- Email: [email protected]
Carpal tunnel syndrome: The role of collagen gene variants
Gene 587 (2016) 53–58
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Research paper
Carpal tunnel syndrome: The role of collagen gene variants Suhail Dada a, Marilize C. Burger a, Franka Massij a, Hanli de Wet b, Malcolm Collins a,⁎ a b
Division of Exercise Science and Sports Medicine, Department of Human Biology, University of Cape Town, South Africa Life Occupational Health, Life Healthcare, Cape Town, South Africa
a r t i c l e
i n f o
Article history: Received 7 March 2016 Accepted 13 April 2016 Available online 14 April 2016 Keywords: CTS Entrapment neuropathy Genetics Collagen Wrist Injury Median nerve
a b s t r a c t Introduction: The direct causes of idiopathic carpal tunnel syndrome (CTS) are still unknown. It is suggested that pathology of the tendons and other connective tissue structures within the carpal tunnel may play a role in its aetiology. Variants in genes encoding connective tissue proteins, such as type V collagen, have previously been associated with CTS. Since variants within other collagen genes, such as type I, XI and XII collagen, have previously been associated with modulating the risk of musculoskeletal soft tissue injuries, the aim of this study was to determine whether variants within COL1A1, COL11A1, COL11A2 and COL12A1 were associated with CTS. Methods: Self-reported Coloured South African participants, with a history of carpal tunnel release surgery (CTS, n = 103) and matched control (CON, n = 150) participants without any reported history of CTS symptoms were genotyped for COL1A1 rs1800012 (G/T), COL11A1 rs3753841 (T/C), COL11A1 rs1676486 (C/T), COL11A2 rs1799907 (T/A) and COL12A1 rs970547 (A/G). Results: The TT genotype of COL11A1 rs3753841 was significantly over-represented in the CTS group (21.4%) compared to CON group (7.9%, p = 0.004). Furthermore, a trend for the T minor allele to be over-represented in the CTS group (p = 0.055) with a significant association when only female participants (p = 0.036) were investigated was observed. Constructed inferred pseudo-haplotypes including a previously investigated COL5A1 variant, rs71746744 (−/AGGG), suggest gene–gene interactions between COL5A1 and COL11A1 modulate the risk of CTS. Discussion: These findings provide further information of the role of the genetic risk factors and the possible role of variations in collagen fibril composition in the aetiology of CTS. Genetic factors could potential be included in models developed to identify indivisuals at risk of CTS. Strategies that target modifiable risk factors to mitigate the effect of non-modifiable risk factors, such as the genetic risk, could be also developed to reduce incidence and morbidity of CTS. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Pathology of the flexor tendons and the surrounding subsynovial connective tissue (SSCT) have been proposed to play a role in the aetiology of carpal tunnel syndrome (CTS)(Lluch, 1992; Shafer-Crane et al., 2005). In support of this, sequence variants within the gene that Abbreviations: ACL, Anterior cruciate ligament; ANOVA, Analysis of variance; BMI, Body mass index; COL1A1, The gene encoding the α1 chain of type I collagen; COL5A1, The gene encoding for the α1 chain of type V collagen; COL11A1, The gene encoding for the α1 chain of type XI collagen; COL11A2, The gene encoding for the α2 chain of type XI collagen; COL12A1, The gene encoding for the α1 chain of type V collagen; CON, Control group; CTS, Carpal tunnel syndrome; FACITs, Fibril Associated Collagens with Interrupted Triple Helices; HWE, Hardy–Weinberg Equilibrium; OA, Osteoarthritis; OR, Odds Ratio; PCR, Polymerase chain reaction; RA, Rheumatoid arthritis; RFLP, Restriction fragment length polymorphism; SNPs, Single Nucleotide Polymorphism(s); SSCT, Subsynovial connective tissue; UTR, Untranslated region; VEGFA, vascular endothelial growth factor A; VEGFA, The gene encoding VEGFA. ⁎ Corresponding author at: Division of Exercise Science and Sports Medicine, PO Box 115, Newlands 7725, South Africa. E-mail address: [email protected] (M. Collins).
http://dx.doi.org/10.1016/j.gene.2016.04.030 0378-1119/© 2016 Elsevier B.V. All rights reserved.
encodes for the α1 chain of type V collagen, a component of the collagen fibril, which is the basic structural unit of tendons, have recently been shown to alter risk of CTS(Burger et al., 2014a). Although the collagen fibril consists predominantly of type I collagen, several other quantitively minor collagens, including types V, XI and XII, playing an important role in regulating the formation and maintaining the structural integrity of the collagen fibril and surrounding matrix. Variants within the genes encoding subunits of types I, V, XI and XII collagen have previously been investigated and/or implicated for their potential role in various musculoskeletal soft tissue injuries(Burger et al., 2014a; Posthumus et al., 2009a; Posthumus et al., 2009b; September et al., 2009; Posthumus et al., 2009c; Hay et al., 2013; September et al., 2008; Posthumus et al., 2010). The most extensively investigated variant within COL1A1, which encodes for the α1 chain of type I collagen, is the functional Sp1 binding site polymorphism (rs1800012, G/T)(Posthumus et al., 2009b; Ficek et al., 2013; Khoschnau et al., 2008; Mann et al., 2001; Tilkeridis et al., 2005). The TT genotype was associated with reduced risk of anterior cruciate ligament (ACL) in several studies(Posthumus et al., 2009b;
54
S. Dada et al. / Gene 587 (2016) 53–58
Ficek et al., 2013; Khoschnau et al., 2008) and shoulder dislocations in a single study(Khoschnau et al., 2008). The TT genotype has however been reported to be associated with increased risk of lumbar disc disease (Mann et al., 2001; Tilkeridis et al., 2005), but not associated with chronic Achilles tendinopathy(Posthumus et al., 2009a) or tennis elbow(Erduran et al., 2014). These findings highlight the possible association of COL1A1 rs1800012 with altered risk of other injuries, such as CTS. It has been proposed that types XI and V collagen interact to regulate fibrillogenesis during tendon development(Wenstrup et al., 2011). This suggests that variants within the genes encoding for type XI collagen may, similar to COL5A1 variants, also modulate the risk of tendon injuries. In support of this, several variants within COL11A1 (rs3753841, T/ C, and rs1676486, C/T) and COL11A2 (rs1799907, T/A) was associated chronic Achilles tendinopathy in South African and Australian cohorts (Hay et al., 2013). In addition, the COL11A1 variants have also been reported to interact with the COL5A1 rs71746744 (− AGGG) variant to modulate the risk of Achilles tendinopathy (Hay et al., 2013). Similar to the types V and XI fibrillar collagen, the type XII nonfibrillar collagen, which belongs to the family of Fibril Associated Collagens with Interrupted Triple Helices (FACITs), is also involved in the regulation of fibrillogenesis (Young et al., 2002). Variants within the COL12A1 gene have been investigated in both Achilles tendinopathy(September et al., 2008) and ACL injuries(Posthumus et al., 2010). The AA genotype of COL12A1 rs975047 (A/G) was reported to be associated with increased risk of ACL injury in females(Posthumus et al., 2010). Therefore the aim of this study was to investigate whether the COL1A1 rs1800012 (G/T), COL11A1 rs3753841 (T/C), COL11A1 rs1676486 (C/T), COL11A2 rs1799907 (T/A) and COL12A1 rs975047 (A/G) variants, previously associated with altered risk of tendon and/ or ligament injuries, were also associated with altered risk of CTS. 2. Methods A case–control genetic association study, following a candidate gene approach, was conducted and reported using recommendations outlined in the genetic association study specific STREGA initiative (Little et al., 2009). 2.1. Participants A total of 103 self-reported Coloured individuals (94 female and 9 male), who had undergone bilateral (n = 50, 50.4%) or unilateral (dominant hand n = 34, 36.6%; non-dominant hand n = 9, 9.7%; right hand of ambidextrous individuals n = 2, 2.1%) carpal tunnel release surgery, (CTS group) were previously recruited (Burger et al., 2014a). Nerve conduction studies were performed in some cases, however since it is not a requirement for the commissioner of worker's compensation in South Africa, the results were not recorded. Furthermore, 150 (133 female and 17 male) apparently healthy, self-reported Coloured individuals with no history of carpal tunnel syndrome-related symptoms (CON group) were also previously recruited (Burger et al., 2014a). The CTS and CON groups had been matched for type of occupation and years of exposure. The majority of participants (CTS n = 29, 28.4%; CON n = 76, 50.7%) were general poultry processing workers or general workers in other industries requiring repetitive movement of the upper limbs. The other major self-reported occupations included administration (CTS n = 18, 17.6%; CON n = 18, 12.0%) and nursing (CTS n = 11, 10.8%; CON n = 12, 8.0%). The remaining participants were from several occupations where a high percentage of the work day required use of the wrists(Burger et al., 2014a). South African populations who self-identify as Coloured have complex genetic inheritance, with racial intermixing over approximately 350 years. This ethnic group within the Western Cape region of South Africa is ancestrally derived from admixtures of immigrants from
