Received: 12 June 2019 | Accepted: 17 July 2019

DOI: 10.1002/jcp.29118

ORIGINAL RESEARCH ARTICLE

Genomic analysis reveals association of specific SNPs with

athletic performance and susceptibility to injuries in
professional soccer players

Raffaele La Montagna1* | Raffaele Canonico2* | Luigi Alfano3 | Enrico Bucci1 |

Silvia Boffo1 | Leopoldo Staiano4 | Beniamino Fulco5 | Enrico D’Andrea1 |
Antonio De Nicola1 | Patrizia Maiorano5 | Costantino D’Angelo5 |
Andrea Chirico1,6 | Alfonso De Nicola1 | Antonio Giordano1,5

1
Sbarro Health Research Organization,
Wayne, Pennsylvania Abstract
2
Unità Operativa Complessa di Dietetica, The development of specific and individualized training programs is a possible way to
Medicina dello Sport e Benessere Psico‐Fisico,
improve athletic performance and minimize injuries in professional athletes. The
Università degli Studi della Campania Luigi
Vanvitelli, Naples, Italy information regarding the sport's physical demands and the athletes’ physical profile have
3
Cell Biology and Biotherapy Unit, Istituto been, so far, considered as exhaustive for the design of effective training programs.
Nazionale Tumori ‐ IRCCS, Fondazione G.
Pascale, Naples, Italy However, it is currently emerging that the genetic profile has to be also taken into
4
Laboratory of Cellular and Developmental consideration. By merging medical and genetic data, it is thus possible to identify the
Biology, Stazione Zoologica Anton Dohrn,
athlete's specific attitude to respond to training, diet, and physical stress. In this context,
Naples, Italy
5
Department of Medical Biotechnologies,
we performed a study in which 30 professional soccer players, subjected to standard sport
University of Siena, Italy medical evaluation and practices, were also screened for genetic polymorphism in five key
6
Department of Psychology of Development genes (ACTN3, COL5A1, MCT1, VEGF, and HFE). This genetic analysis represents the central
and Socialization Processes, “Sapienza”
University of Rome, Italy point of a multidisciplinary method that can be adopted by elite soccer teams to obtain an
improvement in athletic performance and a concomitant reduction of injuries by tailoring
Correspondence
Antonio Giordano MD PhD, Sbarro Health training and nutritional programs. The genetic fingerprinting of single athletes led to the
Research Organization, BioLife Science Bldg. identification of two performance‐enhancing polymorphisms (ACTN3 18705C>T, VEGF‐
Suite 431 1900 N 12th Street, Philadelphia,
PA 19122. 634C>G) significantly enriched. Moreover, we derived a genetic model based on the gene
Email: giordano@[Link] set analyzed, which was tentatively used to reduce athletes’ predisposition to injuries, by
dictating a personalized nutrition and training program. The potential usefulness of this
approach is concordant with data showing that this team has been classified as the
healthiest and least injured team in Europe while covering the highest distance/match with
the highest number of high‐intensity actions/match.

KEYWORDS
athletic performance, genes, injuries, professional soccer players, single‐nucleotide polymorphisms

Abbreviations: ACTN3, actin alpha‐3; COL5A1, collagen Type V alpha 1 chain; HFE, hemostatic iron regulator (high‐Fe); HIF1, hypoxia inducible factor 1; MCT1 (alias SLC16A1), solute carrier
family 16 member 1; MCTs, monocarboxylate transporters; PEP, performance enhancing polymorphism; S.S.C. Naples, società Sportiva Calcio Napoli; VEGF, vascular endothelial growth factor;
VO2 Max, maximal oxygen uptake.

*Raffaele La Montagna and Raffaele Canonico contributed equally to this work.

J Cell Physiol. 2019;1–10. [Link]/journal/jcp © 2019 Wiley Periodicals, Inc. | 1

2 | LA MONTAGNA ET AL.

1 | INTRODUCTION communities (Lega Serie A and UEFA). As a unique physical‐genetic

entity, each athlete was subjected to genetic screening and accurate
Training for elite athletes is a complex task that needs to be adapted physical monitoring during the entire season. The purpose of these
not only to specific sports but also to the individual predisposition of analyses was to explore the connection between gene polymorph-
every single athlete. Soccer is considered a total and very demanding isms considered to be important for athletic performance and the
sport that requires the concomitant training of different body actual training efficacy, with a particular focus on the effect that
systems, such as muscular, cardio‐circulatory, respiratory, and genetic differences have on athletes’ competitiveness, endurance,
nervous system often supported by specific nutrient supplementa- and resistance to injuries.
tion. This multiparametric training fulfils different physical and For athletes’ profiling and coaching, several professionals were
nutritional requirements that, in turn, might improve muscle size involved including physicians, geneticists, nutritionists, and phy-
and strength, endurance, resistance to fatigue, and coordination. siotherapists. Review of data in the literature allowed us to identify a
Although the physical and metabolic features of each athlete are set of five genes that were previously correlated with athletic
considered to be the main parameters to design specific training performance, such as endurance capacity, lactic acid clearance, iron
programs, the impact of genetic traits on the success of a given intake and clearance, and muscle fatigue. These genes have also
training is currently accepted. Advances in science and technology specific polymorphisms that are responsible for increased or
could in principle change the way in which soccer players are decreased gene activity. Specifically, we focused our attention on:
subjected to training programs designed from a personalized MCT1 (alias SLC16A1, Solute Carrier Family 16 Member 1), a lactate
perspective. The inclusion of genetics in the design of training transporter protein (Merezhinskaya, Fishbein, Davis, & Foellmer,
programs (PEP, performance‐enhancing polymorphisms) was trig- 2000); HFE (Homeostatic Iron Regulator) involved in blood and tissue
gered by the identification of a large number of genetic variants, iron homeostasis (Luszczyk et al., 2017); ACTN3 (Actin alpha 3) a
linked to athletic performance (Ostrander, Huson, & Ostrander, protein responsible for fast muscle fiber functioning (Lee, Houweling,
2009). The correlation between specific gene polymorphisms and the North, & Quinlan, 2016), and COL5A1 (Collagen Type V Alpha 1 Chain)
response to physical and nutritional cues can then be explored to mainly found in ligaments and tendons and associated with severity
evaluate athletes’ profiles (Ferioli et al., 2019). This would allow for of musculoskeletal injuries (Massidda et al., 2015a). Moreover, we
the development of tailored programs that maximize athletic analyzed a polymorphism in the VEGF gene, encoding the vascular
performance and reduce the occurrence of indirect injuries. The endothelial growth factor, associated with an improvement in aerobic
medical staff decided to test this new scientific approach and has performance (Prior et al., 2006). The aim of the present study was to
defined a new physical, genetic, and nutritional multiparametric analyze the concordance between the information available on the
approach for soccer training aiming for: (a) Improvement of athletic effects of the selected PEPs on athletic performance and the actual
performance; (b) reduction of injuries; and (c) amelioration of performance of the team. Focusing on a set of genes, which were
athletes’ health. Since the program started in 2013, players trained already validated by other authors on large samples of professional
by this medical staff are considered the “healthiest” in terms of injury athletes, is a way to avoid the risk of finding spurious genetic
susceptibility and performance (Figure 1) by both scientific and sport correlations in the limited population target of the present study.

