Objectives: The human leukocyte antigen-G may have a positive role in graft acceptance in human organ transplant. Several studies have reported an association between the human leukocyte antigen-G-14-base-pair-insertion/deletion polymorphism and risk of developing kidney graft rejection, but the results are inconclusive. We performed a meta-analysis to evaluate this association.
Materials and Methods: We included 5 case-control studies that evaluated the association between human leukocyte antigen-G-14-base-pair-insertion/ deletion polymorphism and risk of developing kidney transplant rejection, including a total 907 patients (rejection, 271 patients; no rejection, 636 patients).
Results: There was no significant association between the human leukocyte antigen-G-14-base-pair-insertion/deletion polymorphism and risk of developing kidney transplant rejection in the allele contrast, homozygous, heterozygous, recessive, or dominant genetic models for all rejection or acute rejection. In 2 studies, there was a significant association between human leukocyte antigen-G-14-base-pair-insertion/deletion polymorphism and chronic graft rejection in the allele contrast model (+14 vs -14: odds ratio, 0.68; 95% confidence interval: 0.48-0.96; P = .618), heterozygous model (+14/-14 vs -14/-14: odds ratio, 0.44; 95% confidence interval: 0.23-0.83; P = .248), and dominant genetic model ([+14/+14 and +14/-14] vs -14/-14: odds ratio, 0.48; 95% confidence interval: 0.30-0.78; P = .355).
Conclusions: There may be no association between 14-base-pair polymorphisms and risk of developing kidney allograft rejection. Additional studies with larger sample size and better study design are justified.
Key words : Chronic kidney disease, Genetics, Immunology, Rejection
Kidney transplant, accompanied by an immuno-suppressive regimen, is a common treatment for end-stage renal failure. Engraftment success is limited by graft rejection. Both acute and chronic rejection may contribute markedly to allograft dysfunction.1 Genetic polymorphisms within transplant recipients may cause decreased or increased immunologic responses. Genetic analysis in transplant patients may help individualize immunosuppressive regimens by identifying alleles that may increase risk or confer protection for immune-mediated complications or by screening for mutations of genes that are important for drug metabolism.2
Human leukocyte antigen-G (HLA-G) is a nonclassic HLA class I molecule that may be expressed selectively at the maternal-fetal interface on cytotrophoblast cells. In addition, HLA-G expression may occur in various tissues under pathologic conditions such as organ transplant, viral infection, malignant transformation, and inflammatory and autoimmune diseases.3 Furthermore, HLA-G may play a positive role in allograft acceptance. The potential tolerogenic role of HLA-G in kidney transplant makes it a possible marker to select patients who may have better allograft acceptance and who may benefit from reduced immunosuppressive treatment. The 14-base-pair (14-bp-) insertion/ deletion polymorphism at the 3’ untranslated region of the HLA-G gene has an important role in the regulation of gene expression because it may affect HLA-G alternative splicing and mRNA stability.3
Several studies have reported an association between the 14-bp-insertion/deletion polymorphism at the 3’ untranslated region and kidney graft rejection risk, but results are conflicting. Therefore, we performed a meta-analysis of 5 eligible case-control studies involving 907 kidney transplant recipients and 887 healthy control subjects. The purpose of this study was to estimate the association between the 14-bp-insertion/deletion polymorphism and risk of developing kidney graft rejection and to address between-study heterogeneity and publication bias.
Materials and Methods
We performed a search in Internet databases (PubMed, Embase, Web of Science, and China Biology Medicine disc) with a combination of the key words “polymorphism,” “HLA-G,” and “kidney transplantation” most recently on March 31, 2013. Additional related studies were identified by searching the references of studies that were identified in the Internet search. In studies that included the same or overlapping data published by the same investigators, we selected the most recent studies that had the largest number of subjects, and no restriction was applied.