Western Europe, slave labourers from West Africa, Indonesia, Madagascar, Java, India and Malaysia and one or more of the indigenous African populations (Khoe- and San-speaking or Bantu-speaking). The term “Coloured” in South Africa is therefore a name that encompasses a wide range of people who are unique to this country (SASHG, 2013). All participants provided written informed consent according to the Helsinki Declaration. Personal and family medical history was ascertained by means of a questionnaire (Burger et al., 2014a; Burger et al., 2015; Burger et al., 2014b). This study has been approved by the Human Research Ethics Committee of the Faculty of Health Sciences within the University of Cape Town (HREC 158/2011). 2.2. Genotyping Blood collection and DNA extraction had been conducted, as previously described, at the Division of Exercise Science & Sports Medicine, University of Cape Town, South Africa (Burger et al., 2014b). All DNA samples were genotyped for the COL1A1 rs1800012 (G/T, Assay ID: C_7477170_30), COL11A1 rs3753841 (T/C, Assay ID: C_2947954_10), COL11A1 rs1676486 (C/T, Assay ID: C_8400671_10) and COL11A2 rs1799907 (T/A, Assay ID: C_25474257_10) gene variants by means of a custom-designed fluorescence-based Taqman® PCR assays (Applied Biosystems, Foster City, California, USA). Gene-specific primers and allele-specific probes were used in conjunction with a pre-made PCR master mix containing AmpliTaq Gold DNA polymerase (Applied Biosystems, Foster City, CA, USA) in a final reaction volume of 8 μl. PCR reactions were carried out in a StepOnePlus Real-Time PCR system following the manufacturers recommended cycling conditions (Applied Biosystems, Foster City, California, USA). COL12A1 rs970547 (A/G) was genotyped by means of a restriction fragment length polymorphism (RFLP) assay as previously described (September et al., 2008). Positive controls DNA-free controls were randomly included on each PCR plate and a subset of samples was genotyped twice for quality control purposes. Genotyping calls were made by two different researchers with no discrepancies observed. Furthermore, samples that failed to genotype twice were excluded. For quality control, subsets of known genotype controls were added to each plate. A total of 87.3% (n = 89) and 96.1% (n = 98) of the cases and 94.6% (n = 139) and 99.3% (n = 146) of the controls were successfully genotyped for COL11A1 rs3753841 and COL11A1 rs1676486, respectively. Ninety-two percent (n = 94) of the cases and 99.3% (n = 146) of the controls were successfully genotyped for COL11A2 rs1799907. For COL1A1 rs1800012, 97.1% (n = 99) of the cases and 96.6% (n = 142) of the controls were successfully genotyped whereas 89.3% (n = 92) of the cases and the controls (n = 134), respectively, were successfully genotyped for COL12A1 rs970547. 2.3. Statistical analysis Quanto (V.1.2.4, USC, CA, USA) was used to determine the statistical power for a given sample size, based on the expected minor allele frequencies for the individual variants. Non-missing data was analysed using STATISTICA (version 11, StatSoft Inc., Tulsa, Oklahoma, USA) and Graphpad Prism (version 5, GraphPad Software, San Diego, CA, USA). A Pearson's chi-squared test or Fisher's exact test were used to determine significant differences in the genotype and allele distributions as well as other categorical data, namely sex and country of birth, between the CTS and CON groups. An analysis of variance (ANOVA) was used to determine significant differences in continuous data. A significant statistical difference was set at p b 0.05. The Hardy–Weinberg equilibrium (HWE) of the groups was established using the program Genepop web version 4.0.10 (http://genepop.curtin.edu.au/). Inferred haplotypes and pseudo-haplotypes were constructed using Chaplin version 1.2.2 (Emory University School of Medicine, Atlanta, Georgia, USA) and
S. Dada et al. / Gene 587 (2016) 53–58
55
HAPSTAT (version 3.0, University of North Carolina at Chapel Hill, NC, USA) (Epstein and Satten, 2003). 3. Results 3.1. General Characteristics The general characteristics of all participants have previously been described in detail (Burger et al., 2014a). In summary, there was no significant difference between the CTS and CON groups, of the participants successfully genotyped for at least one of the investigated variants, for age at surgery (for the CON group, age of recruitment was used), sex, height, weight and country of birth (Table 1). The CTS and CON groups were also similarly matched for BMI after adjusting for the significant difference in age of recruitment (Table 1). Participants within the CTS (27.5%, n = 28) and CON (14.9%, n = 22) groups self-reported similar medical histories suggested to be associated with CTS with no significant difference (p = 0.178), including: diabetes (9.8% CTS, n = 10 vs 7.4% CON, n = 11), osteoarthritis (OA, 7.8% CTS, n = 8 vs 5.5% CON, n = 8), rheumatoid arthritis (RA, 3.9% CTS, n = 4 vs 0.7% CON, n = 1), thyroid disorders (2.0% CTS, n = 2 vs 0.7% CON, n = 1), diabetes and OA (1.0% CTS, n = 1 vs 0.0% CON), RA and OA (0.0% CTS vs 0.7% CON, n = 1), diabetes and RA (1.0% CTS, n = 1 vs 0.0% CON), diabetes and thyroid disorder (1.0% CTS, n = 1 vs 0.0% CON), RA and systemic lupus erythematosus (1.0% CTS, n = 1 vs 0.0% CON).
Fig. 1. Genotype frequency distribution of COL11A1 rs3753841 for all participants in the carpal tunnel syndrome (CTS, black bars) and control (CON, white bars) groups. The number of participants in each group is indicated. Significant differences between the groups are indicated with a solid line and asterisk with the p-value shown.
3.3. Inferred Pseudo-haplotypes Three of the four possible pseudo-haplotypes constructed from COL11A1 rs3753841 and COL11A1 rs1676486 were inferred at a frequency greater than 5% (Fig. 2). The T-C inferred pseudo-haplotype
3.2. Genotypes The TT genotype of COL11A1 rs3753841 was significantly overrepresented in the CTS group (21.4%) compared to CON group (7.9%) (p = 0.004) (Fig. 1). Similarly the minor T allele of COL1A1 rs1800012 was significantly over-represented in the CTS group (10.4%) compared to CON group (4.8%) (p = 0.036) No significant differences were found in the genotype (p = 0.161) distributions between the CTS and CON groups for COL1A1 rs1800012, but a trend for the T minor allele to be over-represented in the CTS group (p = 0.055) was observed. This effect was also observed when the female participants were analysed separately, with the T allele being significantly overrepresented in the CTS group (10.4%) compared to the CON group (4.8%) (p = 0.036) (Table 2). No significant differences were observed in genotype (p = 0.133, p = 0.363 and p = 0.573 respectively) or allele (p = 0.330, p = 0.216 and p = 0.918 respectively) distributions between the CTS and CON groups (Table 2) for COL11A1 rs1676486, COL11A2 rs1799907 and COL12A1 rs970547. Similar genotype and allele distributions were observed when only females were analysed. Additionally, similar genotype and allele distributions were observed when participants with a history of a medical condition thought to be associated with CTS were excluded (Data not shown). Table 1 General characteristics of the carpal tunnel syndrome (CTS) and control (CON) groups.a, b, c
Age at Recruitment (yrs) Age at Surgery (yrs) Sex (% Female) Height (cm) Weight (kg) BMI (kg.m2) Country of Birth (% SA)
CTS (n = 102)
CON (n = 150)
P-value
45.5 ± 10.6 (102) 42.0 ± 10.7 (90) 91.2 (102) 159.9 ± 7.6 (100) 83.0 ± 18.0 (101) 32.5 ± 6.9 (100) 99.0 (98)c
40.3 ± 9.7 (149) 40.3 ± 9.7 (149)a 88.7 (150) 160.3 ± 7.8 (149) 78.5 ± 19.0 (148) 30.4 ± 6.7 (147) 100.0 (141)
b0.001 0.192 0.674 0.648 0.062 0.134b 0.410
Values are expressed as a mean ± standard deviation or a frequency (%).The maximum number (n) of participants with non-missing data in each group is also indicated. Significant p-values are indicated in bold. Yrs., years; cm, centimetres; kg, kilograms; BMI, body mass index; m, meter; SA, South Africa. a Age at recruitment. b Co-varied for age at recruitment. c One participant born in Namibia.
Table 2 Genotype and allele frequency distributions of COL11A1 rs1676486 (C/T), COL11A2 rs1799907 (T/A) and COL1A1 rs1800012 (G/T) in carpal tunnel syndrome (CTS) and control (CON) groups for all participants (All) as well as female participants (Female). All CTS COL1A1 rs1800012 GG genotype GT genotype TT genotype Genotype p-value T minor allele Allele p-value HWE COL11A1 rs1676486 CC genotype CT genotype TT genotype Genotype p-value T minor allele Allele p-value HWE COL11A2 rs1799907 AA genotype AT genotype TT genotype Genotype p-value T minor allele Allele p-value HWE COL12A1 rs970547 AA genotype AG genotype GG genotype Genotype p-value G minor allele Allele p-value HWE
n = 99 81.8 (81) 15.2 (15) 3.0 (3) 10.6 (21) 0.070 n = 98 63.3 (62) 30.6 (30) 6.1 (6) 21.4 (42) 0.370 n = 94 46.8 (44) 40.4 (38) 12.8 (12) 33.0 (62) 0.486 n = 92 51.1 (47) 35.9 (33) 13.0 (12) 31.0 (57) 0.142
Female CON n = 142 89.4 (127) 9.9 (14) 0.7 (1) 0.161 5.6 (16) 0.055 0.362 n = 146 52.7 (77) 43.2 (63) 4.1 (6) 0.133 25.7 (75) 0.330 0.139 n = 136 52.2 (71) 40.4 (55) 7.4 (10) 0.363 27.6 (75) 0.216 1.000 n = 134 47.0 (63) 42.5 (57) 10.5 (14) 0.573 31.7 (85) 0.918 0.842
CTS n = 91 82.4 (75) 14.3 (13) 3.3 (3) 10.4 (19) 0.049 n = 89 61.8 (55) 31.5 (28) 6.7 (6) 22.5 (40) 0.362 n = 85 48.2 (41) 38.8 (33) 12.9 (11) 32.4 (55) 0.322 n = 84 51.2 (43) 34.5 (29) 14.3 (12) 31.5 (53) 0.078
CON n = 125 91.2 (114) 8.0 (10) 0.8 (1) 0.123 4.8 (12) 0.036 0.242 n = 129 54.3 (70) 41.1 (53) 4.7 (6) 0.324 25.2 (65) 0.569 0.365 n = 119 53.8 (64) 40.3 (48) 5.9 (7) 0.210 26.1 (62) 0.183 0.812 n = 120 47.5 (57) 42.5 (51) 10.0 (12) 0.425 31.3 (75) 1.000 1.000
Genotype and allele frequencies are expressed as percentage with the participant number (n) in parentheses. The maximum number (n) of participants with non-missing data in each group is also indicated. Significant p-values are emphasised in bold font. HWE, Harley-Weinberg Equilibrium.
