C H A P T E R

19
Applications of High Performance
Liquid Chromatography in the Analysis
of Herbal Products
Satyajit D. Sarker, Lutfun Nahar
Medicinal Chemistry and Natural Products Research Group, School of Pharmacy and Biomolecular Sciences,
Faculty of Science, Liverpool John Moores University, Liverpool, United Kingdom

O U T L I N E

19.1 Introduction 406 19.5.1 Specific Examples of Step-by-step

Protocols: Single and Multiple
19.2 High Performance Liquid Chromatography
Components Analysis Using Markers 414
(HPLC) 406 [Link] Vidanga (Embelia ribes) 415
19.2.1 Sample Preparation Techniques 406 [Link] Liquorice (Glycyrrhiza glabra) 415
19.2.2 Modes of Separation: Stationary and [Link] Valerian (Valeriana officinalis) 415
Mobile Phases 407 [Link] Compound Hongdoushan
[Link] Normal-Phase HPLC (NP- Capsule 416
HPLC) 407 [Link] Menoprogen Capsules 416
[Link] Reversed-Phase HPLC (RP- [Link] Arjunarishta 417
HPLC) 407 19.5.2 Specific Examples of Step-by-step
[Link] Other HPLC Modes 408 Protocols: Chemical Fingerprinting 417
19.2.3 Isocratic versus Gradient Elution 408 [Link] Curculiginis Rhizoma (Curculigo
19.2.4 Detectors 409 orchioides) 417
19.2.5 Data Analysis 409 [Link] Qianghuo (Notopterygium forbesii
19.2.6 Chemometrics and Principal Component or N. Incium) 418
Analysis [Link] Ayurved Siriraj
409
PrasachandaengdAn
19.3 Herbal Products 410 Antipyretic Product 419
19.3.1 Monoherbal Products 411 [Link] Radix Aconiti (Roots of Aconitum
19.3.2 Polyherbal Products 411 carmichaeli) 419
[Link] Folium Turpiniae 420
19.4 Types of Analysis of Herbal Products 411 [Link] Dioscorea zingiberensis 421
19.4.1 Marker-Based Authentication and Quality [Link] Yuanhu Zhitong Tablet 422
Control of Herbal Products 413
19.4.2 Chemical Profiling or Fingerprinting 19.6 Conclusions 423
of Herbal Products 413 References 423
19.5 Analysis of Herbal Products by High List Of Abbreviations 425
Performance Liquid Chromatography 414

Evidence-Based Validation of Herbal Medicine

[Link] 405 Copyright © 2015 Elsevier Inc. All rights reserved.
406 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

19.1 INTRODUCTION HPLC techniques in the analysis of various herbal prod-

ucts and will present several specific examples of proto-
High performance liquid chromatography (HPLC), cols of such analysis. However, a brief overview of
also known as high pressure liquid chromatography, is available HPLC techniques and methods will be
one of the most popular, modern, powerful, and versa- presented. The examples of analyses of herbal products
tile chromatographic separation techniques that have will primarily be limited to commercially available
been routinely used to separate the components in a formulated products, not necessarily any crude herbs
mixture (e.g., an herbal extract or product), to identify or herbal materials.
each components (or at least as many components as
possible), to quantify separated components, and to
obtain the chemical profile or fingerprint of a crude 19.2 HIGH PERFORMANCE LIQUID
mixture. Any HPLC system fundamentally comprises CHROMATOGRAPHY (HPLC)
a pump (or pumps) to pass a pressurized (50e400 bar)
mobile phase (solvent) through a compact column (typi- HPLC is the most widely used analytical separation
cally 2.1e4.6 mm diameter and 30e250 mm length) technique for qualitative and quantitative determination
filled with a stationary phase, e.g., reversed-phase C18 of compounds in natural products extracts or fractions
silica (typically 2e5 mm particle size), a sample injection as well as finished products. For any analysis of herbal
system (manual or autosampler), at least one suitable products, HPLC, coupled with various modern detec-
detector, e.g., photo-diode-array (PDA) detector, to tion techniques, has now become the method of choice
monitor the separation of any sample mixture, and and is often recommended by various regulatory
nowadays, a PC-based software to control the whole authorities working towards ensuring the quality, effi-
operation and data processing (Figure 19.1). Depending cacy, and safety of herbal products. However, an
on the demand of any particular experiment, a fraction HPLC operation requires careful consideration of a
collector, a column chamber (thermostat or column number of important factors, starting from sample prep-
oven), an online degasser, or a few other additional aration to the choice of appropriate detection technique
modules can also be added to a standard HPLC system. and data processing modes.
While this chapter is not really intended to provide the
details of the composition and function of any HPLC
system, it will certainly focus on the applications of
19.2.1 Sample Preparation Techniques
The fundamental steps in sample preparation for
HPLC analysis are complete dissolution of sample in
Mobile phase reservoir the eluent and filtration of the sample solution using
microfilters (usually 0.45 mm). The quality of the
outcome from any HPLC analysis often depends on
the sample preparation technique used [1e5].The choice
of a sample preparation technique for HPLC analysis
Pumps
depends largely on the nature of samples to be analyzed,
and the HPLC mode to be employed. For example, if the
sample matrix is soluble in the mobile phase, a simple
Manual injector or preparation of sample solution is suitable for HPLC
Autosampler analysis. However, if the sample is not readily soluble
in the mobile phase, for example, powder of an herb, a
suitable extraction protocol will have to be employed
before it can be analyzed by HPLC. The chosen extrac-
Column tion procedures should be rapid but efficient, and they
(With or without a guard column)
should include the total entity of the low molecular
constituents of an herbal product. This is usually
achieved, in most cases, using methanol (MeOH)
or ethanol (EtOH). Additional fingerprints can be
obtained by extraction with petroleum ether/n-hexane
or chloroform (for lipophilic compounds) or water/
Detector Ourput terminal water-acetone mixtures (for tannins, high polymeric
Waste solvent (UV-Vis, PDA or MS) (PC or chart recorder)
procyanidines, and amino acids) as solvents. Polysac-
FIGURE 19.1 HPLC schematic. charides and proteins can be characterized using their
 19.2 HIGH PERFORMANCE LIQUID CHROMATOGRAPHY (HPLC) 407
sugar- or amino acid-fingerprints after enrichment and TABLE 19.1 Commonly Used Stationary Phases and
acidic or enzymatic hydrolysis. Their Associated Modes in HPLC
Depending on the nature of the active ingredient(s)
Stationary phases Modes
used in any formulations or products, e.g., tablets and
capsules, either a simple dissolution or an extraction C6 silica Reversed-phase
step followed by filtration must be introduced prior to C8 silica Reversed-phase
HPLC analysis. A suitable sample preparation step is
C18 silica Reversed-phase
particularly important for the analysis of herbs and
herbal products, because it is necessary to extract the Silica Normal phase
desired chemical components from the herbal materials Diol Normal and reversed-phase
for further separation and characterization. Modern and
Cyano (CN) Normal and reversed-phase
classical sample preparation techniques [4e6] may
include one or more of the following: Benzene sulphonic acid Strong cation exchange

1. Solvent immersion. Polystyrene Size exclusion

2. Solvent partitioning.
3. Refluxing. hydrogen bonding or dipoleedipole type interactions,
4. Ultrasonic extraction. with the sorbent surface. Generally, polar compounds
5. Solid-phase microextraction. in the mixture being passed through the column will
6. Supercritical-fluid extraction. be adsorbed more strongly to the polar silica than
7. Pressurized-liquid extraction. nonpolar compounds, which will pass more quickly
8. Microwave-assisted extraction. through the column and will be eluted quicker than
9. Solid-phase extraction. the polar ones. Increasing the polarity of the eluting
10. Surfactant-mediated extraction. solvent will decrease the elution time, i.e., shorter reten-
An excellent overview on various modern sample tion time.
preparation methods for the extraction, cleanup, and NP-HPLC is suitable for separating lipophilic
concentration of analytes from medicinal plants or herb- compounds, long-chain alkanes, or compounds that
al products for HPLC analysis has been presented by are sparingly soluble in aqueous conditions. As the
Huie [7]. Generally, an ideal sample preparation tech- interaction of various compounds with the stationary
nique should be safe, cost-effective, reproducible, not phase not only depends on the functional groups
too time-consuming, able to extract maximally the target present in the structure of the analyte molecule, but
compounds without changing their structures, and able also on stereochemical features, the effect of steric
to exclude other compounds to minimize any unwanted hindrance on interaction strength allows NP-HPLC
interferences. methods to resolve geometric and positional isomers
[9]. Since the introduction of reversed-phase HPLC
(RP-HPLC) during 1970, NP-HPLC has been super-
19.2.2 Modes of Separation: Stationary and seded by RP-HPLC. The NP-HPLC mobile phase
Mobile Phases consists of a nonpolar solvent such as n-hexane or n-hep-
An HPLC analysis of any crude mixture (e.g., herbal tane mixed with a slightly more polar solvent such as
products) generally requires a normal-phase, reversed- isopropanol, chloroform, or ethyl acetate [10]. Table
phase gel permeation (size exclusion) or an ion exchange 19.2 presents a list of commonly used eluents in
chromatographic mode [8,9]. The use of any particular NP-HPLC. There are four main factors, i.e., solvent
mode relies on the compatibility of the stationary phase strength, localization, basicity, and UV cutoff, involved
with the extract or mixture to be analyzed and also the in the choice of solvents for NP-HPLC.
chemical nature of the crude mixture. Commonly used
stationary phases and their associated modes are [Link] Reversed-Phase HPLC (RP-HPLC)
presented in Table 19.1. Reversed-phase HPLC (RP-HPLC) is the most
commonly used mode of HPLC and, as the name
[Link] Normal-Phase HPLC (NP-HPLC) implies, this mode is just the reverse of NP-HPLC,
Normal-phase HPLC (NP-HPLC), which is not the whereby the stationary phase is more nonpolar than
most popular form of HPLC nowadays, utilizes a polar the eluting solvent. Generally, RP-HPLC has a nonpolar
stationary phase (usually silica) and less polar stationary phase, e.g., C18 silica (Table 19.1), and a
(nonaqueous) eluting solvents, e.g., n-hexane and ethyl moderately polar aqueous mobile phase [8,10]. One
acetate (mobile phase). The separation is based on the common RP-HPLC stationary phase is surface-modified
analyte’s ability to engage in polar interactions, e.g., silica, RMe2SiCl, where R is a straight chain alkyl group
408 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

