Polymorphism of the vascular endothelial growth factor gene (VEGF) and aerobic performance in athletes
The subject of this study is the frequency distribution of alleles of the vascular endothelial growth factor gene (VEGF; the G-634C polymorphism) in athletes (n = 670) and in a control group (n = 1073) and the relationships of genotypes with aerobic performance in rowers (n = 90). Genetic typing was performed using the analysis of restriction fragment length polymorphism. The frequency of the VEGF C allele in the group of endurance athletes (n = 294) was significantly higher than in the control group and increased together with increasing sports qualification. In addition, a correlation of the VEGF C allele with a high aerobic performance of athletes (according to data on the maximal power and maximal oxygen consumption) and with a substantial contribution to the energy supply of aerobic metabolism (according to the values of maximal lactate content) has been found. It is inferred that the G-634C polymorphism of the VEGF gene is associated with physical performance of athletes and plays a key role in sports selection.
ISSN 0362-1197, Human Physiology, 2008, Vol. 34, No. 4, pp. 477–481. © Pleiades Publishing, Inc., 2008.
Original Russian Text © I.I. Ahmetov, A.M. Khakimullina, D.V. Popov, S.S. Missina, O.L. Vinogradova, V. A. Rogozkin, 2008, published in Fiziologiya Cheloveka, 2008, Vol. 34,
No. 4, pp. 97–101.
Enhancement of endurance as a result of systematic
aerobic exercise is determined by numerous adaptive
responses to training stimuli. These responses include
an increase in the number of capillaries around each
muscle ﬁber, which results in improvement of gas and
heat exchange and in acceleration of excretion of the
products of decomposition of nutrients and their
exchange between blood and working muscle ﬁbers.
The improvement of the maximal oxygen consumption
( ) as a result of training is mainly determined by
an increased maximal blood ﬂow and higher density of
muscle capillaries in active tissues . The number of
capillaries and its ratio to the number of muscle ﬁbers
in leg muscles of an athlete engaged in cyclic sports
may be 5–30% and 50% higher, respectively, than in a
sedentary person [2–4].
Individual differences in the degree of adaptive
changes, such as growth of blood vessels of skeletal
muscles and the myocardium, are to a greater extent
accounted for by genetic factors that determine the
genetic predisposition to performing physical exercises
of different intensities and durations . On this basis,
an important problem of muscle performance genetics
is to identify genetic markers associated with the regu-
lation of the growth of vessels; it may help in settling
many questions of optimal sports specialization and
correction of the training process.
One of the main factors directly inﬂuencing the
growth of vessels is vascular endothelial growth factor
(VEGF), the expression of which signiﬁcantly
increases under aerobic physical exercise [6, 7]. VEGF
is a glycoprotein that binds with the cells of blood and
lymphatic vessels and stimulates their proliferation.
The effect of VEGF is mediated by its receptors
located in chromosome 6 (6p12). The expression of
is stimulated by a great number of proangio-
genic factors, including the hypoxia-induced factor
(HIF) and epidermal (EGF) and ﬁbroblast (FGF)
growth factors. In addition, the level of VEGF is inﬂu-
enced by the pH value of the blood and the partial pres-
sure and concentration of oxygen in inhaled air .
Among the studied polymorphisms of the
gene, of special interest are the variants located in the
promoter (regulatory) region. For example, the substi-
tution of cytosine for guanine at position –634 (the
) increases the gene
activity and, accordingly, determines individual differ-
ences in the level of expression . According to data
on subjects not engaged in sports, the
associated with a greater increase in the level
as a result of aerobic physical exercise .
