Introduction
Recurrent pregnancy losses (RPL) due to recurrent miscarriage (RM) are a significant reproductive problem affecting approximately 2% of women in the general population.1 RM is a common complication of pregnancy, accounting for approximately 10–15% of pregnancy losses, with nearly 80% occurring within the first trimester of gestation.2 The latest consensus also defines RPL as the occurrence of two or more successive losses of pregnancy before 28 weeks of gestation.3 Globally, RPL is recognized as a significant reproductive health concern, with more than 50% of couples experiencing RPL having no apparent known cause.4 The intricate regulation between the fetus and the maternal body, involving numerous mediators such as cytokines, histocompatibility antigens, hormones, lifestyle factors, genetic factors, and angiogenic factors,2,5–8 is crucial for viable pregnancy. Genetic polymorphisms that impact maternal and fetal factors are believed to be associated with the risk of RPL.
HLA-G, a non-classical HLA class I antigen, is expressed by invasive cytotrophoblasts and likely plays an essential role in generating a tolerogenic state at the feto-maternal interface.9HLA-G influences maternal acceptance of the semi-allogenic fetus by adapting the maternal immune system during pregnancy, making it an important immune-tolerant factor that inhibits immune cell functions such as natural killer cells, T-lymphocytes, and dendritic cells.10–13HLA-G transcripts are significantly found in placental tissue during the first trimester, predominantly in the extravillous membranes, validating the theory that HLA-G plays an important role in fetal protection.14,15Additionally, HLA-G is thought to be involved in pregnancy complications such as preeclampsia and RPL.16–19
HLA-G gene, present on the short arm of chromosome 6, contains various polymorphic sites in both coding and non-coding regions that potentially influence its biological functions, including the modulation of the immune response.20 Alternative splicing of the primary HLA-G transcript yields seven HLA-G isoforms, with three soluble isoforms (HLA-G 5, -G 6, and -G 7) present in large quantities in the maternal circulation during gestation.15 Studies have shown that plasma levels of soluble HLA-G (s HLA-G) secreted by a class of immune cells are lower in early miscarriage compared to normal pregnancy, indicating that s HLA-G molecules are likely to play a vital role in embryo implantation.21,22 In pregnancies with in vitro fertilization, low levels of s HLA-G in maternal circulation concurrent with unfavorable pregnancy outcomes, compared to embryos that secrete sufficient s HLA-G, result in successful pregnancy.18 Several investigations on HLA-G gene polymorphisms have revealed relationships between different HLA-G alleles and RPL.23,24 The presence of HLA-G*0104 and HLA-G*0105N alleles in either member of the couple has been strongly associated with an elevated risk of RPL.25 Furthermore, evidence from a few investigations has established the potential of HLA-G as a possible target for the therapeutic function.26,27
Several studies have substantiated that elevated serum HLA-G levels are related to decreased risk of RPL.28,29 Our observational study revealed a high prevalence of RM in the Kashmiri population, particularly in the younger age group. Our population strongly favors the custom of consanguineous marriages, which may lead to an increased prevalence of severe genetic disorders, including RPL, over multiple generations. Given the credible role of HLA-G in pregnancy outcomes, the present study was conducted to explore the association between HLA-G genotype and the patients who suffered RPL compared to normal healthy fertile women with successful full-term pregnancies.
Materials and methods
The study enrolled 200 females who experienced ≥2 pregnancy losses due to RM, along with their corresponding male partners for HLA-G (*0103,*0104,*0105N). Additionally, 60 tissue samples from products of conception (POC: abortuses) were examined for HLA-G polymorphism. The study was conducted between 2014 and 2019 at the Advanced Centre for Human Genetics, SK Institute of Medical Sciences (SKIMS), J&K (North India), and all patients were referred from the Department of Obstetrics and Gynecology (SKIMS). Utmost care was taken to ensure that the patients strictly fulfilled the fundamental diagnostic criteria of RM, and all groups of RPL patients were enrolled either before or at the 20th week of gestation. A detailed pedigree analysis was performed to identify patients with a multigenerational history of RPL. The patients’ blood reports, including hormonal, immunological, biochemical profiles, and radiological investigations, were recorded. Patients were recruited after obtaining their written informed consent, and the study was approved by the Institutional Ethics Committee of SKIMS under protocol No. SKIMS Study ref:81/2014. To exclude other causes of recurrent abortion, patient details were recorded for radiological imaging, cytogenetic reports, related infection tests, and immunological profiles. A total of 240 women matched for age and geographical location with at least two live births and without a previous history of abortion were enrolled as the control group. Approximately 2 ml to 3 ml of venous blood from both groups was collected in Ethylenediamine tetraacetic acid (EDTA) vials and stored at −80°C until further processing for DNA extraction. Using nMaster 2.0 statistical software, the sample size of the study was calculated to be >80%. The clinicopathological characteristics of the patients and controls are shown in Table S1.