TABLE 19.2 Commonly Used Mobile Phases in NP-HPLC solvent, and this slows them down on their way through
the column, which means longer retention time. In
Mobile Strength UV-cut-
RP-HPLC the polar molecules travel through the
phases (ε0) Localization Basicity off (nm)
column more quickly.
n-Hexane 0.0 No Not relevant 195 RP-HPLC allows purification of most classes of com-
n-Heptane 0.0 No Not relevant 200 pounds, including compounds present in various herbal
products, and is often the most preferable choice when
Chloroform 0.26 No Not relevant 245
analyzing and attempting to separate and identify
Dichloromethane 0.30 No Not relevant 233 compounds from a complex mixture [12].
Diethylether 0.38 Yes Yes 218
[Link] Other HPLC Modes
Ethylacetate 0.48 Yes No 256
Gel permeation chromatographic HPLC mode (also
Tetrahydrofuran 0.53 Yes Yes 212 called size exclusion chromatography) is generally used
Isopropanol 0.60 Yes Proton 210 for separation of large molecules, e.g., proteins and oligo-
donor saccharides. The stationary phase is made of inherently
Methanol 0.70 Yes Proton 205 hydrophobic and chemically and physically inert poly-
donor styrene/divinylbenzene copolymers (Table 19.1). The
pore size in the particles is strictly controlled, and
compounds are separated by their ability to enter the
such as C18H37 or C8H17. Silica-based reversed-phase pores; the smaller molecules are “trapped” temporarily
sorbents are also known as “bonded-phase” materials in the pores, while larger molecules are not held up
[8,11]. The eluent used in RP-HPLC is usually composed and pass through the column relatively unhindered [9].
of a mixture of water and miscible organic solvents, Ion exchange HPLC utilizes an anionic or cationic
usually acetonitrile (ACN), MeOH, or tetrahydrofuran stationary phase for the separation of acids and amines.
(THF) [9,10] (Table 19.3). Sometimes, buffers, acids, or Compounds with a net charge bind reversibly to the
bases are also added to suppress compound ionization ionizable groups on the stationary phase and are eluted
or to control the degree of ionization of free unreacted through displacement of a stronger ionized species in
silanol groups to reduce peak tailing and improve chro- the eluent [9].
matography [9].
In RP-HPLC there is strong attraction between the
polar solvent and polar molecules in the mixture being
passed through the column, but there is not much attrac- 19.2.3 Isocratic versus Gradient Elution
tion between the hydrocarbon chains attached to the Isocratic analysis is the analysis where the mobile
silica (the stationary phase) and the polar molecules in phase composition, as well as the flow rate, is kept
the solution. Therefore, polar molecules in the mixture constant throughout the total HPLC run, and any re-equi-
spend most of their time moving with the solvent. libration of the column is not required. On the other
Nonpolar compounds in the mixture tend to form attrac- hand, in gradient elution, the mobile phase composition
tions with the hydrocarbon groups because of van der changes over time, either gradually with a steady change
Waals dispersion forces. They are less soluble in the in composition (linear gradient), or a stepwise change in
solvent because of the need to break hydrogen bonds composition (step gradient) [13e15]. While an isocratic
as they squeeze in between the water or methanol elution is suitable for separating a mixture of a few com-
molecules. They spend less time in solution in the pounds, a gradient elution can handle a mixture of
several compounds of wide-ranging polarities [15]. In a
TABLE 19.3 Commonly Used Mobile Phases in gradient operation, the sample is injected onto the col-
RP-HPLC umn in high aqueous conditions, and the organic propor-
tion is increased over time to elute all the compounds
Mobile Polarity index UV-cutoff off the column. For a typical analytical column (4.6 mm
phases (Snyder) (nm)
i.d.
150 mm), the typical flow rate is 1.0 mL/min with
Acetonitrile 6.2 190 the gradient time taking 30 min with a hold at the end
Isopropanol 4.3 210
of the gradient to ensure all compounds have been
eluted [9]. However, the run time can be significantly
Methanol 6.6 205 reduced (five minutes as opposed to 30 min) in the
Tetrahydrofuran 4.2 212e230 recently introduced ultra performance liquid chromatog-
Water 9.0 180
raphy (UPLC), where the diameter of the column and the
particle size of the stationary phase are much smaller
 19.2 HIGH PERFORMANCE LIQUID CHROMATOGRAPHY (HPLC) 409
than conventional analytical HPLC columns [16,17]. In the solvent and buffer used as the mobile phase, the flow
UPLC, the separation efficiency and resolution can be rate, and, of course, the type of interface used, and thus,
improved because of the use of much-reduced particle it often creates difficulties in relation to reproducibility
size in the UPLC columns [18,19]. of information [23].
NMR, albeit probably the least sensitive of all detec-
tion techniques, is also used as a detector for HPLC as
19.2.4 Detectors it offers the most useful structural information towards
A suitable detection device (detector) is used with the structure elucidation of natural products [20]. Often,
an HPLC to detect or monitor compounds separating a UVeVis detector is also used as a primary detector
out while eluting from a column. The most common when an NMR detector is used. NMR has not quite
detector is an UltravioleteVisible (UVeVis) detector. become a widely accepted detector for any HPLC oper-
However, hyphenation between an HPLC and more ation, mainly because of its lower level of sensitivity,
sophisticated detection techniques, e.g., mass spectrom- need for deuterated solvents, and higher cost compared
eter (MS) or nuclear magnetic resonance (NMR) spec- to other available detectors.
trometer, has increased the capability of separating Among the other types of detectors, an evaporative
and solving structural problems of complex natural light scattering detector (ELSD), an infrared (IR) detec-
products [20]. Sometimes, multiple detection techniques tor, an electrochemical detector, and a fluorescent detec-
are employed, e.g., LC-UV-Vis-MS, LC-MS-MS, and tor are also used in special cases. For example, saponins
LC-NMR-MS. with no chromophores may not be detected by conven-
A UVeVis spectroscopic detector is a universal detec- tional UVeVis or PDA, but can be successfully detected
tor for any HPLC system. The photo-diode-array (PDA) by the ELSD. As one-way separation, single detection
detector is another form of UVeVis detector that allows method, or data processing cannot fulfill the needs of
analysis of compounds containing chromophores, the comprehensive quality control of herbal products,
coumarins, flavonoids, isoflavonoids, etc. A PDA detec- multidimensional information-based HPLC technolo-
tor facilitates the analysis of individual HPLC peaks gies (e.g., two dimensional HPLC), use of multidetectors,
after completion of a run and helps to obtain complete and HPLC fingerprinting coupled with multicomponent
UVeVis spectrum of individual compounds. Indepen- quantification can fulfill this demand [24].
dent chromatograms can also be constructed at various
wavelengths to enhance selectivity of the data. PDA is
one of the most popular detectors, but compounds 19.2.5 Data Analysis
with no chromophores are not UVeVis sensitive, and
Most of the data analyses are now performed through
cannot be detected.
various PC-based software that normally come with
Nowadays, an MS detector, generally in combination
HPLC systems, e.g., Chromeleon for Dionex HPLC
with a UVeVis or PDA detector, is probably the most
systems. However, for any HPLC analysis, the retention
sought-after detection method employed for HPLC,
time, peak area, peak height, UVeVis absorption
especially when analyzing complex mixtures of com-
maxima, MS data, and, in some cases, NMR data are
pounds like natural product extracts or herbal products.
generally analyzed for identification and/or quantifica-
When an MS detector is used, separated compounds
tion of separated components from a complex mixture.
emerging from the column can be identified on the basis
Data obtained from HPLC runs can also be analyzed
of their mass spectral data. The ionization techniques
for chemical fingerprinting analysis of various natural
used in HPLC-MS are generally soft ionization tech-
product extracts, including herbal products, especially
niques, e.g., electrospray ionization mass spectrometry
in relation to quality control and authentication
(ESI-MS) that display mainly the molecular ion species
purposes. With the tremendous progress in electronics
with only a few fragment ions are generally used in
and computational techniques, the capabilities of avail-
HPLC-MS. Sometimes, tandem mass spectrometry
able HPLC software have enhanced significantly in
(MSeMS), which provides fragments through colli-
recent years, and various types of data analysis can
sion-induced dissociation of the molecular ions, is also
now be easily performed.
employed [20].
An HPLC-MS system does not necessarily allow a
complete and unambiguous online identification of a 19.2.6 Chemometrics and Principal Component
component, unless it is a well-known natural product
and complementary online spectroscopic information
Analysis
is available in databases for comparison [20e22]. The Chemometrics is not a single tool but a range of
quality of MS response invariably depends on a number methods including basic statistics, signal processing,
of factors, e.g., nature of the compounds to be analyzed, factorial design, calibration, curve fitting, factor
410 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

analysis, detection, pattern recognition, and neural

network. Chemometrics, first introduced by the Swedish
scientist Svante Wold in 1971, is simply the application Time series
of mathematical and statistical techniques to retrieve analyses
more information from the chromatographic data. This Mulyivariate Information
term has been officially defined as the science of relating statistics & technology
pattern
measurements made on a chemical system or process to recognition
the state of the system via application of mathematical or
statistical methods [25e30]. Various areas and principles
that contribute to chemometrics are shown in
Figure 19.2. Chemometrics
Computation
Chemometrics is the tool for extracting information
Decision theory
from multivariate chemical data using tools of statistics
and mathematics. With the advance of computational
techniques, chemometrics has become a leading tool
for faster analysis of results/data and shorter product
development time. It is generally applied for one or Operations Numerical
more of three primary purposes to research analysis

1. Explore patterns of association in data.

2. Track properties of materials on a continuous basis.
3. Prepare and use multivariate classification models. FIGURE 19.2 A flow diagram showing the generic steps involved
in the HPLC-based analysis of herbal products.
This tool has the capacity for analyzing and modeling
a wide variety of data types for an even more diverse set
pretreatment of data is required because of unknown
of applications. Chemometrics are basically classified
components, overlapped peaks, and drifted baselines
into two main categories [26]:
on the HPLC chromatogram. For herbal products data
1. Pattern recognition methods (unsupervised and analysis for quality control and standardization, in addi-
supervised) when a qualitative evaluation is tion to PCA as mentioned above, a number of other che-
considered. mometrics techniques may also be applied, such as
2. Multivariate calibration for quantitative purposes. linear discriminate analysis (LDA), spectral correlative
chromatography (SCC), information theory (IT), local
The design of the experiment, data preprocessing,
least square (LLS), heuristic evolving latent projections
classification, and calibration are the main practical
(HELP), and orthogonal projection analysis (OPA) [30].
steps involved in any chemometrics analysis. Experi-
The commonly used pretreatment of data obtained
mental design primarily screens factors that are impor-
from herbal products analysis are LLS and normaliza-
tant for the success of a process. Selection and
tion. Details on the chemometrics techniques for the
implementation of the optimized conditions under
analysis of herbal products are available in a couple of
which the process will be carried out come next.
excellent review articles published recently [30,32].
Principal component analysis (PCA) is an unsuper-
vised pattern recognition technique used for handling
multivariate data without prior knowledge about the
samples under investigation [31]. The supervised classi- 19.3 HERBAL PRODUCTS
fication procedure using soft independent modeling of
class analogy (SIMCA), based on making a PCA model Herbal medicine refers to medicine that utilizes
to assign unknown samples into the predefined class various plant parts, e.g., seeds, berries, roots, leaves,
model, is also used in chemometrics analysis [30]. The bark, flower buds, or flowers for treating human
central idea of PCA is to reduce the dimensionality of ailments. It is also known as botanical medicine or
a data set consisting of large amounts of interrelated phytomedicine. Herbal products are commercially
variables, yet keeping maximum variation in the data available nutrition (dietary supplements), cosmetics,
set. PCA is generally used to evaluate the discrimination health, or health-related products, including medicines
ability of common components using relative peak areas that contain one or more herbal components. An herbal
of common peaks as input data instead of the full product may be simply dried total plant parts, dried and
fingerprint. ground plant parts, crude extracts of any plant parts, or
In the analysis of herbal products, especially by in a more contemporary form of medicinal formulation,
HPLC, and chemometrics follow-up analysis, e.g., teas, tablets, capsules, solutions, or suspensions.
 19.4 TYPES OF ANALYSIS OF HERBAL PRODUCTS 411
The use of herbs for human consumptions, particu- capsules, is composed of two or more herbs or herbals
larly for medicinal purposes, goes back thousands of components. Most of the TCM products or Ayurvedic
years, as far as 60,000 years ago when the Neanderthal preparations often contain more than two herbal compo-
men appeared to have used herbs for treating ailments. nents. For example, a polyherbal formulation based on
Various forms of major healthcare systems, based on Euphorbia fusiformis (common name: Dhudmul) to in-
the use of herbs and herbal constituents, were developed crease the flow of breast milk in women utilizes 500 g
several thousand years ago; for example, the Traditional of roots of E. fusiformis made into a paste and 20e25
Chinese Medicine (TCM) and the Ayurvedic Medicine grains of Piper nigrum in the form of conventional tablets
systems both are about 5000 years old. It is obvious that [33]. Further examples of some polyherbal formulations
herbal medicine or herbal products have a long tradition are presented in Table 19.4.
of use outside of conventional medicine, and they have
always enjoyed enviable popularity, especially in the
developing world. It is well-known that, according to 19.4 TYPES OF ANALYSIS OF HERBAL
the estimation by the World Health Organization PRODUCTS
(WHO), over 80% of people worldwide rely on herbal
medicines for some part of their primary health care. Most of the herbal products analysis can be catego-
Over the last few decades, the steady growth of the rized into two major categories: (1) chemical profiling
herbal products market, especially in the developed or chemical fingerprinting and (2) authentication and
western countries, has been quite remarkable, and in quality control of herbal products to ensure efficacy
some cases, it has superseded the growth of the so-called and safety of the products. One of the characteristics of
“modern medicine” market. In Germany, about 600e700 herbal preparations or products is that, most often, herb-
plant-based medicinal products are available and are al products, either presenting as a single herb (monoher-
prescribed by a majority (over 70%) of German physi- bal) or as collections of herbs in composite formula
cians. In the past 20 years, in the USA, public dissatisfac- (polyherbal), are extracted with boiling water during
tion with the cost of prescription medications, combined the decoction process, with the exception of more mod-
with an interest in returning to natural or organic ern products (e.g., tablets or capsules of herbal medi-
remedies, has prompted an increase in demands for cines). This sometimes poses the difficulties in quality
herbal medicinal products. However, quality-related control or analysis of herbal products.
problems (lack of consistency, safety, and efficacy) Despite the common perception that herbal products
seem to be overshadowing the potential genuine health are generally safe because of their long-standing use in
benefits of various herbal products; thus, the quality various cultures, several serious adverse effects result-
herbal products can only be made effective and safe if ing from uses of herbal products have been well-docu-
proper standardization methods are followed right mented in the literature. Many of these adverse effects
from the cultivation/collection of the herb to packaging can be implicated to the presence of toxic contaminants,
and storage of the finished product. e.g., heavy metals, microorganisms, microbial toxins,
All herbal products can broadly be divided into two pesticides, and radioactive materials, and various adul-
classes: (1) monoherbal products and (2) polyherbal terations. Some of the herbal components used in the
products. herbal products may also be toxic themselves, especially
when their presence in the product is well above the rec-
19.3.1 Monoherbal Products ommended safety levels. Therefore, appropriate anal-
ysis of herbal products to ensure quality, and thus
Monoherbal products or preparations, as the name ensuring safety, is paramount. This may require compre-
indicates, contain only one herb or herbal component. hensive phytochemical analysis incorporating identifi-
For example, ginseng tea contains the powder of only cation and quantification of all or at least as many
ginseng roots, chamomile tea contains dried leaves of compounds as possible present in any herbal products.
Matricaria chamomilla, Curcuma capsules contain only The major objective of any herbal product analysis is
the powder of rhizomes of Curcuma longa, Milk Thistle generally aimed at overall standardization of the prod-
capsules contain standardized extract of only Milk uct, which is the first step for the establishment of any
Thistle (Silybum marianum), and Ginkgo biloba standard- consistent biological activity, a consistent chemical pro-
ized extract contains the extract of only Ginkgo biloba. file, or simply a quality assurance program for produc-
tion and manufacturing [45]. Standardization can be
defined as the means of adjusting the herbal products
19.3.2 Polyherbal Products to a defined content of a constituent or group of
Again, as the name implies, a polyherbal product substances with defined therapeutic activity. According
or preparation, e.g., decoction, tincture, tablets, and to the European Medicines Agency (EMEA), there is a
412 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