F I G U R E 1 Distribution and frequency of performance‐enhancing polymorphisms in S.S.C. Naples soccer players. (a–e) Genes, alleles, and
genotype frequency in the 30 S.S.C. Naples soccer players. In black, the homozygous wild type genotype; in light gray, the heterozygous and in
the dark gray, the homozygous polymorphic genotype
LA MONTAGNA ET AL. | 3

The data obtained by the genetic analysis and the physical and 40 bases in length (sequences provided in Table 1). The PCR
information of each athlete were correlated, to see whether single reaction was performed as follows: 94°C for 3 min (1 cycle); 94°C for
genetic polymorphisms or a combination thereof could be used to 30 s , −55°C for 1 min, −72°C for 1 min (for 35 cycles), and 72°C for
retrospectively predict the injury rate of each athlete. 10 min. The PCR product was analyzed on the 1% agarose gel and
extracted with the QIAquick Gel Extraction Kit (Qiagen) per the
manufactory's instructions. The samples were then sequenced at the
2 | METHODS Thomas Jefferson Genomic Facility: Sanger sequencing was per-
formed in 100 ng of PCR product, using 3.2 μM the forward primer.
The soccer team's medical staff authorized the use of samples for For the evaluation of SNPs in the HFE, COL5A1, ACTN3, and
screening, as described in the protocol of this project. Athletes VEGF, we performed an allelic discrimination by TaqMan assay using
authorized the use of the results for scientific scope. the following specific reagents: two unlabeled PCR primers (forward
and reverse, Table 1) at 900 nM final concentration; Applied
Biosystems™ VIC™ and Applied Biosystems™ FAM™ dyes (MGB‐
2.1 | Study participants labeled probes), both at 200 nM final concentration. Thermocycling
In total, 30 soccer players with different nationalities were recruited was performed in real‐time PCR instrument platforms [Roche, Light
and included in the study. Cycler 480II], using the specific Applied Biosystems™ TaqMan™
Genotyping Master Mix following standard conditions. In particular,
the reaction is divided into two steps. Initial steps: AmpErase UNG
2.2 | Genomic DNA extraction activation (2 min at 50°C); second step: AmpliTaq Gold DNA
polymerase activation (10 min at 95°C), followed by 40 cycles of
The genomic DNA was extracted from saliva, by using the Oragene‐DNA
PCR (Melt 15 s at 95°C and Anneal/Extend 1 min at 60°C).
(OG‐500) of the DNA Genotek Inc., following manufactory instructions.
Briefly, 2 ml were collected from a single soccer player; 500 μl of the
mixed sample was transferred to the microcentrifuge tube followed by
2.4 | Statistical analysis
adding of 20 μl (1/25th volume) of PT‐L2P (DNA Genotek), mixed by
vortexing and incubated for 10 min on ice. The sample was centrifuged To exclude sampling biases, we analyzed the consistency between
for 5 min at 15,000 × g to pellet the impurities. About 600 μl of 95% the tested genotype frequencies with a Hardy–Weinberg equilibrium
EtOH was added to the 500 μl of cleared supernatant and mixed by by using a χ2 test, verifying whether the probability of obtaining the
inversion ten times. The sample was allowed to stand at room calculated χ2 value by chance only was larger than 0.05.
temperature for 10 min for the DNA precipitation, followed by To test the significance of the difference in frequency of any
centrifugation for 2 min at 15,000 × g. The pellet was washed by adding single genotype compared to a given reference frequency, we used
250 μl of 70% EtOH and centrifugated for 2 min at 15,000 × g. The the χ2 test as recommended by Campbell (2007) and Richardson
genomic DNA was resuspended in 100 μl of TE solution and stored at (2011), assuming alpha = 0.05.
−20°C. To obtain a quantitative predictor of injury risk, based on the
genotype of the five selected genes, we fitted the number of muscular
or ligament injuries occurred for each team player with available
2.3 | Genotyping
genotype information (n = 27) by a logistic (or logit) multinomial model.
For the evaluation of SNP in the MCT1 gene, we amplified the This is a generalization of the logistic model with a variable
genomic DNA by using the PfU ultra HF as manufactory instructions. response on more than two categories. We used this method to
Primers containing the desired mutation were designed between 20 understand the effect of a qualitative variable series (the five selected

T A B L E 1 List of specific primers used for the allelic discrimination by TaqMan assay

SNP ID Gene Context sequence [VIC/FAM]

rs1800562 HFE C28Y CCTGGGGAAGAGCAGAGATATACGT[G/A]CCAGGTGGAGCACCCAGGCCTGGAT
rs1799945 HFE H63D TGACCAGCTGTTCGTGTTCTATGAT[C/G]ATGAGAGTCGCCGTGTGGAGCCCCG
rs1800730 HFE S65C GCTGTTCGTGTTCTATGATCATGAG[A/T]GTCGCCGTGTGGAGCCCCGAACTCC
rs12722 COL5A1 CCGCCCCACGCTCTGTCCACACCCA[C/T]GCGCCCCGGGAGCGGGGCCATGCCT
rs1815739 ACTN3 R577X CAAGGCAACACTGCCCGAGGCTGAC[T/C]GAGAGCGAGGTGCCATCATGGGCAT
rs2010963 VEGF CGCGCGGGCGTGCGAGCAGCGAAAG[C/G]
GACAGGGGCAAAGTGAGTGACCTGC
MCT1 Forward: 5′‐ACACATACTGGGCATGTGGCGTCGT
Reverse: 5′‐AAATCCCATCAATGAACAACTGGTATGATTTCCAC
4 | LA MONTAGNA ET AL.