56
S. Dada et al. / Gene 587 (2016) 53–58
Although a Bonferroni correction, a multiple-comparison correction (Ruigrok et al., 2004), is widely considered to be too conservative for this type of study which follows an a priori approach (Perneger, 1998), the T-C (Fig. 2) and T-C-(AGGG) (Fig. 3) inferred pseudo-haplotypes, remained significant after correction (p b 0.004 for a test to be considered significant at the α = 0.05 level). The inclusion of the COL11A2 rs1799907, COL12A1 rs970547 and previously associated COL5A1 rs13946 variants to the inferred pseudohaplotype analysis did not contribute to the models and are therefore not shown. Similar distributions and associations were found for all inferred pseudo-haplotypes when only female participants were included in the analysis. 4. Discussion
Fig. 2. Inferred pseudo-haplotypes constructed from the COL11A1 rs3753841 and COL11A1 rs1676486 polymorphisms for all participants within the carpal tunnel syndrome (CTS, black bars) and control (CON, white bars) groups. Significant differences are indicated with the p-value.
was significantly over-represented in the CTS (40.1%) group compared to the CON (24.7%) group (p b 0.001). Conversely, the C-C inferred pseudo-haplotype was significantly under-represented in the CTS (38.6%) group compared to the CON (49.6%) group (p = 0.025). Since it has been suggested that type XI collagen interacts with type V collagen during fibrillogenesis in developing tendons(Wenstrup et al., 2011) and gene–gene interactions between the type XI collagen genes and COL5A1 modulate the risk of chronic Achilles tendinopathy(Hay et al., 2013), inferred pseudo-haplotypes were constructed from the COL11A1 rs3753841, COL11A1 rs1676486 variants and the previously investigated COL5A1 rs71746744 (−/AGGG) variant(Burger et al., 2014a). Six of the possible eight pseudo-haplotypes, constructed from COL11A1 rs3753841, COL11A1 rs1676486 and COL5A1 rs71746744, were inferred at a frequency greater than 5% (Fig. 3). The T-C-(−) and T-C-(AGGG) inferred pseudo-haplotypes were significantly overrepresented in the CTS (12.0% and 28%, respectively) compared to the CON (10.5% and 14.2%, respectively) group (p = 0.014 and p b 0.001, respectively). The C-C-(AGGG) inferred pseudo-haplotype was significantly under-represented in the CTS (23.1%) compared to the CON (27.5%) group (p = 0.031).
Previous studies have reported that genetic factors play an important role in the aetiology and susceptibility to CTS (Burger et al., 2014a; Burger et al., 2015; Burger et al., 2014b; Eroğlu et al., 2015; Gragnoli, 2011; Lozano-Calderón et al., 2008). In support of this, associations between variants rs13946 (C/T) and rs12722 (C/T), located within the functional 3′-untranslated region (UTR) of COL5A1, and CTS has previously been reported (Burger et al., 2014a). The aim of this study was to investigate whether variants within other collagen genes, namely COL11A1, COL11A2, COL1A1 and COL12A1, that have previously been associated with other overuse injuries, also modulate the risk of CTS. A main finding of this study was the independent associations of the TT genotype of COL11A1 rs3753841 (T/C) with increased risk of CTS in a South African self-reported Coloured cohort. Although no independent association of the COL11A1 rs1676486 (C/T) variant was observed, the T-C inferred haplotype constructed from rs3753841 and rs1676486 was also associated with increased CTS risk. Interestingly, the T-C-T inferred pseudo-haplotype constructed from both these COL11A1 variants and COL11A2 rs1799907 (T/A) was previously associated with increased risk of chronic Achilles tendinopathy in South African and Australian populations(Hay et al., 2013). The inclusion of the COL11A2 rs1799907 (T/A) variant, which produces distinct isoforms of the α2(XI) chain(Hay et al., 2013) and which was not independently associated with altered risk of CTS, in the inferred pseudo-haplotype analysis did not contribute to the model for modulating CTS risk in this study. The COL11A1 rs3753841 variant, which is a T N C substitution within exon 52, results in a non-synonymous amino acid change of a leucine to
Fig. 3. Inferred pseudo-haplotypes constructed from COL11A1 rs3753841, COL11A1 rs1676486 and COL5A1 rs71746744 polymorphisms for all in the carpal tunnel syndrome (CTS, black bars) and control (CON, white bars) groups.
S. Dada et al. / Gene 587 (2016) 53–58
proline at amino acid position 1323 of the α1(XI) chain whilst the rs1676486 polymorphism within exon 62, results in an amino acid substitution from proline to serine at position 1535 (Mio et al., 2007). As a result, it has been suggested that a combination of alleles from these variants could potentially cause conformational changes in type XI collagen with possible consequences to the structural or functional properties of the arising collagen fibril(Hay et al., 2013). The T allele of COL11A1 rs1676486 has also been reported to be associated with increased mRNA degradation (Mio et al., 2007), suggesting that decreased α1(XI) chain production and by implication type XI collagen synthesis might be involved in the aetiology of CTS(Hay et al., 2013). Since COL11A1-COL5A1 interactions had previously been associated with altered risk of chronic Achilles tendinopanthy(Hay et al., 2013), inferred pseudo-haplotypes were constructed from the COL11A1 rs3753841 and rs1676486 variants investigated in this study and the previously investigated COL5A1 rs71746744 (−/AGGG) insertion variant(Burger et al., 2014a). Similar to previous findings(Hay et al., 2013) the T-C-(AGGG) pseudo-haplotype (rs3753841, rs1676486 and rs71746744) was associated with increased risk of CTS. The α1(XI) and α1(V) chains, encoded by COL11A1 and COL5A1 respectively, share structural and functional similarities resulting in heterotrimer (α1(XI)2α2(V)) formation (Kleman et al., 1992; Brown et al., 1991; Mayne et al., 1993; Niyibizi and Eyre, 1989). We have previously hypothesised that altered mRNA stability is associated with these COL5A1 and COL11A1 polymorphisms, which result in altered types V and XI collagen synthesis(Hay et al., 2013). Since both types V and XI collagen regulate collagen fibril assembly and diameter (fibrillogenesis), these variants could alter the mechanical properties of tendons and other extracellular structures within the carpal tunnel, which in turn could be implicated in the aetiology of CTS. Further studies are required to investigate this hypothesis. The minor T allele of COL1A1 rs1800012 (G/T) was significantly associated with increased risk of CTS in female participants. A trend for the T allele to be associated with increased risk was observed when the male participants were included in the analysis. Because of the lower frequency of the T allele within the population and the small number of male participants included in this study, this results does not necessarily suggest that that rs1800012 is associated with CTS in females only and future work with larger sample sizes is required to investigate this possibility. The substitution of a tyrosine with a guanine nucleotide within the Sp1 binding site of intron 1 of COL1A1 has been proposed to result in an increased binding affinity for the transcription factor Sp1. This results in increased COL1A1 gene expression and the production of a type I collagen homotrimer consisting of three α1(I) chains (Mann et al., 2001; Deak et al., 1985). The inclusion of increased amounts of type I collagen homotrimers in tendons and other connective tissues is believed to alter the tissues properties and susceptibility to injury (Mann et al., 2001) In support of this hypothesis, the TT genotype of this variant has previously been reported to be associated with increased risk of lumbar disc disease(Mann et al., 2001; Tilkeridis et al., 2005) but also to be protective against ACL injury(Posthumus et al., 2009b; Ficek et al., 2013; Khoschnau et al., 2008). A possible explanation for these differences could be the different mechanisms of injury but future research should further investigate these associations. Finally, no independent associations between the non-synonymous COL12A1 rs970547 (A/G) variant and CTS were observed. This variant also did not make a significant contribution to the injury risk model when included in inferred pseudo-haplotypes analysis (data not shown). The COL12A1 rs970547 variant, which was previously associated with increased risk of ACL injury in females(Posthumus et al., 2010), causes a substitution of a serine with a glycine at position 3085 of exon 65 and has been proposed to potentially modulate the function of type XII collagen (September et al., 2008; Kumar et al., 2009). In conclusion, the identification of the independent associations of COL1A1 rs1800012 and COL11A1 rs3753841 with CTS adds to the current knowledge of genetic risk factors in the aetiology of CTS.
