International Journal of Sports Physiology and Performance, 2011, 6, 485-496
Â© 2011 Human Kinetics, Inc.
James C. Brown, Caron-Jayne Miller, Michael Posthumus, and Martin P. Schwellnus are with the UCT/
MRC Research Unit for Exercise Science and Sports Medicine, Department of Human Biology, Faculty
of Health Sciences, University of Cape Town, Cape Town, South Africa. Malcolm Collins is with the
UCT/MRC Research Unit for Exercise Science and Sports Medicine, Department of Human Biology,
Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa, and the South African
Medical Research Council, Cape Town, South Africa.
The COL5A1 Gene, Ultra-Marathon
and Range of Motion
James C. Brown, Caron-Jayne Miller, Michael Posthumus,
Martin P. Schwellnus, and Malcolm Collins
Purpose: Endurance running performance is a multifactorial phenotype that is
strongly associated with running economy. Sit and reach range of motion (SR
ROM) is negatively associated with running economy, suggesting that reduced
SR ROM is advantageous for endurance running performance. The COL5A1 gene
has been associated with both endurance running performance and SR ROM in
separate cohorts. The aim of this study was to investigate whether COL5A1 is
associated with ultra-marathon running performance and whether this relationship
could be partly explained by prerace SR ROM. Methods: Seventy-two runners (52
male, 20 female) were recruited from the 56 km Two Oceans ultra-marathon and
were assessed for prerace SR ROM. The cohort was genotyped for the COL5A1
BstUI restriction fragment length polymorphism, and race times were collected
after the event. Results: Participants with a TT genotype (341 Â± 41 min, N =
21) completed the 56 km Two Oceans ultra-marathon significantly (P = 0.014)
faster than participants with TC and CC genotypes (365 Â± 39 min, N = 50). The
COL5A1 genotype and age accounted for 19% of performance variance. When
the cohort was divided into performance and flexibility quadrants, the T allele was
significantly (P = 0.044) over-represented within the fast and inflexible quadrant.
Conclusion: The COL5A1 genotype was found to be significantly associated with
performance in a 56 km ultra-endurance run. This study confirms previous findings
and it furthers our understanding of the relationships among ROM, COL5A1, and
endurance running performance. We continue to speculate that the COL5A1 gene
alters muscleâ€“tendon stiffness.
Keywords: Type V collagen, endurance, athletic performance, range of motion,
running economy, genetic association
486 Brown et al.
Endurance running performance is a multifactorial phenotype.1 Physiological
factors that are commonly associated with endurance running ability include, but are
not limited to, muscle capillary density, maximal heart rate, anthropometry, substrate
utilization, and aerobic enzyme activity.1,2 Certain genetic3 and flexibility profiles4,5
have also been shown to be associated with endurance running performance.
Range of motion (ROM) is associated with running economy,6â€“8 a measurable
marker of ultra-endurance performance.2 In elite athletes, the sit-and-reach range of
motion (SR ROM) test was negatively associated with running economy.7 Similar
associations have also been shown for subelite athletes. These data6â€“8 suggest that
a reduced range of motion, as measured by the SR ROM test and other tests, is
advantageous for endurance running performance.
We recently reported that the COL5A1 gene was associated with SR ROM.9,10
The COL5A1 gene encodes for the Î±1 chain of type V collagen, a minor fibrillar
collagen that plays an essential role in fibril assembly and lateral fibril growth
within connective tissues (including tendon).11 The CC genotype of the COL5A1
BstUI restriction fragment length polymorphism (RFLP) was significantly associated with increased SR ROM.10 In contrast, Posthumus et al12 reported that the TT
genotype of the COL5A1 BstUI RFLP was associated with improved endurance
running ability in the 42.2 km run of the 226 km South African Ironman triathlon.
This study12 proposed that the association of the COL5A1 BstUI RFLP and endurance running performance is mediated by changes in musculotendinous stiffness.
However, a limitation of the study was that neither musculotendinous stiffness nor
joint ROM were measured.
Owing to the previous association of the COL5A1 BstUI RFLP TT genotype
and improved running performance during the 226 km South African Ironman
triathlon, the primary aim of this study was to investigate whether the TT genotype
of the COL5A1 BstUI RFLP was also associated with improved endurance running
performance in the 56 km Two Oceans ultra-marathon.