Extraction of genomic DNA
DNA extraction from the blood samples of both patients and controls was performed using the phenol-chloroform method and a DNA Extraction Kit (Qiagen).
HLA-G genotyping by PCR-RFLP
Genotyping of HLA-G* 0103, 0104, and 0105N was carried out by PCR-RFLP. Exons 2 and 3 of HLA-G encompassing G*0103, G*0104 and G*0105N alleles were amplified by PCR using the following primers (Biotools, B & M Labs, Madrid, Spain): Exon 2 Forward (F): 5′-TCCATGAGGTATTTCAGCGC-3′, Reverse (R): 5′-CTGGGCCGGAGTTACTCACT-3′; Exon3: F: 5′-CACACCCTCCAGTGGATGAT-3′R: 5′-GGTACCCGCGCGCTGCAGCA-3′. For Polymerase chain reaction (PCR), the optimal reagents were used as follows: DNA (100–250 ng/mL), 1×PCR buffer (100 mM Tris-HCl, pH 8.3; 500 mM KCl; 15 mM MgCl2), deoxyribonucleotide triphosphate (Biotools, B & M Labs, Madrid, Spain): 10 mM each of dATP, dCTP, dGTP, dTTP) (Sigma-Aldrich, USA): 10 pM in sterile deionized water), and 1 unit of Taq DNA polymerase (Biotools, Madrid, Spain). The PCR thermal conditions for HLA-G exon 2 and 3 included 35 cycles: 94°C for 30 sec, annealing at 57°C for exon 2 and 61°C for exon 3 for 30 sec, and 72°C for 30 sec with a final extension at 72°C for 7 minutes. Amplified products (10µL) from exon 2 and exon 3 were digested with the restriction endonuclease Hinf-I for exon 2 and exon 3 with PpuM-I, and BseR-I (Fig. S1). The interpretation of various genotypes is provided in Table S2.
Statistical analysis
All the statistical analyses were performed using IBM Statistics SPSS software (Version-23). Both cases and controls were analyzed by the chi-square test for categorical variables, such as gender and age. To assess whether the polymorphisms were in Hardy-Weinberg equilibrium between patients and controls, we applied a goodness-of-fit chi-square test. The relative risk was estimated in terms of odds ratios (ORs) and 95% confidence intervals (CIs) to observe the association between HLA-G genotypes or other related risk factors and RM. Statistical significance was set at p < 0.05.
Results
The cases and controls were frequency-matched with respect to age and family history (p > 0.05). RM cases with <3 abortions accounted for 46% (92) while 54% (108) accounted for ≥3 miscarriage events. RPL cases with consanguinity and family history were seen in 28.5% and 16% respectively (Table S1).
Genotyping of exon 2 genotype 0103:0103
Among the 200 RM patients and 240 healthy controls with full-term pregnancies, the variant genotype of exon 2 of HLA-G, 0103:0103, was absent in our population. However, the frequency of HLA-G genotype 0104/0105 was 100% in both the RPL patients and the control group (Table 1). Additionally, the spouses of RPL patients showed a 100% frequency of the 0104/0105 HLA-G genotype. Furthermore, genotyping of 60 POC samples for HLA-G exon 2 revealed 100% frequency of the 0104/0105 HLA-G genotype, while the variant genotype 0103/0103 was not found in our population (Table 2).
