Gene locus polymorphisms and expression levels of interleukin-1 in lumbar disc disease: a meta-analysis and immunohistochemical study

Background: To investigate the association between interleukin (IL)-1α (rs1800587), IL-1β (rs1143634) and IL-1 receptor antagonist (RN) variable number tandem repeat (VNTR) polymorphisms, expression levels and lumbar disc disease (LDD). Methods: All relevant articles were searched from 4 databases including PubMed, Embase, Web of Science and China National Knowledge Infrastructure (CNKI). Odds ratios (OR) with 95 % con ﬁ dence intervals (CI) were calculated to evaluate the association between IL-1 gene locus polymorphisms (rs1800587 in IL-1α, rs1143634 in IL-1β, VNTR in IL-1RN) and LDD susceptibility. Statistical analysis was conducted by Review Manager (Revman) 5.31 software. Furthermore, qRT-PCR and immunohistochemistry (IHC) were performed to evaluate IL-1α, IL-1β and IL-1RN expressions in the normal and degenerated disc. Results: A total of 15 case-control studies (1455 cases and 2362 controls) were included in our meta-analysis. The pooled results suggested that IL-1α rs1800587 polymorphism was associated with an increased risk of LDD in overall population (T vs. C, OR=1.21, 95%CI=1.04-1.40, P=0.01). The subgroup analysis found a signi�cant association between IL-1β rs1143634 polymorphism and LDD in Asian population (T vs. C, OR=0.61, 95%CI=0.39-0.96, P=0.03). Results of qRT-PCR and IHC demonstrated that expressions of IL-1α and IL-1-β were signi�cantly increased in the degenerated disc. (all P<0.05 ) Conclusion: IL-1α rs1800587 and IL-1β rs1143634 polymorphisms were signi�cantly associated with LDD in overall population and in Asian population, respectively.


Introduction
Low back pain (LBP) is a most common musculoskeletal disorder.The annual prevalence of LBP ranges from 15% to 45%, and 70-85% of all people have LBP at some time in life [1].The consequences of LBP and related disability are substantial, affecting individuals, families, and health-care systems [2].It is accepted that lumbar disc degeneration (LDD) is a main risk factor for LBP [3].Although the pathogenesis of LDD is not fully understood, immune system has been proved to play an important role in development of LDD [4][5][6].Interleukin (IL)-1 cytokines are key mediators of immune responses and apoptosis of intervertebral disc cells [7,8].Recently, increasing evidence showed that IL-1 gene cluster, including IL-1, IL-1β and IL-1 receptor antagonist (RN), could be responsible for the appearance or the severity of LDD [9,10].

Methods
The meta-analysis conformed to the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines, and registered in PROSPERO International Prospective Register of Systematic Reviews (CRD42019124118).This study was approved by the Ethics Committee of the First A liated Hospital of Guangxi Medical University (2018-KY-NSFC-025).Written informed consent was obtained from all the participants involved in this study.
All procedures were performed in accordance with the guidelines the institutional research committees and with the Declaration of Helsinki.

Search strategy
Four electronic databases, including PubMed, Embase, Web of Science and China National Knowledge Infrastructure (CNKI), were searched by two reviewers (HJ and QX).The following keyword search string was used to identify studies: (IL-1 OR IL 1 OR interleukin-1 OR interleukin 1 OR IL-1RN) AND (polymorphism OR variant OR mutation OR SNPs) AND (disc degeneration OR disc disease OR low back pain).Additional studies were identi ed through a hand search of references listed in the reports and reviews.No language or publication date restrictions were used.The nal literature search was conducted on February 5, 2020.

Inclusion and exclusion criteria
Eligible studies were included in meta-analysis according to the following inclusion criteria: (1) case-control design; (2) LDD diagnosed on the basis of clinical or/and radiologic examinations; (3) the study evaluated the association between IL-1α rs1800587 or IL-1β rs1143634 or IL-1RN (VNTR) polymorphisms and LDD risk; (4) genotype of control group conformed to the Hardy-Weinberg balance; and (5) su cient data were provided to calculate the odds ratios (ORs) and 95% con dence intervals (CI).The exclusion criteria were as follows: (1) comments, reviews, or meta-analysis; (2) repeated publications; (3) study without available data.Based on the inclusion and exclusion criteria, eligible studies were identi ed independently by two reviewers (HJ and QX).The disagreements were resolved by a third reviewer (QW).

Data extraction
According to a standardized form, two investigators (HJ and QX) independently extracted data on outcomes for each study.The following information were collected from the included studies: (1)

Methodological quality assessment
Methodological quality of studies was assessed using Critical Appraisal Skills Programme (CASP) [26].For each question, there were three answers including "no", "can't tell" and "yes", which respectively indicate scored 0, scored 1 and scored 2. The quality score ranged from 0 to 20, and a study with scored 15-20 represented a high-quality study.