genotypes) on the response variable (the number of muscular or and on the differences between elite athletes and the general
ligament injuries occurring to each included player). population. However, the exact genotype plays a role in differentiating
The obtained model fitting to the available data was good, with a among athletes of the same discipline: in a study on 200 Brazilian
R²(McFadden) = 0.953, R²(Cox and Snell) = 0.876 and R²(Nagelk- soccer players, the presence of a CC genotype was associated with
erke) = 0.986. players fastest in short distances and with higher jump potential, while
With a χ value of 56.446, testing the null hypothesis of getting
2
the TT genotype was associated with higher aerobic capacity (Pimenta
such a fitting by chance with the available degrees of freedom et al., 2013). The soccer team analyzed in this project is characterized
yielded a p = .016. by a high frequency of TT homozygotes, which is statistically different
All calculations were performed using the software NodeXL – v. from controls or athletes enrolled in all the above‐mentioned studies.
2018.5. In line with results reported for soccer players (Pimenta et al., 2013),
this might indicate that, on average, athletes have a higher aerobic
capacity. Furthermore, very recently, the TT genotype has been
3 | RES U LTS strongly associated also with a higher incidence of muscle injuries
(Massidda et al., 2017). The overall athletic performances, recognized
Here we report a detailed analysis of the incidence of PEP in the by the Lega Calcio Serie A, agreed with a possible effect of the
genes of interest in 30 soccer players, highlighting the predicted optimized training program (Figure 2a).
effects of a given PEP on athletic performance.
Among the five genes selected for this study, ACTN3, COL5A1,
3.2 | Collagen Type V alpha 1
and VEGF influence athletic performance and predisposition to
indirect injury, independently of any nutritional supply. The effect Collagens are among the main components of connective tissues and
of these PEPs may be enhanced or counteracted, perhaps developing the extracellular matrix. Many diseases, collectively known as
personalized training. However, there are two other genes (MCT1 collagenopathies, are associated with mutations and/or polymorphisms
and HFE) in which PEPs could be used in the coordination of a in different types of collagen (Jobling et al., 2014). Among the 13 types
nutritional intervention along with the development of specific of collagen, Collagen V is mainly enriched in tendons and ligaments
training programs to modify their positive or negative influence. where it accounts for 2% of total collagen (mainly type I). Although
What follows, is a detailed analysis of the PEP genetic makeup found Collagen V is quantitatively a minor component, it has a fundamental
for each gene in the soccer team, and then a discussion of their combined role in the regulation of the assembly of functional collagen fibers. It is
predicting power when a suitable quantitative biomarker model is composed of two α1 chains and two α2 chains, and COL5A1 encodes
applied. specifically for the Collagen type V alpha 1 chain. A single‐nucleotide
polymorphism C to T (rs12722) of COL5A1 gene occurs in the 3′‐
untranslated region of mRNA improving its stability and increasing the
3.1 | Actin alpha‐3
protein amount, consequently (Abrahams, Posthumus, & Collins, 2014;
ACTN3 encodes α‐actinin‐3, a protein that participates to the Bertuzzi et al., 2014). The higher amount of Collagen V alters the
composition of sarcomeres, the functional units of striated muscles; mechanical properties of tendons resulting in an increased predisposi-
it is particularly abundant in the fast type II muscle fibers, which are tion to tendon and ligament injuries. COL5A1 rs12722 polymorphism
involved in the muscle sprinting and strength activities (Blanchard, distribution in the cohort analyzed in this study highlights a prevalence
Ohanian, & Critchley, 1989). The ACTN3 R577X (rs1815739) of the heterozygous (CT) genotype, which accounts for the majority of
polymorphism has been suspected for several years to be associated athletes (44%) compared with the homozygous genotypes CC (22%)
with particular athletic performances in power‐oriented elite athletes and TT (34%) (Figure 1b). Recently, Heffernan et al. (2017) described a
of different ethnicities (Yang et al., 2003). This polymorphism is a small frequency of the CC allele associated with better performance in
genomic C to T transition in position 18705 coding for an arginine to X rugby players. The frequency of CC genotype in the analyzed soccer
substitution in position 577 (R577X). The polymorphism can be players is similar to that of the professional rugby players included in
present in homozygosis (DNA: TT or protein: XX), in heterozygosis the study by Heffernan et al. However, the number of athletes included
(DNA: CT or Protein: RX) or absent (DNA: CC or protein: RR). The X‐ in the present study is too limited for the detection of the expected
allele encodes a shorter variant of ACTN3 that stops at amino acid small effect (insufficient sample power), so that is not possible to draw
577 instead of 901 resulting in a protein with lower performance any significant conclusion on whether the soccer players deviate
capabilities. The CC (RR at protein level) and CT (RX at protein level) significantly from the general population (i.e., the control sample
genotypes were initially found to be increased in frequency in elite reported in the study by Heffernan et al., 2017).
strength athletes, in agreement with the fact that the TT (XX) mutation
was previously associated with impaired muscle performance (Roth
3.3 | Vascular endothelial growth factor
et al., 2008). The enrichment in CC and CT genotypes, however, was
not confirmed by different large‐scale studies (Massidda et al., 2015; VEGF is a member of the growth factor family that plays a pivotal
Rankinen et al., 2016), casting doubts on the original interpretation role in the formation of new and functional blood vessels (Hoeben
LA MONTAGNA ET AL. | 5