Table 1HLA-G Exon 2 and Exon 3 genotypes observed in Recurrent Abortion cases and healthy controls
| Parameters | HLA exon 2 and 3 genotype | Cases females N 200 (%) | Control 240 (%) | p-value |
|---|
| Overall genotype | HLA-G*0103 | 0 | 0 | Ref |
| All but*0103 | 200 (100) | 240 (100) | - |
| HLA-G*0104:0104 | 13 (7) | 2 (0.8) | Ref |
| HLA-G*0104:0105N | 187 (93) | 238 (99.2) | <0.05 |
| HLA-G*0105N: 0105N | 0 | 0 | - |
| All but G*0105N | 200 (100) | 240 (100) | Ref |
| Allele | 0103 | 0 | 0 | - |
| 0104/0105 | 400 (100 | 480 (100) | Ref |
| G*0105N | 187 | 0 | <0.05 |
| G*0104 | 52 | 14 | - |
| G*0103/0104 | 400 | 480 | |
Table 2Genotypic/allelic distribution of HLA-G Exon 3 gene in RM cases and healthy controls with respect to different clinic-pathological characteristics
| Parameters | HLA-G exon G*0104 genotype | RPL cases
| Control N 240 (%) | p-value |
|---|
| Female N 200 (%) | Male N 100 (%) |
|---|
| Specificity | G*0104N | 26 (13) | 7 (7) | 2 (0.8) | Ref |
| All but*0104N | 174 (87) | 93 (93) | 238 (99.2) | <0.05 |
| Allele | 0104 | 52 (13) | 14 (7) | 4 (0.8) | Ref |
| 0103/0105 | 348 (87) | 186 (93) | 480 (99.2) | <0.05 |
| Age | | | | | |
| <30 | G*0104N | 13 (12) | 3 (9) | 1 (0.8) | Ref |
| All but*0104N | 96 (88) | 29 (91) | 115 (99.2) | <0.05 |
| ≥30 | G*0104N | 13 (14) | 4 (6) | 1 (0.8) | Ref |
| All but*0104N | 78 (86) | 64 (94) | 123 (99.2) | <0.05 |
| Miscarriage | | | | | |
| <3 | G*0104N | 11 (12) | | 2 (1.75) | Ref |
| All but G*0104N | 81 (88) | | 112 (99.2) | <0.05 |
| ≥3 | G*0104N | 15 (14) | | 0 | Ref |
| All but G*0104N | 93 (86) | | 126 (100) | - |
| Family History | | | | | |
| Yes | G*0104N | 0 | | 0 | Ref |
| All but*0104N | 32 (100) | | 41 (100) | 1 |
| No | G*0104N | 25 (15) | | 2 (1.0) | Ref |
| All but*0104N | 145 (85) | | 197 (99) | <0.05 |
| Consanguinity | | | | | |
| | | | | |
| Yes | G*0104N | 9 (16) | | 1 (3.3) | Ref |
| All but*0104N | 48 (84) | | 32 (96.7) | 0.05 |
| No | G*0104N | 17 (12) | | 1 (0.09) | Ref |
| All but*0104N | 126 (88) | | 208 (99.1) | <0.05 |
Genotyping of exon 3 Ppum I restriction site (0105:0105 genotype)
For the Ppum I restriction site of exon 3 of HLA-G, as described in Table 1, the same series of RPL patients and full-term healthy controls underwent polymorphic analysis. Genotype 0105:0105 was not detected in either patients or healthy controls. Conversely, the frequency of HLA-G genotype 0104:0103 was 100% in both groups. Moreover, a similar scenario was found in the spouses of the RPL patients (100 samples), which showed the absence of the 0105:0105 genotype in our population, but the frequency of the other variant genotype 0104:0103 was 100% in the same subjects. Similarly, fetal abortuses exhibited the same pattern, wherein all 60 POC samples examined showed a 100% frequency of the 0104:0103 HLA-G genotype, while the other variant (0105:0105) was completely absent (Table 3).