In this connection, it would be particularly interest-
ing to estimate the correlation of the
phism of the
gene with the physical performance
of athletes. It can be assumed that the frequency of the
allele is higher in endurance athletes, with the
genotypes being associated with
higher values of aerobic performance. For testing this
Polymorphism of the Vascular Endothelial Growth Factor
) and Aerobic Performance in Athletes
I. I. Ahmetov
, A. M. Khakimullina
, D. V. Popov
, S. S. Missina
O. L. Vinogradova
, and V. A. Rogozkin
All-Russian Research Institute of Physical Culture and Sports, Moscow, 105005 Russia
St. Petersburg Research Institute of Physical Culture, St. Petersburg, 197110 Russia
Institute of Biomedical Problems, Russian Academy of Sciences, Moscow, 123007 Russia
Received October 31, 2007
—The subject of this study is the frequency distribution of alleles of the vascular endothelial growth
factor gene (
polymorphism) in athletes (
= 670) and in a control group (
= 1073) and the
relationships of genotypes with aerobic performance in rowers (
= 90). Genetic typing was performed using
the analysis of restriction fragment length polymorphism. The frequency of the
allele in the group of
endurance athletes (
= 294) was signiﬁcantly higher than in the control group and increased together with
increasing sports qualiﬁcation. In addition, a correlation of the
allele with a high aerobic performance
of athletes (according to data on the maximal power and maximal oxygen consumption) and with a substantial
contribution to the energy supply of aerobic metabolism (according to the values of maximal lactate content)
has been found. It is inferred that the
polymorphism of the
gene is associated with physical per-
formance of athletes and plays a key role in sports selection.
AHMETOV et al.
hypothesis, the frequency of the
been determined in athletes engaged in cyclic sports
focused on endurance training to some extent. Rowing
was chosen as a model for studying physical perfor-
mance because it is a complex sport that mainly
requires endurance (rowers cover a distance of 2000 m
70% at the expense of aerobic metabolism) .
The goal of this work was to study the distribution
allele frequencies in athletes engaged in
cyclic sports and in a control group of subjects not
engaged in sports and to estimate the correlation of
genotypes with aerobic performance in rowers.
Characteristics of the sample.
The study was per-
formed in 670 athletes engaged in cyclic sports (229
years old, and 441 men,
According to the type of energy supply for a training
load, we divided cyclic sports into three groups within
which physiological characteristics of the subjects
found in training exercises were the same . In addi-
tion to signs characterizing training for endurance,
quickness, and strength, the power of the work per-
formed during training was taken into account, with
division into submaximal, high, and moderate (sprint-
ers and athletes engaged in acyclic sports were excluded
from the research). Group I (moderate power; the main
energy source is fatty acids and glycogen; the time of
competitive exercise performance is over 30 min)
included athletes engaged in biathlon, track cycling, ski
racing (15–50 km), marathon running, swimming (5–
25 km), race walking, and triathlon. Group II (high
power; fatty acids and glycogen; 5–30 min) included ath-
letes engaged in rowing, running (3–10 km), speed skat-
ing (5–10 km), ski racing (5–10 km), and swimming
(800–1500 m). Group III (submaximal power; glyco-
gen and lactate; 45 s to 3–5 min) included athletes
engaged in running (800–1500 m), kayakling, speed
skating (1–3 km), swimming (200–400 m), and short
track speed skating.
At the moment of sampling of biological material
for genotyping, 18 athletes were honored masters of
sport (HMSs), 60 were international masters of sport
(IMSs), 135 were masters of sport (MSs), 210 were
candidates for masters of sport (CMSs), and 247 had
adult sports rankings.
During a physiological examination, indices of aerobic
and anaerobic performance were estimated in 90 rowers,
including 56 men (27 CMSs and 29 MSs) and 34 women
(13 CMSs and 21 MSs).
The control group consisted of 1073 residents of
St. Petersburg, Moscow, and Naberezhnye Chelny
years old, and 488 men,
years old). The main condition for inclusion of sub-
jects into the control group was the absence of experi-
ence in regular sports activity (in questionnaires, the
respondents did not indicate any sports qualiﬁcation).
DNA samples isolated from subjects
by the method of alkaline extraction  or sorbent
method were used for molecular genetic analysis by
means of a DNA-sorb-A kit (Central Research Institute
of Epidemiology, Ministry of Health of the Russian
Federation) as recommended by the manufacturer,
depending on the method of sampling of biological
material (washing out or scraping of epithelial cells of
the oral cavity).