TABLE 19.4 Examples of Some Polyherbal Products

Polyherbal products Compositions Uses References

Arjunarishta Fermented decoction of the bark of Terminalia arjuna Cardioprotective [34]

(2040 g), fruits of Vitis vinifera (1020 g), flowers of
Madhuca indica (408 g), and flowers of Woodfordia
fruticosa (408 g).
Ayurved Siriraj Prasachandaeng Citrus aurantifolia, Bouea macrophylla, Knema globularia, Antipyretic drug [35]
Dracaena loureiri, Myristica fragrans, Caesalpinia sappan,
Conioselinum univitatum, Kaempferia galanga, Mesua
ferrea, Jasminum sambac, Mammea siamensis, and
Nelumbo nucifera
ContudolÒ capsule (1) Boswella serrata (Shallaki) extract (standardized for Improves mobility and reduces [36]
boswellic acid 60%), 200 mg, (2) Curcuma longa (Haldi) pain in the case of arthritis and
extract (standardized for curcumin 20%), 100 mg, (3) other musculoskeletal disorders.
Moringa oleifera (Sahjan) extract (standardized for
tannins 1%, 50 mg, and (4) Zingiber officinale (Sunthi or
Ada) extract (standardized for gingerol), 35 mg
Hongdoushan capsule Panax ginseng, Glycyrrhiza uralensis, and Taxus Chinensis To treat ovarian and breast [37]
cancers
Immunol Extracts of Asparagus racemosus (7.0 mg), Boerhaavaia To stimulate immune system [38]
diffusa (7.0 mg), Tinospora cordifolia (14.5 mg), Terminalia
chebula (3.5 mg), Trigonella foebum-graecum (4.8 mg),
Tylophora asthamatica (4.8 mg), and Withania somnifera
(7.0 mg)
Jessica Each 100 mg contains Emblica offocinalis (7.6 mg), Anxiolytic property: to help with [39]
Nardostachys jatamansi (77.0 mg), and Rauwolfia sleep and relaxation
serpentine (15.4 mg)

Maha Vartikawa Watee pills Composed of 29 herbal ingredients of equal quantities: To treat gastrointestinal disorders [40]
aerial parts of Myristica fragrans, bulbs of Allium satium,
exocarps of Acacia chundra and Ferula assa-foetida,
flower buds of Syzygium aromaticum, fruits of Carum
cavi, Cuminum cyminum, Piper nigrum, Piper longum,
and Trigonella foenum-graecum, heart-wood of Santalum
album, leaves of Piper betel, Trachyspermum ammi, and
Vitex nigundo, pollens of Fumaria parviflora, rhizomes of
Acorus calamus and Zingiber officinale, roots of Solanum
virginianum, seeds of Brassica jancea, Caesalpinia bonduc,
Embelica ribes, Myristica fragrans, Nigella sativa,
Seasamus indicum, stamen of Mesua ferrea, stem barks of
Cinnamomum verum, and total plant of Terminalia
chebula, Terminalia bellarica, and Phyllanthus emblica
Rasayana churna (1) Powder of dried stem of Tinospora cordifolia To enhance general body [41]
(Guduchi), (2) powder of dried fruits of Tribulus resistance, promote longevity, and
terrestris (Gokshur), and (3) powder of dried pericarps relieve stress.
of Emblica officinalis (Amlaki) in equal proportion.

Shrishadi Albezzia lebbeck, Cyprus rotan, and Solanum To treat infectious respiratory [42]
xanthocarpum disorders.

Tisane A Roots of Cocos nucifera and Coix lacryma-jobi, leaves of To treat rheumatoid arthritis [43]
Aphloia theiformis, Antidesma madagascariense, and
Bidens pilosa, and bark of Erythroxylum laurifolium
Triphala powder or tablets Ground dried pericarps (in equal proportion) of (1) For internal cleansing, purgation, [44]
Terminalia chebula (Haritaki), (2) Terminalia bellerica rejuvenation, and detoxification
(Bibhitaki), and (3) Emblica officinalis (Amlaki)
 19.4 TYPES OF ANALYSIS OF HERBAL PRODUCTS 413
clear distinction between constituents with known ther- 19.4.1 Marker-Based Authentication and
apeutic activity (which can be used to standardize any Quality Control of Herbal Products
biological effect) and marker compounds, which allow
standardization on a set amount of the chosen Authentication of the herbal products is the first step
compound. towards its quality control to ascertain efficacy and
A marker compound can be defined as the chemically safety [46]. An herbal product, whether it is a monoher-
defined constituent of an herbal product for quality bal or polyherbal, contains several chemicals, which
control purposes, irrespective of its direct link or not, provide them with specific chemical identities or charac-
to overall therapeutic property of the product. Valerenic teristics. So chemical constituents and their respective
acids in Valeriana officinalis, gingkolides and flavonoids quantities can be analyzed to authenticate, i.e., to
in G. biloba, and hypericin and hyperforin in Hypericum identify, an herbal product and differentiate the real
perfoliatum are classic examples of some marker product from a fake one [32,47,48].
compounds. Generally, one or two markers or pharmacologically
According to the EU definition, there are three active components in herbs or herbal products are
categories of herbal products: (1) those containing con- employed for evaluating the quality and authenticity
stituents (single or group of compounds) with known of herbal products, both in the identification and quanti-
and experienced therapeutic activity that is solely fication of single herb and in multicomponent prepara-
responsible for clinical efficacy; (2) those containing tions [30,45]. According to the definition provided by
chemically defined constituents possessing relevant the EMEA, chemical markers are the chemically defined
pharmacological properties, which are likely to constituents or groups of constituents of an herbal
contribute to the clinical efficacy; and (3) those in which medicinal product, which are of interest for quality
no constituents have been identified as being respon- control purposes regardless whether they possess any
sible for the therapeutic activity. All these three therapeutic activity.
categories need to be covered or considered when The presence and the quantity of a chemical marker are
analyzing any herbal product. A flow diagram to taken as primary indicators of the quality of any herbal
show the generic HPLC-based analysis of herbal prod- product. However, marker-based determination may not
ucts is shown in Figure 19.3. necessarily offer a detailed picture of an herbal product,
as multiple constituents are usually responsible synergis-
tically for its therapeutic efficacy. Also, the chemical con-
stituents in components herbal products may vary
Sample preparation depending on harvest seasons, plant origins, drying pro-
(Extraction or dissolution, and filtration) cesses, and other factors. This is why, nowadays, finger-
printing analysis of herbal products has become popular
to ascertain authenticity and quality of an herbal product.

19.4.2 Chemical Profiling or Fingerprinting

Preparation of standard solutions of Herbal Products
of marker compounds
One of the main modes of analysis of herbal products
is chemical profiling or chemical fingerprinting. Finger-
printing is a quality control tool that builds upon spec-
troscopic and chromatographic technology.
An HPLC-based chemical fingerprint refers to the com-
Analytical HPLC analysis
mon peaks of any HPLC chromatogram, which can be
used to characterize or authenticate any herbal product.
A fingerprint of herbal medicinal products is the profile,
which can illustrate the specific properties of the
finished product after appropriate processing and be
obtained by certain analytical techniques, e.g., HPLC.
A chemical fingerprint usually reflects the overall
Data processing and data analysis
(Fingerprinting and chemometric analysis) nature or “complete information” of the product. It
elevates the quality control target from analysis of a
single component to analysis of all the material in
FIGURE 19.3 Areas and principles that contribute to the product and thus may ensure consistency in
chemometrics. the therapeutic efficacy of the products. Simply,
414 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

fingerprinting can be defined as a type of comprehen- comprehensiveness” and “fuzziness,” which are the two
sive and quantifiable method for authentication; it is basic traits of any fingerprinting approach.
one of the quality control models that is suitable for eval- It has been shown that chemical fingerprinting or
uating the authenticity and stability of herbal products. chemical profiling can help with authentication and
It has now become a globally accepted quality evalua- identification of herbal products [49]. In this, generally
tion model for herbal medicinal products. Figure 19.4 an herbal product is viewed holistically, and the model
presents an overview of the fingerprinting process of of using only one or two marker components may not
herbal products using HPLC. necessarily be the best means of evaluating the quality
The idea of “phytoequivalence” was first introduced in of herbal products. It can be noted that any finger-
Germany in order to ensure consistency of herbal prod- printing protocol must be practical, specific, reproduc-
ucts [32]. As a result, a chemical profile, such as a chro- ible, and fully validated for its use in the quality
matographic fingerprint, for an herbal product under assurance or authentication of any herbal products.
evaluation was constructed and compared with the pro-
file of a clinically proven reference product. A chemical
fingerprint can be defined as a characteristic chemical 19.5 ANALYSIS OF HERBAL PRODUCTS
profile of a sample, which reflects the unique complex BY HIGH PERFORMANCE LIQUID
chemical composition of the sample. The characteristics CHROMATOGRAPHY
can be obtained by spectroscopic, chromatographic, elec-
trophoretic techniques, or a combination of all of them. Herbal products usually contain a large number of
An HPLC coupled with a suitable detector can provide compounds, many of which may be present at low con-
the required characteristic data set, e.g., retention time, centrations, yet are quite significant in terms of the over-
peak area, spectral data, etc., to be used for fingerprinting all quality, safety, and efficacy of the product. To assess
analysis or profiling. It is often expected that this profile is the complete pattern of any herbal product, several
featured by the fundamental attributions of “integrity or chromatographic techniques, particularly HPLC, offers
the desired separation of individual components and
thus constructs a characteristic profile of the sample,
Sample preparation: which is generally known as “chemical fingerprinting.”
Dissolution or extraction HPLC is undoubtedly the most popular separation tech-
nique used for the analysis of herbal products, especially
in relation to authentication, quantification, quality
control, fingerprinting, and standardization because of
its reliability, reproducibility, precision, resolution, selec-
Development of fingerprint: tivity and sensitivity, and ease of hyphenation.
HPLC analysis HPLC emerged as a powerful analytical tool during
the 1970s and, since then, it has advanced to be the
analytical separation method of choice for clinical,
forensic, and pharmaceutical fields including herbal
Peak alignment and warping: products. Increasingly, HPLC analytical techniques
Target peak alignment, have been included in many of the latest monographs
dynamic time warping, on the identification and determination of the plant con-
fuzzy warping, etc.
stituents as well as the adulterants in herbal products.
Applications of HPLC methods in the analysis of herbal
products have been reviewed by various authors
[19,21,50], and also, a good body of literature has
Normalisation:
Fingerprints are organised in become available in recent years. In the following
a data matrix sections, some specific examples with step-by-step
protocols of such analysis are presented.

