Abstract
Objective: Insights gained from studies suggest that genetic factors are major contributors to the onset and progression of intervertebral disc degeneration (IVDD). The aim of this study is to investigate whether fat mass- and obesity-associated (FTO) gene polymorphisms are related to the disease in Han Chinese People. Methods: 118 IVDD cases and 113 healthy subjects were enrolled in this study. Forty-four single nucleotide polymorphisms (SNP) in the FTO gene were tested and analyzed by the VeraCode GoldenGate Genotyping Assay. Results: A novel SNP rs11076008 was identified in the association analysis between the genotype and the phenotype. There was statistical significance in the expression of SNP rs11076008 in an allelic frequency distribution (p=0.003) and genotype (p=0.014) using the method of multi-test correction. Conclusion: SNP rs11076008 of the FTO gene is associated with IVDD and may play an important role in developing IVDD in Han Chinese People. Our findings provide valuable information regarding the genetic etiology of IVDD in the investigated cohort.
Introduction
Recently, several large-scale genome-wide association studies (GWAS) consistently identified the FTO gene in obesity-related traits and obesity risk (Dina et al., 2007; Frayling, et al., 2007). Interestingly, the associations were related to body weight, osteoporosis phenotypes, and lumbar disc disease (Liuke et al., 2005; Samartzis et al., 2011). However, the relationship between the FTO gene variants and IVDD was unknown. Given that genes may play a role in IVDD, 44 tagging single nucleotide polymorphisms (SNPs) in the FTO gene were investigated with association analysis for the risk of the disease.
Patients and Methods
Participants
A population-based case-control study was adopted, and all subjects enrolled are Han populations from North China, including case (IVDD+) and control (IVDD−) subjects. All recruited individuals completed informed consent with a standard risk-factor and disease history questionnaire for the use of their DNA and clinical data. The study was approved by the Institutional Review Board of the Peking Union Hospital.
118 patients with IVDD+ were recruited from October 2005 to May 2008 in department of orthopedics, Peking Union Medical College Hospital. 113 healthy Participants were selected as controls. All these participants were ethnic Han Chinese, whose age, sex, body mass index (BMI) index (<25), and geological distributions were statistically identical to those of the patients. Participants from the control had neither IVDD symptoms nor a history of related back problems. The enrolled criteria of the test group were as follows: between 18 and 60 years of age; evidence of degenerative disc degeneration with BMI index value less than 25; and participants who had spine trauma, spinal deformity, metabolic disease, leg length discrepancy, backbone infection, or oncosis were excluded. The participants in the control group should not have any discogenic low back pain symptoms (O'Neill et al., 2008), and those who were clinically diagnosed or suspected to have a history of tumor, immunological disease, spine, or other organic deformity were excluded. This study was approved by the ethics committee of the Peking Union Medical College Hospital (Beijing, China). It was conducted according to the standards of the Declaration of Helsinki. Written informed consent was obtained from all participants.
Genotyping
Genomic DNA was extracted from the peripheral blood samples of these participants using QIAamp DNA Blood Mini Kit (QIAGEN Company). SNPs of participants' genes were analyzed with VeraCode GoldenGate Genotyping Assay: 96 & 384-plex. This assay may provide one of the most robust and flexible platforms for SNP genotyping, which benefits from the assay's accuracy, sensitivity, and high signal-to-noise ratio. The relevant reagents in the experiment were supplied by Illumina Company (Illumina, Inc.).
Statistical analysis
SAS software (v9.1; SAS institute) was used for statistical analysis. Hardy-Weinberg equilibrium (HWE) was used to assess whether genotypes fell within a standard distribution. Linkage disequilibrium (LD) blocks were ascertained from Haploview v4.0 (www.broad.mit.edu/mpg/haploview) using the Gabriel algorithm by entering genotype information through linkage format. All genotypes fulfilled HWE expectations (p>0.01). Generalized linear models were used to test genotype associations for continuous outcome traits (i.e., height, weight, and BMI). Logistic regression was used to assess the association between each parameter, and one-way analysis of variance was used to examine significant differences in normal continuous variables among the three genotype groups, whereas the chi-square test was conducted to analyze the differences in categorical variables. A p-value of less than 0.05 was considered significant.
To identify epistatic interactions, multifactor dimensionally reduction (MDR) was used. Briefly, we applied the MDR software to our data set comprising an additional nine SNPs as a control for this analysis in 10-fold cross-validation that may determine the best model for the main SNP-SNP interactions.
The examination of pairwise LD correlation (r2) was computed using Haploview beta software; version 4.0. Multilocus haplotype analyses were performed using Haplostats software, which estimated haplotype frequencies using an expectation-maximization algorithm in situations in which the haplotype phase was ambiguous. Global and haplotype-specific score statistics and permutation-based p-values were calculated after adjustments had been made for gender and age. These analyses did not take into account the relatedness of study subjects.
Deviation of allele frequency from HWE was tested for all SNPs. In addition, the LD pattern between the three SNPs near the FTO gene was tested.
Results
HWE test
We analyzed 44 tagging SNPs falling in the range of a 347-kb interval spanning FTO, among which, SNP rs2388405 and rs741300 did not meet the methods we used in the study. The genotypic and allelic distributions of the determined SNPs were provided from HWE (Table 1). All SNPs were examined with similar BMIs among genotypes of all SNPs.