Analysis of restriction fragment length polymor-
polymorphism of the
gene was analyzed using a two-primer system
(forward primer, 5'-GTAGCAAGAGCTCCA-
GAGAGAAGT-3'; reverse primer 5'-TGGAC-
GAAAAGTTTCAGTGCGACG-3') . Amplicons
342 bp in length were hydrolyzed using
endonuclease (SibEnzyme). Fragments 249 and 93 bp in
length corresponded to the
allele, and a fragment
342 bp in length corresponded to the
lengths of restriction fragments of the products were
analyzed by electrophoretic separation in 8% polyacry-
lamide gel followed by ethidium bromide staining and
visualization in transmitted UV light using an ETS Vil-
ber-Lourmat transilluminator (France).
Determination of the indices of aerobic and anaero-
bic capacities in a test with gradually increasing load
The aerobic capacity was determined in a
test with an increasing load using a PM3 mechanical
rowing ergometer (Concept II, United States). The ini-
tial load was 150 W for men and 100 W for women; the
step duration was 3 min; the rest break between steps
was 30 s. The work was performed until failure (a
stroke power decrease by >30 W compared to the preset
power; a respiratory quotient >1.1). Since the athletes
were sometimes unable to perform the exercise for as
long as 3 min at the last step, the following estimated
value was taken as the maximal power (
is the mean power of the last step (W),
is the mean power of the next-to-last step (W), and
the time of work at the last step (s).
Gas exchange indices and heart rate (HR, beats/min)
were recorded continuously (during each breathing
cycle) throughout the test (MetaMax 3B (Cortex, Ger-
many) and Vmax 229 (SensorMedics, United States)
gas analyzers were used). The (l/min) was
determined by the gas exchange values averaged over
the last 30 s of each step of the test.
The lactate content of the blood (
determined electrochemically (Super GL easy, Dr. Muel-
ler, Germany); capillary blood (20
l) was taken from a
ﬁnger after each step and immediately after the end of
POLYMORPHISM OF THE VASCULAR ENDOTHELIAL GROWTH FACTOR 479
The data were statistically processed using the
GraphPad InStat software. The mean value (
dard error of the mean (
), and root mean square
) were determined. The signiﬁcance of dif-
ferences in the frequency of alleles between the com-
pared groups was determined using the
test (for large
groups) or Fisher’s exact test (for small groups). The
groups were compared with respect to quantitative
characters (physiological indices) in an unpaired test.
The differences were considered statistically signiﬁcant
RESULTS AND DISCUSSION
Analysis of the distribution of VEGF genotype and
allele frequencies in athletes and control subjects. The
following results were obtained upon the analysis of the
frequency distribution of genotypes and alleles with
respect to the G-634C polymorphism of the VEGF gene
in the control group (n = 1073) and in the athletes (n =
670). The frequency of the VEGF C allele in the control
group was 24.5%; it was similar in women (25.3%) and
men (23.6%). The distribution of genotypes GG
(57.6%), GC (35.8%), and CC (6.6%) observed in the
control group ﬁt the Hardy–Weinberg equilibrium (χ2 =
0.61; d.f. = 2; p = 0.74) (table).
The VEGF C allele frequency in the total group of
athletes was signiﬁcantly higher than in the control
group (29.2 versus 24.5%; p = 0.0026). The distribution
of genotypes GG (50.6%), GC (40.4%), and CC (9%)
in the group of athletes also ﬁt the Hardy–Weinberg
equilibrium (χ2 = 0.16; d.f. = 2; p = 0.93). The table
shows the data on the distribution of VEGF genotypes
and alleles in athletes with specializations differing in
Distribution of the absolute and relative frequencies of VEGF genotypes and alleles among athletes from different groups and
Group Sport nGenotypes C allele,
GG GC CC
I Biathlon 34 16 16 2 29.4 0.43
Track cycling 110 50 45 15 34.1 0.0025*
Ski racing, 15–50 km 71 34 30 7 31.0 0.1
1 60.0 0.026*
Swimming, 5–25 km 21 12 7 2 26.2 0.94
Race walking 24 11 9 4 35.4 0.12
Triathlon 29 18 10 1 20.7 0.6
Total 294 141 121 32 31.5 0.0008*
II Rowing 192 107 70 15 26.0 0.56
Running, 3–10 km 514
0 40.0 0.44
Skating, 5–10 km 430
1 25.0 0.97
Ski racing, 5–10 km 64 28 31 5 32.0 0.071
Swimming, 800–1500 m 25 14 9 2 26.0 0.94
Total 290 153 114 23 27.6 0.14
III Running, 800–1500 m 11 6 3 2 31.8 0.59
Kayaking 35 20 14 1 22.9 0.86
Skating, 1–3 km 844
0 25.0 0.96
Swimming, 200–400 m 24 9 13 2 35.4 0.12
Short track speed skating 862
0 12.5 0.41
Total 86 45 36 5 26.7 0.57
Total of athletes 670 339 271 60 29.2 0.0026*
Control group 1073 618 384 71 24.5 1.00
* Statistically signiﬁcant differences between the groups of athletes and the control group (p < 0.05).