The secondary aim of this study was to investigate whether the relationship
between the COL5A1 gene and endurance running performance could be explained
by prerace SR ROM. The COL5A1 gene has been previously shown to be associated
with both endurance running ability and SR ROM measurements. Based on these
previous findings, we hypothesize that there will be a significant over-representation
of the TT genotype among those athletes with both a decreased SR ROM (inflexible) and a faster ultra-marathon finishing time.
Seventy-two Caucasian runners (52 male and 20 female) were recruited from the
2009 56 km Two Oceans ultra-marathon running race held during April in Cape
Town, South Africa. The runners were of well-trained club level, with 71 of the
72 recruited athletes finishing the race. The participants of this study represent a
subcohort of a larger study that investigated the association of the COL5A1 BstUI
RFLP with SR ROM measurements.10 The current study therefore represents
all participants within the larger study10 (N = 325), which were recruited at and
completed the Two Oceans ultra-marathon running race. Subjects were required
COL5A1, Running, and Range of Motionâ€ƒ â€ƒ 487
to complete a written informed consent before participating in this trial. SR ROM
was measured at the race registration, which occurred within the three day period
before the event. Sit and reach ROM was assessed by the Canadian Trunk Forward
Flexion Test, which was performed as detailed in the ACSM guidelines for exercise
testing,13 with minor modification.10 Training data were obtained from a detailed
questionnaire completed by 38 (53%) participants. Both weight and height were
only self-reported in 49 (69%) of the participants. Overall race and split times were
obtained from the race website (www.twooceansmarathon.org.za) after the event.
Samples were genotyped anonymously. The recruitment, genotyping, and statistical analyses of this study were performed bearing the guidelines of â€œreplicating
genotype-phenotype associationsâ€ in mind.14 Deviations from these guidelines were
noted in the manuscript. This study was approved by the Human Research Ethics
Committee of the Faculty of Health Sciences within the University of Cape Town,
South Africa, as well as the race organizers of the Two Oceans event.
DNA Extraction and COL5A1 Genotyping
Five milliliters of venous blood was obtained from each participant by venipuncture
of the forearm vein and collected into an EDTA vacutainer tube. Blood samples
were stored at 4Â°C until total DNA extraction was performed. The DNA was
extracted using standard procedures as described by Lahiri and Nurnberg,15 and
modified by Mokone et al.16 All participants were genotyped for the BstUI RFLP
(SNP rs12722) within the 3â€²-untranslated region (UTR) of the COL5A1 gene (TT,
TC, or CC), as previously described.16,17 Whereas a second technology was not
used to verify the genotyping results, the particular RFLP had an internal digestion
control that confirmed whether the enzyme had cut or not. A subset of samples
(10%) were genotyped by two authors independently (JB and MP) to confirm correct genotyping of samples. Furthermore, gels were independently read and the
results confirmed by two investigators.
All statistical analyses were performed only on those who completed the ultramarathon. ANOVA was used to examine differences between genotype (or allele)
groups and continuous data. Where appropriate, a ScheffÃ© post hoc analysis was
used to determine which of the three genotypes were significantly different from
each other. Chi-squared tests were used to examine differences between genotype
(or allele) groups and sex. Leveneâ€™s tests of homogeneity were performed to test for
differences in homogeneity of the data. Where appropriate and hypothesis driven,
the TT genotype group was compared with the combined TC and CC genotype
groups. For genotype effects on running performance, both unadjusted P values
and P values adjusted for weight were calculated.
In addition, the magnitude of changes in performance variables was determined
on a scale of effect sizes where <0.2 = trivial, 0.21â€“0.6 = small, 0.61â€“1.2 = moderate, 1.21â€“2.0 = large, 2.1â€“4.0 = very large, and >4.0 = nearly perfect. A multivariate
analysis was used to determine the model that best predicted race time with factors
that were significantly associated with race time (age, weight, and COL5A1 BstUI
RFLP genotype were included in the model). Hardy-Weinberg equilibrium was
488 Brown et al.
calculated using Genepop version 4.0.10 (http://genepop.curtin.edu.au). Genotype
frequencies were compared with previously published studies and public databases
for Caucasian populations and found to be in accordance with these frequencies.
