microbiota, which, usually, derives from
house microbiotas, flours, and the selective pressure operated by the conventional back
slopping and technological parameters. The composition of the microbiota can be also
modeled by using selected microbial starters.
Advantages in terms of sensory, rheology, shelf life, and multiple nutritional attributes
have been largely demonstrated with respect to any other leavening agents. The scientific
literature of the last decades has largely converged.
Artisanal and industrial bakeries are greatly attracted by the potentialities of sourdough,
and pushed the research toward the development of innovative biotechnological applica-
tions, which embrace the manufacture of novel fermented products and functional foods,
and the valorization of alternative ingredients to wheat, such as cereals, legumes, pseudo-
cereals, and milling by-products.
In this scenario, it is, therefore, necessary to refer to analytical methods capable of
highlighting, monitoring, and comparing the peculiar and qualitative parameters/attributes
of sourdough. An overall and satisfactory characterization of the sourdough inevitably
implies a multitude of microbiological, biochemical, technological, and nutritional features.
After a first overview on the traditional and modern approaches to produce sourdough,
the book describes the state of the art of the analytical tools available for characterizing
sourdough and its performances. For each type of analysis, the book provides the description
of consolidated protocols and, where applicable, a vision of the potential future develop-
ments of the analytical tools and methods.
The book represents the first reference guide for the ever-growing scientific and techni-
cal communities today working with sourdough.

Bolzano, Italy Carlo Giuseppe Rizzello

Rome, Italy Marco Gobbetti

vii
Contents

Preface to the Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

PART I SOURDOUGH MAKING AND PROPAGATION

1 How to Prepare, Propagate, and Use the Sourdough . . . . . . . . . . . . . . . . . . . . . . . . 3

Kashika Arora, Carlo Giuseppe Rizzello, and Marco Gobbetti

PART II MICROBIAL AND BIOCHEMICAL CHARACTERIZATION OF SOURDOUGH

2 Culture-Dependent Estimation of Lactic Acid Bacteria and Yeasts . . . . . . . . . . . . 17

Fabio Minervini
3 Culture-Independent Estimation of Lactic Acid Bacteria and Yeasts . . . . . . . . . . . 29
Erica Pontonio and Carlo Giuseppe Rizzello
4 Shotgun Metagenomic Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Francesco Maria Calabrese and Maria De Angelis
5 Determination of pH and Titratable Acidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Kashika Arora and Raffaella Di Cagno
6 Determination of Lactic and Acetic Acids and Estimation
of Their Molar Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Andrea Polo and Marco Gobbetti
7 Determination of the Content of Free Amino Acids and Their Profiling . . . . . . . 71
Michela Verni
8 Soluble Sugars and Polysaccharides. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Michela Verni and Marco Montemurro

PART III SENSORY, RHEOLOGY AND NUTRITIONAL ATTRIBUTES

OF SOURDOUGH BAKED GOODS

9 Volume Determination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Marco Montemurro
10 Texture Profile Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Marco Montemurro and Erica Pontonio
11 Image Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Michela Verni and Carlo Giuseppe Rizzello
12 Descriptive Sensory Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Hana Ameur and Marco Gobbetti
13 Determination of the Volatile Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Giuseppe Celano and Maria De Angelis

ix
x Contents

14 In Vitro Determination of Protein Nutritional Indexes . . . . . . . . . . . . . . . . . . . . . . 135

Erica Pontonio and Carlo Giuseppe Rizzello
15 In Vitro Determination of the Glycemic Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Alice Costantini, Olga Nikoloudaki, and Raffaella Di Cagno
16 Estimation of Phytic Acid Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Olga Nikoloudaki and Raffaella Di Cagno
17 Phenolic Compounds and In Vitro Antioxidant Activity . . . . . . . . . . . . . . . . . . . . . 165
Rosanna Latronico and Pasquale Filannino

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Contributors

HANA AMEUR • Faculty of Agricultural, Environmental and Food Sciences Free University of
Bozen, NOItech Park, Bolzano, Italy
KASHIKA ARORA • Faculty of Agricultural, Environmental and Food Sciences, Free University
of Bozen, Bolzano, Italy
FRANCESCO CALABRESE • Department of Soil, Plant and Food Science, University of Bari
“Aldo Moro”, Bari, Italy
GIUSEPPE CELANO • Department of Soil, Plant and Food Science, University of Bari “Aldo
Moro”, Bari, Italy
ALICE COSTANTINI • Faculty of Agricultural, Environmental and Food Sciences Free
University of Bozen, NOItech Park, Bolzano, Italy
MARIA DE ANGELIS • Department of Soil, Plant and Food Science, University of Bari “Aldo
Moro”, Bari, Italy
RAFFAELLA DI CAGNO • Faculty of Agricultural, Environmental and Food Sciences, Free
University of Bozen, Bolzano, Italy
PASQUALE FILANNINO • Department of Soil, Plant and Food Science, University of Bari “Aldo
Moro”, Bari, Italy
MARCO GOBBETTI • Faculty of Agricultural, Environmental and Food Sciences, Free
University of Bozen, Bolzano, Italy
ROSANNA LATRONICO • Department of Soil, Plant and Food Science, University of Bari “Aldo
Moro”, Bari, Italy
FABIO MINERVINI • Department of Soil, Plant and Food Science, University of Bari “Aldo
Moro”, Bari, Italy
MARCO MONTEMURRO • National Research Council of Italy, Institute of Science of Food
Production (CNR-ISPA), Bari, Italy
OLGA NIKOLOUDAKI • Faculty of Agricultural, Environmental and Food Sciences Free
University of Bozen, NOItech Park, Bolzano, Italy
ANDREA POLO • Faculty of Agricultural, Environmental and Food Sciences Free University of
Bozen, NOItech Park, Bolzano, Italy
ERICA PONTONIO • Department of Soil, Plant and Food Science, University of Bari “Aldo
Moro”, Bari, Italy
CARLO GIUSEPPE RIZZELLO • Department of Environmental Biology, Sapienza University of
Rome, Rome, Italy
MICHELA VERNI • Department of Environmental Biology, “La Sapienza” University of Rome,
Rome, Italy

xi
 Part I

Sourdough Making and Propagation

 Chapter 1

How to Prepare, Propagate, and Use the Sourdough

Kashika Arora, Carlo Giuseppe Rizzello, and Marco Gobbetti

Abstract
Sourdough fermentation is a traditional method used in breadmaking that enhances flavor, texture, and
shelf life and greatly improves several nutritional attributes of leavened baked goods. The traditional
sourdough preparation involves the spontaneous fermentation of a dough obtained by flour and water
that is subjected to subsequent refreshments (backslopping) to obtain a stable and active community of
lactic acid bacteria and yeasts. Process parameters, such as dough yield, temperature, hydration, and time,
influence the microbial activity and the resulting properties. Propagation of sourdough involves regular
feeding of the culture with fresh flour and water establishing a stable fermentation environment and
promoting consistent baked good quality. Sourdough is widely embraced in the baking industry for its
unique flavor, extended shelf life, and potential health benefits, although innovative production procedures
have been recently developed to simplify its use and to standardize its performances. The acidic environ-
ment of sourdough and the proteolytic activity mainly associated with the lactic acid bacteria activity are
responsible for improved dough structure, increased nutrient availability, and prolonged freshness of the
sourdough products compared to leavened baked goods obtained by baker’s yeast. Understanding the
intricacies of sourdough fermentation enables bakers to harness its full potential in creating artisanal or
industrial baked goods with exceptional sensory and nutritional attributes.

Key words Sourdough fermentation, Propagation, Ready-to-use sourdough, Sourdough microbiota,

Baking industry

1 Introduction

Sourdough preparation was primarily recognized as an art symbo-

lizing the traditions and culture of human society. While sourdough
is being consumed since the second millennium BC, the oldest
leavened bread was acquired 5000 years ago from Switzerland.
Microscopic examinations of ancient bread loaves recuperated
from various archaeological sites of Egypt revealed the presence of
yeasts and lactic acid bacteria [1]. To date, the principal role of this
natural starter as leavening agent as well as its influence on the
baked good characteristics has been extensively studied during the
fermentation of cereals, pseudocereals, and legumes [2]. The term
“sourdough microbiota” has been coined to indicate the

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

3
4 Kashika Arora et al.

consortium of lactic acid bacteria and yeasts inhabiting a mature

sourdough that imparts characteristic flavor, texture, and nutri-
tional attributes when used as a baking ingredient. This chapter
details various processes defined and optimized for sourdough
preparation and propagation at laboratory and industrial scale
along with its subsequent use in bakery products.

1.1 Sourdough Sourdough is a mixture of flour and water fermented by lactic acid
Preparation bacteria and yeasts, which is prepared by single or multi-stage
processes. Consequently, it has been classified into four types, i.e.,
type I, II, III, and IV, based on the approach followed to inoculate
microorganisms for sourdough fermentation from a technological
perspective (Fig. 1) [3]. Among these, type I is the traditional
sourdough spontaneously fermented at temperature between
20 and 30 °C by multiple backslopping procedure. Backslopping
is the process of adding new/fresh flour and water to a previously
fermented dough allowing subsequent fermentations until the
sourdough becomes mature and stable. Lactic acid bacteria and
yeasts that are found at high cell density in mature sourdough
under selective pressure operated by the environmental conditions
of the backslopping procedure (acidic and anaerobic environment)
may derive from different sources, including flours or other dough
ingredients, tap water, the equipment used, and environment in
which the sourdough is produced [4]. The investigation of lactic
acid bacteria and yeast microbiotas inhabiting the sourdoughs used
for breadmaking revealed their median cell densities at 9.01 and
7.30 log CFU/g, respectively, maintaining a ratio ranging from
100:1 to 10:1 for lactic acid bacteria to yeasts [5].

Fig. 1 Different types of sourdough produced by traditional and industrial processes

 Sourdough Fermentation Technology 5

On the contrary, allochthonous starter cultures are used for

single-step fermentation in the industries under controlled condi-
tions to obtain a sourdough (type II) with characteristics influenced
by the selected biotypes. The type II sourdough is usually fermen-
ted at high temperatures (30–35 °C) for a rapid and high acidifica-
tion and to inhibit the growth of endogenous yeasts [6]. For the
same reason, fermentation is commonly prolonged to 24–48 h.
The type II sourdough is industrially dehydrated to produce type
III sourdough with compromised viability of the sourdough micro-
biota due to the high temperature employed. While type III sour-
dough may be activated by rehydration during baking, the addition
of commercial baker’s yeast is usually recommended for obtaining
an adequate leavening power. Until recently, a combinatorial sour-
dough (type IV) is commonly prepared in the laboratories and
artisanal bakeries that is initiated with selected starter cultures
(type II sourdough) followed by multiple backslopping (typical of
the type I propagation) to attain a stable sourdough microbiota
[7]. The backslopping steps allow natural selection of microbial
strains in the sourdough based on their competitive and adaptive
nature in the ecosystem influencing the characteristic features of the
resulting bakery products.
It is worth noting that each type of aforementioned sour-
doughs can be influenced by multiple process parameters that are
endogenous (flour composition and addition of other non-flour
ingredients, such as nuts, fruits, herbs, and sweeteners) or exoge-
nous (dough yield, time, temperature, starter microorganisms,
inoculum size, etc.). Initially, flours from conventional cereals
(wheat, rye, and barley) were employed for sourdough fermenta-
tion due to their worldwide cultivation and inclusion in the regular
human diet. Subsequently, an interest in non-conventional grains
and crops emerged owing to their health benefits and meeting the
need for special dietary requirements. Some of these
non-conventional flours commonly used for sourdough produc-
tion include pseudocereals (amaranth, buckwheat, and quinoa),
leguminous flours (chickpea, lentil, faba bean, etc.), and sprouted
flours (germinated cereals and legumes) [8–10]. Recently, an
increasing demand for protein-rich diet and shift toward sustain-
able food sources has introduced an insect flour (obtained from an
edible house cricket, Acheta domesticus) in the European market as
a nutritional ingredient for sourdough fermentation [11]. Although
the inhabiting microbial composition may differ in each type of
flour, the sourdough ecosystem mainly favors the growth of lactic
acid bacteria and yeasts that define the sourdough microbiota.
However, it has been observed that the sourdough microbiota
continuously undergoes changes at the species and/or strain level
during sourdough propagation that has been highlighted in the
next section. Nevertheless, minimal metabolic changes are
observed in the sourdough if the process parameters for sourdough
propagation remain unchanged [12].
6 Kashika Arora et al.

Dough yield (DY) defines the consistency of the dough based

on the proportion of flour and water that can be calculated using
the following equation:
ðWeight of doughÞ
Dough yield ðDY Þ = × 100
ðWeight of flourÞ
Furthermore, while DY considers the ratio of all the dry and
wet ingredients in the dough composition, hydration percentage
indicates the amount of water relative to the amount of flour in the
dough. Type I sourdoughs are usually prepared and propagated as
firm doughs (DY = 150–160) in artisanal bakeries, whereas indus-
tries prefer semi-liquid sourdoughs (DY = 200–300) for enhanced
acidification and maintain the active state of the starters. Overall,
high DY values allow high level of automation (backslopping and
addition to doughs operated by dosing pumps). A higher DY of
400–500 has been used for applications in sourdough preparations
using rye bran or gluten-free flours due to their high water absorp-
tion capacity [13, 14].
Starters used for sourdough preparation vary in species com-
position depending on the type of sourdough. However, the arti-
sanal bakeries maintain traditional starters in form of doughs that
are daily backslopped (and refrigerated for 1–2 days when neces-
sary) for their consistent metabolic activity. A collection of many
such traditional sourdoughs that are used as starters, collected all
over the world, is maintained in Puratos sourdough library in Saint-
Vith, Belgium [15]. The preparation of type II sourdoughs involves
one-step fermentation with specific strains of lactic acid bacteria
that results in faster acidification of the dough inhibiting the
growth of yeasts in the dough. Consequently, breadmaking using
type II sourdoughs requires the addition of baker’s yeast to allow
the dough leavening. In this case, the inoculum of the selected
starters is performed by using refrigerated pre-cultures, frozen cell
pellets, or dried cells obtained by freeze-drying or spray-drying to
preserve their viability. Several industries employ type III sour-
dough (corresponding to a dried type II sourdough) for bakery
applications. However, type II sourdough dehydration, performed
by using high temperatures, compromises viability of the starter
cells. However, type III sourdoughs can impart typical sensorial
features to the baked goods. As mentioned before, type IV sour-
dough is more commonly prepared in laboratory setting, primarily
to identify and characterize the resistant, dominant, and competi-
tive sourdough microbiota that can eventually be used as starters
for type II sourdough preparation. Research is being conducted to
develop starters for direct use by preserving the viability using wet
granulation technique that had been successfully applied in phar-
maceutical industry [16]. The shift in microbial ecology of sour-
dough is reflected in the technological and organoleptic properties
of baked goods [17], and thus starter selection is considered an
important step during sourdough preparation.
 Sourdough Fermentation Technology 7

The mechanical mixing of ingredients for sourdough prepara-

tion, commonly referred to as kneading, influences the technologi-
cal parameters of the dough and consequently the baked products.
Primarily, the kneading process alone can be affected by kneading
time and speed, dough aeration and temperature, as well as total
water content and its temperature. Cappelli et al. systematically
reviewed the effect of kneading process on dough rheology and
bread characteristics [18]. An optimal kneading for breadmaking is
defined by the formation of a viscoelastic gluten network in the
dough along with the initial gas inclusion for the activity of yeast
during mixing and proofing. Proper mixing during the kneading
step of dough preparation is crucial for the physicochemical inter-
action of dough ingredients. Therefore, many bakers consider
dough mixing as a decisive step in determining the rheological
characteristics of the final baked product. The baking industries
employ different kneading machines, such as vertical spindle,
high-speed horizontal, double-arm, continuous, and planetary
mixers, that are particular to baking applications [19]. The dough
development is usually monitored in the industries by descriptive
rheological tests using a Farinograph or Mixograph. Several meth-
ods used to determine the readiness of the dough have been
reviewed recently that are executed to gain a better control over
dough preparation [20].

1.2 Sourdough While a sourdough can be prepared through the backslopping

Propagation and procedure or by inoculating with starters, the establishment of a
Maintenance mature and stable sourdough primarily relies on sourdough propa-
gation. This refreshment procedure involves the addition of sour-
dough (at the end of incubation) to fresh flour and water in order
to provide an ecological environment for the persistence of sour-
dough starters. The primary objective of sourdough propagation is
to introduce the sourdough microbiota and its functional charac-
teristics into the subsequent dough while maintaining the starter
culture in an active state, which was initially inoculated at the onset
of fermentation. The inoculum percentage during sourdough
propagation varies from 10% to 25% that has an influence on the
fermentation rate, exopolysaccharide synthesis, and sensory and
rheological attributes [2]. It has been observed that a sourdough
prepared with wheat, rye, or spelt reaches equilibrium upon daily
backslopping in up to ten days. The sourdough microbiota under-
goes changes in a three-step process to reach this equilibrium,
beginning from the prevalence of sourdough-atypical LAB to
highly adapted sourdough-typical LAB. These changes arise due
to microbial interactions and competition for available nutrients.
Minervini et al. elaborated on ecological parameters influencing the
sourdough microbial diversity and stability among which sour-
dough propagation also plays a crucial role [17]. Moreover, the
8 Kashika Arora et al.

robustness of inoculated LAB and yeast species is also impacted

during continuous backslopping of sourdough. For instance,
Siragusa et al. demonstrated the persistence of only three strains
of Fructilactobacillus sanfranciscensis during a 10-day long sour-
dough propagation (inoculum 10% w/w) among the nine sour-
doughs singly inoculated with nine different strains of
F. sanfranciscensis [21]. In a similar study, Baek et al. showed the
dominance of F. sanfranciscensis and a decrease in the cell densities
of Latilactobacillus curvatus and Levilactobacillus brevis after the
11th day of propagation (inoculum 25% w/w) [22]. Another study
investigated the robustness and interactions of microbial strains in
firm (DY = 154) and liquid (DY = 330) sourdoughs inoculated
with five different lactobacilli strains and either a Saccharomyces
cerevisiae or a Candida milleri strain during ten consecutive back-
slopping (inoculum 10% w/w) [23]. They observed that the behav-
ior of lactobacilli strains depended on the DY, refreshment time, as
well as the yeast strain. Nevertheless, a recent study on rheological
behavior of type I sourdough suggested that rheological changes
should also be considered as one of the parameters in assessing the
stability and maturity of a sourdough during the propagation [24].
Furthermore, continuous sourdough propagation is funda-
mental to starter maintenance, as usually practiced in artisanal
bakeries and commercial baking industries. The age of the starters
and the frequency at which they are propagated influence the
leavening or fermentative capacity of a sourdough. The time
between the two sourdough propagations may vary from 12 h to
7 days that affects the ability of the sourdough starter inoculum to
adapt to the new ecological environment. Calvert et al. hypothe-
sized that there is a range of propagation frequency that optimizes
the fermentative capacity of a sourdough while maintaining its
microbial diversity [25]. Moreover, while short fermentation time
during sourdough propagation allows the dominance of fast-
acidifying LAB species, longer fermentation times favor more
acid-tolerant species (e.g., Lactiplantibacillus plantarum, Limosi-
lactobacillus reuteri, and Limosilactobacillus fermentum)
[25, 26]. An “ideal” proportion of inoculum and fermentation
time for sourdough propagation is subjective to the bakers’ prefer-
ences and may vary depending on the traditional practices and
recipes used for baking.

1.2.1 Preparation of Type 1. Flour (soft wheat but also durum wheat or whole wheat flour).
I Sourdough 2. Water.
3. Weighing balance.
Ingredients and Equipment
4. Dough mixer.
5. Stainless-steel bowl.
6. Glass beaker (1 L).
 Sourdough Fermentation Technology 9

7. Dough thermometer.
8. Fermentation chamber.
9. pH meter with a food probe.

Procedure The following step-by-step procedure for sourdough preparation is

focused on type I sourdough with a DY of 160 and a final dough
weight of 1000 g. Please note that the amounts mentioned in this
procedure are subject to change and can be modified as per the
amount required for the final mature sourdough.
1. Weigh 625 g of durum wheat flour in a stainless-steel bowl on a
weighing balance and 375 mL of tap water in a measuring
cylinder.
2. Then mix the two ingredients in a dough mixer by adding
100% of flour (625 g) and approximately 60% of tap water
(195–200 mL) in a dough mixer at a slow speed to allow the
blending of ingredients.
3. Add the rest of the quantity of water while continuous mixing
to allow the formation of gluten structure in the dough. The
mixing step usually lasts 7–11 min depending on the dough
mixer.
4. Transfer 750 g of dough to the stainless-steel bowl and cover it
with a plastic wrap or a cloth.
5. Simultaneously, transfer the leftover 250 g of dough to a glass
beaker (1 L) and mark the starting point of fermentation to
control the leavening power of the dough.
6. Measure the pH of the dough in the glass beaker using a pH
meter with food probe and incubate the doughs at 30 °C for
16–24 h (I fermentation).
7. Measure the pH of the dough again and store the dough at
10–12 °C to avoid any microbial activity before next backslop-
ping. The leavening power is measured by determining the
increase in volume after the fermentation. No leavening is
usually observed after the first fermentation.
8. Perform the refreshment of dough by backslopping the previ-
ous dough (mother dough) as inoculum (20% of dough
weight) in fresh flour and water.
9. Weigh 200 g of mother dough (stored after 16–24-h fermenta-
tion) and mix it with 500 g of fresh durum wheat flour and
300 mL of water (added slowly while mixing).
10. Divide the mixed dough into 750 g in a stainless-steel bowl and
250 g in a glass beaker as mentioned previously and measure
the pH before starting the fermentation again at 30 °C for 8 h.
At the end of the 8 h-incubation, the fermented dough must be
stored at 10–12 °C until the next backslopping step.
10 Kashika Arora et al.

11. Repeat the backslopping procedure until a stable pH value and

leavening power is obtained, indicating a mature sourdough.
The pH value for a mature type I sourdough typically ranges
from 4.0 to 4.3 units with a 3–4 times increase in the dough volume
indicating the leavening power. Ideally, backslopping procedure
should be repeated 8–10 times to obtain a stable and mature type
I sourdough. An overview on the use of sourdough in baking
applications is detailed in the forthcoming section.
Sourdough can be stored in refrigerated conditions (preferably
10–12 °C) for 2–3 days if necessary to stop the daily backslopping
procedure (e.g., during the weekend).

1.3 Sourdough Use The mature type I sourdough can be used in baking applications as
an acidifying and leavening agent attributed to the metabolic activ-
ities of lactic acid bacteria and yeasts, respectively. Depending on
the baking recipe, the sourdough can be used directly in the dough
for breadmaking or subjected to refreshments to activate microbial
metabolism. The sourdough can be added at percentages ranging
from 10% to 30% in the dough intended for breadmaking. The
bread dough is allowed to acidify and leaven in a proofing chamber
set at 28–30 °C that lasts for 4–6 h. In some cases, after 4 h of bulk
proofing, the dough may be broken down into smaller dough
pieces for bread loaves that are further proofed to regain the leav-
ened volume. A diagrammatic overview of the use of type I sour-
dough is depicted in Fig. 2. Owing to the long duration required
for fermentation, the baking industries earlier preferred commercial
baker’s yeast over sourdough as a leavening agent.
In order to speed up the sourdough leavening process, a small
amount of brewer’s yeast (0.3–0.8%) can be added to the dough for
breadmaking, together with the sourdough. This operation is also
allowed in some of the protocols for the production of sourdough
breads with recognized protected denomination of origin (PDO).
Moreover, in the recent years, sourdough-based traditional
baked goods, ranging from breads (steamed or flat breads) to
fermented cereal beverages and porridges, are being produced at
an industrial level. In Europe, national regulations have been
defined for the use of sourdough in baked products and hence
the development of sourdough varies from country to country
[12]. The baking industries are continuously aiming to reduce
manpower and working time in order to improve the efficiency of
the sourdough production process. Therefore, automated fermen-
ters have been introduced with controlled monitoring of the pro-
cess parameters. The automation technology is beneficial for
producing large quantities (up to 80,000 kg) of sourdough with
automated addition of ingredients (flour, water, and starters), con-
tinuous mixing or resting, and automated cooling once the desired
 Sourdough Fermentation Technology 11

Fig. 2 Overview of the use of type I sourdough for breadmaking applications

characteristics for the sourdough are achieved. The production

process usually employs the backslopping of 20–30% of flour-
water mixture that acts as a mother dough. The manpower is
further reduced by linking multiple tanks together that allows
circulation of the leavened dough and simultaneous cleaning
through pump systems for a continuous production. However, it
is important to note that these automated fermenters generally are
used to produce liquid or semi-liquid sourdoughs to ease the
process of transfer using the pump systems.
Moreover, sourdoughs are today available in the market as
dried or liquid preparations. The dried sourdoughs are produced
by freeze drying, spray granulation, fluidized bed drying, spray
drying, or drum drying that inactivates the microbial starters and
contains relatively lower flavor compounds as compared to liquid
sourdoughs due to the drying treatment. The dosage of sourdough
in bakery products varies depending on the type of sourdough used
(dried or liquid) and the desired flavor in the final product.
Although lactic acid bacteria strains resistant to drying process are
usually selected for the production of dried sourdoughs, the viabil-
ity of the microbial starters is usually reduced after the dehydration
process due to high-temperature treatments. However, these dried
sourdoughs have the ability for rapid acidification and
12 Kashika Arora et al.

enhancement of texture and aroma that promotes their use in the

industries. The loss of viability, in this case, is compensated by the
addition of baker’s yeast to allow desired leavening power.
A recent study by Caglar et al. demonstrated that the use of
spray-dried or freeze-dried sourdoughs in different blending ratios
(3, 6, 9, and 15%) for breadmaking significantly changed the dough
rheology and reduced the bread volume as compared to baker’s
yeast bread [27]. Another study reported that the sourdough
starter dried by spray drying exhibited better starter viability and
bread properties when rehydrated under specific temperature con-
ditions [28]. Moreover, extensive research has been conducted to
identify sourdough starters suitable for use in salty and sweetened
baked goods. For example, LAB strains tolerant to high sucrose
concentrations (up to 30%) have been identified for their use in
sweetened bakery products [29]. The dried sourdough (type III)
use in bakery products also acts as a natural preservative rendered
by the antifungal activity of the sourdough starters [30]. Neverthe-
less, owing to the nutritional and functional benefits of specific
strains used in the preparation of type II sourdough, investigations
on the use of freeze-dried pure strains to replace type II sourdough
are in progress with promising results in terms of bread quality
[31]. In this context, a recent study exploited the potential of
bioprotective cultures, comprising commercially available freeze-
dried probiotic strains, in combination with type III sourdough for
the production of base-pizza and focaccia with an extended shelf
life [32]. Use of fermented non-conventional flours (e.g., pseudo-
cereals and legumes) in bakery applications is also of increasing
interest over the last decade. Some of the commercially available
gluten-free sourdough breads and sourdough baking mixtures are
summarized by Irigoytia et al. [33]. The sourdough fermentation
contributes to an improved texture and sensory profile of gluten-
free products by the action of sourdough microbiota. However,
further studies are required to gain insights into the rheological and
technological features of gluten-free sourdough that are influenced
by the fermentation process parameters. Progressive research is
being conducted for the use of sourdough fermentation in all
types of baked goods to better exploit the potential of the available
ingredients. Nevertheless, in this era of artificial intelligence, efforts
are being made to deduce mathematical models for a thorough
understanding of a sourdough ecosystem by considering the pro-
cess parameters that affect fermentative activity. Such in-depth
modeling studies will help in developing customized sourdoughs
with desired functionality and quality of baked goods.
 Sourdough Fermentation Technology 13

References
1. Samuel D (1999) A new look at old bread: their impact on the bread texture. LWT 177:
ancient Egyptian baking. Archaeol Int 3:28–31 114566
2. Arora K, Ameur H, Polo A et al (2021) Thirty 14. Vogelmann SA, Hertel C (2011) Impact of
years of knowledge on sourdough fermenta- ecological factors on the stability of microbial
tion: a systematic review. Trends Food Sci associations in sourdough fermentation. Food
108:71–83 Microbiol 28:583–589
3. Siepmann FB, Ripari V, Waszczynskyj N et al 15. The Puratos Sourdough Library, https://
(2018) Overview of sourdough technology: [Link]/en/
from production to marketing. Food Biopro- 16. Blaiotta G, Romano R, Trifuoggi M et al
cess Technol 11:242–270 (2022) Development of a wet-granulated sour-
4. Minervini F, Lattanzi A, Dinardo FR et al dough multiple starter for direct use. Foods
(2018) Wheat endophytic lactobacilli drive 11:1278
the microbial and biochemical features of sour- 17. Minervini F, De Angelis M, Di Cagno R et al
doughs. Food Microbiol 70:162–171 (2014) Ecological parameters influencing
5. Minervini F, Di Cagno R, Lattanzi A et al microbial diversity and stability of traditional
(2012) Lactic acid bacterium and yeast micro- sourdough. Int J Food Microbiol 171:136–
biotas of 19 sourdoughs used for traditional/ 146
typical Italian breads: interactions between 18. Cappelli A, Bettaccini L, Cini E (2020) The
ingredients and microbial species diversity. kneading process: a systematic review of the
Appl Environ Microbiol 78:1251–1264 effects on dough rheology and resulting bread
6. Corsetti A (2013) Technology of sourdough characteristics, including improvement strate-
fermentation and sourdough applications. In: gies. Trends Food Sci 104:91–101
Gobbetti M, G€anzle M (eds) Handbook on 19. Berk Z (2018) Mixing. In: Food process engi-
sourdough biotechnology. Springer, neering and technology. Academic Press
New York, pp 85–103 20. Parenti O, Guerrini L, Mompin SB et al (2021)
7. Vrancken G, Rimaux T, Weckx S et al (2011) The determination of bread dough readiness
Influence of temperature and backslopping during kneading of wheat flour: a review of
time on the microbiota of a type I propagated the available methods. J Food Eng 309:
laboratory wheat sourdough fermentation. 110692
Appl Environ Microbiol 77:2716–2726 21. Siragusa S, Di Cagno R, Ercolini D et al (2009)
8. Montemurro M, Pontonio E, Gobbetti M et al Taxonomic structure and monitoring of the
(2019) Investigation of the nutritional, func- dominant population of lactic acid bacteria
tional and technological effects of the sour- during wheat flour sourdough type I propaga-
dough fermentation of sprouted flours. Int J tion using lactobacillus sanfranciscensis star-
Food Microbiol 302:47–58 ters. Appl Environ Microbiol 75:1099–1109
9. Coda R, Cagno RD, Gobbetti M et al (2014) 22. Baek H-W, Bae J-H, Lee Y-G et al (2021)
Sourdough lactic acid bacteria: exploration of Dynamic interactions of lactic acid bacteria in
non-wheat cereal-based fermentation. Food Korean sourdough during back-slopping pro-
Microbiol 37:51–58 cess. J Appl Microbiol 131:2325–2335
10. Rizzello CG, Calasso M, Campanella D et al 23. Galli V, Venturi M, Pini N et al (2019) Liquid
(2014) Use of sourdough fermentation and and firm sourdough fermentation: microbial
mixture of wheat, chickpea, lentil and bean robustness and interactions during consecutive
flours for enhancing the nutritional, texture backsloppings. LWT 105:9–15
and sensory characteristics of white bread. Int 24. Özülkü G (2019) Rheological behaviour of
J Food Microbiol 180:78–87 type I sourdough during refreshment proce-
11. Vasilica B, (Tătar) B, Chis MS, Alexa E et al dure. Int J Food Sci Nutr 2
(2022) The impact of insect flour on sour- 25. Calvert MD, Madden AA, Nichols LM et al
dough fermentation-fatty acids, amino-acids, (2021) A review of sourdough starters: ecol-
minerals and volatile profile. Insects 13:576 ogy, practices, and sensory quality with applica-
12. Brandt MJ (2007) Sourdough products for tions for baking and recommendations for
convenient use in baking. Food Microbiol 24: future research. PeerJ 9:e11389
161–164 26. De Vuyst L, Van Kerrebroeck S, Harth H et al
13. Arora K, Tlais AZA, Augustin G et al (2023) (2014) Microbial ecology of sourdough fer-
Physicochemical, nutritional, and functional mentations: diverse or uniform? Food Micro-
characterization of gluten-free ingredients and biol 37:11–29
14 Kashika Arora et al.

27. Caglar N, Ermis E, Durak MZ (2021) Spray- index lowering agent in salty muffins. Br Food
dried and freeze-dried sourdough powders: J. ahead-of-print 125:3573
properties and evaluation of their use in bread- 31. Gu Y, Luo X, Qian H et al (2023) Effects of
making. J Food Eng 292:110355 freeze-dried pure strains to replace type II sour-
28. Albagli G, Schwartz I d M, Amaral PFF et al dough in bread production. Food Biosci 53:
(2021) How dried sourdough starter can 102752
enable and spread the use of sourdough 32. Calasso M, Marzano M, Caponio GR et al
bread. LWT 149:111888 (2023) Shelf-life extension of leavened bakery
29. Reale A, Zotta T, Ianniello RG et al (2020) products by using bio-protective cultures and
Selection criteria of lactic acid bacteria to be type-III sourdough. LWT 177:114587
used as starter for sweet and salty leavened 33. Irigoytia KF, Espósito NN, Busch VM et al
baked products. LWT 133:110092 (2023) Fermented gluten-free baked
30. Çetin Babaoğlu H, Arslan Tontul S, Karadu- goods. In: de Escalada Pla MF, Genevois CE
man L et al (2023) The usage of sourdough (eds) Designing gluten free bakery and pasta
powder as the natural preservative and glycemic products. Springer International Publishing,
Cham, pp 163–210
 Part II

Microbial and Biochemical Characterization of Sourdough

 Chapter 2

Culture-Dependent Estimation of Lactic Acid Bacteria

and Yeasts
Fabio Minervini

Abstract
Here we describe how to enumerate lactic acid bacteria and yeasts in sourdough samples, using the plate
count method and selective/elective agar culture media. The results of the analysis, expressed in terms of
cell density, are useful to predict the leavening and acidification performances of the sourdough or to
understand the reasons related to its poor performances. In addition, the plates containing colonies of
presumptive lactic acid bacteria and yeasts might be exploited for isolating pure microbial cultures.

Key words Lactic acid bacteria, Yeasts, Salt peptone water, Agar culture media, Autoclave, Serial
dilutions, Plate, Colonies, Enumeration, Cell density

1 Introduction

Culture-dependent estimation of lactic acid bacteria and yeasts in

sourdough is a method based on the use of various elective or
selective culture media inoculated with diluted sourdough samples
harboring the above-mentioned microorganisms. The method is
used to enumerate the presumptive lactic acid bacteria and yeasts in
any food samples (sourdough included). It is also referred to as
“plate count” because colonies, presumptively belonging to a given
microbial taxon (i.e., taxonomic group), are counted after micro-
bial growth in or on solidified culture media contained in plates
(alias Petri dishes). The method starts with appropriate sampling
procedures, followed by ad hoc preparation of the sample (i.e.,
homogenization) aiming to “extract” microbial cells entrapped in
the matrix. Afterward, given that sourdough may harbor from
hundred thousand to hundred billion microbial cells (summing
lactic acid bacteria and yeasts) [1], the sample undergoes serial
dilutions. Then, the serially diluted sample will be inoculated in
plates containing appropriate agar culture media. Afterward, plates
will be incubated and, finally, colonies, originating from repeated

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

17
18 Fabio Minervini

multiplications of microbial cells harbored in the initial sourdough

sample, will be counted. The enumeration of presumptive lactic
acid bacteria and yeasts will be based on the number of colonies, the
dilution factor of the sample, and the volume of sample inoculated
in plates. Several elective or selective agar media can be used in this
method for a given microbial group, be it lactic acid bacteria or
yeasts [2, 3], but it seems that no medium guarantees a satisfactory
recovery of all the members. This is probably due to different
metabolic capabilities among microorganisms, although belonging
to the same microbial group and living in the same ecosystem
[3]. In this regard, the reader will find here details about four and
two culture media used for enumerating lactic acid bacteria and
yeasts, respectively, although the list of culture media that can be
used could include more than 13 media. A worthy question could
be, “How many culture media and which shall I use?” The answer
depends on the degree of knowledge of the microbial community
harbored in a given sourdough. In case the composition of bacterial
and yeast populations is totally unknown, it would be advisable to
use at least three media for lactic acid bacteria: (1) media elective for
lactobacilli and other genera (i.e., Leuconostoc, Pediococcus, and
Weissella) with similar nutritional requirements (i.e., de Man
Rogosa Sharpe, MRS); (2) media that allow to enumerate those
very fastidious and sourdough-adapted lactobacilli such as Fructi-
lactobacillus sanfranciscensis (i.e., Sour Dough Bacteria, SDB); and
(3) media elective for lactococci and streptococci (i.e., M17 with
glucose), which are genera of lactic acid bacteria usually playing the
role of subdominant microorganisms. At least two media should be
used for enumeration of yeasts, considering that some media (e.g.,
Malt Extract) contain maltose (not usable by all the yeast species) as
the main carbohydrate, whereas other media (e.g., Sabouraud Dex-
trose) contain glucose, which is usable by virtually all the yeast
species. Overall, the use of more than one culture medium for
enumerating one microbial group allows to describe a microbial
diversity that approaches to the one really characterizing a certain
sourdough. Indeed, it is frequent to find in traditional sourdough
samples more than one microbial genus and species and character-
ized by partially different nutritional requirements [3]. On the
other hand, in case the microbial community of a certain sour-
dough is known, it is more practical and cheaper to use just two
culture media, one for lactic acid bacteria and one for yeasts. This
can be the case of a sourdough prepared with a defined starter
culture or that of a traditional sourdough whose microbial commu-
nity has been described through an ad hoc study. Overall, some
media may allow to differentiate between two bacterial taxa. For
instance, MRS 5 allows to distinguish between F. sanfranciscensis
and Lactiplantibacillus plantarum, two sourdough key bacterial
species, based on their colony morphology. Indeed, colonies of
F. sanfranciscensis on MRS 5 have a diameter less than 1 mm and
 Enumerating Lactic Acid Bacteria and Yeasts 19

are whitish, whereas those of L. plantarum have a diameter of

1–2 mm, have entire margins, and are white [4]. Another example
is represented by the use of modified Chalmers medium, which may
be used to differentially enumerate yeasts, Leuconostoc sp., Entero-
coccus sp., Lactococcus sp., homofermentative lactobacilli, and het-
erofermentative lactobacilli harbored in the same sourdough [5].

2 Materials

Prepare all the solutions and media and sterilize them in autoclave
(121 °C, 15 min). After sterilization, they can be stored at room
temperature. However, it is advisable to store them under
refrigeration.

Salt Peptone Water Dissolve 8.5 g of peptone and 1.0 g of NaCl

in 900 mL of distilled or demineralized water. After having
obtained a clear solution, bring to the volume of 1000 mL (see
Note 1). Physiological solution (9.0 g of NaCl in 1000 mL of
distilled or demineralized water) may be used as well, instead of
salt peptone water.

Fresh Yeast Extract This represents an ingredient for some cul-

ture media, such as SDB and modified MRS and shall be prepared
one day before the preparation of the medium. Suspend 60 g of
compressed baker’s yeast (see Note 2) in 300 mL of distilled water.
Sterilize the suspension in autoclave (121 °C, 30 min), which will
not only inactivate yeast cells but also cause cell lysis and conse-
quent release of compounds from yeast cells that will be used by
microorganisms for their growth in culture media. After steriliza-
tion, wait for the settling of dead/lysed cells (preferably at 4 °C)
over night. Then centrifuge (aseptically) (5520 × g, 20 min, 4 °C)
and finally collect, under aseptic conditions, the supernatant and
dispense it in a sterile bottle.

Glucose Solution Prepare this solution if you intend to use M17

as culture medium for enumerating lactococci and streptococci.
Solubilize 10 g of glucose in 100 mL of distilled/demineralized
water and sterilize (autoclave).

Carbohydrates Solution Prepare this solution if you intend to

use MRS 5 as culture medium for enumerating lactic acid bacteria.
Solubilize 10 g of maltose, 5 g of glucose, and 5 g of fructose in
100 mL of distilled/demineralized water and sterilize (autoclave).

Vitamins Solution Prepare this solution if you intend to use MRS

5 as culture medium for enumerating lactic acid bacteria. Solubilize
20 Fabio Minervini

20 mg each of cobalamin, folic acid, nicotinic acid amide, pantothe-

nic acid, pyridoxal phosphate, and thiamine in 100 mL of distilled/
demineralized water and sterilize by using membrane filters (pore
diameter ≤ 0.22 μm).

Cycloheximide Solution Dissolve 1 g of cycloheximide in

100 mL of distilled/demineralized water and sterilize by using
membrane filters (see Note 3). This solution will be added, at a
dose of 10 mL/L, in the culture media intended for lactic acid
bacteria after having sterilized the media and after having
conditioned the agar media at a temperature not higher than 55 °
C. The addition of the cycloheximide solution must be performed
under aseptic conditions.

Modified MRS (mMRS) Agar Some companies sell powdered

preparations of MRS (in some cases also containing agar). Just
follow the manufacturer’s directions for dissolving the powder,
but before adding ca. 700 mL of distilled/demineralized water,
add 5 g of maltose (in case you want to prepare 1 L of medium).
Solubilize and add 50 ml of fresh yeast extract, then nearly bring to
final volume (1 L) (see Note 4), adjust pH to 5.6 and bring to final
volume [6] (see Note 5).

SDB Agar For 1 L of medium, dissolve maltose (7.5 g), dextrose

(7.5 g), yeast extract (3.0 g), tryptone (6.0 g), agar n° 3 (or technical
agar) (12.0 g) in ca. 250 mL of distilled/demineralized water and
add 500 mL of fresh yeast extract. Nearly bring to final volume
(e.g., 950 mL), adjust pH to 5.6 and bring to final volume [7] (see
Note 5).

MRS 5 Agar For 1 L of medium, dissolve agar n° 3 (12 g),

tryptone (10 g), meat extract (or “LabLemco powder,” 5 g),
yeast extract (5 g), C2H3NaO2 ∙ 3H2O (5 g), ammonium chloride
(3 g), K2HPO4 ∙ 3H2O (2.6 g), KH2PO4 ∙ 3H2O (4 g), MgSO4 ∙
7H2O (0.2 g), MnSO4 ∙ 4H2O (0.05 g), cysteine-HCl (0.5 g),
Tween 80 (1 mL) in ca. 700 mL of distilled/demineralized water.
Nearly bring to final volume (e.g., 880 mL), paying attention to
recover all the agar, which is almost insoluble at room temperature
and tends to precipitate, adjust pH to 5.8, and bring to final volume
of 889 mL. Sterilize the medium, bring the temperature of the
medium to 50–55 °C, and add, under aseptic conditions, 100 mL
of carbohydrates solution and 1 mL of vitamins solution [8] (see
Note 5).

M17—Glucose Agar Some companies sell powdered prepara-

tions of M17 (in some cases also containing agar). Just follow the
manufacturer’s direction for dissolving the powder, and remember
 Enumerating Lactic Acid Bacteria and Yeasts 21

that after sterilization, you will have to add the glucose solution.
This means that the final volume before sterilization shall consider
the volume of glucose solution. For instance, if you intend to
prepare 1 L of M17— glucose agar, 50 mL of the medium will be
represented by the glucose solution. Sterilize the medium, bring
the temperature of the medium to 50–55 °C, and add, under
aseptic conditions, the glucose solution (50 mL/L of medium)
(see Note 5).

Sabouraud Dextrose Agar Some companies sell powdered pre-

parations of Sabouraud Dextrose (in some cases also containing
agar). Just follow the manufacturer’s direction for dissolving the
powder. Although not mandatory, it is advisable, before dissolving
the powder, to add chloramphenicol (at a dose of 0.1 g/L) in order
to inhibit growth of bacteria in or on the medium (see Note 6).

Malt Extract Agar Some companies sell powdered preparations of

Malt Extract (in some cases also containing agar). Just follow the
manufacturer’s direction for dissolving the powder. Although not
mandatory, it is advisable, before dissolving the powder, to add
chloramphenicol (at a dose of 0.1 g/L) in order to inhibit growth
of bacteria in or on the medium (see Note 6).

3 Methods

3.1 Sampling the 1. Use disposable gloves for picking at least 50 g of the sourdough
Sourdough sample.
2. Put the sample in a sterile bag that can be sealed to avoid
contamination and insert the bag in a refrigerated container.
3. Bring the sample in the laboratory and store it at 4 °C for a
maximum of 24 h before starting the analysis (see Note 7).

3.2 Homogenization 1. Weigh 10–40 g of sample, using a sterile spatula, in a sterile

of the Sourdough plastic bag for blenders/homogenizers.
Sample 2. Add 90–360 mL (see Note 8) of salt peptone water in the
same bag.
3. Insert the bag in a peristaltic homogenizer (e.g., Stomacher®)
and treat the sample for 3 min (see Note 9). After treatment,
you have obtained the so-called “initial suspension” (see Note
10).

3.3 Serial Dilutions 1. Transfer 1 mL of the initial suspension (or, in case of liquid
of the Sourdough sourdough, of sample) into a labeled (“10-2” or, in case of
Sample liquid sourdough sample, “10-1”) bacteriology tube
22 Fabio Minervini

containing 9 mL of salt peptone water (or another diluent such

as physiological solution).
2. Place the tube on a vortex mixer for 3 s.
3. Transfer 1 mL of the “10-2” dilution into a labeled (“10-3” or,
in case of liquid sourdough sample, “10-2”) bacteriology tube
containing 9 mL of salt peptone water.
4. Repeat the steps 2 (see Note 11) and 3, proceeding toward the
highest dilution (see Note 12).

