Effect of gene polymorphisms on the mechanical properties of human tendon structures

Recent studies showed that polymorphisms in alpha 1 chains of types I (COL1A1) and V (COL5A1) collagen, growth and differentiation factor 5 (GDF5), and matrix metalloproteinase 3 (MMP3) genes were associated with injuries in tendons and ligaments (e.g., September et al. (Br J Sports Med 43: 357–365 2009)). In the present study, we aimed to investigate the effects of injury-associated polymorphisms within these four genes on the mechanical properties of human tendon structures in vivo. One hundred Japanese males participated in this experiment. The mechanical properties of tendon structures in knee extensors and plantar flexors were measured using ultrasonography. All subjects were genotyped for COL1A1 rs1800012, COL5A1 rs12722, GDF5 rs143383, and MMP3 rs679620 single nucleotide polymorphisms. For COL1A1, all subjects had a GG genotype. For COL5A1, maximal tendon elongation and strain of individuals with a CC genotype were significantly greater than individuals with other genotypes (combined TT and CT) for knee extensors, but not for plantar flexors. For GDF5 and MMP3, there were no differences in the mechanical properties of tendon structures in knee extensors and plantar flexors among the three genotypes. The present study demonstrated that subjects with a CC genotype of the COL5A1 gene had more extensible tendon structures than those of subjects with other genotypes (combined TT and CT) for knee extensors, but not for plantar flexors. The results presented in this study need to be confirmed in a larger cohort of subjects.


Introduction
Recent studies showed that polymorphisms within alpha 1 chains of types I (COL1A1) and V (COL5A1) collagen, growth and differentiation factor 5 (GDF5), and matrix metalloproteinase 3 (MMP3) genes were associated with tendon and/or ligament injuries (Posthumus et al. 2009a(Posthumus et al. ,b, 2010Raleigh et al. 2009;September et al. 2009). On the other hand, the mechanical properties of tendons and ligaments would be expected to be one of the risk factors for these injuries. More recently, Collins et al. (2009) and Brown et al. (2011) demonstrated that the COL5A1 rs12722 single nucleotide polymorphism was related to range of motion in the lower limb. Furthermore, Kato et al. (2010) suggested that an increase in range of motion due to static stretching was attributable to a change in tendon, not muscle, stiffness. Considering these points, the mechanical properties, such as maximal elongation and stiffness, of tendons and ligaments would be associated with gene polymorphisms mentioned above.
For the last decade, several reports have used ultrasonography to investigate the relationship between tendon properties and performances during stretch-shortening cycle exercises (Kubo et al. 1999(Kubo et al. , 2000(Kubo et al. , 2011Stafilidis and Arampatzis, 2007). In addition, some previous studies have demonstrated the effects of resistance training on the mechanical properties of human tendons in vivo (Kongsgaard et al. 2007;Kubo et al. 2001Kubo et al. , 2007Kubo et al. , 2009Reeves et al. 2003). According to these previous findings, we have no means of enhancing the extensibility of tendon structures, i.e., tendon properties change to be suitable for stretch-shortening cycle exercises, except for bed rest (Kubo et al. 2004;Reeves et al. 2005) and detraining (Kubo et al. 2010). Furthermore, crosssectional studies demonstrated that tendon structures were more compliant in excellent sprinters compared to inferior sprinters and untrained subjects for knee extensors, but not for plantar flexors (Kubo et al. 2000(Kubo et al. , 2011Stafilidis and Arampatzis, 2007). Accordingly, it has been assumed that these compliant tendon structures in excellent sprinters are partly determined by genetic factors. In particular, this tendency would be found more clearly in knee extensors than in plantar flexors.
In the present study, we aimed to investigate the effects of single nucleotide polymorphisms within COL1A1, COL5A1, GDF5, and MMP3 genes previously shown to be associated with tendon and/or ligament injuries (Posthumus et al. 2009a(Posthumus et al. ,b, 2010September et al. 2009) on the mechanical properties of human tendon structures (outer tendon and aponeurosis) in vivo. In addition, we also examined whether site-differences in these relationships were found between knee extensors and plantar flexors.