The COL5A1 BstUI RFLP genotype distribution (29% TT, n = 21; 46% TC, n =
33; 25% CC, n = 18) within this study was in Hardy-Weinberg equilibrium (P =
0.485). Age (P = .328), height (P = .369), and sex (P = .204) were similar between
the three genotype groups (Table 1). However, weight (P = .002) and BMI (P =
.027) were significantly different between genotype groups. Similar significant
differences were observed when individuals with a TT genotype were compared
with individuals with a TC or CC genotype. The average weekly distance run (km/
week) during the 15 wk before the event were similar among the three genotype
groups. Although the SR ROM was significantly different (P = .034) between
the three genotypes, there were no significant differences when ScheffÃ© post hoc
analysis was performed on these data. In addition, there were no significant differences (P = .749) when the TT genotype was compared with the combined TC
and CC genotypes.
The overall level of the race competitors, along with that of the competitors
who formed part of the current study, is provided with the following summary:
â€¢ 5824 of the athletes completed the 2009 ultra-marathon within the 7 h cut-off
time, of which 188 (3.2%) finished the event in under 4 h and 924 (15.9%)
completed the race in between 4 and 5 h;
â€¢ 2400 (41.2%) and 2312 (39.7%) of the athletes completed the race in between
5 and 6 h, and 6 and 7 h, respectively (www.twooceansmarathon.org.za);
â€¢ the majority (n = 38, 53%) of the 71 subjects who participated in this study
finished the 56 km race in between 5 and 6 h, at an average time of 5:49 min/
km (Â± 0:29 min, range: 4:31 to 6:47 min); and
â€¢ the remainder of the subjects completed the ultra-marathon in between 6 and
7 h, at an average time of 7:02 min/km (Â± 19 s, range: 4:27 to 6:40 min).
The COL5A1 Gene and Running Performance
Time to complete the 56 km ultra-marathon had a tendency to be different (P =
0.053, weight-adjusted P = .046) among the three genotype groups (Table 1). On
average, individuals with a TT genotype (341 Â± 41, n = 21) were significantly (P
= 0.014, weight-adjusted P = 0.013) faster overall than individuals with a TC or
CC genotype (365 Â± 39, n = 50) (Figure 1). The magnitude of the change in performance between the individuals with a TT genotype and individuals with either
a TC or CC genotype was considered â€œmoderateâ€ (effect size = 0.61).
Table 1 General characteristics of the 56 km Two Oceans ultra-marathon athletes when
divided into the three COL5A1BstUI restriction fragment length polymorphism (RFLP)
genotype (TT, TC, and CC) groups
COL5A1 BstUI RFLP Genotype Groups
P-valuea P-valueb TT TC CC
Age (y) 38.5 Â± 9.6 (21) 42.5 Â± 9.3 (32) 40.6 Â± 9.2 (18) 0.328 0.181
Sex (% males) 76.2 (16) 78.1 (25) 55.6 (10) 0.204 0.597
Height (m) 1.76 Â± 0.09 (16) 1.79 Â± 0.07 (20) 1.75 Â± 0.07 (13) 0.369 0.554
Weight (kg) 69.9 Â± 8.1 (17) 79.8 Â± 12.4 (21) 72.5 Â± 8.6 (14) 0.002 0.005
BMI (kg/m2) 22.5 Â± 1.7 (16) 24.5 Â± 2.5 (20) 23.8 Â± 2.5 (13) 0.027 0.008
Training (km/week) 58 Â± 2 (11) 58 Â± 14 (17) 49 Â± 23 (9) 0.500 0.657
SR ROM (mm) 249 Â± 91 (21) 231 Â± 95 (32) 304 Â± 94 (18) 0.034c 0.749
Finishing Time (min) 341 Â± 41 (21) 369 Â± 34 (32) 357 Â± 47 (18) 0.053 0.014
Note. Age, height, weight, body mass index (BMI), training (during the 15 wk period before the event), and race time are expressed as an
average Â± standard deviation, while sex is expressed as a frequency. The number of subjects with nonmissing data (N) is in parentheses.
