Correlations Between the Genetic Variations in the COL1A1, COL5A1, COL12A1, and β-fibrinogen Genes and Anterior Cruciate Ligament Injury in Chinese Patients^a

The study involved 101 patients with ACLI (73 males and 28 females) and 110 healthy control participants without ACLI (69 males and 41 females). The average ages were 33.62 ± 11.18 years of the ACLI patients and 35.57 ± 12.78 years of the control participants. A total of 327 patients with signs and symptoms of knee injury underwent arthroscopic examination in our hospital; of these, 41 ACLI patients with left-sided injuries and 60 with right-sided injuries were selected for this study. Control individuals with normal knees were randomly selected from the general population, and magnetic resonance imaging confirmed they had no organic knee-joint injury such as ACLI, meniscal tear, or fracture. Control recruits were matched by age and sex to the ACLI participants. Demographic information is shown in Table 1. The trial was approved by our hospital and the Clinical Research Ethics Committee of First People's Hospital of Kunming University. All participants signed the consent form. Inclusion criteria were (1) clinical physical examination indicated knee instability; (2) magnetic resonance imaging examination confirmed fiber discontinuity by coronal and axial imaging, the empty notch sign on coronal imaging, thickening and edema of the ACL (increased signal intensity on T2 or intermediate-weighted sequences), the fibers being completely absorbed- or the residual ACL stump adherent to the synovial envelope covering the posterior cruciate ligament; (3) arthroscopic examination of the ACLI; (4) nonmechanical damage. Exclusion criteria were (1) ACLI combined with ligament injury in another part of the knee joint; (2) patients with previous knee disease such as rheumatism or rheumatoid arthritis, septic arthritis, or knee-joint tuberculosis; or (3) patients with previous knee surgery.

A total of 5 mL of peripheral venous blood of all participants was collected and anticoagulated with ethylene diamine tetra-acetic acid (Tianjin Flourish Chemical Co, Ltd, Hebei District, Tian Jin, China). Genomic DNA was extracted from samples using the AxyPrep Blood Genomic DNA Mini Kit (Shanghai Haorui Biotechnology Co, Ltd, Shanghai, China).

The SNPs detected were COL1A1 gene intron rs1800012, COL5A1 gene 3′-UTR rs12722 and rs13946, COL12A1 gene exons rs970547 and rs240736 and the rs1800787, rs1800788, rs1800789, rs1800790, rs1800791, and rs2227389 in the β-fib gene promoter region.

Polymerase chain reaction (PCR) mixtures were prepared according to the instructions supplied with the AmpliTaq Gold kit (Applied Biosystems, Foster City, CA), which contained 200 μM of deoxynucleoside triphosphate, 2.5 μL of 10 × reaction buffer (100 mM Tris-HCl at pH 8.3, 500 mM KCl), 1.5 mM MgCl2, a 0.1 μM concentration of the primers with 2.5 U of Taq DNA polymerase, and 5 ng of template DNA. The volume of this mix was adjusted to 25 μL with sterile water. The PCR reaction conditions were as follows: an initial denaturation at 94°C for 3 minutes, followed by 30 cycles of denaturation at 94°C for 30 seconds, annealing at 59°C for 30 seconds, extension at 72°C for 120 seconds, and a final extension step at 72°C for 10 minutes. The COL5A1 (rs12722, rs13946) and COL12A1 (rs970547, rs240736) amplicons were subsequently digested using BstUI (NEB), DpnII (NEB), AluI (NEB), and BsrI (NEB) for RFLP analysis. Primers used for PCR RFLP in this study are shown in Table 2.

The amplicons of COL1A1 (rs1800012) and COL12A1 (rs970547, rs240736) and the rs1800787, rs1800788, rs1800789, rs1800790, rs1800791, and rs2227389 of the β-fib gene promoter region were sequenced to detect SNPs (Table 2). (Primers were provided by Shanghai Bioengineering Co, Ltd, Shanghai, China.) The haplotypes were generated using the 2 SNPs of rs12722 and rs13946 and 3 SNPs of rs1800790, rs1800791, and rs2227389 in the ACLI patients and control individuals.