HUMAN PHYSIOLOGY Vol. 34 No. 4 2008
AHMETOV et al.
the type of energy supply. As can be seen in the table,
only group I, with mostly endurance training, had a sig-
niﬁcantly higher frequency of the VEGF C allele as
compared to the control group (31.5 versus 24.5%; p =
The analysis of the distribution of VEGF gene alle-
les between the sexes showed no differences either in
the pooled sample of athletes or in athletes of group I
(all athletes: women, 29%; men, 29.3%; group I:
women, 28.3%; men, 32.9%).
The assessment of the distribution of allele frequen-
cies depending on sports qualiﬁcation showed that the
frequency of the VEGF C allele in athletes of group I
(long-distance racers) had a tendency towards an
increase with increased qualiﬁcation: it was 26.3% in
the MS group (n = 38), 28.8% in the IMS group (n =
40), and 50% in the HMS group (n = 10) (p = 0.09).
The higher frequency of the VEGF C allele in long-
distance racers compared to the control and its increase
with increased sports qualiﬁcation suggests that the
VEGF C allele favors the development and expression
Relationship between physiological indices and
VEGF genotypes in athletes. The ﬁnding of substantial
differences in some physiological indices between row-
ers of different sexes and qualiﬁcations made it neces-
sary to perform separate analyses of the relationship
between phenotypes and genotypic data. Hence, 90 ath-
letes were divided into four subgroups (male CMSs
(n = 27), male MSs (n = 29), female CMSs (n = 13), and
female MSs (n = 21)).
The mean Wmax values were signiﬁcantly higher in
male MSs with the VEGF C allele as compared with
carriers of the VEGF GG genotype (CC + GC, 401 (31)
W; GG, 371 (47) W; p = 0.049). In addition, the VEGF
C allele was associated with higher values in
men (CC, 5.5 (0.6) l/min; GC, 5.1 (0.5) l/min; GG, 4.9
(0.5) l/min; p = 0.09) and with a greater contribution to
the energy supply of aerobic metabolism in female MSs,
which is indirectly evidenced by the minimal values of
lactate content upon failure to continue the exercise (CC,
4.4 µM; GC, 8.4 (0.9) µM; GG, 8.7 (1.6) µM; p = 0.08).
The results are in agreement with previously pub-
lished data on 148 volunteers, 50–75 years old, leading
a sedentary life . The increase in after 24
weeks of aerobic training was signiﬁcantly higher in
carriers of VEGF haplotypes with the –634°C allele.
The same study showed that the VEGF C allele in a cul-
ture of human myoblasts was expressed to a greater
extent than the G allele. The researchers supposed that
the functional signiﬁcance of the G-634C polymor-
phism of the VEGF gene is determined by the localiza-
tion of the polymorphic region at the site of binding of
the promoter with various transcription factors regulat-
ing the gene activity.
The high expression of the VEGF C allele suggests
a greater adaptive growth of capillaries in response to
aerobic physical exercise. Consequently, carriers of the
VEGF C allele may have some advantage in increasing
the aerobic performance, which has been revealed in
this and previous studies. This assumption is supported
by the higher frequency of the VEGF C allele in the
group of long-distance racers compared to the control.
This phenomenon underlies the process of sports selec-
tion: the alleles that favor the development of endur-
ance and the achievement of high sports results are
accumulated in the subgroups of long-distance racers
as their sports qualiﬁcation increases.
