CTLA-4 Gene Polymorphism in Women with Idiopathic Recurrent Pregnancy Loss

Cytotoxic T lymphocyte associated antigen-4 (CTLA-4) is considered as a negative regulator of T cell activation and its role in maintaining immune tolerance is well established. The present case-control study aimed to investigate the CTLA-4 +49 A/G, -1661 A/G, -318 C/T and -1722 T/C single nucleotide polymorphisms (SNPs) and predisposition to recurrent pregnancy loss (RPL) in Gaza Strip Palestine. The study was performed on 200 women with a history of two or more pregnancy losses (case group) and 200 control women with at least two live births and without any previous history of abortion. PCR-based restriction fragment length polymorphism (RFLP-PCR) method was used for genotyping CTLA-4 polymorphisms. Study results revealed that there is no significant association between the allele/genotype frequencies of the four investigated CTLA-4 SNPs and RPL. This trend remained true under dominant, co-dominant and recessive models. The A\G genotype of -1661 A\G polymorphism was higher in patients (45%) as compared to controls (39.5%) but without statistical significance. The minor allele frequencies (MAFs) of the CTLA-4 gene polymorphisms in the patient/control group were as follows: +49A>G: 0.22/0.22, -318 C>T: 0.15/0.11, -1661 A>G: 0.30/0.26 and -1722 T>C: 0.08/0.08. The four investigated CTLA-4 polymorphisms do not contribute to the risk of RPL in the study population. Testing other CTLA-4 gene polymorphisms and the level of CTLA-4 expression in RPL patients is recommended.


Introduction
Recurrent pregnancy loss, the occurrence of 2 or more consecutive miscarriages, is one of the most common pregnancy complications with a prevalence of 1-2% among pregnant women in reproductive life [1].
Despite years of effort to determine the factors involved in miscarriage, the cause remains unclear in around 50% of the cases hence, this reflects the heterogeneous nature of this malady [2].
Several mechanisms have been proposed to function actively in the protection of the semi-allogeneic fetus from maternal immune system. The presence of regulatory T cells (Tregs) and the expression of immunomodulatory molecules at the fetal maternal interface have been identified as crucial factors for fetomaternal tolerance [3].
Cytotoxic T-Lymphocyte Antigen 4 (CTLA-4) is constitutively expressed in Foxp3+ Tregs and induced in conventional T cells following activation [4]. CTLA-4 has a suppressive effect on T cell activation and might be involved in establishing immune tolerance by blocking CD28dependent T cell activation through interactions with its ligand CD80/86 on antigen presenting cells in the decidua. The CTLA-4/B7 complex can compete with the CD28/B7 complex and convey an inhibitory influence to the T cell affecting T cell development and functions [5,6]. CTLA-4 dysregulation, therefore, has the potential to affect fetal tolerance through altered activation of T cells to fetal antigens. CTLA-4 gene is located on the chromosomal region 2q33 and contains more than 100 polymorphic sites. Distinct single nucleotide polymorphisms (SNPs) such as +49A/G (rs231775), -318C/T (rs5742909), -1661A/G (rs4553808) and -1722T/C (rs733618) have been associated with autoimmune diseases and preeclampsia [7]. CTLA-4 expression in placental fibroblasts and deciduas supports its role in the maintenance of pregnancy and fetomaternal tolerance [1]. Therefore, it is suggested that abnormal expression of CTLA-4, effected by genetic polymorphism(s), may be associated with RPL. Indeed, some CTLA-4 single nucleotide polymorphisms (SNPs) have been implicated as potential risk factors for RPL in certain populations [8,9,10].
The present study was designed in order to investigate the association between four CTLA-4 gene polymorphisms namely, +49A/G, -318C/T, -1661A/G and -1722T/C and RPL in a group of Palestinian women.

Study Subjects
The study group (n=200) included women aged 20-35 years who had experienced at least two unexplained spontaneous abortions before 20th week of gestation. The control group (n=200) consisted of women who had delivered at least one healthy child and had no previous history of pregnancy loss. Controls were matched with study subjects for all other possible characteristics. None of the individuals included in the study population used oral contraceptives, hormonal, or any serious medication affecting body vital functions. Individuals with known causes of RPL were excluded from the study group.

Genotyping
DNA extraction and polymorphism determination About 2.0 ml of venous blood were drawn into sterile EDTA tubes under quality control and safety procedures. Genomic DNA was isolated from blood using Wizard Genomic DNA Purification Kit (Promega, USA) following the manufacturer instructions.
The four SNPs were genotyped using PCR-RFLP technique. The specific PCR primers were designed using online Primer3 software (http://primer3.ut.ee/) based on the genomic sequence deposited in gene bank and the sequence of each SNP was retrieved from NCBI-SNP database (http://www.ncbi.nlm.nih.gov/snp/). Then restriction enzymes required for the PCR-RFLP identification of each SNP were selected from new England Biolabs database by using NEB cutter software (http://nc2.neb.com/NEB Cutter2/) ( Table 1).
PCR primers and conditions, restriction enzyme digestion and results interpretation were done as indicated in (Tables 1  & 2).