Table 3Genotypic/allelic distribution of HLA-G Exon 2 gene polymorphism in Recurrent Abortion cases and POC
| Parameters | HLA exon 2 genotype | POC N = 60 | Cases females N = 200 (%) | p-value |
|---|
| Over all genotype | | | | |
| G*0103 | 0 | 0 | Ref |
| All but G*0103 | 60 (100) | 200 (100) | - |
| G*0105N | 0 | 0 | Ref |
| All but*0105N | 60 (100) | 200 (100) | |
| G*0104N | 4 (6.7) | 26 (13.0) | Ref |
| All but*0104N | 56 (93.3) | 173 (86.5) | 0.03 |
| Allele | | | | |
| 0103 | 0 | 0 | Ref |
| 0104/0105 | 120 (100) | 400 (100) | - |
| 0105 | 0 | 0 | Ref |
| 0104/0103 | 120 (100) | 400 (100) | - |
| 0104 | 4 (3) | 52 (13) | Ref |
| 0103/0105 | 116 (97) | 348 (87) | 0.003 |
Genotyping of exon 3 BseRI restriction site (0104:0104 genotype)
As shown in Table 2, for exon 3 restriction site BseRI, the frequency of HLA-G 0104:0104 was 13% in RPL patients vs. 0.8% in controls, whereas the frequency of the HLA-G 0103:0105 variant was 87% in patients compared to 99.2% in controls, with a significant association between the two groups (p<0.00001). Similarly, among spouses belonging to corresponding RPL patients, 93% showed the 0103:0105 HLA-G variant, with a significant difference compared to controls (p<0.05). Interestingly, the control population exhibited a frequency of 99.2% for the 0103:0105 genotype, while the other genotype, 0104:0104 was present in only 0.2% of the patients. In exon 3, all genotypes except G0105N were significantly associated with RPL (100%).
Overall, exon 2 and exon 3 HLA-G variant genotypes G*0103 and G*0105, respectively, did not exist in our population and thus have no role in the etiopathogenesis of RPL. In contrast, the frequency of the exon 3 HLA-G variant G*0104N was significantly greater in RPL patients than in their spouses but was negligible in the control group (p<0.05). The presence of the 13% HLA-G variant genotype G*0104N (exon 3) in RPL patients and 7% in their male partners together was significantly greater frequency than that in controls, which can pose a substantial risk for pregnancy losses (p<0.05). All parameters related to RPL were stratified, and their associations are depicted in Table 2.
Discussion
HLA-G is a distinctive immunomodulator expressed in various tissues, and its polymorphic variations have been associated with a spectrum of diseases, from autoimmune diseases to cancer.30–32HLA-G is exclusively expressed at the maternal-fetal interface in fetal tissues and plays a crucial role in the functional activities of the local maternal immune response. This distinctive expression pattern, particularly in HLA genes such as HLA-G, is believed to be significant for the preservation and maintenance of pregnancy.33–36HLA-G is considered vital in obstetric complications, primarily including RM and preeclampsia.13–16 In this study, we analyzed the role of HLA-G to observe the association between HLA-G genotypes and RPL patients, wherein the HLA-G G*0104N (exon 3) variant genotype was found to pose a substantial risk for pregnancy losses.
The current study identified the HLA-G*0104 allele as a significant predictor of the risk of RPL. The frequency of HLA-G*0104 was 13% in RPL patients compared to 0.8% in healthy controls. This finding aligns with Aldrich et al.,21 who demonstrated a novel relationship between the HLA-G*0104 allele and RPL. Located in the α2-domain, G*01040x alleles are non-synonymous polymorphic variants of HLA-G that have shown an impact on the pregnancy outcome in our study, with an estimated frequency of 13% in females and 7% in their male partners, compared to a negligible presence (0.8%) in healthy controls. However, Matter et al.37 reported a difference in the frequency of HLA-G*0104 between RPL patients and controls (17.5% and 12.5%, respectively) but did not observe significant differences between the two groups. Conversely, some reports have shown no association between RPL and HLA-G*0104 alleles in studies conducted on different ethnic populations worldwide.22HLA-G*0104 allele is found at a frequency of 34.0% in the Korean population and other ethnic groups, such as Japan, China Han, and African Shona.38,39HLA-G*0104 was also evaluated in POC samples in the present study, confirming that the frequency of transmission of this allele to the offspring was 6.7%. Thus, HLA-G*0104 appears to be an important risk factor for RPL in our population compared to other populations.