F I G U R E 2 Analysis of athletic performances and injury rate of S.S.C. Naples athletes (adapted from Ranking Serie A‐Report Fisico (a) and
UEFA elite club injury study (b), (c)). (a) In the graph the report of the athletic performance in the 38 matches of the Italian championship
2017–2018 is presented. The black line indicates the number of high‐intensity actions and the red line indicates the distance covered at high
speed ( > 15 km/hr) by the 20 teams of Serie A. (b), (c) The graphs presented are adapted from the mid‐season report (2017–2018) of the UEFA
Elite club injury study. In (b) the data regarding the number of training sessions/month and the incidence of injuries/1,000 hr of training are
presented. It is clear that S.S.C. Naples is the Team, among the other participants to the UEFA Champions League, with the highest number of
training session/month and with the lowest incidence of indirect injuries. In (c) the data regarding the number of matches/month and the
incidence of injuries/1,000 hr of match are presented. It is clear that S.S.C. Naples is one of the teams with the highest number of matches/
month and with the lowest incidence of indirect injuries. These analysis were conducted by Professor Jan Ekstrand and are reported in The
UEFA Elite Club Injury Study (ECIS), funded and supported by UEFA. Data reported above correspond to 2017/2018 middle season analysis.
This trend has also been confirmed in the 2017/18 Final Season Report performed by Professor Jan Ekstrand (data not shown) [Color figure can
be viewed at [Link]]
6 | LA MONTAGNA ET AL.

et al., 2004). VEGF is expressed in a variety of human tissues, tight connection that is needed among athletes, medical staff and
including skeletal muscles, and it has been recently demonstrated to geneticists to allow for a tailored training program for each athlete
control the angiogenic processes that follow aerobic exercises that takes into account both physical and genetic features.
(Gustafsson, Puntschart, Kaijser, Jansson, & Sundberg, 1999;
Richardson et al., 2000). Aerobic exercise causes the development
3.4 | Monocarboxylate transporter 1
of a hypoxic environment in the muscles (Richardson, Noyszewski,
Kendrick, Leigh, & Wagner, 1995; Thompson, Pratley, & Ossowski, Lactate is widely considered as a negative influencer of athletic
1996) that induces the synthesis of hypoxia‐inducible‐factor 1 (HIF1) performance and it is often correlated with muscle fatigue and pain
which, in turn, upregulates the expression of VEGF, through the after exercise (Westerblad, Allen, & Lannergren, 2002). However, it is
prolonged mRNA stability, resulting in an increased protein function becoming more and more clear that lactate is an extremely important
(Liu et al., 2002). Being tightly linked to aerobic exercise and athletic fuel for a particular type of muscle fibers: the type I oxidative muscle
performance, angiogenesis thus represents a critical factor in the fibers. In general, during intense exercise, the glycolytic fibers
adaptation to the training‐induced increased oxygen consumption produce lactate and protons as end products of glycolysis. These
(VO2 Max) since it can foster the formation of new capillaries in the by‐products need to be excreted from the muscle fibers and released
muscles. Neo‐angiogenesis facilitates the delivery of oxygen by the into the blood. The excretion of lactate and protons mainly occur
red blood cells to the muscles, making it available to foster muscle through the monocarboxylate transporters (MCTs), a family of 14
metabolism. Previous data show that the amount of capillaries per members with different tissue localizations and expression levels
muscle fiber, whose formation is triggered by the activity of VEGF, is (Halestrap & Wilson, 2012). MCTs can mediate the inward/outward
increased in endurance athletes, indicating the requirement for an transfer of lactate/proton according to their chemical gradient (Juel
improved oxygen transport system to support aerobic activity & Halestrap, 1999). The most characterized MCTs are MCT1 and
(Brodal, Ingjer, & Hermansen, 1977). Although some aerobic training MCT4, with the latter controlling the export of lactate from glycolytic
has been demonstrated to control the generation of new capillaries fibers (Cupeiro et al., 2016). MCT1 is, by contrast, responsible of the
through VEGF upregulation, it is worth mentioning that genetic uptake of lactate by type I oxidative fibers and its activity has the
factors may also contribute to the success of angiogenesis in elite double beneficial effect of (a) reducing the circulating levels of lactate
athletes (Wahl et al., 2014). Variations of the DNA sequence in the and (b) increasing the levels of lactate in the muscle fibers that can
promoter of VEGF have been correlated with different levels of use it as fuel (Van Hall, 2000). Interestingly, MCT1 expression levels
VEGF expression (Al‐Habboubi, Sater, Almawi, Al‐Khateeb, & are increased upon high‐intensity training, enforcing its role in
Almawi, 2011). It has been reported that a C to G polymorphism in enhancing the athletic performance through lactate reabsorption. A
the VEGF promoter at position −634 is causative, when in PEP (rs1049434) in the MCT1 gene has been identified in position
homozygosis (GG), of a reduced expression of VEGF in hypoxic 1470, corresponding to A to T transversion (A1470T) which results in
conditions and this causes, in turn, a reduced maximal oxygen a glutamic acid to aspartic acid substitution in position 490 of the
consumption (Prior et al., 2006). However, in the same study, the protein (Glu490Asp). Compared with carriers of the A allele, the
authors have also demonstrated that whenever the polymorphism presence of the minor allele (T) is causative of a two‐fold reduction in
occurs in heterozygosis (CG), the VEGF expression under hypoxia is lactate transport by MCT1 (Merezhinskaya et al., 2000) and has been
null and VO2 Max is not affected. In light of these findings, we correlated with higher levels of blood lactate during high‐intensity
screened 30 soccer players for the rs2010963 polymorphism training (Cupeiro, Benito, Maffulli, Calderon, & Gonzalez‐Lamuno,
(−634C > G). The three possible genotypes CC, CG, and GG are 2010). These findings highlight the impact that MCT1 A1470T
represented as follows: the heterozygous (CG) is the single most polymorphism may have on endurance performances and on athlete's
diffuse genotype (46%), as compared with the homozygous geno- response to specific training. It has been demonstrated however that
types CC (22%) and GG (32%; Figure 1c). In a large study on 670 the frequency of the T allele of MCT1 was high in elite sprinters,
Russian athletes of different disciplines, a small but significant excess suggesting that reduced lactate uptake by type I oxidative fibers is
of the C allele was correlated to the higher aerobic performance of not adverse to anaerobic activities (Kikuchi et al., 2017; Sawczuk
athletes and to a better lactic acid metabolism (Akhmetov et al., et al., 2015). By contrast, the T‐allele may be deleterious for the
2008). As for the case of the ACTN gene, soccer players exhibit a aerobic and endurance activities (such as soccer players). We
significant excess of haplotype conferring better aerobic capacities, analyzed the occurrence of MCT1 A1470T polymorphism in the 30
with the CC homozygotes exceeding in a statistically significant soccer players; the three possible genotypes AA, AT, and TT turned
proportion the frequency measured both for the control and the out to be distributed with a prevalence of heterozygous (AT)
athlete populations in the Russian study, indirectly confirming the genotype, that accounts for most athletes (46%) as compared to
association of G‐634C polymorphism with physical performance of the homozygous genotypes AA (23%) and GG (31%; Figure 1d). In a
the athletes. This genetic makeup of the team agrees with the fact large study on 323 Russian athletes and 467 control individuals, a
that the soccer team under investigation has been the Italian soccer small excess of the AA genotype frequency in the athlete population
team, which covered the highest distance/match with the highest has been associated with higher in endurance (Fedotovskaya,
number of high‐intensity actions/match (Figure 2a). This supports the Mustafina, Popov, Vinogradova, & Ahmetov, 2014). However, the
LA MONTAGNA ET AL. | 7