Subjects
One hundred Japanese males (age: 22.0 ± 3.3 yrs, height: 172.6 ± 5.5 cm, body mass: 67.9 ± 10.4 kg, mean ± SD) participated in this experiment. They were undergraduate and graduate students of three universities. When data were collected, subjects were involved in recreational sports activity on average not more than twice per week or 1 hour per week in the past 3 years. None of the subjects reported any current or recent lower limb injuries in the 3 years before testing. Subjects were fully informed of the procedures to be utilized as well as the purpose of this study. Written informed consent was obtained from all subjects. This study was approved by the office of the Department of Sports Sciences, University of Tokyo, and complied with their requirements for human experimentation.

Elongation and stiffness of tendon structures
Maximal voluntary isometric contraction (MVC) was measured by means of specially designed dynamometers (Applied Office, Tokyo, Japan) for knee extension and plantar flexion, respectively. All measurements were performed on the right lower limb. During each task, subjects exerted isometric torque from zero (relax) to MVC within 5 s. Torque signals were amplified and sampled at 1 kHz using a 16-bit A/D converter (PowerLab/16SP, AD Instruments, Australia). During the knee extension task, the hips and back were held tightly in the seat using adjustable lap belts. The right ankle was firmly attached to the lever arm of the dynamometer with a strap and fixed with the knee joint flexed at an angle of 90 deg (full extension = 0 deg). During the plantar flexion task, subjects lay prone on a test bench and the waist and shoulders were secured by adjustable lap belts and held in position. The ankle joint was set at 90 deg with the knee joint at full extension and the right foot was securely strapped to a footplate connected to the lever arm of the dynamometer.
Elongations in tendon structures (outer tendon and aponeurosis) of knee extensors and plantar flexors were assessed during isometric contractions. An ultrasonic apparatus (SSD-6500, Aloka, Tokyo, Japan) with an electronic linear array probe (7.5-MHz wave frequency with 80 mm scanning length; UST 5047-5, Aloka) was used to obtain longitudinal ultrasonic images of vastus lateralis and medial gastrocnemius muscles by procedures described previously (Kubo et al. 2007(Kubo et al. , 2009). Two measured sites were selected for measurements: at 50% of the distance between the greater trochanter and the lateral epicondyle of the femur for vastus lateralis muscle and at 30% of the distance between the popliteal crease and the centre of the lateral malleolus for medial gastrocnemius muscle. Ultrasonic images were recorded on videotape at 30 Hz and synchronized with recordings of a clock timer for subsequent analysis. The point at which one fascicle was attached to the aponeurosis was visualized on ultrasonic images. The displacement of this point is considered to indicate lengthening of the deep aponeurosis and distal tendon. To correct measurements taken for tendon and aponeurosis elongation, additional measurements were taken under passive conditions (Kubo et al. 2007(Kubo et al. , 2009. For each subject, the displacement of each site obtained from ultrasonic images could be corrected for that attributed to joint rotation alone. In this study, only values corrected for angular rotation were reported. The tendon elongation value (L) was converted to strain by the following equation (Kubo et al. 1999): where TL is the length of the tendon structure at rest. We measured the distance between the measurement site for L and the insertion of the patella and Achilles tendons (confirmed using ultrasonography).
Torque (TQ) measured during isometric contractions was converted to muscle force (Fm) by the following equation (Kubo et al. 2007(Kubo et al. , 2009: where k is the relative contribution of physiological crosssectional area in each vastus lateralis muscle within knee extensors and medial gastrocnemius muscle within plantar flexors, and MA is the moment arm length in each quadriceps femoris muscles at 90 deg and triceps surae muscle at 90 deg, which was estimated from the limb length of each subject. In this study, Fm and L above 50% of MVC were fitted to a linear regression equation, the slope of which was adopted as stiffness (Kubo et al. 2007(Kubo et al. , 2009.
In a preliminary study, the repeatability of the tendon properties measurement was investigated on 2 separate days with 10 male among all subjects. The coefficient of variation was 5.8% for maximal strain and 6.3% for stiffness.