Study population
Based on our previous study [27], we collected degenerative disc tissues (n=34) and normal disc tissues (n=21) from the LDD patients and the control subjects, respectively (Table S1).These disc samples were used to evaluate gene expressions via qRT-PCR and IHC.LDD patients were diagnosed as lumbar disc herniation by physical examination and MRI scan.The nal diagnosis was veri ed by histopathology.The control subjects were the patients with traumatic lumbar vertebral fracture, who had no history of low back pain.According to Schneiderman's classi cation [28], MRI evaluation of the control subjects showed no signi cant disc damage and degeneration before surgery (Schneiderman's classi cation, Grade 1: 19 cases; Grade 2: 2 cases).

Statistical analysis
The association between IL-1α rs1800587, IL-1β rs1143634 and IL-1RN (VNTR) gene polymorphisms and LDD were estimated by odds ratio (OR) and 95% confidence intervals (CI).We evaluated the pooled ORs in ve different genetic models.The heterogeneity of statistics was calculated by chi-square-based Q statistics and I 2 statistics.Heterogeneity was considered to be effective, when P was less than 0.10 and I 2 was greater than 50%.If there was signi cant heterogeneity (I 2 >50%), the random effect model was used.Otherwise, the xed effect model was applied.Subgroup analysis of ethnicity was conducted to identify the source of heterogeneity.HWE was detected in control groups using the chi-square test.We performed the sensitivity test to assess the possible in uence of one study on the pooled OR.In sensitivity test, studies were removed, in turn, from the overall analysis.Publication bias was tested by Begg's funnel plot and Egger's test.The qRT-PCR and IHC assay were calculated by the mean±standard error.Statistical difference between two groups was evaluated using unpaired Student t test.P<0.05 was considered to be statistically signi cant.Statistical analysis was performed by STATA 12.0 software (Stata, College Station, TX) and Revman 5.31 software (Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Denmark).

Characteristics of included studies
A ow chart of article selection process is described in Figure 1.A total of 15 studies that met the inclusion criteria were identi ed in our meta-analysis, including 1455 cases and 2362 controls (IL-1α rs1800587: 10 studies, 922 cases and 1351 controls), (IL-1β rs1143634: 10 studies, 1056 cases and 1747 controls), and (IL-1RN VNTR: 3 studies, 290 cases and 507 controls).LDD was diagnosed by MRI scan in all the studies.The characteristics of included studies were shown in Table 1 and 2. All eligible studies were categorized as high quality, with scores >15.IL-1α rs1800587, IL-1β rs1143634 and IL-1RN (VNTR) polymorphisms and LDD susceptibility The meta-analysis of IL-1α rs1800587, IL-1β rs1143634 and IL-1RN (VNTR) polymorphisms are presented in Table 3.The pooled result showed rs1800587 was signi cantly associated with LDD risk in overall population (T vs. C, OR=1.21, 95%CI=1.04-1.40,P=0.01) (Figure 2).This signi cance was across the ethnicity.It was proposed that IL-1α rs1800587 T allele increased the susceptibility to LDD, and in contrast, IL-1α rs1800587 C allele was protective.However, IL-1β rs1143634, IL-1RN (VNTR) polymorphisms were not associated with LDD susceptibility in overall population (all P>0.05).After strati cation by ethnicity, the result showed C allele of IL-1β rs1143634 may be a risk allele for LDD in Asian population (T vs. C, OR=0.61, 95%CI=0.39-0.96,P=0.03) (Figure 3).Sensitivity analysis was performed to examine the impact of each study on the pooled ORs by removing each study in turn (Table 4).For rs1143634, when the study reported by Mu et al. was omitted in turn, the heterogeneity was obviously reduced under allele contrast genetic models [13,19,22].For IL-1RN (VNTR), when we omitted the study reported by Ye et al. [14], the heterogeneity was signi cantly reduced under allele contrast genetic models.When the studies deviated from HWE were excluded, our results were robust and consistent.Funnel plots showed no signi cant evidence of publication bias (Figure 4a and 4b) (all P>0.05) IL-1α, IL-1β and IL-1RN expressions and LDD Hematoxylin & eosin stained sections were shown in Figure 5.In contrast to control group, IL-1α and IL-1β mRNA levels were increased 2.6 fold and 1.7 fold in LDD group, respectively (Figure 6d and 6h).IL-1RN mRNA levels were not signi cant difference between the two groups (Figure 6l) (all P>0.05).IHC analysis showed that signi cantly higher IL-1α and IL-1β expression levels in the LDD group than those in the control group (Figure 6a-c and Figure 6e-g) (IL-1α immunopositive cells: 43.8±4.9% vs. 22.5±2.8%,P<0.001; IL-1β immunopositive cells: 36.4±3.7% vs. 21.6±2.3%,P<0.001).However, there were no signi cant differences in IL-1RN expression between the two groups (Figure 6i-k) (IL-1RN immunopositive cells: 21.4±3.3% vs. 24.8±2.3%, P=0.09).