3.4 Inoculating Agar 1. Vortex mix the diluted samples and transfer 1 mL of the most
Culture Media with diluted one (i.e., 10-9) into a Petri dish (diameter of the dish:
Diluted Sourdough 90 mm), previously bottom-labeled with type of medium (e.g.,
Sample SDB) and dilution factor (e.g., “-9”).
2. Repeat step 1 two more times to get three analytical replicates.
3. Repeat steps 1 and 2 for the other culture media you intend
to use.
4. Vortex mix the diluted samples and transfer 1 mL of the
one-tenth less-diluted one (i.e., 10-8) into a Petri dish, previ-
ously bottom-labeled with type of medium (e.g., SDB) and
dilution factor (e.g., “-8”).
5. Repeat step 4 two more times to get three analytical replicates.
6. Repeat steps 4 and 5 for the other culture media you intend
to use.
7. Vortex mix the diluted samples and transfer 1 mL of the
one-tenth less-diluted one (i.e., 10-7) into a Petri dish, previ-
ously bottom-labeled with type of medium (e.g., SDB) and
dilution factor (e.g., “-7”).
8. Repeat step 7 two more times to get three analytical replicates.
9. Repeat steps 7 and 8 for the other culture media you intend
to use.
10. Pour ca. 15 mL of each molten agar culture medium,
pre-conditioned at 50 °C, in the plates labeled with that
medium (see Note 13).
11. Gently move each plate on the horizontal plan, ideally drawing
the 1 symbol (see Note 14).
12. If you inoculated plates under a laminar flow hood, leave the
top (lid) of the plate slightly misaligned to the bottom, so that
vapor will not condense on the lid.
13. After solidification of the media, invert the plates (bottom
upward, top downward), insert them in a plastic bag, close
the bag with adhesive tape, and incubate plates at 30 °C for
48 h (see Notes 15 and 16).
 Enumerating Lactic Acid Bacteria and Yeasts 23

3.5 Counting 1. Exclude the “non-countable” plates, i.e., those plates contain-
Colonies and ing more than 300 colonies; exclude as well those plates with
Calculating Cell very few colonies (e.g., less than 30).
Densities 2. Count the colonies for each medium (see Note 17), for all the
dilutions you inoculated (excluded those indicated at step 1),
and for all the analytical replicates (see Note 18); register the
number in a table (e.g., Table 1).
3. Multiply each number by the reciprocal of the dilution factor.
For instance, considering the rows of Table 1 containing the
number of colonies observed on SDB plates:
Replicate 1: 30 × 1/10-9 = 30 × 109 colony forming units
(CFU)/g of sourdough.
291 × 1/10-8 = 291 × 108 CFU/g
Replicate 2: 28 × 1/10-9 = 28 × 109 CFU/g.
246 × 1/10-8 = 246 × 108 CFU/g
Replicate 3: 25 × 1/10-9 = 25 × 109 CFU/g.
230 × 1/10-8 = 230 × 108 CFU/g
4. Average the numbers of each replicate. For instance:
Replicate 1: 30 × 109 = 3 × 1010 CFU/g; 291 × 108 =
2.91 × 1010 CFU/g Average = [(3 + 2.91)/2] ×
1010 CFU/g = 2.955 × 1010 CFU/g.
Replicate 2: 28 × 109 = 2.8 × 1010 CFU/g; 246 × 108 =
2.46 × 1010 CFU/g Average = [(2.8 + 2.46)/2] ×
1010 CFU/g = 2.63 × 1010 CFU/g.
Replicate 3: 25 × 109 = 2.5 × 1010 CFU/g; 230 × 108 =
2.3 × 1010 CFU/g Average = [(2.5 + 2.3)/2] ×
1010 CFU/g = 2.4 × 1010 CFU/g.

Table 1
Dilution table for registering the number of colonies observed in plates

Dilutions

Medium 10-9 10-8 10-7 10-6 10-5

SDB agar—replicate 1 30 291 NCa NTb NT
SDB agar—replicate 2 28 246 NC NT NT
SDB agar—replicate 3 25 230 NC NT NT
Sabouraud Dextrose agar—replicate 1 NT 2 29 300 NT
Sabouraud Dextrose agar—replicate 2 NT 3 25 280 NT
Sabouraud Dextrose agar—replicate 3 NT 1 23 259 NT
a
NC, non-countable
b
NT, not tested
24 Fabio Minervini

5. Average the numbers of the three replicates, obtaining the

estimated cell density in your sourdough sample. For instance:
[(2.955 + 2.63 + 2.4)/3] × 1010 = 2.6 × 1010 CFU/g is the
estimated cell density of presumptive lactobacilli in the
sourdough sample.

4 Notes

1. This solution may be used both for homogenization of sour-

dough sample (solid or semi-solid) and serial dilutions of the
sample. If this solution is intended to be used for serial dilu-
tions, it can be dispended in bacteriology tubes (9 mL/tube),
before being sterilized. However, after sterilization, a decrease
of volume could be observed in each tube, which represents a
bias for estimating microbial cell densities. It is advisable to
measure the amount of volume lost upon sterilization, so that
one will dispense in each tube 9 mL of solution + this extra
volume (usually in the order of 0.1–0.9 mL). In this way, after
sterilization, each tube will contain exactly 9 mL of solution.
An alternative way could be sterilization of empty tubes and to
dispense, under aseptic conditions, 9 mL of sterile solution in
each tube.
2. Not all the brands of baker’s yeast will give the same results in
terms of clearness of the fresh yeast extract. It is advisable to
select the brand of baker’s yeast that allows to get a fresh yeast
extract that is as clear as possible. This would help the user to
understand whether the fresh yeast extract is still good to be
used (i.e., with no contaminating microorganisms), even if
some weeks are passed by from preparation.
3. The use of this antibiotic, targeted to eukaryotic cells (such as
yeast cells), is not mandatory. However, if you do not include
this antibiotic in the culture medium for enumerating lactic
acid bacteria, you might observe on/in the media a few or
more yeast colonies (usually with larger diameter), besides
those of lactic acid bacteria (usually with smaller diameter
compared to yeast colonies). In this case, a catalase test
and/or observation of cells composing the colony through an
optical microscope must be carried out. If antibiotic has not
been added and test and/or colonies have not be carried out,
the cell density of lactic acid bacteria population in your sour-
dough sample can be overestimated.
4. For all the agar culture media, pay attention, before bringing to
final volume, to recover all the agar, because the latter is almost
insoluble at room temperature and tends to precipitate.
 Enumerating Lactic Acid Bacteria and Yeasts 25

5. Please consider that if you are going to use the cycloheximide

solution, the final volume shall be reached only with the addi-
tion of the cycloheximide solution.
6. Lactic acid bacteria in sourdough usually are at a higher (at least
one log10 cycle) cell density than yeasts and therefore their
growth in agar media elective for yeasts could make difficult
to distinguish between bacterial and yeast colonies. Neverthe-
less, sourdough lactic acid bacteria are often not able to grow in
media elective for yeasts. That is why the addition of the anti-
bacterial chloramphenicol is advisable, but not mandatory.
7. Consider that prolonged refrigeration of the sourdough sam-
ple, before analysis, could not totally stop microbial growth or,
contrarily, could induce a decline phase of some microbial
populations. In both cases, the estimated cell density of lactic
acid bacteria and yeasts could result significantly different from
that obtained analyzing the sample immediately or after few
hours of refrigerated storage.
8. The volume of salt peptone water depends on the weight of the
sample. You shall choose that, in order to keep the ratio salt
peptone water: sample = 9: 1. For instance, in case you have
weighed 10 g of sample, you should add 90 mL of salt peptone
water.
9. Time of treatment could be less than 3 min (e.g., 1 or 2 min),
but also could be more (e.g., 5 min), depending on the texture
of the sourdough sample. Firm and tough sourdough samples
could require up to 5 min of homogenization treatment.
10. In case of liquid sourdough samples (e.g., those with dough
yield > 280), the homogenization treatment could be skipped.
11. It is understood that, after having used the dilution 10-(n), this
is not more used for further dilutions; instead, you shall trans-
fer the sample from the dilution 10-(n + 1) to another tube, thus
obtaining the dilution 10-(n + 2).
12. The obtainment of the highest dilution, determining the com-
pletion of the serial dilutions procedure, depends on the
expected cell density of microorganisms in sourdough sample.
Usually, stopping after having reached the 10-9 dilution could
be enough to count sourdough samples very rich in microor-
ganisms. In that case, the Petri dishes inoculated with a 10-9
dilution might contain tens colonies, easily countable. In case
of sourdough samples that have been stored under refrigerated
conditions or frozen for long, without having been back-
slopped, the highest dilution could be 10-7 or even lower.
13. Avoid pouring the medium directly on the suspension, because
the relatively high (ca. 50 °C) temperature of the medium
could damage some microbial cells.
26 Fabio Minervini

14. Considering that the medium will solidify at 45 °C, you may
keep on gently moving the plate for ca. 30 s. This allows an
even distribution of microbial cells, which will ease counting
colonies.
15. Most of sourdough-related microorganisms will be able to
form colonies after 48 h of incubation at 30 °C. However, if
the sourdough sample is usually fermented at relatively low
(e.g., ≤ 15 °C) or high (e.g., ≥ 40 °C) temperatures (which is
quite unusual, but a few examples are known), you may con-
sider choosing a temperature of incubation approaching the
sourdough fermentation temperature. If you choose to incu-
bate plates at 15 °C, you will probably have to incubate for
more than 48 h in order to observe colonies.
16. Sourdough yeasts and lactic acid bacteria are typically faculta-
tive anaerobic and aerotolerant anaerobic/microaerophilic
microorganisms, respectively. Therefore, there is no need to
incubate plates in an atmosphere whose gas composition differs
from that of laboratory. However, if a spread-plate inoculation
has been used, instead of the pour-plate inoculation described
here, it could be preferable to incubate plates containing selec-
tive/elective media for lactic acid bacteria in a microaerophilic
or strictly anaerobic atmosphere.
17. Colonies of lactic acid bacteria are usually 1–3 mm of diameter
and are white to whitish in color. Yellow colonies on media for
lactic acid bacteria usually are staphylococcal colonies. Colonies
of yeasts are usually 2–5 mm of diameter and are cream to
whitish in color. However, at least for those marketed media,
it is advisable to refer to the colony description available on the
manufacturer’s manual. It is understood that colonies with a
too different aspect (e.g., staphylococci) from the typical colo-
nies shall not be considered during counting.
18. The most accurate estimation of cell density requires the anal-
ysis of three biological replicates of sourdough sample. To do
that, you shall just simply repeat the procedure from 3.2 to 3.5
for two more aliquots of sample. It is understood that averag-
ing the estimated cell densities of the three biological replicates
will provide you with the most reliable results.

References
1. Gobbetti M (1998) Interactions between lactic 3. Minervini F, Di Cagno R, Lattanzi A, De
acid bacteria and yeasts in sourdoughs. Trends Angelis M, Antonielli L, Cardinali G et al
Food Sci Technol 9:267–274 (2012) Lactic acid bacterium and yeast micro-
2. Vera A, Rigobello V, Demarigny Y (2009) Com- biotas of 19 sourdoughs used for traditional/
parative study of culture media used for sour- typical Italian breads: interactions between
dough lactobacilli. Food Microbiol 26:728–733 ingredients and microbial species diversity. Appl
Environ Microbiol 78:1251
 Enumerating Lactic Acid Bacteria and Yeasts 27

4. Dinardo FR, Minervini F, De Angelis M, (species Triticum durum and Triticum aestivum)
Gobbetti M, G€anzle MG (2019) Dynamics of sourdoughs of Southern Italy. Int J Food Micro-
Enterobacteriaceae and lactobacilli in model biol 64:95–104
sourdoughs are driven by pH and concentra- 7. Kline L, Sugihara TF (1971) Microorganisms of
tions of sucrose and ferulic acid. LWT-Food Sci the San Francisco sour dough bread process.
Technol 114:108394 II. Isolation and characterization of undescribed
5. Pepe O, Villani F, Coppola S (2001) Differential bacterial species responsible for the souring
viable count of mixed starter cultures of lactic activity. Appl Microbiol 21:459–465
acid bacteria in doughs by using modified Chal- 8. Meroth CB, Walter J, Hertel C, Brandt MJ,
mers medium. Microbiol Res 155:351–354 Hammes WP (2003) Monitoring the bacterial
6. Corsetti A, Lavermicocca P, Morea M, population dynamics in sourdough fermentation
Baruzzi F, Tosti N, Gobbetti M (2001) Pheno- processes by using PCR- Denaturing gradient
typic and molecular identification and clustering gel electrophoresis. Appl Environ Microbiol
of lactic acid bacteria and yeasts from wheat 69:475–482
 Chapter 3

Culture-Independent Estimation of Lactic Acid Bacteria

and Yeasts
Erica Pontonio and Carlo Giuseppe Rizzello

Abstract
The population balances inside microbial consortia are highly variable from one sourdough ecosystem to
another. In general, growth rate and yield of microorganisms are governed by a multitude of ecological
factors. Indeed, in-depth studies of the biodiversity of the microbiota of traditional sourdough products are
interesting from an academic and industrial point of view. Since not all sourdough microorganisms are easily
cultivated in common laboratory media, modern molecular approaches such as the omics methodologies
may help to further unravel sourdough fermentation processes. The application of metagenetics confirmed
the dominance of species belonging to genera of Leuconostoc, Weissella, and ex-Lactobacillus for lactic acid
bacteria and Saccharomyces and Kazachstania for yeasts. Here, detailed culture-independent methods
commonly used to study the sourdough microbiota are described.

Key words Lactic acid bacteria, Yeast, Sourdough, Microbiological characteristics

1 Introduction

Lactic acid bacteria and yeast, considered the dominant sourdough

microbiota, affect, through their metabolic pathways, the techno-
logical, nutritional, and functional properties as well as the organo-
leptic profiles and shelf-life of sourdough products. Since the
microbial composition and dynamics through preparation, propa-
gation, and storage of sourdough are influenced by several ecologi-
cal determinants, the investigation of the main microbial species is
of main importance to better address technological parameters as
well as tailored selection of microbial starters. Several methods,
based on culture-dependent and independent approaches, have
been developed and optimized to characterize microbial consortia.
Nevertheless, classical plating methods to monitor microorganisms
are laborious and time-consuming, especially when investigation of

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

29
30 Erica Pontonio and Carlo Giuseppe Rizzello

the microbial populations dynamics throughout time is needed.

Most importantly, culture-dependent methods are not effective if,
for any reason, microorganisms are not cultivable. Thus, culture
independent and molecular DNA or RNA-based approaches have
become the essential complement for the identification of the
sourdough population. These methods guarantee a high taxonomic
resolution at species up to the strain level. In this regard, culture-
independent methods allow to study the microorganisms of a given
ecosystem not as cells, but as nucleic acids present. Indeed, instead
of isolating, purifying, and then identifying the microorganisms, in
a culture-independent approach, we will proceed directly with the
extraction of the microbial DNA or RNA from the matrix and the
taxonomic identification, and eventual monitoring will be carried
out starting from the nucleic acids without recourse to culture
media. This type of approach, born essentially in environmental
and soil microbiology, has been transferred and successfully applied
for more than ten years also in food microbiology. There also have
been several applications of culture-independent methods for the
study of the microbial ecology of lactic acid bacteria and yeasts in
sourdough.

2 Materials

2.1 DGGE To perform DGGE analysis, the following solutions are needed:
1. Acrylamide solutions to make lactic acid bacteria and yeast
DGGE gel using 30–60% or 35–50% and 30–60% or 30–55%
gradient gel, respectively:
• 30–35% (100 mL): made by 7.5 or 15 mL of acrylamide/
bisacrylamide (40%), 1 or 2 mL Tris–acetate–EDTA buffer
(TAE) (50×), 12 or 14 mL of formamide, and 12.6 or
14.7 g of urea.
• 50–60% (100 mL): made by 15 mL of acrylamide/bisacry-
lamide (40%), 2 mL Tris–acetate–EDTA buffer (TAE)
(50x), 20 or 24 mL of formamide, and 21 or 25.2 g of urea.
2. Running buffer, circa 7.5 L of TAE (1x).
3. Staining solution: final volume of 500 mL (each gel needs
200 mL of staining solution): mix 5 mL of Tris–HCl 1 M and
1 mL EDTA 0.5 M and fill up to 500 mL in a graduated flask.
Cover with aluminum paper. Add 20 μL of syber green
(orange) immediately before using the solution. Cover each
gel with 200 mL of staining solution and leave under stirring
condition for 30 min.
 Culture-Independent Estimation of Lactic Acid Bacteria and Yeasts 31

3 Methods

3.1 Electrophoretic Molecular fingerprinting techniques are based on the amplification

Techniques of nucleic acids, by PCR, which are then separated by electropho-
retic techniques. The most common technique is polymerase chain
reaction-denaturing gradient gel electrophoresis (PCR-DGGE).
This technique that spread in the 1990s of the last centuries is
based on the separation of PCR amplicons of the same length,
belonging to different microbial species. The literature describing
the application of PCR-DGGE in food microbiology is very exten-
sive. Through the DGGE, DNA fragments of the same length, but
with a different nucleotide base sequence, can be separated. In
general, the target of PCR amplification for fingerprinting complex
microbial communities is represented by variable sequences of
genes encoding ribosomal RNA (rRNA) subunits. In the case of
bacteria, different species have different 16S ribosomal gene
sequences, and for this reason, 16S amplicons of different species
migrate differently in polyacrylamide gel electrophoresis with urea-
formamide gradients (DGGE) or when subjected to a gradient of
thermal denaturation (TGGE). Similarly, 26S amplicon is used for
the yeast microbiota characterization and monitoring. PCR-DGGE
is widely applied in microbial ecology of foods, as it allows to obtain
a fingerprint of the microbial community from a food sample,
photographing its diversity in microbial species without resorting
to isolation. The application of the technique envisages, initially, a
direct extraction of microbial DNA (or RNA) from a food, obtain-
ing a mixture containing DNA/RNA of the various recurrent
microbial species. Subsequently, the DNA/RNA mixture is used
for PCR amplification of one or more variable regions of the gene
coding for the rRNA, obtaining a mixture of amplicons all the same
length, but with different sequences depending on the microbial
species. The separation in DGGE is based on the decrease of the
electrophoretic mobility of partially denatured double-stranded
DNA molecules in the polyacrylamide gel. The denaturation takes
place by means of increasing concentrations of urea and formamide
in the denaturing polyacrylamide gel and by means of the running
temperature, which can vary between 55 and 65 °C. The fusion of
DNA fragments proceeds in areas called “melting domains”: an
extension of base pairs with an identical melting temperature. Once
a melting domain with the lower melting temperature reaches its
denaturing condition, the double helix transitions to a partially
denatured molecule, causing the migration of the molecule to a
particular location in the polyacrylamide gel to slow down. A
variation in the sequence within these domains causes a different
melting behavior. For this reason, molecules with different
sequences stop in different positions of the gel, being effectively
separated through DGGE. The separation is generally improved by
32 Erica Pontonio and Carlo Giuseppe Rizzello

the addition of a GC-rich sequence, so-called GC-clamp, to one

end of the DNA fragment. A sequence of guanine and cytosine is
added to the 5′ end of one of the PCR primers, co-amplified, and
thus introduced into the amplified DNA fragments. The GC-clamp
represents a high fusion domain (three hydrogen bonds are
involved between guanine and cytosine, while only two are present
between adenine and thymine) preventing the two DNA strands
from completely dissociating into single strands in a short time and
thus favoring a longer electrophoretic run with better band resolu-
tion. The result of the DGGE is an electrophoretic pattern for
which the number of bands is related to the number of species
present in the analyzed microbial community. The DNA bands in
the DGGE pattern can be visualized by staining with ethidium
bromide or other nucleic acid markers. The identity of each species
can be obtained by purification and sequencing of the individual
bands. The result for each band is a sequence of the specific variable
region of the ribosomal RNA, which can be compared with the
commonly available databases obtaining, as a result, a homology
percentage which will indicate the microbial species to which the
band corresponds. By repeating this procedure for each of the
bands present in the fingerprint, it will be possible to obtain a
complete picture of the microbial species present in the food sam-
ple, without having carried out the isolation and characterization of
the isolates. Disadvantages of the DGGE are as follows: (i) the
presence of multiple and slightly variable 16S rRNA operons pro-
vides complex and different patterns for particular strains of a single
species; (ii) the detection limit is restricted and strain-dependent;
and (iii) the intensity of bands does not necessarily correspond with
total counts present in the ecosystem.

Example of Protocol
• Extraction of total bacterial and fungal DNA or RNA from the
sourdough samples using commercial kits [e.g., Wizard Geno-
mic DNA Purification Kit (Promega); Power Soil DNA Isolation
kit (MO-BIO Laboratories); Nucleo Spin Food (Macherey-
Nagel, Germany); DNeasy Plant Mini Kit (Qiagen); FastDNA®
Pro Soil-Direct Kit (MP Biomedicals, CA, USA)].
• Amplification of the extracted nucleic acids using specific pri-
mers for the amplification of 16S (lactic acid bacteria) and
25-28S (yeast) rRNA genes (Table 1): Amplification for lactic
acid bacteria can be made in 50 μL final volume containing
10 μL of buffer 5X, 1 μL of dNTP 10 mM, 0.5 μL of each
primer (100 μM), 0.4 μL of Taq (5 U/μL), 4 μL of MgCl2
(25 mM), 31.6 μL of sterile water, and 2 μL of nucleic acid
sample. The PCR core program is as follows: 95 °C for 3 min,
followed by 30 cycles of 95 °C for 20 s, 65 °C (annealing
Table 1
Not-exhaustive list of the primers used to amplify total sourdough DNA or RNA for DGGE analysis

Primer Sequence (5′ – 3′) Target gene Reference

Lactic acid bacteria
HDA6-fa AAACCGGAGGAAGGTGGGGA 16S [2]
L1395-r CCCGGGAACGTATTCACCG
P1(F)a GCGGCGTGCCTAATACATGC [3]
P2(R) TTCCCCACGCGTTACTCACC
F357a ATTACCGCGGCTGCTGG [4–6]
518R ATTACCGCGGCTGCTGG
Lac1 AGCAGTAGGGAATCTTCCA [7]
Lac2 ATTYCACCGCTACACATG
P1V1a GCGGCGTGCCTAATACATGC [8]
P2V1 TTCCCCACGCGTTACTCACC
V3f CCGGGGGGCGCGCCCCGGGCGGGGCGGGGACGGGGGGCCTACGGGAGGCAGCAG [9]
Uni-0515r ATCGTATTACCGCGGCTGCTGCTGGCA
Yeast
U1 (F)a GTGAAATTGTTGAAAGGGAA 26S [10, 11]
U2 (R) GACTCCTTGGTCCGTGTT
NL1(F)a GCCATATCAATAAGCGGAGGAAAAG [7, 12, 14]
LS2 (R) ATTCCCAAACAACTCGACTC
NL4 (R) GGTCCGTGTTTCAAGACGG [15]
LIEV-fa TTGTTGAAAGGGAAGGG [16]
LIEV-r CATTACGCCAGCATCCT
5.8S-Fa GTGAATCATCGAATCTTTGAAC 18S [17, 18]
28S-1- TATGCTTAAGTTCAGCGGGTA
Culture-Independent Estimation of Lactic Acid Bacteria and Yeasts

F forward, R reverse
a
A GC clamp was attached to the 5′ end of the primer
33
34 Erica Pontonio and Carlo Giuseppe Rizzello

temperature depends on primers used) for 45 s, 72 °C for 1 min,

and 72 °C for 7 min. The elongation step at 72 °C depends on
the length of the DNA/RNA fragment to be amplified. Ampli-
fication for yeast can be made in 50 μL final volume containing
10 μL of buffer 5X, 1 μL of dNTP 10 mM, 0.5 μL of each primer
(100 μM), 0.4 μL of Taq (5 U/μL), 4 μL of MgCl2 (25 mM),
5 μL of KCl, 26.6 μL of sterile water, and 2 μL of nucleic acid
sample. The PCR core program is as follows: 95 °C for 3 min,
followed by 30 cycles of 95 °C for 1 min, 52 °C (annealing
temperature depends on primers used) for 45 s, 72 °C for
1 min, and 72 °C for 7 min. The elongation step at 72 °C
depends on the length of the DNA/RNA fragment to be
amplified.
• Preparation of acrylamide gels and solidification (circa 3–4 h).
• Separation through electrophoresis in running buffer TAE (1×)
for 10 min at 20 V and 15 h at 75 V and 10 min at 20 V and 4 h at
120 V for lactic acid bacteria and yeast, respectively.
• Gel Staining and DGGE profile visualization thorough UV
illumination (Fig. 1).

3.2 Metagenomic More recently, DGGE has been replaced by metagenomics. Also in
this case, the technique is based on the study of the genetic material
(DNA or RNA) directly extracted from the matrix. Starting from
the extracted genetic material, two different approaches can be
followed:

Fig. 1 Example of denaturing gradient gel electrophoresis (DGGE) profiles of sourdoughs. Primers Lac1/Lac2-
GC and NL1-GC/LS1, targeting the region of 16S and 26S rRNA gene of lactic acid bacteria (PANEL A) and
yeasts (PANEL B), were used, respectively. Lanes: M, low-range DNA ladder (Di Cagno et al., 2014)
Culture-Independent Estimation of Lactic Acid Bacteria and Yeasts 35

• Amplicon-Based or Amplicon-Targeted
This approach involves a PCR amplification step of portions of
genes of taxonomic interest (generally, genes coding for ribosomal
subunits). High-throughput sequencing (HTS) methods have
replaced traditional clone library analysis and fingerprinting
techniques.
Although several HTS techniques are known, the workflow for
bacteria and yeast (16S/ITS) meta-amplicon sequencing protocol
contains four steps: (i) template preparation (by isolating and pur-
ifying the original DNA fragments and then creating a DNA
library); (ii) clonal amplification (forming multiple copies by load-
ing the library onto a flow cell and then amplifying the fragments
into clusters); (iii) parallel sequencing (simultaneously sequencing
DNA templates without the requirement for physical separation);
and (iv) bioinformatics and data analysis.
Bioinformatic is considered a multidisciplinary field featuring
knowledge from biology, computer science, mathematics, statistics,
and medicine. It refers to the computer methods and programs
used to understand and use biological and biomedical data. These
covers acquiring, storing, analyzing, and interpreting biological
data (e.g., DNA sequences) [1]. Further details are reported in
Chap. 4.
1. Roche/454 Pyrosequencing
One of the HTS methods is via the Roche/454 pyrosequencer,
which is one of the earliest high-throughput DNA sequencing
techniques. The method utilizes pyrosequencing, which allows for
sequencing-by-synthesis as the sequence read out can be achieved
at the same time as the sequence is extended. Therefore, electro-
phoresis, as used in Sanger sequencing, is not needed to generate a
nucleotide read out of the output. During pyrosequencing, one
nucleotide at a time is washed over copies of the sequence being
determined, with the complimentary nucleotides being
incorporated onto the template strand. The nucleotide additions
release a light signal that can then detect the location and sequence
of the nucleotide being incorporated. While this type of sequencing
is faster and cheaper than Sanger sequencing, there is a known issue
of homopolymer errors where there is a difficulty in distinguishing a
run of bases in a sequence that are identical such as the sequence
GGGG (i.e., the guanine quartet).
2. Illumina Genome Analyzer
The Illumina Genome Analyzer also has a sequencing-by-syn-
thesis concept where the reaction is stopped after each base, a
fluorescent dye is used to read the base label, and the sequence
reaction is then continued with the next base. During clonal ampli-
fication, the new strand is covalently bound to the flow cell. This
36 Erica Pontonio and Carlo Giuseppe Rizzello

new strand can bend and attach to an oligonucleotide that is

complementary to the adaptor sequence at the free end of the
new strand. A second covalently bound reverse strand can then be
synthesized, which is called bridge amplification and can be
repeated to form clusters. More than 200 million clusters per run
can be formed and 150 nucleotides can be sequenced from both
ends of a fragment. This is accomplished by washing away the
synthesized sequence, repeating the bridge amplification cycle for
the reverse of the strand, removing the starting strand, and adding a
new sequencing primer for the second read. This allows for twice
the amount of sequenced data to be generated.

Protocol
• Total DNA/RNA extraction using commercial kit [e.g., Fas-
tDNA® SPIN kit (MP Biomedicals); DNeasy PowerSoil Kit
(Qiagen); OMEGA DNA isolation kit (Omega); RiboPure bac-
terial kit (Ambion RNA, Life Technologies Co); PowerSoil
DNA Isolation Kit (MO BIO laboratories); GenElute Bacterial
Genomic DNA Kit (Sigma-Aldrich); ZymoBIOMICS DNA
Miniprep Kit (Zymo research)] following the manufacturer’s
instructions.
• Amplification of ITS (fungi) 16S (bacteria) DNA using specific
primers (Table 2). Amplification for lactic acid bacteria can be
made in 50 μL final volume containing 10 μL of buffer 5X, 1 μL
of dNTP 10 mM, 0.5 μL of each primer (100 μM), 0.4 μL of Taq
(5 U/μL), 4 μL of MgCl2 (25 mM), 31.1 μL of sterile water, and
2.5 μL of nucleic acid sample. The PCR core program is as
follows: 94 °C for 5 min, followed by 30 cycles of 94 °C for
45 s, 55 °C (annealing temperature depends on primers used)
for 1 min, 72 °C for 1 min, and 72 °C for 7 min. The elongation
step at 72 °C depends on the length of the DNA/RNA fragment
to be amplified. Amplification for yeast can be made in 50 μL
final volume containing 10 μL of buffer 5X, 1 μL of dNTP
10 mM, 0.5 μL of each primer (100 μM), 0.4 μL of Taq (5 U/
μL), 4 μL of MgCl2 (25 mM), 1 μL of KCl, 31.1 μL of sterile
water, and 1.5 μL of nucleic acid sample. The PCR core program
is as follows: 94 °C for 4 min, followed by 36 cycles of 94 °C for
1 min, 52 °C (annealing temperature depends on primers used)
for 1 min, 72 °C for 1 min, and 72 °C for 7 min. The elongation
step at 72 °C depends on the length of the DNA/RNA fragment
to be amplified.
• Libraries preparation and sequencing on the Illumina MiSeq/
Roche/454.
• Bioinformatics for data elaboration.
Table 2
Not-exhaustive list of the primers used to amplify total sourdough DNA or RNA for pyrosequencing analysis

Primer Sequence (5′ – 3′) Target gene Region Platform Reference

Bacteria
F CCTACGGGAGGCAGCAG 16S V3 – V5 454 GS FLX + system [19]
R TCCTCCGCTTATTGATATGC

27F CGTATCGCCTCCCTCGCGCCATCAGXXXXXXXXXXAGAGTTTGAT V1 – V3 [20]

CCTGGCTCAG
534R CTATGCGCCTTGCCAGCCCGCTCAGXXXXXXXXXXATTACCGCG
GCTGCTGG

Gray28F TTTGATCNTGGCTCAG [13, 21–23]

Gray519r GTNTTACNGCGGCKGCTG

28F GAGTTTGATCNTGGCTCAG Illumina MiSeq [24]

388R TGCTGCCTCCCGTAGGAGT

341F CCTACGGGRSGCAGCAG V3 – V4 [25]

806R GGACTACHVGGGTWTCTAAT

F TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGAGACAGCCTAC [26]
GGGNGGCWGCAG
R GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHV
GGGTATCTAATCC

338-F ACTCCTACGGGAGGCAGCAG [27, 28]

806-R GGACTACNNGGGTATCTAAT

D-Bact-0341-b-S-17 CCTACGGGNGGCWGCAG [29]

D-Bact-0008-a-S-16 AGAGTTTGATCMTGGC

515F GTGCCAGCMGCCGCGGTAA V4 [30]

806R GGACTACHVGGGTWTCTAAT

(continued)
Culture-Independent Estimation of Lactic Acid Bacteria and Yeasts
37
Table 2
38

(continued)

Primer Sequence (5′ – 3′) Target gene Region Platform Reference

8F GCCTCCCTCGCGCCATCAGNNNNCTGCTGCCTYCCGTA V1 – V2 454 GS FLX + system [31]

357R GCCTTGCCAGCCCGCTCAGNNNNAGAGTTTGATCCTGGCTCAG
Tru27F CCATCTCATCCCTGCGTGTCTCCGACTCAGNNNNNNNNNNagrg Illumina MiSeq [32]
tttgatymtggctcag
Tru354R CCTATCCCCTGTGTGCCTTGGCAGTCTCAGtgctgcctcccgtaggagt

FUNGI
ITS3F GCATCGATGAAGAACGCAGC ITS 1 Illumina MiSeq [24]
ITS4R TCCTCCGCTTATTGATATGC

SSU-0817 TTAGCATGGAATAATRRAATAGGA [27]

SSU-1196 TCTGGACCTGGTGAGTTTCC

NSA3 AAACTCTGTCGTGCTGGGGATA 454 GS FLX titanium [33]

58A2R CTGCGTTCTTCATCGAT

BITS ACCTGCGGARGGATCA V7 – V8 Illumina MiSeq [29]

Erica Pontonio and Carlo Giuseppe Rizzello

B58S3 GAGATCCRTTGYTRAAAGTT

ITS1F CTTGGTCATTTAGAGGAAGTAA 2 n.s. [34]

2043R GCTGCGTTCTTCATCGATGC

ITS1-1F-F CTTGGTCATTTAGAGGAAGTAA 1 Illumina MiSeq [30]

ITS1-1F-R GCTGCGTTCTTCATCGATGC

FR1 AICCATTCAATCGGTAIT 18S [29]

FF390 CGATAACGAACGAGACCT

F forward, R reverse, n.s. not-specified

 Culture-Independent Estimation of Lactic Acid Bacteria and Yeasts 39

Due to the large number of sequences obtained from a single

plate (or a small portion of a plate such as one-quarter or
one-eighth) and the increasing read-length of sequences obtained
HTS approaches offer an increase in the depth of coverage and the
ability to detect particularly rare phylotypes, potentially providing
more reliable estimates of relative abundance and assessment of
diversity indices, such as richness and evenness, compared to tradi-
tional clone libraries. A mixture of amplicons will be obtained from
the various microbial species present in the sample (defined as
operational taxonomic units or OTUs). After sequencing, it will
be possible to obtain the nucleotide sequence of the various ampli-
cons and therefore to trace the taxonomic identification of the
species present after comparison with specific databases.
Unlike approaches based on electrophoresis, it will also be
possible to obtain a relative quantification of the different species:
the number of sequences obtained for each species will be propor-
tional to the number of copies of the amplified gene present in the
extracted DNA (and therefore indirectly to the number of cells of a
specific microbial species present in the sample). Therefore, it will
be possible to obtain the relative abundance of each species in the
analyzed sample (Fig. 2). Another advantage is the greater sensitiv-
ity compared to electrophoretic methods.

Fig. 2 Principal coordinates analysis based on weighted Unifrac distances of 16S RNA gene sequences of
doughs (A – B) (after mixing and before fermentation) and sourdoughs after 1, 2, 5, 7, and 14 days of
propagation
40 Erica Pontonio and Carlo Giuseppe Rizzello

• Shotgun Approach ( for Details See Chap. 2.3).

This involves the fragmentation of all the genetic material
extracted from the sample. The genomic fragments will be
sequenced directly, without performing selection by PCR. After
sequencing, nucleotide sequences will be obtained which can be
assembled, reconstructing the microbial genomes initially present
in the sample and, also in this case, it will be possible to identify and
quantify the microbial species present using specific databases. This
technique has innumerable advantages: it allows contextually to
identify all the microorganisms present (bacteria, archaea, fungi,
protozoa, etc.); having reconstructed the genomes, it is potentially
possible to identify the biotype of the microorganism; and it allows
to obtain information on the functional potential of the microbial
community, reconstructing genes, and metabolic pathways.
Other techniques have been developed to study the microbial
composition and changes during sourdough microbiota. Among
these, the genome-probing microarray (GPM) was developed for
quantitative, high-throughput monitoring of community dynamics
of lactic acid bacteria by depositing 149 microbial genomes as
probes on a glass slide. DNA microarrays are genomic technologies
that are commonly applied to the exploration of genome-wide
transcriptional profiles. Their application has also been extended
into the realms of environmental microbiology and microbial ecol-
ogy as a substitute for existing molecular tools studied the micro-
bial dynamics during sourdough fermentation, using back-slopping
procedure.
Although the knowledge on molecular-based approaches for
the detection and quantification of microorganisms, and their
applicability to sourdough system, is improving, the researchers
agree on the combination of culture-dependent and culture-
independent methods. This approach would provide a more com-
plete overview of a complex system such as the sourdough.

References

1. Luscombe NM, Greenbaum D, Gerstein M DNA probes. Appl Environ Microbiol 57:
(2001) What is bioinformatics? A proposed 3390–3393
definition and overview of the field. Methods 4. Van der Meulen R, Scheirlinck I, Van Schoor A
Inf Med 40:346–358 et al (2007) Population dynamics and metabo-
2. Gatto V, Torriani S (2004) Microbial popula- lite target analysis of lactic acid bacteria during
tion changes during sourdough fermentation laboratory fermentations of wheat and spelt
monitored by DGGE analysis of 16S and 26S sourdoughs. 2007. Appl Environ Microbiol
rRNA gene fragments. Ann Microbiol 54:31– 73:4741–4750
42 5. Gafan GP, Spratt DA (2005) Denaturing gra-
3. Klijn N, Weerkamp AH, de Vos WM (1991) dient gel electrophoresis gel expansion
Identification of mesophilic lactic acid bacteria (DGGEGE) – an attempt to resolve the limita-
by using polymerase chain reaction-amplified tions of co-migration in the DGGE of complex
variable regions of 16S rRNA and specific
 Culture-Independent Estimation of Lactic Acid Bacteria and Yeasts 41

polymicrobial communities. FEMS Microbiol 18. Zhang G, Sadiq FA, Zhu L et al (2015) Inves-
Lett 253:303–307 tigation of microbial communities of Chinese
6. Baek HW, Bae JH, Lee YG et al (2021) sourdoughs using culture-dependent and
Dynamic interactions of lactic acid bacteria in DGGE approaches. J Food Scie 80:M2535–
Korean sourdough during back-slopping pro- M2542
cess. J Appl Microbiol 131:2325–2335 19. Liu T, Li Y, Chen J et al (2016) Prevalence and
7. Di Cagno R, Pontonio E, Buchin S et al (2014) diversity of lactic acid bacteria in Chinese tradi-
Diversity of the lactic acid bacterium and yeast tional sourdough revealed by culture depen-
microbiota in the switch from firm- to liquid- dent and pyrosequencing approaches. LWT –
sourdough fermentation. Appl Environ Micro- Food Scie Technol 68:91–97
biol 80:3161–3172 20. Lhomme E, Orain S, Courcoux P et al (2015)
8. Reale A, Di Renzo T, Boscaino F et al (2019) The predominance of Lactobacillus sanfrancis-
Lactic acid bacteria biota and aroma profile of censis in French organic sourdoughs and its
Italian traditional sourdoughs from the Irpi- impact on related bread characteristics. Int J
nian area in Italy. Front Microbiol 10:1621 Food Microbiol 213:40–48
9. Ruiz Rodrı́guez L, Pingitore EV, Rollan G et al 21. Ercolini D, De Filippis F, La Storia A et al
(2016) Biodiversity and technological- (2012) “Remake” by high-throughput
functional potential of lactic acid bacteria sequencing of the microbiota involved in the
isolated from spontaneously fermented quinoa production of water buffalo mozzarella cheese.
sourdoughs. J Appl Microbiol 120:1289–1301 Appl Environ Microbiol 78:8142–8145
10. Sandhu GS, Kline BC, Stockman L et al (1995) 22. Ercolini D, Pontonio E De Filippis F et al
Molecular probes for diagnosis of fungal infec- (2013) Microbial ecology dynamics during
tions. J Clin Microbiol 33:2913–2919 rye and wheat sourdough preparation. Appl
11. Meroth BC, Hammes WP, Hertel C (2003) Environ Microbiol 79: 7827–7836
Identification and population dynamics of 23. Minervini F, Lattanzi A, De Angelis M et al
yeasts in sourdough fermentation processes by (2012) Influence of artisan bakery- or
PCR-denaturing gradient gel electrophoresis. laboratory-propagated sourdoughs on the
Appl Environ Microbiol 69:7453–7461 diversity of lactic acid bacterium and yeast
12. Cocolin L, Bisson LF, Mills DA (2000) Direct microbiotas. Appl Environ Microbiol 78:
profiling of the yeast dynamics in wine fermen- 5328–5340
tations. FEMS Microbiol Lett 189:81–87 24. Coda R, Kianjam M, Pontonio E et al (2017)
13. Rizzello CG, Cavoski I, Turk J et al (2015) Sourdough-type propagation of faba bean
Organic cultivation of Triticum turgidum flour: dynamics of microbial consortia and bio-
subsp. durum is reflected in the flour- chemical implications. Int J Food Microbiol
sourdough fermentation-bread axis. Appl 248:10–21
Environ Microbiol 81:3192–3204 25. Menezes LAA, Savo Sardaro ML, Duarte RTD
14. Syrokou MK, Themeli C, Paramithiotis S et al et al (2020) Sourdough bacterial dynamics
(2020) Microbial ecology of Greek wheat sour- revealed by metagenomic analysis in Brazil.
doughs, identified by a culture-dependent and Food Microbiol 85:103302
a culture-independent approach. Foods 9: 26. Yang Q, Rutherfurd-Markwick K, Mutukumira
1603 AN et al (2021) Identification of dominant
15. Iacumin L, Cecchini F, Manzano M et al lactic acid bacteria and yeast in rice sourdough
(2009) Description of the microflora of sour- produced in New Zealand. J Curr Res Food Sci
doughs by culture-dependent and culture- 4:729–736
independent methods. Food Microbiol 26: 27. Suo B, Nie W, Wang Y et al (2020) Microbial
128–135 diversity of fermented dough and volatile com-
16. Valmorri S, Tofalo R, Settanni L et al (2010) pounds in steamed bread prepared with tradi-
Yeast microbiota associated with spontaneous tional Chinese starters. LWT- Food Scie
sourdough fermentations in the production of Technol 126:109350
traditional wheat sourdough breads of the 28. Li H, Li Z, Qu J et al (2017) Bacterial diversity
Abruzzo region (Italy). Antonie Van Leeuwen- in traditional Jiaozi and sourdough revealed by
hoek 97:119–129 high-throughput sequencing of 16S rRNA
17. Khot PD, Ko DL, Fredricks DN (2009) amplicons. LWT- Food Scie Technol 81:319–
Sequencing and analysis of fungal rRNA oper- 325
ons for development of broad-range fungal 29. Katsi P, Kosma IS, Michailidou S et al (2021)
PCR assays. Appl Environ Microbiol 75: Characterization of artisanal spontaneous sour-
1559–1565 dough wheat bread from Central Greece:
42 Erica Pontonio and Carlo Giuseppe Rizzello

evaluation of physico-chemical, microbiologi- 32. Oshiro M, Tanaka M, Momoda R et al (2021)

cal, and sensory properties in relation to con- Mechanistic insight into yeast bloom in a lactic
ventional yeast leavened wheat bread. Foods acid bacteria relaying-community in the start of
10:635 sourdough microbiota evolution. Microbiol
30. Wang X, Zhu X, Bi Y et al (2020) Dynamics of Spectr 9:e00662–e00621
microbial community and changes of metabo- 33. Urien C, Legrand J, Montalent P et al (2019)
lites during production of type I sourdough Fungal species diversity in French bread sour-
steamed bread made by retarded sponge- doughs made of organic wheat flour. Front
dough method. Food Chem 330:127316 Microbiol 10:201
31. Bessmeltseva M, Viiard E, Simm J et al (2014) 34. Yan B, Sadiq FA, Cai Y (2019) Microbial diver-
Evolution of bacterial consortia in spontane- sity in traditional type I sourdough and jiaozi
ously started rye sourdoughs during two and its influence on volatiles in Chinese
months of daily propagation. PLoS One 9: steamed bread. LWT- Food Scie Technol 101:
e95449 764–773
 Chapter 4

Shotgun Metagenomic Approaches

Francesco Maria Calabrese and Maria De Angelis

Abstract
The most recent advancements in the knowledge of sourdough ecologic meta-community come from the
uncultured approach based on shotgun metagenomics sequencing and the downstream application of a
series of bioinformatics/statistics tools properly concatenated in analytical pipelines. The bioinformat-
ics analysis of raw read sequences has the aim of obtaining a precise quantification of gene annotation
and taxonomic assignments. The inspection of this data allows for disclosing the contribution of species
into this evolving, heterogeneous, and complex community made by bacteria and yeasts. The comprehen-
sive dataset of taxa and their functional genomic potential is indicative both of metabolic active pathways
and possible contaminations due to environmental factors or to human sourdough making processes. A
dynamic interplay between major and minor taxa in terms of genomic contribution underlies those
processes implying sourdough stability and resilience.
In this view, this chapter elucidates concepts related to those taxa identified as mainly responsible for the
resilience and robustness of this ecological niche. Moreover, following a theoretical workflow, this chapter
provides an up-to date list including the most recent tools commonly used to check read sequence quality,
the taxonomic/functional annotation, and the reconstruction of contigs used to infer the completeness of
the metabolic pathways fundamental in the maintenance of the functional, organoleptic, and nutritional
properties of sourdoughs.