Classification and discrimination: 19.5.1 Specific Examples of Step-by-step

Similarity, PCA, clustering etc. Protocols: Single and Multiple
Components Analysis Using Markers
FIGURE 19.4 An overview of the fingerprinting process of herbal It is quite common to use a single or multiple marker
products using HPLC. compounds (bioactive or nonbioactive) as an attempt
 19.5 ANALYSIS OF HERBAL PRODUCTS BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY 415
to assure quality of any herbal product using an [Link] Liquorice (Glycyrrhiza glabra)
HPLC-based method [51,52]. Most often, especially for Liquorice (roots of G. glabra) is a popular herbal
a monoherbal preparation, a single marker compound medicinal product, where glycyrrhizin is the main active
is used. However, for polyherbal preparations, because component (2e15% w/w). It is also used as a natural
of the complexity of the mixture, it is necessary to use sweetener because of the presence of glycyrrhizin,
more than one marker compound to assess the quality. which is 200 times sweeter than sucrose. Dried roots of
For example, 13 marker compounds have been used to liquorice are commercially sold as an herbal product
assess the quality of a Chinese polyherbal product, as well as a food additive. A semi-preparative HPLC-
Gwakhyangjeonggi-san [53]. There are numerous PDA-based method using glycyrrhizin as the marker
publications describing various HPLC-based validated compound has recently been published [62]. A sum-
quality control protocols incorporating single or mary of the protocol is presented below.
multiple marker compounds [41,53e60]. The following
are just a few simple examples of protocols for 1. Nine samples of liquorice were obtained from various
HPLC analysis of herbal products using marker global markets.
compounds. 2. Ground samples (15 g each) were Soxhlet-extracted,
sequentially, with n-hexane and MeOH (400 mL
each). The extracts were filtered and evaporated
under a vacuum to dryness. The residue from the
[Link] Vidanga (Embelia ribes)
extracts of each sample was redissolved in
Vidanga (dried ripe fruits of E. ribes) is a popular MeOH (6.05 mL) and filtered before HPLC analysis
Ayurvedic herbal medicinal product used traditionally (or TLC in the case of n-hexane extracts). As the
in India as an anthelmintic, carminative, and stimulant. n-hexane extract did not contain any glycyrrhizin (as
An HPLC-based method using a single marker checked by the TLC), the MeOH extracts of all nine
compound, embelin, for the quality assurance of samples were subjected to HPLC analysis.
Vidanga was proposed by Sudani et al. [61]. The proto- 3. Glycyrrhizin (20 mg/mL) in MeOH was used as the
col is summarized below. reference marker, and further dilutions (15, 10, 5, and
1. Three commercial samples of Vidanga were obtained 1 mg/mL) were made with MeOH. Each dilution of
from a local Indian market. stock was injected into HPLC in triplicates to
2. Samples were macerated with chloroform (CHCl3) at construct the calibration curve.
room temperature for 48 h in the dark. The extracts 4. Each MeOH extract (100 mL) was analyzed by an
were filtered and concentrated over a water bath until optimized and validated semi-preparative HPLC-
dry. PDA method using a reversed-phase ACE 10C18-HL
3. Embelin (1 mg/mL) in MeOH was used as the (150
10 mm, 10 mm) with a C18 guard column
reference sample (marker). ACE3310110GD (10
10 mm, 10 mm), setting the
4. Dried CHCl3 extracts of test samples (10 mg detection/monitoring wavelength at 254 nm and
equivalent of embelin) were resuspended in MeOH column temperature at 25  C, and using a binary
and sonicated for 10 min, filtered, and the volume gradient solvent system (30e100% B in A over 30 min)
was made to 10 mL with MeOH, and filtered again for with a flow rate of 3.00 mL/min, where solvent A
HPLC analysis. consisted of 0.1% v/v TFA in water and solvent B was
5. Each sample (20 mL) was analyzed by an optimized 0.1% v/v of TFA in MeOH. A Dionex 3000 LC System
and validated HPLC-PDA method using a reversed- coupled with Dionex 3000 semi-preparative pumps, a
phase Chromatopak Peerless basic C18 column Dionex 3000 autosampler, a Dionex 3000 column
(4.6
250 mm, 5 mm), setting the detection/ chamber, and a Dionex PDA-3000 detector was used.
monitoring wavelength at 291 nm and column Data were analyzed using the ChromeleonÒ
temperature at 25  C, employing solvent A: MeOH Chromatography Data System.
and solvent B: phosphate buffer pH 3.0 (adjusted with 5. While the retention time and the UV spectra were used
5% v/v acetic acid), using an isocratic elution with to identify glycyrrhizin, peak area was used to quantify
10% B in A and at a flow rate of 1.4 mL/min. A the amounts of glycyrrhizin present in each sample.
PerkineElmer Series 200 HPLC system with PDA was The glycyrrhizin percentage level in the samples was in
used. the range of 0.177e0.688% w/w of dry materials.
6. Presence of embelin in commercial samples was
determined by direct comparison of the relative
retention time and the UVeVis absorption maxima of [Link] Valerian (Valeriana officinalis)
corresponding peak. Relative peak area was used to Valerian (V. officinalis) is a well-known herbal medic-
quantify embelin in the samples. inal product, traditionally used to promote sleep and as
416 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

an anxiolytic agent for nervous unrest, to treat neuralgia, for ginsenoside Rg1, 488.92 mg/mL for ginsenoside
epilepsy, and to relieve digestive and other spasms of Re, 429.74 mg/mL for ginsenoside Rb1, 556.93 mg/mL
smooth muscle [63]. An HPLC-based method using a for glycyrrhizin, 340.47 mg/mL for liquiritin,
single bioactive marker compound, valerinic acid, for 100.52 mg/mL for glycyrrhetinic acid, 149.30 mg/mL
qualitative and quantitative analysis of some commer- for 10-DAB III, 120.51 mg/mL for baccatin III,
cial brands of Valeriana products, including tablets, 1314.02 mg/mL for taxol, and 122.80 mg/mL for
caplets, capsules, and drops, aiming at their quality con- cephalomannine.
trol was reported by Ghafari et al. [63]. The protocol is 3. These stock solutions were mixed to obtain the
summarized below. combined solutions and then were diluted to yield a
series of standard working solutions with different
1. Commercial samples (Neurogol tablets, Valerian
concentrations for linear validation.
capsules, antimigraine herbal drops, etc.) of Valerian
4. Sample preparation of raw herbs: Three component
products were obtained from local pharmacies in
herbs of this preparation, Hongdoushan, RenShen,
Tehran, Iran.
and GanCao (10 g each), were dried, ground, and
2. Samples (solid dosage forms) were macerated with
extracted to obtain extracts as outlined for the
80% aqueous MeOH (20e60 mL) at room
formula of Hongdoushan capsules. The extracts were
temperature. For analysis, four to six tablets, two
dissolved in MeOH to prepare sample solutions.
caplets, and three capsules were used. The
5. Sample preparation of Hongdoushan capsules: A
supernatants were filtered through a 0.45 mm filter.
sample (0.3 g) was placed into a centrifuge tube and
The extraction process was repeated twice. A portion
extracted ultrasonically (2
30 min) with MeOH
(5 mL) of each extract was diluted with 20 mL of
(10 mL) at room temperature. The pooled extract was
MeOH and again filtered through a 0.45 mm filter.
centrifuged for five minutes at 9000 g, the supernatant
3. Valerenic acid in MeOH was used as the reference
was removed into a 25 mL volumetric flask, and the
sample, and a calibration curve of the standard was
constant volume was obtained with MeOH. All the
made using various dilutions. Valerenic acid was
solutions were filtered through a 0.45 mm membrane
identified within the samples by comparing its retention
filter for HPLC analyses.
time with the internal standard of valerenic acid.
6. Each sample (5 mL) was analyzed by an optimized
4. Each sample (10 mL) was analyzed by an optimized
and validated HPLC-PDA method using a reversed-
and validated HPLC-UV method using a reversed-
phase SHIMADZU C18 column (4.6
250 mm, 5 mm),
phase VP-ODS C18 column (4.6
250 mm, 5 mm),
coupled with a Phenomenex guard column
setting the detection/monitoring wavelength at
(4.0
3.0 mm), setting the temperature at 25  C and at
220 nm and at a flow rate of 0.8 mL/min. A Shimadzu
a flow rate of 0.8 mL/min and utilizing a gradient
LC-10ADVP HPLC system with UVeVis detector
mobile phase comprising acetonitrile (component A)
was used.
and water (component B). The elution program was
[Link] Compound Hongdoushan Capsule set as follows: 0e35 min kept at 19%A, 35e50 min
linear increased from 19% to 55% A, 50e70 min kept
This is a classic example of using HPLC for the analysis
at 55% A, 70e85 min linear decreased from 55% to
of a polyherbal product utilizing multiple marker com-
19% A. An HP1200 Agilent HPLC system equipped
pounds. Zhu et al. [37] used 10 bioactive markers, e.g.,
with a dual pump, an autosampler, and a PDA
glycyrrhetinic acid, liquiritin, glycyrrhizin, baccatin III,
detector were used. Detection wave lengths were at
10-deacetylbaccatin III, cephalomannine, taxol, ginseno-
0e30 min (276 nm), 30e50 min (203 nm), 50e85 min
side Rg1, ginsenoside Re, and ginsenoside Rb1, to
(227 nm), and PDA spectra were recorded from 200 to
analyze the compound Hongdoushan capsule, a widely
600 nm.
known compound herbal preparation, which is often
7. Retention times and associated UVeVis spectral
used for the treatment of ovarian and breast cancers
characteristics of the marker compounds were used to
and to stimulate the body immunity. A Hongdoushan
identify the presence of these compounds in the
capsule contains more than one herbal ingredient (Panax
samples (six batches of Hongdoushan capsule).
ginseng, Glycyrrhiza uralensis, and Taxus Chinensis) and
thus can be quite difficult to control or assure the quality
of the product. A summary of the protocol is presented [Link] Menoprogen Capsules
below.
The Menoprogen capsule is a well-known traditional
1. All samples were obtained from local markets in Chinese medicine formula generally prescribed for the
China. management of menopause. Wang et al. [64] reported
2. Stock solutions of the marker compounds were an HPLC-PDA-based quality control method for this
prepared in MeOH at concentrations of 362.36 mg/mL formula, utilizing more than one marker compound
 19.5 ANALYSIS OF HERBAL PRODUCTS BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY 417
and incorporating a fingerprinting approach. A sum- the temperature at 25  C and at a flow rate of 1.5 mL/
mary of this method is presented below. min, and utilizing a binary step-gradient mobile phase
consisting of 0.5% aqueous acetic acid (solvent A) and
1. Ten batches of Menoprogen capsules were obtained
acetonitrile with 20% A (solvent B) over 50 min:
from Nanjing Moresoft Manufacturing Company,
0e10 min 10% B, 10e20 min 10e20% B, 20e30 min
Nanjing, China.
20e40% B, 30e40 min 40e60% B, 40e45 min 60e70%
2. The content of the capsules of each batch (1 g each)
B, and 45e50 min 70e10% B. A Shimadzu HPLC
was extracted with MeOH (10 mL) with sonication for
system comprising LC-10AT pump, automated
60 min at room temperature, filtered through a
gradient controller, Shimadzu SPD-M10 A (Class VP-
0.45 mm membrane filter before HPLC analysis.
series, version 6.10) and Rheodyne 7725 I manual
3. Chlorogenic acid, rutin, hyperoside, quercetin, and
injector was used. Detection wave length was 210 nm.
kaempferol were used as marker compounds.
4. While the marker compounds were identified by the
4. Each sample (5 mL) was analyzed by an optimized
retention times and respective UVeVis spectra, peak
and validated HPLC-PDA method using a reversed-
area was used for quantification.
phase Kromasil ODS C18 column (4.6
250 mm,
5 mm), setting the temperature at 25  C and at a flow
rate of 1.0 mL/min and utilizing a step-gradient 19.5.2 Specific Examples of Step-by-step
mobile phase comprising 0.1% phosphoric acid Protocols: Chemical Fingerprinting
(solvent A) and acetonitrile (solvent B) over 135 min:
There are several examples available in the literature
0e50 min 10e20% B, 50e86 min 20e48% B,
[21,65e76] that describe HPLC-based fingerprinting of
86e135 min 48e100% B. An HP1100 Agilent HPLC
various herbal products, both mono- and polyherbals.
system consisting of a vacuum degasser,
While probably because of low cost and simplicity, the
thermostated column compartment, and a photo-
use of HPLC-PDA seems to be the most common of all,
diode-array detector was used. Detection wave
many other sophisticated hyphenated techniques, e.g.,
length was 205 nm.
LC-PDA-MS [21,67e69,74,76,77], are also used. For
5. While the marker compounds were identified by the
example, HPLC and LC-DAD-MS/MS were applied for
retention times and respective UVeVis spectra and
the qualitative and quantitative analyses of Compound
quantified by peak area, the fingerprinting approach
Kushen Injection, a well-known Chinese herbal product
identified 21 “common peaks” representing the
[74,78]. In that study, a simple fingerprinting approach
characteristic chemicals of this herbal product.
was used; eight peaks were selected as the characteristic
This multiple marker-based HPLC-PDA finger- peaks in the HPLC chromatograms of 27 different batches
printing approach demonstrated a comprehensive qual- of this herbal product. Peak identification was carried out
ity evaluation method for Menoprogen capsules and for by HPLC and LC-MS/MS analyses. A total of 21 chro-
the chemical standardization and batch-to-batch consis- matographic peaks were identified by comparison of
tency of this product. retention time and UV spectra with authentic compounds
as well as by summarized MS fragmentation rules. Some
[Link] Arjunarishta variations in the total amounts of marker compounds
were observed in different batches of this product.
Arjunarishta is an Ayurvedic cardioprotective poly-
1.5.2 Some simple examples of step-by-step protocols
herbal formulation. It is prepared by fermenting the
of chemical fingerprinting of herbal products are pre-
decoction of the bark of Terminalia arjuna, fruits of Vitis
sented below.
vinifera, and flowers of Madhuca indica using flowers of
Woodfordia fruticosa. Lal et al. [34] reported an HPLC- [Link] Curculiginis Rhizoma (Curculigo
PDA-based standardization method for this formula- orchioides)
tion, utilizing more than one marker compound. A
Curculiginis Rhizoma, the dry rhizome of C. orchi-
protocol of this method is outlined below.
oides, is a well-known Chinese herbal medicinal product
1. The formulation (fermented decoction, 1 mL) was and also an important Ayurvedic product used for the
dried under vacuum, suspended in MeOH (5 mL), treatment of pain, inflammation, immune and hepatic
sonicated for 10 min, centrifuged at 3000 rpm, and the disorders, and as an aphrodisiac [79]. Curculigoside, a
supernatant (1 mL) was filtered (0.45 mm). phenolic glycoside, is considered to be the major contrib-
2. Gallic acid, ethyl gallate, ellagic acid, quercetin, and utor to the medicinal properties of this herbal product.
kaempferol were used as marker compounds. However, as it is a well accepted fact that the overall
3. Sample (20 mL) was analyzed by an optimized and therapeutic activity (and/or toxicity) of any herbal prod-
validated HPLC-PDA method using a reversed-phase uct is not necessarily owing to just one compound, but
Phenomenex C18 column (4.6
250 mm, 5 mm), setting most often a result of synergistic effect of a number of
418 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