Discussion
Our meta-analysis 15 studies, involving 1455 cases and 2362 controls, found the statistically signi cant associations between IL-1α rs1800587, IL-1β rs1143634 polymorphisms and LDD.The pooled analysis indicated that T allele of rs1800587 was signi cant associated with increased risk of LDD in overall population.This signi cance was across the ethnicity.Subgroup analysis revealed that IL1-β rs1143634 polymorphism was associated with LDD in Asian population but not in Caucasian population.The C allele of rs1143634 was identi ed as a risk allele in the patients with LDD.Furthermore, the results of qRT-PCR and IHC analysis demonstrated that increased IL-1α and IL1-β expression levels were found in the degenerated disc.Wang et al. initially reported a metaanalysis on IL-1α rs1800587 and IL-1β rs1143634 polymorphisms and LDD, which suggested that IL-1α rs1800587 polymorphism was signi cantly associated with the risk of disc degeneration [29].There were several weaknesses in the previous meta-analysis.First, some data extraction errors were found in meta-analysis, such as the allele and genotype frequencies from Solovieva's and Aparicio's studies [11,18].Second, the result of pooled analysis was limited by the marginal p values.This may lead to in ate the chance of a false positive association.More importantly, three genetic studies focusing on this topic were published in recent years [23][24][25], which were not included in Wang's study.Based on the Cochrane guidelines [30], an overlapping meta-analysis is necessary to be updated in time with latest studies.To the best of our knowledge, the current study is the largest sample size of meta-analysis to investigate the association between IL-1 gene locus polymorphisms and LDD.
IL-1 has extensively been studied among proin ammatory mediators, and is believed to play a critical role in the etiology of LDD [8,31].There are 3 major members in the IL-1 gene family: IL-1α, IL-1β, and IL-1RN.IL-1α and IL-1β have a strong in uence on apoptosis of intervertebral disc cells, whereas IL-1RN suppresses the effect of IL-1 by competitively inhibiting the binding of IL-1 to the IL-1 receptor [32,33].The genetic control of the cytokine function may have an impact on the occurrence and the severity of LDD [8].We performed an in silico analysis for evaluating the possible functional implication of IL-1α rs1800587, IL-1β rs1143634 and IL-1RN (VNTR) polymorphisms by using rSNPase (http://rsnp3.psych.ac.cn/).The results indicated that rs1800587 and rs1143634 were located within the promoter and exon 5 of IL-1 gene, which may affect the normal production, secretion or function of IL-1.The C to T polymorphism at position-889 of IL-1α gene (rs1800587) could increase gene expression at mRNA and at protein levels by enhanced promoter activity [34].
Regarding the IL-1β rs1143634 (T/C) polymorphism, earlier data suggested that a shift from T to C lead to a disruption of the TATA box in exon 5 sequences.The C allele conferred higher expression of the IL1β gene compared to the T allele [35,36].However, the relationship between IL-1α, IL-1β polymorphisms and IL-1α, IL-1β expressions in LDD patients has not been reported.The results of qRT-PCR and IHC analysis demonstrated that elevated IL-1α and IL-1β expression levels were found in the degenerated disc.Compared with other genotypes, TT genotype of rs1800587 and CC genotype of rs1143634 were associated with higher expression levels of IL-1α and IL-1β respectively.Thus we postulated that rs1800587 TT and rs1143634 CC genotypes were the genetic risk factors for the progression of LDD, probably by increasing the expression of IL-1α and IL-1β [37][38][39].Mutations in the introns may affect gene expression and function via effects on mRNA splicing or RNA stability.IL-1RN (VNTR) gene polymorphism, located within a non-regulatory region (intron 2), could not be part of RNAbinding protein site.Thus the functional in uence of IL-1RN (VNTR) polymorphism remains unclear.The molecular mechanisms of IL-1 gene locus polymorphisms and expressions in LDD are likely to be more complicated, which need be investigated in the future.
Some limitations in our study should be taken into consideration.First, the sample size is relatively small, which may exert an impact on the statistical power.There is much to be done to ensure the accuracy of this result.Second, we only acquired suitable studies published in English or Chinese.The potential publication bias could not be eliminated as the exclusion of unpublished articles, or articles published in another language.Third, the function information provided by qRT-PCR and IHC is limit to assess the exact mechanisms of disc degeneration.It is necessary to accumulate further evidence to clarify the impact of IL-1 gene locus polymorphisms on LDD.

Conclusion
IL-1α rs1800587 polymorphism was signi cantly associated with LDD in overall population, while IL-1β rs1143634 polymorphism was signi cantly associated with LDD in Asian population but not in Caucasian population.IL-1α rs1800587 T allele and IL-1β rs1143634 C allele could increase the susceptibility to LDD.

Declarations Ethical Approval and Consent to participate
The histology study was approved by the Ethics Committee of the First A liated Hospital of Guangxi Medical University.All the experimental protocol and the methods were carried out in accordance with the relevant guidelines and regulations, and complied with the principles of the Declaration of Helsinki.Written informed consent was achieved from each participant.

Consent for publication
Not applicable.There was no signi cant publication bias (P>0.05).
Hematoxylin & eosin stained sections from the control group (a) and the LDD group (b).(Scale bar=20 μm)

Figures Figure 1
Figures

Table 1 .
Characteristics of the case-control studies included in meta-analysis

Table 2 .
Genotype frequency of IL-1 gene locus polymorphisms in case-control studies

Table 3 .
Association test and heterogeneity test of IL-1 gene locus