57
Furthermore, this study provides evidence suggesting that the genes that encoding the structural and functionally related type XI and type V collagens, interact with one another in part to collectively modulate the risk for CTS. Although the selected variants within COL11A2 and COL12A1 were not independently associated with CTS risk, this does not exclude the possibility that other variants within these genes are associated and potentially involved in the aetiology of CTS. Further research is required to understand the functional mechanisms underlying the complex genetic associations found as well as the nature of the gene–gene interactions in modulating risk of this injury. Conflict of Interest The authors declare no conflict of interest. Acknowledgements The authors would like to thank Dr. Ajmal Ikram, Dr. Adriaan Smit and Dr. David Rodseth, as well as the nurses from the various clinics, for their help in participant identification. This study was supported in part by funds from the National Research Foundation (NRF) of South Africa, University of Cape Town, and the South African Medical Research Council - Grant numbers 78990 and 85374. The grant holder acknowledges that opinions, findings and conclusions or recommendations expressed in any publication generated by the NRF supported research are that of the author(s), and that the NRF accepts no liability whatsoever in this regard. Author MC and A/Prof Alison September have filed patents on the application of specific sequence variations, which include those in this manuscript, related to risk assessment of Achilles tendon and anterior cruciate ligament injuries. References Brown, K.E., Lawrence, R., Sonenshein, G.E., 1991. Concerted modulation of alpha 1(XI) and alpha 2(V) collagen mRNAs in bovine vascular smooth muscle cells. J. Biol. Chem. 266 (34), 23268–23273. Burger, M., De Wet, H., Collins, M., 2014a. The COL5A1 gene is associated with increased risk of carpal tunnel syndrome. Clin. Rheumatol. 34 (4), 767–774. Burger, M.C., De Wet, H., Collins, M., 2014b. The BGN and ACAN genes and carpal tunnel syndrome. Gene 551 (2), 160–166 (Available at: http://www.ncbi.nlm.nih.gov/ pubmed/25173489. Accessed October 16, 2014). Burger, M.C., de Wet, H., Collins, M., 2015. Interleukin and growth factor gene variants and risk of carpal tunnel syndrome. Gene 564 (1), 67–72 (Available at: http://linkinghub. elsevier.com/retrieve/pii/S037811191500339X). Deak, S.B., van der Rest, M., Prockop, D.J., 1985. Altered helical structure of a homotrimer of alpha 1(I)chains synthesized by fibroblasts from a variant of osteogenesis imperfecta. Coll. Relat. Res. 5 (4), 305–313. Epstein, M.P., Satten, G.A., 2003. Inference on haplotype effects in case–control studies using unphased genotype data. Am. J. Hum. Genet. 73 (6), 1316–1329 (Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1180397&tool= pmcentrez&rendertype=abstract). Erduran, M., Altinisik, J., Meric, G., Ates, O., Ulusal, A.E., Akseki, D., 2014. Is Sp1 binding site polymorphism within COL1A1 gene associated with tennis elbowα. Gene 537 (2), 308–311 (Available at: http://dx.doi.org/10.1016/j.gene.2013.12.014.). Eroğlu, P., Erkol İnal, E., Sağ, Ş.Ö., Görükmez, Ö., Topak, A., Yakut, T., 2015. Associations analysis of GSTM1, T1 and P1 Ile105Val polymorphisms with carpal tunnel syndrome. Clin. Rheumatol. 1 (Available at: http://link.springer.com/10.1007/s10067-014-2855-0). Ficek, K., Cieszczyk, P., Kaczmarczyk, M., et al., 2013. Gene variants within the COL1A1 gene are associated with reduced anterior cruciate ligament injury in professional soccer players. J. Sci. Med. Sport 16 (5), 396–400 (Available at: http://dx.doi.org/10. 1016/j.jsams.2012.10.004). Gragnoli, C., 2011. Proteasome modulator 9 and carpal tunnel syndrome. Diabetes Res. Clin. Pract. 94 (2), e47–e49 (Available at: http://www.ncbi.nlm.nih.gov/pubmed/ 21862167. Accessed February 7, 2013). Hay, M., Patricios, J., Collins, R., et al., 2013. Association of type XI collagen genes with chronic Achilles tendinopathy in independent populations from South Africa and Australia. Br. J. Sports Med. 47 (9), 569–574 (Available at: http://www.ncbi.nlm. nih.gov/pubmed/23624467. Accessed May 17, 2013). Khoschnau, S., Melhus, H., Jacobson, A., Rahme, H., Bengtsson, H., Ribom, E., 2008. Type I collagen α 1 Sp1 polymorphism and the risk of cruciate ligament ruptures of shoulder dislocations. Am. J. Sports Med. 36 (12), 2432–2436. Kleman, J.P., Hartmann, D.J., Ramirez, F., Van Der Rest, M., 1992. The human rhabdomyosarcoma cell line A204 lays down a highly insoluble matrix composed mainly of ??1 type-XI and ??2 type-V collagen chains. Eur. J. Biochem. 210 (1), 329–335.
58
S. Dada et al. / Gene 587 (2016) 53–58
Kumar, P., Henikoff, S., Ng, P.C., 2009. Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm. Nat. Protoc. 4 (8), 1073–1082 (Available at: http://www.ncbi.nlm.nih.gov/pubmed/19561590). Little, J., Higgins, J.P.T., Ioannidis, J.P.A., et al., 2009. STrengthening the REporting of Genetic Association studies (STREGA) - an extension of the STROBE statement. Eur. J. Clin. Invest. 39 (4), 247–266 (Available at: http://doi.wiley.com/10.1111/j.1365-2362. 2009.02125.x. Accessed September 9, 2013). Lluch, A., 1992. Thickening of the synovium of the digital flexor tendons: cause or consequence of the carpal tunnel syndrome? J. Hand Surg. (Br.) 17 (2), 209–212. Lozano-Calderón, S., Anthony, S., Ring, D., 2008. The quality and strength of evidence for etiology: example of carpal tunnel syndrome. J. Hand. Surg. [Am.] 33 (4), 525–538 (Available at: http://www.ncbi.nlm.nih.gov/pubmed/18406957. Accessed November 30, 2011). Mann, V., Hobson, E.E., Li, B., et al., 2001. A COL1A1 Sp1 binding site polymorphism predisposes to osteoporotic fracture by affecting bone density and quality. J. Clin. Invest. 107 (7), 899–907. Mayne, R., Brewton, R.G., Mayne, P.M., Baker, J.R., 1993. Isolation and characterization of the chains of type V/type XI collagen present in bovine vitreous. J. Biol. Chem. 268 (13), 9381–9386. Mio, F., Chiba, K., Hirose, Y., et al., 2007. A functional polymorphism in COL11A1, which encodes the alpha 1 chain of type XI collagen, is associated with susceptibility to lumbar disc herniation. Am. J. Hum. Genet. 81 (December), 1271–1277. Niyibizi, C., Eyre, D.R., 1989. Identification of the cartilage alpha 1(XI) chain in type V collagen from bovine bone. FEBS Lett. 242 (2), 314–318. Perneger, T.V., 1998. What's wrong with Bonferroni adjustments. BMJ 316, 1236–1238 (Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid= 1112988&tool=pmcentrez&rendertype=abstract). Posthumus, M., September, A.V., Schwellnus, M.P., Collins, M., 2009a. Investigation of the Sp1-binding site polymorphism within the COL1A1 gene in participants with Achilles tendon injuries and controls. J. Sci. Med. Sport 12 (1), 184–189 (Available at: http:// www.ncbi.nlm.nih.gov/pubmed/18353721. Accessed June 26, 2011). Posthumus, M., September, A.V., Keegan, M., et al., 2009b. Genetic risk factors for anterior cruciate ligament ruptures: COL1A1 gene variant. Br. J. Sports Med. 43 (5), 352–356 (Available at: http://www.ncbi.nlm.nih.gov/pubmed/19193663. Accessed June 10, 2011).
Posthumus, M., September, A.V., O'Cuinneagain, D., Merwe van der, W., Schwellnus, M.P., 2009c. The COL5A1 Gene is associated with increased risk of anterior cruciate ligament ruptures in female participants. Am. J. Sports Med. 37 (11), 2234–2240. Posthumus, M., September, A.V., O'Cuinneagain, D., van der Merwe, W., Schwellnus, M.P., Collins, M., 2010. The association between the COL12A1 gene and anterior cruciate ligament ruptures. Br. J. Sports Med. 44 (16), 1160–1165 (Available at: http:// www.ncbi.nlm.nih.gov/pubmed/19443461. Accessed August 23, 2011). Ruigrok, Y.M., Seitz, U., Wolterink, S., Rinkel, G.J.E., Wijmenga, C., Urban, Z., 2004. Association of Polymorphisms and Haplotypes in the elastin Gene in Dutch patients with sporadic aneurysmal subarachnoid hemorrhage. Stroke 35 (9), 2064–2068. SASHG, 2013. Guidelines from the SASHG Committee for publication purposes regarding: Nomenclature for South African populations. ((June):1). September, V., Posthumus, M., van der Merwe, L., Schwellnus, M., Noakes, T.D., Collins, M., 2008. The COL12A1 and COL14A1 genes and Achilles tendon injuries. Int. J. Sports Med. 29 (3), 257–263 (Available at: (http://www.ncbi.nlm.nih.gov/pubmed/ 17960519. Accessed September 8, 2011)). September, A.V., Cook, J., Handley, C.J., Merwe van der, L., Schwellnus, M.P., Collins, M., 2009. Variants within the COL5A1 gene are associated with Achilles tendinopathy in two populations. Br. J. Sports Med. 43, 357–365. Shafer-Crane, G.A., Meyer, R.A., Schlinger, M.C., Bennett, D.L., Robinson, K.K., Rechtien, J.J., 2005. Effect of occupational keyboard typing on magnetic resonance imaging of the median nerve in subjects with and without symptoms of carpal tunnel syndrome. Am. J. Phys. Med. Rehabil. 84 (4), 258–266 (Available at: http://content.wkhealth. com/linkback/openurl?sid=WKPTLP:landingpage&an=00002060-20050400000004. Accessed February 7, 2013). Tilkeridis, C., Bei, T., Garantziotis, S., Stratakis, C.a., 2005. Association of a COL1A1 polymorphism with lumbar disc disease in young military recruits. J. Med. Genet. 42 (7), e44. Wenstrup, R.J., Smith, S.M., Florer, J.B., et al., 2011. Regulation of collagen fibril nucleation and initial fibril assembly involves coordinate interactions with collagens V and XI in developing tendon. J. Biol. Chem. 286, 20455–20465. Young, B.B., Zhang, G., Koch, M., Birk, D.E., 2002. The roles of types XII and XIV collagen in fibrillogenesis and matrix assembly in the developing cornea. J. Cell. Biochem. 87 (2), 208–220.