Criteria for studies included in the current meta-analysis were (1) evaluation of the 14-bp-insertion/deletion polymorphism of HLA-G gene and kidney transplant rejection; (2) case-control study design; and (3) sufficient published data to enable estimation of odds ratio (OR) and 95% confidence interval (CI). Major exclusion criteria were (1) absence of a control group; (2) absence of genotype frequency; and (3) duplication of previous publications. Study quality was assessed with Newcastle-Ottawa quality assessment scale.
Data were independently extracted in duplicate by 2 investigators (WH and ZJ) using a standardized protocol and data collection form. For conflicting evaluations, agreement was reached after discussion. Included characteristics were the name of the first author, year of publication, country of study origin, source of control group, sample size, and Hardy-Weinberg equilibrium (Table 1).4-9
All statistical analyses were performed with software (Stata 11.0, StataCorp, College Station, TX, USA) using 2-sided tests. The chi-square test was used to compare genotype (likelihood ratio P value) and allele frequencies (continuity correction P value) between the kidney recipient and healthy control groups. Statistically significant differences in the frequencies between the kidney transplant recipient and healthy control groups were defined by P ≤ .05.
The strength of association between 14-bp-insertion/deletion polymorphism and kidney rejection risk was assessed by summary OR and 95% CI. The significance of the pooled OR was determined by Z-test and statistical significance was defined by P ≤ .05. For the 14-bp-insertion/deletion polymorphism, the meta-analysis compared association between 14-bp-insertion allele (+14) and rejection risk versus 14-bp-deletion allele (-14) and rejection risk (allele contrast, +14 vs -14). In addition, homozygous (+14/+14 vs -14/-14), heterozygous (+14/-14 vs -14/-14), recessive (+14/+14 vs [+14/-14 and -14/-14]), and dominant ([+14/+14 and +14/-14] vs -14/-14) models for allele +14 were used. Furthermore, subgroup analyses were performed based on acute and chronic rejection.
Heterogeneity was evaluated with Cochran Q test based on the chi-square test and the I2% to evaluate the variation in study outcomes between different studies; I2% < 0.25 indicated a lack of heterogeneity among the studies. The pooled OR estimate of each study was calculated by the fixed-effects model when the effects were assumed homogeneous. Moderate heterogeneity was defined by 0.25 < I2% < 0.75, and in this case the random-effects model was used. When I2% > 0.75, we considered it not proper to determine a pooled OR. Hardy-Weinberg equilibrium was tested using the chi-square test for the distribution of frequencies of 14-bp polymorphism.4
A Begg funnel plot was constructed and the degree of asymmetry of the plot was evaluated using Egger test to evaluate publication bias. Statistically significant publication bias was defined by P ≤ .05.
Study identification and characteristics
There were 45 studies identified in the initial search, of which 18 studies were unrelated to the topic of interest (Figure 1). In the 27 relevant studies, 22 studies were excluded because of duplicated publication, meeting abstracts duplicated with journal articles, overlapping data published by the same investigators, and absence of data about the 14-bp-insertion/deletion polymorphisms (Figure 1). Therefore, 5 studies that included a total of 907 kidney transplant recipients (rejection, 271 patients; no rejection, 636 patients) and 887 healthy control subjects were included in the meta-analysis (Table 1).5-9 According to the Newcastle-Ottawa quality assessment scale, all the studies were of high quality (Table 2).
Comparison between kidney recipient and healthy control groups
There were no significant differences in genotype or allele frequencies between transplant recipient and healthy control groups except for a significant difference for allele frequency between the 2 groups in 1 study (Table 3).8 Therefore, the kidney recipient group in the 5 studies came from a general population in terms of HLA-G gene characteristics, enabling further study of genotype and allele distribution within the recipient group.
There were 5 different genetic models to test the 14-bp polymorphism and risk of graft rejection (overall, acute, and chronic) (Table 4). With the recessive genetic model, the P for heterogeneity and I2% indicated moderate heterogeneity, and the random-effects model was used; the overall OR for the recessive genetic model was 1.06, suggesting that there was no association between this genetic model and graft rejection in the recessive genetic model (Table 4, Figure 2).