compounds present in the product, an HPLC-based and an anodyne. The roots and rhizomes of Notoptery-
chemical fingerprinting method aiming at the quality gium forbesii and N. incium, both from the Apiaceae
evaluation of this herbal product has recently been re- family, are included in the Chinese Pharmacopoeia
ported by Bian et al. [79]. A summary of the protocol under the same name (Qianghuo in Chinese). An
is presented below. HPLC-PDA based fingerprinting quality control method
was reported by Jiang et al. [80]. Evaluation System for
1. Ten different samples of Curculiginis Rhizoma were
Chromatographic Fingerprint of Traditional Chinese
collected from China.
Medicine (Version 2004 A), as in the above example
2. Each sample (2 g) was refluxed with MeOH (45 mL)
(see Section 5.2.1), was employed to normalize the
for 60 min, and the resulting mass was centrifuged at
chromatographic peaks, to calculate the correlation
3000 rpm for 15 min.
coefficients between entire chromatographic profiles,
3. The supernatant was transferred to a 50 mL
and to perform quantitative standard fingerprint/chro-
volumetric flask, MeOH was added to bring it to the
matogram for a group of chromatograms. Besides, the
50 mL mark, and finally it was filtered through a
relative retention time and relative peak area of each
cellulose acetate membrane (0.45 mm) for HPLC
characteristic peak related to the reference peak were
analysis.
calculated for quantification of the chemical properties
4. Each sample (10 mL) was analyzed by an optimized
in the chromatographic pattern of the product. A step-
and validated HPLC-PDA method using an Altima
by-step protocol of this method is summarized below.
C18 ODS column (4.6
250 mm, 5 mm), setting the
detection/monitoring wavelength at 220 nm and
1. Eighteen different samples of Qianghuo, 15
temperature at 30  C, employing solvent A:
containing N. forbesii and three containing N. incium,
acetonitrile with 0.02% TFA and solvent B: water with
were collected from China.
0.5% TFA in a step-wise gradient (7e30% A in B over
2. Each sample (2 g) was extracted with distilled water
35 min) and a flow rate of 1.0 mL/min. An Agilent
(25 mL) using an ultrasonic water bath for 15 min. The
1100 liquid chromatography system, equipped with
extraction was repeated twice, and extracts were
the G1311A Quatpump solvent delivery system,
mixed and filtered.
UVeVis photodiode array detector G1315B, and
3. The filtered extract was precipitated by addition of
degasser G1322A was used.
ethanol (final concentration was 50% v/v) at 4  C for
5. Similarity analysis to establish similarities of different
12 h and filtered again, and diluted by water to the
chromatograms was carried out by Similarity
final volume of 100 mL. The final extract solution
Evaluation System for Chromatographic Fingerprint
was filtered again (0.45 mm filter) prior to HPLC
of Traditional Chinese Medicine (Version 2004A), as
analysis.
recommended by the State Food and Drug
4. Each sample (10 mL) was analyzed by an optimized
Administration of China (SFDA). Correlation
and validated HPLC-PDA method using a reversed-
coefficient was used as a measure for similarity
phase Thermo C18 column (4.6
250 mm, 5 mm),
analysis.
setting the detection/monitoring wavelength at
6. Out of 20 separated peaks, 11 peaks were found in all
310 nm and column temperature at 30  C, employing
samples and were termed as “common peaks.”
solvent A: water with 0.1% phosphoric acid (H3PO4)
It can be noted that for any HPLC-based finger- and solvent B: acetonitrile in a step-linear gradient
printing analysis, all HPLC conditions have to be opti- (10e30% B in A over 12 min; 30e100% B in A in
mized prior to any sample run, and all chromatograms 12e25 min) and at a flow rate of 1.0 mL/min. An
must be standardized. The process of standardization Agilent HP 1100 series HPLC-DAD system consisting
generally involves the selection of “common peaks” in of a vacuum degasser, thermostated column chamber,
chromatograms and the normalization of retention times and a DAD was used. Ferulic acid, imperatorin, and
of all the common peaks. Relative retention time and isoimperatorin were used as standards.
relative peak area of each characteristic peak related to 5. Similarities of different chromatograms by
the reference peak are usually calculated for the quanti- calculating the correlative coefficient and/or cosine
tative parameters of chemical properties in the chro- value of vectorial angle was determined by Similarity
matographic pattern and for similarity analysis of Evaluation System for Chromatographic Fingerprint
herbal products. of Traditional Chinese Medicine (Version 2004A), as
recommended by the State Food and Drug
[Link] Qianghuo (Notopterygium forbesii Administration of China (SFDA).
or N. Incium) 6. A total of 10 peaks were separated, and the highest
Qianghuo is a popular Tibetan and Chinese herbal peak (ferulic acid) in the chromatogram was assigned
medicinal product used as a diaphoretic, an antifebrile, as the reference peak.
 19.5 ANALYSIS OF HERBAL PRODUCTS BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY 419
The selection of detection wavelength (310 nm) was 1.0 mL/min. The UV spectra were recorded over the
one of the key factors that contributed to the reliability range of 200e400 nm.
and reproducibility of Qianghuo. The relative retention 6. Similarity of the chromatographic pattern among
time and relative peak area for all common peaks were batches was determined by the presence of
determined in comparison with the ferulic acid peak mathematical parameters in the range of 80e125% of
(reference peak). the average.

[Link] Ayurved Siriraj PrasachandaengdAn

Antipyretic Product [Link] Radix Aconiti (Roots of Aconitum
carmichaeli)
For the quality control of the Thai traditional medici-
nal polyherbal product, Ayurved Siriraj Prasachan- A fingerprinting approach was introduced by means
daeng, HPTLC and HPLC-based fingerprinting of UPLC-ESI/MSn (ultra performance liquid chroma-
approaches were adopted [35]. This product, sold as a tography-electrospray ionization/mass spectrometry)
powder form, contains 12 medicinal herbs. The exam- for the quality control of the Chinese herbal product
ples shown in Sections 5.2.1 and 5.2.2 are monoherbal Radix Aconiti, a widely used toxic traditional herbal
products and are certainly much easier to deal with, medicine [76]. This approach was founded on two pro-
especially in relation to fingerprinting, than a polyherbal cessing methods as recorded in the Chinese Pharmaco-
product like this one. Relative retention time and poeia for the purpose of reducing the toxicity and
applied information content were evaluated in the ensuring the therapeutic efficacy. Various data process-
HPLC fingerprints. Gallic, caffeic, and vanillic acids ing approaches, e.g., similarity evaluation, hierarchical
were used as qualitative and quantitative markers in cluster analysis, and principal component analysis
HPLC. Similarity of the chromatographic pattern among (PCA), were carried out to assess the similarity and vari-
batches was determined by the presence of mathemat- ation among the samples. A summary of the protocol of
ical parameters in the range of 80e125% of the average. this fingerprinting method is presented below.
Twelve medicinal herbal components and three batches
1. Sample processing and extraction: all samples were
of the product were analyzed. The chromatograms from
processed according to the Chinese Pharmacopoeia
three batches were regarded as the original fingerprint
2010. Briefly, Radix Aconiti was cleaned and soaked
of this product. A summary of steps involved in the
with water until it was wet thoroughly, water was
fingerprinting of this product is presented below.
drained and replaced with fresh water, and the water
1. Three batches of Ayurved Siriraj Prasachandaeng and was exchanged with fresh water and boiled for one to
its 12 components (see Table 19.4), a total of 15 six hours or steamed for one to eight hours. The
samples, were obtained from the Herbal Medicines samples boiled for four, five, and six hours and those
and Products Manufacturing Unit, Ayurved Siriraj, steamed for six, seven, and eight hours were
Faculty of Medicine, Siriraj Hospital, Mahidol, designated as the qualified Radix Aconiti. The other
Thailand. Radix Aconiti samples were unqualified ones. Each of
2. All 15 samples were extracted separately in 80% the boiled or steamed samples was sliced, dried
ethanol and lyophilized to dry powder. (50  C), ground, and sieved (0.45 mm). Ground sample
3. Each powder was dissolved separately in 50% (1.0 g) was placed in a sealed vessel by adding 0.5 mL
aqueous MeOH for HPLC and then filtered through a of 10% aqueous ammonia solution, and diethyl ether
0.2 mm membrane filter. (50 mL) was added and extracted using an ultrasonic
4. Reference markers, gallic acid, caffeic acid, and bath for 30 min at 25  C. The extraction was repeated
vanillic acid (1 mg each) were dissolved separately in three times, pooled together, dried at 50  C, and the
50% aqueous MeOH. Each solution was vortexed for dried extract was stored at 80  C. For HPLC
two minutes and filtered through a 0.2 mm membrane analyses, the extracts were dissolved in MeOH-water
filter. (1:1, 5 mL) and 0.02 mg/mL of reserpine was added
5. Each sample (10 mL) was analyzed by an optimized as an internal standard. All solutions were filtered
and validated HPLC-PDA method using a reversed- through a membrane (0.22 mm).
phase C18 column (4.6
250 mm, 5 mm), setting the 2. Standard solutions: stock solutions of aconitine-type
column temperature at 30  C, employing solvent A: alkaloids (aconitine, hypaconitine, mesaconitine,
o-phosphoric acid (0.1% v/v) and solvent B: benzoylmesaconine, benzoylhypacoitine, and
acetonitrile in an isocratic condition of 95% A and 5% benzoylaconine) were prepared in dichloromethane
B for the first four minutes, a step-gradient elution of and kept at 80  C. An internal standard solution
95-85% A at 4e8 min; 85e0% A at 8e13 min and comprising 0.02 mg/mL reserpine in 1:1 MeOH-
0e95% A at 13e16 min and at a flow rate of water was used for further dilution of all standards.
420 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