limited sample used in the present study did not allow us to find any carrying the HFE H63D polymorphism have been carefully followed
significant difference between the AA frequency among the soccer by the medical staff that periodically checked the serum iron,
players and the control population used in that study, thus transferrin, and ferritin levels to control the iron homeostasis and
preventing any conclusion to be drawn. However, in a recent study execute nutritional interventions to avoid the variation of these
on a population on 73 male soccer players, the AA genotype was parameters above a certain threshold detrimental for the health and
found to be associated with a higher risk of muscular injuries the performance of the athletes. For HFE polymorphism, we obtained
(Massidda et al., 2015). The frequency of the AA genotype found in a nonsignificant difference with the data reported for a large, general
the published study involving 73 soccer players is statistically population study and for a smaller control population in a different
identical to that observed in the team, meaning that the genetic study (Chicharro et al., 2004; Katsarou et al., 2016). In contrast, the G
risk of muscular problems due to this variant is not different in the haplotype resulted significantly less present in S.S.C. Naples athletes
soccer team as compared with other teams in the same country. than in a sample of 65 endurance athletes, with a significant
Thus, the observed lower frequency of muscular accidents for this difference between the overall number of individuals carrying at
team must be due to reasons other than the genetic MCT1 makeup. least one H63D C > G mutation [p = .0067; χ2 test as recommended
Among the possible reasons, the superior performance of the studied by (Campbell, 2007) and (Richardson, 2011)]. In this respect, we
team can be connected to the special training programs and conclude that S.S.C. Naples soccer players do not differ from the
nutritional intervention (i.e., the specific formulation of sports drinks general control population.
to carbohydrate and fluid supplementation) developed by the
medical staff with the help of professional nutritionists to improve
the athletic performance of the athletes (Figure 2a).
3.6 | Combined analysis
From the single‐gene analysis discussed so far, it appeared that the
3.5 | Hemostatic iron regulator
S.S.C. Naples team is predicted to be negatively influenced at the
The hereditary hemochromatosis gene (HFE; High‐Fe) encodes a genetic level by the measured frequency of ACTN3 haplotypes,
transmembrane protein that plays a key role in iron absorption which predisposes the players to muscular injuries, and by specific
(Barton, Edwards, & Acton, 2015). Although HFE is not an iron MCT1 haplotypes positively affecting the iron transport, which
transporter, it interacts and regulates the activity of other proteins turned out to be less abundant in comparison to other teams playing
such as: the transferrin receptor (being a negative regulator of it), in Italy. However, the team is also well disposed because of its VEGF
that directly controls iron uptake from blood circulation (Parkkila superior genetic makeup, predicted to confer high aerobic capacities
et al., 2001); the ferroportin, a transmembrane protein that clears and in agreement with the reported team's highest distance/match
iron from the cell, excreting it into the blood (Ganz, 2005). HFE is covered and the highest number of intensity actions/match
thus able to control the homeostasis of iron (Barton et al., 2015). Loss performed.
of function of HFE results in a genetic disease known as To unravel the combined effects of the studied polymorph-
hemochromatosis, characterized by excessive accumulation of iron isms on the number of muscular injuries occurring to each player,
in the organism (Barton & Edwards, 1993; Barton et al., 2015). we resorted to a multinomial fitting procedure. The overall
However, high iron content, below the pathological threshold, may be number of muscular injuries occurred to each player is reported
beneficial for athletic performance. The HFE C187G polymorphism in Table 2.
results, at the protein level, in the substitution of a Histidine with an Complete genetic information was available only for 27 players.
aspartic acid in position 63 (H63D; rs1799945) with the subsequent Assuming as a regression variable the number of muscular incidents,
reduction of HFE function (Yang, Ferec, & Mura, 2011). HFE H63D we performed a multinomial fitting of a logistic model based on the
has a reduced ability to bind to the transferrin receptor resulting in player’ genotype. After the regression, an almost complete separa-
increased transport of iron in blood and cells (Feder et al., 1998). tion of observations was detected, with the following very good
Although a clear association of HFE H63D polymorphism with elite determination coefficients: R²(McFadden) = 0.953, R²(Cox and
athletic performance has not been established, it is worth noting that Snell) = 0.876, R²(Nagelkerke) = 0.986. The goodness of the fit is
in several studies a higher frequency of HFE H63D has been found in demonstrated by a χ2 = 56.446; with the available degrees of
endurance athletes such as cyclists (Chicharro et al., 2004; Deugnier freedom, the probability of finding a value of χ2 equal or superior in
et al., 2002). Iron is extremely important for oxygen transport and case the null hypothesis is true (i.e., there is no correlation between
delivery to the muscles; thus, it is reasonable that any impairment of the genetic makeup and the number of injuries for any given player)
iron uptake would result in an increase of circulating iron and an is p = .016, thus below the usual significance alpha threshold of
increase in oxygen availability, conditions that are both important for p = .05.
the execution of aerobic performance. The HFE H63D polymorphism, In agreement with such a good and significant fitting, the players’
exclusively in heterozygosis with the wild type allele (CG genotype) reclassification table obtained by applying the identified model to the
has been found in 22% of the soccer players (Figure 1e). The players 27 considered players is reported in Table 3.
8 | LA MONTAGNA ET AL.