Key words Shotgun sequencing, Sourdough metagenomics, Taxonomic assignment of contigs,

Estimation of taxa abundances, Metabolic pathway reconstruction

1 Introduction

Natural cereal-fermenting metacommunity is made of a heteroge-

neous consortium that includes bacteria and yeasts cohabiting
under specific conditions and that act in a synergistic way.
More specifically, from an ecologic point of view, sourdoughs
from different countries proved to have an associated portfolio of
taxa which may vary in terms of presence/absence and relative
abundances [1].
These ecological niches are shaped by different solicitations
(environmental or introduced with the making process) that push
toward a continuous adaptation of taxa members that act as a

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

43
44 Francesco Maria Calabrese and Maria De Angelis

whole, resilient, and robust organism in response to external

stresses.
In this scenario, specialized taxa members grant for species-
specific functions involved in the maintenance of unique metabolic
pathways. At the same, multiple (redundant) genomic traits, shared
by different community members, ensuring an increased and mas-
sive response of the whole microbiota as a unique entity. The
availability of multiple gene copies from different species constitu-
tes the potential genomic network required in the case of key stone
functions, as it happens for the metabolism of carbohydrates and
amino acids.
The possibility to obtain DNA gene sequences from many
genomes at a time, and in a single run, was something unheard of
until a short time ago. The last decade faced a tremendous increase
in high-throughput sequencing techniques applied to biologic and
food matrices. In fact, nowadays, new chemistries and technologi-
cal improvements bring down the costs for a complete metage-
nomics sequencing in a time-effective way. Further, the
downstream steps directed at data processing and bioinformatics
analyses are usually handled by specific pipelines that consecutively
iterate a series of scripts/packages and result in compact and easy
output formats.
In uncultured approaches, the bioinformatics analysis of raw
reads obtained from shotgun sequencing leads to taxonomic anno-
tation of genes and at the same time allows for a more specific
identification of less abundant taxa when compared with 16S
rRNA target sequencing. The genome coverage relative to
thousands of genes assembled in large genomic pieces, named as
contigs, inevitably impact the reliability and accuracy of similarity
searching methods that are advisedly used to infer the proper
taxonomy. More in details, the taxa belonging is inferred by explor-
ing the sequence similarity versus ad hoc customized databases. The
more complete and non-redundant the database, the faster and
accurate the search of DNA strings.
The sourdough metacommunity of fermented sourdough,
namely the “fermentome,” has been recently reconstructed based
on a multiomics approach combining the metagenomics with the
culturomics, i.e., the screening and isolation of cultivable bacteria
and yeasts in eight spontaneous sourdoughs [2].
Thanks to the convergence between culturomics and metage-
nomics analyses, the classification of newly emerging taxa into
different bins became feasible. This reliable resource extends the
known spectrum of colonizing bacteria and yeasts to uncultivable
bacteria [3]. Genomic details from uncultured taxa hold significant
perspective that helps scientists to better cultivate them, leading to
valuable insights crucial in filling the gap of this environmental
“uncultured majority.”
 Shotgun Metagenomics on Sourdough Matrices 45

The result of metagenomics technological advancement finds a

fertile ground in the annotation of genes, at the species taxo-
nomic level and, where possible, down to the strain level. Sour-
dough autochthonous microbiota is mainly composed of diverse
LAB species, proteobacteria, staphylococci, and yeasts [4]. The
presence/absence criteria applied to taxa from both bacteria and
yeast in mature sourdoughs allow for identifying core and dispens-
able species and, based on the number of gene and transcripts
related to the main pathways relevant in maintaining sourdough
key properties, these two classes further give origin to additional
sub-branches of dominant, subdominant, and satellite members
[2]. Noteworthy, the metagenomics approach is of fundamental
importance in elucidating the contribution of satellite members
that, although marked with a minor content of genes that could
be potentially transcribed, may be metabolically active and
equipped with unique genes and enzymes.
As a final remark, it is possible to retrieve individual genomes
from the batch of identified genes based on metagenomics data
[5]. This reconstruction completeness is dependent on the cover-
age of concatenated fragments. If compared with marker gene
approach (metagenetic—16S rRNA), the higher certainty of
assigning a species based on larger genomic pieces is a finely
tuned and well-established practice that leads to species identifica-
tion with dedicated parameters including completeness (occu-
pancy), uniqueness, and contamination of bins.
This chapter in the methods section provides a schematic list of
the most recently developed software and their relative alternatives
useful for the processing of metagenomics short reads raw sequenc-
ing data in terms of a bioinformatics workflow.

2 Materials

DNA Commercial Extraction Kits (Alternatives)

• FastDNA® SPIN kit (MP Biomedicals).
• DNeasy PowerSoil Kit (Qiagen).
• OMEGA DNA isolation kit (Omega).
• PowerSoil DNA Isolation Kit (MO BIO laboratories).
• GenElute Bacterial Genomic DNA Kit (Sigma-Aldrich).
• ZymoBIOMICS DNA Miniprep Kit (Zymo research).
46 Francesco Maria Calabrese and Maria De Angelis

Kit for High-Quality DNA Isolation from Samples

Host DNA removal can be obtained using one of the following

alternatives:
• Maxwell® RSC PureFood Pathogen Kit (Promega).
• MoBio PowerPlant Pro® DNA Isolation Kit (MoBio).
• Qiagen DNeasyR Plant Mini Kit (Qiagen).
• MoBio PowerSoil™ DNA Purification Kit (MoBio).

DNA Library Construction

A pre-step-based mechanical or enzymatic shearing allowed for the

obtainment of fragmented DNA samples for NGS (Next Genera-
tion Sequencing) high-throughput sequencing. A multi-step pro-
tocol accounts for the following sub-steps/materials, each one with
its specific requirements:
• Enzymes.
• Buffers for end preparation.
• Adapter ligation.
• Adapter concentration test, ligation efficiency assay.
• Library amplification.
Library preparation kits (alternatives) for shotgun metagenomic
sequencing:
• Nextera XT DNA library prep kit (Illumina).
• xGen™ DNA Lib Prep EZ (IDT).
Bioinformatics resources.
In a classic bioinformatics approach, specific software are dedi-
cated to the necessary steps going from read quality checking
(low-quality reads are removed) to gene annotation: concatenated
paired reads are processed in order to remove adaptors and
barcodes.
Moreover, as an additional procedure, specific software can
handle the combining of genes in clusters with a biologic rationale,
i.e., the “binning procedure”, that operate through the creation of
condensed metabolic pathway objects. By way of example, the
flowchart reported in Fig. 1 summarizes the chained steps of a
customizable pipeline that is based on the connection of inputs
and outputs in a logical way. To ensure the elaboration of huge size
raw sequencing files (from tens to hundreds of Gb) and considering
the high consumption capacity required to handle bioinformatics
computing processes, dedicated servers must necessarily be
equipped with dozens of CPU threads and Gb of RAM memory.
 Shotgun Metagenomics on Sourdough Matrices 47

Fig. 1 Step-by-step bioinformatics workflow for the analysing of metagenomics data. Raw data from the
sequencing are processed thanks to three main steps accounting for assembly, classification and annotation
of genes and their relative taxonomy

Moreover, powerful partitions and (internal or external) hard disks,

with a massive storage power in the order of terabytes (Tb), are
necessary to archive raw, intermediate, and post-processed output
files.
Importantly, the skills of a high-preforming multiprocessor
server include the scalability (increasing the computing capacity in
response to workload processing demand) and the possibility to
split processes on multiple CPU nodes. Increased capacity will
result in time-saving analyses.

3 Methods

Two main steps are required to prepare DNA before sequencing.

Regardless of the type of samples, preparing procedures mainly
consist of DNA extraction and the relative library preparation.
Several efforts have been spent toward the standardization of pro-
cedures and are aimed at reducing the great diversity of available
protocols. At the same time, sequencing companies are providing
users with both standard-based solutions and advanced extraction
methods working on specific needs. Nevertheless, specific adjust-
ments improve the yield of kits originally designed for DNA extrac-
tion from soil.
48 Francesco Maria Calabrese and Maria De Angelis

DNA Extraction and Quality Control

As reported in the previous chapter (e.g., 2.2), DNA can be
extracted from samples using commercial kits [e.g., FastDNA®
SPIN kit (MP Biomedicals); DNeasy PowerSoil Kit (Qiagen);
OMEGA DNA isolation kit (Omega); RiboPure bacterial kit
(Ambion RNA, Life Technologies Co); PowerSoil DNA Isolation
Kit (MO BIO laboratories); GenElute Bacterial Genomic DNA Kit
(Sigma-Aldrich); ZymoBIOMICS DNA Miniprep Kit (Zymo
research)] optimized to work with food sample types including
sourdough matrices. The setup of best performing kits is based
on DNA yield and quality evaluation. Specifically designed extrac-
tion kits act by capturing DNA with magnetic beads and excluding
organic inhibitors. Bead beating, heating, and chemical lysis are the
classic methods used to prepare DNA. Samples are practically sub-
jected to iterate cycles of vortexing and heating.

Moreover, in order to increase the quality of microbial

taxa detected, host contaminant sequences, that in the specific
case of sourdough derived from plant genomes, need to be filtered.
Although downstream bioinformatics procedures are capable of
recognizing host-derived assignments based on a similarity search
approach, pre-sequencing filtering protocols are needed to remove
undesired host DNA and enrich for microbial DNA.
Osmotic lysis followed by propidium monoazide (PMA) treat-
ment is an effective way for enriching microbial sequence data in
shotgun metagenomics. In the case of food microbiota sequencing,
host (plant) DNA depletion can be obtained by using various kits
[e.g., Maxwell® RSC PureFood Pathogen Kit (Promega); MoBio
PowerPlant Pro® DNA Isolation Kit (MoBio); Qiagen DNeasyR
Plant Mini Kit (Qiagen); MoBio PowerSoil™ DNA Purification Kit
and MoBio UltraClean® Soil DNA Isolation Kit].
Once extracted, its concentration and quality are checked
before the library preparation. To estimate DNA concentration,
yield, and purity, the available methods utilized include absorbance,
agarose gel electrophoresis, and fluorescence techniques based on
dyes. In the most commonly used absorbance method, the purity is
obtained as the ratio between the absorbance at 260 nm, preva-
lently due to DNA, divided by the reading at 280 nm (mainly due
to aromatic amino acids). Anyway, due to other molecules that
absorb at 260 nm (mainly RNA and guanidine), this read can be
considered an overestimation. A good value of this ratio oscillates
between 1.7 and 2.0 and needs to be corrected for turbidity (absor-
bance at 320 nm).
 Shotgun Metagenomics on Sourdough Matrices 49

DNA Library Preparation for High-Throughput Sequencing

Library preparation kits [e.g., Nextera XT DNA library prep kit
(Illumina); HiPurA soil DNA isolation kit (Himedia)] are used to
barcode and mix DNA from different samples. The mix (multi-
plexed DNA) is then sequenced using Illumina technolo-
gies (Hiseq and Nextera Illumina sequencing platforms) for
paired end reads.

The number of multiplexed samples is dependent on the sam-

ple type (host DNA) or complex communities.
The output of a sequencing run is stored in compact file with
specific fastQ format. Each read sequence is provided with a relative
quality string that reports the per base error probability derived
from the sequencing step.
Concatenated paired reads are processed in order to remove
adaptors, barcodes, and low-quality reads, as well as contaminant
sequences (typically, host DNA).

Sequencing Depth of Coverage

One of the crucial questions linked to the sequencing is the esti-
mate of the required read depth. Noteworthy, among the factors to
be considered, we must take into account, first of all, the matrix
species diversity. To do this, we should have an idea of the number
of the putative species; each taxa should have to be accounted for
individually. Other important factors are the genomic size and the
relative abundance of each species. In the case of de novo sequenc-
ing (the metagenome is sequenced for the first time without any a
priori knowledge), it is difficult to have any understanding of all
these factors.

Just as a simplistic calculation, the below reported equation

reports the number of total Gb that will be occupied on the
sequencing flow cell in the case of a metagenome composed of
50 (most representative) bacterial species, with an estimated
genome size of 2 Mb and a coverage read depth of 50x (each
sequence will be present in the output 50 times).
50 most representative bacterial species × 50x × 2 Mb = 5Gb

Bioinformatics Data Processing

Following an up-to date inventory of the most used software and
databases for metagenomics analyses, we here provide a list inclu-
sive of the more recent alternatives, for each of the main step.

The link address for downloading the software source code and
the relative paper is also provided.
50 Francesco Maria Calabrese and Maria De Angelis

Comprehensive sequence database for similarity searching:

taxonomic assignment
GenBank [6]
functional assignment !
eggNOG [7]
KEGG [8]
Pfam [9]

Raw sequence quality check:

FastQC ! [Link]
jects/fastqc/
fastp ! [Link] [10]
Trimmomatic ! [Link] [11]
SOAPnuke ! [Link] [12]
Assembly software working on short reads:
MetaVelvet-DL ! [Link]
[13]
metaSPAdes ! [Link] [14]
Megahit ! [Link] [15]
Contig statistics and quality control read mapping:
prinseq ! [Link] [16]
Picard ! [Link]
RNAs prediction:
Barrnap [Link]
TopHat ! [Link] [17]
16S rRNA taxonomic classification:
RDP classifier ! [Link] [18]
QIIME2 ! [Link] [19]
Mothur ! [Link] [20]
SINTAX ! [Link]
[Link] [21]
SPINGO ! [Link] [22]
IDTAXA ! [Link]
html [23]
Kraken2 ! [Link] [24]
 Shotgun Metagenomics on Sourdough Matrices 51

tRNA/tmRNA sequence prediction:

Aragorn ! [Link]
ter/aragorn1.2.36.c [25]
tRNAscan-SE ![Link] [26]
ORFs prediction:
Prodigal ! [Link] [27]
Balrog ! [Link] [28]
Meta-MFLD ! [Link]
[29]
CNN-MGP ! [Link] [30]
Similarity searches versus databases:
Diamond ! [Link] [31]
MMseqs2 ! [Link] [32]
HMM homology searches:
HMMER3 ! [Link] [33]
Read mapping against contigs:
Bowtie2 ! [Link]
shtml [34]
HISAT2 ! [Link] [35]
Contig binning:
VAMB ! [Link] [36]
MAGO ! [Link] [37]
MaxBin2 ! [Link] [38]
Metabat2 ! [Link]
master/ [39]
MyCC ! [Link]
MyCC/
SolidBin ! [Link] [40]
METAMVGL ! [Link]
METAMVGL [41]
Bin3C ! [Link] [42]
HiCBin ! [Link] [43]
GraphBin ! [Link] [44]
Combination of binning results:
DAS Tool ! [Link] [5]
MetaBinner ! [Link] [45]
MetaWRAP ! [Link] [46]
52 Francesco Maria Calabrese and Maria De Angelis

References
1. Comasio A, Verce M, Van Kerrebroeck S, De 2114–2120. [Link]
Vuyst L (2020) Diverse microbial composition formatics/btu170
of sourdoughs from different origins. Front 12. Chen Y, Chen Y, Shi C et al (2018) SOAPnuke:
Microbiol 11:1212. [Link] a MapReduce acceleration-supported software
3389/fmicb.2020.01212 for integrated quality control and preproces-
2. Calabrese FM, Ameur H, Nikoloudaki O et al sing of high-throughput sequencing data.
(2022) Metabolic framework of spontaneous GigaScience 7:1–6. [Link]
and synthetic sourdough metacommunities to 1093/gigascience/gix120
reveal microbial players responsible for resil- 13. Liang K, Sakakibara Y (2021) MetaVelvet-DL:
ience and performance. Microbiome 10:148. a MetaVelvet deep learning extension for de
[Link] novo metagenome assembly. BMC Bioinfor-
01301-3 matics 22:427. [Link]
3. Liu S, Moon CD, Zheng N et al (2022) s12859-020-03737-6
Opportunities and challenges of using metage- 14. Nurk S, Meleshko D, Korobeynikov A, Pevz-
nomic data to bring uncultured microbes into ner PA (2017) metaSPAdes: a new versatile
cultivation. Microbiome 10:76. [Link] metagenomic assembler. Genome Res 27:
org/10.1186/s40168-022-01272-5 824–834. [Link]
4. De Vuyst L, Comasio A, Kerrebroeck SV 213959.116
(2023) Sourdough production: fermentation 15. Li D, Liu C-M, Luo R et al (2015) MEGA-
strategies, microbial ecology, and use of HIT: an ultra-fast single-node solution for
non-flour ingredients. Crit Rev Food Sci Nutr large and complex metagenomics assembly via
63:2447–2479. [Link] succinct de Bruijn graph. Bioinforma Oxf Engl
10408398.2021.1976100 31:1674–1676. [Link]
5. Sieber CMK, Probst AJ, Sharrar A et al (2018) informatics/btv033
Recovery of genomes from metagenomes via a 16. Schmieder R, Edwards R (2011) Quality con-
dereplication, aggregation and scoring strategy. trol and preprocessing of metagenomic data-
Nat Microbiol 3:836–843. [Link] sets. Bioinforma Oxf Engl 27:863–864.
10.1038/s41564-018-0171-1 [Link]
6. Clark K, Karsch-Mizrachi I, Lipman DJ et al btr026
(2016) GenBank. Nucleic Acids Res 44:D67– 17. Kim D, Pertea G, Trapnell C et al (2013)
D72. [Link] TopHat2: accurate alignment of transcrip-
7. Huerta-Cepas J, Szklarczyk D, Forslund K et al tomes in the presence of insertions, deletions
(2016) eggNOG 4.5: a hierarchical orthology and gene fusions. Genome Biol 14:R36.
framework with improved functional annota- [Link]
tions for eukaryotic, prokaryotic and viral 18. Wang Q, Garrity GM, Tiedje JM, Cole JR
sequences. Nucleic Acids Res 44:D286– (2007) Naive Bayesian classifier for rapid
D293. [Link] assignment of rRNA sequences into the new
gkv1248 bacterial taxonomy. Appl Environ Microbiol
8. Kanehisa M, Goto S (2000) KEGG: Kyoto 73:5261–5267. [Link]
encyclopedia of genes and genomes. Nucleic AEM.00062-07
Acids Res 28:27–30. [Link] 19. Bokulich NA, Kaehler BD, Rideout JR et al
1093/nar/28.1.27 (2018) Optimizing taxonomic classification of
9. Finn RD, Coggill P, Eberhardt RY et al (2016) marker-gene amplicon sequences with QIIME
The Pfam protein families database: towards a 2’s q2-feature-classifier plugin. Microbiome 6:
more sustainable future. Nucleic Acids Res 44: 90. [Link]
D279–D285. [Link] 0470-z
gkv1344 20. Introducing Mothur: Open-Source, Platform-
10. Chen S, Zhou Y, Chen Y, Gu J (2018) Fastp: Independent, Community-Supported Soft-
an ultra-fast all-in-one FASTQ preprocessor. ware for Describing and Comparing Microbial
Bioinformatics 34:i884–i890. [Link] Communities | Applied and Environmental
org/10.1093/bioinformatics/bty560 Microbiology. [Link]
11. Bolger AM, Lohse M, Usadel B (2014) Trim- full/10.1128/AEM.01541-09. Accessed
momatic: a flexible trimmer for Illumina 10 May 2023
sequence data. Bioinforma Oxf Engl 30:
 Shotgun Metagenomics on Sourdough Matrices 53

21. Edgar R (2016) SINTAX: a simple 33. Eddy SR (2009) A new generation of homol-
non-Bayesian taxonomy classifier for 16S and ogy search tools based on probabilistic infer-
ITS sequence ence. Genome Inform Int Conf Genome
22. Allard G, Ryan FJ, Jeffery IB, Claesson MJ Inform 23:205–211
(2015) SPINGO: a rapid species-classifier for 34. Langmead B, Salzberg SL (2012) Fast gapped-
microbial amplicon sequences. BMC Bioinfor- read alignment with Bowtie 2. Nat Methods 9:
matics 16:324. [Link] 357–359. [Link]
s12859-015-0747-1 1923
23. Murali A, Bhargava A, Wright ES (2018) 35. Kim D, Paggi JM, Park C et al (2019) Graph-
IDTAXA: a novel approach for accurate taxo- based genome alignment and genotyping with
nomic classification of microbiome sequences. HISAT2 and HISAT-genotype. Nat Biotech-
Microbiome 6:140. [Link] nol 37:907–915. [Link]
1186/s40168-018-0521-5 s41587-019-0201-4
24. Lu J, Salzberg SL (2020) Ultrafast and accurate 36. Nissen JN, Johansen J, Allesøe RL et al (2021)
16S rRNA microbial community analysis using Improved metagenome binning and assembly
Kraken 2. Microbiome 8:124. [Link] using deep variational autoencoders. Nat Bio-
org/10.1186/s40168-020-00900-2 technol 39:555–560. [Link]
25. Laslett D, Canback B (2004) ARAGORN, a 1038/s41587-020-00777-4
program to detect tRNA genes and tmRNA 37. Murovec B, Deutsch L, Stres B (2020)
genes in nucleotide sequences. Nucleic Acids Computational framework for high-quality
Res 32:11–16. [Link] production and large-scale evolutionary analy-
nar/gkh152 sis of metagenome assembled genomes. Mol
26. Chan PP, Lowe TM (2019) tRNAscan-SE: Biol Evol 37:593–598. [Link]
searching for tRNA genes in genomic 1093/molbev/msz237
sequences. Methods Mol Biol Clifton NJ 38. Wu Y-W, Simmons BA, Singer SW (2016)
1962:1–14. [Link] MaxBin 2.0: an automated binning algorithm
4939-9173-0_1 to recover genomes from multiple metage-
27. Hyatt D, Chen G-L, LoCascio PF et al (2010) nomic datasets. Bioinforma Oxf Engl 32:605–
Prodigal: prokaryotic gene recognition and 607. [Link]
translation initiation site identification. BMC ics/btv638
Bioinform 11:119. [Link] 39. Kang DD, Li F, Kirton E et al (2019) Meta-
1186/1471-2105-11-119 BAT 2: an adaptive binning algorithm for
28. Sommer MJ, Salzberg SL (2021) Balrog: a robust and efficient genome reconstruction
universal protein model for prokaryotic gene from metagenome assemblies. PeerJ 7:e7359.
prediction. PLoS Comput Biol 17:e1008727. [Link]
[Link] 40. Wang Z, Wang Z, Lu YY et al (2019) SolidBin:
1008727 improving metagenome binning with semi-
29. Zhang S-W, Jin X-Y, Zhang T (2017) Gene supervised normalized cut. Bioinformatics 35:
prediction in metagenomic fragments with 4229–4238. [Link]
deep learning. Biomed Res Int 2017: formatics/btz253
4740354. [Link] 41. Zhang Z, Zhang L (2021) METAMVGL: a
4740354 multi-view graph-based metagenomic contig
30. Al-Ajlan A, El Allali A (2019) CNN-MGP: binning algorithm by integrating assembly
convolutional neural networks for metage- and paired-end graphs. BMC Bioinform 22:
nomics gene prediction. Interdiscip Sci Com- 378. [Link]
put Life Sci 11:628–635. [Link] 04284-4
1007/s12539-018-0313-4 42. DeMaere MZ, Darling AE (2019) bin3C:
31. Buchfink B, Xie C, Huson DH (2015) Fast and exploiting Hi-C sequencing data to accurately
sensitive protein alignment using DIAMOND. resolve metagenome-assembled genomes.
Nat Methods 12:59–60. [Link] Genome Biol 20:46. [Link]
1038/nmeth.3176 1186/s13059-019-1643-1
32. Steinegger M, Söding J (2017) MMseqs2 43. Du Y, Sun F (2022) HiCBin: binning metage-
enables sensitive protein sequence searching nomic contigs and recovering metagenome-
for the analysis of massive data sets. Nat Bio- assembled genomes using Hi-C contact maps.
technol 35:1026–1028. [Link] Genome Biol 23:63. [Link]
1038/nbt.3988 1186/s13059-022-02626-w
54 Francesco Maria Calabrese and Maria De Angelis

44. Mallawaarachchi V, Wickramarachchi A, Lin Y genomes from complex microbial commu-

(2020) GraphBin: refined binning of metage- nities. Genome Biol 24:1. [Link]
nomic contigs using assembly graphs. Bioinfor- 10.1186/s13059-022-02832-6
matics 36:3307–3313. [Link] 46. Uritskiy GV, DiRuggiero J, Taylor J (2018)
1093/bioinformatics/btaa180 MetaWRAP—a flexible pipeline for genome-
45. Wang Z, Huang P, You R et al (2023) Meta- resolved metagenomic data analysis. Micro-
Binner: a high-performance and stand-alone biome 6:158. [Link]
ensemble binning method to recover individual s40168-018-0541-1
 Chapter 5

Determination of pH and Titratable Acidity

Kashika Arora and Raffaella Di Cagno

Abstract
Acidification of sourdough is a distinguishing parameter that can vary depending on the ingredients and the
fermentation conditions. This chapter describes a step-by-step procedure to determine the acidity of
sourdough in terms of hydrogen ions (pH) and the buffering capacity defined as total titratable acidity
(TTA). The selection of pH electrode is critical to reproducing the pH measurements at different positions
within the sourdough. This reproducible method can determine the pH and TTA using as low as 10 g of
sourdough per measurement.

Key words Sourdough, Acidity, pH electrode, Titration, Buffering capacity

1 Introduction

The fermentation of a mixture of flour and water, known as

“dough,” gives it a characteristic “sour” (or acidic) taste, hence
the term “sourdough.” The metabolically active lactic acid bacteria
(LAB) during sourdough fermentation secrete organic acids, such
as lactic and acetic acids, in high concentrations with a median value
of 75 and 20 mM, respectively [1], which greatly affect the acidity
of sourdough. Additionally, the type of flour, time, and tempera-
ture of fermentation are the technological parameters that affect
the acidic nature of the sourdough. Consequently, determination
of acidity is critical to the performance of sourdough [1]. Different
techniques for the measurement of acidity in foods was approved by
Food and Drug Administration (FDA) and detailed in the code
21CFR114.90 [2]. Among all the described methods, acidity
determination using pH meters and titration of acids (titratable
acidity) are the most reliable and commonly used. Theoretically,
according to Arrhenius, an acid combines with water (H2O) to
produce hydronium ion (H3O+), whereas a base produces hydroxyl

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

55
56 Kashika Arora and Raffaella Di Cagno

ion (OH-). A pH meter precisely measures the concentration of

H3O+ ions in an acidic environment using a pH electrode. The
selection of pH electrode is crucial to a sensitive and accurate
assessment of acidity in solid/semi-solid food matrices such as
sourdough [3]. On the other hand, titratable acidity, also identified
as total titratable acidity (TTA), refers to the total acidity in foods
determined by titrating the intrinsic acids with a standard base such
as sodium hydroxide (NaOH). The relevance and chemistry of pH
and titratable acidity in analysis of foods have been detailed enor-
mously [4–6]. In simpler words, pH values measure the power of
acidity, while TTA gives information about the influence of any acid
on the overall taste of the foods. Moreover, the buffering capacity
of sourdough, as determined by TTA, is influenced by the type of
flour used in sourdough preparation [7]. Therefore, it is important
to assess the acidity of sourdough in terms of pH as well as TTA.
While the pH measurements are well standardized by the use of
specific pH electrodes, the methodology to estimate titratable acid-
ity in sourdough needs to be refined for the following parameters:
• Amount of distilled water needed to make the dough
suspension.
• Molarity of NaOH for titration.
• Exact pH value to define the end point of titration.
In this chapter, we provide a detailed procedure for measuring
pH and titratable acidity of sourdough prepared with any type of
flour. Considering that an active and well-performing sourdough
will rapidly increase the acidity, the pH and TTA should be always
measured at the start and end of sourdough fermentation. Here, it
is important to note that the pH and TTA are inversely propor-
tional to each other as a low pH (or high acidity) requires a higher
quantity of conjugate base (NaOH) for titration (TTA).

2 Materials

The pH electrode for solids should have a spear tip that can pierce
through the sourdough and measure pH values in a sensitive man-
ner. This type of pH electrode is also easy to clean any sticky dough
particles. Distilled water is used for cleaning the pH electrodes as
well as for preparing the sodium hydroxide (0.1 M NaOH) solu-
tion. Wear gloves throughout the procedure. Take precautions
while handling NaOH solution and dispose the waste in the defined
chemical waste containers. Unused NaOH solution in the titration
burette can be transferred in a glass bottle and stored at room
temperature until further use.
 pH and Titratable Acidity 57

2.1 Common 1. An electronic pH meter.

Reagents 2. pH calibration solutions (pH 4.0, 7.0, and 10.0 units)
provided with the pH meter.
3. Kimwipes or any other soft paper towels.
4. Distilled water in a squeeze bottle.
5. Waste beaker.

2.2 Additional pH meter with a pH electrode for solids (see Note 1).
Reagents for pH
Measurements

2.3 Titratable Acidity 1. 0.1 M NaOH: Dissolve 2 g of NaOH in 500 mL of distilled

water. Store at room temperature up to 3 months.
2. Titration equipment set with a 50 mL glass burette, stand, and
burette clamp.
3. Magnetic stirrer and a magnetic stir bar.
4. pH meter with a pH electrode for liquids (see Note 1).
5. pH calibration solutions (pH 4.0, 7.0, and 10.0 units)
provided with the pH meter.
6. Weighing balance.
7. Kimwipes or any other soft paper towels.
8. 250-mL glass beaker.
9. 100-mL measuring cylinder.
10. Distilled water in a squeeze bottle.
11. Waste beaker.

3 Methods

3.1 pH 1. Calibrate the pH meter using appropriate calibration solutions

(see Note 2).
2. Dry the pH electrode using Kimwipes or soft paper towels
before use for pH measurement.
3. Put the pH electrode in the sourdough with probe completely
covered with the sample. Stabilize the electrode for accurate
measurements and avoid any vibrations around the pH meter.
4. Start the pH measurement at the “Stable” mode and wait until
a stable pH value is displayed on the pH meter screen.
5. Clean the probe using Kimwipes and remove any particles from
the probe before the next measurements.
58 Kashika Arora and Raffaella Di Cagno

Fig. 1 Schematic illustration of experimental setup for total titratable acidity (TTA) determination

6. Record the pH values at three different positions in the sour-

dough (see Note 3).
7. The pH value is expressed as “units.”

3.2 Total Titratable 1. Weigh 10 g of dough in a 250-mL glass beaker.

Acidity (TTA) 2. Add 90 mL of distilled water to it, and stir using a magnetic stir
bar on a magnetic stirrer for 2–5 min.
3. Set the position of burette apparatus, pH meter, and magnetic
stirrer such that it is possible to measure the pH while stirring
and adding the NaOH from the burette drop-by-drop (see
Fig. 1).
4. Now carefully add 0.1 M NaOH in the glass burette up to the
mark of 0 (see Note 4), known as the initial volume.
5. Calibrate the pH meter using appropriate calibration solutions
(see Note 2).
 pH and Titratable Acidity 59

6. After the mixture of dough and water is homogenized (step 2),

measure the pH (initial pH) using a suitable pH electrode in
the “Stable” mode.
7. Keep the pH electrode immersed in the solution in a stable
position to continuously measure the pH when the NaOH
solution is being added from the burette.
8. Once the initial pH is recorded, set the pH meter in
“Continuous” mode.
9. Start adding 0.1 M NaOH drop-by-drop very slowly to titrate
the acidity of the homogenized sourdough. Notice the increase
in pH measurement with the addition of NaOH (see Note 5).
10. Stop the addition of 0.1 M NaOH solution once the pH value
reaches 8.3 units.
11. Check the volume of 0.1 M NaOH left in the burette, known
as the final volume.
12. Calculate the total volume of 0.1 M NaOH added during
titration by subtracting the initial volume from the final
volume.
13. The total titratable acidity (TTA) is expressed as the volume of
0.1 M NaOH/10 g of dough.

4 Notes

1. The pH electrode for solids can be purchased from any com-

pany and should be compatible with the pH meter. The spear
tip of the pH electrode is critical to pH measurements for solid
and semi-solid samples. Handle the pH electrodes with care
and immerse them in the correct solutions when not in use.
2. Usually, a pH meter is calibrated in the order of pH 4.0, 7.0,
and 10.0.
3. You can stir the sourdough with the pH electrode in case of
semi-solid texture for replicate measurements.
4. It is important to note the initial volume of 0.1 M NaOH in the
burette, which should be minimum 20 mL for titrating the
acidity of a sourdough.
5. Continuous stirring of the homogenized sourdough is impor-
tant for titration of acids.
6. pH values should be measured immediately.
7. These procedures can be adopted for any other acidic and solid
food matrices as well.
60 Kashika Arora and Raffaella Di Cagno

References
1. Arora K, Ameur H, Polo A et al (2021) Thirty 5. Tyl C, Sadler GD (2017) pH and titratable
years of knowledge on sourdough fermentation: acidity. In: Nielsen SS (ed) Food analysis.
a systematic review. Trends Food Sci Technol Springer International Publishing, Cham, pp
108:71–83. [Link] 389–406
2020.12.008 6. Nielsen SS (2017) Standard solutions and titrat-
2. CFR – Code of Federal Regulations Title 21. In: able acidity. In: Nielsen SS (ed) Food analysis
h t t p s : // w w w. f d a . g o v / . h t t p s : // w w w. laboratory manual. Springer International Pub-
[Link]/scripts/cdrh/cfdocs/cfcfr/ lishing, Cham, pp 179–184
[Link]?fr=114.90. Accessed 7. Salovaara H, Valjakka T (2007) The effect of
23 Dec 2021 fermentation temperature, flour type, and
3. Thermo Scientific (2014) pH measurement. In: starter on the properties of sour wheat bread.
Handbook Prago Lab, pp 15–30 [Link]
4. Sadler GD, Murphy PA (2010) pH and titratable TB00527.X
acidity. In: Nielsen SS (ed) Food analysis.
Springer, Boston, pp 219–238
 Chapter 6

Determination of Lactic and Acetic Acids and Estimation

of Their Molar Ratio
Andrea Polo and Marco Gobbetti

Abstract
Lactic and acetic acids concentrations and their molar ratio (defined as quotient of fermentation, QF) are
important indicators to describe the sourdough features. Such organic acids, produced by lactic acid
bacteria (LAB) during fermentation, are responsible for lowering the pH, and, consequently, for sourdough
structure, sensory attributes, antifungal properties, and preservation. QF has also relevant effects on sensory
characteristics, flavor, and shelf life of leavened baked products such as bread. In the chapter introduction,
different available methods to determine lactic and acetic acids in sourdough are briefly listed highlighting
advantages and limitations of each approach. Then, the procedure for the detection and quantification of
lactic and acetic acids through high performance liquid chromatography (HPLC) analysis and for the
calculation of QF is detailed.

Key words Lactic acid, Acetic acid, Sourdough, High performance liquid chromatography (HPLC),
Quotient of fermentation

1 Introduction

Biochemical parameters represent important indicators to describe

the sourdough performance. In this section, the method to deter-
mine lactic and acetic acids in sourdough is described, and the
calculation of the deriving quotient of fermentation (QF), that is
defined as the molar ratio between lactic and acetic acids, is also
detailed.
It is known that acidity is an important feature of sourdough.
Organic acids, namely lactic and acetic acids, produced by lactic
acid bacteria (LAB) during fermentation are responsible for lower-
ing the pH. The concentration of lactic acid is considered favorable
to a more elastic gluten structure of sourdough [1]. The concen-
tration of acetic acid is another important functional trait as it
contributes to the enhancement of sensory properties, has antifun-
gal property, and exerts a control effect against molds. On the other

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

61
62 Andrea Polo and Marco Gobbetti

hand, at the common sourdough pH (3.7–4.0), acetic acid is

present mainly as undissociated lipophilic and membrane-diffusible
form, and, combined with ethanol, it can have a negative effects on
yeast growth [2]. Traditionally, it is accepted that higher the levels
of acetic acid, the better the resulting flavor of sourdough. There-
fore, actions that favor the acetate kinase pathway are ordinary in
sourdough processing [3]. The last thirty years of research on
sourdough fermentation depicted that acetic and lactic acids have
concentrations between 1 and 50 mM (with a median value of
20 mM) and between 15 and 150 mM (with a median value of
75 mM), respectively.
The QF has also a significant effect on the sensory character-
istics, flavor, and shelf life of leavened baked products such as bread.
The QF depends on the structure and activity of microbial com-
munities in sourdough. For instance, the ratio between homo- and
heterofermentative LAB is considered to likely affect the QF,
resulting in marked variations of its final values. Yeast activities
can also impact the ratio. In particular, yeasts play a role on fructose
liberation in sourdough through invertase activity, which, as a
consequence, results in acetic acid production by LAB, with final
decrease of QF [4]. Although the literature of last decades high-
lighted that QF values may range between 0.25 and 20, the average
value was estimated to be 4.4, while the most recommended value
is below 5.0 [3].
Overall, the determination of lactic and acetic acids in food
matrices can be achieved through different approaches: mainly
spectrophotometric, enzymatic, and chromatographic methods.
Spectrophotometric methods are based on the reaction of the
organic acid with a certain substance, resulting in a compound or
colored complex that can be detected and measured at a certain
wavelength. However, for these analyses, organic acids need to be
isolated (for instance by precipitation, ionic exchange resins, etc.)
to avoid problems due to interferences. Enzymatic methods evalu-
ate the increase/decrease in absorbance of the coenzymes
nicotinamide–adenine dinucleotide, reduced form (NADH), or
nicotinamide–adenine dinucleotide phosphate, reduced form
(NADPH). The measurement is usually made at 340 nm through
a spectrophotometer. The main advantage of these methods is the
high specificity. On the other hand, generally only one organic acid
is determined in each assay [5]. Nowadays, chromatographic meth-
ods are the most frequently used for the determination of acetic and
lactic acids in sourdough. Separation and quantification of organic
acids are achieved through high performance liquid chromatogra-
phy (HPLC) or ion chromatography (IC). Here the HPLC method
is detailed as it is the most widely adopted in sourdough [2, 6]. Two
main steps can be distinguished in this method: the preparation and
filtration of water/salt-soluble extracts (WSE) and the HPLC
 Lactic and Acetic Acids and Quotient of Fermentation 63

analysis. The preparation of standards and the construction of

calibration curves are also presented. Finally, the calculation of
QF is described.

2 Materials

2.1 Preparation and The following solutions and materials are needed to prepare WSE
Filtration of WSE from sourdough samples: falcon tubes, spoons, 10 mL pipettes,
50 mM Tris–HCl (pH 8.8), a pH meter, a stirrer (see Note 1), a
centrifuge, 10 mL syringes, 0.22 μm filters, HPLC vials, and a
freezer (see Note 2). At least 5 g of each studied sourdough is
needed. Moreover, a chemical hood and personal protective equip-
ment (like goggles, gloves, and lab coat) are necessary.

2.2 HPLC Analysis Lactic and acetic acids contained in WSE are determined by HPLC.
Instruments by several brands are available and suitable. We use a
Dionex Ultimate 3000 (Thermo Scientific) equipped with an Ami-
nex HPX-87H column 300 × 7.8 mm (ion exclusion, Biorad) (see
Note 3), a refractometer, and a UV detector operating at 210 and
190 nm. Other necessary materials are lactic and acetic acid stan-
dards for HPLC (purity 99.9%), distilled water, 2.0 or 1.5 ml tubes,
pipettes, tips, bottles, and 5 mM H2SO4 solution as mobile phase.

3 Methods

3.1 Preparation and Under the chemical hood, 5 g of sourdough is collected with a
Filtration of WSE spoon, transferred into a falcon tube, and mixed to 10 mL of
50 mM Tris–HCl (pH 8.8). The mixture is incubated at 4 °C for
1 h under stirring conditions (150 rpm) and then centrifuged at
12,000 × g for 20 min (see Note 4). The supernatant is collected
with a 10 mL syringe, filtered by using a 0.22 μm filter, and
transferred to a HPLC vial (see Note 5). The WSE can now be
immediately analyzed by HPLC or stored at -20 °C till the analy-
sis. All steps are performed by using the personal protective
equipment.

3.2 Preparation of To detect lactic and acetic acids (target compounds of this analysis),
Standards, the RT of both compounds have to be defined (see Note 6). For this
Determination of purpose, a solution of the pure compound has to be prepared for
Retention Times (RT), both organic acids by diluting each standard in distilled water (see
and Construction of Notes 7 and 8). Solutions are then transferred to HPLC vials (see
Calibration Curves Note 5) and subjected to HPLC run that is performed as described
in the next section (see Note 9).
A calibration curve for both organic acids must also be built to
quantify them. A calibration curve is a graphical representation of
the amount and response data (i.e., area of peak) for a single analyte
64 Andrea Polo and Marco Gobbetti

obtained from one or more calibration samples. The curve is usually

constructed by injecting an aliquot of the calibration standard
solution of known concentration and measuring the peak area
obtained. Therefore, the curve is a plot of how the instrumental
response (i.e., area of peak) changes with the concentration of the
analyte. A range of analyte concentrations exists in which the curve
can be described by a linear equation (y = ax + b). The coefficient of
determination (R2) of the resulting linear equation is the propor-
tion of the variation in the dependent variable that is predictable
from the independent variable. In other words, it provides a mea-
sure of how well observed outcomes are replicated by the calibra-
tion curve, based on the proportion of total variation of outcomes
explained by the linear equation. To be a good calibration curve, R2
should be close to 1. The output from an unknown sample (peak
area measured using a data system) is then related, through the
calibration curve, to those of a calibration sample to calculate the
amount in the unknown. For this purpose, several solutions of
lactic and acetic acids with final concentration ranging from 0.1
and 15 g/L are prepared by diluting standards in distilled water (see
Notes and 11). Solutions are then transferred to HPLC vials (see
Note 5) and subjected to HPLC analysis that is performed as
described in the next section (see Note 12).

3.3 HPLC Analysis The structure of HPLC apparatus is shown in Fig. 1. It consists of
different modules: a pump, an autosampler (containing the
sequence of vials to be analyzed), an oven containing the chroma-
tography column, UV detectors, and refractometer. The bottle of
mobile phase is first filled with 5 mM solution of H2SO4 and a
purge of instrument is done before the analysis by flowing the
mobile phase through the system. The Aminex HPX-87H column
(300 × 7.8 mm) can now be connected (if it is not yet). The flow
rate is then increased gradually to the maximum flow rate in order
to exclude the presence of leakages (see Note 13). If no leakages are
present, a new purge of instrument is done. The flow rate is
increased (for instance to 0.2 or 0.3 mL/min, according to the
value of maximum flow rate reported in the column manufacturer
instruction) and then the elution temperature is also increased in
the oven to 70 °C (see Notes 14 and 15). The detector can now
turn on by flagging the UV lamp. Samples are analyzed both with
UV detector operating at 210 and 190 nm, and with refractometer
(IR) (see Note 16). The analytical parameters must now be set. We
use 0.6 mL/min as flow rate, 20 μL of injective volume, 70 °C oven
temperature (see Note 17), 30 min acquisition time, and isocratic
run with only (i.e., 100%) 5 mM solution of H2SO4 as mobile
phase. The sequence of analyses should now be prepared, and
vials containing WSE, calibration standard solutions, and at least
one pure solution containing only one organic acid standard (for
 Lactic and Acetic Acids and Quotient of Fermentation 65

Fig. 1 The structure of HPLC apparatus. It includes the bottle with the mobile
phase, a pump, an autosampler system (containing the sequence of vials to be
analyzed), an oven containing the chromatography column, and the UV and
refractometer detectors

RT determination) are placed accordingly in the autosampler (see

Note 18). A blank sample (consisting of mobile phase only) should
also be considered.
Resulting spectra are analyzed for lactic and acetic acids detec-
tion and quantification. Either the spectra at 210 or 190 nm or IR
can be chosen for this purpose (see Note 19). Peaks resulting in the
chromatogram of unknown samples are assigned based on the RT
of the standards. After the peaks have been identified and
integrated, quantification is performed. For each analyzed sample,
the calibration curves and the peak areas of the target compounds
in the WSE with unknown concentration are used to calculate the
concentrations of the target compounds (see Note 20). With this
output and knowing how the WSE was prepared (i.e., 5 g of
sourdough in 10 mL of 50 mM Tris–HCl at pH 8.8), the concen-
tration of lactic and acetic acids in the original sourdough sample
can be calculated. In this case, the concentration of each target
compound in original sourdough corresponds to X/500, where X
is the concentration of the target compound in WSE.
Finally, the overall QF of sourdough can be calculated as molar
ratio between lactic and acetic acids concentrations. For instance, if
66 Andrea Polo and Marco Gobbetti

Fig. 2 Flowchart for the determination of lactic and acetic acids and the quotient
of fermentation in sourdough by HPLC method

concentrations of 93.3 ± 0.3 and 20.3 ± 0.2 mmol/kg were

determined for lactic and acetic acid, respectively, the
corresponding QF is 4.6 ± 0.2 [7].
Figure 2 shows the flowchart of the overall method.