The 351 and 316 bp DNA fragments generated by rs12722 BstUI RFLP suggested that they were homozygous for TT, and the 316-, 271-, and 80-bp fragments suggested a CC homozygote. The 418-, 194-, 40-, and 15-bp fragments produced by rs13946 DpnII RFLP were homozygous for TT, and the 612-, 40-, and 15-bp fragments were homozygous for CC. The 460-, 139-, and 16-bp fragments produced by rs970547 AluI RFLP were defined as AA homozygotes, and the 599- and 16-bp fragments were homozygous for GG. The 480-bp fragment produced by rs240736 BsrI RFLP digestion was TT homozygous, and the 322- and 158-bp fragments represented CC homozygotes.

The χ² test was used to analyze the frequency of allele, genotype, and haplotype distributions. Hardy-Weinberg equilibrium testing for genotypes was performed using the online calculator (http://www.oege.org/software/hwe-mr-calc.shtml). Haplotype analysis was carried out using http://analysis2.bio-x.cn/myAnalysis.php.¹⁴  Statistical significance was set at P < .05.

Magnetic resonance imaging confirmed 101 patients with ACLI and 110 control participants without ACLI. Age, sex, height, and weight did not differ between the groups (P > .05; Table 1).

As shown in Table 3, the allele frequency was not statistically different between the patients and the control individuals (P > .05). The genotype frequency distribution was consistent with the Hardy-Weinberg equilibrium, except for rs1800791 (P < .01). In comparing genotype frequencies, we found that rs1800787 TT was less frequent in the patients than in the control group (P < .05). In contrast, rs1800788 CT, rs1800790 AG, and rs2227389 CT were higher in the patients than in the control group (P < .05). Other genotypes, such as rs1800012, rs12722, rs13946, rs970547, rs240736, and rs1800789, showed no difference between the groups (P > .05). We suggest that the changes of the rs1800787, rs1800788, rs1800790, and rs2227389 genotypes are associated with ACLI.

The linkage disequilibrium (LD) analysis of the SNPs of the collagen genes showed that rs12722 and rs13946 LD were higher in patients with ACLI than in control participants. The remaining sites rs1800012, rs970547, and rs240736 were less linked between the patients and control individuals (Figure 1). The β-fib gene promoter regions rs1800790, rs1800791, and rs2227389 had higher LD in the ACLI patients (Figure 2). The remaining regions rs1800787, rs1800788, and rs1800789 in the patient group were not associated with those in the control group.

Further analysis showed that the frequencies of the rs970547 A and G alleles in males were different between the patient group and the control group (P = .019). The frequencies of A and G in the patients were 71.9% and 28.1%, respectively, and in the control participants were 58.8% and 41.2%, respectively. The frequencies of the rs970547 AA, AG, and GG genotypes in the patient group were 49.3%, 45.2%, and 5.5%, respectively, and in the control group were 27.5%, 62.7%, and 9.8%, respectively. The rs970547 AA genotypes were greater in the ACLI patients (P = .026), indicating that male individuals carrying the rs970547 A allele and AA genotype had an increased risk of ACLI. No differences were found in the remaining SNPs. The analysis of SNPs in female patients did not show any statistical differences (Table 4).

According to the results of the LD analysis, the rs12722 and rs13946 haplotypes (Table 5) and rs1800790, rs1800791, and rs2227389 haplotypes (Table 6) were constructed. No statistical differences in rs12722 and rs13946 haplotype distributions were present between the groups (Table 5; P > .05). However, differences were noted between the haplotypes of TGG*, CGA*, and TGA* of rs1800790, rs1800791, and rs2227389. The detection rates of haplotypes TCG* and CGA* of rs1800790, rs1800791, and rs2227389 in patients were significantly lower than in control participants (Table 6; P < .001), whereas haplotype TGA* was higher in the patients than in the control group (Table 6; P = .0228), suggesting that this combination of SNPs was likely to be the genetic background of Chinese patients with ACLI.

