Gene polymorphisms and expression levels of interleukin-6 and interleukin-10 in lumbar disc disease: a meta-analysis and immunohistochemical study

This meta-analysis was conducted in accordance with a prespecified protocol registered with PROSPERO International Prospective Register of Systematic Reviews (CRD42019124118). Two authors (YW Guan and ST Wang) searched all major databases up to February 28, 2019, including PubMed, Embase, Web of Science, Cochrane Library, and China National Knowledge Infrastructure (CNKI). The following keywords were used for searching: (“LDD” OR “lumbar disc disease” OR “lumbar disc degeneration” OR “intervertebral disc degeneration”) AND (“IL-6” or “interleukin-6” OR “IL-10” OR “interleukin-10”) AND (“SNP” OR “polymorphisms”). In addition, we checked the reference lists of the included studies for further relevant literature. No publication date or language restrictions were implemented. The authors strictly adhered to the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines [20] throughout the study.

To include in all the eligible articles and exclude ineligible articles, the inclusion and exclusion criteria were established. The studies were included in the review if they (1) recruited the experimental subjects which were diagnosed as LDD by clinical or/and radiological examination, (2) were cohort-based or case-control studies that assessed the associations of IL-6 and IL-10 gene polymorphisms with LDD, (3) evaluated the genotype of the control group and conformed to the Hardy-Weinberg balance, and (4) calculated the odds ratios (OR) to assess the association. Exclusion criteria were (1) repeated publications; (2) conference abstracts, letters to editor, and unpublished studies; and (3) incomplete data on allele and genotype frequencies. Articles published in languages other than English were translated. Based on the inclusion and exclusion criteria, two authors (YW Guan and ST Wang) independently screened the titles and abstracts of references and obtained the full text for reference. Disagreements between two authors were resolved by discussion, and if necessary, a third author (H Jiang) was consulted.

Based on a standardized form, two authors (YW Guan and ST Wang) independently extracted data on outcomes for each study. For disagreements, a consensus was reached by a third author (H Jiang). Extracted parameters included the following: (1) first author, (2) publication year, (3) study population (country, ethnicity), (4) study design, (5) numbers of cases and controls, (6) characteristics of participants (age and gender), (7) diagnostic criteria, (8) source of controls, (9) allele or/and genotype frequencies, and (10) results of Hardy-Weinberg equilibrium (HWE) test. Hardy-Weinberg equilibrium was checked in study controls using the χ² goodness-of-fit test as a quantitative assessment for potential selection bias and confounding.

Methodological quality of studies was independently assessed by two investigators according to a quality evaluation form (Critical Appraisal Skills Programme for case- control study, CASP), which is based on ten questions associated with information given by each article [21]. Each question has three degrees behind, “yes” (scored 2), “cannot tell” (scored 1), and “no” (scored 0). The maximum total score is 20. Studies could be divided into three grades: grade A (high quality, scored 15–20), grade B (medium quality, scored 8–14), and grade C (low quality, scored 0–7).

Based on our previous study [22], we collected degenerative disc tissues (n = 34) and normal disc tissues (n = 21) from the LDD patients and the control subjects. LDD patients were diagnosed with lumbar disc herniation by physical examination and MRI scan. The control subjects were the patients with traumatic lumbar vertebral fracture, who had no history of low back pain. Based on Schneiderman’s classification [23], MRI evaluation showed no significant disc damage and degeneration before surgery (Schneiderman’s classification, grade 1, 19 cases; grade 2, 2 cases). These disc samples were used to evaluate IL-6 and IL-10 expressions via IHC and RT-PCR. This study was approved by the institute’s ethics committee for human studies (2018-KY-NSFC-025). Informed consent was obtained from all the participants in this study.

The intervertebral disc tissue was fixed in 10% neutral buffered formalin within 1 h after surgical excision and was embedded in paraffin for serial sectioning at 3 μm. After routinely xylene dewaxing and gradient ethanol hydration, the sections were incubated in 0.01 mol/L citrate buffer (pH = 6). The sections were immersed in 3% hydrogen peroxide for 10 min and subsequently rinsed three times with PBS buffer solution for 3 min each time at room temperature. Nonspecific binding was blocked with 3% normal goat serum in a phosphate-buffered saline solution (pH = 7.4) for 15 min at room temperature. Primary antibody (polyclonal rabbit anti-IL-6/anti-IL-10, ab6672/ab34843, Abcam, Cambridge, UK) was diluted at 1:400 and incubated at 4 °C overnight. After washing, it was followed by secondary antibody (Biotin-labeled Goat Anti-Rabbit IgG, SP-9001, ZYMED, USA) for 15 min at room temperature. Then, the sections were incubated in Horseradish Peroxidase Streptavidin for 15 min. DAB reagent was added to the section that was examined by microscope and incubated for 10 min at room temperature. Sections were counterstained with hematoxylin for 10 min, dehydrated through several baths of graded hydrochloric alcohol and xylene dehydration, and then mounted using Neutral Balsam. The results of immunohistochemistry (IHC) were obtained using an Olympus BX43 upright microscope (Olympus Optical, Tokyo, Japan). In addition, fresh intervertebral disc tissues were collected to obtain the total RNA. The total mRNA was extracted with TRIzol (Invitrogen Life Technologies, CA, USA). One microgram of the total RNA was used to synthesize complementary DNA (cDNA) using an iScript cDNA Synthesis kit (Quanta Biosciences, MD, USA). Subsequently, real-time PCR amplification was performed using specific primers of target genes and a SYBR Green real-time PCR kit (Quanta Biosciences, MD, USA). The primer sequences were selected according to previous study [24].