T A B L E 2 Overall number of muscular injuries occurred in S.S.C. 4 | CO NCL USION

Naples soccer players (30) during the season 2014–2018
Player Injuries occurred Player Injuries occurred This study is based on the hypothesis that the combination of
1. xxxxx 0 16. xxxxx 2 appropriate training, adapted to the genetic and nutritional profile of

2. xxxxx 0 17. xxxxx 0 each athlete, can in principle be a way to achieve elite athletic status
and performance.
3. xxxxx 1 18. xxxxx 0
Five genes were selected for the screening of performance‐
4. xxxxx 3 19. xxxxx 1
enhancing polymorphisms in elite athletes, so as to cover the
5. xxxxx 0 20. xxxxx 0
polymorphisms in genes that intrinsically predispose toward a
6. xxxxx 0 21. xxxxx 4
specific activity (sprint vs. endurance) or to injuries and in genes
7. xxxxx 2 22. xxxxx 1 that can be externally modulated, for instance through specific
8. xxxxx 0 23. xxxxx 0 nutrient supplementation.
9. xxxxx 1 24. xxxxx 0 The evaluation of the status of each of the selected five genes in
10. xxxxx 0 25. xxxxx 0 the S.S.C. Naples soccer team uncovered two potential genetic
11. xxxxx 0 26. xxxxx 0 hindrances and one genetic enhancer of athletic performance. Given

12. xxxxx 1 27. xxxxx 0 the robust association of these genes to performance and injury risk,
it is possible to infer that the potential genetic hindrances were
13. xxxxx 1 28. xxxxx 0
minimized at least in part by the training and nutritional program
14. xxxxx 0 29. xxxxx 2
designed by the team's medical staff, thus explaining the excellent
15. xxxxx 1 30. xxxxx 0
performance obtained by the team (as recognized by Lega Calcio Serie
A and UEFA) in terms of athletic performance and injury rate.
Despite the overall reduced injuries occurring for the team's
As evident from the table, the prediction of the number of injuries
players, a residual genetic effect on the individual risk can be
based on the obtained model is wrong in only a single case (a player who
observed when combining all the five considered genes. The
was injured only once is back‐predicted to have incurred no injuries).
simultaneous effect of these five genes is indeed strongly correlated
This is the most accurate athletic injury predictive model we are
to the number of injuries experienced by the single components of
aware of, relying on the thoughtful selection of five genes already
the S.S.C. Naples soccer team. The obtained regression indicates that
related to the athletic performance, which were never tested
the identified five genes‐performance association could, at least in
simultaneously for their predictive power. With the help of this
part, explain why an athlete is more susceptible to injuries than
model, the S.S.C. Naples medical staff or the staff of other soccer
another, so that, after a proper validation in a larger population, it
teams can now in principle focus their efforts on players at risk,
can in principle be used to identify athletes who need special care
tailoring their training programs and giving them useful insights, so as
during training sessions and special attention when taking the field
to optimize the performance of the athletes.
and could support the development of tailored injury prevention
As a note of caution, we should, however, mention that the
programs and/or targeted therapeutic interventions. Moreover, our
obtained regressive model, albeit very good in retro‐predicting the
data can be helpful for the soccer team to define the peculiar role of
occurrence of injuries for each considered player in the S.S.C. Naples
the players, as defender or attacker, respect to their genetic profile.
soccer team, is promising but vexed by a potential overfitting
Finally, an interesting point of view can be the analysis of the possible
problem, due to the limited sample size used and to the lack of an
correlation of the gene SNPs variability and the role predisposition of
external, large sample of genetic information on soccer players to
the female soccer players.
test it. A thorough validation of the predictor needs further testing,
which is however not in the scope of the present study.

AC KNO WL EDG M EN TS
T A B L E 3 Prediction of number of injuries for S.S.C. Naples soccer
players (27), based on the new model The authors would like to thank Marie Basso for editorial and writing
From / to 0 1 2 3 4 Total % correct support and the Medical Staff of S.S.C. Napoli. This study was supported

0 16 0 0 0 0 16 100.00 by the Commonwealth of Pennsylvania, Department of Health for Sbarro

Health Research Organization, S.H.R.O. ([Link]).
1 1 6 0 0 0 7 85.71
2 0 0 2 0 0 2 100.00
3 0 0 0 1 0 1 100.00
4 0 0 0 0 1 1 100.00 CON F LI CT OF IN TE RES T S

Total 17 6 2 1 1 27 97.14 The authors have no conflict of interests.