3. HPLC-PDA analysis: each sample (5 mL) was and simultaneous quantification of the chemical constit-
analyzed by an optimized and validated HPLC-PDA uents of Folium Turpiniae has recently been reported
as well as a UPLC-ESI/MS method using a reversed- [84]. Nine marker compounds were quantified simulta-
phase Waters ACQUITY UPLC BEH C18 column neously in 10 batches of Folium Turpiniae collected
(2.1
50 mm, 1.7 mm), setting the column from different regions. Hierarchical clustering analysis
temperature at 30  C, employing solvent A: water (HCA), which is a useful multivariate statistical
containing 5 mM ammonium bicarbonate and technique to assign a data set into groups by creating a
adjusted to pH 10.5 with ammonia, and solvent B: cluster tree or dendrogram, and principal component
MeOH in a gradient elution mode: 0e5 min from 65% analysis (PCA, based on area of common peaks in
A to 57% A, 5e10 min 57% A to 55% A, 10e30 min chromatograms) revealed notable similarities (or differ-
from 55% A to 40% A, 30e55 min from 40% A to 10% ences) among various samples. A summarized protocol
A, and 55e75 min from 10% A to 5% A, and at a flow is outlined below.
rate of 0.3 mL/min. The UVeVis spectra were
recorded over the range of 200e600 nm, but the 1. Ten batches of samples of Folium Turpiniae were
chromatogram was monitored at 235 nm, which was obtained from different parts of China.
suitable for aconitine-type alkaloids. An Accela UPLC 2. Nine compounds, ellagic acid-3-O-a-L-
system equipped with an Accela 1250 pump, rhamnopyranoside, apigenin-7-O-(20 -rhamnosyl)-
autosampler, DAD, and an LTQ ion trap mass gentiobioside, luteolin-7-O-b-D-neohesperidoside,
spectrometer was used. ligustroflavone, apigenin-7-O-b-D-(60 -rhamnosyl)
4. UPLC-MS analysis: The ESI source of the mass glucoside, 40 -O-methylellagic acid-3-O-a-L-
spectrometer was connected to the UPLC system via a rhamnopyranoside, rhoifolin, neobudofficide, and
capillary to the UV cell outlet. Spray voltage was set apigenin-7-O-b-D-(20 -O-a-rhamnosyl)glucuronide, all
at 4.5 kV in the positive ion mode, sheath gas flow rate with >98% purity, were used as marker compounds.
at 40.5 L/h, and aux gas flow rate at 90 L/h. The 3. Each sample powder (0.5 g) was transferred to a
capillary temperature was maintained at 250  C. The 100 mL round-bottom flask and mixed with 50%
scan range was m/z 300e1000 Da. Collision energies MeOH (50 mL), refluxed for 15 min, and cooled to
for the MS2 analyses ranged from 25e40 eV, room temperature. MeOH was added to top it up to
depending on the mass of the precursor ion. Helium 50 mL and filtered through a 0.45 mm membrane
(He) was used as the collision gas for the MSn. before injection into the HPLC.
5. For chemometric analysis of the data obtained from the 4. Each sample (10 mL) was analyzed by an optimized
HPLC runs, similarity evaluation (to assess the and validated HPLC method using a reversed-phase
consistency) [81,82], hierarchical clustering analysis Hanbon Phecda C18 column (4.6
250 mm, 5 mm)
(to classify samples based on the similarities of their with a standard guard column, setting the column
chemical properties) [19,24], and principal component temperature at 35  C, employing solvent A: 1% formic
analysis (to feature extraction and dimensionality acid (containing 2 mmol/L ammonium acetate) and
reduction aiming at reduction of data dimensionality solvent B: acetonitrile, in a gradient elution mode:
and provision of an overview of class separation, 3e9% B at 0e10 min, 9e12% B at 10e26 min, 12e14%
clustering, and outliers) were carried out [83]. B at 26e32 min, 14e17% B at 32e40 min, 17% B at
40e48 min, 17e22% B at 48e58 min, 22e38% B at
In this UPLC-UV and UPLC-ESI/MSn fingerprint
58e70 min, and 38e60% B at 70e74 min, and at a flow
study, 39 characteristic MS peaks and 34 UV chromato-
rate of 1.0 mL/min. The UVeVis spectra were
gram peaks in the common pattern were identified to
recorded over the range of 200e600 nm, but the
further characterize the chromatographic fingerprint
chromatogram was monitored at 262 nm. A
and contribute to the quality control of this Radix
Shimadzu LC2010 HPLC system coupled with a
Aconite herbal product through successful application
quaternary pump, online vacuum degasser,
of chemometrics.
autosampler, thermostat column chamber, PDA
detector, and LC solution software was used.
[Link] Folium Turpiniae 5. The HPLC-PDA system was interfaced to an Agilent
Folium Turpiniae, a well-known traditional Chinese 6520 Q-TOF time-of-flight (TOF) mass spectrometer
medicine derived from Turpinia arguta (Staphyleaceae), equipped with an electrospray interface operating
is used for the treatment of abscesses, fevers, gastric under the chromatographic conditions mentioned
ulcers, and inflammations [84]. A strategy incorporating above. The optimized MS operating conditions were
HPLC-PDA and quadrupole TOF-MS detection, as negative and positive ion mode, scan spectra from m/
well as phytochemical and chemometrics analysis z 100e1000, drying gas (N2) with a flow rate of 8.0 L/
(and fingerprinting) for the characterization, isolation, min, drying gas temperature of 325  C, nebulizer
 19.5 ANALYSIS OF HERBAL PRODUCTS BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY 421
pressure of 40 psi, capillary voltage of 3500 V, was added to top it up to the 25 mL mark and it was
skimmer of 65 V, and fragmentor voltage of 120 V. The filtered through a 0.22 mm membrane before injection
sample collision energy was set at 35 V. Data were into the HPLC.
processed using MassHunter Workstation Data 6. A fully validated HPLC fingerprinting analysis [86] of
Acquisition Software Ver. A.01.00. each sample (10 mL) was performed on a Waters
6. Samples were analyzed by HCA and PCA using SPSS Alliance 2695 HPLC equipped with an online vacuum
software (SPSS 11.5, SPSS) and the areas of 31 degasser, a high-pressure quaternary pump, an
“common peaks” in chromatograms were selected as autosampler, and an Alltech 2000 evaporative light
the variables. scattering detector (ELSD), utilizing a Welchrom C18
column (250
4.6 mm, 5 mm) with a binary gradient
Using the hyphenated technique HPLC-DAD-Q-
elution as follows: acetonitrile as solvent A and water
TOF-MS (where multiple detectors were used) and
as solvent B, initial 25% A, 0e5 min linear gradient
phytochemical and chemometrics analyses, it was
25e30% A, 5e20 min at 30% A, 20e35 min from 30 to
possible to develop qualitative and quantitative analyt-
35% A, 35e45 min 35e47% A, 45e47 min 47e70% A,
ical methods for Folium Turpiniae herbal Medicinal
47e60 min at 70% with a flow rate of 1 mL/min. The
products from different regions.
effluent was introduced into the Alltech 2000 ELSD,
[Link] Dioscorea zingiberensis in which the drift tube temperature was 90  C and the
gas flow rate at 2.8 L/min. Data analyses were carried
Dioscorea zingiberensis is a Chinese medicinal plant,
out on an Empower Workstation.
which is the major component of a traditional Chinese
7. A Varian 212-LC equipped with a Q-TOF Premier, a
medicinal product. Steroidal saponins have already
quadrupole, and orthogonal acceleration time-of-
been shown to be the major bioactive compounds
flight tandem mass spectrometer, linked to an
responsible for the therapeutic efficacy of this product
electrospray ionization interface, was used for the
[85]. HPLC-based quality control involving saponins
HPLC-ESI-Q/TOF analyses of the samples. MS data
as the markers is not an easy task, but Zhang et al. [86]
were recorded by a Varian MS Workstation software.
have managed to establish a quality control method
High purity N2 gas was used as the nebulizer and
for D. zingiberensis utilizing a fingerprint approach
auxiliary gas and argon as the collision gas. The same
involving HPLC coupled with evaporative light scat-
HPLC conditions, including the column, elution
tering detector (HPLC-ELSD) and the simultaneous
program, and flow rate as outlined above were
characterization of the steroid saponins by HPLC
applied. A portion of the column effluent (0.2 mL/
coupled with electrospray ionization-mass spectrometry
min) was delivered into the ion source of the mass
and quadrupole tandem time-of-fight mass analyzers
spectrometer after a micro split. The conditions of the
detection (HPLC-ESI-Q/TOF). A summarized protocol
ESI source were drying gas (N2) flow rate 9.0 L/min,
is outlined below.
drying gas temperature 350  C, nebulizer 35 psig,
1. Twenty batches of samples of D. zingiberensis were capillary voltage 5000 V, and spectral scanning range
obtained from different parts of China. from m/z 100 to 1500.
2. Twelve saponins (purity >98%), 10 previously 8. Data processing for the fingerprinting analysis was
isolated and identified by the authors [86] and two achieved by using the software named the Similarity
purchased from a commercial supplier, were used as Evaluation System for Chromatographic Fingerprint
marker compounds. of Traditional Chinese Medicine (Version 2004A)
3. Stock solution (5 mg/mL) of each of these markers according to the recommendation made from the
was prepared in MeOH, several dilutions were State Food and Drug Administration (SFDA) of
carried out in MeOH, and the resulting solutions China, which is mainly applied in the similarity study
were kept at 4  C for subsequent use. of chromatographic and spectral patterns. This
4. Each sample powder (5 g) was refluxed three times software helped to synchronize the chromatographic
with 50 mL of 70% ethanol at 80  C. Pooled extracts peaks and to calculate the correlation coefficients
were evaporated to dryness under a vacuum, and the between entire chromatographic profiles, as well as to
residue was redissolved in water (30 mL) and compute and generate the mean chromatogram as a
centrifuged. representative standard fingerprint chromatogram
5. The supernatant obtained after centrifugation was from a group of chromatograms. The similarities of
passed through aD-101 macroporous resin column different chromatographic patterns were evaluated
(2
15 cm) eluting first with water, then with 20% among samples.
ethanol until the effluent was colorless, and finally
with 70% ethanol (80 mL), which was concentrated The chromatographic patterns obtained from the
and transferred to a 25 mL volumetric flask. MeOH HPLC-ELSD analyses of 20 different samples were
422 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