Age, height, weight, and training data were self-reported. BMI was calculated as kilograms per meter squared. Boldface indicates a
significant difference (P-value < 0.05).
aTT vs TC vs CC.
bTT vs TC + CC.
cNo significant differences for post hoc analyses (ScheffÃ©).
490 Brown et al.
Contributors to Variance in Race Finishing Time
For the overall running performance (56 km); age, sex, weight, and COL5A1
BstUI RFLP genotype (TT vs TC + CC) were assessed for their contribution to
race time variance using a casewise multiple regression model. The best model,
which accounted for 35% (P < .001, standard error of the estimate = 32.76) of
the variance included age, weight, and COL5A1 BstUI RFLP genotype. Of these,
only weight (P = 0.018) and genotype (P = .038) contributed significantly to the
overall race time model. When weight was excluded from the analysis, age (P
= .018) and genotype (P = .047) contributed significantly to the variance in race
time (Table 2). Together with sex, these three variables accounted for 19% of the
variance in performance.
Relationship Between Running Performance, COL5A1 BstUI
RFLP Genotype and SR ROM
There was no correlation between overall finish time and prerace SR ROM (r =
â€“0.104, n = 71, P = 0.390). However, for the purposes of this study, the genotype
frequency between quadrants were analyzed (Figure 2). Quadrants were defined
by the median value of SR ROM and time to complete the ultra-marathon (performance). Participants who had more ROM than the median were termed flexible,
Figure 1 â€” Running performance, expressed as the time (in minutes) to complete the
56 km Two Oceans ultra-marathon race. Mean time, with an error bar to indicate standard
deviation, is compared between participants with the COL5A1 BstUI restriction fragment
length polymorphism (RFLP) TT genotype and those with a TC and CC genotype.
COL5A1, Running, and Range of Motionâ€ƒ â€ƒ 491
whereas those with less ROM than the median were termed inflexible. Similarly,
those faster than the median were termed fast and those slower than the median
were termed slow (Figure 2). All participants were divided into quadrants based
on these categories: inflexible-fast (group 1), inflexible-slow (group 2), flexiblefast (group 3), and flexible-slow (group 4). To reduce the limitation of the small
sample size, the T and C allele frequencies were compared between the quadrants.
The T allele was significantly over-represented in participants within group 1
when compared with the remaining participants (T, 69% vs 47%; P = 0.044). To
determine a linear trend, groups 2 and 3 were combined. There was a significant
linear trend for the T allele to be over-represented within inflexible-fast individuals,
whereas the C allele was over-represented in the flexible-slow individuals (group 1,
Table 2 Multivariate analysis for the 56 km Two Oceans ultramarathon overall finishing time (N = 71)
Î² B P-value
Age (y) 0.281 1.23 0.018
Sex 0.149 13.5 0.199
COL5A1 Genotype (TT vs TC + CC) 0.226 20.2 0.047
Note. Î², partial correlation coefficient, and B, parameter estimate. R = .436, R2 = .191, standard error of
the estimate = 37.73, P < .00257. Boldface indicates a significant difference (P-value < 0.05).
Figure 2 â€” The COL5A1 BstUI restriction fragment length polymorphism (RFLP) allele
frequency distribution in four physiological quadrants, based on median values for the cohort,
using sit and reach range of motion (SR ROM) (y-axis) and overall race time performance
(x-axis) as parameters: group 1, inflexible-fast; group 2, flexible-fast; group 3, flexibleslow; and group 4, inflexible-slow. The groups are divided as indicated by the dashed line,
which represents the median line. Deviations on the median line indicate which quadrant
participants were incorporated into if they were the median values, based on the definition.
492 Brown et al.
69%, vs group 2+3, 51%, vs group 4, 36%; P = 0.037). Similar trends were found
when the genotypes were compared between the four groups; however, owing to
the small sample size, the significance of this analysis could not be determined
(data not shown).
The main finding of this study was that the COL5A1 BstUI RFLP TT genotype
was associated with improved endurance running performance in a 56 km ultramarathon. While there was only a trend for performances difference among the
three genotypes groups, the TT genotype were significantly faster when compared
with the C allele (TC and CC genotype combined). The TC and CC genotypes
were grouped together based on our a priori hypothesis and small sample size. This
finding confirms that of a previous study, which reported an association between
the COL5A1 TT genotype and faster performance during the 42.2 km run of an
Ironman triathlon.12 An additional novel finding of this study was that the T allele
was significantly over-represented among inflexible-fast (group 1) individuals.