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Research paper
Carpal tunnel syndrome: The role of collagen gene variants Suhail Dada a, Marilize C. Burger a, Franka Massij a, Hanli de Wet b, Malcolm Collins a,⁎ a b
Division of Exercise Science and Sports Medicine, Department of Human Biology, University of Cape Town, South Africa Life Occupational Health, Life Healthcare, Cape Town, South Africa
a r t i c l e
i n f o
Article history: Received 7 March 2016 Accepted 13 April 2016 Available online 14 April 2016 Keywords: CTS Entrapment neuropathy Genetics Collagen Wrist Injury Median nerve
a b s t r a c t Introduction: The direct causes of idiopathic carpal tunnel syndrome (CTS) are still unknown. It is suggested that pathology of the tendons and other connective tissue structures within the carpal tunnel may play a role in its aetiology. Variants in genes encoding connective tissue proteins, such as type V collagen, have previously been associated with CTS. Since variants within other collagen genes, such as type I, XI and XII collagen, have previously been associated with modulating the risk of musculoskeletal soft tissue injuries, the aim of this study was to determine whether variants within COL1A1, COL11A1, COL11A2 and COL12A1 were associated with CTS. Methods: Self-reported Coloured South African participants, with a history of carpal tunnel release surgery (CTS, n = 103) and matched control (CON, n = 150) participants without any reported history of CTS symptoms were genotyped for COL1A1 rs1800012 (G/T), COL11A1 rs3753841 (T/C), COL11A1 rs1676486 (C/T), COL11A2 rs1799907 (T/A) and COL12A1 rs970547 (A/G). Results: The TT genotype of COL11A1 rs3753841 was significantly over-represented in the CTS group (21.4%) compared to CON group (7.9%, p = 0.004). Furthermore, a trend for the T minor allele to be over-represented in the CTS group (p = 0.055) with a significant association when only female participants (p = 0.036) were investigated was observed. Constructed inferred pseudo-haplotypes including a previously investigated COL5A1 variant, rs71746744 (−/AGGG), suggest gene–gene interactions between COL5A1 and COL11A1 modulate the risk of CTS. Discussion: These findings provide further information of the role of the genetic risk factors and the possible role of variations in collagen fibril composition in the aetiology of CTS. Genetic factors could potential be included in models developed to identify indivisuals at risk of CTS. Strategies that target modifiable risk factors to mitigate the effect of non-modifiable risk factors, such as the genetic risk, could be also developed to reduce incidence and morbidity of CTS. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Pathology of the flexor tendons and the surrounding subsynovial connective tissue (SSCT) have been proposed to play a role in the aetiology of carpal tunnel syndrome (CTS)(Lluch, 1992; Shafer-Crane et al., 2005). In support of this, sequence variants within the gene that Abbreviations: ACL, Anterior cruciate ligament; ANOVA, Analysis of variance; BMI, Body mass index; COL1A1, The gene encoding the α1 chain of type I collagen; COL5A1, The gene encoding for the α1 chain of type V collagen; COL11A1, The gene encoding for the α1 chain of type XI collagen; COL11A2, The gene encoding for the α2 chain of type XI collagen; COL12A1, The gene encoding for the α1 chain of type V collagen; CON, Control group; CTS, Carpal tunnel syndrome; FACITs, Fibril Associated Collagens with Interrupted Triple Helices; HWE, Hardy–Weinberg Equilibrium; OA, Osteoarthritis; OR, Odds Ratio; PCR, Polymerase chain reaction; RA, Rheumatoid arthritis; RFLP, Restriction fragment length polymorphism; SNPs, Single Nucleotide Polymorphism(s); SSCT, Subsynovial connective tissue; UTR, Untranslated region; VEGFA, vascular endothelial growth factor A; VEGFA, The gene encoding VEGFA. ⁎ Corresponding author at: Division of Exercise Science and Sports Medicine, PO Box 115, Newlands 7725, South Africa. E-mail address: [email protected] (M. Collins).
http://dx.doi.org/10.1016/j.gene.2016.04.030 0378-1119/© 2016 Elsevier B.V. All rights reserved.
encodes for the α1 chain of type V collagen, a component of the collagen fibril, which is the basic structural unit of tendons, have recently been shown to alter risk of CTS(Burger et al., 2014a). Although the collagen fibril consists predominantly of type I collagen, several other quantitively minor collagens, including types V, XI and XII, playing an important role in regulating the formation and maintaining the structural integrity of the collagen fibril and surrounding matrix. Variants within the genes encoding subunits of types I, V, XI and XII collagen have previously been investigated and/or implicated for their potential role in various musculoskeletal soft tissue injuries(Burger et al., 2014a; Posthumus et al., 2009a; Posthumus et al., 2009b; September et al., 2009; Posthumus et al., 2009c; Hay et al., 2013; September et al., 2008; Posthumus et al., 2010). The most extensively investigated variant within COL1A1, which encodes for the α1 chain of type I collagen, is the functional Sp1 binding site polymorphism (rs1800012, G/T)(Posthumus et al., 2009b; Ficek et al., 2013; Khoschnau et al., 2008; Mann et al., 2001; Tilkeridis et al., 2005). The TT genotype was associated with reduced risk of anterior cruciate ligament (ACL) in several studies(Posthumus et al., 2009b;
54
S. Dada et al. / Gene 587 (2016) 53–58
Ficek et al., 2013; Khoschnau et al., 2008) and shoulder dislocations in a single study(Khoschnau et al., 2008). The TT genotype has however been reported to be associated with increased risk of lumbar disc disease (Mann et al., 2001; Tilkeridis et al., 2005), but not associated with chronic Achilles tendinopathy(Posthumus et al., 2009a) or tennis elbow(Erduran et al., 2014). These findings highlight the possible association of COL1A1 rs1800012 with altered risk of other injuries, such as CTS. It has been proposed that types XI and V collagen interact to regulate fibrillogenesis during tendon development(Wenstrup et al., 2011). This suggests that variants within the genes encoding for type XI collagen may, similar to COL5A1 variants, also modulate the risk of tendon injuries. In support of this, several variants within COL11A1 (rs3753841, T/ C, and rs1676486, C/T) and COL11A2 (rs1799907, T/A) was associated chronic Achilles tendinopathy in South African and Australian cohorts (Hay et al., 2013). In addition, the COL11A1 variants have also been reported to interact with the COL5A1 rs71746744 (− AGGG) variant to modulate the risk of Achilles tendinopathy (Hay et al., 2013). Similar to the types V and XI fibrillar collagen, the type XII nonfibrillar collagen, which belongs to the family of Fibril Associated Collagens with Interrupted Triple Helices (FACITs), is also involved in the regulation of fibrillogenesis (Young et al., 2002). Variants within the COL12A1 gene have been investigated in both Achilles tendinopathy(September et al., 2008) and ACL injuries(Posthumus et al., 2010). The AA genotype of COL12A1 rs975047 (A/G) was reported to be associated with increased risk of ACL injury in females(Posthumus et al., 2010). Therefore the aim of this study was to investigate whether the COL1A1 rs1800012 (G/T), COL11A1 rs3753841 (T/C), COL11A1 rs1676486 (C/T), COL11A2 rs1799907 (T/A) and COL12A1 rs975047 (A/G) variants, previously associated with altered risk of tendon and/ or ligament injuries, were also associated with altered risk of CTS. 2. Methods A case–control genetic association study, following a candidate gene approach, was conducted and reported using recommendations outlined in the genetic association study specific STREGA initiative (Little et al., 2009). 2.1. Participants A total of 103 self-reported Coloured individuals (94 female and 9 male), who had undergone bilateral (n = 50, 50.4%) or unilateral (dominant hand n = 34, 36.6%; non-dominant hand n = 9, 9.7%; right hand of ambidextrous individuals n = 2, 2.1%) carpal tunnel release surgery, (CTS group) were previously recruited (Burger et al., 2014a). Nerve conduction studies were performed in some cases, however since it is not a requirement for the commissioner of worker's compensation in South Africa, the results were not recorded. Furthermore, 150 (133 female and 17 male) apparently healthy, self-reported Coloured individuals with no history of carpal tunnel syndrome-related symptoms (CON group) were also previously recruited (Burger et al., 2014a). The CTS and CON groups had been matched for type of occupation and years of exposure. The majority of participants (CTS n = 29, 28.4%; CON n = 76, 50.7%) were general poultry processing workers or general workers in other industries requiring repetitive movement of the upper limbs. The other major self-reported occupations included administration (CTS n = 18, 17.6%; CON n = 18, 12.0%) and nursing (CTS n = 11, 10.8%; CON n = 12, 8.0%). The remaining participants were from several occupations where a high percentage of the work day required use of the wrists(Burger et al., 2014a). South African populations who self-identify as Coloured have complex genetic inheritance, with racial intermixing over approximately 350 years. This ethnic group within the Western Cape region of South Africa is ancestrally derived from admixtures of immigrants from