Thus, the G-634C polymorphism of the VEGF gene
is associated with physical performance of athletes and
plays a key role in sports selection. The results of this
research are of both fundamental and applied signiﬁ-
cance for understanding the molecular mechanisms of
adaptation of the cardiovascular system to aerobic exer-
cise, selecting an optimal sports specialization, and
professional training of athletes.
1. Saltin, B. and Rowell, L.B., Functional Adaptations to
Physical Activity and Inactivity, Fed. Proc., 1980,
vol. 39, p. 1506.
2. Ingjer, F., Capillary Supply and Mitochondrial Content
of Different Skeletal Muscle Fiber Types in Untrained
and Endurance-Trained Men: A Histochemical and
Ultrastructural Study, Eur. J. Appl. Physiol. Occup.
Physiol., 1979, vol. 40, no. 3, p. 197.
3. Shono, N., Urata, H., Saltin, B., et al., Effects of Low
Intensity Aerobic Training on Skeletal Muscle Capillary
and Blood Lipoprotein Proﬁles, J. Atheroscler. Thromb.,
2002, vol. 9, p. 78.
4. Hermansen, L. and Wachtlova, M., Capillary Density of
Skeletal Muscle in Well-trained and Untrained Men,
J. Appl. Physiol., 1971, vol. 30, p. 860.
5. Brodal, P., Ingjer, F., and Hermansen, L., Capillary Sup-
ply of Skeletal Muscle Fibers in Untrained and Endur-
ance-Trained Men, Am. J. Physiol. Heart Circ. Physiol.,
1977, vol. 232, H705.
6. Gustafsson, T., Puntschart, A., Kaijser, L., Jansson, E.,
and Sundberg, C.J., Exercise-Induced Expression of
Angiogenesis-Related Transcription and Growth Factors
in Human Skeletal Muscle, Am. J. Physiol. Heart Circ.
Physiol., 1999, vol. 276, p. 679.
7. Richardson, R.S., Wagner, H., Mudaliar, S.R., Sau-
cedo, E., Henry, R., and Wagner P.D., Exercise Adapta-
tion Attenuates VEGF Gene Expression in Human Skel-
etal Muscle, Am. J. Physiol. Heart Circ. Physiol., 2000,
vol. 279, p. 772.
8. Liu, L.X., Lu, H., Luo, Y., et al., Stabilization of Vascular
Endothelial Growth Factor mRNA by Hypoxia-Induc-
ible Factor 1, Biochem. Biophys. Res. Comm., 2002,
vol. 281, p. 908.
9. Watson, C.J., Webb, N.J., Bottomley, M.J., and Brench-
ley, P.E., Identiﬁcation of Polymorphisms within the
Vascular Endothelial Growth Factor (VEGF) Gene: Cor-
HUMAN PHYSIOLOGY Vol. 34 No. 4 2008
POLYMORPHISM OF THE VASCULAR ENDOTHELIAL GROWTH FACTOR 481
relation with Variation in VWGF Protein Production,
Cytokine, 2000, vol. 12, p. 1232.
10. Prior, S.J., Hagberg, J.M., Paton, C.M., et al., DNA
Sequence Variation in the Promoter Region of the VEGF
Gene Impacts VEGF Gene Expression and Maximal
Oxygen Consumption, Am. J. Physiol. Heart Circ. Phys-
iol., 2006, vol. 290, p. 1848.
11. Hagerman, F.C., Applied Physiology of Rowing, Sports
Med., 1984, vol. 1, no. 4, p. 303.
12. Ahmetov, I.I., Astratenkova, I.V., and Rogozkin, V.A.,
Association of a PPARD Polymorphism with Human
Physical Performance, Mol. Biol., 2007, vol. 41, no. 5,
13. Bolla, M.K., Haddad, L., Humphries, S.E., et al., A
Method of Determination of Hundreds of APOE Geno-
types Utilizing Highly Simpliﬁed, Optimized Protocols
and Restriction Digestion Analysis by Microtitre Array
Diagonal Gel Electrophoresis (MADGE), Clin. Chem.,
1995, vol. 41, p. 1599.