DNA extraction and genotyping
Total DNA was isolated from saliva (2 ml) using Orangene DNA (DNA Genotek, Ottawa, Ontario, Canada). Saliva samples were stored at room temperature until total DNA extraction. Genotypes of four polymorphisms {COL1A1 rs1800012 (G/T), COL5A1 rs12722 (T/C), GDF5 rs143383 (T/C), and MMP3 rs679620 (G/A)} were determined at G&G Science (Fukushima, Japan) by a method that combines PCR and sequence-specific oligonucleotide probes with suspension array technology (Luminex, Austin, Texas, USA). Primers and probes for genotyping are shown in Table 1. Detailed genotyping methodology was described previously (Itoh et al. 2005).

Statistics
Descriptive data are represented as the means ± SD. Any significant differences in measured variables among the three-genotype groups were tested by a one-way ANOVA. When the overall F value was significant, a Tukey's honest significance post hoc test was used to determine specific differences. The level of significance was set at p<0.05.

Results
For COL1A1 rs1800012 (G/T), all subjects had a GG genotype. For COL5A1 rs12722 (T/C), GDF5 rs143383 (T/C), and MMP3 rs679620 (G/A), there were no significant differences in age, height, or body mass between the three genotype groups of each single nucleotide polymorphism (Table 2).
For COL5A1, the subjects of TT and CT genotypes combined, since the number of subjects with a TT genotype was only two. In both knee extensors and plantar flexors, there were no significant differences (p>0.05) in the MVC values between COL5A1 (Table 3), GDF5 (Table 4), and MMP3 (Table 5) genotype groups. For COL5A1, maximal tendon elongation and strain of individuals with a CC genotype were significantly greater than individuals with other genotypes (combined TT and CT) for knee extensors (p=0.012 for maximal elongation, p=0.008 for maximal strain), but not for plantar flexors (both p>0.05) (Table 3). Similarly, the stiffness of individuals with a CC genotype was significantly lower compared to other genotypes (combined TT and CT) in knee extensors only (p=0.013). For GDF5 (Table 4) and MMP3 (Table 5), there were no significant differences (p>0.05) in the mechanical properties of tendon structures among the three genotype groups of each single nucleotide polymorphism.