Western Europe, slave labourers from West Africa, Indonesia, Madagascar, Java, India and Malaysia and one or more of the indigenous African populations (Khoe- and San-speaking or Bantu-speaking). The term “Coloured” in South Africa is therefore a name that encompasses a wide range of people who are unique to this country (SASHG, 2013). All participants provided written informed consent according to the Helsinki Declaration. Personal and family medical history was ascertained by means of a questionnaire (Burger et al., 2014a; Burger et al., 2015; Burger et al., 2014b). This study has been approved by the Human Research Ethics Committee of the Faculty of Health Sciences within the University of Cape Town (HREC 158/2011). 2.2. Genotyping Blood collection and DNA extraction had been conducted, as previously described, at the Division of Exercise Science & Sports Medicine, University of Cape Town, South Africa (Burger et al., 2014b). All DNA samples were genotyped for the COL1A1 rs1800012 (G/T, Assay ID: C_7477170_30), COL11A1 rs3753841 (T/C, Assay ID: C_2947954_10), COL11A1 rs1676486 (C/T, Assay ID: C_8400671_10) and COL11A2 rs1799907 (T/A, Assay ID: C_25474257_10) gene variants by means of a custom-designed fluorescence-based Taqman® PCR assays (Applied Biosystems, Foster City, California, USA). Gene-specific primers and allele-specific probes were used in conjunction with a pre-made PCR master mix containing AmpliTaq Gold DNA polymerase (Applied Biosystems, Foster City, CA, USA) in a final reaction volume of 8 μl. PCR reactions were carried out in a StepOnePlus Real-Time PCR system following the manufacturers recommended cycling conditions (Applied Biosystems, Foster City, California, USA). COL12A1 rs970547 (A/G) was genotyped by means of a restriction fragment length polymorphism (RFLP) assay as previously described (September et al., 2008). Positive controls DNA-free controls were randomly included on each PCR plate and a subset of samples was genotyped twice for quality control purposes. Genotyping calls were made by two different researchers with no discrepancies observed. Furthermore, samples that failed to genotype twice were excluded. For quality control, subsets of known genotype controls were added to each plate. A total of 87.3% (n = 89) and 96.1% (n = 98) of the cases and 94.6% (n = 139) and 99.3% (n = 146) of the controls were successfully genotyped for COL11A1 rs3753841 and COL11A1 rs1676486, respectively. Ninety-two percent (n = 94) of the cases and 99.3% (n = 146) of the controls were successfully genotyped for COL11A2 rs1799907. For COL1A1 rs1800012, 97.1% (n = 99) of the cases and 96.6% (n = 142) of the controls were successfully genotyped whereas 89.3% (n = 92) of the cases and the controls (n = 134), respectively, were successfully genotyped for COL12A1 rs970547. 2.3. Statistical analysis Quanto (V.1.2.4, USC, CA, USA) was used to determine the statistical power for a given sample size, based on the expected minor allele frequencies for the individual variants. Non-missing data was analysed using STATISTICA (version 11, StatSoft Inc., Tulsa, Oklahoma, USA) and Graphpad Prism (version 5, GraphPad Software, San Diego, CA, USA). A Pearson's chi-squared test or Fisher's exact test were used to determine significant differences in the genotype and allele distributions as well as other categorical data, namely sex and country of birth, between the CTS and CON groups. An analysis of variance (ANOVA) was used to determine significant differences in continuous data. A significant statistical difference was set at p b 0.05. The Hardy–Weinberg equilibrium (HWE) of the groups was established using the program Genepop web version 4.0.10 (http://genepop.curtin.edu.au/). Inferred haplotypes and pseudo-haplotypes were constructed using Chaplin version 1.2.2 (Emory University School of Medicine, Atlanta, Georgia, USA) and
S. Dada et al. / Gene 587 (2016) 53–58
55
HAPSTAT (version 3.0, University of North Carolina at Chapel Hill, NC, USA) (Epstein and Satten, 2003). 3. Results 3.1. General Characteristics The general characteristics of all participants have previously been described in detail (Burger et al., 2014a). In summary, there was no significant difference between the CTS and CON groups, of the participants successfully genotyped for at least one of the investigated variants, for age at surgery (for the CON group, age of recruitment was used), sex, height, weight and country of birth (Table 1). The CTS and CON groups were also similarly matched for BMI after adjusting for the significant difference in age of recruitment (Table 1). Participants within the CTS (27.5%, n = 28) and CON (14.9%, n = 22) groups self-reported similar medical histories suggested to be associated with CTS with no significant difference (p = 0.178), including: diabetes (9.8% CTS, n = 10 vs 7.4% CON, n = 11), osteoarthritis (OA, 7.8% CTS, n = 8 vs 5.5% CON, n = 8), rheumatoid arthritis (RA, 3.9% CTS, n = 4 vs 0.7% CON, n = 1), thyroid disorders (2.0% CTS, n = 2 vs 0.7% CON, n = 1), diabetes and OA (1.0% CTS, n = 1 vs 0.0% CON), RA and OA (0.0% CTS vs 0.7% CON, n = 1), diabetes and RA (1.0% CTS, n = 1 vs 0.0% CON), diabetes and thyroid disorder (1.0% CTS, n = 1 vs 0.0% CON), RA and systemic lupus erythematosus (1.0% CTS, n = 1 vs 0.0% CON).
Fig. 1. Genotype frequency distribution of COL11A1 rs3753841 for all participants in the carpal tunnel syndrome (CTS, black bars) and control (CON, white bars) groups. The number of participants in each group is indicated. Significant differences between the groups are indicated with a solid line and asterisk with the p-value shown.
3.3. Inferred Pseudo-haplotypes Three of the four possible pseudo-haplotypes constructed from COL11A1 rs3753841 and COL11A1 rs1676486 were inferred at a frequency greater than 5% (Fig. 2). The T-C inferred pseudo-haplotype
3.2. Genotypes The TT genotype of COL11A1 rs3753841 was significantly overrepresented in the CTS group (21.4%) compared to CON group (7.9%) (p = 0.004) (Fig. 1). Similarly the minor T allele of COL1A1 rs1800012 was significantly over-represented in the CTS group (10.4%) compared to CON group (4.8%) (p = 0.036) No significant differences were found in the genotype (p = 0.161) distributions between the CTS and CON groups for COL1A1 rs1800012, but a trend for the T minor allele to be over-represented in the CTS group (p = 0.055) was observed. This effect was also observed when the female participants were analysed separately, with the T allele being significantly overrepresented in the CTS group (10.4%) compared to the CON group (4.8%) (p = 0.036) (Table 2). No significant differences were observed in genotype (p = 0.133, p = 0.363 and p = 0.573 respectively) or allele (p = 0.330, p = 0.216 and p = 0.918 respectively) distributions between the CTS and CON groups (Table 2) for COL11A1 rs1676486, COL11A2 rs1799907 and COL12A1 rs970547. Similar genotype and allele distributions were observed when only females were analysed. Additionally, similar genotype and allele distributions were observed when participants with a history of a medical condition thought to be associated with CTS were excluded (Data not shown). Table 1 General characteristics of the carpal tunnel syndrome (CTS) and control (CON) groups.a, b, c
Age at Recruitment (yrs) Age at Surgery (yrs) Sex (% Female) Height (cm) Weight (kg) BMI (kg.m2) Country of Birth (% SA)
CTS (n = 102)
CON (n = 150)
P-value
45.5 ± 10.6 (102) 42.0 ± 10.7 (90) 91.2 (102) 159.9 ± 7.6 (100) 83.0 ± 18.0 (101) 32.5 ± 6.9 (100) 99.0 (98)c
40.3 ± 9.7 (149) 40.3 ± 9.7 (149)a 88.7 (150) 160.3 ± 7.8 (149) 78.5 ± 19.0 (148) 30.4 ± 6.7 (147) 100.0 (141)
b0.001 0.192 0.674 0.648 0.062 0.134b 0.410
Values are expressed as a mean ± standard deviation or a frequency (%).The maximum number (n) of participants with non-missing data in each group is also indicated. Significant p-values are indicated in bold. Yrs., years; cm, centimetres; kg, kilograms; BMI, body mass index; m, meter; SA, South Africa. a Age at recruitment. b Co-varied for age at recruitment. c One participant born in Namibia.
Table 2 Genotype and allele frequency distributions of COL11A1 rs1676486 (C/T), COL11A2 rs1799907 (T/A) and COL1A1 rs1800012 (G/T) in carpal tunnel syndrome (CTS) and control (CON) groups for all participants (All) as well as female participants (Female). All CTS COL1A1 rs1800012 GG genotype GT genotype TT genotype Genotype p-value T minor allele Allele p-value HWE COL11A1 rs1676486 CC genotype CT genotype TT genotype Genotype p-value T minor allele Allele p-value HWE COL11A2 rs1799907 AA genotype AT genotype TT genotype Genotype p-value T minor allele Allele p-value HWE COL12A1 rs970547 AA genotype AG genotype GG genotype Genotype p-value G minor allele Allele p-value HWE
n = 99 81.8 (81) 15.2 (15) 3.0 (3) 10.6 (21) 0.070 n = 98 63.3 (62) 30.6 (30) 6.1 (6) 21.4 (42) 0.370 n = 94 46.8 (44) 40.4 (38) 12.8 (12) 33.0 (62) 0.486 n = 92 51.1 (47) 35.9 (33) 13.0 (12) 31.0 (57) 0.142
Female CON n = 142 89.4 (127) 9.9 (14) 0.7 (1) 0.161 5.6 (16) 0.055 0.362 n = 146 52.7 (77) 43.2 (63) 4.1 (6) 0.133 25.7 (75) 0.330 0.139 n = 136 52.2 (71) 40.4 (55) 7.4 (10) 0.363 27.6 (75) 0.216 1.000 n = 134 47.0 (63) 42.5 (57) 10.5 (14) 0.573 31.7 (85) 0.918 0.842
CTS n = 91 82.4 (75) 14.3 (13) 3.3 (3) 10.4 (19) 0.049 n = 89 61.8 (55) 31.5 (28) 6.7 (6) 22.5 (40) 0.362 n = 85 48.2 (41) 38.8 (33) 12.9 (11) 32.4 (55) 0.322 n = 84 51.2 (43) 34.5 (29) 14.3 (12) 31.5 (53) 0.078
CON n = 125 91.2 (114) 8.0 (10) 0.8 (1) 0.123 4.8 (12) 0.036 0.242 n = 129 54.3 (70) 41.1 (53) 4.7 (6) 0.324 25.2 (65) 0.569 0.365 n = 119 53.8 (64) 40.3 (48) 5.9 (7) 0.210 26.1 (62) 0.183 0.812 n = 120 47.5 (57) 42.5 (51) 10.0 (12) 0.425 31.3 (75) 1.000 1.000
Genotype and allele frequencies are expressed as percentage with the participant number (n) in parentheses. The maximum number (n) of participants with non-missing data in each group is also indicated. Significant p-values are emphasised in bold font. HWE, Harley-Weinberg Equilibrium.