generally consistent and similar in terms of the presence 6. A fully validated HPLC quantification and
of similar chemical compounds (saponins), albeit the fingerprinting analysis [78] of each sample (5 mL) was
intensity of the peaks was variable. Out of 12 marker performed on an Agilent 1260 series HPLC, utilizing
saponins, 10 could be easily determined by the ESI-MS an Agilent Eclipse plus C18 column (250
4.6 mm,
analyses. Among the samples, a total of 68 “common 5 mm) with a binary gradient elution as follows: 0.4%
characteristic” peaks, including 22 new steroid saponins ammonium acetate aqueous solution (pH adjusted to
with eight aglycone skeletons in the fingerprint, were 6.0 with glacial acetic acid) as solvent A and
detected. Zhang et al. [86] demonstrated that trivial acetonitrile as solvent B, initial 17% B, 0e25 min linear
HPLC-based fingerprinting, when combined with the gradient 17e19% B, 25e55 min at 19% B, 55e70 min
online HPLC-MS method, could become a powerful from 19 to 25% B, 70e80 min 25e28% B, 80e95 min
tool in the quality control of herbal products and in 28e34% B, 95e120 min 34e35% B, 120e140 min
the analysis of their chemical constituents. However, a 35e42% B and 140e160 min 42e50% B; flow rate
similar approach was previously reported by Man 1 mL/min; column temperature at 30  C; quantitative
et al. [87]. detection wavelength at 254 nm (xanthotoxin,
bergapten, imperatorin and isoimperatorin), 270 nm
[Link] Yuanhu Zhitong Tablet (berberine), 280 nm (protopine and
tetrahydropalmatine) or 345 nm (jatrorrhizine,
The Yuanhu Zhitong tablet is a Chinese polyherbal
coptisine, and palmatine), while the wavelength of
product consisting of Angelica dahurica (223 g) and
fingerprinting analysis was set at 280 nm.
Rhizoma Corydalis (445 g) processed with vinegar,
7. The above Agilent HPLC system was interfaced
and it has been used to treat gastralgia, costalgia, head-
with an Agilent 6460 Triple Quadrupole mass
ache, and dysmenorrhea in China [88]. Tang et al. [78]
spectrometer in a post-column splitting ratio of 4:1.
have recently reported a quality control strategy for
The conditions of ESI source were source voltage
quantitative and qualitative analysis of “common
3000 V; drying gas (N2) flow rate10.0 L/min; drying
peaks” in the chemical fingerprint of Yuanhu Zhitong
gas temperature 320  C; and nebulizer 25 psi.
tablets, using photo-diode-array detector and tandem
The MS data were acquired from m/z 100 to1000
mass spectrometry (HPLC-DADeMS/MS). A brief
in positive ion modes. LC-ESI-MS/MS was used
outline of this approach is presented below.
to identify “common peaks” in each sample. In the
1. Twelve batches of Yuanhu Zhitong tablets were full scan mass spectra, most of the constituents
obtained from various pharmaceutical companies in exhibited their pseudomolecular ions [M þ H]þ in
China [78]. positive ion mode under the soft electrospray
2. Ten marker compounds (purity >98%) [78] were ionization mode. Precursor ions were subjected to
purchased from the National Institute for the Control collision-induced dissociation to generate the
of Pharmaceutical and Biological Products (Beijing, fragment ions, and the fragmentation patterns were
China). proposed for the structural identification of
3. Coatings of the tablet samples were removed, and the constituents.
content was ground into fine powder. Each ground 8. Data analysis was carried out by the software,
sample (1.0 g) was extracted with MeOH (35 mL) Similarity Evaluation System for Chromatographic
ultrasonically for 30 min, volume was made to 50 mL Fingerprint of TCM (Version 2004A). The relative
with MeOH, filtered, and evaporated to dryness on a retention time and relative peak area of each
water bath (70  C). The residue was redissolved in “common peak” related to the marker peaks were
MeOH (5 mL) and centrifuged (15,000 rpm) for calculated for quantitative expression of the chemical
10 min. The supernatant was filtered (0.45 mm filter) properties in the chromatographic pattern of the
and transferred to an autosampler vial for HPLC- tablet samples. Based on this, the correlation
DAD-ESI-MS/MS analysis. coefficients of entire chromatographic profiles of
4. According to the prescription and preparation samples were calculated, while the simulative mean
protocol of Yuanhu Zhitong tablets recorded in China chromatogram was generated.
Pharmacopoeia, two negative control samples
without Radix, Angelica dahurica or Rhizoma A total of 40 peaks were assigned as the “common
Corydalis, were prepared to validate the specificity of peaks.” For quantification of “common peaks,” the
the method. detection wavelength was set at 254, 270, 280, or
5. Stock solution of each of the 10 markers was prepared 345 nm [78]. Ten analytes, e.g., protopine, jatrorrhizine,
in MeOH, or MeOH-water (1:1). Several dilutions coptisine, palmatine, berberine, xanthotoxin, bergapten,
were carried out in MeOH, and the resulting tetrahydropalmatine, imperatorin, and isoimperatorin,
solutions were kept at 4  C for subsequent use. were simultaneously determined. For qualification
 REFERENCES 423
of “common peaks,” 33 compounds including 10 quan- [8] Kirkland JJ. Development of some stationary phases for reversed-
titative analytes were identified or tentatively character- phase high-performance liquid chromatography. J Chromatogr A
2004;1060:9e21.
ized using LC-MS/MS. [9] Latif Z, Sarker S. Isolation of natural products by preparative high
performance liquid chromatography (prep-HPLC). In: Natural
products isolation. 3rd ed. USA: Humana Press e Springer-Verlag;
19.6 CONCLUSIONS 2012.
[10] Moldoveanu SC. Mobile phases and their properties. In:
Hyphenated techniques, originating from the Moldoveanu MC, David V, editors. Essentials in modern HPLC
separations. Elsevier; 2013. p. 363e447.
coupling between HPLC and various detection tech- [11] Eurby MR, Petersson P. Chromatographic classification and com-
niques, e.g., UVeVis, PDA, MS, and NMR, have become parison of commercially available reversed-phase liquid chro-
indispensable tools for analyzing various complex matographic columns using principle component analysis.
matrices, including herbal products. Much of the appli- J Chromatogr A 2003;994:13e26.
cations of HPLC-based analyses of herbal products are [12] Snyder LR, Kirkland JJ, Glajch JL. Practical HPLC method devel-
opment. 2nd ed. New York: John Wiley and Sons; 1997.
associated with authentication, standardization, and [13] Dolan JW. Starting out right, part 3-the role of the solvent in con-
quality assurance/control of various herbal products trolling selectivity. LC-GC Europe 2000;13:148e56.
and often provide better outcome than other conven- [14] Dolan JW. Starting out right, part 6 e the scouting gradient
tional techniques. Quality control of herbal products alternative. LC-GC Europe 2000;13:388e94.
aims to ensure its quality, safety, and efficacy. HPLC [15] Dolan JW. How much is enough? LC-GC Europe 2003;16:740e5.
[16] Pratima NA, Shraddha B, Zibran S. Review of ultra performance
analyses involving chemical markers, as well as finger- liquid chromatography and its applications. Int J Res Pharm Sci
printing with chemometrics, are pivotal in the current 2013;3:19e40.
practice of quality control of such products. Nowadays, [17] Qi Y, Li S, Pi Z, Song F, Lin N, Liu S, et al. Chemical profiling of
to obtain a comprehensive understanding of the charac- Wu-tou decoction by UPLC-Q-TOF-MS. Talanta 2014;118:21e9.
teristic components of an herbal product, chemical [18] Novakova L, Matysova L, Solich P. Advantages of application of
UPLC in pharmaceutical analysis. Talanta 2006;68:908e18.
fingerprints from hyphenated techniques, especially [19] Yang G-L, Yang L-W, Li Y-X, Cao H, Zhou W-L, Fang Z-J, et al.
involving HPLC, are strongly recommended for the Applications of ultra performance liquid chromatography to
purpose of quality control, since they might represent traditional Chinese Medicines. J Chromatogr Sci 2010;48:19e21.
appropriately the “chemical integrities” of the herbal [20] Sarker S, Nahar L. Hyphenated techniques and their applications
products and therefore be used for authentication and in natural products analysis. In: Natural products isolation. 3rd
ed. USA: Humana Press e Springer-Verlag; 2012.
identification. With the continuing progress in computa- [21] Li M, Hou X-F, Zhang J, Wang S-C, Fu Q, He L-C. Applications of
tional technology, hyphenation efficiency, and afford- HPLC/MS in the analysis of traditional Chinese medicines.
ability, it is envisaged that HPLC-based analysis of J Pharm Anal 2011;1:81e91.
herbal products will become even more attractive to [22] Shaw L-H, Lin L-C, Tsai T-H. HPLC-MS/MS analysis of a tradi-
the analysts dealing with efficacy and safety of herbal tional Chinese medical formulation of Bu-Yang-Huan-Wu-Tang
and its pharmacokinetics after oral administration to rats. Plos
products in the years to come. One 2012;7:e43848.
[23] McMaster MC. LC-MS: a practical user’s guide. New Jersey: John
Wiley & Sons; 2005.
References [24] Yang DZ, Yin XX, Ong CN, Tang DQ. Multidimensional
[1] Choi CK, Dong MW. Sample preparation for HPLC analysis of information-based HPLC technologies to evaluate traditional
drug products. Sep Sci Technol 2005;6:123e44. Chinese medicine. J Chromatogr Sci 2013;51:716e25.
[2] Waksmundzha-Hajnos M, Oniszczuk A, Szewczyk K, [25] Wold S. Chemometrics: what do we mean with it, and what do we
Wianowska D. Effect of sample preparation methods on the want from it? Chemom Intell Lab Sys 1995;30:109e15.
HPLC quantification of some phenolic acids in plant materials. [26] Brenton RG. Chemometrics: data analysis for the laboratory and
Acta Chromatogr 2007;19:227e37. chemical plant. Chichester: John Wiley and Sons; 2003.
[3] Sadek PC. Sample preparation techniques for HPLC and GC [27] Gemperline P. Practical guide to chemometrics, CRC. Boca Raton:
analyses. Chichester: John Wiley and Sons; 2008. Florida; 2006.
[4] Pan J, Zhang C, Zhang Z, Lim G. Review of online coupling of [28] Jing D, Linfang H. Application of chemometrics in quality evalu-
sample preparation techniques with liquid chromatography. ation of medicinal plants. J Med Plants Res 2006;5:4001e8.
Anal Chim Acta 2014;815:1e15. [29] Lavine B, Workman J. Chemometrics. Anal Chem 2010;82:
[5] Wen Y, Chen L, Li J, Liu D, Chen L. Recent advances in solid-phase 4699e711.
sorbents for sample preparation prior to chromatographic [30] Bansal A, Chhabra V, Rawal RK, Sharma S. Chemometrics: a new
analysis. Trends Anal Chem 2014;59:26e41. [Link] scenario in herbal drug standardization. J Pharm Anal 2014;4:
10.1016/[Link].2014.03.011. 223e33. [Link]
[6] Li DQ, Zhao J, Xie J, Li SP. A novel sample preparation and on-line [31] Jolliffe IT. Principal component analysis. 2nd ed. Germany:
HPLC-DAD-MS/MS-BCD analysis for rapid screening and char- Springer-Verlag; 2002.
acterization of specific enzyme inhibitors in herbal extracts: case [32] Gad HA, El-Ahmady SH, Abou-Shoer MI, Al-Azizi MM. Applica-
study of a-glucosidase. J Pharm Biomed Anal 2014;88:130e5. tion of chemometrics in authentication of herbal medicines: a
[7] Huie CW. A review of modern sample-preparation techniques for review. Phytochem Anal 2012;24:1e24.
the extraction and analysis of medicinal plants. Anal Bioanal [33] Chowdhury HR, Mondal S. Morphological and ethno-
Chem 2002;373:23e30. medicinal consideration of Euphorbia fusiformis Buch.-Ham.
424 19. APPLICATIONS OF HIGH PERFORMANCE LIQUID CHROMATOGRAPHY IN THE ANALYSIS OF HERBAL PRODUCTS