Discussion
The main finding of this study was that subjects with a CC genotype of the COL5A1 gene had more extensible tendon structures compared to subjects with other genotypes (combined TT and CT) for knee extensors, but not for plantar flexors. To our knowledge, this is the first study to demonstrate the relationship between any mechanical properties of tendon structures and a gene polymorphism in vivo.
This study suggested the possibility that tendon structures of individuals with a COL5A1 rs12722 CC genotype were more extensible than individuals with other genotypes (combined TT and CT). A previous study suggested that the COL5A1 gene was associated with benign joint hypermobility syndrome (Grahame, 1999). More recently, Collins et al. (2009) and Brown et al. (2011) reported that the COL5A1 rs12722 single nucleotide polymorphism was associated with range of motion in the lower limb. Several researchers have suggested that the major factor contributing to range of motion, i.e., flexibility, is the extensibility of muscles and tendons (Jewell and Wilkie, 1958;Kato et al. 2010;McHugh et al. 1998). Therefore, the present result was supported by the findings of Brown et al. (2011) and Collins et al. (2009). On the other hand, Goncalves-Neto et al. (2002) and Satomi et al. (2008) reported that damaged and pathological tendons contained relatively higher proportion of collagen type III and V, and these alterations were accompanied by a reduction in type I collagen. According to previous findings (Birk, 2001;Roulet et al. 2007), type V collagen expression levels are critical in determining fiber diameter and strength, although type V collagen is a quantitatively minor fibril-forming collagen. In addition, type V collagen gene expression can be, at least in part, determined by polymorphisms within the 3'-UTR of COL5A1 (Laguette et al. 2011). Therefore, we may say that COL5A1 gene expression, and by implication type V collagen production, is one of the factors that determine the mechanical properties of human tendon structures. On the other hand, there were no differences in tendon properties in plantar flexors among the three genotypes of COL5A1 (Table 3). This implied that the degree of genetic effects on tendon properties is different between knee extensors and plantar flexors. Crosssectional studies demonstrated that tendon structures were more compliant in excellent sprinters than that in inferior sprinters and untrained subjects for knee extensors, but not for plantar flexors (Kubo et al. 2000(Kubo et al. , 2011Stafilidis and Arampatzis, 2007). In addition, according to longitudinal studies (e.g., Kubo et al. 2007), we have no training protocol to enhance the extensibility of tendon structures. Considering these points, it has been assumed that these compliant tendon structures for knee extensors in excellent sprinters are partly determined by genetic factors. In addition, our previous study showed that age-associated muscle thickness loss in plantar flexors was less than that in knee extensors (Kubo et al. 2003). The reasons for the differences in the declines in muscle thickness with aging were unclear, but several possibilities exist, i.e., postnatal and genetic factors. In particular, these discrepancies may be due to differences in the daily activity levels between knee extensors and plantar flexors. Indeed, some previous studies indicated that the relative activation level and exerted torque of plantar flexors are higher than those of knee extensors during normal walking (DeVita et al. 1996;Ericson et al. 1986). Considering these points, it seems reasonable to suppose that the mechanical properties of tendon structures for plantar flexors are affected greatly by a postnatal factor.
In previous studies between gene polymorphisms and tendon injuries (Posthumus et al. 2009a(Posthumus et al. ,b, 2010Raleigh et al. 2009;September et al. 2009), South African and Australian and Caucasian populations were investigated. To date, no studies have investigated this theme in Japanese populations. For all gene polymorphisms (COL1A1, COL5A1, GDF5, and MMP3), the distribution of each gene polymorphism in the present study was different from previously reported distributions in Caucasian populations (Posthumus et al. 2009a(Posthumus et al. ,b, 2010Raleigh et al. 2009;September et al. 2009). We hypothesized that the genotype of COL1A1 rs180002 single nucleotide polymorphism, in which more than one study previously reported the relationship between gene polymorphism and injuries (Posthumus et al. 2009a,b), is associated with tendon mechanical properties. Unfortunately, the genotype of this gene (rs180002) was the same among subjects in the present study. Genotype distributions of COL5A1, GDF5, and MMP3 polymorphisms were, however, similar to the distributions reported in public  In the present study, we must draw the attention to the limitations and assumptions of the methodology followed. Firstly, we measured the tendon elongation at the one point of insertion of a fascicle into the aponeurosis. Two measured sites were selected for measurements: at 50% of the thigh length for vastus lateralis muscle and at 30% of the lower leg length for medial gastrocnemius muscle. Therefore, we may say that these measured sites were relatively same among all subjects. Furthermore, our previous study demonstrated that there was no difference in maximal strain of tendon structures among the proximal, central, and distal sites (Kubo et al. 2002). Therefore, we considered that this point did not affect the main results of this study. Secondly, we must confirm that there was no difference in activity level or loading history between the groups with the different genotypes. The subjects in the present study had engaged in recreational sports activity on average not more than twice per week or 1 hour per week in the past 3 years. In addition, there were no differences in MVC (Tables 3, 4 and 5) and muscle thickness (not showing these data) among the genotype groups.
Therefore, we considered that there was no difference in activity level or loading history among the genotype groups. Thirdly, the present study was performed on a small sample size. Nevertheless, the present results showed that tendon structures in knee extensors of subjects with a CC genotype of the COL5A1 gene were more extensible than those with the other genotypes. In a future study, the results presented in this study need to be confirmed in a larger cohort of subjects.
In conclusion, the present study demonstrated that the COL5A1 rs12722 genotype, but none of the three other variants investigated, was associated with the mechanical properties of human tendon structures in vivo. In knee extensors only, the tendon structures of subjects with a CC genotype of the COL5A1 gene were more extensible than those with the other genotypes (combined TT and CT). In a future study, these conclusions await additional data for clarification in a larger cohort of subjects. Furthermore, according to predictive genomics DNA profiling for athletic performance, knowledge of genetic suitability in respect to physical function (e.g., speed, endurance) may be useful for the selection of appropriate sporting event (Kambouris et al. 2012). Therefore, it is possible that the extensibility of tendon structures,