4 Notes

1. The stirring should be at 150 rpm and 4 °C. If the stirrer does
not allow a temperature control, it can be placed in fridge at 4 °
C.
2. A freezer is needed only if samples cannot be analyzed immedi-
ately after the preparation. In this case, they have to be stored at
-20 °C till the analysis.
 Lactic and Acetic Acids and Quotient of Fermentation 67

3. The column is the main component in HPLC because it is

responsible for the separation of the sample components. The
sample passes through the column with the mobile phase and
separates in its components when it comes out from the
column.
4. As an alternative, in case a stirrer is not available, the mixture
can be vortexed for 1 min every 5 min of incubation at 4 °C.
5. The quantity inside the vial should be enough to allow the
needle to be immersed in the liquid.
6. The RT is a measure of the time taken for a solute to pass
through a chromatography column. It is calculated as the time
from injection to detection and, if other analytical parameters
are fixed, it is characteristic of a substance. However, the RT of
a substance can vary depending on the kind of mobile phase,
flow rate, features of chromatography column, temperature of
column. Therefore, an RT can be considered valid for the
qualitative detection of a target substance only if the analysis
is performed with the same parameters that were adopted to
determine the RT of such target substance. If analytical para-
meters are identical, an RT can be adopted to detect the same
substance in different analyses. If analytical parameters change,
also the new RT must be defined for each target substance.
7. It is important that each solution contains only one target
substance. This is necessary to define the RT as it is characteris-
tic of each target substance.
8. It is important that such solutions contain a final concentration
of target compounds that is within the range of detection for
the instrument. In general, for such kind of instruments and
compounds, it is between 0.1 and 15 g/L.
9. Samples can be immediately analyzed or stored at -20 °C till
the analysis. Moreover, they can be loaded in the same
sequence with WSE samples and calibration standard solutions,
or in a separate sequence.
10. The range of concentrations depends on the detection limits of
the instrument. To prepare a good calibration curve, calibra-
tion standard solutions at 0.1, 0.5, 1, 5, 10, and 15 g/L could
be used for both lactic and acetic acids.
11. Calibration standard solutions can also be prepared by mixing
lactic and acetic acids in the same solution. In this way, a lower
amount of mobile phase is consumed for the analysis. In fact, in
the resulting spectra, lactic and acetic acids can be distin-
guished based on the different RT. For this purpose, it is
necessary that the RT characteristic for each organic acid is
determined by injecting a solution containing only one target
substance, for both target compounds. Such solutions can run
before or together with calibration standard solutions and
unknown samples.
68 Andrea Polo and Marco Gobbetti

12. Samples can be immediately analyzed or stored at -20 °C till

the analysis. Moreover, they can be loaded in the same
sequence with WSE samples and pure standard solutions, or
in a separate sequence.
13. The maximum flow rate is reported in the column manufac-
turer instruction.
14. To avoid damages of column, the temperature should be
increased when the flow is running. Therefore, the flow must
be first increased and then the temperature can also be
increased.
15. Small adjustment of elution temperature could be necessary
depending on other analytical parameters adopted for the anal-
ysis and/or if the equipment used for the analysis has different
characteristics (e.g., different column, different instrument,
etc.). In this case, the optimal temperature for the method
must be set up and tested.
16. Three spectra are collected for each sample (operating at
210 and 190 nm, and with IR). All of them can be used to
detect and quantify lactic and acetic acids in samples by using
the corresponding calibration curve (i.e., the curve built by
using spectra of the standard solutions obtained operating at
the same wavelength).
17. The values of these parameters are suggested by the column
manufacturer instruction based on the characteristics of the
chromatography column and validated/adjusted during the
method development. Different methods should be set up for
different target compounds or in case a different column
is used.
18. The sequence of vials, as it is in the autosampler, must be
reported in the files of analysis in order to assign each spectrum
to the corresponding sample.
19. The outputs resulting from the wavelength corresponding to
the best quality can be selected. In order to compare the data
from different samples/analyses, the spectra obtained at the
same wavelength should be used. Moreover, in order to have a
valid quantification, the reference calibration curves must also
be built with the spectra obtained operating at the same
wavelength.
20. To obtain a valid estimation by comparing the unknown sam-
ple response to that of the known standard, the data must be
acquired and processed under identical conditions (i.e., same
method and identical analytical parameters).
 Lactic and Acetic Acids and Quotient of Fermentation 69

References
1. Gobbetti M (1998) The sourdough microfora: 5. Boehringer Mannheim GmbH (1995) Methods
interactions of lactic acid bacteria and yeasts. of enzymatic bioanalysis and food analysis using
Trends Food Sci Technol 9:267–274 test-combinations. Boehringer Mannheim
2. De Luca L, Aiello A, Pizzolongo F, Blaiotta G, GmbH, Biochemicals, Mannheim. https://
Aponte M, Romano R (2021) Volatile organic [Link]/books/about/Methods_of_
compounds in breads prepared with different Enzymatic_Bioanalysis_and_Foo.html?id=
sourdoughs. Appl Sci:11 gmcLSQAACAAJ&pgis=1
3. Arora K, Ameur H, Polo A, Di Cagno R, Riz- 6. Rizzello CG, Nionelli L, Coda R, De Angelis M,
zello CG, Gobbetti M (2021) Thirty years of Gobbetti M (2010) Effect of sourdough fer-
knowledge on sourdough fermentation: a sys- mentation on stabilisation, and chemical and
tematic review. Trends Food Sci Technol 108 nutritional characteristics of wheat germ. Food
(November):71–83. [Link] Chem 119(3):1079–1089. [Link]
1016/[Link].2020.12.008 1016/[Link].2009.08.016
4. Corsetti A (2013) Technology of sourdough 7. Rizzello CG, Portincasa P, Montemurro M, di
fermentation and sourdough applications. In: Palo DM, Lorusso MP, de Angelis M et al
Gobbetti M, G€anzle M (eds) Handbook on (2019) Sourdough fermented breads are more
sourdough biotechnology. Springer, Boston, pp digestible than those started with baker’s yeast
85–103 alone: an in vivo challenge dissecting distinct
gastrointestinal responses. Nutrients 11(12)
 Chapter 7

Determination of the Content of Free Amino Acids and Their

Profiling
Michela Verni

Abstract
When assessing sourdough maturity, the quantification of free amino acids (FAA) is of utmost importance,
since their concentration is one of the biochemical indicators commonly used to describe its performances.
The difficulty of analyzing amino acids with conventional chromatographic systems is mainly due to the
complex nature of samples and unique chemical properties of amino acids, which require chemical
derivatization, either before or after chromatographic separation, to improve their detection. For good
chromatographic separations, the right combination of resins, buffer, pH, and temperature is necessary,
which is why automated systems like the Biochrom amino acid analyzer have been developed. In the
Biochrom system, amino acids are separated by ion exchange chromatography and quantified using dual
wavelength photometric detection following ninhydrin post-column derivatization. Notwithstanding the
high reproducibility and the possibility to quantify up to 56 amino acids and derivatives, the long time of
analysis can be inconvenient if large numbers of samples are to be analyzed or if the goal is to just quantify
total FAA. In such cases, rapid spectrophotometric assays like the ninhydrin method are available.
This chapter describes the determination of FAA in sourdough according to two methods: (i) the
spectrophotometric quantification of total free amino acids with the cadmium-ninhydrin assay, which is
rapid and does not require elaborated and expensive equipment, and (ii) the chromatographic technique
using the Biochrom system, which, although more time consuming than the former, allows the quantifica-
tion of the single free amino acids.

Key words Proteolysis, Free amino acids, Cadmium-ninhydrin assay, Cation exchange chromatogra-
phy, Amino acid analyzer, Post-column derivatization

1 Introduction

Proteolysis, one of the key phenomena occurring during sour-

dough fermentation, is the result of a combination of both cereal
and microbial proteases. Primary proteolysis (proteins to peptides)
in sourdoughs is mainly attributable to endogenous cereal pro-
teases, activated by the acidification; whereas secondary proteolysis,
which releases free amino acids and small peptides, is exclusively
operated through peptidase activity of lactic acid bacteria

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

71
72 Michela Verni

[1]. Proteolysis affects the overall quality of both sourdough and

sourdough-containing foods and is of crucial importance for baked
goods sensory, structural, and nutritional features [2]. Amino acids
and peptides affect the taste of fermented foods and are precursors
of volatile compounds. For example, the roasty flavor of bread crust
is caused by acetyl pyrroline, which originates during baking from
ornithine, a nonproteinogenic amino acid deriving from arginine
by microbial metabolism during fermentation [3, 4]. Depending
on the amount of sourdough in the dough recipe and on the extent
of proteolysis, the degradation of gluten structure can improve the
viscoelastic properties of the dough, whereas the proteolytic system
of sourdough lactic acid bacteria can also contribute to the accu-
mulation of bioactive peptides having antioxidant, anti-
hypertensive, and antifungal activities [1, 5].
When assessing sourdough maturity, the quantification of free
amino acids (FAA) is of utmost importance, since their concentra-
tion is one of the biochemical indicators commonly used to
describe its performances. It is, however, difficult to assign prees-
tablished ranges of both single amino acids and total concentrations
since they are markedly influenced by the type of flour, parameters
of fermentations, and strains used. Nevertheless, a recent systematic
review on sourdough indicated that FAA concentration after fer-
mentation can vary from 390 to 5000 mg/kg [6].
While the determination of the total amino acid content of
foods requires protein hydrolysis, free amino acids can be simply
quantified after deproteinization of the protein extracts.
The first step is the extraction of the soluble protein fraction
which is often performed using a method originally described by
Osborne [7] for wheat flour proteins. The original procedure
consisted of a sequential extraction of wheat protein fractions:
(i) salt-soluble proteins (albumins and globulins), (ii) aqueous
alcohol-extractable proteins (gliadins), and (iii) SDS/DTT-
extractable proteins (glutenins). Yet the first extraction was later
modified [8] and applied to several matrices of both cereal and
non-cereal origin [9–12] for the extraction of amino acids due to
their high solubility in water solutions. Extractions directly with
water alone or in combination with 7% perchloric acid, which
deproteinizes the sample, as well as 96% ethanol, have been pro-
posed [3, 4], yet many more studies report extraction with Tris–
HCl, the water/salt-soluble solution proposed by Osborne [7],
and deproteinization with sulphosalicylic acid [9–14].
Modern methods for separation and quantitation of amino
acids either before or after protein hydrolysis include high perfor-
mance liquid chromatography (HPLC), gas chromatography, and
capillary electrophoresis. Ion exchange chromatography is cur-
rently the most widely used analytical technique; however, the
difficulty of analyzing amino acids with conventional HPLC sys-
tems is mainly due to the complex nature of the samples and the
 Free Amino Acid Determination 73

unique chemical properties of the amino acids, ionic compounds

with no specific absorption wavelength [15]. Hence, their chemical
derivatization, either before (pre-column) or after (post-column)
chromatographic separation, is required to create derivatives with
properties, such as fluorescence, that improve their detection. Post-
column derivatization after cation-exchange chromatography is
more repeatable than pre-column derivatization followed by
reversed-phase chromatography separation because any interfering
substance in the sample is separated prior to derivatization, which
occurs on the pure amino acids [15]. Ninhydrin, o-phthalaldehyde
(OPA), and fluorescamine are some of the post-column derivatiz-
ing agents more often used in ion exchange chromatography with
detection ranges of the order of pmol [15, 16]. The most common
method is to form a dye complex with reduced ninhydrin, which is
then measured colorimetrically. Although OPA is 10 times more
sensitive than ninhydrin for primary ammine, it has a low response
to cystine and none to proline and hydroxyproline. Fluorescamine,
similarly to OPA, only reacts with imino acids after they have been
oxidized and it is also expensive [16].
Anion-exchange chromatography in combination with
integrated pulsed amperometric detection, a method often used
for the detection of carbohydrates and amines, was also proposed
for the determination of FAA in sourdough [4]. It is a very sensitive
method that does not require derivatization, but if the concentra-
tion of sugars is much higher than that of FAA, interferences of
glucose, fructose, and sucrose can occur, since they coelute with
asparagine, alanine, threonine, and serine [4].
Overall, for good separations in ion exchange chromatography,
the right combination of resins, buffer, pH, and temperature is
necessary; even the slightest change can negatively affect the sepa-
ration. For this reason, automated systems like the Biochrom
amino acid analyzer have been developed. The possibility to pur-
chase the appropriate chemical kit (buffers and reagent), subjected
to rigorous control procedures under the ISO 13485 Quality
System, is one of the advantages of the automated system, as it
guarantees accurate and reproducible results, avoiding any errors.
In the Biochrom system, which is considered the gold standard
for amino acid analysis, amino acids are separated by ion exchange
chromatography and quantified using dual wavelength photomet-
ric detection following ninhydrin post-column derivatization, as
required by the AOAC method and the Commission Directive
98/64/EC [17]. The Biochrom amino acid analyzer was devel-
oped based on the system first proposed by Spackman et al. [18]
and then improved to obtain a completely automated and sensitive
system. Two precision pumps deliver constant volumes against
modest back pressures. One pump pushes buffered eluents through
a cation exchange column and the second pump adds ninhydrin to
the flow of separated amino acids exiting the column. The positively
charged amino acids are bound to the resin which is negatively
74 Michela Verni

Table 1
Composition of elution buffers for the separation of amino acids as reported in the Decree n. 6890 of
April 28, 2014, of the Italian Ministry of Agricultural, Food and Forestry Policies

Buffer 1 Buffer 2 Buffer 3 Buffer 4 Buffer 5 Buffer 6 Loading buffer

pH 2.84 3.00 3.15 3.50 3.58 – 2.20
+
Li (mol/L) 0.2 0.3 0.5 0.9 1.62 0.3 0.2
Lithium hydroxide 8.4 8.4 8.4 4.2 7.0 12.59 8.4
monohydrate (g)
Citric acid (g) 9.6 9.6 9.6 9.6 21.0 – 9.6
Lithium cloride (g) – 4.25 12.72 34.0 61.5 – –
Pentachlorophenol (mL) 0.1 0.1 0.1 0.1 0.1 0.1 0.1
Thiodiglycol 25% (mL) 8 8 8 – – – 8
2-propanol (mL) 15 15 – – – – –
Hydrochloric acid (mL) ~ 14 ~ 13 ~ 12 ~3 – – ~ 14

charged. The conditions are then altered to increase pH, tempera-

ture, and concentration of the buffer counter ion. An example of
elution buffers composition used for the separation of amino acids
is shown in Table 1. When the isoionic point of an amino acid is
being reached, the ionic attraction to the resin is lost and the amino
acid elutes from the column [15, 17]. Once separated, amino acids
reach the reaction coil where ninhydrin oxidizes them, releasing the
amine group as ammonium. Since high reaction temperatures are
required to accelerate the reaction with ninhydrin, the reaction
happens at 135 °C [15]. Hydrindantin, the reduced form of ninhy-
drin, reacts with the free amines as well as any urea present to
produce a purple derivative that absorbs at 570 nm (Ruhemann’s
purple). Secondary amines, such as proline and hydroxyproline
which do not have free α-amino groups, react similarly but produce
a yellow chromophore that is measured at 440 nm [15, 17]. After
each analysis, the column is regenerated with a strong base (buffer
6) and then reequilibrated for the following run with buffer 1.
Notwithstanding the high reproducibility of retention time and
peak area of the automated system, as well as the ability to allow
both qualitative and quantitative analysis of up to 56 amino acids
and derivatives [17], the long time of analysis (up to 3 h per sample)
can be inconvenient if large numbers of samples are to be analyzed
or if the goal is to just quantify total FAA.
In such cases, rapid spectrophotometric assays like the ninhy-
drin method are available. Although originally developed for amino
acids derivatization after chromatographic elution, the ninhydrin
method [19], introduced in the late 1940s, has been later adapted
for the determination of amino group–containing compounds in
various foods [20]. Considerable variations in the experimental
 Free Amino Acid Determination 75

conditions have been performed. Heating times (from 5 to 45 min)

and temperatures (from 85 to 100 °C), buffer systems (phosphate,
sodium acetate, lithium acetate, potassium acetate, sodium citrate,
etc.), pH values [5–8] of buffer solutions and solvents (dimethyl
sulfoxide, absolute alcohol, acetone, ethylene glycol, water) are
among the main variables considered [20]. Toxic cadmium salt
was also used in some studies [21, 22], and although not always
preferred by some researchers, the cadmium-ninhydrin assay is
among the most used for the quantification of total free amino
acids in sourdough or, indirectly, the proteolytic activity of sour-
dough lactic acid bacteria [23–25].
In this chapter, the determination of FAA in sourdough will be
described according to two methods: i) the spectrophotometric
quantification of total free amino acids with the cadmium-
ninhydrin assay, which is rapid and does not require elaborated
and expensive equipment, and ii) the chromatographic technique
using the Biochrom system, which although more time consuming
than the former, allows the quantification of the single free amino
acids.

2 Materials

Prepare solutions using ultrapure water (obtained by purifying

deionized water with Milli-Q® system) and analytical grade
reagents. Diligently follow all waste disposal regulations when
disposing waste materials.

2.1 Extraction and • Extraction buffer: 0.05 M Tris–HCl, pH 8.8. Place about
Pre-treatment 800 mL water to a 1 L glass beaker. Add 6.057 g Trizma
Base® while stirring. When the powder is completely dissolved,
adjust the pH with HCl (see Note 1). Transfer to a cylinder,
bring to volume (1 L) with water, and mix. Store at 4 °C.
• Laboratory refrigerated orbital shaker incubator or refrigerator
equipped with a socket for an orbital shaker.
• Conical tubes (1.5–2 and 50 mL) and respective centrifuges.
• Analytical balance.
• 5-Sulfosalicylic acid (MW:254.21 g/mol).
• Micro-pipettors, e.g., Gilson Pipetman® (1000 μL).

2.2 • In a conical tube, place the sample/standard and the cadmium-

Spectrophotometric ninhydrin reagent in a ratio of 1:2.
Assay for Total FAA • Place the tubes in the floating racks and hold at 84 °C in a water
Quantification bath for 5 min.
76 Michela Verni

Table 2
Program optimized to separate unusual amino acids using the Biochrom
system

Ninhydrin
No. Time Temperature (°C) Buffer Buffer pump (flow rate 20 mL/h)
1 01:00 70 1 25.0 mL/h ON
2 00:00 70 1 25.0 mL/h ON
3 01:00 70 1 25.0 mL/h ON
4 02:30 70 1 25.0 mL/h ON
5 21:30 28 2 25.0 mL/h ON
6 22:00 38 3 25.0 mL/h ON
7 03:00 67 3 25.0 mL/h ON
8 27:00 67 4 25.0 mL/h ON
9 40:00 85 5 25.0 mL/h ON
10 06:00 85 6 25.0 mL/h ON
11 06:00 85 1 25.0 mL/h ON
12 02:00 50 0 OFF OFF
13 30:00 50 1 32.1 mL/h OFF
14 06:00 70 1 25.0 mL/h OFF

• After cooling in ice, place the tubes content in the cuvettes and
read the absorbance at 507 nm against a blank (water and
reagent) (see Note 7).
• Elaborate the data using the calibration curve equation and
apply potential dilution factors to quantify total free amino
acids in the samples.

2.3 Chromatographic • For the chromatographic analysis, the supernatant containing

Separation and the amino acids must be filtered through the syringe filter
Quantification of FAA (0.22 μm) and placed into the glass vials (see Note 8).
• Prepare dilutions of the standard so that each amino acid has a
concentration of 500 μmol/L except cystine, which will be
250 μmol/L (see Note 9).
• Prepare the ninhydrin reagent by purging with nitrogen for
10 min the Ultrasolve solution, add the UltraNinhydrin solu-
tion, and purge for 10 more minutes while stirring.
• Set the instrument following manufacturer instructions and
launch the run. An example of the elution gradient for the
chromatographic run is shown in Table 2, and the respective
amino acid standard chromatogram is shown in Fig. 1.
 Free Amino Acid Determination 77

Fig. 1 Example of chromatographic separation of an amino acid standard using the Biochrom amino acid
analyzer set up on the program shown in Table 2

• When each run is completed, extrapolate FAA concentration

from the EZChrom software and apply all correction and/or
dilution factors to quantify single and total amino acids.

3 Notes

1. Concentrated HCl (5–10 N) can be used at first to narrow the

gap from the starting pH to the required pH. From then on, it
would be better to use a series of HCl (e.g., 0.1–1 N) with
lower ionic strengths to avoid a sudden drop in pH below the
required pH.
2. Due to its high toxicity, wear protective gloves when using
ninhydrin. If all the glassware used for the ninhydrin reagents
are not already shielded, cover them with tin foil to prevent
oxidation.
3. When purchasing the standard, make sure it contains all the
amino acids of interest, if not, a solution providing all the
missing amino acids should be prepared separately. A calibra-
tion standard containing most amino acids can be purchased by
several suppliers, as a solution in 0.1 N HCl matrix. Since
dilutions are necessary, use the loading buffer provided in
the kit.
78 Michela Verni

4. It is ideal to keep the nitrogen flow continuously on, to avoid

oxidation of the reagents, even if the instrument is shutdown.
By doing so, nitrogen levels can be monitored and potential
leaks detected. An amino acid analyzer well kept can be consid-
ered a close system and leaks are nearly absent. Note that if the
nitrogen pressure is above the limits recommended by the
manufacturer, security valves will throw out the excess.
5. The columns available for the Biochrom system can exchange
either lithium or sodium ions which means buffers must be
chosen accordingly. The difference among the two types of
columns is mostly the number of amino acids that they are
able to separate (more than 50 and 24 for lithium and sodium,
respectively). The lithium system is often recommended for
physiological fluids, whereas the sodium one for food matrices;
however, this is not a strict alternative. Indeed, the official
method for the detection of amino acids in cheese involves
the use of a lithium system.
6. Depending on the sample, it might be necessary to homoge-
nize the sourdough and the buffer in advance. While sour-
doughs having high dough yield tend to dissolve easily in the
extraction buffer, for sourdough with low dough yield, it might
be necessary to use a stomacher blender and then transfer the
content in a conical tube. In the latter condition, it is suitable to
weigh at least 10 g of sample.
7. If the absorbance values do not fit within the calibration curve,
repeat the reaction using an extract diluted in water (depending
on the type of sourdough, 1:10 or 1:100 dilution ratios might
be necessary).
8. If required, store the sample at 4 °C for short term (no more
than 48 h). Storage at -20 °C is usually suitable up to 2 weeks.
However, sensitive amino acids such as glutamine or arginino-
succinic acid continue to degrade slowly. For longer storage, -
80 °C is recommended.
9. As for the samples, the calibration standard solution should be
treated with sulfosalicylic acid to ensure that both are loaded for
the analysis at the same pH and give identical retention times for
corresponding peaks, especially for aspartic acid. Otherwise, if
the loading pH of the sample is lower than the standard, the
retention times of the initial amino acids may be increased.

References

1. G€anzle M, Gobbetti M (2013) Physiology and 2. G€anzle MG, Loponen J, Gobbetti M (2008)
biochemistry of lactic acid bacteria. In: Proteolysis in sourdough fermentations:
Gobbetti M, G€anzle M (eds) Handbook on mechanisms and potential for improved bread
sourdough biotechnology. Springer, Boston, quality. Trends Food Sci Technol 19:513–521
MA, pp 183–216
 Free Amino Acid Determination 79

3. Thiele C, G€anzle MG, Vogel RF (2002) Con- chemical and nutritional characteristics of
tribution of sourdough lactobacilli, yeast, and wheat germ. Food Chem 119:1079–1089
cereal enzymes to the generation of amino 14. De Pasquale I, Verni M, Verardo V, Gómez-
acids in dough relevant for bread flavor. Cereal Caravaca AM, Rizzello CG (2021) Nutritional
Chem 79:45–51 and functional advantages of the use of fermen-
4. Thiele C, G€anzle MG, Vogel RF (2002) Sam- ted black chickpea flour for semolina-pasta for-
ple preparation for amino acid determination tification. Foods 10:182
by integrated pulsed amperometric detection 15. Peace RW, Gilani GS (2005) Chromatographic
in foods. Anal Biochem 310:171–178 determination of amino acids in foods. J AOAC
5. Rizzello CG, Tagliazucchi D, Babini E, Rutella Int 88:877–887
GS, Saa DLT, Gianotti A (2016) Bioactive 16. Walker V, Mills GA (1995) Quantitative meth-
peptides from vegetable food matrices: ods for amino acid analysis in biological fluids.
research trends and novel biotechnologies for Ann Clin Biochem 32:28–57
synthesis and recovery. J Funct Foods 27:549– 17. Lolia A (2007) Accelerated buffer system for
569 amino acid analysis. Biochrom Ltd, Cam-
6. Arora K, Ameur H, Polo A, Di Cagno R, Riz- bridge, East Anglia
zello CG, Gobbetti M (2021) Thirty years of 18. Spackman DH, Stein WH, Moore S (1958)
knowledge on sourdough fermentation: a sys- Automatic recording apparatus for use in chro-
tematic review. Trends Food Sci Technol 108: matography of amino acids. Anal Chem 30:
71–83 1190–1206
7. Osborne T (1907) The proteins of the wheat 19. Moore S, Stein WH (1948) Photometric nin-
kernel. Carnegie Inst. Washington Publ. 84. hydrin method for use in the chromatography
Judd and Detweiler, Washington of amino acids. J Biol Chem 176:367–388
8. Weiss W, Vogelmeier C, Görg A (1993) Elec- 20. Sun SW, Lin YC, Weng YM, Chen MJ (2006)
trophoretic characterization of wheat grain Efficiency improvements on ninhydrin method
allergens from different cultivars involved in for amino acid quantification. J Food Compost
bakers’ asthma. Electrophoresis 14:805–816 Anal 19:112–117
9. Coda R, Rizzello CG, Gobbetti M (2010) Use 21. Doi E, Shibata D, Matoba T (1981) Modified
of sourdough fermentation and pseudo-cereals colorimetric ninhydrin methods for peptidase
and leguminous flours for the making of a assay. Anal Biochem 118:173–184
functional bread enriched of γ-aminobutyric
acid (GABA). Int J Food Microbiol 137:236– 22. Fisher GH, Arias I, Quesada I, D’aniello S,
245 Errico F, Di Fiore MM, D’aniello A (2001) A
fast and sensitive method for measuring pico-
10. Curiel JA, Coda R, Centomani I, Summo C, mole levels of total free amino acids in very
Gobbetti M, Rizzello CG (2015) Exploitation small amounts of biological tissues. Amino
of the nutritional and functional characteristics Acids 20:163–173
of traditional Italian legumes: the potential of
sourdough fermentation. Int J Food Microbiol 23. Rashmi BS, Gayathri D, Vasudha M, Prashant-
196:51–61 kumar CS, Swamy CT, Sunil KS, Somaraja PK,
Prakash P (2020) Gluten hydrolyzing activity
11. Verni M, De Mastro G, De Cillis F, of bacillus spp isolated from sourdough.
Gobbetti M, Rizzello CG (2019) Lactic acid Microb Cell Factories 19:1–11
bacteria fermentation to exploit the nutritional
potential of Mediterranean faba bean local bio- 24. Dallagnol AM, Pescuma M, De Valdez GF,
types. Food Res Int 125:108571 Rollán G (2013) Fermentation of quinoa and
wheat slurries by Lactobacillus plantarum CRL
12. Palla M, Agnolucci M, Calzone A, 778: proteolytic activity. Applied Microbiol
Giovannetti M, Di Cagno R, Gobbetti M, Riz- Biotechnol 97:3129–3140
zello CG, Pontonio E (2019) Exploitation of
autochthonous Tuscan sourdough yeasts as 25. Wehrle K, Crowe N, van Boeijen I, Arendt EK
potential starters. Int J Food Microbiol 302: (1999) Screening methods for the proteolytic
59–68 breakdown of gluten by lactic acid bacteria and
enzyme preparations. Eur Food Res Technol
13. Rizzello CG, Nionelli L, Coda R, De 209:428–433
Angelis M, Gobbetti M (2010) Effect of sour-
dough fermentation on stabilisation, and
 Chapter 8

Soluble Sugars and Polysaccharides

Michela Verni and Marco Montemurro

Abstract
During sourdough fermentation, important changes in the carbohydrate fraction occur due to enzymatic
and metabolic reactions of both yeasts and lactic acid bacteria involved in the process. Indeed, the
production of several metabolites depends on the availability of soluble carbohydrates either initially present
in the flour or resulting from the hydrolysis of starch or other polysaccharides.
For the determination of soluble sugars and polysaccharides, enzymatic kits, often based on official
methods of analysis, recognized by the scientific community are available. The high specificity of enzymes
enables the analysis of complex matrixes without complicated sample preparation techniques and little
interferences which, on the contrary, might occur during chromatographic coelution. Nevertheless,
although enzymatic assays favor the analysis of several samples in relatively little amount of time, the higher
costs of analysis per sample compared to lab-developed methods tip the scale in favor of the latter.
Whereas soluble sugars can be easily extracted and separated through high performance liquid chroma-
tography, starch quantification needs to be preempted by its hydrolysis to glucose. As for dietary fibers, the
hydrolysis of proteins and starch precedes filtration of insoluble fiber and precipitation of soluble fiber in
ethanol. Their content is then evaluated by weighing after drying. In this chapter, the chromatographic
method for mono-, di-, and oligosaccharides, the enzymatic hydrolysis of starch, and the analysis of dietary
fibers according to official methods will be described.

Key words Carbohydrates, Mono-, di-, oligosaccharides, Starch, Dietary fiber, Enzymatic hydrolysis,
HPLC, Official method

1 Introduction

Even though the analysis of sugars is not a direct mean to assess

sourdough maturity and its composition highly varies depending
on the flours used (e.g., α-galactosides are typically found in
legumes-based sourdoughs but not in those made of wheat), dur-
ing fermentation, important changes in the carbohydrate fraction
occur due to enzymatic and metabolic reactions of both yeasts and
lactic acid bacteria involved in the process [1, 2]. Indeed, the
production of several metabolites depends on the availability of
soluble carbohydrates either initially present in the flour (ferment-
able carbohydrates) or resulting from the hydrolysis of starch or

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

81
82 Michela Verni and Marco Montemurro

other polysaccharides [2]. The soluble carbohydrates remaining

after microbial fermentation then participate in browning reactions
during baking, contributing to the organoleptic characteristics of
bread. In addition, sugars enhance the freshness and texture of
breads, while low-molecular-weight dextrins play a role in interfer-
ing with bread staling [3], which is why the use of amylases is
considered the most successful enzymatic approach in reducing
the rate of bread staling [4].
Among polysaccharides, starch is the most abundant and
important in all cereal-based products, representing the main
source of energy in the human diet. It is present not only in cereals
but also in other grains, legumes, root vegetables, and fruits. The
molecular composition and properties of starch are mainly influ-
enced by the presence of the two glucose polymers: amylose and
amylopectin. Amylose is chemically defined as a glucan polymer
with mainly α-1,4 links and relatively few α-1,6 branch points.
Conversely, amylopectin has a highly branched structure with a
significant amount of α-1,6-linked D-glucan. Amylose-rich starches
are resistant to amylolysis more so than waxy or regular starches
[5]. Amylases hydrolyze native starches in vitro relatively slowly and
to a limited extent [6]. During processing, gelatinization greatly
boosts the in vitro rate of amylolysis [7]. Thus, the more the
gelatinization of starch, the quicker it will be digested [8]. In
1992, Englyst and colleagues [9] established the terms rapidly
digestible starch (RDS), slowly digestible starch (SDS), and
amylase-resistant starch (RS) to describe the rate of starch digestion
in vivo. They also supported previous finding that the physiological
form of the meal and the type of the starch are significant factors in
starch digestion rate, setting up a detection method for non-starch
polysaccharides (NSP) and, for the first time, amylase-resistant
starch. Indeed, NSP provided the most accurate assessment of
dietary fiber (DF), after the complete hydrolysis of digestible
starch. The potential of sourdough processing to reduce starch
digestibility has been attributed to many processes. It is believed
that the impact is mostly caused by the creation of organic acids,
particularly lactic acid, during fermentation. The physiological pro-
cesses behind the acute effects of acids appear to vary; while lactic
acid slows the pace of starch digestion in bread [10], acetic and
propionic acids appear to slow the rate of stomach emptying
[11]. Moreover, it has been hypothesized that chemical changes
occurring during sourdough fermentation reduce the degree of
starch gelatinization [8], which would partially explain the reduced
digestibility of sourdough-fermented cereal diets. Usually, lactic
acid bacteria are not able to use directly starch for acids production.
However, the amylolytic properties of specific lactic acid bacteria
and the cooperation involving yeasts in sourdough microbiota can
transform starch into sugars and subsequently into organic acid.
 Quantification of Carbohydrates 83

Some lactic acid bacteria strains (mainly belonging to Weissella

genus) can also be used in sourdough fermentation to produce
exopolysaccharides (EPS) from free sugars (e.g., glucose and fruc-
tose) [12], whereas sourdough fermentation was correlated to the
increase of oligosaccharides and soluble dietary fiber [13–
15]. These, also including EPS, are mainly evaluated by the
enzymatic-gravimetric approach which includes the complete solu-
bilization and hydrolysis of digestible carbohydrates and proteins,
and then the precipitation with cold ethanol. Nevertheless, EPS
quantification will not be a subject of this chapter.
For the determination of soluble sugars and polysaccharides,
enzymatic kits from several manufacturers (e.g., Megazyme, Sigma-
Aldrich, and Bio-Rad), often based on official methods of analysis,
are available and recognized by the scientific community [16–
18]. The first step in the assay is the conversion or degradation of
the carbohydrate to glucose or glucose derivate, and then glucose
concentration is determined according to the two different detec-
tion systems: glucose oxidase or hexokinase in combination with
the glucose-6-phosphate dehydrogenase reaction. Hence, the con-
centration of the compound of interest is measured quantifying
spectrophotometrically the nicotinamide adenine dinucleotide
(NADH/NAD+) coenzyme system or H2O2 production whose
quantity is stoichiometric to that of glucose [19]. Overall, the
high specificity of enzymes enables the analysis of complex matrixes
without complicated sample preparation techniques and little inter-
ferences which, on the contrary, might occur during chro-
matographic coelution. Nevertheless, if on the one hand,
enzymatic assays allow the analysis of several samples in relatively
little amount of time, the higher costs of analysis per sample com-
pared to chromatographic techniques tip the scale in favor of the
latter.
Chromatographic techniques, liquid chromatography
(LC) more than gas chromatography (GC), are usually applied for
the qualitative and quantitative analysis of carbohydrates. Although
more appropriate for small and thermally stable metabolites like
volatile compounds, GC can be used for the analysis of sugars as
methylsilyl derivatives [2, 20]. Still, derivatization has its challenges
and, despite the advantages of GC, one of its limitations includes
the mass range of ca. 50–600 Da, hence excluding higher molecular
weight oligomers [20]. On the contrary, mass range covered by LC
systems is wider compared to that of GC platforms making high-
performance liquid chromatography (HPLC) the technique of
choice for the analysis of carbohydrates. As for the detection
method after separation, although refractive index (RI) is the
most employed for sourdough samples [21–23], it is not very
sensitive for low concentrations [24]. More recently, high-
performance anion exchange chromatography coupled with
pulsed-amperometric detection (PAD) was proposed for
84 Michela Verni and Marco Montemurro

legume-based sourdough [25, 26]. PAD is highly selective and

sensitive because only reactive compounds will give response and
at very low concentrations [24].
The determination of soluble sugars in sourdough using HPLC
often happens simultaneously with organic acid and ethanol using a
chromatographic system equipped with ion-exclusion columns
such as Aminex HPX-87H [2, 27, 28] [see Chapter 6]. Simulta-
neous detection of carboxylic acids and carbohydrates using UV
and RI detection was first optimized and described by Doyon et al.
[29] adjusting mobile phase and column temperature to change the
selectivity of the solutes while avoiding overlapping detection situa-
tions. The Aminex HPX-87H is the column of choice for the
analysis of monosaccharides found in solution with carboxylic
acids, volatile fatty acids, alcohols, ketones, and many neutral meta-
bolites, providing separation of di-, tri-, and tetrasaccharides as
well. This column allows the use of simple isocratic methods and
elution with water, organically modified water, or diluted acid,
usually 5–10 mM H2SO4 [27]. Since a preliminary treatment of
the extract is necessary to achieve the extraction of the compounds
of interest and eliminate interferences, perchloric acid is often used
for protein precipitation. Nevertheless, perchloric acid does not
allow the quantification of sucrose and other oligofructans because
these are hydrolyzed to glucose and fructose in the presence of 3.5%
perchloric acid [30] and, hence, this method cannot be recom-
mended for samples containing sucrose. Alternatively, extractions
directly with the eluent to be used for separation (H2SO4) have
been proposed [18, 31], as well as simple extraction in water
followed by heat treatment to inactivate enzymes and microbes
[25, 26]; still, most of the studies favor the use of aqueous ethanol
followed by normal-phase chromatography and RI detection [21–
23, 32] as optimized by Hernandez et al. [33].
As for polysaccharides, a variety of techniques could be used to
measure starch involving a three-step quantification. Firstly, starch
is hydrolyzed using chemicals (e.g., potassium hydroxide) or
enzymes (e.g., amylases) to produce a range of linear and branched
dextrins. The hydrolysis is then completed using amyloglucosi-
dases, which releases glucose, later quantified by spectroscopy or
chromatography. Soluble and insoluble dietary fibers, instead, are
often determined following the official methods of analysis as
described by Prosky and colleagues [34]. The principle is based
on the hydrolysis of proteins and starch using α-amylase, protease,
and amyloglucosidase, filtration of insoluble fiber (IDF), and pre-
cipitation of soluble fiber (SDF) in ethanol. IDF and SDF contents
are then evaluated by weighing after drying.
Based on the above considerations, in this chapter, the analysis
of dietary fibers (according to official method AOAC n. 991.43 and
AACC n. 32-07.01), the enzymatic hydrolysis of starch and the
 Quantification of Carbohydrates 85

chromatographic method for mono-, di-, and oligosaccharides will

be described.

2 Materials

Prepare solutions using ultrapure water (obtained by purifying

deionized water with Milli-Q® system) and analytical grade
reagents. Diligently follow all waste disposal regulations when dis-
posing waste materials.

2.1 Extraction and 1. Chromatographic system (e.g., Agilent 1200 Series HPLC
Chromatographic apparatus comprising degasser system, quaternary pump, col-
Determination of umn heater, refractive index detector, and a Rheodyne
Soluble Sugars injector loop).
2. Normal phase column for HPLC (e.g., Spherisorb Amino
(NH2) column 8 nm, 5 μm, 4.6 mm × 250 mm (Waters,
Milford, MA, USA)).
3. Extraction solution: aqueous ethanol 80%. In a flask, add
400 mL of 99.9% v/v ethanol (gradient grade for liquid chro-
matography) and 100 mL of water, cover the container, and
store at room temperature until use (see Note 1).
4. Mobile phase: acetonitrile solution 82%. In a graduated glass
cylinder, add 820 mL of 100% v/v acetonitrile (gradient grade
for liquid chromatography) and bring to volume (1 L) with
water. Transfer in a bottle and cover with a screw cap equipped
with a hose connection set. Store at room temperature until use
(see Note 1).
5. HPLC-grade sugars to be used as standard (e.g., fructose,
glucose, sucrose, maltose, raffinose, stachyose, verbascose).
6. Conical tubes (50 mL) and centrifuge.
7. Analytical balance.
8. Laboratory vortex or orbital shaker.
9. Rotavapor or SpeedVac™ concentrator.
10. Micro-pipettors (e.g., 1000 μL Gilson Pipetman®).

2.2 Determination of 1. Sodium hydroxide solution: 1.7 M. Weigh 68 g NaOH and

Starch add it slowly to 900 mL of deionized water in a 2 L beaker
under continuous stirring (see Note 2). When all the NaOH is
dissolved, make up to 1 L with deionized water in a glass
cylinder. Store in a glass bottle at room temperature.
2. Sodium acetate buffer: 100 mM, pH 5.0, supplemented with
calcium chloride (5 mM). Use a glass pipette to collect 5.8 mL
of glacial acetic acid and add to 900 mL of distilled water in a
beaker. Adjust the pH to 5.0 with 1 M NaOH. Add 0.74 g of
86 Michela Verni and Marco Montemurro

calcium chloride dihydrate and after complete dissolution, in a

cylinder, make up to 1 L with deionized water. Store at 4 °C in
a glass bottle.
3. Sodium acetate buffer: 600 mM, pH 3.8, supplemented with
calcium chloride (5 mM). Use a glass pipette to collect 34.8 mL
of glacial acetic acid and add to 900 mL of distilled water in a
beaker. Adjust the pH to 3.8 with 1 M NaOH. Add 0.74 g of
calcium chloride dihydrate and after complete dissolution, in a
cylinder, make up to 1 L with deionized water. Store at 4 °C in
a glass bottle.
4. Solution of thermostable a-amylase (3000 U/mL) (see Note
3).
5. Solution of amyloglucosidase (AMG) (3300 U/mL).
6. Water bath set at 50 °C.
7. Materials needed for glucose quantification with a glucose
colorimetric assay kit (e.g., Megazyme, Invitrogen) or by
using chromatographic determination as described in this
chapter.

2.3 Determination of 1. Crucible, Pyrex® 50 mL, pore size coarse (40–60 μm).
Soluble and Insoluble 2. Micro 90 or Deacon 90 cleaning solution.
Fibers
3. Celite, acid washed; for use in total dietary fiber assay
procedures.
4. Purified protease from Bacillus licheniformis ~ 350 tyrosine
U/mL (see Note 4).
5. Solution of thermostable a-amylase (3000 U/mL) (see Note
3).
6. Solution of amyloglucosidase (3300 U/mL) in sodium acetate
buffer (100 mM, pH 5.0, supplemented with calcium chloride
5 mM).
7. Ethanol, 95% v/v, and ethanol, 78% (821 mL of 95% v/v
ethanol in 1 L).
8. MES/TRIS buffer: 0.05 M each, pH 8.2. Dissolve 19.52 g 2
(N-morpholino)ethanesulfonic acid (MES) and 12.2 g tris
(hydroxymethyl)aminomethane (TRIS) in 1.7 L deionized
water. Adjust the pH to 8.2 with 6.0 N NaOH and dilute to
2 L with water.
9. Hydrochloric acid solution, 0.561 N. Add 93.5 mL of 6 N HCl
to approximately 800 mL of water in 1 L volumetric flask and
then dilute to 1 L with water.
10. Acetone.
11. Analytical balance and shaking water bath.
12. Muffle furnace for ash quantification.
 Quantification of Carbohydrates 87

13. Equipment and materials for the determination of proteins

using the Kjeldahl.

3 Methods

3.1 Extraction and 1. Weigh 1 g of freeze-dried sample in a conical tube.

Quantification of 2. Add 10 mL of ethanol solution.
Soluble Sugars
3. Homogenize for 1 min at room temperature.
4. Centrifuge for 5 min at 500× g.
5. Decant the supernatant and repeat the procedure on the pellet
twice as illustrated in Fig. 1.
6. Evaporate to dryness the combined supernatants and resus-
pend the dried extract in 1 mL of mobile phase (see Note 5).
7. Prepare a calibration curve dissolving the soluble sugars of
interest in the mobile phase (see Note 6).
8. Set the method using the following parameters: isocratic elu-
tion (35 min) with 82% acetonitrile, flow rate 1.2 mL/min,
column temperature set at 28 °C, 20 μL injection loop (see
Note 7).
9. Inject the sugar standard in the HPLC to determine retention
times and do the same with the samples.
10. Elaborate the data using the calibration curve equation and
apply potential dilution factors to quantify soluble sugars.