Our study revealed that the exon 2 genotype of HLA-G, 0103:0103 was not detected in our population, neither in RPL patients nor in the control group. Furthermore, 60 POC samples genotyped for HLA-G*0103 also lacked the 0103/0103 allele. The rarity or negligible occurrence of HLA-G*0103 has been reported in several other populations in the world, with frequencies ranging from 0% to 2%.28 In contrast to these reports, Park et al.40 reported an HLA-G*0103 frequency of ∼24.2% among the Korean population, comprising the main HLA-G allele. Our region (Kashmir, North India) has different ethnic backgrounds and differs completely from the results reported by Park et al.41 and Abbas et al.36 Thus, HLA-G*0103 seems to exhibit considerable genetic variation across different ethnic populations. Additionally, HLA-G*0105N was not detected in either the case series or control group, indicating the absence of this allele in our population, a finding supported by Abbas et al.36
Furthermore, our study revealed HLA-G*0104:0105N alleles implicated in comparable frequencies in both the RPL patient and controls, at 93% and 99.2%, respectively. These alleles were found to be significantly more prevalent in the controls compared to the reference HLA-G*0104, suggesting their non-pathogenic nature in RPL patients in our population. A similar scenario was observed by Matter et al.,37 wherein although the two alleles HLA-G*0104:0105N were more frequent in the RPL group, but they did not achieve statistical significance.35
Inconsistencies have been observed among various studies due to the presence of different HLA-G genotypes and their pathogenic nature with respect to RPL. This variability could be attributed to several factors, including genetic variations unrelated to explored HLA alleles, polymorphic variants positioned elsewhere in the HLA gene, such as in the 5′, 3’UTR or intronic regions, and linkage disequilibrium to different polymorphic variants that lie in close proximity to the HLA locus. In summary, our findings suggest that certain HLA-G alleles have a substantial connection with an increased risk of RPL, and the combination of a few alleles can impact RPL patients.
Conclusions
In conclusion, the presence of the HLA-G*0104 allele at a higher frequency in both partners strongly indicates a significant risk for RPL within our population. We further conclude that there is no role for HLA-G *0103 and *0105 in our population.
Supporting information
Supplementary material for this article is available at https://doi.org/10.14218/GE.2023.00144 .
Table S1
Demographic details of the patients and controls.
(DOCX)
Table S2
Identification of various genotypes in HLA-G.
(DOCX)
Fig. S1
Amplified products (10µL) from exon 2 and exon 3 were digested with the restriction endonuclease Hinf-I for exon2 and exon3 with PpuM-I, and BseR-I. (a) Representative gel picture of HLA-G exon 2 PCR Amplification. Lane M:100bp marker, Lane 1-6:281bp Amplicon; (b) Representative gel picture of HLA-G exon 2 Digestion. Lane 1-4: 175+106bp (wild), Lane M: 25bp Marker; (c) HLA-G exon 3 Digestion by Ppum I, Lanes 1-6: 168+108bp, Lane M: 100 bp DNA Marker; (d) HLA-G exon 3 Digestion by BseR I, Lanes 1,3,4,5:276+236+40bp, Lane 6: 276 bp, Lane M: 25 bp DNA Marker.
(TIF)
Declarations
Acknowledgement
The authors are grateful to the patients for their support of participation in the study. We duly acknowledge the technical support provided by the multidisciplinary research unit (MRU) of SKIMS.
Ethical statement
The study was approved by the local Institutional Ethics Committee SKIMS-Institutional Ethics Committee under protocol No. SKIMS Study ref: 81/2014. Patients were recruited after obtaining written informed consent. The procedures done involving human participants were as per the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments.
Data sharing statement
The data supporting the results of this study are available within the paper and it’s Supplementary Information
Funding
This study was funded by the Department of Biotechnology, India (No. BT/PR11769/MED/97/229/2014).
Conflict of interest
All authors declare no conflict of interests.