Odds ratio (OR) and 95% confident intervals (CI) were used to assess the strength of associations between IL-6 and IL-10 gene polymorphisms with LDD susceptibility which were conducted under five genetic models. Chi-square-based Q test and I² test were used to check the heterogeneity among the included studies. A fixed-effect model was used while no heterogeneity existed (P > 0.10, I² < 50%). Otherwise, a random-effects model was used (P < 0.10, I² > 50%). For comparisons which have significant heterogeneity, we have performed a sensitive analysis to evaluate the effect of one study on the pooled OR. The Hardy-Weinberg equilibrium of controls was calculated using the HWE Version 1.20 program. Potential publication bias was assessed using Begg’s funnel plot and Egger’s tests. IHC and RT-PCR data were presented as the mean ± standard error. Statistical difference between the two groups was evaluated using unpaired Student’s t test. Statistical significance was set at P < 0.05. Data analysis was conducted using SPSS 20.0 statistical software (SPSS Inc., Chicago, IL, USA) and Review Manager 5.31 (Nordic Cochrane Center: http://​ims.​cochrane.​org/​revman/​ download).

The flow chart for the identification of the studies is presented in Fig. 1. A total of 6 studies were finally identified, including 1456 cases and 1611 controls (rs1800795: 4 studies, 638 cases, and 879 controls; rs1800796: 5 studies, 1136 cases, and 1342 controls; rs1800797: 3 studies, 371 cases, and 579 controls; rs2069849: 2 studies, 221 cases, and 333 controls; rs13306435: 2 studies, 221 cases, and 333 controls; rs1800896: 3 studies, 742 cases, and 748 controls; rs1800871: 2 studies, 587 cases, and 569 controls). The main characteristics of included studies are presented in Table 1 and Table 2. For quality assessment, all selected studies are categorized as grade A according to scores ranging from 15 to 19 (Table 1).

The meta-analysis of IL-6 and IL-10 gene polymorphisms is presented in Table 3. There was a significant association between IL-6 polymorphisms (rs1800795 and rs1800797) and LDD predisposition in five genetic models (G vs. C, OR 1.39, 95% CI 1.15–1.67, P = 0.0005; G vs. A, OR 1.35, 95% CI 1.12–1.63, P = 0.002) (Figs. 2 and 3). For subgroup analysis, IL-6 rs1800795 polymorphism was negatively associated with LDD risk in the European population in all genetic models except the dominant model (CC+CG vs. GG). In addition, no significant association of other IL-6 and IL-10 gene polymorphisms (rs1800796, rs13306435, rs2069849, rs1800871, and rs1800896) with LDD was observed (all P > 0.05).

Sensitivity analysis was performed by excluding one study at a time to analyze the heterogeneity. For rs1800896, the overall results were not altered in all genetic models after omitting the study by Huang et al. [14], and the heterogeneity was obviously reduced after omitting Huang’s study (Table 4). For rs1800796, the result of the allelic contrast model suggested a negative association between rs1800796 and LDD after excluding the study reported by Eskola et al. [11], and the heterogeneity was significantly reduced after excluding Eskola’s study (Table 5). According to Begg’s funnel plot and Egger’s regression tests, the result indicated no significant publication bias under all genetic models (all P > 0.05 for all models tested) (Fig. 4).

The proteins of IL-6 and IL-10 expressed in the cytoplasm were detected in the nucleus pulpous of the intervertebral disc. Compared with the control group, the LDD group had significantly higher expression of IL-6 (IL-6 immunopositive cells, 73.68 ± 10.99% vs. 37.23 ± 6.42%; P < 0.001). However, there were no significant differences in IL-10 expression between the two groups (IL-10 immunopositive cells, 41.18 ± 23.56% vs. 32.79 ± 20.56%; P = 0.325). For RT-PCR, there were higher IL-6 mRNA levels in the LDD group than those in the control group (P < 0.001). No significant differences in IL-10 expression levels were found in the two groups (P = 0.112) (Fig. 5).