3.2 Determination of 1. Weigh 100 mg in duplicate in the bottom of 50 mL centrifuge

Starch tube recording the exact weight.
2. Add 0.2 mL of ethanol/water solution (80/20 v/v) and stir
the tubes on a vortex mixer to completely disperse the sample
in the solution.
3. Add 2 mL of cold 1.7 M sodium hydroxide solution and vortex
it for at least 15 seconds before all the lumps are dissolved.
Place the tubes in an ice-water bath and shake it for 15 min. At
5 min intervals, vortex the tube vigorously for 15 sec and put it
again in the water bath before the end of incubation time.
4. Add 8 mL of sodium acetate buffer (600 mM, pH 3.8 supple-
mented with calcium chloride). Vortex the tube and check the
pH. If it is not in between 4.8 and 5.2, adjust it within the
range (see Note 8).
5. In one of the tubes, promptly add 0.1 mL of thermostable
α-amylase solution and then 0.1 mL of AMG (3300 U/mL).
Use the second tube as a control adding only 0.2 mL of sodium
acetate buffer (100 mM, pH 5.0 8, supplemented with calcium
chloride).
88 Michela Verni and Marco Montemurro

Fig. 1 Extraction of soluble sugars in sourdough samples

6. Vortex all the tubes for 3 s and incubate at 50 °C for 30 min.

7. Cool down the tubes at room temperature.
8. Mix well and use this solution for glucose quantification (see
Note 9).

3.3 Determination of 1. Crucible preparation (to be performed at least 1 day before

Soluble and Insoluble fiber determination as described in Fig. 2): Burn all the residues
Fibers overnight at 525 °C in a muffle furnace before the use and then
 Quantification of Carbohydrates 89

Fig. 2 Determination of soluble and insoluble fibers in sourdough samples

90 Michela Verni and Marco Montemurro

remove all the materials by using a vacuum. Soak in 2% Micro

90 cleaning solution (or Deacon 90) at room temperature for
1 h. Wash the crucible accurately with deionized water with the
last rinse by using 15 mL acetone and air dry. Add approxi-
mately 1.0 g of Celite to dried crucibles and dry at 130 °C to
constant weight. Cool crucible in a desiccator for 1 h and note
down the weight.
2. Accurately weigh 1 g of dried (e.g., lyophilized) sample into
500 mL lab flask (see Note 10). Carry out a blank using a flask
without sample.
3. Add 40 mL of MES-TRIS (pH 8.2) to each flask and stir with a
magnetic stirrer until the sample is thoroughly homogenized.
4. Add 50 μL of the amylase solution and incubate in a shaking
water bath with continuous agitation at 98–100 °C for 30 min.
5. Remove all sample from the water bath and cool to approxi-
mately 60 °C.
6. Recover all the sample from the flask side wall with a spatula
and use 10 mL of deionized water to be sure that all the
samples from side wall and spatula are at the bottom of the
flask.
7. Adjust temperature of water bath to 60 °C (see Note 11).
8. Add 100 μL protease solution to each sample and incubate in
shaking water bath with continuous agitation at 60 °C for
30 min.
9. Remove sample from the water bath and add 5 mL of 0.561 N
HCl solution while stirring. Check if the pH of the samples is in
the range 4.1–4.8, if not adjust it with 5% NaOH or 5% HCl
solutions.
10. Add 200 μL of AMG while stirring and incubate in shaking
water bath with continuous agitation at 60 °C for 30 min.
11. Use approximately 5 mL of distilled water to redistribute the
celite in the crucible and use a filtration flask to dry the
celite bed.
12. Filter the sample obtained after enzymatic treatments
with AMG.
13. Add 10 mL of distilled water (heated at 70 °C in the water
bath) in the empty flask to rinse all the samples residues. Repeat
the procedure if required. Filter the water as well, by using the
filtration flask, note down the final volume of the permeate,
and transfer it in a 500 mL volume bottle.
14. For the separation of IDF, wash the retentate in the crucible
with 10 mL of ethanol 95% v/v and subsequently acetone.
Repeat the washing twice.
 Quantification of Carbohydrates 91

15. For the separation of SDF, add 4 volumes of pre-heated (60 °

C) ethanol 95% v/v to the permeate obtained in the bottle after
step 13 and incubate at room temperature for 1 h.
16. Take another crucible with celite and use approximately 15 mL
of ethanol 78% v/v to redistribute the celite in the crucible.
Use a filtration flask to dry the celite bed.
17. Filter the sample obtained after fiber precipitation (step 15).
18. Wash the retentate in the crucible obtained in the step 17 with
10 mL of ethanol 78% v/v, 10 mL of ethanol 95% v/v, and
subsequently 10 mL of acetone. Repeat the washing twice.
19. For the quantification, dry crucible at step 14 (IDF) and
18 (SDF) overnight in 103 °C oven.
20. Cool the crucible in a desiccator for around 1 h. Weigh the
crucible containing dietary fiber residue and Celite. To get the
residual weight, deduct the tare weight, which is the weight of
the dry crucible and Celite obtained in the step 1.
21. From the retentate residues (for both IDF and SDF), evaluate
protein and ash content using the Kjeldahl method and the
muffle furnace at 525 °C for 5 h, respectively. Subtract the
values of protein and ash from the fiber weight. Subtract the
values of the blank to obtain the final concentrations.

4 Notes

1. Prepare the solution under a fume hood and use personal

protective equipment (e.g., gloves).
2. Sodium hydroxide is corrosive, and to avoid irritation or other
issues, use gloves and other safety devices. NaOH is highly
soluble in water and the dissolution is a highly exothermic
reaction where a large amount of heat is liberated; hence,
make sure to slowly add small amount of powder to water
when preparing the solution.
3. A ready-to-use solution can be bought; otherwise, it is possible
to prepare it with the powdered enzyme (another name of the
thermostable a-amylase is Termamyl 120). When preparing it
from scratch, make sure to check the enzymatic unit of the
powder and accordingly prepare the solution to get the final
enzyme concentration.
4. Protease could be prepared by using powdered enzyme (e.g.,
protease from Bacillus licheniformis P3910, Sigma) according
to the unit/mg reported in the product data sheet or bought as
a ready-to-use solution (e.g., Protease, Subtilisin A from Bacil-
lus licheniformis E-BSPRT, Megazyme).
92 Michela Verni and Marco Montemurro

5. It is crucial that all the extraction solvents are removed. The

dried extract must be dissolved in the mobile phase to avoid
water/ethanol interferences during the chromatographic run.
Indeed, the incomplete evaporation of all ethanol will result in
a peak detected in the first few minutes of the run which might
shift the whole chromatogram.
6. Use standard solutions freshly prepared and keep them refri-
gerated until use. Concentration ranges and type of sugars
should be chosen based on those expected in the sample,
since highly related to type of flour and protocol used.
7. It is recommended to carry out an isocratic elution of
acetonitrile-water 82:18 at a flow rate of 1.2 mL/min for
monosaccharides detection, whereas oligosaccharides are bet-
ter separated at 68:32 ratio with a flow rate of 0.8 mL/min.
8. Be careful and try to clean well the pH-meter probe before and
after the measurement to avoid sample contamination or loss.
9. Dilution is usually needed for glucose quantification depending
on sample type. Usually, a 1/5 or 1/10 dilution is needed for
sourdough samples.
10. Lab flask should be always covered by aluminum foil during
incubations to avoid evaporation. Alternatively, glass bottle
could be used.
11. If the water bath does not have a refrigerator incorporated,
drain some hot water, and use ice to faster the temperature
decrease. Wait until the temperature is stable before incubating
the samples.

References

1. G€anzle MG, Vermeulen N, Vogel RF (2007) 5. Asp NG, Björck I (1992) Resistant starch.
Carbohydrate, peptide and lipid metabolism of Trends Food Sci Technol 3:111–114
lactic acid bacteria in sourdough. Food Micro- 6. Björck I, Asp NG (1994) Controlling the
biol 24:128–138 nutritional properties of starch in foods—a
2. Lefebvre D, Gabriel V, Vayssier Y, Fontagne- challenge to the food industry. Trends Food
Faucher C (2002) Simultaneous HPLC deter- Sci Technol 5:213–218
mination of sugars, organic acids and ethanol 7. Lauro M, Poutanen K, Forssell P (2000) Effect
in sourdough process. LWT-Food Sci Technol of partial gelatinization and lipid addition on
35:407–414 α-amylolysis of barley starch granules. Cereal
3. Durán E, León A, Barber B, Benedito de Bar- Chem 77:595–601
ber C (2001) Effect of low molecular weight 8. Östman E (2003) Fermentation as a means of
dextrins on gelatinization and retrogradation optimizing the glycaemic index-food mechan-
of starch. Eur Food Res Technol 212:203–207 isms and metabolic merits with emphasis on
4. Chen Y, Eder S, Schubert S, Gorgerat S, lactic acid in cereal products. Lund University
Boschet E, Baltensperger L, Boschet E, 9. Englyst HN, Kingman SM, Cummings JH
St€adeli C, Kuster S, Fischer P, Windhab EJ (1992) Classification and measurement of
(2021) Influence of amylase addition on nutritionally important starch fractions. Eur J
bread quality and bread staling. ACS Food Sci Clin Nutr 46:S33–S50
Technol 1:1143–1150 10. Liljeberg HG, Lönner CH, Björck IM (1995)
Sourdough fermentation or addition of
 Quantification of Carbohydrates 93

organic acids or corresponding salts to bread chemical and nutritional characteristics of

improves nutritional properties of starch in wheat germ. Food Chem 119:1079–1089
healthy humans. J Nutr 125:1503–1511 22. Coda R, Kianjam M, Pontonio E, Verni M, Di
11. Liljeberg H, Björck I (1998) Delayed gastric Cagno R, Katina K, Rizzello CG, Gobbetti M
emptying rate may explain improved glycaemia (2017) Sourdough-type propagation of faba
in healthy subjects to a starchy meal with added bean flour: dynamics of microbial consortia
vinegar. Eur J Clin Nutr 52:368–371 and biochemical implications. Int J Food
12. Ripari V (2019) Techno-functional role of exo- Microbiol 248:10–21
polysaccharides in cereal-based, yogurt-like 23. Rizzello CG, Cavoski I, Turk J, Ercolini D,
beverages. Beverages 5:16 Nionelli L, Pontonio E, De Angelis M, De
13. Manini F, Brasca M, Plumed-Ferrer C, Filippis F, Gobbetti M, Di Cagno R (2015)
Morandi S, Erba D, Casiraghi MC (2014) Organic cultivation of Triticum turgidum
Study of the chemical changes and evolution subsp. durum is reflected in the flour-
of microbiota during sourdough like fermenta- sourdough fermentation-bread axis. Appl
tion of wheat bran. Cereal Chem 91:342–349 Environ Microbiol 81:3192–3204
14. Mao M, Wang P, Shi K, Lu Z, Bie X, Zhao H, 24. Giannoccaro E, Wang YJ, Chen P (2008)
Zhang C, Lv F (2020) Effect of solid state Comparison of two HPLC systems and an
fermentation by Enterococcus faecalis M2 on enzymatic method for quantification of soy-
antioxidant and nutritional properties of bean sugars. Food Chem 106:324–330
wheat bran. J Cereal Sci 94:102997 25. Perri G, Coda R, Rizzello CG, Celano G,
15. Zhao HM, Guo XN, Zhu KX (2017) Impact of Ampollini M, Gobbetti M, De Angelis M,
solid state fermentation on nutritional, physical Calasso M (2021) Sourdough fermentation of
and flavor properties of wheat bran. Food whole and sprouted lentil flours: In situ forma-
Chem 217:28–36 tion of dextran and effects on the nutritional,
16. Siragusa S, Di Cagno R, Ercolini D, texture and sensory characteristics of white
Minervini F, Gobbetti M, De Angelis M bread. Food Chem 355:129638
(2009) Taxonomic structure and monitoring 26. Perri G, Rizzello CG, Ampollini M, Celano G,
of the dominant population of lactic acid bac- Coda R, Gobbetti M, De Angelis M, Calasso M
teria during wheat flour sourdough type I (2021) Bioprocessing of barley and lentil grains
propagation using Lactobacillus sanfranciscen- to obtain in situ synthesis of exopolysaccharides
sis starters. Appl Environ Microbiol 75:1099– and composite wheat bread with improved tex-
1109 ture and health properties. Foods 10:1489
17. Gulati P, Sabillón L, Rose DJ (2018) Effects of 27. Hamad SH, Böcker G, Vogel RF, Hammes WP
processing method and solute interactions on (1992) Microbiological and chemical analysis
pepsin digestibility of cooked proso millet of fermented sorghum dough for Kisra produc-
flour. Food Res Int 109:583–588 tion. Appl Microbiol Biotechnol 37:728–731
18. Minervini F, Dinardo FR, Celano G, De 28. Dinardo FR, Minervini F, De Angelis M,
Angelis M, Gobbetti M (2018) Lactic acid Gobbetti M, G€anzle MG (2019) Dynamics of
bacterium population dynamics in artisan sour- Enterobacteriaceae and lactobacilli in model
doughs over one year of daily propagations is sourdoughs are driven by pH and concentra-
mainly driven by flour microbiota and nutri- tions of sucrose and ferulic acid. LWT – Food
ents. Front Microbiol 9:1984 Sci Technol 114:108394
19. Al-Mhanna NM, Huebner H, Buchholz R 29. Doyon G, Gaudreau G, St-Gelais D,
(2018) Analysis of the sugar content in food Beaulieu Y, Randall CJ (1991) Simultaneous
products by using gas chromatography mass HPLC determination of organic acids, sugars
spectrometry and enzymatic methods. Foods and alcohols. Can Inst Food Sci Technol J 24:
7:185 87–94
20. Adebo OA, Oyeyinka SA, Adebiyi JA, Feng X, 30. Thiele C, G€anzle MG, Vogel RF (2002) Sam-
Wilkin JD, Kewuyemi YO, Abrahams AM, ple preparation for amino acid determination
Tugizimana F (2021) Application of gas by integrated pulsed amperometric detection
chromatography–mass spectrometry in foods. Anal Biochem 310:171–178
(GC-MS)-based metabolomics for the study 31. De Angelis M, Minervini F, Siragusa S, Riz-
of fermented cereal and legume foods: a review. zello CG, Gobbetti M (2019) Wholemeal
Int J Food Sci Technol 56:1514–1534 wheat flours drive the microbiome and func-
21. Rizzello CG, Nionelli L, Coda R, De tional features of wheat sourdoughs. Int J Food
Angelis M, Gobbetti M (2010) Effect of sour- Microbiol 302:35–46
dough fermentation on stabilisation, and
94 Michela Verni and Marco Montemurro

32. Verni M, De Mastro G, De Cillis F, evaporative light-scattering detection and

Gobbetti M, Rizzello CG (2019) Lactic acid refractive index detection. J Chromatogr Sci
bacteria fermentation to exploit the nutritional 36:293–298
potential of Mediterranean faba bean local bio- 34. Prosky L, Asp NG, Schweizer TF, DeVries JW,
types. Food Res Int 125:108571 Furda I (1988) Determination of insoluble,
33. Hernandez JL, González-Castro MJ, Alba IN, soluble, and total dietary fibre in foods and
De La Cruz GC (1998) High-performance food products. J Assoc Off Anal Chem 71:
liquid chromatographic determination of 1017
mono-and oligosaccharides in vegetables with
 Part III

Sensory, Rheology and Nutritional Attributes of Sourdough

Baked Goods
 Chapter 9

Volume Determination
Marco Montemurro

Abstract
Volume determination is an important and easy-to-do procedure for assessing the impact of inclusion of
ingredients or process setup in bread and in other bakery products. Rapeseed displacement procedure is the
oldest and widely diffused procedure used for this analysis. However, it involves the use of volumeter, which
is usually substituted with beaker and cylinders, and this allows to perform the volume determination of
bread without any other specific instruments. Moreover, two different official methods based on laser
scanning were proposed in the last 20 years.

Key words Bread volume, Specific volume, Rapeseed displacement

1 Introduction

The volume of bread is the most significant exterior attribute. It is

an essential element in the quality control program and consumer
appeal evaluation for bread baking. Typically, this is a sign of a well-
aerated crumb and outstanding texture, as well as adequate formu-
lation and ingredient quality, dough management, gas retention,
and processing conditions. Indeed, higher volume per unit weight
is generally connected with airier crumb and improved texture.
A particularly powerful flour may result in a loaf with a tiny
volume, indicating the necessity for a longer fermentation time
during which the gluten will mature and become more extensible.
This volume information may then be utilized to adjust the bread
dough mixture to make bread of the desired quality. Indeed, high
specific volume of the loaf is usually correlated with soft wheat
bread and higher consumer’s acceptability [1]. In this scenario,
sourdough biotechnology represents a useful tool to increase the
rheological properties of bread [2], especially when exopolysac-
charides (EPS)-producing strains are used as microbial starters
[3, 4]. Moreover, the inclusion of flours other than soft wheat in
bread formulation is usually coupled with the sourdough

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

97
98 Marco Montemurro

biotechnology, overcoming the disadvantages of the presence of

higher amount of fiber and lower amount of gluten. Indeed, sour-
dough biotechnology was positively applied to design a gluten-free
bread without the inclusion of additives in the recipe, which are
usually used to counteract the lack of gluten [5].
Conversely, Burton and Lightowler [6] demonstrated that the
decrease in loaf volume, but constant macronutrients content, is
correlated to lower plasma glucose levels and glycemic index
(GI) values.
The volume can be determined by using different methods.
However, three official methods are usually used: the AACC Inter-
national Method 10-05.01 as well-known as the rapeseed displace-
ment procedure [7] and the AACCI Approved Method 10-14.01
involving the laser scanning [8] then improved by the AACC
Standard Method 10-16.01 which includes also the dimensional
profile of the products [9].
In the AACC International Method 10-05.01, the quantity of
rapeseed packing the bread loaf in a conventional container has
been used to determine the volume of baked products. Therefore,
the first method needs relatively easy-to-find apparatus, while the
second and third ones need the presence of a specific instrument in
the laboratory (e.g., Volscan Profiler). However, the need for
repeated calibration, operator dependence, loss of rapeseed due to
spillage/adherence of seeds to the product or to static electricity,
seed clumping due to moisture absorption, periodic sieving of the
seed to remove food crumbs, the potential crushing of soft pro-
ducts, and the manual recording of results are common drawbacks
affecting the measurement’s efficiency and accuracy. Because of
these factors and the potential for implementing novel technolo-
gies, other two methods were set up and tested giving the possibil-
ity to choose among different official methods.
In this chapter, the procedure of rapeseed displacement
method with some modification (e.g., without the use of volume-
ter) is described and compared to the official AACC International
Method 10-05.01, aiming at the possibility of determining bread
volume when volume determination cannot be performed by using
laser (Fig. 1).

2 Materials

After baking, the volume determination should be performed when

the product is most firm. A good time slot for the analysis is few
minutes after baking, less than 5 min, which allows to perform the
analysis when the crust is still hard. Otherwise, the analysis can be
performed 24 h after baking, considering the moisture equilibra-
tion in the samples completed.
 Volume Determination 99

Rapeseed in the beaker

Quantification of rapeseed (V1)

Bread/calibration block in the

beaker

Rapeseed in the beaker with

bread

Quantification of rapeseed (V2)

Fig. 1 Volume determination by using rapeseed displacement procedure

2.1 Materials for 1. Block with similar shape of the bread to be evaluated with a
Calibration and Bread volume in the median point of the one of the bread loafs (see
Evaluations Note 1).
2. Graduated cylinder and two graduated beakers. Beakers and
cylinders (or the volumeter) must cover all the volumes of the
samples to be tested (see Note 2).
3. Rapeseeds (see Note 3).

3 Methods

3.1 Calibration 1. Put the rapeseed in the graduated beaker till they reach the
highest volume line (or in the volumeter in the upper housing).
2. Check carefully that the amount of rapeseed is the right one by
evaluating the surface of the beaker (or closing the gate of the
housing in the volumeter).
3. Remove the excess of the rapeseed or add other seeds if they
were not enough.
4. Measure the volume of the rapeseed in graduated
cylinder (V1).
5. Put a layer of rapeseed in the graduated beakers.
100 Marco Montemurro

6. Add the calibration block in the beaker (or in the lower housing
of the volumeter).
7. Pour the rapeseeds in the beaker till they get the highest
volume line (V2, volume of rapeseed in the beaker containing
calibration block). Be careful in the evaluation of the surface of
the seeds in the beaker.
8. Measure the volume of the rapeseeds that are not in the evalu-
ation beaker and record the value. This corresponds to the
difference between the initial volume of rapeseed (V1, as
described in step 4) and the volume of the rapeseed in the
beaker during the calibration block evaluation (V2, as
described in step 7):

Calibration block volume = V 1 - V 2

9. Do the calibration at least three times (see Note 4)

3.2 Samples For samples evaluation, use the bread loaf instead of the calibration
Evaluation block and follow steps from 5 to 8 of the method for the calibra-
tion. Repeat the passages from 1 to 4 of the calibration in the
between of the samples (see Note 4).
Specific volume can be evaluated after weighing the samples
(W) in the analytical balance and it is calculated by the following
equation:
V1-V2
Specific volume =
W
The specific volume is usually expressed as cm3/g value.

4 Notes

1. The block should be in metal; please avoid old bread loafs or

other sticky materials that could decrease the amount of rape-
seed during the analysis.
2. In the official method is described the use of the volumeter.
This is a specific tool to be used for the volume determination
of bread in which two housing chambers (one for the bread and
the other one for the rapeseed) communicate by a graduated
cylinder [10]. Instead of volumeter, two beakers and a cylinder
can also be used. If the beakers and the cylinder are used, be
careful in transferring the rapeseed from a tool to the other
avoiding seeds losses.
3. Instead of rapeseed, other seeds with similar particle size can
also be used [11].
4. Calibration is necessary to avoid error due to different rapeseed
displacement in the evaluation beaker and to check the
 Volume Determination 101

rapeseed loss during the measurement. In the latter case, add

rapeseed to get the correct volume again. Moreover, bread
particles can be mixed with rapeseed after analysis; therefore,
the rapeseed should be periodically sieved to remove both small
and large particles.

References
1. Pasqualone A, Caponio F, Pagani MA, Guidelines for measurement of volume by
Summo C, Paradiso VM (2019) Effect of salt rapeseed displacement. First approval October
reduction on quality and acceptability of 17, 2001. Cereals & Grains Association,
durum wheat bread. Food Chem 289:575– St. Paul
581 8. Anderson S, Purhagen JK, Bason ML (2014)
2. Arora K, Ameur H, Polo A, Di Cagno R, Riz- AACCI approved methods technical commit-
zello CG, Gobbetti M (2021) Thirty years of tee report: collaborative study on bread volume
knowledge on sourdough fermentation: a sys- determination by laser topography using a
tematic review. Trends Food Sci Technol 108: bread volume meter. Cereal Foods World 59:
71–83 294–296
3. Katina K, Maina NH, Juvonen R, Flander L, 9. Smewing J (2016) AACCI approved methods
Johansson L, Virkki L, Tenkanen M, Laitila A technical committee report on the collabora-
(2009) In situ production and analysis of Weis- tive study for a new AACCI method
sella confusa dextran in wheat sourdough. (10-16.01): volumetric and dimensional pro-
Food Microbiol 26:734–743 file determination of baked products using laser
4. Chen XY, Levy C, G€anzle MG (2016) topography-volscan profile. Cereal Foods
Structure-function relationships of bacterial World 61:18–23
and enzymatically produced reuterans and dex- 10. Vanhamel S, Vandenende L, Darius PL, Del-
tran in sourdough bread baking application. cour JA (1991) A volumeter for breads
Int J Food Microbiol 239:95–102 prepared from 10 grams of flour. Cereal
5. Montemurro M, Pontonio E, Rizzello CG Chem 68:170–172
(2021) Design of a “clean-label” gluten-free 11. Mohd Roby BH, Muhialdin BJ, Abadl MMT,
bread to meet consumers demand. Foods 10: Mat Nor NA, Marzlan AA, Lim SAH, Musta-
462 pha NA, Meor Hussin AS (2020) Physical
6. Burton P, Lightowler HJ (2006) Influence of properties, storage stability, and consumer
bread volume on glycaemic response and sati- acceptability for sourdough bread produced
ety. Br J Nutr 96:877–882 using encapsulated kombucha sourdough
7. AACC International. Approved methods of starter culture. J Food Sci 85:2286–2295
analysis, 11th Ed. Method 10–05.01,
 Chapter 10

Texture Profile Analysis

Marco Montemurro and Erica Pontonio

Abstract
Texture profile analysis (TPA) is the most common test used for analyzing the structural properties of bread.
The characteristics of bread prepared according to different recipes are exploited by the evaluation of both
crust and crumb. Moreover, the staling process of bread during storage is directly correlated with changing
in firmness, springiness, and resilience values.

Key words Texture profile analysis, TPA, Bread, Bread slice, Bread staling

1 Introduction

The texture profile analysis (TPA) method is a test for determining

the textural properties of foods simulating a two bites compression,
thus providing insight into how samples behave when chewed. The
result is a set of values describing different characteristics of the
food sample. Before 1956, the textural properties of food were
evaluated using different instruments, one for each parameter,
and the MIT’s Strain Gauge Denture Tenderometer was the first
successful attempt to measure a set of textural properties in one
time. However, only in 1963, the General Foods Texturometer
(GF Texturometer) was designed, which correlated instrumental
measure with sensory judgments defining all the attributes used
and classifying mechanical characteristics of food in principal and
secondary parameters [1]. Hardness, cohesiveness, viscosity, elas-
ticity, and adhesiveness were grouped into the first category, while
brittleness, chewiness, and gumminess in the second one [1]. How-
ever, the definition of textural parameters was constantly evolving,
indeed elasticity evolved into springiness since it had already rheo-
logical and engineering definitions, while brittleness evolved into
fracturability, which is considered to be a more accurate term to
define the breakage of the samples during the test [2].

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

103
104 Marco Montemurro and Erica Pontonio

Bread TPA is commonly used to evaluate differences among

samples obtained by using different ingredients or processes. To
better understand the meaning of each attribute, a correlation
between the definition and the results of a typical graph of TPA is
necessary. Hence, the definitions of the parameters in bread char-
acterization and how to obtain the corresponding values are
described in the following text and in Fig. 1.
Hardness is usually used to describe the crust properties and is
correlated to the maximum force required to press the bread at the
first bite, thus defining it as the force necessary to attain a given
deformation [3]. The hardness value corresponds to the higher
peak force in Newton (N) that occurs during the first compression.
Although force could be also measured as kilogram of force, the SI
unit is N with the conversion as follows: N = kg of force × 9.8 m/
sec2. Moreover, if the crust breaks during the compression, it is
possible to evaluate the fracturability that corresponds to the maxi-
mum force applied to the crust before breaking. The fracturability
point occurs where the force falls off in the first compression peak.
Firmness is usually used to describe a loss of softness in the
bread crumb, but it could also be used to evaluate the crust. In the
latter case, low values are considered negatively if the bread is a
“crusty” one. This parameter is usually used to evaluate the staling
rate of bread during to storage [4]. The idea for considering this
parameter as a shelf-life indicator of bread is the redistribution of
moisture between the dry crust (0.05–0.10 g moisture/100 g
product) and relatively moist crumb (0.4 g moisture/100 g prod-
uct) [5] and the recrystallization of starch mainly involving amylo-
pectin fraction [6]. The firmness of bread could be confused with
the hardness, which is usually evaluated in bread by setting the
compression between 30 and 50% [7]. However, the official
method considers the firmness as the force registered after 25% of
slice compression [4]. Like hardness, firmness is also measured in
Newton. Moreover, it is possible to evaluate the shelf-life of a bread
by calculating the firming rate. Therefore, it is possible to evaluate
the firmness at different times of storage and calculate the slope of
the linear regression by using the measured firmness values as a
function of storage time [8].
Springiness describes the way a product’s crumb springs back
and returns to its undeformed state after the end of the first
compression force. This is an important parameter to evaluate the
staling process of bread during storage [9] considering high values
related to freshness of the products. The springiness is calculated
dividing the time (in seconds or millimeters) necessary to reach the
highest point of the peak in the second compression and in the first
one. To evaluate the springiness, it is important to set properly the
analysis parameter because the time between the two compressions
should agree with the time between two chews. Springiness is a
 Texture Profile Analysis 105

Fig. 1 Graphical results of a texture profile analysis (TPA) of bread crumb. The figure shows how the results of
different parameters are calculated

ratio between two values (time or distances, depending on the

parameter of the x axis); therefore, no unit of measure is needed.
Like springiness, resilience is also related to the return to the
undeformed state of sample and represents how well a product
regains its original height. Although TPA is a double compression
analysis, resilience can be measured with a single compression;
however, the withdrawal speed should be the same as the compres-
sion speed, which is calculated by dividing the upstroke area of the
first compression peak by the downstroke one.
Adhesiveness is the force necessary to separate food from any
surface with which it has contact. The adhesiveness can be quanti-
fied by the negative peak force and expressed as a negative number
(N ×mm), thus increasing proportionally with the absolute value.
In bread analysis, the compression occurs without adding saliva.
Therefore, the adhesiveness is usually reported as zero because of
the absence of viscous or sticky materials that affect the textural
characteristics of most foods [10], which also increase the
adhesiveness.
Cohesiveness is considered as a positive characteristic of bread
[11] and represents how well the product withstands a second
deformation taking into account its resistance during the first
deformation. It is related to the strength of the internal bonds
making up the body of the product [3] and is calculated by dividing
areas of work during the second compression by that of the
first one.
Gumminess is mutually exclusive with chewiness, the first one is
usually used for semi-solid while the latter for solid products
[2]. Both describe the energy required to disintegrate food ready
106 Marco Montemurro and Erica Pontonio

for swallowing, but for bread evaluation, the parameter that better
describes this attitude is the latter one. The chewiness is calculated
by multiplying hardness, cohesiveness, and springiness.
Bread texture profile analysis is usually performed within sev-
eral hours of storage after baking, avoiding long storage that could
be correlated with changes in bread structure. Indeed, most of the
changes in the bread crumb occurs during 24 h after baking.
However, TPA of bread should not be performed directly after
baking and a period of storage is necessary. Bread cooling is neces-
sary, and at least 2 h is necessary to perform the TPA, while storage
longer than 24 h could affect the analysis results. The storage of the
bread after baking should be in plastic bags and at room tempera-
ture (approximately 25 °C), and the packaging should be per-
formed at least after 2 h of cooling. Storage, packaging, and
handling of sample before testing are considered parts of variable
conditions under which bread is tested; therefore, it is important to
maintain constants and report them for comparison purpose
among samples.

2 Materials

The weight of the bread loaf should be approximately of 200 g. All

bread samples should be prepared using the same water; therefore,
ultrapure water could be used, although tap water is usually
included for dough preparation. The optimal water absorption
and the time needed for obtaining the optimum consistency should
be evaluated before preparing bread by using a farinograph. Dough
preparation should be performed by a proofer. No chemicals are
used for performing the TPA; however, it is a destructive sample
analysis. Therefore, diligently follow all waste disposal regulations
when disposing waste materials. Aiming at evaluating bread tex-
ture, TPA could be performed either on the loaf or on the bread
slice.

2.1 Bread Samples 1. Bread loaf of 200 g.

and Equipment 2. Bread knife or automatic bread slicer (only for bread slice
analysis, see Note 1).
3. Plastic bags.
4. Room with temperature of approximately 25 °C or incubator.
5. Textural profile analyzer ([Link] texture analyser or similar
one) with load cell of 500 N (for bread loaf analysis, see Note
2) or 100 N (for bread slice analysis, see Note 3).
6. Cylindrical probe of 36 mm in diameter (see Note 4).
7. Laptop with management software for TPA data acquisition
and evaluation.
 Texture Profile Analysis 107

3 Methods

3.1 TPA of Bread 1. Use the bread after cooling for 2 h. If the textural profile
Loaf analysis is not performed within 4 h, put bread loafs in plastic
bags (not more than 24 h) before analysis.
2. Install the 36 mm cylindrical probe to the instrument and use
the TPA analyzer software for setting the correct probe
(depending on texture profile analyzer and software).
3. Set the zero point between the plate and the plunger (depend-
ing on texture profile analyzer and software).
4. Put the sample on the center of the plate. Be careful to evaluate
the position of the compression plunger, avoiding any irregular
or not representative areas of bread crust (see Note 5).
5. Set the analysis for the double compression test (this depends
on texture profile analyzer and software).
6. Evaluate the height of the sample and set the correct limit to
the plate considering a compression of 30% (see Note 6).
7. Place upper crosshead limit so that the compression plunger is
5 mm above the center surface of sample.
8. Set crosshead speed (rate of compression) at 1 mm/sec (see
Note 7).
9. Set the return speed at 5 mm/sec.
10. Start the analysis and prevent the plunger from becoming
trapped in the bread crust when it is going back after the first
compression (see Note 8).
11. Get the results and assess the manner in which they are pre-
sented (depending on texture profile analyzer and software)
(see Note 9).

3.2 TPA of Bread 1. After cooling for 2 h, cut bread into slices of 25 mm thickness
Slice with a bread knife or a bread slicer (this is preferred considering
the high reproducibility of the slice thickness). Use six central
slices from each loaf to perform textural analysis (see Note 10).
If the textural profile analysis is not performed within 4 hours,
put bread slice in plastic bags (not more than 24 hours) before
analysis.
2. Install the 36 mm cylindrical probe to the instrument and use
the TPA analyzer software for setting the correct probe
(depending on texture profile analyzer and software).
3. Set the zero point between the plate and the plunger (depend-
ing on texture profile analyzer and software).
108 Marco Montemurro and Erica Pontonio

4. Put the sample on the center of the plate. Be careful to evaluate

the position of the compression plunger, avoiding any irregular
or not representative areas of bread slices (see Note 11).
5. Set the analysis for the double compression test (this depends
on texture profile analyzer and software).
6. Place upper crosshead limit so that compression plunger is
1 mm above center surface of sample.
7. Place lower crosshead limit at 40% compression (10 mm com-
pression depth).
8. Set crosshead speed (rate of compression) at 100/min (see
Note 12).
9. Set chart speed at 500 mm/min (5:1 ratio of chart to crosshead
extension).
10. Start the analysis.
11. Get the results and and assess the manner in which they are
presented (this depends on texture profile analyzer and soft-
ware) (see Note 9).
12. For firmness evaluation, see the force reading at 25% compres-
sion which corresponds to the force registered after 31 mm
from the start of the force recording.

4 Notes

1. It is also possible to cut the bread crust off before the compres-
sion test. Usually, for central bread slice from 200 g loaf of soft
bread, the crust does not interfere with compression by a
36 mm diameter plunger. However, if the samples analyzed
are crusty bread with low amount of crumb, or the plunger is
larger than 36 mm, the crust may be removed from the bread
sample before testing to avoid any resistance during
compression.
2. The load cell is an important parameter to evaluate before
performing the TPA. In order to make the optimal selection,
it is crucial to determine the greatest force that may be reached
during the analysis. Indeed, the evaluation of textural proper-
ties of the crusty bread implies higher force, and a 50 kg load
cell (or 500 N load cell) might be the best option. However, if
softer products should be evaluated, load cells with lower
maximum force could also be used, which increases the accu-
racy of the test.
3. The load cell is an important parameter to evaluate before
performing the TPA. The evaluation of textural properties of
the bread slice usually implies lower force compared to bread
 Texture Profile Analysis 109

crust evaluation, and a 10 kg load cell (or 100 N load cell) is

usually the best option. See also Note 2.
4. The typical probe for bread evaluation is aluminum cylinder
with a diameter of 36 mm, which is the standardized one
developed by the American Association of Cereal Chemists
(AACC) for the measurement of bread crumb and other bakery
products. However, other cylindrical probes with a diameter of
30–40 mm could also be used if it is not possible to use the
standard one.
5. The bread crust has usually an irregular surface. Therefore, it is
important to choose the samples with regular surface. More-
over, place the plunger in the proximity of the highest part of
the crust, thus avoiding that the side parts of the plunger get in
contact with samples before the center.
6. The compression is usually set to 30% deformation for the
texture profile analysis of the bread loaf. The compression
could be changed according to the type of bread crust, which
mainly affects the results of this analysis. Moreover, the settings
of texture profile analyzer could differ based on the analyzer
and software. Usually, it is possible to previously define, by the
software, the percentage of compression, while in other cases,
the proximity limit to the plate is to be manually evaluated. In
the latter case, the height of the samples should be measured,
and the limit to the plate values should be calculated according
to the compression percentage.
7. Different speed could be also applied; however, the speed
should be between 1 and 2 mm/sec.
8. When the plunger goes back after the first compression, the
crust holds the plunger. In this case, try to separate them
manually before the beginning of the second compression.
9. Some software report the results as the force registered during
the analysis; other software report them related to a surface of
1 meter square (calculating the surface area of the plumber).
10. Total sample thickness is 25 mm; therefore, for evaluating
commercially available sliced bread, put two or three slices
stacked together. Always use the central slice of the loaf.
11. Check if the crust needs to be cut off. This additional operation
immediately before the compression test is also considered a
variable testing procedure. This is usually not necessary but,
based on different type of breads, the crust could resist the
compression.
12. A speed of 100 mm/min is used in the official method, which
corresponds approximately at 1.7 mm/sec. However, a speed
of 2 mm/sec could also be used.
110 Marco Montemurro and Erica Pontonio

References
1. Szczesniak AS (1963) Classification of textural 7. Kaszab T, Csima G, Lambert-Meretei A,
characteristics. J Food Sci 28(4):385–389 Fekete A (2002) Food texture profile analysis
2. Overview of Texture Profile Analysis. https:// by compression test. Corvinus University of
[Link]/resources/texture- Budapest, Faculty of food Science
profile-analysis#:~:text=Springiness%20is%20 8. Jekle M, Fuchs A, Becker T (2018) A normal-
now%20expressed%20as,by%20the%20original ized texture profile analysis approach to evalu-
%20compression%20distance. (Available ate firming kinetics of bread crumbs
online) independent from its initial texture. J Cereal
3. Szczesniak AS (2002) Texture is a sensory Sci 81:147–152
property. Food Qual Prefer 13(4):215–225 9. Tian YQ, Li Y, Jin ZY, Xu XM, Wang JP, Jiao
4. Method 74–09.01. Measurement of Bread AQ et al (2009) β-Cyclodextrin (β-CD): a new
Firmness by Universal Testing Machine, approach in bread staling. Thermochim Acta
AACC International Approved Methods 489(1–2):22–26
5. Chhanwal N, Anandharamakrishnan C (2014) 10. Nishinari K, Fang Y, Rosenthal A (2019)
Temperature- and moisture-based modeling Human oral processing and texture profile
for prediction of starch gelatinization and analysis parameters: bridging the gap between
crumb softness during bread-baking process. J the sensory evaluation and the instrumental
Texture Stud 45(6):462–476 measurements. J Texture Stud 50(5):369–380
6. Goesaert H, Slade L, Levine H, Delcour JA 11. Young LS (2012) Applications of texture anal-
(2009) Amylases and bread firming–an ysis to dough and bread. In: Cauvain S
integrated view. J Cereal Sci 50(3):345–352 (ed) Breadmaking. Woodhead Publishing, pp
562–579
 Chapter 11

Image Analysis
Michela Verni and Carlo Giuseppe Rizzello

Abstract
Bread structure is a determinant of loaf volume and resilience, as well as the sensation perceived during
eating, and its analysis comprises the study of the fluid phase (air), also named as void, cells or pores, and the
solid phase of bread crumb (cell wall material).
Acquisition of two-dimensional images by flatbed scanning is the most commonly employed method to
perform image analysis of bread as it is fast, easy to use, economical, robust, independent of the external
light conditions, and accurate. To enhance the contrast between the two phases, once acquired, the image is
subjected to cell segmentation, a process that separates or classifies objects of interest from its background,
typically yielding a binary image, set at a specific threshold. Thresholding is a critical step in ensuring a
successful partition of crumb cells from the background and assumes that the object and background pixels
can be distinguished by selection of an optimal gray level value. Hundreds of segmentation techniques are
described in the literature, yet the Otsu method, based on an algorithm that minimizes the intraclass
variance of the segmented region, or manual thresholding, is those most reported.
Once the binary image is obtained, the software processes it, and crumb grain properties such as number
of cells, number of cells/cm2, mean cell area, cell-total area ratio, and cell wall thickness can be calculated.

Key words Bread, Crumb structure, Crumb cells, resolution, Black/white pixels, Thresholding

1 Introduction

The analysis of bread structure has been traditionally performed

based on the experience of the baker who evaluates parameters such
as size, shape, uniformity, and wall thickness of crumb bubbles, all
of them representing, at macroscopic level, the fluid phase (air),
also named as void, cells or pores, and the solid phase of bread
crumb (cell wall material) [1, 2].
The relationship between crumb structure and its appearance
may be self-evident, yet crumb structure is a determinant of loaf
volume and resilience and of the sensation perceived during eating
and even its taste [3]. Therefore, studying the structure that defines
crumb appearance allows us to predict quality attributes of bread
from the knowledge of the ingredients and how they are processed

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

111
112 Michela Verni and Carlo Giuseppe Rizzello

into the cellular structure on which bread quality is established

[3]. For example, optimal gluten development during kneading is
vital to the development of crumb structure, and the density of the
resulting structure is markedly affected by insufficient disaggre-
gation of the native gluten polymers [4]. Similarly, proofing time
highly affects crumb structure; indeed, the purpose of the final
proof is to allow the dough after panning to regain an extensible
character and a high level of aeration to yield a loaf with a desired
volume and crumb grain [4].
Many analytical techniques are studied to describe the structure
of baked products: light microscopy, electron microscopy, scanning
probe microscopy, atomic force microscopy, magnetic resonance
imaging, dynamic magnetic resonance microscopy, confocal laser
scanning microscopy, X-ray computerized microtomography, ultra-
sonic reflectance spectroscopy, low frequency acoustic spectrome-
ter, and flatbed scanning [5]. Nevertheless, the acquisition of
two-dimensional images by flatbed scanning is the most commonly
employed method to perform image analysis of bread as it is fast,
easy to use, economic, robust, independent of the external light
conditions, and accurate [5].
To acquire a digital image of the cut surface of the bread
crumb, three elements are necessary: source of illumination, sam-
ple, and image sensing device [3]. Although few studies have
reported on the utilization of digital cameras to acquire bread
image [4, 6], most of them have used flatbed scanners [7–
12]. The use of digital scanners to capture bread crumb high-
resolution images coupled with image analysis software has been
used to quantify crumb features such as cell size, cell size distribu-
tion, number of cells per unit area, and cell wall thickness [2]. How-
ever, a standardized technique to evaluate differences among the
image analysis methodologies is yet to be found. Some of the
differences among the reported methodologies are about the scan-
ning resolution; it is possible to find reports using 150, 200,
300, 350, 550, 600 dpi or, in some cases, the resolution is not
even reported [1].
When examining bread texture, particularly if one is trying to
obtain quantitative information on its structural organization, the
goal is to enhance the contrast between the two phases in the
image, which is why, once acquired, the image is processed in
order to extract information on crumb grain features (Fig. 1). To
achieve this, images are usually subjected to cell segmentation, a
process that separates or classifies objects of interest within an
image from its background, typically yielding a binary image
[3]. Thresholding is a critical step in ensuring a successful partition
of crumb cells from the background, and consequent computation
of crumb grain features, and assumes that the object and back-
ground pixels can be distinguished by selecting an optimal gray
level value [13]. Hundreds of segmentation techniques are
 Image Analysis of Bread 113

Fig. 1 Representation of the crumb structure of wheat bread (left) and sourdough bread (right) converted from
true color (a) to grayscale level (b) and subjected to cell segmentation using the UTHSCSA Image Tool program
2.0 (c) where the white part is the fluid phase and the black part is the solid phase
114 Michela Verni and Carlo Giuseppe Rizzello

described in the literature [14], and few of them were also com-
pared in a study of bread image analysis [13]; however, the Otsu
method or manual thresholding are the most reported. While the
Otsu’s algorithm searches for the threshold that minimizes the
intraclass variance of the segmented region, the manual one locates
the threshold by finding each of the peaks of the frequencies
histogram and then the valleys between them or, in other words,
the gray level threshold is subjectively varied until the binary image
retains the essential features of the monochrome counterpart
[1, 13, 15]. The Otsu method provides a more reliable representa-
tion of bread crumb structure, whereas the manual method can be
unreliable for the quantification of cell size uniformity and particu-
larly for void fraction [1, 16]. Hence, the threshold selection can
strongly affect bread crumb features values and must be carefully
chosen.
The literature describes that it is not the voids themselves,
rather the space they occupy and the extent to which they influence
the surrounding matrix that define the three-dimensional structural
characteristics of bread. The surrounding matrix, or cell wall struc-
ture, is therefore a significant factor in the mechanical strength and
structural properties of the baked product [11].
Once the binary image is obtained, the software processes it,
and crumb grain properties can be calculated and extracted. Num-
ber of cells, number of cells/cm2, mean cell area, cell-total area
ratio, and cell wall thickness are among the most common para-
meters obtained with image analysis of bread crumb. The number
of cells/cm2 represents the density of cells and is independent of
the magnitude of the cross-sectional area of the field of view; higher
values indicate finer crumb structure [15]. The cell-total area ratio,
instead, corresponds to the crumb section counted as cells, and it is
calculated as the sum of all cell areas divided by the total area
surveyed. Whereas cell wall thickness is calculated as the average
value between the boundaries of a given cell and the boundaries of
cells detected at regular intervals in its neighborhood [15].
Depending on the level of incorporation as well as its intrinsic
properties, the addition of sourdough to bread affects both texture
and crumb grain characteristics of the baked product, and the
nature and extent of these effects change accordingly. Contradic-
tory results about the role of sourdough in improving loaf bread
volume have been reported. While some studies have shown that
the use of sourdough increases bread volume [12, 17, 18], others
claimed crumb of sourdough bread is often open and non-uniform,
an aspect which gives the bread a natural rustic appearance and
smaller volume [10, 19, 20]. Fois et al. [7], instead, observed that
the addition of 30% sourdough, prepared using Saccharomyces cer-
evisiae, Fructilactobacillus sanfranciscensis, and Lactiplantibacillus
plantarum as starters, led to a distribution of the small-to-medium
gas cells similar for sourdough and baker’s yeast leavened bread.
 Image Analysis of Bread 115

Nevertheless, bread volume was higher in baker’s yeast leavened

bread, except in the case of samples with very strong and elastic
gluten network [7]. Similar considerations concluded the work of
Alfonzo et al. [9] who evaluated through image analysis the tech-
nological parameters of several semolina flours, from old and mod-
ern genotypes cultivated in Southern Italy. The flours were then
used to produce sourdough breads which showed highly different
results in terms of void fraction, cell density, and mean cell area
[9]. On the other hand, one aspect the literature agrees on is the
fact that the addition of pseudo-cereals or legume flours, as well as
milling by-products to bread, worsens crumb structure due to
lower or absence of gluten and higher fiber content, and conse-
quent weaker gluten network, which results in a poorer ability to
retain gases. Sourdough fermentation of those flours improved
crumb porosity leading to, although more uneven than baker’s
yeast leavened control bread, higher gas cell total area than the
corresponding bread added of unfermented flours [10, 20–22];
the same could be said for gluten-free breads [8, 23].
In this framework, this chapter will describe the method for
image analysis of bread as reported by most studies, evaluating the
effect of sourdough on the overall quality of bread. However, a
more in-depth schematization of the most common software
employed, as well as the parameters used in bread image analysis,
is reported in Table 1.