56
S. Dada et al. / Gene 587 (2016) 53–58
Although a Bonferroni correction, a multiple-comparison correction (Ruigrok et al., 2004), is widely considered to be too conservative for this type of study which follows an a priori approach (Perneger, 1998), the T-C (Fig. 2) and T-C-(AGGG) (Fig. 3) inferred pseudo-haplotypes, remained significant after correction (p b 0.004 for a test to be considered significant at the α = 0.05 level). The inclusion of the COL11A2 rs1799907, COL12A1 rs970547 and previously associated COL5A1 rs13946 variants to the inferred pseudohaplotype analysis did not contribute to the models and are therefore not shown. Similar distributions and associations were found for all inferred pseudo-haplotypes when only female participants were included in the analysis. 4. Discussion
Fig. 2. Inferred pseudo-haplotypes constructed from the COL11A1 rs3753841 and COL11A1 rs1676486 polymorphisms for all participants within the carpal tunnel syndrome (CTS, black bars) and control (CON, white bars) groups. Significant differences are indicated with the p-value.
was significantly over-represented in the CTS (40.1%) group compared to the CON (24.7%) group (p b 0.001). Conversely, the C-C inferred pseudo-haplotype was significantly under-represented in the CTS (38.6%) group compared to the CON (49.6%) group (p = 0.025). Since it has been suggested that type XI collagen interacts with type V collagen during fibrillogenesis in developing tendons(Wenstrup et al., 2011) and gene–gene interactions between the type XI collagen genes and COL5A1 modulate the risk of chronic Achilles tendinopathy(Hay et al., 2013), inferred pseudo-haplotypes were constructed from the COL11A1 rs3753841, COL11A1 rs1676486 variants and the previously investigated COL5A1 rs71746744 (−/AGGG) variant(Burger et al., 2014a). Six of the possible eight pseudo-haplotypes, constructed from COL11A1 rs3753841, COL11A1 rs1676486 and COL5A1 rs71746744, were inferred at a frequency greater than 5% (Fig. 3). The T-C-(−) and T-C-(AGGG) inferred pseudo-haplotypes were significantly overrepresented in the CTS (12.0% and 28%, respectively) compared to the CON (10.5% and 14.2%, respectively) group (p = 0.014 and p b 0.001, respectively). The C-C-(AGGG) inferred pseudo-haplotype was significantly under-represented in the CTS (23.1%) compared to the CON (27.5%) group (p = 0.031).
Previous studies have reported that genetic factors play an important role in the aetiology and susceptibility to CTS (Burger et al., 2014a; Burger et al., 2015; Burger et al., 2014b; Eroğlu et al., 2015; Gragnoli, 2011; Lozano-Calderón et al., 2008). In support of this, associations between variants rs13946 (C/T) and rs12722 (C/T), located within the functional 3′-untranslated region (UTR) of COL5A1, and CTS has previously been reported (Burger et al., 2014a). The aim of this study was to investigate whether variants within other collagen genes, namely COL11A1, COL11A2, COL1A1 and COL12A1, that have previously been associated with other overuse injuries, also modulate the risk of CTS. A main finding of this study was the independent associations of the TT genotype of COL11A1 rs3753841 (T/C) with increased risk of CTS in a South African self-reported Coloured cohort. Although no independent association of the COL11A1 rs1676486 (C/T) variant was observed, the T-C inferred haplotype constructed from rs3753841 and rs1676486 was also associated with increased CTS risk. Interestingly, the T-C-T inferred pseudo-haplotype constructed from both these COL11A1 variants and COL11A2 rs1799907 (T/A) was previously associated with increased risk of chronic Achilles tendinopathy in South African and Australian populations(Hay et al., 2013). The inclusion of the COL11A2 rs1799907 (T/A) variant, which produces distinct isoforms of the α2(XI) chain(Hay et al., 2013) and which was not independently associated with altered risk of CTS, in the inferred pseudo-haplotype analysis did not contribute to the model for modulating CTS risk in this study. The COL11A1 rs3753841 variant, which is a T N C substitution within exon 52, results in a non-synonymous amino acid change of a leucine to
Fig. 3. Inferred pseudo-haplotypes constructed from COL11A1 rs3753841, COL11A1 rs1676486 and COL5A1 rs71746744 polymorphisms for all in the carpal tunnel syndrome (CTS, black bars) and control (CON, white bars) groups.
S. Dada et al. / Gene 587 (2016) 53–58
proline at amino acid position 1323 of the α1(XI) chain whilst the rs1676486 polymorphism within exon 62, results in an amino acid substitution from proline to serine at position 1535 (Mio et al., 2007). As a result, it has been suggested that a combination of alleles from these variants could potentially cause conformational changes in type XI collagen with possible consequences to the structural or functional properties of the arising collagen fibril(Hay et al., 2013). The T allele of COL11A1 rs1676486 has also been reported to be associated with increased mRNA degradation (Mio et al., 2007), suggesting that decreased α1(XI) chain production and by implication type XI collagen synthesis might be involved in the aetiology of CTS(Hay et al., 2013). Since COL11A1-COL5A1 interactions had previously been associated with altered risk of chronic Achilles tendinopanthy(Hay et al., 2013), inferred pseudo-haplotypes were constructed from the COL11A1 rs3753841 and rs1676486 variants investigated in this study and the previously investigated COL5A1 rs71746744 (−/AGGG) insertion variant(Burger et al., 2014a). Similar to previous findings(Hay et al., 2013) the T-C-(AGGG) pseudo-haplotype (rs3753841, rs1676486 and rs71746744) was associated with increased risk of CTS. The α1(XI) and α1(V) chains, encoded by COL11A1 and COL5A1 respectively, share structural and functional similarities resulting in heterotrimer (α1(XI)2α2(V)) formation (Kleman et al., 1992; Brown et al., 1991; Mayne et al., 1993; Niyibizi and Eyre, 1989). We have previously hypothesised that altered mRNA stability is associated with these COL5A1 and COL11A1 polymorphisms, which result in altered types V and XI collagen synthesis(Hay et al., 2013). Since both types V and XI collagen regulate collagen fibril assembly and diameter (fibrillogenesis), these variants could alter the mechanical properties of tendons and other extracellular structures within the carpal tunnel, which in turn could be implicated in the aetiology of CTS. Further studies are required to investigate this hypothesis. The minor T allele of COL1A1 rs1800012 (G/T) was significantly associated with increased risk of CTS in female participants. A trend for the T allele to be associated with increased risk was observed when the male participants were included in the analysis. Because of the lower frequency of the T allele within the population and the small number of male participants included in this study, this results does not necessarily suggest that that rs1800012 is associated with CTS in females only and future work with larger sample sizes is required to investigate this possibility. The substitution of a tyrosine with a guanine nucleotide within the Sp1 binding site of intron 1 of COL1A1 has been proposed to result in an increased binding affinity for the transcription factor Sp1. This results in increased COL1A1 gene expression and the production of a type I collagen homotrimer consisting of three α1(I) chains (Mann et al., 2001; Deak et al., 1985). The inclusion of increased amounts of type I collagen homotrimers in tendons and other connective tissues is believed to alter the tissues properties and susceptibility to injury (Mann et al., 2001) In support of this hypothesis, the TT genotype of this variant has previously been reported to be associated with increased risk of lumbar disc disease(Mann et al., 2001; Tilkeridis et al., 2005) but also to be protective against ACL injury(Posthumus et al., 2009b; Ficek et al., 2013; Khoschnau et al., 2008). A possible explanation for these differences could be the different mechanisms of injury but future research should further investigate these associations. Finally, no independent associations between the non-synonymous COL12A1 rs970547 (A/G) variant and CTS were observed. This variant also did not make a significant contribution to the injury risk model when included in inferred pseudo-haplotypes analysis (data not shown). The COL12A1 rs970547 variant, which was previously associated with increased risk of ACL injury in females(Posthumus et al., 2010), causes a substitution of a serine with a glycine at position 3085 of exon 65 and has been proposed to potentially modulate the function of type XII collagen (September et al., 2008; Kumar et al., 2009). In conclusion, the identification of the independent associations of COL1A1 rs1800012 and COL11A1 rs3753841 with CTS adds to the current knowledge of genetic risk factors in the aetiology of CTS.