Ex D. Don: some new observations from district Birbhum, West [52] Wang B, Shen L, Cong WJ, Lin X, Feng Y, Zhu YY, et al. A simple
Bengal, India. Int J Res Ayurveda Pharm 2012;3:113e6. HPLC method for simultaneous analysis of 7 bioactive constitu-
[34] Lal UR, Tripathi SM, Jachak SM, Bhutani KK, Singh IP. HPLC ents for Liuwei Dihuang pill and its application in quality consis-
analysis and standardisation of Arjunarishta e an Ayurvedic car- tency evaluation. Anal Methods 2013;5:2384e90.
dioprotective formulation. Sci Pharm 2009;77:605e16. [53] Kim JH, Shin HK, Seo CS. Simultaneous determination of 13
[35] Akarasereenont P, Thitilertdecha P, Chotewuttakom S, Palo T, chemical marker compounds in Gwakhyangjeonggi-san, an herb-
Seubnooch P, Wattanarangsan J, et al. Chromatographic finger- al formula, with validated analytical methods. Nat Prod Commun
print development for quality assessment of “Ayurved Siriraj Pra- 2014;9:65e9.
sachandaeng” antipyretic drug. Siriraj Med J 2010;62:4e8. [54] Lazarowych NJ, Pekos P. Use of fingerprinting and marker com-
[36] Rane R, Gangolli D, Salkar K, Chotalia C, Suthar A. Application of pounds for identification and standardisation of botanical drugs:
classical and instrumental methods for evaluation of polyherbal strategies for applying pharmaceutical HPLC analysis to herbal
capsule formulation. Int J Res Ayurveda Pharm 2014;5:23e5. products. Drug Inf J 1998;32:497e512.
[37] Zhu LC, Yang X, Tan J, Wang BC, Zhang X. A validated high [55] Yuan D, Yamamoto K, Bi K, Zhang P, Liu F, Kano Y. Studies on the
performance liquid chromatograph-photodiode array method marker compounds for standardisation of traditional Chinese
for simultaneous determination of 10 bioactive components in medicine “Polyporus sclerotium”. Yakugaku Zasshi 2003;123:
compounds hongdoushan capsule. Pharmacogn Mag 2014;10: 53e62.
83e8. [56] Tang W, Wan M, Zhu Z, Chen G, Huang X. Simultaneous determi-
[38] Sulaiman SM, Rajashekhar G, Prakash PJ, Singh DS, Saleem C. nation of eight major bioactive compounds in Dachengqi Tang
Evaluation of safety and immunotoxicity of Immunol, a polyher- (DT) by high-performance liquid chromatography. Chin Med
bal formulation in rats. J Pharm Toxicol 2010;5:262e74. 2008;3:1e6.
[39] Tejasvi PR, Satyavati D, Begum SA. Anxiolytic activity of Jessica e [57] Kushwaha SKS, Kushwaha N, Maurya N, Rai A. Role of markers
a polyherbal formulation. Int J Pharm Pharm Sci 2013;5:384e8. in the standardisation of herbal drugs: a review. Arch Appl Sci Res
[40] Hewavithana T, Ranaweera KKDS, Tissera MHA, Yapa PAJ. Use 2010;2:225e9.
of physicochemical properties, chromatographic and spectropho- [58] Kwok K-Y, Xu J, Ho H-M, Chen H-B, Li M, Lang Y, et al. Quality
tometric measurements in the standardisation of Srilankan poly- evaluation of commercial Huang-Lian-Jie-Du-Tang based on
herbal formulation “Maha Varthikava Watee”. Intl J Res simultaneous determination of fourteen major chemical constitu-
Ayurveda Pharm 2012;3:493e6.22. ents using high performance liquid chromatography. J Pharm
[41] Soni H, Patgiri B, Bhatt S. Quantitative determination of three con- Biomed Anal 2013;85:239e44.
stituents of Rasayana Churna (a classical Ayurvedic formulation) [59] Malasoni R, Srivastava A, Pandey RR, Srivastava PK, Swivedi AK.
by reversed phase HPLC. Int J Res Ayurveda Pharm 2014;5:17e22. Development and validation of improved HPLC method for
[42] Kajaria DK, Gangwar M, Kumar D, Sharma AK, Tilak R, Nath G, the quantitative determination of curcuminoids in herbal
et al. Evaluation of antimicrobial activity and bronchodialator ef- medicament. J Sci Ind Res 2013;72:88e91.
fect of a polyherbal drug-Shrishadi. Asian Pac J Trop Biomed [60] Weon JB, Ma JY, Yang HJ, Lee B, Yun B-R, Ma CJ. Qualitative and
2012:905e9. quantitative analysis of nine majot compounds in the
[43] Neergheen-Bhujun VS, Munogee N, Coolen V. Antioxidant and Bozhougyiqi-Tang using a high-performance liquid chromatog-
anti-inflammatory efficacies of polyherbal formulations and raphy coupled with a photodiode array detector and electrospray
elixirs traditionally used in Mauritius for the treatment of rheu- ionization mass spectrometer. Pharmacogn Mag 2013;9:271e82.
matoid arthritis. J Herbal Med 2014;4:1e9. [61] Sudani RJ, Akbari BV, Vidyasagar G, Sharma P. Quantitative and
[44] Chouhan B, Kumawat RC, Kotecha M, Ramamurthy A, Nathani S. chromatographic fingerprint analysis of Embelia ribes churna for-
Triphala: a comprehensive Ayurvedic review. Int J Res Ayurveda mulations by HPLC method. Int J Pharm Biol Arch 2011;2:657e63.
Pharm 2014;4:612e7. [62] Basar N, Talukdar AD, Nahar L, Stafford A, Kushiev H, Kan A,
[45] Balammal G, Babu MS, Reddy PJ. Analysis of herbal medicines by et al. A simple semi-preparative recersed-phase HPLC/PDA
modern chromatographic techniques. Int J Preclin Pharm Res method for separation and quantification of glycyrrgizin in nine
2012;3:50e63. samples of Glycyrrhiza glabra root collected from different
[46] Liang Y-Z, Xie P, Chan K. Quality control of herbal medicines. geographical origins. Phytochem Anal 2014;25:399e404. http://
J Chromatogr B 2004;812:53e70. [Link]/10.1002/pca.2507.
[47] Gao W, Yang H, Qi LW, Liu EH, Ren MT, Yan YT, et al. Unbiased [63] Ghafari S, Esmaelli S, Aref H, Naghibi F, Mosaddegh M. Qualita-
metabolite profiling by liquid chromatography-quadrupole time- tive and quantitative analysis of some brands of Valerian pharma-
of-flight mass spectrometry and multivariate data analysis for ceutical products. Ethnomedicine 2009;3:61e4.
herbal authentication: classification of seven Lonicera species [64] Wang GL, Wei M, Wang J, Lu Y, Mahady GB, Liu D. High-
flower buds. J Chromatogr A 2012;1245:109e16. performanceliquid chromatography with photo diode array
[48] Wang LB, Wang XB, Kong LY. Automatic authentication and (HPLC-PDA) quality control of menoprogen, a Traditional Chi-
distinction of Epimedium joreanum and Epimedium wushanense nese Medicine (TCM) formula used for the management of
with HPLC fingerprint analysis assisted by pattern recognition menopause. Int J Med Plants Res 2013;2:146e51.
techniques. Biochem Syst Ecol 2012;40:138e45. [65] Zhou J, Qi L, Li P. Quality control of Chinese herbal medicines
[49] Gao X, Yang XW, Marriott PJ. Evaluation of Coptidis Rhizoma- with chromatographic fingerprint. Chin J Chromatogr 2008;26:
Euodiae Fructus couple and Zuojin products based on HPLC 153e9.
fingerprint chromatogram and simultaneous determination of [66] Feng C, Ruan J-L, Cai Y-L. Study of high-performance liquid chro-
main bioactive constituents. Pharm Biol 2013;51:1384e92. matography fingerprint for traditional Chinese medicine Yigong-
[50] Zou H-B, Du A-Q, Zhang X-L, Wei P-H, Lu W-J, Yang G-S, et al. ningxue oral liquid. J Chem Soc Pak 2010;32:34e9.
Quality control methodology and their application in analysis [67] Zhou D, Zhu YR, Guan Y, Quan W, Li YW, Guo C, et al. Chemical
on HPLC fingerprint spectra of herbal medicine. Chromatogr fingerprint and metabolic fingerprint of Danhong injection by
Res Int 2012;2012:1e12. HPLC-UV and HPLC-MS. Asian J Chem 2013;25:6285e92.
[51] Reyes GCDL, Koda RT. Determining hyperforin and hypericin [68] Chen F, Li H-L, Tan Y-F, Zhang J-Q. Quantitative analysis
content in eight brands of St John’s wort. Am J Health-Syst Pharm of the major constituents in Chinese medicinal preparation Suo-
2002;59:545e7. Quan formulae by ultra fast high performance liquid
 LIST OF ABBREVIATIONS 425
chromatography/quadrupole tandem mass spectrometry. Chem [85] Tang SR, Yang RT, Pan FS, Zhao AM, Pang ZJ. Steroidal saponin
Cent J 2013;7:131e8. and steroidalsapogenin in Chinese Dioscorea L. J Plant Resour
[69] Chen F-F, Qi H-Y, Shi Y-P. Fingerprint analysis of Codonopsis Radix Environ 2007;16:64e72.
by HPLC coupled with chemometrics analysis. Chin Herbal Med [86] Zhang X, Liang J, Liu J, Zhao Y, Gao J, Sun W, et al. Quality control
2013;5:307e12. and identification of steroid saponins from Dioscoreazingiberen-
[70] Goodarzi M, Russell PJ, Heyden V. Similarity analyses of chro- sis C. H. Wright by fingerprint with HPLC-ELSD and HPLC-
matographic herbal fingerprints: a review. Anal Chim Acta ESI-Quadrupole/Time-of-fight tandem mass spectrometry.
2013;804:16e28. J Pharm Biomed Anal 2014;91:46e59.
[71] Feng X, Kong WJ, Wei JH, Ou-Yang Z, Yang M. HPLC fingerprint [87] Man S, Gao W, Zhang Y, Wang J, Zhao J, Huang L, et al. Qualita-
analysis combined with chemometrics for pattern recognition of tive and quantitative determination of major saponins in Paris and
ginger. Pharm Biol 2014;52:362e7. Trillium by HPLC-ELSD and HPLCeMS/MS. J Chromatogr B
[72] Funari CS, Carneiro RL, Andrade AM, Hilder EF, Cavalheiro AJ. 2010;878:2943e8.
Green chromatographic fingerprinting: an environmentally [88] National Commission of Chinese Pharmacopoeia. Pharmacopoeia
friendly approach for the development of separation methods of peoples Republic of China. Beijing: Chemical Industry Press;
for fingerprinting complex matrices. J Sep Sci 2014;37:37e44. 2010. 525e526.
[73] Kiazolu JB, Zhang LY, Intisar A, Wang Y, Zhang RS, Wu ZP, et al.
RP-HPLC separation and statistical data processing of different
batches of aerial parts of Jologbo. J Liq Chromatogr Relat Technol
2014;37:48e60.
[74] Ma Y, Gao HM, Liu J, Chen LM, Zhang QW, Wang ZM. Identifica-
tion and determination of the chemical constituents in a herbal LIST OF ABBREVIATIONS
preparation, compound Kushen injection, by HPLC and LC-
DAD-MS/MS. J Liq Chromatogr Rela Technol 2014;37:207e20. ACN Acetonitrile
[75] Xiong HS, Yu LX, Qu HB. A weighting approach for chromato- ELSD Evaporative light scattering detector
graphic fingerprinting to ensure the quality consistency of botan- EMEA European Medicines Agency
ical drug products. Anal Methods 2014;6:476e81. ESI-MS Electrospray ionization mass spectrometry
[76] Zhu HB, Wang CY, Qi Y, Song FR, Liu ZQ, Liu SY. Fingerprint EtOH Ethanol
analysis of Radix Aconiti using ultra-performance liquid EU European Union
chromatography-electrospray isonisation/tandem mass HCA Hierarchical clustering analysis
spectrometry (UPLC-ESI/MSn) combined with stoichiometry. HELP Heuristic evolving latent projections
Talanta 2013;103:56e65. HPLC High performance liquid chromatography
[77] Zhang J-I, Cui M, Yu H-I, Guo D-A. Chemical fingerprint and HPTLC High performance thin layer chromatography
metabolic fingerprint analysis of Danshen injection by HPLC-UV IR Infrared
and HPLC-MS methods. J Pharm Biomed Anal 2005;36:1029e35. IT Information theory
[78] Tang D-Q, Zheng X-X, Chen X, Chen X, Yang D-Z, Du Q. Quanti- LC Liquid chromatography
tative and qualitative analysis of common peaks in chemical LDA Linear discriminate analysis
fingerprint of Yuanhu Zhitong tablet by HPLC-DAD-MS/MS. LLS Local least square
J Pharm Anal 2014;4:96e106. MeOH Methanol
[79] Bian Q, Yang H, Chan C-O, Mok DK-W, Chen S. Fingerprint anal- MS Mass spectrometry
ysis and simultaneous determination of phenolic compounds in NMR Nuclear magnetic resonance
extracts of Curculiginis Rhizoma by HPLC-Diode Array NP Normal-phase
detector. Chem Pharm Bull 2013;61:802e8. OPA Orthogonal projection analysis
[80] Jiang F, Tao Y, Shao Y. Fingerprinting quality control of Qianghuo PCA Principal component analysis
by high performance liquid chromatography-photodiode array PDA Photo diode array
detection. J Ethnopharmacol 2007;111:265e70. RP Reversed-phase
[81] Lu HM, Liang YZ, Chen S. Identification and quality assessment SCC Spectral correlative chromatography
of Houttuynia cordata injection using GCeMS fingerprint: a stan- SFDA State Food and Drug Administration
dardization approach. J Ethnopharmacol 2006;105:436e40. SIMCA Supervised classification procedure using soft independent
[82] Xu CJ, Liang YZ, Chau FTV, Heyden YV. Pretreatments of chro- modeling of class analogy
matographic fingerprints for quality control of herbal medicines. TCM Traditional Chinese Medicine
J Chromatogr A 2006;1134:253e9. TFA Tri-fluoro-acetic acid
[83] Theodoridis GA, Gika HG, Want EJ, Wilson ID. Liquid chroma- TLC Thin layer chromatography
tographyemass spectrometry based global metabolite profiling: TOF Time-of-flight
a review. Anal Chim Acta 2012;711:7e16. UPLC Ultra performance liquid chromatography
[84] Li L, Zhao Y, Liu W, Feng F, Xie N. HPLC with quadrupole TOF- UPLCeESI/MSn Ultra performance liquid chromatography-
MS and chemometrics analysis for the characterisation analysis electrospray ionization/mass spectrometry
for the characterisation of Folium Turpiniae from different UV Ultraviolet
regions. J Sep Sci 2013;36:2552e61. WHO World Health Organization