This indicates a potential association between endurance running performance,
COL5A1, and SR ROM.
Endurance performance is a multifactorial phenotype resulting from a poorly
understood complex interaction of environmental and genetic factors.18 The
majority of genes that have been associated with endurance performance to date
encode proteins involved in metabolic pathways and skeletal muscle biology.3 To
our knowledge, this is the first extracellular matrix protein encoding gene that has
been associated with running performance. Of further interest is that the genotype
association with performance was evident in the running, but not the swimming or
cycling events of the Ironman triathlon in the previous study.12 The greater amount
of loading placed on the body may explain why this association is seen only with
endurance running performance, and not the swimming or cycling. Furthermore,
the magnitude of the effect of COL5A1 on endurance running performance was
calculated as being â€œmoderateâ€ in this study. Owing to the polygenic nature of
the endurance phenotype,3 it is highly unlikely that a single gene would have any
greater magnitude of effect. Based on the current study design and the limited available training data, it was not possible to determine whether this gene-performance
interaction has a direct effect on running performance during the event or enables
athletes to train harder in preparation for the race. Both mechanisms are plausible
and further research is required to elucidate this question. As with other biological
systems, connective tissue is encoded by multiple genes. It is therefore likely that
other extracellular matrix encoding genes need to be considered for their contribution to the performance phenotype.
Together with the COL5A1 genotype (TT vs TC + CC), age and sex accounted
for 19% of the variance in running performance for the ultra-marathon. Although
the contribution of age and sex to endurance running performance was previously
described, the contribution of COL5A1 genotype to performance is novel and worth
exploring in future studies.
The COL5A1 genotype alone accounted for approximately 7% of the variance
in performance and contributed significantly to the model (P = .027). The fact
that weight was significantly associated with the COL5A1 BstUI RFLP genotype
COL5A1, Running, and Range of Motionâ€ƒ â€ƒ 493
in this cohort cannot be explained, and it should be reiterated that not all subjects
reported their height and weight. This finding may therefore be spurious and should
be interpreted with caution owing to the small sample size. In addition, there is
no evidence for a COL5A1 BstUI RFLP genotype-weight association from the
published literature of several larger diverse cohorts.10,12,16,19,20
We recently reported10 that the CC genotype was associated with greater SR
ROM in a larger cohort consisting of 325 subjects. The participants of the current study represent part of a larger study that investigated the association of the
COL5A1 BstUI RFLP with SR ROM measurements. The current study included
only the participants who completed the 2009 Two Oceans 56 km ultra-marathon,
and therefore it was not an objective of this particular study to investigate the
relationship between SR ROM and the COL5A1 gene. However, this subcohort
enabled us to investigate the previously proposed relationship between running
performance, SR ROM, and COL5A1 genotype.12
In the current study, we did not find a direct association between SR ROM
measurements (flexibility) and time to complete the 56 km ultra-marathon race.
We believe that Figure 3 explains this lack of association due to both ROM and
endurance running performance being complex multifactorial phenotypes.1,21
Although there are associated factors that are common to both endurance running
performance and joint ROM (eg, age, sex, and the COL5A1 BstUI genotype),9,12
there are also unique factors, which are associated with only one or the other
phenotype (eg, slow-twitch muscle fiber proportion for endurance performance
and limb dominance for ROM) (Figure 3).22,23 We propose that muscle-tendon
stiffnessâ€”an additional intrinsic factor common to both running performance and
SR ROM24,25â€”is the most plausible biological explanation for the findings of this
study and previously published associations with the COL5A1 gene.9,10,12,20 In support of this proposed mechanism, COL5A1 haplo-insufficiency in mice (Â± mutants)
leads to a significantly different connective tissue elasticity modulus (stiffness).26
This finding therefore supports a role of COL5A1 in modulating the biomechanical
properties of muscle-tendon units. Moreover, SR ROM and running performance
are both determined by, among other factors, the biomechanical properties of the
musculoskeletal system. Increased muscle-tendon stiffness is associated with a
reduced ROM24 and increased running performance.25 It is, therefore, not surprising that in our study, there was an over-representation of the T allele within the
â€œinflexible-fastâ€ athletes. Conversely, there was an over-representation of the C
allele within the â€œflexible-slowâ€ athletes. Because of the small sample sizes, this
finding should be interpreted with caution and repeated in a larger cohort.