57
Furthermore, this study provides evidence suggesting that the genes that encoding the structural and functionally related type XI and type V collagens, interact with one another in part to collectively modulate the risk for CTS. Although the selected variants within COL11A2 and COL12A1 were not independently associated with CTS risk, this does not exclude the possibility that other variants within these genes are associated and potentially involved in the aetiology of CTS. Further research is required to understand the functional mechanisms underlying the complex genetic associations found as well as the nature of the gene–gene interactions in modulating risk of this injury. Conflict of Interest The authors declare no conflict of interest. Acknowledgements The authors would like to thank Dr. Ajmal Ikram, Dr. Adriaan Smit and Dr. David Rodseth, as well as the nurses from the various clinics, for their help in participant identification. This study was supported in part by funds from the National Research Foundation (NRF) of South Africa, University of Cape Town, and the South African Medical Research Council - Grant numbers 78990 and 85374. The grant holder acknowledges that opinions, findings and conclusions or recommendations expressed in any publication generated by the NRF supported research are that of the author(s), and that the NRF accepts no liability whatsoever in this regard. Author MC and A/Prof Alison September have filed patents on the application of specific sequence variations, which include those in this manuscript, related to risk assessment of Achilles tendon and anterior cruciate ligament injuries. References Brown, K.E., Lawrence, R., Sonenshein, G.E., 1991. Concerted modulation of alpha 1(XI) and alpha 2(V) collagen mRNAs in bovine vascular smooth muscle cells. J. Biol. Chem. 266 (34), 23268–23273. Burger, M., De Wet, H., Collins, M., 2014a. The COL5A1 gene is associated with increased risk of carpal tunnel syndrome. Clin. Rheumatol. 34 (4), 767–774. Burger, M.C., De Wet, H., Collins, M., 2014b. The BGN and ACAN genes and carpal tunnel syndrome. Gene 551 (2), 160–166 (Available at: http://www.ncbi.nlm.nih.gov/ pubmed/25173489. Accessed October 16, 2014). Burger, M.C., de Wet, H., Collins, M., 2015. Interleukin and growth factor gene variants and risk of carpal tunnel syndrome. Gene 564 (1), 67–72 (Available at: http://linkinghub. elsevier.com/retrieve/pii/S037811191500339X). Deak, S.B., van der Rest, M., Prockop, D.J., 1985. Altered helical structure of a homotrimer of alpha 1(I)chains synthesized by fibroblasts from a variant of osteogenesis imperfecta. Coll. Relat. Res. 5 (4), 305–313. Epstein, M.P., Satten, G.A., 2003. Inference on haplotype effects in case–control studies using unphased genotype data. Am. J. Hum. Genet. 73 (6), 1316–1329 (Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1180397&tool= pmcentrez&rendertype=abstract). Erduran, M., Altinisik, J., Meric, G., Ates, O., Ulusal, A.E., Akseki, D., 2014. Is Sp1 binding site polymorphism within COL1A1 gene associated with tennis elbowα. Gene 537 (2), 308–311 (Available at: http://dx.doi.org/10.1016/j.gene.2013.12.014.). Eroğlu, P., Erkol İnal, E., Sağ, Ş.Ö., Görükmez, Ö., Topak, A., Yakut, T., 2015. Associations analysis of GSTM1, T1 and P1 Ile105Val polymorphisms with carpal tunnel syndrome. Clin. Rheumatol. 1 (Available at: http://link.springer.com/10.1007/s10067-014-2855-0). Ficek, K., Cieszczyk, P., Kaczmarczyk, M., et al., 2013. Gene variants within the COL1A1 gene are associated with reduced anterior cruciate ligament injury in professional soccer players. J. Sci. Med. Sport 16 (5), 396–400 (Available at: http://dx.doi.org/10. 1016/j.jsams.2012.10.004). Gragnoli, C., 2011. Proteasome modulator 9 and carpal tunnel syndrome. Diabetes Res. Clin. Pract. 94 (2), e47–e49 (Available at: http://www.ncbi.nlm.nih.gov/pubmed/ 21862167. Accessed February 7, 2013). Hay, M., Patricios, J., Collins, R., et al., 2013. Association of type XI collagen genes with chronic Achilles tendinopathy in independent populations from South Africa and Australia. Br. J. Sports Med. 47 (9), 569–574 (Available at: http://www.ncbi.nlm. nih.gov/pubmed/23624467. Accessed May 17, 2013). Khoschnau, S., Melhus, H., Jacobson, A., Rahme, H., Bengtsson, H., Ribom, E., 2008. Type I collagen α 1 Sp1 polymorphism and the risk of cruciate ligament ruptures of shoulder dislocations. Am. J. Sports Med. 36 (12), 2432–2436. Kleman, J.P., Hartmann, D.J., Ramirez, F., Van Der Rest, M., 1992. The human rhabdomyosarcoma cell line A204 lays down a highly insoluble matrix composed mainly of ??1 type-XI and ??2 type-V collagen chains. Eur. J. Biochem. 210 (1), 329–335.
58
S. Dada et al. / Gene 587 (2016) 53–58
Kumar, P., Henikoff, S., Ng, P.C., 2009. Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm. Nat. Protoc. 4 (8), 1073–1082 (Available at: http://www.ncbi.nlm.nih.gov/pubmed/19561590). Little, J., Higgins, J.P.T., Ioannidis, J.P.A., et al., 2009. STrengthening the REporting of Genetic Association studies (STREGA) - an extension of the STROBE statement. Eur. J. Clin. Invest. 39 (4), 247–266 (Available at: http://doi.wiley.com/10.1111/j.1365-2362. 2009.02125.x. Accessed September 9, 2013). Lluch, A., 1992. Thickening of the synovium of the digital flexor tendons: cause or consequence of the carpal tunnel syndrome? J. Hand Surg. (Br.) 17 (2), 209–212. Lozano-Calderón, S., Anthony, S., Ring, D., 2008. The quality and strength of evidence for etiology: example of carpal tunnel syndrome. J. Hand. Surg. [Am.] 33 (4), 525–538 (Available at: http://www.ncbi.nlm.nih.gov/pubmed/18406957. Accessed November 30, 2011). Mann, V., Hobson, E.E., Li, B., et al., 2001. A COL1A1 Sp1 binding site polymorphism predisposes to osteoporotic fracture by affecting bone density and quality. J. Clin. Invest. 107 (7), 899–907. Mayne, R., Brewton, R.G., Mayne, P.M., Baker, J.R., 1993. Isolation and characterization of the chains of type V/type XI collagen present in bovine vitreous. J. Biol. Chem. 268 (13), 9381–9386. Mio, F., Chiba, K., Hirose, Y., et al., 2007. A functional polymorphism in COL11A1, which encodes the alpha 1 chain of type XI collagen, is associated with susceptibility to lumbar disc herniation. Am. J. Hum. Genet. 81 (December), 1271–1277. Niyibizi, C., Eyre, D.R., 1989. Identification of the cartilage alpha 1(XI) chain in type V collagen from bovine bone. FEBS Lett. 242 (2), 314–318. Perneger, T.V., 1998. What's wrong with Bonferroni adjustments. BMJ 316, 1236–1238 (Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid= 1112988&tool=pmcentrez&rendertype=abstract). Posthumus, M., September, A.V., Schwellnus, M.P., Collins, M., 2009a. Investigation of the Sp1-binding site polymorphism within the COL1A1 gene in participants with Achilles tendon injuries and controls. J. Sci. Med. Sport 12 (1), 184–189 (Available at: http:// www.ncbi.nlm.nih.gov/pubmed/18353721. Accessed June 26, 2011). Posthumus, M., September, A.V., Keegan, M., et al., 2009b. Genetic risk factors for anterior cruciate ligament ruptures: COL1A1 gene variant. Br. J. Sports Med. 43 (5), 352–356 (Available at: http://www.ncbi.nlm.nih.gov/pubmed/19193663. Accessed June 10, 2011).
Posthumus, M., September, A.V., O'Cuinneagain, D., Merwe van der, W., Schwellnus, M.P., 2009c. The COL5A1 Gene is associated with increased risk of anterior cruciate ligament ruptures in female participants. Am. J. Sports Med. 37 (11), 2234–2240. Posthumus, M., September, A.V., O'Cuinneagain, D., van der Merwe, W., Schwellnus, M.P., Collins, M., 2010. The association between the COL12A1 gene and anterior cruciate ligament ruptures. Br. J. Sports Med. 44 (16), 1160–1165 (Available at: http:// www.ncbi.nlm.nih.gov/pubmed/19443461. Accessed August 23, 2011). Ruigrok, Y.M., Seitz, U., Wolterink, S., Rinkel, G.J.E., Wijmenga, C., Urban, Z., 2004. Association of Polymorphisms and Haplotypes in the elastin Gene in Dutch patients with sporadic aneurysmal subarachnoid hemorrhage. Stroke 35 (9), 2064–2068. SASHG, 2013. Guidelines from the SASHG Committee for publication purposes regarding: Nomenclature for South African populations. ((June):1). September, V., Posthumus, M., van der Merwe, L., Schwellnus, M., Noakes, T.D., Collins, M., 2008. The COL12A1 and COL14A1 genes and Achilles tendon injuries. Int. J. Sports Med. 29 (3), 257–263 (Available at: (http://www.ncbi.nlm.nih.gov/pubmed/ 17960519. Accessed September 8, 2011)). September, A.V., Cook, J., Handley, C.J., Merwe van der, L., Schwellnus, M.P., Collins, M., 2009. Variants within the COL5A1 gene are associated with Achilles tendinopathy in two populations. Br. J. Sports Med. 43, 357–365. Shafer-Crane, G.A., Meyer, R.A., Schlinger, M.C., Bennett, D.L., Robinson, K.K., Rechtien, J.J., 2005. Effect of occupational keyboard typing on magnetic resonance imaging of the median nerve in subjects with and without symptoms of carpal tunnel syndrome. Am. J. Phys. Med. Rehabil. 84 (4), 258–266 (Available at: http://content.wkhealth. com/linkback/openurl?sid=WKPTLP:landingpage&an=00002060-20050400000004. Accessed February 7, 2013). Tilkeridis, C., Bei, T., Garantziotis, S., Stratakis, C.a., 2005. Association of a COL1A1 polymorphism with lumbar disc disease in young military recruits. J. Med. Genet. 42 (7), e44. Wenstrup, R.J., Smith, S.M., Florer, J.B., et al., 2011. Regulation of collagen fibril nucleation and initial fibril assembly involves coordinate interactions with collagens V and XI in developing tendon. J. Biol. Chem. 286, 20455–20465. Young, B.B., Zhang, G., Koch, M., Birk, D.E., 2002. The roles of types XII and XIV collagen in fibrillogenesis and matrix assembly in the developing cornea. J. Cell. Biochem. 87 (2), 208–220.