At present, the genetic factors of ACLI are mostly focused on the polymorphism of collagen genes in different regions. In the Caucasian population of South Africa, Khoschnau et al¹⁹  observed that the polymorphism of the COL1A1 Sp1 binding site was associated with ACLI in Caucasians. In 2009, Posthumus et al confirmed that the TT type of the Sp1 binding site was significantly less in South African patients with ACLI.¹¹  Population genetics showed that the frequency of the rs1800012 T allele was 20% and of the G allele was 80% in European populations. In the American population, T was 16% and G was 84%. In the Asian population, T was only 1% and G was 99%.¹²  We did not observe an association between ACLI and the Sp1 binding sites in the Yunnan Han population, indicating that the polymorphism of rs1800012 in the Chinese population was insufficient and that it was not a site related to ACLI. Mokone et al²⁰  studied 111 cases of Achilles tendon injury in the Caucasian population of South Africa, 39 patients with spontaneous tendon ruptures, and 129 asymptomatic sports enthusiasts and found that only the rs12722-BstUI polymorphism of COL5A1 gene was associated with chronic Achilles tendinitis. Among 85 Australian Caucasian patients with Achilles tendon injury, the rs12722-BstUI and rs13946-DpnII polymorphisms were associated with Achilles tendon injury.²¹  The frequency of the rs12722 BstUI CC genotype in female patients with ACLI was higher than in the control group.¹²  This investigation demonstrated that the frequencies of rs12722 C and T alleles in the Yunnan Han population were consistent with the allelic frequency distribution previously reported in Asian populations, but they were not genetic risk factors in the Chinese population. Additionally, the findings for the COL12A1 gene were different. Posthumus et al¹³  reported that, in the Caucasian population, the risk of ACLI was increased in females with rs970547 AluI RFLP AA. However, September et al²²  did not find that the COL12A1 gene was associated with Achilles tendon injury. We observed no difference in the rs970547 A and rs240736 C and T alleles compared with the patient group. Further analysis showed that male patients with ACLI carried more rs970547 A alleles and AA genotypes, but statistical differences were present in female patients.

At the time of ACLI and after surgery, fibrinogen levels often increase significantly.^23–25  The β-fib gene determines the synthesis of the fibrin β polypeptide chain, which in turn affects most of the changes in plasma fibrinogen levels. Deficient fibrinogen in anterior cruciate ligament wound sites has been demonstrated.²⁰  In this study, rs1800787 TT was lower in the patient group than in the control group. Also, rs1800788 CT, rs1800790 AG, and rs2227389 CT were higher in the patients than in the control participants, suggesting that changes in rs1800787, rs1800788, rs1800790, and rs2227389 genotypes were associated with ACLI. Here, rs1800787 TT seemed to confer protection against ACLI. The detection rate of haplotypes of TGG* and CGA* of rs1800790, rs1800791, and rs2227389 in the patients with ACLI was lower than that in the control group, whereas the TGA* was higher in the patients, indicating that this combination of changes is likely to be the genetic background of the ACLI pathogenesis in Chinese individuals.

The COL1A1 gene Sp1 binding site rs1800012, as well as the rs12722 and rs13946 sites, were not susceptible sites for ACLI in the Chinese Han population. The COL12A1 gene rs970547 and rs240736 had a correlation with ACLI in Chinese men. Male individuals carrying the COL12A1 gene rs970547 A allele and AA genotype may have an increased risk of ACLI. The sites of the promoter region of the β-fib gene rs1800788 CT, rs1800790 AG, and rs2227389 CT and haplotypes of TGG*, CGA*, and TGA* of rs1800790, rs1800791, and rs2227389 may be associated with the genetic background of ACLI. The rs1800787 TT polymorphism may offer genetic protection against ACLI among Chinese individuals.