Evaluation of publication bias with Begg funnel plot and Egger test for the included studies showed an asymmetric funnel plot (Figure 3). The Begg test for overall effect Z value was -0.49 (P = .624), and the Egger linear regression test for t value was -0.86 (P = .451); therefore, both tests showed no significant publication bias in this model. Comparison for the other 4 genetic models to compare the overall rejection and control groups showed no significant association for these 4 models (Table 4). There were no significant associations between gene polymorphisms and acute rejection (Table 4). The 14-bp polymorphisms were associated with lower chronic rejection risk in the allele contrast, heterozygous, and dominant genetic models but not the homozygous or recessive genetic models (Table 4). We did not perform stratified analysis by ethnicity, country, or sex because of limited data and number of studies.
The HLA-G antigen is an immunomodulatory and tolerogenic molecule and is important in fetal-maternal physiologic immunotolerance.10 The HLA-G gene is located within the major histocompatibility complex at the telomeric part of the 6p21.3 chromosomal region and has few polymorphisms compared with other genes located in this region.11 In healthy subjects, HLA-G is expressed in few tissues such as extravillous cytotrophoblasts, oocytes, embryos, amnion, and thymic epithelial cells.3 In patients who have transplant, cancer, viral infection, multiple sclerosis, or inflammatory diseases, the HLA-G molecule may be induced in tissues affected by disease.3 There are minimum 7 HLA-G isoforms, including 4 membrane-bound proteins (HLA-G1 to HLA-G4) and 3 soluble proteins (HLA-G5 to HLA-G7).12 The HLA-G molecule exerts immuno-modulatory and tolerogenic functions by inhibiting cytolytic activity of natural killer cells, antigen-specific cytotoxic T cells, allogeneic proliferative response of CD4+ T cells, and maturation of dendritic cells; HLA-G also induces tolerogenic antigen-presenting cells or suppressor T and natural killer cells. The HLA-G molecule performs these functions by binding the inhibitory receptors ILT2 (CD85j/LILRB1), ILT4 (CD85d/LILRB2), and KIR2DL4 (CD158d).3,10 The expression of the HLA-G gene is regulated by epigenetic mechanisms at transcriptional and posttranscriptional levels. The varied immunomodulatory effects of HLA-G, including detrimental effects in tumors and viral infections but beneficial effects for embryo implantation and graft survival, suggest that HLA-G expression is precisely regulated.13
The 14-bp-insertion/deletion polymorphism at the 3’ untranslated region has an important role on the regulation of gene expression because of its role on HLA-G alternative splicing and mRNA stability.14 An additional splicing step occurs for transcripts with the 14-bp sequence that removes 92 bases encompassing the region where this sequence is located. This deletion increases mRNA stability.14 Therefore, homozygous +14/+14 individuals may have higher HLA-G expression because of the more stable transcript products. However, the mRNA stability conferred by the 14-bp insertion may exert a negative effect on HLA-G expression, and the 14-bp insertion is associated with lower soluble HLA-G levels.15 The 14-bp insertion also is associated with the generation of a more stable mRNA, but these transcripts have minimal effects on overall HLA-G expression. Conversely, decreased protein expression is associated with the insertion allele. The paradox between the stability of the transcripts originated by the +14 alleles and the observed low HLA-G levels associated with this variant in vivo has been named “the 14-bp polymorphism paradox.”13 This paradox may be caused by other unknown characteristics of this polymorphism or other nearby polymorphisms that may be in linkage disequilibrium with the 14-bp polymorphism.