Table 1
Image software and processing parameters often used for sourdough bread image analysis

Image software Picture parameters References

Image-Pro Plus 4.5 (Media Cybernetics Inc., USA) 600 dpi, converted from true color [8, 12, 24,
to 256 level grayscale 25]
UTHSCSA ImageTool program 2.0 (University of 300 dpi, analyzed in grayscale (0 to [10, 20–
Texas Health Science Centre, San Antonio, Texas 255, 8 bit) 22, 26,
Sub-image: 500 × 500 dpi 27]
Manual threshold
SigmaScan Pro software (Jandel Corporation, USA) 150 dpi, converted from true color [2, 7, 11,
to 256 level grayscale 28, 29]
C-cell Bread Imaging system (Calibre Control Analyzed in grayscale (0–255) [23, 30]
International Ltd., UK)
ImageJ software (National Institutes Health, 150–350 dpi, converted from true [9, 31, 32]
Bethesda, Md, USA) color to 256 level grayscale (8 bit)
Sub-image: 207 × 207 dpi.
Otsu’s threshold algorithm
116 Michela Verni and Carlo Giuseppe Rizzello

2 Materials

1. Loaf of bread.
2. Flatbed scanner.
3. Image analysis software, e.g., UTHSCSA ImageTool
program 3.0.

3 Methods

1. Cut a loaf of bread in slices (see Note 1).

2. Scan the surface of the slice using a using a flatbed image
scanner with a high resolution (see Note 2).
3. Obtain a sub-image of the overall slice image so that all images
have the same dimension, e.g., 500 × 500 pixel (see Note 3).
4. Acquire 8-bit gray-level images, saving them in TIFF format
without any image compression.
5. Convert the grayscale image into a binary image (black and
white pixel) setting a threshold either manually or using one of
the methods reported in the literature.
6. Analyze the image using the software and extrapolate the para-
meters of interest.

4 Notes

1. When cutting bread loaf, make sure the slice surface is even, to
avoid interferences of shadows in the scanned image. Usually,
slices width ranges from 1 to 2 cm, with 1.5 cm being the most
common.
2. During the experiment, scanner brightness and contrast para-
meters should be kept constant for each sample analyzed and in
the default value (zero). Most of the published papers use a
resolution of 300 dpi.
3. The binarization of sub-images is more suitable for detecting
bubbles of samples more precisely than the binarization of a
whole image, because background inequality caused by uneven
surface of the samples can occur.
 Image Analysis of Bread 117

References
1. Farrera-Rebollo RR, de la Paz S-CM, Cha- 12. Chiavaro E, Vittadini E, Musci M, Bianchi F,
nona-Pérez J, Gutiérrez-López GF, Alamilla- Curti E (2008) Shelf-life stability of artisanally
Beltrán L, Calderón-Domı́nguez G (2012) and industrially produced durum wheat sour-
Evaluation of image analysis tools for charac- dough bread (“Altamura bread”). Food Sci
terization of sweet bread crumb structure. Technol 41:58–70
Food Bioproc Technol 5:474–484 13. Gonzales-Barron U, Butler F (2006) A com-
2. Pourfarzad A, Mohebbi M, Mazaheri-Tehrani parison of seven thresholding techniques with
M (2012) Interrelationship between image, the k-means clustering algorithm for measure-
dough and Barbari bread characteristics; use ment of bread-crumb features by digital image
of image analysis to predict rheology, quality analysis. J Food Eng 74:268–278
and shelf life. Int J Food Sci Technol 47:1354– 14. Fu KS, Mui JK (1981) A survey on image
1360 segmentation. Pattern Recogn 13:3–16
3. Scanlon MG, Zghal MC (2001) Bread proper- 15. Sapirstein HD, Roller R, Bushuk W (1994)
ties and crumb structure. Food Res Int 34: Instrumental measurement of bread crumb
841–864 grain by digital image analysis. Cereal Chem
4. Zghal MC, Scanlon MG, Sapirstein HD 71:383–391
(1999) Prediction of bread crumb density by 16. Scheuer PM, Ferreira JAS, Mattioni B,
digital image analysis. Cereal Chem 76:734– Miranda MZD, Francisco AD (2015) Optimi-
742 zation of image analysis techniques for quality
5. Esteller MS, Zancanaro O, Palmeira CNS, da assessment of whole-wheat breads made with
Silva Lannes SC (2006) The effect of kefir fat replacer. Food Sci Technol 35:133–142
addition on microstructure parameters and 17. Clarke CI, Schober TJ, Arendt EK (2002)
physical properties of porous white bread. Eur Effect of single strain and traditional mixed
Food Res Technol 222:26–31 strain starter cultures on rheological properties
6. Rouillé J, Valle GD, Devaux MF, Marion D, of wheat dough and on bread quality. Cereal
Dubreil L (2005) French bread loaf volume Chem 79:640–647
variations and digital image analysis of crumb 18. Corsetti A, Gobbetti M, Balestrieri F,
grain changes induced by the minor compo- Paoletti F, Russi L, Rossi J (1998) Sourdough
nents of wheat flour. Cereal Chem 82:20–27 lactic acid bacteria effects on bread firmness
7. Fois S, Fadda C, Tonelli R, Sanna M, Urgeghe and stalin. J Food Sci 63:347–351
PP, Roggio T, Catzeddu P (2012) Effects of 19. Kulp K (2003) Baker’s yeast and sourdough
the fermentation process on gas-cell size technologies in the production of U.S. bread
two-dimensional distribution and rheological product. In: Kulp K, Lorenz K (eds) Hand-
characteristics of durum-wheat-based doughs. book of dough fermentations. Marcel Dekker,
Food Res Int 49:193–200 New York
8. Rinaldi M, Paciulli M, Caligiani A, Scazzina F, 20. Rizzello CG, Nionelli L, Coda R, Di Cagno R,
Chiavaro E (2017) Sourdough fermentation Gobbetti M (2010) Use of sourdough fermen-
and chestnut flour in gluten-free bread: a ted wheat germ for enhancing the nutritional,
shelf-life evaluation. Food Chem 224:144–152 texture and sensory characteristics of the white
9. Alfonzo A, Urso V, Corona O, Francesca N, bread. Eur Food Res Technol 230:645–654
Amato G, Settanni L, Di Miceli G (2016) 21. Rizzello CG, Calasso M, Campanella D, De
Development of a method for the direct fer- Angelis M, Gobbetti M (2014) Use of sour-
mentation of semolina by selected sourdough dough fermentation and mixture of wheat,
lactic acid bacteria. Int J Food Microbiol 239: chickpea, lentil and bean flours for enhancing
65–78 the nutritional, texture and sensory character-
10. Coda R, Varis J, Verni M, Rizzello CG, Katina istics of white bread. Int J Food Microbiol 180:
K (2017) Improvement of the protein quality 78–87
of wheat bread through faba bean sourdough 22. Coda R, Rizzello CG, Gobbetti M (2010) Use
addition. Food Sci Technol 82:296–302 of sourdough fermentation and pseudo-cereals
11. Crowley P, Schober TJ, Clarke CI, Arendt EK and leguminous flours for the making of a
(2002) The effect of storage time on textural functional bread enriched of γ-aminobutyric
and crumb grain characteristics of sourdough acid (GABA). Int J Food Microbiol 137:236–
wheat bread. Eur Food Res Technol 214:489– 245
496
118 Michela Verni and Carlo Giuseppe Rizzello

23. Axel C, Röcker B, Brosnan B, Zannini E, 28. Parenti A, Guerrini L, Granchi L, Venturi M,
Furey A, Coffey A, Arendt EK (2015) Applica- Benedettelli S, Nistri F (2013) Control of mix-
tion of Lactobacillus amylovorus DSM19280 in ing step in the bread production with weak
gluten-free sourdough bread to improve the wheat flour and sourdough. J Agric Eng 44.
microbial shelf life. Food Microbiol 47:36–44 [Link]
24. Hayta MEHMET, Ertop MH (2018) Evalua- 29. Scheuer PM, Southgate ANN, Martelli MF,
tion of microtextural properties of sourdough Dias C, da Silva ME, Coelho AA, de Francisco
wheat bread obtained from optimized formu- A (2020) Quality properties of a bread made
lation using scanning electron microscopy and with levain and cocoa waste. J Culin Sci Tech-
image analysis during shelf life. J Food Sci nol:20:409-420
Technol 55:1–9 30. Wolter A, Hager AS, Zannini E, Czerny M,
25. Rinaldi M, Paciulli M, Caligiani A, Sgarbi E, Arendt EK (2014) Impact of sourdough fer-
Cirlini M, Dall’Asta C, Chiavaro E (2015) mented with Lactobacillus plantarum FST 1.7
Durum and soft wheat flours in sourdough on baking and sensory properties of gluten-free
and straight-dough bread-making. J Food Sci breads. Eur Food Res Technol 239:1–12
Technol 52:6254–6265 31. Abedfar A, Sadeghi A (2019) Response surface
26. Coda R, Di Cagno R, Rizzello CG, Nionelli L, methodology for investigating the effects of
Edema MO, Gobbetti M (2011) Utilization of sourdough fermentation conditions on
African grains for sourdough bread making. J Iranian cup bread properties. Heliyon 5:
Food Sci 76:M329–M335 e02608
27. Rizzello CG, Curiel JA, Nionelli L, 32. Xi J, Xu D, Wu F, Jin Z, Xu X (2020) Effect of
Vincentini O, Di Cagno R, Silano M, Na2CO3 on quality and volatile compounds of
Gobbetti M, Coda R (2014) Use of fungal steamed bread fermented with yeast or sour-
proteases and selected sourdough lactic acid dough. Food Chem 324:126786
bacteria for making wheat bread with an inter-
mediate content of gluten. Food Microbiol 37:
59–68
 Chapter 12

Descriptive Sensory Analyses

Hana Ameur and Marco Gobbetti

Abstract
Sourdough technology is widespread throughout the world and is considered as the most suitable and
sustainable process to enhance the nutritional, functional, rheology, and sensory attributes of bread in its
many varieties. To obtain a specific, reproducible sourdough and bread quality, a specific knowledge on the
process parameters and on the most appropriate methods and protocols to characterize the sourdough is
required. Descriptive sensory analyses are a powerful tool to describe the sensory properties of sourdough
baked goods. However, it is necessary to establish a complete methodology for the evaluation. Here, we
describe a protocol that is valid for sensory analysis of sourdough bread and closely related products. This
protocol follows the general guidance for the application of sensory analysis in quality control developed by
ISO (the International Organization for Standardization), and it is adapted from the standard methodology
for the sensory analysis of bread developed by the Spanish Center of Baking Technology. The protocol
outlined included 27 sensory attributes, a trained panel, and the optimal procedure for the evaluation.

Key words Sourdough, Flavor, Texture, Aroma, Sensory profile, Bread quality, Panel test

1 Introduction

Nowadays, consumers are increasingly demanding in terms of over-

all quality, freshness, and nutritional and sensory properties of
baked goods. In fact, sensory perception plays an important role
in consumer preference and purchase of bread [1]. In this context,
the modern biotechnology of leavened baked goods largely uses
sourdough as natural leavening agent because of the many advan-
tages it offers over baker’s yeast, e.g., in the development of the
characteristic flavor of bread, resulting in a product with high
sensory quality [2]. A recent systematic survey [3] showed that
sourdough is an irreplaceable starter for improving the sensory,
rheology, and shelf-life attributes of baked goods because of its
unique and rich microbial composition. Fermentation is the
major route for volatile formation in sourdough and bread
crumb. It produces mainly organic acids, alcohols, aldehydes,
esters, and ketones [4]. The degradation of the cereal proteins in

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

119
120 Hana Ameur and Marco Gobbetti

wheat and rye sourdough fermentations strongly affects the quality

of bread. Acidification and the reduction of disulfide bonds of
gluten by heterofermentative lactobacilli increase the activity of
cereal proteases and substrate accessibility [5]. Bacterial proteolysis
during sourdough fermentation was shown to contribute much
more to the development of typical sourdough flavors of baked
breads when compared to breads produced from chemically acid-
ified or yeasted doughs [2]. Moreover, sourdough confers a unique
and superior flavor and taste, especially because of the liberation of
free amino acids (FAA) during fermentation, which act as precur-
sors of volatile compounds (VOC) or directly affect the flavor
intensity [3]. The sensory qualities that are most described regard-
ing sourdough breads include their taste, aroma, texture, and visual
characteristics, which are key drivers in the liking and perception of
food products [6]. Sensory evaluation is a powerful tool with a wide
range of applications in the bakery industry. Sensory quality of
bread can be evaluated by affective (subjective) methods, such as
consumers’ acceptance test, and by analytical (objective) methods.
Regarding the use of analytical methods, a trained panel of judges is
required to evaluate the products and describe them qualitatively
and quantitatively in terms of the sensory attributes. This creates a
sensory profile of the product [7]. Thus, a standard sensory meth-
odology is an interesting tool for the industry in the development
of sourdough bakery goods to achieve not only the best technolog-
ical quality but also to consistently meet consumers’ expectations.
Furthermore, the use of a standard sensory methodology facilitates
communication among research and producer groups. A wide vari-
ety of descriptive analysis techniques are available, ranging from the
long-established methods such as the flavor profile and the texture
profile methods, Spectrum™ Method, quantitative descriptive
analysis, methods that enable a faster output, for example free
choice profiling and flash profiling, to those methods that are
designed to measure dynamic changes in attribute intensity such
as time intensity, progressive profiling, and temporal dominance of
sensations. Although the exact approach differs depending on
method, there are generic steps that are relevant, to a greater or
lesser extent, to all [8]. Here we describe a protocol that is valid for
sensory analysis of sourdough bread and closely related products.
This protocol follows the general guidance for the application of
sensory analysis in quality control developed by ISO (the Interna-
tional Organization for Standardization), and it is adapted from the
standard methodology for the sensory analysis of bread developed
by the Spanish Center of Baking Technology [1]. This a relatively
rapid descriptive analysis method, requiring only about 4 weeks
from start to finish to recruit, screen, train, and evaluate. Moreover,
there is no need to maintain a vast standardized reference library or
to calibrate panelists with subjective numerical values. Finally, this
 Sensory Analyses 121

method can provide extensive data analysis. The method involves

formation and training of the panel, generation of descriptors and
their definitions, sample preparation, distribution, and evaluation.

2 Materials

Control the test conditions. Prepare and present all bread samples
in a standardized and consistent manner to ensure that no sources
of bias are introduced which could have an adverse effect on the end
test result [9].

2.1 Formation and 1. Conduct a preliminary screening in your research group to

Training of the Panel select members with sensory acuity and previous experience
in sourdough technology and baked goods.
2. Select 10–12 panelists ideally (male and female, mean age:
35 years) for the basic training sessions (see Note 1).
3. Conduct standardized tests to identify basic tastes, olfactory
substance recognition, and taste ranking and for describing
texture [10].

2.2 Generation of 1. Define the vocabulary before the test.

Descriptors and Their 2. Comparing sourdough vs. baker’s yeast breads and merging
Definitions research articles that used the same descriptive approach, we
found 27 sensory attributes that made possible the differentia-
tion [3]. The list of sensory attributes selected for profiling and
definitions for each attribute is given in Table 1.
3. All sensory panels must use the same set of attributes (see
Note 2).

2.3 Sample 1. Prepare sourdough bread samples according to your own

Preparation experimental design.
2. Prepare a control bread fermented with baker’s yeast and with-
out the addition of sourdough.
3. After baking, cool bread loaves for 1 h at room temperature and
then pack it into polyethylene bags.
4. Store sample loaves at -18 °C within 2 h of baking until
required for sensory evaluation.
5. Before the sensory evaluation, remove bread loaves from frozen
storage and thaw them at room for 5–6 h (see note 3).
6. Cut each bread loaf using a bread slicer or a sharp knife into
slices of 1.5 cm thickness.
7. Cut each slice into four pieces with each piece including both
crust and crumb to evaluate the overall sensory characteristics.
122 Hana Ameur and Marco Gobbetti

Table 1
List of the 27 sensory attributes selected for profiling and definitions for each attribute

Descriptors Definitions

Acidic taste and smell The sharp taste and smell like that elicited by vinegar

Adhesiveness The degree to which the product adheres to the palate after the compression between the
tongue and palate

Attractiveness The quality of being pleasing or appealing to the senses

Bitterness The basic taste factor produced by caffeine

Buttery The aromatics commonly associated with natural, fresh, unsalted butter

Cereal flavor The typical flavor of grains, for example, wheat, maize, or rye

Chewiness The number of chews required to achieve a consistency suitable for swallowing

Crust crispness The low-pitch sound produced on crust fracture during mastication

Dryness The sensation of dryness due to lack of saliva and absence of water

Elasticity The rapidity and degree of recovery of the crumb after the deforming force, exerted by
fingers, is removed

Flavor intensity The measurement of the strength of the flavor

Freshness The state of being recently made or obtained or not having decayed

Fruitiness The aromatic impression of dark fruit that may include sweet, slightly brown, overripe, or
somewhat sour

Greasiness The appearance of bread, which is covered with oil

Hardness The force required to achieve a deformation, penetration of the product by the molars

Descriptors Definitions

Hay-like The aroma associated with the freshly cut grass, legumes, or other herbaceous plants

Intense aftertaste and aroma The flavors and aroma sensations that linger after bread is no longer in the mouth

Nutty The aromatic characteristics of mixed nuts, e.g., walnuts, hazelnuts, and pine nuts

Overall taste The overall gustatory perceptions

Porosity The air pockets in the leavened bread

Pronounced crumb and crust The degree of perceived color characterizing the crumb and the crust
color

Roasted The typical aroma of roasted bread

Salty The umami taste–elicited by glutamic acid or NaCl solution

Sourness The sour taste, like that elicited by citrus fruits

Stale The loss of freshness and palatability

Sweetness A basic taste factor produced by sugars

Yeast-like The sour umami taste, reminiscent of the presence of inactive yeast
 Sensory Analyses 123

3 Methods

Carry out the descriptive panel in a standard tasting room [11]

equipped with individual taste booths separated by screens high
and wide enough to isolate the different panels.

3.1 Sample 1. Serve the samples at room temperature (about 20 °C) on black
Distribution dishes covered with plastic wrap and labeled with randomized
three-digit codes.
2. Each panelist should receive two pieces from the same sliced
sample to do the evaluation in two replicates.

3.2 Sample Intensity of the sensory attributes is rated on a scale of 10 cm in

Evaluation length and verbally anchored at each end, the left side of the scale
corresponding to the lowest intensity (value 0) and the right side to
the highest intensity (value 10) of the attribute [1].
1. The panel is instructed to sniff each sample prior to tasting it,
and they are requested to swallow the samples. In details, the
panel starts the analysis with the olfactory phase. They should
open the packet containing the sample, bring the sample close
to the nose, and smell it with long and strong inhalations at the
same time as squeezing it gently twice to release the aromatic
compounds. After the olfactory phase, the following phases are
carried out in the order: appearance phase, texture phase
(by hand and in mouth), and flavor phase [9].
2. Provide each panelist with filtered water and unsalted crackers
to cleanse their palate between tastings (see Note 4).
3. Collect and record the data using a computerized data system.
4. Calculate the final scores for each attribute as the means of the
data collected in three independent evaluations.
5. Use Friedman’s nonparametric test for the statistical treatment
of the results.
An example of the evaluation is given in the Spider web chart of
the sensory analysis data of sourdough vs. baker’s yeast breads as
assessed by trained panelists (Fig. 1).

4 Notes

1. Descriptive panels, in which the panelists have received exten-

sive training and performance, may consist of five or more
panelists. It is wise to have more than five people on a trained
panel, however, because if one panelist misses an evaluation,
the influence on the data can be extreme. Ten to twelve trained
panelists on a trained panel is more satisfactory to provide more
statistical confidence in the data.
124 Hana Ameur and Marco Gobbetti

Baker's yeast bread Sourdough bread

Acidic taste and smell

Yeast-like 10 Adhesiveness
Sweetness 9 Attractivness
Stale 8 Bitterness
7
Sourness 6 Buttery
5
Salty 4 Cereal flavor
3
Roasted 2 Chewiness
1
Pronounced crumb and crust 0
Crust crispness
color

Porosity Dryness

Overall taste Elasticity

Nutty Flavor intensity

Intense aftertaste and aroma Freshness

Hay-like Fruitiness
Hardness Greasiness

Fig. 1 Representation of a Spider web chart of the sensory analysis data of sourdough vs. baker’s yeast breads
as assessed by trained panelists

2. Give each member of the panel the list of attributes including

definitions to aid them during sample evaluation.
3. Testing time periods should be scheduled when the panelists
will be most receptive to product characteristics and have ade-
quate attention to the task at hand. Do not schedule the session
immediately after a meal or coffee break to avoid sensory error.
4. To reduce the likelihood of carryover, a 2 min interval is
allowed between each sample assessment.

References
1. Elı́a M (2011) A procedure for sensory evalua- Technol 108:71–83. [Link]
tion of bread: protocol developed by a trained 1016/[Link].2020.12.008
panel. J Sens Stud 26:269–277. [Link] 4. Limbad M, Gutierrez Maddox N, Hamid N,
org/10.1111/j.1745-459X.2011.00342.x Kantono K (2020) Sensory and physicochemi-
2. Corsetti A, Settanni L (2007) Lactobacilli in cal characterization of sourdough bread
sourdough fermentation. Food Res Int 40: prepared with a coconut water kefir starter.
539–558. [Link] Foods 9:1165. [Link]
foodres.2006.11.001 foods9091165
3. Arora K, Ameur H, Polo A et al (2021) Thirty 5. G€anzle MG, Loponen J, Gobbetti M (2008)
years of knowledge on sourdough fermenta- Proteolysis in sourdough fermentations:
tion: a systematic review. Trends Food Sci mechanisms and potential for improved bread
 Sensory Analyses 125

quality. Trends Food Sci Technol 19:513–521. 9. Callejo MJ, Vargas-Kostiuk M-E, Rodrı́guez-
[Link] Quijano M (2015) Selection, training, and val-
6. Calvert MD, Madden AA, Nichols LM et al idation process of a sensory panel for bread
(2021) A review of sourdough starters: ecol- analysis: influence of cultivar on the quality of
ogy, practices, and sensory quality with applica- breads made from common wheat and spelt
tions for baking and recommendations for wheat. J Cereal Sci 61:55–62. [Link]
future research. PeerJ 9:e11389. [Link] org/10.1016/[Link].2014.09.008
org/10.7717/peerj.11389 10. ISO (International Organization for Standardi-
7. Callejo MJ (2011) Present situation on the zation) 8586-1:1993. Sensory analyses. Gen-
descriptive sensory analysis of bread. J Sens eral guidance for the selection, training, and
Stud 26:255–268. [Link] monitoring of assessors. Part 1: Selected
j.1745-459X.2011.00341.x assessors
8. Muñoz AM, Kemp SE, Hollowood T, Hort J 11. ISO (International Organization for Standardi-
(2018) Comparison of descriptive analysis zation) 8589:1998. Sensory analyses. General
methods. In: Kemp SE, Hort J, Hollowood T guidance for the design of test rooms
(eds) Descriptive analysis in sensory evaluation.
Wiley, Chichester, pp 679–709
 Chapter 13

Determination of the Volatile Components

Giuseppe Celano and Maria De Angelis

Abstract
The metabolic activity of sourdough microbial community is exploited not only to leaven dough but also to
produce a complex array of volatile and nonvolatile flavor compounds that influence the sensory properties
of the final bread product. The sourdough and the processing parameters applied in the production of
sourdough bread affect both the diversity of microorganisms and their metabolic activity and flavor
generation during fermentation. Volatile organic compounds (VOCs) are important contributors to the
characteristic flavors and off-flavors of baked goods, affecting the overall acceptability to consumers. A wide
range of volatile compounds have been identified in sourdough bread, including acids, alcohols, aldehydes,
esters, ketones, lactones, hydrocarbons, pyrroles, pyrazines, and sulfur compounds. The gas chromatogra-
phy — mass spectrometry (GC—MS) analysis have been described for VOC detection, by using a method
for volatile extraction based on headspace solid phace microextraction (HS—SPME). This chapter will
provide a description of HS-SPME GC-MS analysis for the evaluation of volatile compounds in sourdough
bread including sample treatment, HS-SPME conditions, GC-MS conditions, as well as quantification
methods.

Key words Volatile organic compounds, Sourdough breads, GC-MS, HS-SPME

1 Introduction

Volatile organic compounds (VOCs) are important contributors to

the characteristic flavors and off-flavors of baked goods, affecting
the overall acceptability to consumers. The volatile compounds are
extensively studied in bakery products and more than 540 volatiles
have been reported [1]. Among these, only a small number of them
have a real impact on the final sensory profile [2, 3].
The most important chemical classes of VOCs include alde-
hydes, alcohol, ketones, esters, acids, and pyrazines. Furans, hydro-
carbons, and lactones are also present but contribute to a lesser
extent to the sensory profile. These volatile compounds may origi-
nate from the crumb, crust, or both. In the crumb, the volatile
fraction is generated by enzymatic reactions during dough knead-
ing [4] and, principally, during the long fermentation required for

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

127
128 Giuseppe Celano and Maria De Angelis

sourdough production [5, 6]. The aromatic profile of sourdough

bread is characterized by the high content of organic acids,
in particular lactic and acetic acids, deriving from the lactic acid
fermentation of lactic acid bacteria (LAB) and associated metabo-
lisms [7]. Moreover, 2-methylpropanol, 2-methylbutanol, 3-
methylbutanol, ethyl acetate, and ethyl lactate detection in bread
is directly linked to their presence in sourdough [8, 9]. Aldehydes,
ketones, and alcohols may also be engendered from lipid oxidation.
In the crust, the volatile fraction is formed by thermal reactions
occurring during baking, such as the non-enzymatic Maillard reac-
tions (including Strecker degradation of carbonyl compounds) and
caramelization of sugars [10]. Moreover, there could be odorant
transfers during crust formation from the crumb to the crust and
vice versa [11].
Collectively, VOCs act in a synergistic way that serves as stimuli
for bread flavor. Although it is often difficult to establish a direct
link between a given compound and a single-generation pathway
[12], to understand the real impact of sourdough on bread aroma is
of utmost importance to assess the qualitative and quantitative
profile of VOCs arising from fermentation processes linked to lactic
acid bacteria (LAB) and yeasts metabolic activities.
The evaluation of the volatile compound profile requires the
preparation of sample, extraction, identification, and quantification
of the compounds. For the preparation of sample, the crumb and
crust must be carefully separated from each other. The crumb
and/or crust will then be frozen with liquid nitrogen for the
purpose of terminating further biochemical evolution as well as
trapping the volatile compounds of interest [4, 13]. The frozen
sample should be then grounded to powder by a mortar with a
pestle or with a blender. The obtained powder can be spiked with
an internal standard (IS) prior to extraction, in order to compensate
for changes in extraction efficiency, detector response due to sample
loss during other sample preparation steps, or changes in sample
matrix composition.
Depending on the study aims, the extraction methods are
based on solvent extraction, supercritical fluid extraction (SFE),
and headspace extraction. The main extraction methods have
been recently reviewed [10, 14], and the relative advantages and
some limitations are reported in Table 1. Among these, headspace
solid-phase micro-extraction (HS-SPME) has been widely used for
sourdough and bread analysis and seems to be the most convenient
method to develop a standardized extraction method. HS-SPME is
a simple and inexpensive method, and the dependent parameters
(i.e., fiber coating, sample preparation, equilibrium time, equilib-
rium temperature, extraction time, extraction temperature) have
been optimized for bread aroma profiling [15–17].
The extracted volatile compounds are characterized by a gas
chromatography (GC) analysis coupled with mass spectrometrys
 Volatile Organic Compounds in Sourdough Bread 129

Table 1
Limitations and advantages of extraction methods to evaluate VOCs in bread

Extraction technique Limitation Advantages

Headspace Static headspace Strongly depend on temperature, Provide accurate information
methods (SHS) low sensitivity, rarely employed in of composition perceived by
bread aroma analysis, collect only odor, establish direct
the most volatile and major measurement of ratio of each
compounds volatile compound in
gaseous phase
Dynamic Composition of volatile fraction is Most utilized for bread aroma
headspace different from that in the air above analysis, better sensitivity
extraction the food, difficult to know the than SHS, collect lots of
(DHE) exact concentration of each volatile volatile compounds
compound that passes to the
gaseous phase, could be affected
by bread matrix
Multiple Possible error due to a high binding Useful in approaching the real
headspace of analyte to matrix concentration, not affected
extraction by bread matrix
(MHE)
Headspace-solid- Limitations for minor volatile Most preferred option for
phase compounds, need careful method bread aroma analysis;
microextraction development for good solvent-free, rapid sampling,
(HS-SPME) reproducibility, selectivity, and low cost, high sensitivity,
sensitivity and selectivity; improved
precision when combined
with solvent extraction
results
Solvent Soxhlet extraction Slow process, high solvent volume Can extract nonvolatiles with
extraction needed, may need a higher boiling points, and
preconcentration step before tightly bounded and
encapsulated volatiles
Distillation or For preventing interference in GC
sublimation separation and dirty GC column is
required the isolation of fractions
Simultaneous Elevated temperature required: No need for prior solvent
steam artifact formation, form nonnative extraction, continuous
distillation aroma compounds due to the supply of fresh solvent, fast
extraction presence of sugars and free amino and simple, good
(SDE) acids repeatability and recovery
Solvent-assisted No artifact formation,
flavor versatile, direct isolation of
evaporation volatiles from complex food
(SAFE) matrices
Supercritical fluid extraction Possible low recovery of volatile Environmentally friendly
(SFE) compounds mobile phase, shorter
extraction time, good
alternative to solvent
extraction for lipid analysis,
reduce coextracted materials

(continued)
130 Giuseppe Celano and Maria De Angelis

Table 1
(continued)

Extraction technique Limitation Advantages

Electronic nose Low sensitivity. Use limited to Low cost, rapid, noninvasive;
quality control due to low able to distinguish key bread
sensitivity aromas during baking; good
prediction for bread
staleness and suitability for
consumption immediately
after baking and during
storage
Artificial mouth Only analyze retronasal compounds Efficient extraction method for
but not orthonasal volatiles volatiles participate in
retronasal perception of
bread; representative of
bread aroma
Adapted from Dong et al. [14]

(MS) detector and qualified using statistical and bioinformatics

approaches. The GC-MS analysis allowed the separation of volatile
compounds through a capillary column whose properties depend
on the column’s dimensions (length, diameter, film thickness) as
well as stationary phase. Due to the polarity and boiling point of
target compounds in sourdough bread, polar stationary phases are
more suitable for crumb aroma analysis, while low polar stationary
phases are for crust aroma analysis [14]. The chromatography
parameters including temperature, time, and rate of gradient tem-
perature would depend on the target compounds. Concerning the
MS detector source, the electron ionization (EI) mass spectrometry
combined with GC allows for the acquisition of volatile and ther-
mally stable compounds through their breakage into molecular and
fragment ions [18]. The EI fragmentation is the most common
ionization technique used in organic mass spectrometry for analyz-
ing low- to medium-polarity, non-ionic organic compounds of
molecular weights limited to about 800 Dalton. Due to its high
reproducibility, chromatographic peak resolution, and the existence
of libraries of mass spectra, GC-EI-MS is regarded as the gold
standard for metabolomics research in food matrix.
Following the acquisition process, the aim is to extract infor-
mation on as many VOCs as possible in an untargeted approach.
Different software packages have been developed to handle the
acquired data, including AMDIS (Automatic Mass spectral Decon-
volution and Identification System), ChromaTOF (Leco Inc.,
USA), and PARADISe (PARAFAC2-based Deconvolution and
Identification System). These programs differ in how the calcula-
tions are performed on the samples and the amount of user input
 Volatile Organic Compounds in Sourdough Bread 131

required. AMDIS performs a deconvolution calculation of spectra

based on mass profiles. Deconvolution is the process of computa-
tional separation of co-eluting components that creates a pure
spectrum for each component. By linking to a mass spectral library
(e.g., National Institute of Standards and Technology—NIST
GC-Ms database), the deconvoluted compounds can be identified
generating a report that includes the compound identification, the
relative abundance in each sample, the resolved mass spectra for
each compound, and the spectral library hit information. The
identification of the deconvoluted mass spectra and calculated
retention indices are matched against the NIST database and/or
can be confirmed by comparing it to standard compounds.
This chapter will provide a description of HS-SPME GC-MS
analysis for the evaluation of volatile compounds in sourdough
bread including sample treatment, HS-SPME conditions (sample
amount, fiber type, equilibrium, extraction temperature, and time),
GC-MS conditions (injection settings, column type, temperature
program, MS settings, and scanning mode), as well as quantifica-
tion methods.

2 Materials

2.1 Sample • Blender or mortar with a pestle.

Pre-treatment • Analytical balance.
• 20 mL glass vials.
• 4-Methyl-2-pentanol (98% Sigma-Aldrich; USA) (see Note 1).
• Micro-pipettors, e.g., Gilson Pipetman® (10 μL).
• Polytetrafluoroethylene (PTFE)-coated silicone rubber septa
(20 mm diameter).
• Cooling tray (Peltier Thermostat, CTC analytics, Switzerland).

2.2 Headspace Solid • Autosampler (PAL COMBI-xt, CTC Analytics AG;

Phase Microextraction Switzerland).
(HS-SPME) • Conditioned fiber (270 °C for 30 min) divinylbenzene/car-
boxen/polydimethylsiloxane (DVB/CARB/PDMS)
50/30 μm, Stableflex, 23 Ga, for autosampler (Supelco; USA)
(see Note 2).
• PSS quartz liner with an inner diameter of 1 mm for Program-
mable Split/Splitless Injector (PerkinElmer, Beaconsfield, UK).

2.3 GC-MS Analysis • Grade 5.5 helium (99.9995% purity) regulated to 5 bar.
and Data Analysis • GC Clarus 680 (PerkinElmer, Beaconsfield UK).
• Capillary column Rtx-WAX column (30 m × 0.25 mm i.d.,
0.25 μm film thickness) (Restek Superchrom, Milano, Italy).
132 Giuseppe Celano and Maria De Angelis

• A single-quadrupole mass spectrometer Clarus SQ8MS (Perki-

nElmer) with an Electrone Ionization (EI) source.
• Vacuum system: turbine pump (Edwards TIC Pumping Station
from BOC Edwards; England).
• TurboMass software (Version 6.1.0, Perkinelmer).

3 Methods

3.1 Sample Weigh an amount of 0.750 g (± 0.0050 g) of finally ground

Pre-treatment crumb/crust of bread into a 20 mL vial and sealed with a magnetic
screw cap provided with PTFE/silicone septa. 10 μL of internal
standard (4-methyl-2-pentanol) must be included in each sample
(final concentration of 33 ppm). The vials should be placed on a
cooling tray keeping them at constant temperature of 4 °C.

3.2 Headspace Solid The following parameters have been optimized for HS-SPME/
Phase Microextraction GC-MS method for the analysis of sourdough bread volatile flavor
(HS-SPME) compounds (see Note 3). To reach the equilibrium between the
sample matrices and the headspace, each vial should be kept for
15 min at 60 °C under stirring condition (280 g). After the equili-
bration step, a conditioned fiber of DVB/CAR/PDMS 50/30 μm
is exposed to the headspace of the sample for 60 min at the same
temperature. Desorption is carried out for 3 min in a GC injector
equipped with a quartz liner operating in splitless mode, under a
temperature of 230 °C and helium flow of 1 mL/min. After
desorption, fiber is exposed to 250 °C for additional 2 min.

3.3 GC-MS Analysis The separation of compounds is achieved on a polar capillary

and Data Analysis column Rtx-WAX with polyethylene glycol as the stationary phase
(Restek Superchrom, Milano, Italy). The column temperature is
initially set at 35 °C for 8 min; then increased to 60 °C at 4 °C/min,
to 160 °C at 6 °C /min, and finally to 230 °C at 20 °C /min; and
held for 15 min. Helium is used as the carrier gas at a flow rate of
1 mL/min. A single-quadrupole mass spectrometer Clarus SQ8MS
(PerkinElmer) is used to detect the different compounds, and the
source and transfer line temperatures are kept at 250 and 230 °C,
respectively. The MS detector system under vacuum (≃6.95e-6
Torr) operates in scan mode with mass-to-charge ratio interval of
30–350 Da. The electronic ionization (EI) mode is used with
ionization energy of 70 eV. Chromatographic data are obtained in
full scan mode and analyzed using the program TurboMass soft-
ware. Each chromatogram is analyzed for peak identification by
operating the deconvolution and comparing (i) retention time
(RT) of detected compound with those of provided pure standard
for HPLC (Sigma-Aldrich, St. Louis, MO, USA) and
(ii) experimental mass spectra with those of the National Institute
 Volatile Organic Compounds in Sourdough Bread 133

of Standards and Technology database (NIST/EPA/NIH Mass

Spectral Library with Search Program, data version NIST 05, soft-
ware version 2.0 d). A peak area threshold of > 000000 and a
match criterion of >85% are used for VOC identification followed
by manual visual inspection of the fragment patterns when
required. The relative concentrations of the identified compounds
are normalized to the 4-methyl-2-pentanol internal standard prior
to further analysis using the following equation:
Ac
C= × C is
A is
where C is the normalized concentration of the compound of
interest; Ac is the relative concentration of the compound of inter-
est; Ais is the relative concentration of the internal standard; and Cis
is the concentration of the added internal standard.

4 Notes

1. The limitation of the ISs mentioned above arises from the fact
that it would only provide accurate data when the stability and
physical properties of the compound to be analyzed and that of
the IS are very similar. In this line, it is possible to employ
3-heptanol as IS for both crumb and crust analyses, and
2-methyl-3-heptanone has been also reported as an IS for
crust [4, 19].
2. For the extraction procedure of VOCs in bread, several types of
fiber have been tested including PDMS/DVB, CAR/PDMS,
DVB/CAR/PDMS, and polyacrylate [15].
3. Among the useful parameters to be optimized are the extrac-
tion temperature, which should not generate a Maillard reac-
tion, and the equilibrium time, which must be sufficient to
obtain volatile compound partition in the headspace. It is
important to underline that each parameter needs to be care-
fully set because it directly impacts the extraction optimization.

References

1. Quı́lez J, Ruiz JA, Romero MP (2006) Rela- compounds in white bread by aroma extract
tionships between sensory flavor evaluation dilution analysis, quantitation, and sensory
and volatile and nonvolatile compounds in eval- uation experiments. J Food Proc Preserv
commercial wheat bread type baguette. J 43(5):e13933
Food Sci 71(6):S423eS427. [Link] 4. Pico J, Bernal J, Gomez M (2015) Wheat bread
10.1111/j.1750-3841.2006.00053.x aroma compounds in crumb and crust – a
2. Cho IH, Peterson DG (2010) Chemistry of review. Food Res Int 75:200e215
bread aroma: a review. Food Sci Biotechnol 5. Bianchi F, Careri M, Chiavaro E, Musci M,
19(3):575–582 Vittadini E (2008) Gas chromatographic-
3. Pu D, Zhang H, Zhang Y, Sun B, Ren F, Chen mass spectrometric characterisation of the Ital-
H (2019) Characterization of the key aroma ian Protected Designation of Origin ‘Altamura’
134 Giuseppe Celano and Maria De Angelis

bread volatile profile. Food Chem 110:787– 13. Moskowitz MR, Bin Q, Elias RJ, Peterson DG
793 (2012) Influence of endogenous ferulic acid in
6. Birch AN, Petersen MA, Hansen ÅS (2013) whole wheat flour on bread crust aroma. J
The aroma profile of wheat bread crumb influ- Agric Food Chem 60:11245–11252
enced by yeast concentration and fermentation 14. Dong Y, Karboune S (2021) A review of bread
temperature. LWT- Food Sci Technol 50:480– qualities and current strategies for bread bio-
488 protection: flavor, sensory, rheological, and
7. De Vuyst L, Comasio A, Kerrebroeck SV textural attributes. Compr Rev Food Sci Food
(2021) Sourdough production: fermentation Saf 20(2):1937–1981
strategies, microbial ecology, and use of 15. Pico J, Antolı́n B, Román L, Gómez M, Bernal
non-flour ingredients. Crit Rev Food Sci Nutr J (2018) Analysis of volatile compounds in
63:1–33 gluten-free bread crusts with an optimised
8. Hansen A, Schieberle P (2005) Generation of and validated SPME-GC/QTOF methodol-
aroma compounds during sourdough fermen- ogy. Food Res Int 106:686–695
tation: applied and fundamental aspects. 16. Drakula S, Mustač NČ, Novotni D, Voučko B,
Trends Food Sci Technol 16:85–94 Krpan M, Hruškar M, Ćurić D (2022) Optimi-
9. Hu Y, Zhang J, Wang S, Liu Y, Li L, Gao M zation and validation of a HS-SPME/GC–MS
(2022) Lactic acid bacteria synergistic fermen- method for the analysis of gluten-free bread
tation affects the flavor and texture of bread. J volatile flavor compounds. Food Anal Methods
Food Sci 87(4):1823–1836 15:1–16
10. Pétel C, Onno B, Prost C (2017) Sourdough 17. Niçin R, Özdemir N, Şimşek Ö, Çon AH
volatile compounds and their contribution to (2022) Production of volatiles relation to
bread: a review. Trends Food Sci Technol 59: bread aroma in flour-based fermentation with
105–123 yeast. Food Chem 378:132125
11. Onishi M, Inoue M, Araki T, Iwabuchi H, 18. Halket JM, Waterman D, Przyborowska AM,
Sagara Y (2011) Characteristic coloring curve Patel RK, Fraser PD, Bramley PM (2005)
for white bread during baking. Biosci Biotech- Chemical derivatization and mass spectral
nol Biochem 75(2):255–260 libraries in metabolic profiling by GC/MS
12. Saa DLT, Nissen L, Gianotti A (2019) Meta- and LC/MS/MS. J Exp Bot 56(410):219–243
bolomic approach to study the impact of flour 19. Scott KB, Turko IV, Phinney KW (2015)
type and fermentation process on volatile pro- Quantitative perfor- mance of internal standard
file of bakery products. Food Res Int 119:510– platforms for absolute protein quantification
516 using multiple reaction monitoring-mass spec-
trometry. Anal Chem 87(8):4429–4435
 Chapter 14

In Vitro Determination of Protein Nutritional Indexes

Erica Pontonio and Carlo Giuseppe Rizzello

Abstract
Most of the recent studies investigating the effects of sourdough on baked goods have highlighted its role in
improving different nutritional features. Some of the nutritional changes caused by sourdough fermenta-
tion seem to be strictly related to the proteolytic activity of the lactic acid bacteria. Indeed, it has been
demonstrated that endogenous enzymes, activated by biological acidification, together with lactic acid
bacteria peptidases, are responsible for the hydrolysis of the native flour proteins in peptides and free amino
acids, thus leading to the increase of the protein bio-availability. Several methods, summarized in this
chapter, have been set up and optimized in order to quantify in vitro and on simulated in vivo conditions the
nutritional effects of sourdough fermentation on the protein fraction.

Key words Nutritional value, In vitro protein digestibility

1 Introduction

The quality of protein depends on the amino acid composition, the

digestibility, and the absorption of the hydrolysis products pro-
duced in the human gastrointestinal tract (GIT). It is, for example,
possible that a protein has a very good amino acid profile but
cannot be well digested and/or absorbed. Policy makers, to make
recommendations on protein requirements, should consider both
factors, i.e., amino acid composition and digestibility. According to
a recent systematic review [1], where the analysis of 527 research
articles revealed the main nutritional attributes of sourdough
breads, the protein digestibility is one of the main issue consistent
with the sourdough biotechnology discussed through years.
Empirical and in vitro scientific evidence agree that sourdough
fermentation is associated with improved bread digestibility, mainly
related to proteins.

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

135
136 Erica Pontonio and Carlo Giuseppe Rizzello

Almost all in vitro investigations concluded that the in vitro

digestibility of protein (IVPD), expressing the stability of protein
hydrolysates and how they withstand digestive processes, increased
with sourdough fermentation. This is apart from the flours and
products.
Since the 1990s, in vitro digestion models have been developed
as a tool to design new food products for human health [2]. It is a
laboratory-based methodology that attempts to simulate digestive
processes that occur in humans or animals. It is less expensive and
slow alternative to in vivo digestibility, which analyzes the differ-
ences in composition between ingested food and feces after the
digestion process. During the last decades, the study of the diges-
tive system, due to its connection with various fields of scientific
research such as nutrition, toxicology, pharmacology, and microbi-
ology, is becoming highly important. To study the complex multi-
stage process of human digestion (the oral, gastric, and small
intestinal phases), a wide range of in vitro systems that are flexible,
accurate, and reproducible from single static systems to dynamic
multi-compartmental models have been proposed [3]. Most
dynamic models have been shown to be suitable for simulating
the digestion of foods and pharmaceutical products in different
population groups and for different purposes. However, these
models are relatively complex and expensive to set up and maintain
and, therefore, may not be available to most food researchers.
Owing to their simplicity, static models, which use a constant
ratio of food to enzymes and electrolytes, and a constant pH for
each digestive phase, have been widely used for many decades for
food, animal feed, and pharmaceutical purposes. Static in vitro
digestion models have been shown to be very useful in predicting
outcomes of in vivo digestion. Nevertheless, these methods can
hardly fully mimic the various digestive parameters (i.e., pH, diges-
tion time, enzyme concentration) which are needed to allow com-
parison of results between different models [4].