LA MONTAGNA ET AL. | 9

A UT HO R C ONT RI BU TIO NS Brodal, P., Ingjer, F., & Hermansen, L. (1977). Capillary supply of skeletal
muscle fibers in untrained and endurance‐trained men. American
Raffaele La Montagna and Raffaele Canonico conceived the study. Journal of Physiology‐Heart and Circulatory Physiology, 232, H705–H712.
Luigi Alfano, Silvia Boffo, Leopoldo Staiano and Beniamino Fulco, Campbell, I. (2007). Chi‐squared and Fisher‐Irwin tests of two‐by‐two
Patrizia Maiorano, Costantino D’Angelo, and Andrea Chirico colla- tables with small sample recommendations. Statistics in Medicine, 26,
3661–3675.
borated on the collection of data. Enrico Bucci performed the
Chicharro, J. L., Hoyos, J., Gomez‐Gallego, F., Villa, J. G., Bandres, F., Celaya,
combined analysis. Alfonso De Nicola, Raffaele Canonico, and Enrico P., Jimenez, F., Alonso, J. M., Cordova, A., & Lucia, A. (2004). Mutations
D’Andrea as part of the Società Sportiva Calcio Napoli and of the in the hereditary haemochromatosis gene HFE in professional
S.S.C. Napoli Medical Staff supervised the screening. Raffaele La endurance athletes. British Journal of Sports Medicine, 38, 418–421.
Cupeiro, R., Benito, P. J., Maffulli, N., Calderón, F. J., & González‐Lamuño, D.
Montagna, Luigi Alfano, and Silvia Boffo analyzed the data, conceived
(2010). MCT1 genetic polymorphism influence in high intensity circuit
the figures and wrote the manuscript. Antonio Giordano originated training: A pilot study. Journal of Science and Medicine in Sport, 13,
the idea and supervised the project. All authors commented on the 526–530.
manuscript and approved the final version. Cupeiro, R., Pérez‐Prieto, R., Amigo, T., Gortázar, P., Redondo, C., &
González‐Lamuño, D. (2016). Role of the monocarboxylate transporter
MCT1 in the uptake of lactate during active recovery. European Journal
E TH ICS S TA T EM E NT of Applied Physiology and Occupational Physiology, 116, 1005–1010.
Deugnier, Y., Loreal, O., Carre, O., Duvallet, E., Zoulim, F., Vinel, J. P., …
We obtained the signed informed consent from all the involved Guinot, M. (2002). Increased body iron stores in elite road cyclists.
Medicine & Science in Sports & Exercise, 34, 876–880.
subjects; after performing the DNA extraction and analysis, all data
Feder, J. N., Penny, D. M., Irrinki, A., Lee, V. K., Lebron, J. A., Watson, N., …
were anonymized. Dott. De Nicola and Dott. Canonico are Schatzman, R. C. (1998). The hemochromatosis gene product
responsible for the processing of all personal data. This study did complexes with the transferrin receptor and lowers its affinity for
not involve any additional pharmacological or therapeutic interven- ligand binding. Proceedings of the National Academy of Sciences United
States of America, 95, 1472–1477.
tion, as compared with the routine procedures adopted by the SSC
Fedotovskaya, O. N., Mustafina, L. J., Popov, D. V., Vinogradova, O. L., &
Calcio Napoli (Naples soccer team). Ahmetov, I. I. (2014). A common polymorphism of the MCT1 gene and
athletic performance. Int J Sports Physiol Perform, 9, 173–180.
Ferioli, M., Zauli, G., Maiorano, P., Milani, D., Mirandola, P., & Neri, L. M.
OR CID (2019). Role of physical exercise in the regulation of epigenetic
mechanisms in inflammation, cancer, neurodegenerative diseases, and
Luigi Alfano [Link]
aging process. Journal of Cellular Physiology, 234, 14852–14864.
Enrico Bucci [Link] Ganz, T. (2005). Cellular iron: Ferroportin is the only way out. Cell
Andrea Chirico [Link] Metabolism, 1, 155–157.
Antonio Giordano [Link] Gustafsson, T., Puntschart, A., Kaijser, L., Jansson, E., & Sundberg, C. J.
(1999). Exercise‐induced expression of angiogenesis‐related transcrip-
tion and growth factors in human skeletal muscle. American Journal of
REFERENC ES Physiology‐Heart and Circulatory Physiology, 276, H679–H685.
Halestrap, A. P., & Wilson, M. C. (2012). The monocarboxylate transporter
Abrahams, S., Posthumus, M., & Collins, M. (2014). A polymorphism in a family‐‐role and regulation. IUBMB Life, 64, 109–119.
functional region of the COL5A1 gene: Association with ultraendur- Van Hall, G. (2000). Lactate as a fuel for mitochondrial respiration. Acta
ance‐running performance and joint range of motion. International Physiologica Scandinavica, 168, 643–656.
Journal of Sports Physiology and Performance, 9, 583–590. Heffernan, S. M., Kilduff, L. P., Erskine, R. M., Day, S. H., Stebbings, G. K.,
Akhmetov, A. M., II, Khakimullina, D. V., Popov, S. S., Missina, O. L., Cook, C. J., … Williams, A. G. (2017). COL5A1 gene variants previously
Vinogradova, & Rogozkin, V. A. (2008). Polymorphism of the vascular associated with reduced soft tissue injury risk are associated with
endothelial growth factor gene (VEGF) and aerobic performance in elite athlete status in rugby. BMC Genomics, 18, 820.
athletes. Fiziologiia Cheloveka, 34, 97–101. Hoeben, A., Landuyt, B., Highley, M. S., Wildiers, H., Van Oosterom, A. T., &
Al‐Habboubi, H. H., Sater, M. S., Almawi, A. W., Al‐Khateeb, G. M., & De Bruijn, E. A. (2004). Vascular endothelial growth factor and
Almawi, W. Y. (2011). Contribution of VEGF polymorphisms to angiogenesis. Pharmacological Reviews, 56, 549–580.
variation in VEGF serum levels in a healthy population. European Jobling, R., D’souza, R., Baker, N., Lara‐Corrales, I., Mendoza‐Londono, R.,
Cytokine Network, 22, 154–158. Dupuis, L., … Kannu, P. (2014). The collagenopathies: Review of clinical
Barton, J.C., & Edwards, C.Q. 1993. HFE hemochromatosis. In gene phenotypes and molecular correlations. Current Rheumatology Reports, 16,
reviews((R)). M.P. Adam, H.H. Ardinger, R.A. Pagon, S.E. Wallace, L.J.H. 394.
Bean, K. Stephens, and A. Amemiya, editors, Seattle (WA). Juel, C., & Halestrap, A. P. (1999). Lactate transport in skeletal muscle ‐
Barton, J. C., Edwards, C. Q., & Acton, R. T. (2015). HFE gene: Structure, role and regulation of the monocarboxylate transporter. Journal of
function, mutations, and associated iron abnormalities. Gene, 574, Physiology (Cambridge, United Kingdom), 517(Pt 3), 633–642.
179–192. Katsarou, M. S., Latsi, R., Papasavva, M., Demertzis, N., Kalogridis, T.,
Bertuzzi, R., Pasqua, L. A., Bueno, S., Lima‐Silva, A. E., Matsuda, M., Marquezini, Tsatsakis, A. M., … Drakoulis, N. (2016). Population‐based analysis of
M., & Saldiva, P. H. (2014). Is the COL5A1 rs12722 gene polymorphism the frequency of HFE gene polymorphisms: Correlation with the
associated with running economy? PLoS One, 9, e106581. susceptibility to develop hereditary hemochromatosis. Molecular
Blanchard, A., Ohanian, V., & Critchley, D. (1989). The structure and Medicine Reports, 14, 630–636.
function of α‐actinin. Journal of Muscle Research and Cell Motility, Kikuchi, N., Fuku, N., Matsumoto, R., Matsumoto, S., Murakami, H.,
10, 280–289. Miyachi, M., & Nakazato, K. (2017). The association between MCT1
10 | LA MONTAGNA ET AL.