The main limitations of this study were the small sample sizes within genotype
groups and that not all participants had height and weight measurements recorded.
The fact that weight was significantly associated with genotype emphasizes the
latter limitation. Nevertheless, running performance between the genotype groups
remained significant after adjusting for the available weight measurements. Furthermore, the COL5A1 genotype significantly contributed to the multiple regression
model irrespective of whether weight was included or excluded from the analysis.
Future biomechanical studies are required to further explore the relationships
detailed in this study. Direct measurements of muscle-tendon stiffness are required
to gain a clearer insight into the mechanism of the association between an extracellular matrix encoding gene and endurance running performance.
494 Brown et al.
In conclusion, the COL5A1 TT genotype was associated with improved endurance running performance in a cohort of ultra-endurance athletes. The sit-and-reach
ROM of this genotype did not explain this association with performance, but the T
allele was significantly over-represented in the inflexible-fast athletes within this
study. Larger cohort studies are required before the relationship between endurance running performance and the COL5A1 BstUI RFLP can be fully understood.
The study confirms a previous novel finding12 that the BstUI RFLP TT genotype
of COL5A1 is associated with faster endurance running performance. While there
was no direct relationship between prerace SR ROM and running performance,
the T allele was significantly over-represented within the fast-inflexible athletes.
Figure 3 â€” A hypothetical diagram emphasizing the proposed relationship of unique and
common factors associated with endurance running performance and joint range of motion
(ROM). Both joint ROM measurements (flexibility) and endurance running ability are complex
multifactorial phenotypes. Although there are unique factors that are associated only with
one or the other phenotype, there are also factors associated with both phenotypes (such as
muscle-tendon stiffness and the COL5A1 genotype). The intrinsic and extrinsic factors that
are associated with both endurance running performance and ROM are indicated between
the diagonal dashed lines. The solid black and white shaded individuals represent those who
are inflexible-fast and flexible-slow, respectively. Their endurance running performance and
ROM are predominately influenced by common factors. Individuals whose improved running
performance is influenced predominately by unique factors, which do not influence ROM, have
black shaded legs. Conversely, individuals whose reduced ROM is influenced predominately
by unique factors, which do not influence running performance, have black shaded torsos.
COL5A1, Running, and Range of Motionâ€ƒ â€ƒ 495
Weight, height, and training data were not available for all the athletes; nevertheless, age and COL5A1 genotype accounted for a substantial proportion (19%) of
running performance in this cohort. As result, this study emphasizes the importance
of considering genes that encode for connective tissue proteins within musculoskeletal soft tissues when examining endurance running performance. Furthermore,
the over-representation of the COL5A1 T allele in the fast-inflexible athletes supports our hypothesis that there is a relationship between the COL5A1 gene, joint
ROM, and endurance running performance. This relationship should be further
investigated within a larger cohort, using direct measures of muscle-tendon stiffness and running economy.
This work was supported, in part, with funds from the South African National Research
Foundation (grant number IFR2009021600020 to MC), the South African Medical Research
Council and the University of Cape Town. Brown was supported by postgraduate funding
from the University of Cape Town, as well as funding from the Department of Science and
Technology and the National Research Foundation in South Africa. The authors would also
like to thank all the staff and students within the UCT/MRC Research Unit for Exercise
Science and Sports Medicine who assisted with data collection during the Two Oceans
1. Joyner MJ, Coyle EF. Endurance exercise performance: the physiology of champions.
J Physiol. 2008;586(1):35â€“44.
2. Laursen PB, Rhodes EC. Factors affecting performance in an ultraendurance triathlon.
Sports Med. 2001;31(3):195â€“209.
3. Bray MS, Hagberg JM, Perusse L, et al. The human gene map for performance and
health-related fitness phenotypes: the 2006-2007 update. Med Sci Sports Exerc.
4. Gleim GW, McHugh MP. Flexibility and its effects on sports injury and performance.
Sports Med. 1997;24(5):289â€“299.