2 Materials

2.1 Chemicals Hydrochloric acid (HCl), pepsin (e.g., pepsin from porcine pan-
creas, P7012-25G, Sigma Aldrich), sodium hydroxide (NaOH)
(0.2 M), pancreatin (e.g., from porcine pancreas, P7545-100G,
Sigma-Aldrich), sodium phosphate buffer (pH 8.0), merthiolate
(50 ppm), picric acid, toluene, trichloroacetic acid acetate buffer,
amylase (e.g., fungal source, Megazyme; dosage: 2.5 U/g freeze-
dried bread sample), bovine serum albumin, sodium chloride,
sodium bicarbonate, potassium phosphate buffer, amyloglucosi-
dase (3260 U/mL; E-AMGDF, Megazyme, Bray, Ireland), and
2,4,6- trinitrobenzene sufonic acid (TNBS).
 In Vitro Determination of Protein Nutritional Indexes 137

3 Methods

3.1 Protein Overall, protein digestion begins in the stomach with pepsin action,
Digestibility continuing in the intestine with the digestion of trypsin and chy-
motrypsin, and is completed by the action of proteases on the
intestinal surface. Despite this, due to the complexity of the
in vitro digestion simulation, some authors include, in addition to
the gastric and intestinal phase, an oral phase (with amylases or
artificial saliva). Depending on the phases included in the in vitro
stimulation, the results can vary significantly. The enzymes used are
usually of commercial origin. There are no major variations
between studies, mainly using α-amylase in the oral phase, pepsin
in the gastric phase, and pancreatin in the intestinal phase. Accord-
ing to the most used methods, the digestion procedure consists of
three distinct steps: (i) sample preparation; (ii) the digestion proce-
dure involving oral, gastric, and intestinal digestion using simulated
salivary fluid (SSF), simulated gastric fluid (SGF), and simulated
intestinal fluid (SIF) at standard times, temperatures, and condi-
tions; and (iii) sample processing and analysis [5]. Here, the details
related to the most used protocols for the digestion of sourdough
and related products preparatory for the quantification of the
in vitro protein digestibility (IVPD).
Method #1 proposed by Akeson and Stahmann [6]
1. Pepsin hydrolysis with 15 mL of 0.1 M HCl at pH 2.0 and
1.5 mg pepsin (pepsin from porcine pancreas, P7012-25G,
Sigma Aldrich) incubated for 24 h at 37 °C in water bath;
0.2 M NaOH is used to adjust the pH to 8.0.
2. Pancreatin digestion with 4 mg of pancreatin (from porcine
pancreas, P7545-100G, Sigma-Aldrich) dissolved with 7.5 mL
sodium phosphate buffer (pH 8.0), incubated for 24 h at 37 °C
with constant shaking.
3. Merthiolate (50 ppm) is added to the digestion mixture to
prevent growth of microorganisms.
4. A volume of 10 mL of the digestion mixture is added with
50 mL of picric acid solution (1%) and centrifuged for 30 min
at 1000× g to remove undigested protein and larger peptides.
5. Fifty milliliters of supernatant is passed through a column
containing anion exchange resin in chloride form into a
250 mL lyophilizing bottle.
6. After rinsing the column with three 5 mL portions of 0.02 N
hydrochloric acid, the samples are dried by lyophilization.
7. The dried samples are dissolved and diluted to 10 mL with
pH 2.2 buffer.
138 Erica Pontonio and Carlo Giuseppe Rizzello

Notes: Amino acid analysis of the samples is performed by

using ion exchange method. Basic amino acids were separated on
a 10 cm column using pH 5.28 buffer. The acidic and neutral
amino acids were separated on a 159 cm column using pH 3.25
buffer followed by pH 4.25 buffer after 8 h and 30 min from zero
time. The total amino acid content of the samples was determined
on acid hydrolysates.
Method #2, proposed by Akeson and Stahmann [6] and
modified by Rizzello et al. [7]
1. One gram of each sample is incubated with 1.5 mg of pepsin in
15 mL of 0.1 M HCl at 37 °C for 3 h.
2. After neutralization with 2 M NaOH and addition of 4 mg of
pancreatin, in 7.5 mL of phosphate buffer (pH 8.0), 1 mL of
toluene is added to prevent microbial growth, and the solution
is incubated for 24 h at 37 °C.
3. After 24 h, the enzyme is inactivated by addition of 10 mL of
trichloroacetic acid (20%, wt/vol), and the undigested protein
is precipitated.
4. The volume is made up to 100 mL with distilled water and
centrifuged at 5000 rpm for 20 min.
5. The concentration of protein of the supernatant is determined
by the Bradford method [8]. The precipitate is subjected to
protein extraction, according to Weiss et al. [9], and the con-
centration of protein is determined. The IVPD is expressed as
the percentage of the total protein, which is solubilized after
enzyme hydrolysis.
Method#3, proposed by Ibrahim et al. [10] and modified by
Ferreya et al. [11]
1. Incubate the bread samples with 7.5 mg pepsin (250 UI mg-1)
in 7.5 mL HCl (0.1 M) at 37 °C for 3 h.
2. Add 3.75 mL of 0.2 M NaOH and 2 mg of pancreatin in pH 8
phosphate buffer.
3. Incubate for 24 h at 37 °C.
4. Interrupt the enzymatic reaction by adding 25 mL of 10%
trichloroacetic acid to 5 mL of sample.
Nitrogen in the supernatant is determined by Kjeldahl method
and employed to calculate the degree of hydrolysis. IVPD is calcu-
lated according to Ibrahim et al. [10].
Method#4, proposed by Lappi et al. [12] using a sequential
amylase and pepsin treatment to mimic hydrolysis reactions in the
mouth and stomach, respectively
 In Vitro Determination of Protein Nutritional Indexes 139

1. Carry out pre-hydratation of 100 mg of freeze-dried milled

bread sample for 15 min at 40 °C in 0.5 mL 0.2 M acetate
buffer, pH 5.
2. Add amylase (fungal source, Megazyme; dosage: 2.5 U/g
freeze-dried bread sample) to the pre-hydrated suspension
and incubate for 10 min.
3. Adjust the pH of the suspension to 2.0 and add pepsin (2000
FIP U/g, Merck; dosage: 0.4 FIP U/g freeze-dried bread
sample).
4. Collect samples after pepsin incubation times of 0, 5, 15, 30,
60, 90, 120, and 150 min.
5. Adjust the pH of the suspension to 5.0 to terminate the enzy-
matic reaction.
6. Centrifuge the samples to remove the solids.
The solubilized proteins in the supernatant are determined by
the Bio-Rad DC protein assay kit (Bio-Rad, Richmond, CA, USA)
using bovine serum albumin as standard.
Method#5, proposed by Gulati et al. [13]
1. Suspend 120 mg of dried, ground bread in 4 mL of simulated
gastric fluid (SGF; 0.5 mol/L NaCl adjusted to pH 2.5 with
HCl) and incubate at 37 °C for 10 min.
2. Mix the content with pepsin (284 U mg - 1 solid; P-7000,
Sigma-Aldrich, St Louis, MO, USA) dissolved in SGF to give
an activity of 200 U of pepsin mg - 1 of protein in the bread
and incubate for 2 h.
3. Stop gastric digestion by increasing the pH to 7 using 0.5 mol/
L sodium bicarbonate.
4. Intestinal digestion mixing pepsin-hydrolyzed samples with
1 mL simulated intestinal fluid (SIF) (0.05 mol/L potassium
phosphate buffer, pH: 7.0) and warming at 37 °C. Three
milliliters of pancreatin (P-7545, Sigma-Aldrich) in SIF
(2 mg/mL) containing 2 μL amyloglucosidase (3260 U/mL;
E-AMGDF, Megazyme, Bray, Ireland) was added to the sample
mixture and incubated for 4 h.
5. Stop intestinal digestion by plunging the tubes into a boiling
water bath for 5 min and then immediately cool.
6. Centrifuge the digested samples and analyze the supernatants
for primary amine groups with 2,4,6-trinitrobenzene sufonic
acid (TNBS) using leucine as the standard.
Notes: Enzyme blanks, that is tubes containing enzymes and
buffers without bread samples, are maintained and analyzed to
correct for primary amines generated by the enzymes alone and
subtracted from the sample results.
140 Erica Pontonio and Carlo Giuseppe Rizzello

A consensus standardized and practical static digestion method

based on physiologically relevant conditions that can be applied for
various endpoints was developed within the COST Action InfoGest
(Fig. 1). The method comprises up to three stages that mimic the
oral, gastric, and small intestinal phases of digestion in vivo. At each
stage, the duration and physical and biochemical environment are
described and the reasons for their selection given. The enzymes

Fig. 1 A simplified version of the flow diagram describing the InfoGest digestion method as proposed by
Mackie and Rigby [14]
 In Vitro Determination of Protein Nutritional Indexes 141

recommended for inclusion are described using their IUBMB

Enzyme Nomenclature and the method has been written in such
a way as to allow the sourcing of material from any suitable
supplier [14].
Since the study of digestibility in vitro can be compromised
sometimes by failing to mimic the dynamic aspects of human
digestion, such as mechanical forces and fluid dynamics, the study
of in vitro digestibility needs to be complemented through more
sophisticated in vitro dynamic models [4]. Indeed, dynamic models
can be used for a more sophisticated simulation of in vivo condition
to study the in vitro digestibility. In this way, greater control of
digested meal transport, digestive and physiological fluid secretion,
and pH changes over time allows thoroughly simulated conditions
in the human GIT [15]. Among dynamic in vitro systems, different
models exist such as dynamic gastric model (DGM), human gastric
simulator (HGS), ARtificial COLon-ARCOL (monocompartmen-
tal systems), DIDGI® (consecutive compartments simulating the
stomach and the small intestine), Engineered Stomach and small
INtestinal (ESIN) system, and SIMulator of the GIT—simgi®
(multi-compartmental dynamic in vitro models of the human stom-
ach and small intestine), the TNO gastrointestinal model (TIM),
and the Simulator of the Human Intestinal Microbial Ecosystem-
SHIME®. Among them, TIM and SHIME® model represent a real
multi-compartmental artificial gastrointestinal system [16]. TIM-1
simulates the stomach, duodenum, jejunum, and ileum, later com-
prising a large intestine model (TIM-2) providing a full simulation
of GIT [4]. However, the literature reported that the SHIME
(Ghent University-Prodigest, Belgium) [17] is the first multi-
compartmental in vitro digestion system that represents the whole
GIT (except mouth). The stomach (ST), small intestine (SI),
ascending colon (AC), transverse colon (TC), and descending
colon (DC) are simulated by five double-jacketed vessels main-
tained at body temperature (37 °C).

3.2 Protein The well-established substantial difference in the nutritional value

Nutritional Value of proteins from different sources [18] has been captured in a range
of protein quality scoring methods. One of the most popular
scoring methods is the protein digestibility corrected amino acid
score (PDCAAS) [18, 19]. PDCAAS is calculated as:
mg of limiting amino acid in 1 g of test protein
PDCAAS = × fecal true digestibility
mg of same amino acid in 1 g of reference protein
A more recent protein quality score is the digestible indispens-
able amino acid score (DIAAS) that compares the content of all
digestible essential amino acids in a protein to the level of these
digestible amino acids in a reference protein. The reference protein
has an indispensable amino acid sequence that is like the profile
required by a 0.5- to 3-year-old child. This newer score is
142 Erica Pontonio and Carlo Giuseppe Rizzello

considered superior to PDCAAS as it utilizes the true ileal digest-

ibility instead of the fecal digestibility of the proteins. Amino acids,
peptides, and proteins that are not absorbed in the small intestine
end up in the colon where they are metabolized by the microbiota.
Furthermore, the truncation of PDCAAS to 1.0 means that impor-
tant information on highly nutritional proteins is discarded. DIAAS
is calculated as follows:
mg of digestible dietary indispensable AA in 1 g dietary protein
DIAAS = 100 ×
mg of the same dietary indispensable AA in 1 g reference protein
The indispensable amino acid in the DIAAS equation is the
amino acid that has the lowest reference ratio [20]. DIAAS values
for a variety of (cooked) cereals and pseudocereals have shown that
none of the cereals can be a good protein source.
Other indexes are also used to estimate the protein quality in
sourdough and related products starting from the digested protein
fraction obtained after the in vitro protein digestion protocol (e.g.,
Subheading 3.1).
1. Chemical Score (CS): This estimates the amount of protein
required to provide the minimal essential amino acids (EAA)
pattern for adults, which was recently redefined by the FAO
(Food and Agriculture Organization) in 2007. It is calculated
using the following equation [21]:

CS = ðmg AA=g sample proteinÞ * 100=ðmg AA=g egg protein or FAOÞ

The sequence of limiting essential amino acids corresponds to
the list of EAA, having the lowest chemical score [21].
2. Protein Score: This indicates the chemical score of the most
limiting EAA that is present in the test protein.
3. Essential Amino Acids Index (EAAI): This estimates the
quality of the test protein, using its EAA content as the crite-
rion. EAAI is calculated according to the procedure of Oser
[22]. It considers the ratio between EAA of the test protein and
EAA of the reference protein, according to the following
equation:

ðEAA * 1 100ÞðEAA * 2 100Þð. . .ÞðEAA * n 100Þ½sample]

EAAI =
ðEAA * 1 100ÞðEAA * 2 100Þð. . .ÞðEAA * n 100Þ½reference]

4. Biological Value (BV): This indicates the utilizable fraction of

the test protein. BV is calculated using the following equation
[22]:

BV = ð½1:09 × EAAI]–11:70Þ
 In Vitro Determination of Protein Nutritional Indexes 143

5. Protein Efficiency Ratio (PER): This estimates the protein

nutritional quality based on the amino acid profile after hydro-
lysis. PER is determined using the model developed by Ihekor-
onye [23]:

PER = ð- 0:468 þ ð0:454 × ½Leucine]Þ–ð0:105 * ½Tyrosine]Þ

6. Nutritional Index (NI): This normalizes the qualitative and
quantitative variations of the test protein compared to its nutri-
tional status. NI was calculated using the equation of Crisan
and Sands [24], which considers all the factors with an equal
importance:
EAAI × Protein ð%Þ
NI =
100

References

1. Arora K, Ameur H, Polo A et al (2021) Thirty 9. Weiss W, Vogelmeier C, Görg A (1993) Elec-
years of knowledge on sourdough fermenta- trophoretic characterization of wheat grain
tion: a systematic review. Trends Food Sci allergens from different cultivars involved in
Technol 108:71–83 bakers’ asthma. Electrophoresis 14:805–816
2. Ferrua MJ, Singh RP (2011) Understanding 10. Ibrahim F, Babiker E, Yousif N et al (2005)
the fluid dynamics of gastric digestion using Effect of fermentation on biochemical and sen-
computational modeling. Procedia Food Sci sory characteristics of sorghum flour supple-
1:1465–1472 mented with whey protein. Food Chem 92:
3. Guerra A, Etienne-Mesmin L, Livrelli V et al 285–292
(2012) Relevance and challenges in modeling 11. Ferreyra LS, Verdini RA, Soazo M (2021)
human gastric and small intestinal digestion. Impact of whey protein addition on wheat
Trends Biotechnol 30:591–600 bread fermented with a spontaneous sour-
4. Li C, Yu W, Wu P et al (2020) Current in vitro dough. Int J Food Sci Technol 56:4738–4745
digestion systems for understanding food 12. Lappi J, Selinheimo E, Schwab U et al (2010)
digestion in human upper gastrointestinal Sourdough fermentation of wholemeal wheat
tract. Trends Food Sci Technol 96:114–126 bread increases solubility of arabinoxylan and
5. Brodkorb A, Egger L, Alminger M et al (2019) protein and decreases postprandial glucose and
INFOGEST static in vitro simulation of gastro- insulin responses. J Cereal Sci 51:152–158
intestinal food digestion. Nat Protoc 14:991– 13. Gulati P, Brahma S, Graybosch RA et al (2020)
1014 In vitro digestibility of proteins from historical
6. Akeson WR, Stahmann MA (1964) A pepsin and modern wheat cultivars. J Sci Food Agric
pancreatin digest index of protein quality eval- 100:2579–2584
uation. J Nutr 83:257–261 14. Mackie A, Rigby M (2015) InfoGest consensus
7. Rizzello CG, Lorusso A, Montemurro M et al method. In: Verhoeckx K et al (eds) The impact
(2016) Use of sourdough made with quinoa of food bio-actives on gut health.
(Chenopodium quinoa) flour and autochtho- Springer, Cham
nous selected lactic acid bacteria for enhancing 15. Minekus M, Alminger M, Alvito P et al (2014)
the nutritional, textural and sensory features of A standardised static in vitro digestion method
white bread. Food Microbiol 56:1–13 suitable for food-an international consensus.
8. Bradford MM (1976) A rapid and sensitive Food Funct 5:1113–1124
method for the quantitation of microgram 16. Dupont D, Alric M, Blanquet-Diot S et al
quantities of protein utilizing the principle of (2019) Can dynamic in vitro digestion systems
protein-dye binding. Anal Biochem 72:248– mimic the physiological reality? Crit Rev Food
254 Sci Nutr 59:1546–1562
144 Erica Pontonio and Carlo Giuseppe Rizzello

17. Molly K, Vande Woestyne M, Verstraete W 21. Block RJ, Mitchel HH (1946) The correlation
(1993) Development of a 5-step multi-cham- of the amino acid composition of protein with
ber reactor as a simulation of the human intes- their nutritive value. Nutr Abstr Rev 16:249–
tinal microbial ecosystem. Appl Microbiol 278
Biotechnol 39:54–258 22. Oser BL (1959) An integrated essential amino
18. Schaafsma G (2000) The protein digestibility- acid index for predicting the biological value of
corrected amino acid score. J Nutr 130:1865S– proteins. In: Albanese AA (ed) Protein and
1867S amino acid acids in nutrition. Academic,
19. Sarwar G (1997) The protein digestibility- New York
corrected amino acid score method overesti- 23. Ihekoronye AI (1981) A rapid enzymatic and
mates quality of proteins containing antinutri- chromatographic predictive model for the
tional factors and of poorly digestible proteins in-vivo rat-based protein efficiency ratio. Uni-
supplemented with limiting amino acids in rats. versity of Missouri
J Nutr 127:758–764 24. Crisan EV, Sands A (1978) Nutritional
20. Han F, Han F, Wang Y et al (2019) Digestible value. In: Chang ST, Hayes WA (eds) The biol-
indispensable amino acid scores of nine cooked ogy and cultivation of edible mushrooms. Aca-
cereal grains. Br J Nutr 121:30–41 demic, New York
 Chapter 15

In Vitro Determination of the Glycemic Index

Alice Costantini, Olga Nikoloudaki, and Raffaella Di Cagno

Abstract
The use of sourdough as leavening agent, among other nutritional attributes, is preferable for making bread
and other leavened baked goods with medium or low glycemic index (GI).
In vitro methods mimicking in vivo human digestion are simple, reliable, accurate, and cost-effective to
predict and estimate the glycemic response to a food.
Here, we describe a procedure for starch hydrolysis—in vitro digestion. In the first part of the process, the
native starch of the sample is broken down into D-glucose and determined. Then, the sample is subjected to
an enzymatic treatment. The rate of starch hydrolysis is defined as the percentage of starch hydrolyzed at
different time points, and the GI index may be predicted using a mathematical model. This process can be
largely used in the evaluation of the predicted GI (pGI) in starchy products, where fermented cereals and
legumes flour dough are most commonly used.

Key words Baked products, Enzymatic treatmen, In vitro digestion, Predicted glycemic index (pGI),
Starch hydrolysis, Sourdough bread

1 Introduction

Starch represents the main source of carbohydrates in bread

[1]. The concept of the glycemic index (GI) was developed in
1981 to classify foods based on their human glycemic response
[2]. The GI is used to estimate the blood glucose response of a
test food to a reference food (anhydrous glucose, white bread, or
other standard) [3]. The postprandial glucose level following con-
sumption of starchy foods is correlated with the starch hydrolysis
[4, 5].
The use of sourdough as leavening agent has been shown to
decrease the GI in bread and other leavened baked products
[6, 7]. This is mostly associated with acidification process, the
release of organic acids, such as lactic and acetic acids, and related
mechanisms [8]. These effects may be related to the increased

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

145
146 Alice Costantini et al.

content of organic acids and soluble fibers from sourdough, which

increases the rate of gastric emptying and glucose transport
through the intestinal mucosa [9].
In vivo methods that require human involvement are expensive
and time-consuming [10]. Hence, in vitro methods are a suitable,
inexpensive, and a fast alternative for estimating glycemic response
following consumption of staple foods.
The in vitro digestion of starch comprises a sequential enzy-
matic treatment through incubation with pepsin and pancreatic
α-amylase [5, 11, 12]. Glucoamylase derives from a mixture of
Aspergillus niger and A. orzyzae enzymes, which are used to catalyze
the hydrolysis of a-D-(l->4) and a-D-(l->6) linkages. Starch is
completely converted to D-glucose and the rate of starch hydrolysis
can be quantified indirectly using the glucose-to-starch conversion
factor 0.9. Soluble starch is calculated using the conversion factor
of 0.9 considering anhydrous glucose units as starch subunits
[MW of starch monomer/MW of glucose (162/180 = 0.9)]
[5, 13, 14].
The predicted glycemic index (pGI) can be estimated using a
mathematical equation from the value of starch hydrolysis index
(HI) [5].
Glucose can be determined by a variety of methods that are
grouped into two main categories: enzymatic and no-enzymatic
methods, such as high performance liquid chromatography cou-
pled with UV detector (HPLC-UV) and gas chromatography cou-
pled with flame ion detector (GC-FID) [15].
Commonly, an HPLC-UV method detects other simple or
complex carbohydrates during the chromatographic separation.
This is an advantage in the application; however, it requires trained
technicians to properly initialize and set up the method [16]. Gas
chromatography is also a helpful tool for quantifying glucose. In
this scenario, the procedure is extensive because of the low solubil-
ity of glucose in air, requiring derivatization for neutral sugars. This
involves the formation of trimethylsilyl (TMS) ethers, TMS oximes,
and aldonitrile acetates [17].
Conversely, the enzymatic methods are specific for determining
glucose. The most common enzymatic methods exploit the activity
of glucose oxidase (GOx: EC [Link]) and hexokinase (HK:
EC2.7.1.1). HK in the hexokinase method is coupled to the glu-
cose 6-phosphate dehydrogenase (G6PDH) reaction. HK catalyzes
the reaction between glucose and adenosine triphosphate (ATP) to
give glucose 6-phosphate (G6P). G6PDH catalyzes the reaction
between glucose 6-phosphate and nicotinamide-adenine dinucleo-
tide phosphate (NADP+). The reduced nicotinamide-adenine
dinucleotide phosphate (NADPH) resulting from the reaction can
be quantified by spectrophotometric assay at a wavelength of
340 nm [18]. Methods using the coupled reaction of HK and
 Glycemic Index In Vitro 147

Step 1 Weight sample

Homogenize

Incubate at RT
Add buffer
Adjust the pH
Add of amyloglucosidase
Incubate at 60°C
TOTAL STARCH Step 2
DETERMINATION
Glucose
quanitification Weight sample

Add salivary D amylase

Add pepsin
Adjust the pH
Incubate at 37°C

Adjust the pH
Add pancreatic D amylase

Dyalisis procedure
HI
AND pGI
DETERMINATION
Aliquot at Glucose
different times quanitification

Fig. 1 Workflow of in vitro glycemic index determination

G6P-DH are recognized by the AOAC Method 985.09 and are

available in commercial kit form with manufacturer’s instructions.
The method for pGI presented here is a procedure mimicking
the in vivo digestion of starch and its quantification, which is based
on different in vitro procedures [5, 11, 12, 19] with some
modifications.
After total starch determination, samples are subjected to enzy-
matic in vitro digestion by a sequential treatment with salivary
α-amylase, pepsin, and pancreatic α- amylase. Each reaction is
carried out with maximum efficiency targeting the optimal pH of
each enzyme, which reflects the normal conditions of in vivo diges-
tion. Amyloglucosidases are used to hydrolyze digested starch into
glucose. We have selected an enzymatic method to determine the
concentration of glucose, as described in AOAC Method 985.09
[20]. The protocol workflow is described in Fig. 1.

2 Materials

Prepare and store reagents and buffers according to the manufac-

turer’s instructions. Enzymatic solutions need to be prepared as
fresh as possible (see Note 1). Use ultrapure water with a resistivity
of ≥18 MΩ at 25 °C to prepare solutions and buffers. Potentially
hazardous substances will be used. Therefore, before each step,
148 Alice Costantini et al.

check the safety datasheet and wear recommended personal protec-

tive equipment (PPE).

2.1 Sample 1. Homogenizer Stomacher®.

Treatment for the 2. Potassium hydroxide (KOH) 2 M: add about 500 mL of water
Determination of Total in a glass beaker. Weigh 112.22 g of KOH (MW = 56.11 g/
Starch mol) and transfer into the glass beaker (see Note 2). Dissolve it
completely while stirring and adjust the final volume of 1 L
with water. Store at room temperature (RT).
3. Buffer 1: sodium acetate buffer 0.4 M, pH 4.75 (0.2 μm
filtered). Store at RT.
4. Amyloglucosidase solution: amyloglucosidase from A. niger
(2760 U/mL) in water. Store at -20 °C.

2.2 Sample 1. Shaking incubator with temperature control/controller shak-

Enzymatic Digestion ing water bath.
2. Dialysis tubing cellulose membrane (MW cut off 14.000).
3. Buffer 2: Na-K phosphate buffer (PBS) 0.05 M (adjust the pH
at 6.9 with HCl 1 M). Store at 2–8 °C.
4. Sodium azide solution: 0.05% (w/v) sodium azide in water (see
Note 3). Store at 2–8 °C.
5. Dialysis solution: Add 400 mL of Na-K phosphate buffer in a
1 L graduated glass cylinder. Add 800 μL of sodium azide
solution, 80 mg of chloramphenicol, and 4 mL of 2-propanol
(IPA, ≥ 99.5%). Make up volume of 800 mL using Na-K
phosphate buffer (Buffer 2).
6. α-Amylase solution 1: Salivary α-amylase (5250 U/mL) in
water.
7. Pepsin solution: 50 mg of pepsin (from porcine mucosa, 2000
FIP-U/g) in 6 mL of Na-k phosphate buffer (0.05 M,
pH 6.9)).
8. α-Amylase solution 2: 56 mg/mL pancreatic α-amylase
(ca. 50 U/mg) in water.

2.3 Glucose The enzymatic assay follows the enzymatic method AOAC Official
Determination Method 985.09 (see Note 4).
1. Spectrophotometer.
2. Quartz cuvettes (1 cm light path, 3.5 mL).
3. Solution A: Prepare solutions of triethanolamine hydrochloride
buffer 0.94 mol/L, pH 7.6 (plus magnesium sulfate
0.0125 M/L); NADP-Na2 (10 mg/mL); ATP disodium salt
trihydrate (50 mg/mL) plus NaHCO3 (50 mg/mL). Mix
triethanolamine hydrochloride buffer, NADP-Na2, and ATP
solution (8:1:1). Store at 4 °C for a maximum of 1 week.
 Glycemic Index In Vitro 149

4. Solution B: In a glass tube, prepare the required volume of

(NH4)SO4, 3.2 mol/L, pH 6. Add hexokinase (2 mg/mL,
~140 U/mg) and glucose-6-phosphate dehydrogenase
(1 mg/mL, ~140 U/mg).

3 Method

3.1 Determination of 1. Blend the sample (bread or other baked goods) into fine par-
Total Starch ticles (see Note 5).
2. Weigh 1 g of sample and transfer it into a stomacher bag.
3. Add 100 mL KOH 2 M into the stomacher bag.
4. Homogenize well using a homogenizer stomacher® (30 min).
5. Take 6 mL of the suspension and transfer into 50 mL falcon
tube (see Note 6).
6. Wait for 60 min at RT (see Note 7).
7. Add 3 mL of sodium acetate buffer (Buffer 1) and adjust the
pH at 4.75 with HCl (2 M) and NaOH (0.5 M) (see Note 8).
8. Add 400 μL of amyloglucosidase solution.
9. Incubate at 60 °C for 45 min in a controller shaking water bath.
10. Take note of the final volume of the sample using a graduated
pipette (see Note 9).
11. Dilute the sample 10X or 100X using water (see Note 10) and
consider dilution factor (Fd) (see Note 11).
12. Quantify glucose as described below in paragraph 3.3.
13. For determination of total starch, consider the following cal-
culation factors:
Fd: Multiply by dilution factor (1:10 o 1:100)
Fc1: Glucose-to-starch conversion factor, glucose ×0.9
Fc2: Convert g/L to mg/ml. Consider the sample volume.
Convert mg/ml to mg/mg weight of sample. Convert in g
of starch /g of sample.

3.2 Determination of 1. Weigh a sample portion containing 1 g starch into a 50 mL

Starch Hydrolysis and falcon tube.
Dialysis Procedure 2. Add 100 μL of salivary α-amylase solution 1 plus 7 mL of Na-k
phosphate buffer (0.05 M, pH 6.9) and vortex for 1 min.
3. Add the 6 mL of pepsin solution and vortex gently for 1 min.
4. Adjust the pH at 1.5 (optimum for pepsin) using HCl (2 M)
and NaOH (0.5 M).
5. Incubate the sample at 37 °C for 30 min while continuously
shaking.
150 Alice Costantini et al.

Fig. 2 Process of dialysis after enzymatic digestion for glucose release. Sample is placed inside a dialysis
tubing (14 kDa cut-off) into a beaker containing the dialysis solution. Aliquots of dialysate (2 ml) taken at
different time points (30, 60, 120, 180 min, and 16 h) are used to estimate the kinetics of starch hydrolysis

6. Adjust the pH at 6.9 using NaOH (0.5 M).

7. Add 100 μL of pancreatic α-amylase solution 2.
8. Adjust the volume until 30 mL with Na-K phosphate buffer.
9. Cut a piece of dialysis membrane (ca. 60 cm), boil in water for
1 min and hydrate internally.
10. Tie a knot and transfer the entire sample suspension into the
dialysis membrane.
11. Put the filled dialysis membrane into glass beaker with 800 mL
of dialysis solution prewarmed at 37 °C. Incubate at 37 °C
(Fig. 2.)
12. Take an aliquot of 2 mL at 0, 30, 60, 90, 120, 180 min, and
16 h (see Note 12).
13. Add to each aliquots 3 mL of Na-K phosphate buffer.
14. Adjust the pH at 4.75 using HCl (2 M) and NaOH (0.5 M)
(see Note 8).
15. Add 80 μL of amyloglucosidase solution.
16. Step 9, step 10, step 12 as reported above (Subheading 3.1).
17. For determination of digested starch, consider the following
calculation factors:
HI: is expressed as percentage.
Fc1: Glucose-to-starch conversion factor, glucose × 0.9.
 Glycemic Index In Vitro 151

Fc2: Convert to mg/2 mL. Consider the sample volume.

Consider dialysis solution volume. Convert to g/mL
digested starch.

3.3 Enzymatic Assay 1. Prepare always a blank cuvette (b) and a sample cuvette.
AOAC Official Method 2. Set the spectrophotometer at 340 nm. Press autozero.
985.09
3. Add 1 mL of Solution A into the cuvettes (blank (b) and sample
cuvette (s)).
4. Add 0.1 mL of sample into the sample cuvette(s). Instead add
0.1 mL of water for blank cuvette (b).
5. Add 1.9 mL of water.
6. Mix both cuvettes using a film layer and wait for 5 min at
20–25 °C.
7. Read the absorbance at 340 nm for both cuvettes (A1) against
air blank.
8. Add 0.02 mL of Solution B in both cuvettes to start the
reaction.
9. Mix using a film layer and wait for almost 12 min at 20–25 °C
until the reaction stops.
10. Determine A2 for both solutions reading the absorbance at
340 nm.
11. Determine glucose concentration based on following general
formula (see Note 13).

ΔA glucose = A 2ðs Þ - A 1ðs Þ - A2ðb Þ - A 1ðb Þ

V × MW
C ðg=L Þ = = 5:441 × ΔA glucose 6:3
ε × t × v × 1000

3.4 Starch 1. Determine the rate of starch digestion as percentage of total

Hydrolysis Index (HI) available starch hydrolyzed at the five different times point
(30, 60, 90, 120, and 180 min).
3.4.1 Kinetics of Starch
Digestion 2. Use a statistical software (see Note 14) to obtain the hydrolysis
curve as a first-order equation (see Note 15) [5]:

C = C 1 1 - e - kt
3. Determine AUC (area under the hydrolysis curve) of the
kinetic equation of starch digestion (Fig. 3):
AUC test sample
HI =
AUC reference sample
152 Alice Costantini et al.

Fig. 3 Schematic representation of the rate of starch hydrolysis in wheat breads

after in vitro enzymatic sequential treatment at the five time points (30, 60,
90, 120, and 180 min). Soft wheat baker’s yeast bread is used as reference
(diamonds). Soft wheat sourdough bread (squares). Durum wheat sourdough
bread (triangles)

3.4.2 Ratio of Digested 1. Determine the rate of starch hydrolyzed as the ratio of grams
Starch digested at 3 and 16 h.

Starch hydrolized ð3 hÞ
Sample starch hydolized ð%Þ = × 100
Starch hydrolized ð16 hÞ
2. Determine the hydrolysis index (HI) as the percentage ratio of
the sample-digested starch and the reference-digested starch.
HI = Sample starch hydrolized ð%ÞWheat bread starch hydrolized ð%Þ × 100

3.5 Predict Glycemic 1. Use the predictive model [5] to estimate glycemic index:
Index (pGI)
pGI = 39:71 þ 0:549 ðHI Þ

4 Notes

1. Optimal dissolution of enzymes in aqueous solution requires

approximately 5–10 min at controlled pH to limit enzyme
activity.
2. The dissolution of KOH in water generates a strongly exother-
mic reaction. Place the glass beaker in a vessel with ice before
adding the KOH, until it is manageable (ca. 8–10 min).
3. Substances are toxic. Take adequate precaution.
4. Kits based on this enzymatic assay (AOAC Official Method
985.09) are available.
 Glycemic Index In Vitro 153

5. Soft wheat bread is always to be considered as reference. It is

recommended to always analyze a wheat bread together with
the samples.
6. Six milliliter corresponds to 0.06 g of starch present in the
sample.
7. Wait for around 60 min for complete starch solubilization to
take place.
8. The pH rises sharply while adjusting it. It is strongly recom-
mended to add drop by drop the acid or base (HCl (2 M)
and/or NaOH (0.5 M)) until the desired value is reached.
9. The volume of each sample could be different due to the pH
adjustment carried out during previous steps. The sample vol-
ume has to be considered to determine the final concentration
of starch based on the weighted sample.
10. The concentration of glucose should be less than 0.5 g/L. The
estimation of glucose content needs to meet the specification
of the AOAC Official Method 985.09.
11. The sample after the enzymatic treatment is diluted 10X or
100X depending on the glucose estimates concentration
present.
12. Sample aliquots can be kept in the refrigerator (4 °C) during
the incubation time. Aliquot can be stored at -20 °C until
further use.
13. V is the final volume in the cuvette (mL); v is the sample
volume (mL); MW is the molecular weight of glucose
(180.156 g/mol); d is the light path (cm); } is the coefficient
of NADPH at 340 nm (6.3).
14. We recommend using a statistical software dedicated to these
analyses. Statistica 8.0 (StatSoft Inc., Tulsa, USA) or later
version of this software is suitable.
15. C is the concentration at t time, C1 is the equilibrium con-
centration, k is the kinetic constant, and t is the time point.

References

1. Hug-Iten S, Handschin S, Conde-Petit B et al 3. Jenkins DJA, Wolever TMS, Jenkins AL et al

(1999) Changes in starch microstructure on (1983) The glycaemic index of foods tested in
baking and staling of wheat bread. LWT – diabetic patients: a new basis for carbohydrate
Food Sci Technol 32(5):255–260. https:// exchange favouring the use of legumes. Diabe-
[Link]/10.1006/fstl.1999.0544 tologia 24(4):257–264. [Link]
2. Jenkins DJA, Wolever TMS, Taylor RH et al 1007/BF00282710
(1981) Glycemic index of foods: a physiologi- 4. Chung HJ, Shin DH, Lim ST (2008) In vitro
cal basis for carbohydrate exchange. Am J Clin starch digestibility and estimated glycemic
Nutr 34(3):362–366. [Link] index of chemically modified corn starches.
1093/ajcn/34.3.362 Food Res Int 41(6):579–585
154 Alice Costantini et al.

5. Goñi I, Garcia-Alonso A, Saura-Calixto F 13. Bordoloi A, Singh J, Kaur L (2012) In vitro

(1997) A starch hydrolysis procedure to esti- digestibility of starch in cooked potatoes as
mate glycemic index. Nutr Res 17(3):427–437 affected by guar gum: microstructural and rhe-
6. Scazzina F, Del Rio D, Pellegrini N et al (2009) ological characteristics. Food Chem 133(4):
Sourdough bread: starch digestibility and post- 1206–1213
prandial glycemic response. J Cereal Sci 49(3): 14. Dartois A, Singh J, Kaur L et al (2010) Influ-
419–421. [Link] ence of guar gum on the in vitro starch
2008.12.008 digestibility-rheological and microstructural
7. De Angelis M, Rizzello CG, Alfonsi G et al characteristics. Food Biophys 5(3):149–160
(2007) Use of sourdough lactobacilli and oat 15. Galant AL, Kaufman RC, Wilson JD (2015)
fibre to decrease the glycaemic index of white Glucose: detection and analysis. Food Chem
wheat bread. Br J Nutr 98(6):1196–1205 188:149–160
8. Gobbetti M, De Angelis M, Di Cagno R et al 16. Shaw PE, Wilson CW III (1983) Separation of
(2019) Novel insights on the functional/nutri- fructose, glucose and sucrose in fruit by high
tional features of the sourdough fermentation. performance liquid chromatography using UV
Int J Food Microbiol 302:103–113. https:// detection at 190 nm. J Sci Food Agric 34(1):
[Link]/10.1016/[Link].2018.05.018 109–112
9. Fardet A, Leenhardt F, Lioger D et al (2006) 17. Sweeley CC, Bentley R, Makita M et al (1963)
Parameters controlling the glycaemic response Gas-liquid chromatography of Trimethylsilyl
to breads. Nutr Res Rev 19(1):18–25. https:// derivatives of sugars and related substances. J
[Link]/10.1079/NRR2006118 Am Chem Soc 85(16):2497–2507
10. Magaletta RL, DiCataldo SN, Liu D et al 18. Duxbury M (2004) An enzymatic clinical
(2010) In vitro method for predicting glycemic chemistry laboratory experiment incorporating
index of foods using simulated digestion and an introduction to mathematical method com-
an artificial neural network. Cereal Chem parison techniques. Biochem Mol Biol Educ
87(4):363–369 32(4):246–249
11. Liljeberg H, Åkerberg A, Björck I (1996) 19. Brodkorb A, Egger L, Alminger M et al (2019)
Resistant starch formation in bread as influ- INFOGEST static in vitro simulation of gastro-
enced by choice of ingredients or baking con- intestinal food digestion. Nat Protoc 14(4):
ditions. Food Chem 56(4):389–394. https:// 991–1014. [Link]
[Link]/10.1016/0308-8146(95)00199-9 s41596-018-0119-1
12. Granfeldt Y, Bjorck I, Drews A et al (1992) An 20. AOAC (2005) Official method of analysis,
in vitro procedure based on chewing to predict 18th edn. AOAC Press, Maryland
metabolic response to starch in cereal and
legume products. Eur J Clin Nutr 46(9):
649–660
 Chapter 16

Estimation of Phytic Acid Content

Olga Nikoloudaki and Raffaella Di Cagno

Abstract
In plant seeds, phytic acid acts as a storage compound of phosphorus, which can form complexes with
minerals and trace elements reducing their bioavailability, thereby having a negative impact on human
nutrition. Further, phosphorus in this form is unavailable for absorption in monogastric animals and
humans. Dephosphorylation of phytic acid is carried out by enzymes called phytases, which in turn
makes phosphorus available back for absorption. Sourdough fermentation has shown reduced amounts
of phytic acid even by 70% due to the combined action of acidification process and endogenous microbial
phytases. Hence, sourdough can be used to naturally reduce the amount of phytic acid in cereal
(or legumes) breads and improve bioavailability with significant importance in the food industry. Determi-
nation of phytic acid in unprocessed foods and feeds has been extensively studied using diverse methods,
including chromatographic separation, precipitation, and spectroscopy. However, most of these methods
are laborious, costly, are of low throughput, and require specialized analysts. Here, we describe a rapid
colorimetric method, which involves acid extraction of phytic acid, followed by dephosphorylation with
phytase and alkaline phosphatase. The phosphate released is measured with a molybdenum-based blue assay
and calculated as phytic acid or total phosphorus content in the original sample.

Key words Alkaline phosphatase, Ammonium molybdate, Colorimetry, Inorganic Phosphate, Min-
eral bioavailability, Myo-inositol, Phytase, Sourdough, Total phosphorus

1 Introduction

Phytic acid (phytate; myo-inositol 1,2,3,4,5, 6-hexakisphosphate;

InsP6) is a natural component found in legumes, cereals, and
pseudo-cereals, where it forms insoluble complexes with minerals
and other compounds, thus decreasing the dietary bioavailability/
bioaccessibility of trace elements (e.g., Ca2+, Zn2+, Mg2+, Mn2+,
and Fe2+/3+) [1], proteins, lipids, and carbohydrates [2]. Humans
and most animals, except the ruminants, cannot metabolize phytic
acid [3]. Therefore, specific enzymes called phytases are added to
animal feed [4], whereas specific microorganisms that can synthe-
size phytic acid are added to human food to increase bioavailability
of minerals. Phytases (myo-inositol hexakisphosphate

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

155
156 Olga Nikoloudaki and Raffaella Di Cagno

phosphohydrolase; EC [Link]/EC [Link]) hydrolyze phytic

acid, which sequentially releases soluble inorganic phosphate (Pi),
low-size inositol phosphate, and myo-inositol [5]. In the past
decade, research explored the residual content of phytic acid or
minerals, mainly iron, and their bioavailability in wheat flour
doughs and breads subjected to sourdough fermentation [6]. The
sourdough acidification process increased the bioavailability of
minerals indirectly by activating the flour endogenous phytases
and the microbial enzyme activities. The level of acidification can
decrease the phytic acid content above 70% in sourdough [7] and
bread, which demonstrates the importance of having a rapid
method available.
Different methods have been used since early 1990s for the
determination of phytic acid from food matrixes, but mostly grains.
The methods included an extraction process step and quantification
of the analytes. Based on these two principal steps, methods were
divided into classical (precipitation techniques, potentiometric
titrations), separation techniques (liquid-, gas-, ion-, thin layer-
chromatography, and electrophoresis), spectroscopy (colorimetry,
fluorescence, nuclear magnetic resonance (NMR)), inductively
coupled plasma (ICP)), and sensors (electrochemical biosensors
and fluorescence probes) [8]. Even though so many methods
exist, they have numerous disadvantages such as complex extraction
protocols, intensive labor, high cost, and low throughput. Precipi-
tation methods principally include extraction by hydrochloric acid
and titration of the extract with acidic solution of FeCl3. Generally,
precipitation techniques show an inconsistent stochiometric ratio
between the iron and phytic acid in the precipitate, while the
method is non-selective as the titration does not allow the distinc-
tion between phytate, its dephosphorylated analogs, and Pi causing
overestimation [9]. Potentiometric titrations require a precise stan-
dardization of phytic acid prior its use as a standard or for calibra-
tion, which is dependent on the purity of commercially available
phytate salts [8]. Chromatographic methods included in the sepa-
ration techniques have been widely used to overcome the limita-
tions of the precipitation techniques. Despite the fact that
chromatographic methods have shown higher sensitivity, they are
time consuming, have low throughput efficiency, and require
sophisticated equipment [8, 10]. Electrophoretic analysis enables
the characterization of unknown pyrophosphorylated inositol moi-
eties; however, abundance of inorganic phosphates in certain kind
of samples results in underestimation of inositol phosphates
[8]. NMR and ICP are other analytical methods applied for the
determination of phytic acid in sourdough breads within the cate-
gory of spectroscopic techniques [11, 12]. These methods are quite
useful and provide even more insights into the dephosphorylation
mechanism and the lower myo-inositol-, di-, tri-, tetra-, pentakis-
phosphate forms (InsP2-InsP5); however, they are expensive and
 Phytic Acid 157

require highly skilled analysts. AOAC Official MethodSM 986.11

[13], the most common accepted method, applied in foods, animal
feeds, and cereals, has also limitations. This method requires a
separate anion exchange chromatographic (AEC) column for each
sample, followed by acid hydrolysis of phytic acid and colorimetric
determination of phosphorus. The main drawback is that the
method assumes that only phytic acid is purified by the AEC,
which is only true for unprocessed grains or legumes, but not
processed foods [14]. Processed foods, normally depending on
their processing steps, already contain proportions of lower myo-
inositol phosphate forms [15], which results in overestimation of
phytic acid.
The most frequent method used to determine phytate in flour,
sourdough, and breads along with chromatographic based meth-
ods is the colorimetric method. These methods are more rapid and
simple than precipitation and separation methods; however, they
have the same limitations as the AOAC method. Here, we describe
a colorimetric, rapid, high-throughput method that enables the
estimation of phytic acid and total phosphorus content as described
elsewhere [16] with some modifications in the extraction process.
The method involves acid extraction of phytate and myo-inositol
phosphates from sourdough samples, accompanied by enzymatic
dephosphorylation by phytase and further enzymatic dephosphor-
ylation by alkaline phosphate (ALP). ALP guarantees the liberation
of the final phosphate moiety (InsP1) and inorganic phosphate,
which finally ensures that all the myo-inositol phosphates are
released from phytic acid. The coloring agent ammonium molyb-
date reacts with inorganic phosphate formed, and under acidic
environment, molybdenum blue is measured at 655 nm. The
amount of phosphorus is equivalent to the amount of Pi measured
and can be calculated as phytic acid content of the original sample.
The principle of the assay is shown in Fig. 1.