T1470A polymorphism and power‐oriented athletic performance. Rankinen, T., Fuku, N., Wolfarth, B., Wang, G., Sarzynski, M. A., Alexeev, D.
International Journal of Sports Medicine, 38, 76–80. G., … Bouchard, C. (2016). No evidence of a common DNA variant
Lee, F. X. Z., Houweling, P. J., North, K. N., & Quinlan, K. G. R. (2016). How profile specific to world class endurance athletes. PLoS One, 11,
does α‐actinin‐3 deficiency alter muscle function? Mechanistic in- e0147330.
sights into ACTN3, the ‘gene for speed. Biochimica et Biophysica Acta Richardson, R. S., Wagner, H., Mudaliar, S. R. D., Saucedo, E., Henry, R., &
(BBA) ‐ Molecular Cell Research, 1863, 686–693. Wagner, P. D. (2000). Exercise adaptation attenuates VEGF gene
Liu, L. X., Lu, H., Luo, Y., Date, T., Belanger, A. J., Vincent, K. A., … Jiang, C. expression in human skeletal muscle. American Journal of Physiology‐
(2002). Stabilization of vascular endothelial growth factor mRNA by Heart and Circulatory Physiology, 279, H772–H778.
hypoxia‐inducible factor 1. Biochemical and Biophysical Research Commu- Richardson, M., & Reischman, D. (2011). The magical number 7. Teaching
nications, 291, 908–914. Statistics, 33, 17–19.
Luszczyk, M., Kaczorowska‐Hac, B., Milosz, E., Adamkiewicz‐Drozynska, Richardson, R. S., Noyszewski, E. A., Kendrick, K. F., Leigh, J. S., &
E., Ziemann, E., Laskowski, R., … Antosiewicz, J. (2017). Reduction of Wagner, P. D. (1995). Myoglobin O2 desaturation during exercise.
skeletal muscle power in adolescent males carrying H63D mutation in Evidence of limited O2 transport. Journal of Clinical Investigation, 96,
the HFE gene. BioMed Research International, 2017, 1–7. 1916–1926.
Massidda, M., Eynon, N., Bachis, V., Corrias, L., Culigioni, C., Piras, F., … Calò, Roth, S. M., Walsh, S., Liu, D., Metter, E. J., Ferrucci, L., & Hurley, B. F. (2008).
C. M. (2015). Influence of the MCT1 rs1049434 on indirect muscle The ACTN3 R577X nonsense allele is under‐represented in elite‐level
disorders/injuries in elite football players. Sports Medicine ‐ Open, 1, 33. strength athletes. European Journal of Human Genetics, 16, 391–394.
Massidda, M., Bachis, V., Corrias, L., Piras, F., Scorcu, M., Culigioni, C., … Sawczuk, M., Banting, L. K., Cięszczyk, P., Maciejewska‐Karłowska, A.,
Calò, C. M. (2015). ACTN3 R577X polymorphism is not associated Zarębska, A., Leońska‐Duniec, A., … Eynon, N. (2015). MCT1 A1470T:
with team sport athletic status in Italians. Sports Medicine ‐ Open, 1, 6. A novel polymorphism for sprint performance? Journal of Science and
Massidda, M., Voisin, S., Culigioni, C., Piras, F., Cugia, P., Yan, X., … Calo, C. Medicine in Sport, 18, 114–118.
M. (2017). ACTN3 R577X polymorphism is associated with the Thompson, D. B., Pratley, R., & Ossowski, V. (1996). Human primary myoblast
incidence and severity of injuries in professional football players. cell cultures from non‐diabetic insulin resistant subjects retain defects in
Clinical Journal of Sport Medicine, 29(1), 57–61. insulin action. Journal of Clinical Investigation, 98, 2346–2350.
Massidda, M., Bachis, V., Corrias, L., Piras, F., Scorcu, M., & Calo, C. M. Wahl, P., Jansen, F., Achtzehn, S., Schmitz, T., Bloch, W., Mester, J., &
(2015a). Influence of the COL5A1 rs12722 on musculoskeletal Werner, N. (2014). Effects of high intensity training and high volume
injuries in professional soccer players. Journal of Sports Medicine and training on endothelial microparticles and angiogenic growth factors.
Physical Fitness, 55, 1348–1353. PLoS One, 9, e96024.
Merezhinskaya, N., Fishbein, W. N., Davis, J. I., & Foellmer, J. W. (2000). Westerblad, H., Allen, D. G., & Lännergren, J. (2002). Muscle fatigue: Lactic
Mutations in MCT1 cDNA in patients with symptomatic deficiency in acid or inorganic phosphate the major cause? Physiology, 17, 17–21.
lactate transport. Muscle and Nerve, 23, 90–97. Yang, N., MacArthur, D. G., Gulbin, J. P., Hahn, A. G., Beggs, A. H., Easteal, S., &
Ostrander, E. A., Huson, H. J., & Ostrander, G. K. (2009). Genetics of North, K. (2003). ACTN3 genotype is associated with human elite athletic
athletic performance. Annual Review of Genomics and Human Genetics, performance. The American Journal of Human Genetics, 73, 627–631.
10, 407–429. Yang, Y., Férec, C., & Mura, C. (2011). SNP and haplotype analysis reveals
Parkkila, S., Niemelä, O., Britton, R. S., Fleming, R. E., Waheed, A., Bacon, B. new HFE variants associated with iron overload trait. Human Mutation,
R., & Sly, W. S. (2001). Molecular aspects of iron absorption and HFE 32, E2104–E2117.
expression. Gastroenterology, 121, 1489–1496.
Pimenta, E. M., Coelho, D. B., Veneroso, C. E., Barros Coelho, E. J., Cruz, I.
R., Morandi, R. F., … De paz fernández, J. A. (2013). Effect of ACTN3
gene on strength and endurance in soccer players. Journal of Strength How to cite this article: La Montagna R, Canonico R, Alfano L,
and Conditioning Research, 27, 3286–3292. et al. Genomic analysis reveals association of specific SNPs
Prior, S. J., Hagberg, J. M., Paton, C. M., Douglass, L. W., Brown, M. D., with athletic performance and susceptibility to injuries in
McLenithan, J. C., & Roth, S. M. (2006). DNA sequence variation in the
professional soccer players. J Cell Physiol. 2019;1–10.
promoter region of the VEGF gene impacts VEGF gene expression
and maximal oxygen consumption. American Journal of Physiology‐ [Link]
Heart and Circulatory Physiology, 290, H1848–H1855.