HW, Hardy-Weinberg; ObsHET, marker's observed heterozygosity; PredHET, marker's predicted heterozygosity; HWpval: Hardy-Weinberg equilibrium p-value; MAF, minor allele frequencies.
The values of observation and expectation, haplogene number, and the chi-square test for genetic frequency of candidate genes were determined with a p-value expressed in the control group. It was clear that there were three tag SNP locations with a p-value below 0.05, which was discrepant from HWE and they were rs12931934, rs1136002, and rs12600060 (Table 1).
Case-control association analysis
Based on the multiple detection and correction data, there may be correlations in the following locations: rs11076008, rs12446047, rs12933928, rs1121980, rs9940128, rs9939973, rs9926289, and rs9923147. Among them, rs11076008 was of the most statistically significant difference (p=0.0003353 and 0.0001553), (Table 2). To correct the false-positive results, permutation test (107 permutations) was performed (Table 3). The p-value was arranged from small to large one after the correction had been distributed in Figure 1. The results in Table 3 indicated that there was a strikingly significant correlation between rs11076008 and IVDD (p=0.0058). To correct the type I error, one more conservative correction method—Bonferroni—was adopted. The results also showed that there was a strikingly significant relationship between rs11076008 and IVDD (p=0.014 and 0.0065, or significant level was 0.05/42=0.0012).
Illustration of chi-square distribution of the result from permutation test.
LD analysis and haplotyping
The FTO gene is located in human chromosome 16q12.2; after spanning the SNP alleles from 52295376 to 52705882 regions, using LD (Fig. 1) and tagging SNP data (Fig. 3), we identified five haplotype blocks (Fig. 2) that had a significant relationship between TA and IVDD (p=0.00332).
Linkage disequilibrium (LD) plot for 44 single nucleotide polymorphisms (SNPs) in the FTO-associated gene indicated five blocks. Genotypes were entered by linkage format into Haploview v4.0. The numbers inside the diamonds represent the D’ for pairwise analysis. The darker the square, the higher the LD between two variants. Bold polymorphisms are contained inside the LD blocks.
Tagging SNP data showed SNP rs11076008 located in human chromosome 16q12.2 after spanning the SNP alleles from 52295376 to 52705882 regions.
Multifactor-clinical genotyping
It was clear that there were no correlations among the locations presented earlier, using multifactor dimensionality reduction analysis. The rs11076008 location showed a higher main effect (Fig. 4A). Comparatively, the consistency of the two-factor effecting model rs9923147+rs806158 is poor (CV4/10) (Fig. 4B).
The results regarding epistatic interaction of intervertebral disc degeneration showed that SNP rs9923147 and rs8061518 were involved in the event
Discussion
We explored the relationship between SNP rs11076008 in the FTO gene and IVDD, which was motivated by multiple lines of evidence showing that the genetic factor may be associated with IVDD.
Liuke et al. (2005) first reported that body weight had a positive correlation with IVDD, while Frayling et al. (2007) found that the FTO gene played an important role in obesity and body weight. Since both IVDD and obesity were highly correlated phenotypically and genetically, the relativity of the gene to obesity may also be considered a candidate for IVDD. Thus, we hypothesized that the FTO gene variants might exert certain effects that are linked to obesity as well as IVDD.
In the present study, we investigated the association of FTO gene polymorphism (rs11076008) with intervertebral degeneration traits in a case-control study containing 330 Han Chinese subjects. Although several studies have been focused on the associations about several gene polymorphisms in patients with IVDD, the study clearly showed that SNP rs11076008 in the FTO gene may be an important etiological factor in the subjects with IVDD. Furthermore, we also revealed that there may be an association relationship between FTO haplotyping Block 5 and IVDD, and there was no detailed correlation in multiple locations on the SNPs within the FTO gene in our study. Consequently, SNP rs11076008 might become useful in clinical applications with the potential to develop genetic screening for the individuals with IVDD as well as gene therapy for the disease.
A mechanism(s) by which the FTO genetic polymorphism may be associated with IVDD is unknown. Further studies are needed to elucidate it. Another point mentioned was whether ethnic discrepancy existed in the FTO genetic variants associated with IVDD. Li et al. (2008) reported that the variants in the FTO gene were not associated with obesity in a population-based study containing 3210 subjects, which was inconsistent with our studies which suggested that the FTO gene polymorphisms were significantly related to obesity, adipose, and body weight index (Do et al., 2008; Rampersaud et al., 2008; Qi et al., 2008; den Hoed et al., 2009). Different genetic architecture and allele frequencies in Chinese people may produce the discrepancy between Chinese people and other populations. Thus, further studies, especially with a larger sample size, should be conducted to examine whether SNP rs11076008 of the FTO gene is related to IVDD in other ethnic groups, and an etiological mechanism(s) needs to be investigated.
Conclusions
Our results indicate that SNP rs11076008 polymorphism of the FTO gene is highly associated with IVDD patients, suggesting that it plays an important role in occurrence and progression of IVDD in the investigated cohort.
Footnotes
Author Disclosure Statement
No competing financial interests exist.