5. Davison RR, Van Someren KA, Jones AM. Physiological monitoring of the Olympic
athlete. J Sports Sci. 2009;27(13):1433â€“1442.
6. Gleim GW, Stachenfeld NS, Nicholas JA. The influence of flexibility on the economy
of walking and jogging. J Orthop Res. 1990;8(6):814â€“823.
7. Jones AM. Running economy is negatively related to sit-and-reach test performance
in international-standard distance runners. Int J Sports Med. 2002;23(1):40â€“43.
8. Craib MW, Mitchell VA, Fields KB, Cooper TR, Hopewell R, Morgan DW. The association between flexibility and running economy in sub-elite male distance runners.
Med Sci Sports Exerc. 1996;28(6):737â€“743.
(9) Collins M, Mokone GG, September AV, van der Merwe L, Schwellnus MP. The COL5A1
genotype is associated with range of motion measurements. Scand J Med Sci Sports.
10. Brown JC, Miller C-J, Schwellnus MP, Collins M. Range of motion measurements
diverge with increasing age for COL5A1 genotypes. Scand J Med Sci Sports.
11. Riley G. The pathogenesis of tendinopathy. A molecular perspective. Rheumatology
496 Brown et al.
12. Posthumus M, Schwellnus MP, Collins M. The COL5A1 Gene: A Novel Marker of
Endurance Running Performance. Med Sci Sports Exerc. 2011;43:584â€“589.
13. American College of Sports Medicine. ACSMâ€™s guidelines for exercise testing and
prescription. 8th ed. Lippincott Williams & Wilkins; 2010.
14. Chanock SJ, Manolio T, Boehnke M, et al. Replicating genotype-phenotype associations. Nature. 2007;447(7145):655â€“660.
15. Lahiri DK, Nurnberger JI, Jr. A rapid non-enzymatic method for the preparation of
HMW DNA from blood for RFLP studies. Nucleic Acids Res. 1991;19(19):5444.
16. Mokone GG, Schwellnus MP, Noakes TD, Collins M. The COL5A1 gene and Achilles
tendon pathology. Scand J Med Sci Sports. 2006;16(1):19â€“26.
17. Greenspan DS, Pasquinelli AE, Bst UI, Dpn II. RFLPs at the COL5A1 gene. Hum Mol
18. Davids K, Baker J. Genes, environment and sport performance: why the nature-nurture
dualism is no longer relevant. Sports Med. 2007;37(11):961â€“980.
19. September AV, Schwellnus MP, Collins M. Tendon and ligament injuries: the genetic
component. Br J Sports Med. 2007;41(4):241â€“246.
20. Posthumus M, September AV, Oâ€™Cuinneagain D. van der MW, Schwellnus MP, Collins
M. The COL5A1 gene is associated with increased risk of anterior cruciate ligament
ruptures in female participants. Am J Sports Med. 2009;37(11):2234â€“2240.
21. Corbin CB. Flexibility. Clin Sports Med. 1984;3(1):101â€“117.
22. Macedo LG, Magee DJ. Differences in range of motion between dominant and
nondominant sides of upper and lower extremities. J Manipulative Physiol Ther.
23. Suminski RR, Mattern CO, Devor ST. Influence of racial origin and skeletal muscle properties on disease prevalence and physical performance. Sports Med. 2002;32(11):667â€“
24. Magnusson SP, Simonsen EB, Aagaard P, Boesen J, Johannsen F, Kjaer M. Determinants
of musculoskeletal flexibility: viscoelastic properties, cross-sectional area, EMG and
stretch tolerance. Scand J Med Sci Sports. 1997;7(4):195â€“202.
25. Kubo K, Tabata T, Ikebukuro T, Igarashi K, Yata H, Tsunoda N. Effects of mechanical
properties of muscle and tendon on performance in long distance runners. Eur J Appl
26. Wenstrup RJ, Florer JB, Davidson JM, et al. Murine model of the Ehlers-Danlos syndrome. col5a1 haploinsufficiency disrupts collagen fibril assembly at multiple stages.
J Biol Chem. 2006;281(18):12888â€“12895.
Copyright of International Journal of Sports Physiology & Performance is the property of Human Kinetics
Publishers, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder’s express written permission. However, users may print, download, or email articles for