In the present meta-analysis, we pooled data from 5 eligible published studies including 907 kidney transplant recipients and 887 healthy control subjects, and all included studies had good quality (Table 2). The current meta-analysis did not show an association between the 14-bp-insertion/deletion polymorphism and allograft rejection risk in kidney transplant recipients. The ORs of the 14-bp allele and 5 different genotype models showed that the 14-bp allele and genotype did not correlate with risk of developing kidney graft rejection. When stratified by acute and chronic rejection, a significant association was found in recipients who had chronic rejection (Table 4). The allele contrast, heterozygous, and dominant genetic models had a protective role in allograft kidney chronic rejection (Table 4). However, the validity of this conclusion may be limited because only 2 studies were included in the analysis.7,9 Furthermore, the 3 genetic models that had significant association underscored the role of the +14 allele, which has a negative effect on HLA-G expression. Considering the immunosuppressive role of the HLA-G molecule, it is reasonable to conclude that the 14-bp insertion may be a risk factor for allograft rejection. This is inconsistent with the finding that the 14-bp insertion was a protective factor. Therefore, the conclusion of the meta-analysis about chronic rejection may be a false positive result. The 14-bp polymorphism may have conflicting and complex effects on HLA-G expression, and we could not use 14-bp genotype or allele to predict the risk of graft rejection.
Limitations of this meta-analysis included the small number of studies in this analysis, and the pooled sample size was small and possibly insufficient for a valid conclusion. These case-control studies were performed in different populations, and the limited number of studies did not enable stratified analysis by ethnicity, which usually is an important factor in genetic studies. In addition, there was heterogeneity between the included studies as shown by I2% (Table 4). Furthermore, the HLA-G gene may contain many more gene polymorphisms other than the 14-bp polymorphism that may affect the findings. The same polymorphism may function differently in patients who have different genetic backgrounds. Therefore, the results of the present meta-analysis should be interpreted with caution, and additional studies in patients with different ethnicity are justified with larger sample size and better study design.
In summary, the present meta-analysis suggests that the 14-bp polymorphisms are not significantly associated with allograft kidney rejection. The 14-bp polymorphisms have conflicting and complex effects on HLA-G expression, and we recommend not using 14-bp genotype or allele to predict the risk of graft rejection until further study is performed.
Volume : 12
Issue : 2
Pages : 89 - 94
DOI : 10.6002/ect.2013.0175
From the Department of Urology, Peking Union Medical College Hospital,
Chinese Academy of Medical Sciences, and Peking Union Medical College, Beijing,
Acknowledgements: The authors have no conflicts of interest to disclose. Great thanks go to Yuan Wu for his moral support and technical assistance in the preparation of this meta-analysis. And we would thank all authors of the original studies included in this meta-analysis, especially Daniela Piancatelli and Roberto Littera, with whom we have contacted to obtain further data materials.
Corresponding author: Zhigang Ji, Department of Urology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, 1#, Shuaifuyuan, Dongcheng District, Beijing, 100730, China
Phone: +86 10 6915 6031
Fax: +86 10 6915 6030
Table 1. Main Characteristics of Studies Included in the Meta-Analysis of Human Leukocyte Antigen-G-14-Base-Pair-Insertion/Deletion Polymorphism and Graft Survival in Kidney Transplant Recipients*
Table 2. The Newcastle-Ottawa Quality Assessment Scale for Studies Included in the Meta-Analysis
Figure 1. Flow Diagram of the Study Selection For the Meta-Analysis
Table 3. The Newcastle-Ottawa Quality Assessment Scale for Studies Included in the Meta-Analysis
Figure 2. Forest Plot For the Meta-Analysis of the Association Between the Recessive Genetic Model of the Human Leukocyte Antigen-G-14-Base-Pair Polymorphism and Allograft Kidney Rejection (+14/+14 vs [-14/-14 and -14/+14])
Figure 3. Funnel Plot for Publication Bias in Selection of Studies About Human Leukocyte Antigen-G-14-Base-Pair-Insertion/Deletion Polymorphism (+14/+14 vs [-14/-14 and -14/+14])
Table 4. Pooled Odds Ratios and Stratification Analysis of the Human Leukocyte Antigen-G-14-Base-Pair-Insertion/Deletion Polymorphism and Graft Survival in Kidney Transplant Recipients