2 Materials

Prepare all materials using distilled water and analytical grade

reagents. Store reagents at room temperature unless otherwise
indicated. When you waste material and solutions, follow the dis-
posal regulations. Follow safety regulations, use glasses and gloves,
and work under chemical wood when using concentrated acids and
volatile reagents.

2.1 Preparation of 1. Solution 1: Ascorbic acid (10%, w/v)/sulfuric acid (1 M). Add
Color Reagent for 10 g ascorbic acid in a beaker and dissolve it by stirring in
Phosphorus 90 mL of distilled water. Afterward, add 5.35 mL of concen-
Determination trated sulfuric acid (12 M). Add distilled water and make up to
a final volume 100 mL. Store at 4 °C for no more than 1 week
(see Note 1).
158 Olga Nikoloudaki and Raffaella Di Cagno

Principle of phytic acid assay

Enzyme phytase hydrolyses phytic acid (phytate; InsP6; myo-inositol hyxakisphosphate) into lower myo-inositol phosphates (di-, tri-, tetra-, pentakis- phosphates;
InsP2-InsP5) and inorganic phosphate (Pi) (I)

(phytase)
(I) Phytic acid (InsP6) + H2O myo-inositol phosphates (InsP2-InsP5) + Pi

Enzymatic dephosphorylation by alkaline phosphatase (ALP) releases the final phosphate moiety; inositol mono-phosphate (InsP1) (II)

(ALP)
(II) myo-inositol phosphates (InsP2-InsP5) + H2O myo-inositol mono-phosphate (InsP1) + Pi

Pi and ammonium molybdate ((NH4)2MoO4) react to form 12-molybdophosphoric acid (H3Mo12O40P), which is further reduced to molybdenum blue H2MoO2-2 (III, IV)

(III) Pi + (NH4)2MoO4 H3Mo12O40P

(IV) H3Mo12O40P + H2SO4 /ascorbic acid H2MoO2-2

The amount of Pi in the sample is equal to the amount of molybdenum blue formed and is recorded at absorbance 655nm. From the phosphorus calibration curve the Pi
can be quantified and expressed as phytic acid content of the original sample.

Fig. 1 Principle of phytic acid determination assay

2. Solution 2: Ammonium molybdate (5%, w/v). Add 1.25 g

ammonium molybdate to 20 mL distilled water and dissolve.
Make up to a final volume of 25 mL with distilled water and
store at 4 °Cfor no more than 1 month.
3. Color reagent. Add one part of Solution 2 to five parts of
Solution 1 (e.g., add 1 mL of Solution 2 to 5 mL of Solution
1). Prepare 0.6 mL per sample to be analyzed. Prepare fresh on
the day of use only.

2.2 Preparation of 1. Phytase assay buffer: 200 mM sodium acetate, pH 5.5. Add
Phytase and Alkaline 27.2 g of sodium acetate trihydrate acid in a beaker and dissolve
Phosphatase Assay in 900 mL distilled water. Adjust the pH to 5.5 by drop-wise
Buffers addition of concentrated hydrochloric acid (12 M). Adjust to
final volume with distilled water (1 L).
2. Alkaline phosphatase (ALP) buffer: 400 mM glycine, 4 mM
magnesium chloride, and 0.4 mM zinc sulfate, pH 10.4. Add
30 g of glycine, 813 mg of magnesium chloride hexahydrate,
and 115 mg zinc sulfate in a beaker and dissolve in 900 mL
distilled water while stirring. Adjust the pH to 10.4 by adding
sodium hydroxide pellets. Adjust to final volume with distilled
water (1 L).

2.3 Other Reagent 1. Trichloroacetic acid (TCA): 50%, w/v. Add 50 g of TCA to
Solutions 60 mL of distilled water and dissolve. Make up to a volume of
100 mL with distilled water and store at 4 °C.
 Phytic Acid 159

2. Hydrochloric acid (HCl): 0.66 M. Add 54.5 mL of HCl (37%,

v/v) to 945.5 mL of distilled water and mix. Store at room
temperature.
3. Sodium hydroxide: 0.75 M. Add 6 g of sodium hydroxide
pellets to 180 mL of distilled water and dissolve. Make up to
a volume of 200 mL with distilled water. Store at room
temperature.
4. Phytic acid standard solution: Dissolve dipotassium salt of
phytic acid (5 mg, 10 mg, 15 mg, 25 mg, and 30 mg) in a
final volume of 1 mL of distilled water. Use of pure phytic acid
dipotassium salt is recommended to test the linearity of the
enzymatic dephosphorylation.
5. Phytase: 12,000 U/mL. Use directly as it is without prior
preparations.
6. Alkaline phosphatase (ALP): 400 U/mL. Use directly as it is
without prior preparations.
7. Phosphorus standard solution: 50 μg/mL. To prepare the
phosphorus standard solutions for the calibration curve, add
0 mL (STD0), 0.05 mL (STD1), 0.25 mL (STD2), 0.50 mL
(STD3), and 0.75 mL (STD4) of the phosphorus standard
solution to 15 mL falcon tubes. Then add distilled water to
each of the falcon tubes until the final volume for each is 5 mL.
The corresponding concentration per point of the calibration
curve should be 0 μg, 0.5 μg, 2.5 μg, 5 μg, and 7.5 μg. The
solutions are stable for 1 week at 4°C.

3 Methods

Carry out procedures at room temperature. Use protective gloves

and lab coat throughout the process.

3.1 Sample 1. Add the lyophilized sourdough (1 g) (see Note 2) to 20 mL

Extraction HCl (0.66 M) and mix with vigorous stirring for a minimum of
3 h and up to 24 h at room temperature (see Note 3).
2. Transfer the extract (ca. 1 mL) to a 1.5 mL microfuge tube and
centrifuge at 11,000× g for 10 min. Immediately after, transfer
0.5 mL of the extract supernatant to a new 1.5 mL microfuge
tube and neutralize by adding 0.5 mL of sodium hydroxide
solution (0.75 M).

3.2 Enzymatic Each sample extracted from the step above will be used for “total
Dephosphorylation of phosphorus” reaction and “free phosphorus” reaction, which are
Phytic Acid by Phytase done simultaneously in different microfuge tubes (1.5 mL).
and ALP
160 Olga Nikoloudaki and Raffaella Di Cagno

1. Total phosphorus reaction: For the total phosphorus reaction,

add 0.20 mL phytase assay buffer, 0.05 mL sample extract,
0.02 mL phytase (12,000 U/mL), and distilled water
(0.60 mL). Final volume is 1.39 mL.
2. Free phosphorus reaction: For the free phosphorus reaction,
add 0.20 mL phytase assay buffer, 0.05 mL sample extract, and
0.62 mL distilled water.
3. Mix thoroughly the microfuge tubes and incubate at 40 °C for
10 min.
4. After incubation, add 0.20 mL of ALP assay buffer and
0.02 mL (80 U/mL) of ALP to the total phosphorus mixture.
Then, add 0.20 mL of ALP assay buffer and 0.02 mL of
distilled water into the free phosphorus reaction mixture.
5. Mix thoroughly the microfuge tubes and incubate at 40 °C for
15 min.
6. After incubation, stop all reactions by adding 0.3 mL TCA
(50%, w/v). Mix thoroughly by vortexing and centrifuge at
11,000× g for 10 min.
7. Transfer the supernatant (1 mL) into a new microfuge tube for
colorimetric determination of phosphorus assay.

3.3 Colorimetric 1. Add color reagent 0.5 mL to 1 mL supernatant from the

Determination of mixtures prepared in previous step (Subheading 3.2) in a
Phosphorus 1.5 mL microfuge tube.
2. Mix thoroughly using a vortex mixer and incubate in a water
bath set at 40 °C for 1 h.
3. After incubation, mix thoroughly all reactions using a vortex
mixer. Transfer approximately 1 mL to a 1 cm path-length
microcuvette and record the absorbance of each solution at
655 nm. Measure also the phosphorus standard solutions
prepared for the calibration curve (see Note 4).
4. The absorbance values of samples and phosphorus standard
solutions will be used in the calculation of total phosphorus
and phytic acid.

3.4 Linearity of the 1. To check the linearity of the enzymatic dephosphorylation of

Phytic Acid Assay phytic acid reaction, prepare phytic acid solutions by adding up
to 30 mg/mL of distilled water and phytic acid. Read each
solution at 655 nm.
2. To check the linearity of the colorimetric determination of
phosphorus assay, use the phosphorus standard solution previ-
ously prepared (up to 7.5 μg/mL) and read absorbance at
655 nm (see Note 5).
 Phytic Acid 161

3.5 Calculation of 1. Subtract the absorbance value of STD0 obtained from the
Total Phosphorus and colorimetric determination of phosphorus assay from the
Phytic Acid absorbance values of the other phosphorus standard solutions
(STD1–STD4) to obtain ΔAphosphorus for each standard.
2. The slope, M (μg/ΔAphosphorus), for each individual phospho-
rus standard solution and the mean M are calculated as follows:

P ðμg Þ
M μg=ΔA phosphorus =
ΔA phosphorus
P(μg): Value of absorbance for each one of the phosphorus
standard solutions
ðMSTD1 þ MSTD2 þ MSTD3 þ MSTD4Þ
mean M =
4

3. Total phosphorus content is calculated as follows

mean M × V × F
Total phosphorous ðg=100g Þ = × ΔAphosphorus
10, 000 × w × v
V: original sample extract volume (mL)
F: dilution factor of the sample in the extraction and the
dephosphorylation reaction
ΔAphosphorus: the absorbance change of sample
10,000: conversion factor from μg/g to g/100 g
w: original weight of sample material used in the extraction (g)
v: the sample volume used in the colorimetric determination of
phosphorus assay (mL)
4. Phytic acid content is calculated as follows:
phosphorus ðg=100g Þ
Phytic acid ðg=100g Þ =
0:282

0.282: the factor used to convert the measured phosphorus

content to phytic acid content (see Notes 6 and 7).

Example of Calculations
Calibration curve calculations are given in Table 1 and the calibra-
tion curve in Fig. 2.
The absorbance of free and total phosphorus of a sourdough
sample (1 g) was 0.21 and 0.928, respectively. Subtraction of the
free phosphorus absorbance from the total phosphorus gives the
ΔAphosphorus, which is 0.718.
162 Olga Nikoloudaki and Raffaella Di Cagno

Table 1
Calculations of phosphorus calibration curve

Phosphorus standard P(μg) A655 ΔAphosphorus μg/ΔAphosphorus Mean M

STD0 0 0.063 0.000 0.000 5.063
STD1 0.5 0.162 0.099 5.051
STD2 2.5 0.558 0.495 5.051
STD3 5 1.062 0.999 5.005
STD4 7.5 1.520 1.457 5.148

1.6

1.4
y = 0.1953x + 0.0675
1.2 R² = 0.9997
Absorbance, 655nm

0.8

0.6

0.4

0.2

0
0 1 2 3 4 5 6 7 8
Phosphorus, µg/assay

Fig. 2 Calibration curve (signal vs known concentration) of standard phosphorus solutions using pure
phosphorus. R2 is the coefficient of determination, and the equation y = mx + b is the slope intercept form
of a straight line. Where, m is the slope of the line and b is the y-intercept of the line

The concentration of phosphorus is then calculated as follows:

5:063 × 20 × 55:6
Total phosphorous ðg=100g Þ = × 0:718
10, 000 × 1 × 1

Total phosphorous ðg=100g Þ = 0:404

0:404
Phytic acid ðg=100g Þ =
0:282

Phytic acid ðg=100g Þ = 1:433

 Phytic Acid 163

4 Notes

1. All principal reagent solutions can be added with sodium azide

(0.02%, w/v) as a preservative, which can extend their stability
up to 2 years when stored at 4 °C. Exceptions are the color
agent, ammonium molybdate, HCl, and sodium hydroxide.
2. Lyophilized sourdough must be ground to fine particles so to
increase the efficiency of extraction.
3. Extraction process requires at least 3 h. Interrupting the pro-
cess after 3 h does not affect the concentration of phytic acid
extracted. For convenience, the extraction can be performed
overnight.
4. It is important to prepare the phosphorus calibration curve
concurrently with the batch of samples to ensure the use of
the same batch of color reagent for both sample and calibration
curve aliquots.
5. Linearity of phytic acid assay is dependent on the (i) extraction
process; (ii) enzymatic dephosphorylation of phytic acid reac-
tion; (iii) colorimetric determination of phosphorus; and
(iv) limitation of spectrophotometer used to read the absor-
bance above a certain level. Using the protocol described, it is
possible to obtain linearity of the enzymatic dephosphorylation
of phytic acid reaction using phytic acid up to 30 mg/mL,
which is equivalent to a maximum phytic acid content of 3 g
phytic acid/100 g when 1 g of solid sample is applied. Further,
the linearity of the colorimetric determination of phosphorus
assay is achievable up to a phosphorus concentration of 7.5 μg/
mL. Finally, the most important factor influencing the dynamic
range of phytic acid assay is the accuracy of spectrophotometer
above a certain absorbance. Moreover, it is recommended to
test the linearity of phytic acid assay to adapt the method to
peculiar experimental conditions (different types of samples or
use of 96 multi-well plates).
6. The calculation of phytic acid content assumes that the amount
of phosphorus measured is exclusively released from phytic acid
(InsP6) (that comprises 28.2% of phytic acid) and not from any
other phosphate esters including the lower myo-inositol phos-
phates (InsP1–5).
7. When the absorbance obtained for a sample is below 0.100,
repeat the standard assay procedure using 4 g of sample mate-
rial for the extraction, and when it is between 0.05 and 0.10,
use 2.5 g of sample. When the absorbance obtained is above
that of STD4, repeat the standard assay procedure using 1 g of
sample material in 100 mL of HCl (0.66 M). Then include the
new volume of extracted sample in the total phosphorus
equation.
164 Olga Nikoloudaki and Raffaella Di Cagno

References
1. Torres J, Domı́nguez S, Cerdá MF et al (2005) 9. Duniere L, Xu S, Long J et al (2017) Bacterial
Solution behaviour of myo-inositol hexaki- and fungal core microbiomes associated with
sphosphate in the presence of multivalent small grain silages during ensiling and aerobic
cations. Prediction of a neutral pentamagne- spoilage. BMC Microbiol 17(1):50. https://
sium species under cytosolic/nuclear condi- [Link]/10.1186/s12866-017-0947-0
tions. J Inorg Biochem 99(3):828–840. 10. Kwanyuen P, Burton JW (2005) A simple and
[Link] rapid procedure for phytate determination in
12.011 soybeans and soy products. J Am Oil Chem Soc
2. Kumar V, Sinha AK, Makkar HPS et al (2010) 82(2):81–85. [Link]
Dietary roles of phytate and phytase in human s11746-005-1046-9
nutrition: a review. Food Chem 120(4): 11. Reale A, Mannina L, Tremonte P et al (2004)
945–959. [Link] Phytate degradation by lactic acid bacteria and
foodchem.2009.11.052 yeasts during the wholemeal dough fermenta-
3. Noureddini H, Malik M, Byun J et al (2009) tion: a 31P NMR Study. J Agric Food Chem
Distribution of phosphorus compounds in 52(20):6300–6305. [Link]
corn processing. Bioresour Technol 100(2): 1021/jf049551p
731–736 12. Wu P, Tian JC, Walker CE (Chuck) et al (2009)
4. Rao D, Rao KV, Reddy TP et al (2009) Molec- Determination of phytic acid in cereals – a brief
ular characterization, physicochemical proper- review. Int J Food Sci Technol 44(9):
ties, known and potential applications of 1671–1676. [Link]
phytases: an overview. Crit Rev Biotechnol 1365-2621.2009.01991.x
29(2):182–198 13. Official method of Analysis (2000) AOAC
5. Nagashima T, Tange T, Anazawa H (1999) international 17th edition. Method 986:11
Dephosphorylation of phytate by using the 14. Phillyppi BQ, Johnston MR, Tao SH et al
aspergillus Niger phytase with a high affinity (1988) Inositol phosphates in processed
for phytate. Appl Environ Microbiol 65(10): foods. J Food Sci 53(2):496–499. https://
4682–4684. [Link] [Link]/10.1111/j.1365-2621.1988.
65.10.4682-4684.1999 tb07740.x
6. Arora K, Ameur H, Polo A et al (2021) Thirty 15. Schlemmer U, Frølich W, Prieto RM et al
years of knowledge on sourdough fermenta- (2009) Phytate in foods and significance for
tion: a systematic review. Trends Food Sci humans: food sources, intake, processing, bio-
Technol 108:71–83. [Link] availability, protective role and analysis. Mol
1016/[Link].2020.12.008 Nutr Food Res 53(S2):S330–S375. https://
7. Kumar A, Singh B, Raigond P et al (2021) [Link]/10.1002/mnfr.200900099
Phytic acid: blessing in disguise, a prime com- 16. McKie V, McCleary B (2016) A novel and
pound required for both plant and human rapid colorimetric method for measuring total
nutrition. Food Res Int 142:110193. https:// phosphorus and phytic acid in foods and animal
[Link]/10.1016/[Link].2021.110193 feeds. J AOAC Int 99. [Link]
8. Marolt G, Kolar M (2021) Analytical methods 5740/jaoacint.16-0029
for determination of phytic acid and other ino-
sitol phosphates: a review. Molecules 26:174
 Chapter 17

Phenolic Compounds and In Vitro Antioxidant Activity

Rosanna Latronico and Pasquale Filannino

Abstract
Phenolic compounds (PCs) are mainly responsible for the antioxidant capacity of bread, which is affected by
several factors such as the type of flour, microbial dynamics, endogenous and exogenous enzymes, proces-
sing conditions, and interactions between PCs and other dough components. During sourdough fermen-
tation, PCs undergo significant modifications via endogenous cereal enzymes and microbial metabolism.
This results in increased free PCs, hydrolysis of esters of phenolic acids, and conversion of flavonoid
glycosides to the corresponding aglycone. In vitro colorimetric assays are simple and versatile tools for
measuring the amount of PCs in bread and their antioxidant capacity. They are also utilized for investigating
the effects of fermentation and baking on PCs profile. This chapter will describe the most common assays
implemented for bread analysis, focusing on the association between antioxidant activity and PCs. This
chapter will also describe the methods for extracting free and bound PCs from bread, as well as Folin–
Ciocalteu assay, DPPH and ABTS+ scavenging assays, and ferric ion reducing antioxidant power (FRAP)
assay on microplate or cuvette scales.

Key words Phenolics, ABTS +, DPPH, FRAP, Folin-Ciocalteu

1 Introduction

Phenolic compounds (PCs) in cereal-based products are often con-

sidered anti-nutritional compounds because they can inhibit diges-
tive enzymes leading to poor digestibility of proteins and other
nutrients and are responsible for the bitter or astringent taste. On
the other hand, PCs act as precursors for the formation of aroma
compounds and are ascribed with antioxidant, anti-inflammatory,
anti-carcinogenicity, and antimicrobial properties [1–3].
PCs comprise several hundred molecules whose structure
includes one or more aromatic rings substituted by, at least, one
hydroxyl group. They are associated with various sugar moieties
and organic acids and can bind to biopolymers, such as proteins and
polysaccharides, either covalently or non-covalently. The main PCs
in wheat and rye are phenolic acids, which are mostly found in
bound form and as dimers, rather than as free phenolic acids.

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough,
Methods and Protocols in Food Science, [Link]
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

165
166 Rosanna Latronico and Pasquale Filannino

Ferulic acid is the most abundant phenolic acid, mainly esterified

with arabinoxylans, but other phenolic acids such as caffeic, dihy-
drobenzoic, and sinapic acids are also present. Wholemeal flours
contain more phenolic acids than refined flours. Millet and sor-
ghum have higher levels of PCs than wheat, barley, or rye. In
particular, sorghum contains phenolic acids and their esters mostly
in bound form, as well as flavonoids, condensed tannins, and deox-
yanthocyanidins [1]. Moreover, sourdough fermentation is a com-
mon strategy to recycle various agro-food by-products, which
greatly enhance the quality and quantity of the PCs profile in flours
typically used for baking [4, 5]. Besides this, the contribution of
PCs from unconventional ingredients and flours (e.g., sprouted
seeds, pseudo-cereals, legumes) that are used to fortify bread
should also be taken into account [4].
During sourdough fermentation, PCs undergo modifications
via endogenous cereal enzymes and microbial metabolism, result-
ing in (i) increased free PCs, indicating release of bound com-
pounds, (ii) partial hydrolysis of esters of phenolic acids, and (iii)
partial conversion of flavonoid glycosides to the corresponding
aglycone. Lactic fermentation modulates the phenolic fraction dur-
ing sourdough fermentation due to the enzymatic repertoire of the
lactic acid bacteria, which may include, depending on the species
and strain, (i) feruloyl esterases, which hydrolyze feruloylated sugar
esters; (ii) esterases with specificity for galloyl ester bonds in gallo-
tannins; (iii) glycosyl hydrolases releasing flavonoid aglycons which
show higher bioactivity in humans than their precursor glycosides
and may act as anti-inflammatory adjuvants; and (iv) reductases and
decarboxylases with specificity for phenolic acid [1, 2]. Concerning
the latter enzyme activities, the metabolism of phenolic acids may
shift from decarboxylase to reductase activity, depending on the
fermentation substrate. In fact, free hydroxybenzoic and hydroxy-
cinnamic acids may be decarboxylated to the corresponding phenol
or vinyl derivatives. But lactic acid bacteria display also inducible
phenolic acid reductases, able to hydrogenate the double bond of
hydroxycinnamic acids and their vinyl derivatives. Overall, phenolic
acid derivatives exert higher biological activities than their precur-
sors [1, 2].
PCs affect microbial growth and viability in different ways,
depending on their chemical structure and concentration. Low
concentrations of phenolics may have a stimulatory effect on the
growth of lactic acid bacteria. Conversely, high concentrations
impair the cell wall and membrane integrity, cause the pH gradient
dissipation, and delay the carbohydrate metabolism. The physio-
logical and ecological meaning of the metabolism of PCs in lactic
acid bacteria has been mainly explained as a mechanism to detoxify
such compounds, because some metabolic derivatives of PCs have a
reduced antimicrobial activity compared to their precursors.
Another hypothesis regards PC metabolism by lactic acid bacteria
as a strategy to preserve cellular energy balance [1, 2].
 Phenolics and Antioxidant Activity 167

The concentration of PCs in wheat or rye sourdoughs is several

orders of magnitude lower than the inhibitory concentration for
lactic acid bacteria, whereas in sorghum sourdoughs, it exceeds the
inhibitory concentration for sensitive lactobacilli. For instance,
Fructilactobacillus sanfranciscensis is inhibited by PCs in sorghum,
while other lactobacilli (Lacticaseibacillus casei and Lentilactobacil-
lus parabuchneri) can persist, suggesting that the concentration and
type of PCs can drive the microbial dynamics during sourdough
fermentation [1, 2, 6].
PCs are highly reactive toward other radical components,
which results in a high antioxidant capacity [3]. The type of flours
(e.g., compositions and particle size), microbial dynamics, addition
of exogenous enzymes, chemical environment, processing condi-
tions, and interactions between PCs and carbohydrates or proteins
are crucial for the antioxidant properties and bioavailability of PCs
in bread [3, 7]. In general, fermentation and baking increase the
content of free PCs and their antioxidant activity. The lower pH
due to sourdough fermentation enhances the activity of hydrolases
and esterases, which break the glycoside and ester bonds and release
simple PCs. Other factors such as water content, oxygen, and
temperature affect the fermentation dynamics and the solubility
of PCs. Higher temperature or longer fermentation and baking
change the antioxidant capacity of the PCs fraction by
(i) destroying some antioxidant molecules and creating new ones
(e.g., Maillard products and quinones); (ii) forming covalent bonds
between molecules; (iii) reducing the bound PCs content and
increasing the free PCs fraction [3].
In vitro colorimetric assays are simple and versatile tools for
measuring the PCs content in bread and their antioxidant capacity.
They are also utilized for investigating the effects of fermentation
and baking on PCs profile. The advantages of these assays are not
only the ease and low cost of execution but also the ability to
provide a comprehensive picture in one simple measure, which
can be useful for large screenings with many samples, and to auto-
mate the procedure if needed [8, 9]. Although in vitro methods are
widely used to assess the health-promoting effects of baked pro-
ducts, their reliability is questionable due to (i) the poor correlation
between in vitro and in vivo results, as these assays do not account
for synergistic and antagonistic interactions in biological tissues;
(ii) the interference of many factors unrelated to the concentration
of antioxidant molecules (e.g., metals and pigments) with the
measurement; (iii) the low inter-laboratory reproducibility, as
results are highly dependent on the extraction procedure and the
adopted methodologies [8, 9]. In light of these considerations, it is
clear that each methodology has its advantages and disadvantages
and that the application and interpretation of the results must be
consistent with the context of the specific research.
168 Rosanna Latronico and Pasquale Filannino

Several reviews have previously described in detail the reaction

mechanisms, chemical kinetics, and thermodynamics underlying
the methods for in vitro testing of the antioxidant activity of food
matrices [8–10]. This chapter will describe the most common
assays implemented for bread analysis, focusing on the association
between antioxidant activity and PCs. After describing the extrac-
tion of free and bound PCs from bread [11, 12], this chapter
presents Folin–Ciocalteu assay [7], DPPH· and ABTS·+ scavenging
assays [7, 13], and ferric ion reducing antioxidant power (FRAP)
assay [13] on microplate or cuvette scales.
PCs extraction is one of the critical steps in the implementation
of these assays, because it must ensure the complete solubilization
of total PCs in a solvent. The effectiveness of the extraction
depends on the interplay between the solubilization ability of the
solvent, relative solubility of PCs, and bread microstructure. The
polarity of the solvent is particularly critical since it greatly deter-
mines the selectivity of the extraction. Since PCs are mainly polar
compounds, aliphatic alcohols (e.g., methanol and ethanol) and
polar organic solvents (e.g., acetone and ethyl acetate) are the most
common choices of solvents used in the extraction of PCs from
food matrices, although their polarity and affinity can vary depend-
ing on the number and arrangement of the phenol groups [11].
The Folin–Ciocalteu assay is used to determine the total PCs
content in bread, which is based on single electron transfer reac-
tions between the Folin–Ciocalteu reagent and PCs. At the pH
value of the reaction mixture (pH 10), the dissociation of a pheno-
lic proton leads to the formation of a phenolate ion, which causes
the reduction of the Folin–Ciocalteu reagent, whose intense yellow
color turns into a blue color [9, 10]. However, other substances
(e.g., reducing sugars and ascorbic acid) can also interfere with the
reaction by reducing the reagent [9, 10].
2,2-Diphenyl-1-picrylhydrazyl (DPPH·) is a stable radical that
can be neutralized by accepting either a hydrogen atom or an
electron from antioxidant molecules, and thus the initial purple
color gradually decolorizes into pale yellow. The long reaction
time (30 min) and incubation at room temperature allow even
weakly antioxidant or thermolabile molecules to be tested. How-
ever, DPPH· radical is only soluble in organic solvents, tends to
react with other radicals, and is sensitive to Lewis bases and light
[9, 10]. The 2,2-azino-bis(3-ethylbenzothiazoline-6-sulphonic
acid) (ABTS·+) radical is obtained by reacting ABTS with ammo-
nium or potassium persulfate. When the ABTS·+ radical accepts an
electron from PCs, the color changes from blue-green to pale blue.
The reaction time may range from 1 to 30 min. The ABTS·+ radical
is soluble in both organic and aqueous solvents, and it is stable for a
long period in the dark [9, 10].
The FRAP assay is based on the reduction of ferric-
tripyridyltriazine (Fe3+-TPTZ) at low pH to an intense blue color
 Phenolics and Antioxidant Activity 169

ferrous-tripyridyltriazine complex (Fe2+-TPTZ). However, this

assay is non-specific, and any compound with a suitable redox
potential will cause Fe3+-TPTZ reduction. The cupric reducing
antioxidant capacity (CUPRAC) assay is conceptually similar to
the FRAP assay but is based on the reduction of Cu2+ ions in the
presence of 2,9-dimethyl-1,10-phenanthroline at pH 7 [9, 10].

2 Materials

2.1 Free and Bound • Ultrasonic bath.

PCs Extraction • Rotary evaporator.
• Sodium hydroxide.
• Nitrogen gas.
• Ethanol.
• Methanol.
• Hydrochloric acid.
• Diethyl ether.
• Ethyl acetate.

2.2 Folin–Ciocalteu • Folin-Ciocalteu reagent.

Assay • Distilled water.
• Sodium carbonate.
• Cuvettes (4 mL).
• Spectrophotometer.
• Ferulic acid.

2.3 DPPH· Radical • Methanol.

Scavenging Assay • DPPH.·.
• Distilled water.
• Cuvettes (1 mL).
• Spectrophotometer.
• Trolox.

2.4 ABTS·+ Radical • ABTS.

Scavenging Assay • Distilled water.
• Potassium persulfate.
• Sodium acetate.
• Microplates.
• Microplate reader.
• Trolox.
170 Rosanna Latronico and Pasquale Filannino

2.5 FRAP Assay • 2,4,6-Tripyridy-s-triazine (TPTZ).

• Hydrochloric acid.
• Distilled water.
• Iron trichloride (FeCl3).
• Acetate buffer (300 mM, pH 3.6).
• FeSO4.
• Microplates.
• Microplate reader.

3 Methods

3.1 Free and Bound • Extract twice dried bread (4 g) with methanol/water (70:30 v/
PCs Extraction v) in an ultrasonic bath for 10 min.
• Collect the supernatants, evaporate the solvent at 40 °C in a
rotary evaporator, and reconstitute the free PCs with 2 mL of
methanol/water (70:30 v/v). Store the extract at -20 °C
until use.
• Digest the residues of free PCs extraction with 300 mL of
NaOH 2 M at room temperature overnight by shaking under
nitrogen gas. At the end of the incubation, acidify the mixtures
(pH 2–3) with hydrochloric acid and extract PCs with diethyl
ether/ethyl acetate (1:1 v/v). Pool the fractions and evaporated
to dryness at 40 °C in a rotary evaporator. Reconstitute bound
PCs in 2 mL of methanol/water (70:30 v/v) (see Notes 1
and 2).

3.2 Folin–Ciocalteu • Mix 0.2 mL of PCs extract (appropriately diluted in order to

Assay achieve absorbance values within the calibration curve range)
and 0.4 mL of Folin-Ciocalteu reagent (1:5 H2O) and wait to
equilibrate for 3 min (see Note 3).
• Add 2 mL of Na2CO3 (100 g/L) and allow reaction for 30 min
at 30 °C in darkness.
• Measure absorbance at 725 nm through a spectrophotometer.
• The amount of free or bound PCs was calculated using a ferulic
acid curve and expressed as μg ferulic acid equivalent/g of dried
sample.

3.3 DPPH· Radical • Mix 0.5 mL of DPPH· solution (0.2 mM) with 0.5 mL of PCs
Scavenging Assay extract (appropriately diluted in order to achieve absorbance
values within the calibration curve range).
• Allow reaction for 30 min in darkness at room temperature.
• Measures absorbance at 517 nm through a spectrophotometer.
 Phenolics and Antioxidant Activity 171

• The DPPH• scavenging activity was calculated using a Trolox

curve and expressed as μmol Trolox equivalent/g of dried sam-
ple (see Notes 3 and 4).

3.4 ABTS·+ Radical • Prepare the ABTS·+ solution by mixing (1:1 vol/vol) the ABTS
Scavenging Assay stock solution (7 mM in water) with potassium persulfate
(High-Throughput (2.45 mM).
Methodology for • Keep the mixture overnight in the dark and then dilute with
Microplates) sodium acetate (20 mM pH 4.5) to reach an absorbance of
0.70 ± 0.01 at 734 nm.
• Mix 198 μL of the solution with 2 μL of PCs extract (appropri-
ately diluted in order to achieve absorbance values within the
calibration curve range) and keep the mixture at room tempera-
ture in the dark for 30 min, before reading the absorbance at
734 nm through a microplate reader.
• The ABTS·+ scavenging activity was calculated using a Trolox
curve and expressed as μmol Trolox equivalent/g of dried sam-
ple (see Note 4).

3.5 FRAP Assay • Prepare the FRAP reagent solution by mixing 1 mL of TPTZ
(High-Throughput (10 mM) in 40 mM HCl, 1 mL of iron trichloride (FeCl3)
Methodology for (20 mM), and 10 mL of acetate buffer (300 mM, pH 3.6) (see
Microplates) Note 5).
• Mix 10 μL of PCs extract (appropriately diluted in order to
achieve absorbance values within the calibration curve range)
with 300 μL FRAP reagent.
• Keep the mixture at room temperature for 10 min.
• Measure the absorbance at 593 through a microplate reader.
• FeSO4 solution is used to construct the calibration curve.
Results are expressed as μmol Fe2+/g of dried sample (see Note
4).
• Trolox can also be used as positive control, and results can be
expressed as μmol Trolox equivalent/g of dried sample, calcu-
lated from a standard curve.

4 Notes

1. Since PCs content and their antioxidant capacity can be

affected by several factors (bread recipe, fermentation and bak-
ing conditions, etc.) the weight of bread sample and the dilu-
tion rate of the reconstituted PCs extract can be adjusted in
order to achieve absorbance values within the calibration curve
range.
2. The methods for evaluating the antioxidant capacity in vitro
depend on the extraction conditions, which can vary among
172 Rosanna Latronico and Pasquale Filannino

different laboratories and implemented protocols. Novel pro-

tocols have been proposed that allow the evaluation to be
performed directly on the solid food, without any extraction
step [14].
3. The inter-laboratory reproducibility of these in vitro assays may
prove to be low, due to the use of heterogeneous protocols. To
overcome this problem, it may be useful to refer to official
methods [15, 16].
4. The estimated antioxidant activity can differ depending on the
assay used and is susceptible to interference. For instance, the
DPPH· and FRAP assays were shown to be less sensitive for
kaempferol glycosides. Moreover, the ABTS· assay is particu-
larly suitable for colored extracts because the wavelength
absorption at 734 nm can remove the color interference.
Thus, to prevent inaccurate interpretation of radical scavenging
activity, it is recommended to perform different assays in order
to compare the results obtained [17, 18].
5. When preparing the FRAP reagent, scientists recommend to
follow a specific order: first, add the acetate buffer, next, FeCl3,
and finally, TPTZ. This order is important to prevent TPTZ
from reducing FeCl3 [9].

References
1. G€anzle MG (2014) Enzymatic and bacterial 7. Lin S, Jin X, Gao J, Qiu Z, Ying J, Wang Y,
conversions during sourdough fermentation. Dong Z, Zhou W (2022) Impact of wheat bran
Food Microbiol 37:2–10 micronization on dough properties and bread
2. Filannino P, Di Cagno R, Gobbetti M (2018) quality: part II–Quality, antioxidant and nutri-
Metabolic and functional paths of lactic acid tional properties of bread. Food Chem 396:
bacteria in plant foods: get out of the labyrinth. 133631
Curr Opin Biotechnol 49:64–72 8. Pellegrini N, Vitaglione P, Granato D,
3. Schefer S, Oest M, Rohn S (2021) Interactions Fogliano V (2020) Twenty-five years of total
between phenolic acids, proteins, and antioxidant capacity measurement of foods and
carbohydrates—influence on dough and bread biological fluids: merits and limitations. J Sci
properties. Foods 10(11):2798 Food Agric 100(14):5064–5078
4. Arora K, Ameur H, Polo A, Di Cagno R, Riz- 9. Bibi Sadeer N, Montesano D, Albrizio S,
zello CG, Gobbetti M (2021) Thirty years of Zengin G, Mahomoodally MF (2020) The ver-
knowledge on sourdough fermentation: a sys- satility of antioxidant assays in food science and
tematic review. Trends Food Sci Technol 108: safety—chemistry, applications, strengths, and
71–83 limitations. Antioxidants 9(8):709
5. Tlais AZA, Fiorino GM, Polo A, Filannino P, 10. Amorati R, Valgimigli L (2015) Advantages
Di Cagno R (2020) High-value compounds in and limitations of common testing methods
fruit, vegetable and cereal byproducts: an over- for antioxidants. Free Radic Res 49(5):
view of potential sustainable reuse and exploi- 633–649
tation. Molecules 25(13):2987 11. Gil-Martı́n E, Forbes-Hernández T,
6. Dinardo FR, Minervini F, De Angelis M, Romero A, Cianciosi D, Giampieri F, Battino
Gobbetti M, G€anzle MG (2019) Dynamics of M (2022) Influence of the extraction method
Enterobacteriaceae and lactobacilli in model on the recovery of bioactive phenolic com-
sourdoughs are driven by pH and concentra- pounds from food industry by-products. Food
tions of sucrose and ferulic acid. LWT 114: Chem 378:131918
108394
 Phenolics and Antioxidant Activity 173

12. Perri G, Coda R, Rizzello CG, Celano G, 15. AOAC 2012.04-2012 antioxidant activity in
Ampollini M, Gobbetti M, De Angelis M, foods and beverages
Calasso M (2021) Sourdough fermentation of 16. AOAC 2017.13-2017 Total phenolic content
whole and sprouted lentil flours: in situ forma- in extracts
tion of dextran and effects on the nutritional, 17. Klopsch R, Baldermann S, Voss A, Rohn S,
texture and sensory characteristics of white Schreiner M, Neugart S (2019) Narrow-
bread. Food Chem 355:129638 banded UVB affects the stability of secondary
13. Nissen L, Samaei SP, Babini E, Gianotti A plant metabolites in kale (Brassica oleracea var.
(2020) Gluten free sourdough bread enriched sabellica) and pea (Pisum sativum) leaves being
with cricket flour for protein fortification: anti- added to lentil flour fortified bread: a novel
oxidant improvement and Volatilome charac- approach for producing functional foods.
terization. Food Chem 333:127410 Foods 8(10):427
14. Gökmen V, Serpen A, Fogliano V (2009) Direct 18. Li HB, Wong CC, Cheng KW, Chen F (2008)
measurement of the total antioxidant capacity of Antioxidant properties in vitro and total phe-
foods: the ‘QUENCHER’approach. Trends nolic contents in methanol extracts from
Food Sci Technol 20(6–7):278–288 medicinal plants. LWT 41(3):385–390
 INDEX

A 2,2-diphenyl-1-picrylhydrazyl (DPPH).............. 168–172

DNA extraction............................................................... 47
ABTS·+, 2,2-azino-bis(3-ethylbenzothiazoline-6- DNA library........................................................ 35, 46, 49
sulphonic acid)................................. 168, 169, 171 Dough yield (DY) ................................................ 6, 25, 78
Acetic acid............................................. 55, 61–68, 85, 86,
128, 145 E
Alkaline phosphatase (ALP) ................................ 157–160
Amino acid analyzer ........................................... 73, 77, 78 Electrophoretic techniques............................................. 31
Ammonium molybdate............................... 157, 158, 163 Enumeration.................................................................... 18
Antioxidant activity ..................................... 167, 168, 172 Enzymatic hydrolysis ...................................................... 84
Aroma .................................. 12, 120, 122, 128–130, 165 Enzymatic treatment......................................90, 146, 153
Essential amino acids index (EAAI)............................. 142
B
F
Backslopping ..................................................4–6, 8, 9, 11
Bioinformatics resources................................................. 46 Fermentation quotient.................................................... 66
Biological value (BV) .................................................... 142 Ferric ion reducing antioxidant power
Breadmaking........................................4, 6, 7, 10–12, 165 (FRAP)...................................................... 168–172
Bread staling .................................................................... 82 Flavour ...................................................... 4, 11, 119, 120,
Bread volume .......................................... 12, 98, 114, 115 122, 123, 128, 132, 165
Folin–Ciocalteu .................................................... 168–170
C Free amino acids (FAAs).................................. 71–77, 120

Cadmium-ninhydrin ....................................................... 75 G
Carbohydrates ............................................ 18–20, 44, 73,
81–84, 145, 146, 155, 166, 167 Genes annotation ............................................................ 46
Cation exchange chromatography ................................. 73 Glucose ................................................. 18–21, 73, 82–86,
Cell density ...................................................4, 24–26, 115 88, 92, 98, 145–147, 149, 150, 153
Colonies........................................................17, 18, 23–26
H
Colorimeter ........................... 73, 86, 156, 160, 161, 167
Colorimetry ................................................................... 156 Headspace gas chromatography - mass
Crumb cells ................................................................... 112 spectrometry ............................................. 130–132
Crumb structure .................................................. 111–115 High performance liquid chromatography
Culture-dependent estimation ....................................... 17 (HPLC).......................62–67, 72, 83–85, 87, 146
Culture-independent estimation ..............................29–40 High-throughput sequencing (HTS) ............... 35, 39, 44
Homogenization ......................................... 17, 21, 24, 25
D
I
Data processing ............................................................... 44
Denaturing gradient gel electrophoresis Illumina genome analyzer .............................................. 35
(DGGE) .........................................................30–34 Image analysis...............................................112, 114–116
Dietary fiber (DF) ....................................... 82, 83, 86, 91 Inoculum .............................................................. 5, 6, 8, 9
Digestible indispensable amino acid score Inorganic phosphate ............................................ 156, 157
(DIAAS).................................................... 138, 142 In vitro digestion........................136, 137, 141, 146, 147

Marco Gobbetti and Carlo Giuseppe Rizzello (eds.), Basic Methods and Protocols on Sourdough, Methods and Protocols in Food
Science , [Link]
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer
Nature 2024

175
 BASIC METHODS AND PROTOCOLS ON SOURDOUGH
176 Index
L Pyrosequencing .........................................................35, 37

Lactic acid ........................................... 3–6, 10, 11, 17–20, R

24–26, 30, 32, 34, 35, 40, 55, 61, 71, 72, 75,
81–83, 128 Radical scavenging activity ........................................... 172
Lactic acid bacteria (LAB) ..........................17–20, 24–26, Rapeseed displacement ...........................................98–100
29, 30, 32, 34, 35, 40, 45, 55, 61, 62, 71, 72, 75,
S
81–83, 128, 166, 167
Sabouraud dextrose agar ..........................................21, 23
M Sampling ...........................................................17, 21, 129
Malt extract agar ............................................................. 21 SDB agar....................................................................20, 23
Maltose ............................................................... 18–20, 85 Sensory attributes................................................. 120–123
Metabolic pathways..................................... 29, 40, 44, 46 Sensory profile................................................12, 120, 127
Metagenomic................................................34–40, 43–50 Serial dilutions.................................................... 17, 24, 25
M17–glucose agar .....................................................20, 21 Shotgun metagenomic..............................................46, 48
Mineral bioavailability................................................... 155 Shotgun sequencing........................................................ 44
Monosaccharides .......................................................84, 92 Simulator of the Human Intestinal Microbial Ecosystem
MRS agar ......................................................................... 18 (SHIME)............................................................ 141
MRS 5 agar................................................................18–20 Sourdough.............................................. 3–12, 17–19, 21,
Myo inositol ...................................................155–157, 163 22, 24–26, 29, 30, 32–34, 37, 39, 40, 43–45, 48,
55–59, 61–63, 65, 66, 71–73, 75, 78, 81–84, 88,
N 89, 92, 97, 98, 113–115, 119–121, 123, 124,
127, 128, 130–132, 135–137, 142, 145, 146,
Nutritional index (NI) .................................................. 143 152, 156, 157, 159, 161, 163, 166, 167
Specific volume....................................................... 97, 100
O
Starch ........................................................ 81, 82, 84, 104,
Oligosaccharides................................................. 83, 85, 92 145–147, 149–153
Starch hydrolysis index ................................................. 146
P
T
Panel test ....................................................................... 123
Peptides ................................................... 71, 72, 137, 142 Taxa abundances ............................................................. 43
pH ................................................................. 9, 20, 55, 61, Taxonomy..................................................................44, 47
73, 85, 147, 158 Texture....................................................4, 12, 25, 59, 82,
Phenolic compounds (PCs).......................................... 165 97, 103, 105–109, 112, 114, 120, 121, 123
pH-meter ................................................. 9, 55–59, 63, 92 Texture profile analysis (TPA) ............................. 103–109
Phytase .................................................................. 155–160 Titratable acidity (TTA).............................. 55, 56, 58, 59
Phytic acid ....................................................155–161, 163 Type-I sourdough .............................................. 4–6, 8–11
Plate count....................................................................... 17
Post-column derivatization ............................................ 73 V
Predicted glycemic index (pGI) ................................... 146 Volatile organic compounds (VOCs).................. 120, 127
Propagation ................................................. 4, 5, 8, 29, 39
Protein digestibility .............................135, 137, 138, 165 Y
Protein digestibility corrected amino acid score
(PDCAAS) ................................................ 138, 142 Yeasts................................................... 3–8, 10, 12, 17–20,
Protein efficiency ratio (PER) ...................................... 143 24–26, 29–32, 34, 35, 43–45, 62, 81, 82, 114,
Protein nutritional value ...................................... 141–143 115, 119, 121–124, 128, 152
Proteolysis.........................................................71, 72, 120