Ethical Considerations
Informed consent was taken from all the subjects who participated in the study. The objective of the study was fully explained to all participants and their consent was taken.

Statistical Analyses
The genotype, allele frequency in RPL patients and the controls were analyzed by standard Chi-square test and odds ratio (OR) for risk of RPL at 95% confidence intervals (CI). Hardy-Weinberg equilibrium (HWE) was tested in the control group using a freely available software (http://www. oege. org/software/hwe-mr-calc.shtml). P-values of 0.05 or less were regarded as statistically significant. Table 3 illustrates genotypes and alleles frequencies, odds ratio, 95% confidence intervals and P values for the four tested CTLA-4 polymorphisms among RPL patients and controls.

Results
Statistical analyses of genotypic and allelic frequencies for the tested SNPs revealed no significant (all P values are > 0.05) difference between RPL patients and controls.
Moreover, statistical analyses of the four SNPs genotypes under recessive, dominant, and co-dominant models indicated no significant difference between the two study groups.

Hardy-Weinberg equilibrium for investigated SNPs
The distribution of the genotypes of -318C\T, -1661 A\G and -1722 T\C SNPs conformed with Hardy-Weinberg equilibrium as there was no significant difference between the expected and the observed genotypes. CTLA-4 +49A\G genotypes, however, showed a significant deviation from Hardy-Weinberg equilibrium with a P-value =0.0001.

Discussion
Idiopathic RPL is a heterogeneous condition and its complexity might be due to the additive effect of several genes, and their interactions with each other and with environmental factors. From the immunology of pregnancy standpoint, major efforts are concerned with the role of sequence variants (e.g., SNPs) of immune tolerance related genes (e.g., CTLA-4) in the etiology of RPL.
There have been several studies that couple CTLA-4 deficiency to adverse pregnancy outcomes such as recurrent pregnancy loss (RPL), placental abrubtion, and pre-eclampsia (PE). Moreover, polymorphisms in the CTLA-4 gene have been associated with dysregulated CTLA-4 production and function. For instance, heterozygosity of the CTLA-4 A49G allele, has been reported as a predisposing factor to severe PE and placental abruption in some populations [8,11]. Moreover, stretches of AT repeats in the 3'-untranslated region of the CTLA-4 gene has been suggested to affect mRNA stability and fetal survival in humans [12].
The four polymorphisms targeted in this study affect CTLA-4 expression. The exon 1 +49A/G leads to less efficient glycosylation and reduced expression of membrane CTLA4 [13,14,16]. The -318 C/T, -1661 A/G and -1722 T/C polymorphisms lie in the promoter region and affect the level of transcription of CTLA-4 [15,16,7]. Reduced CTLA-4 expression and/or change in CTLA-4 activity result in uncontrolled T-cell regulation.
Results of the present work showed that no statistically significant differences were evident between RPL cases and controls in terms of the allelic and genotypic distributions of CTLA-4 "-318 C/T, -1661 A\G, -1722 T\C and +49A\G" polymorphisms.
Interestingly, the homozygous "GG" genotype was not observed in the present study. The absence of this genotype in our population may be a consequence of selection against this genotype. This also explains the observed deviation of this SNP genotypes from Hardy-Weinberg equilibrium.
As for CTLA-4 "-1661 A/G" SNP, although the difference was not significant between the control women and the RPL patients, the A/G genotype of this SNP was higher in the RPL patients relative to the control group (45%, 39.5%, respectively). In harmony with our results, (Chaili, 2010) showed that CTLA-4 "-1661 A/G" polymorphism is not associated with RPL [17].
For, CTLA-4 "-1722 T/C" SNP, the comparison of genotype and allele frequencies in the case and control groups did not show significant differences. Similar results were reported by (Chaili, 2010) who also showed a lack of association between CTLA-4 "-1722 T\C " polymorphism and RPL [17]. Moreover, the CC homozygote genotype of this SNP was not encountered in any of the study subjects. The explanation for the absence of the CC genotype is mainly due to the uncommon occurrence of the C-allele (MAF ~ 0.08) in our population.
Contradictory results are a common place in genetic association studies performed on different populations. Possible explanations for discrepant results include one or more of the following: differences in the ethnicity (genetic background), the sample size, the inclusion and exclusion criteria of the study subjects, presence of nucleotide polymorphism(s) somewhere else in the examined gene, epigenetic alterations, linkage disequilibrium to other sequence variants in the vicinity of the studied locus, and prevailing environmental conditions.
Lack of association of the investigated SNPs does not exclude the importance of CTLA-4 in RPL. Research is ongoing to examine other CTLA-4 polymorphisms and correlate those polymorphisms with the level of CTLA-4 in RPL women.