Genetic Polymorphisms and Recurrent Spontaneous Abortions: An Overview of Current Knowledge

Authors


Correspondence

Silvia Daher, PhD, Department of Obstetrics, Sao Paulo Federal University, Rua Bela Cintra 1920/41- CEP: 01415-002, São Paulo, SP, Brazil. E-mail: silviadaher@hotmail.com

Abstract

The relevance of gene polymorphisms in the development of unexplained recurrent spontaneous abortion is still unclear. Cytokines, angiogenic mediators, and hormones are involved in all stages of reproduction and pregnancy outcome. Impaired production and/or unbalanced ratios of these mediators have been implicated in the pathogenesis of unexplained recurrent spontaneous abortion. Functional polymorphism influence gene activity and therefore can interfere with the expression of mediators. Several studies have been carried out to evaluate the relationship between cytokines, angiogenic mediators, and hormones gene polymorphisms and unexplained recurrent spontaneous abortion. The results of these studies are mostly contradictory, and few significant associations have been identified. Up to present time, the evidence is insufficient to support the evaluation of cytokines, angiogenic mediators, and hormones gene polymorphism in routine workup in all cases of recurrent pregnancy loss, and these tests are not included in any of the major obstetric guidelines.

Introduction

Recurrent spontaneous abortion (RSA) has been associated with multiple causes, but in almost half of the cases, etiologies remain unknown. It is defined as the loss of three or more consecutive pregnancies prior to the 20th week of gestation and affects 1% of couples.[1, 2] Several studies suggest that gene polymorphisms encoding for different mediators may represent a susceptibility factor for unexplained RSA.

Evidence for genetic contributions to obstetric disorders comes from candidate gene-based association studies, in which investigators have evaluated particular polymorphic variants of genes selected primarily because of their potential roles in pregnancy.[3-7] Most of the studied polymorphisms are considered functional, because they have been proven to influence the expression of mediators defining different phenotypes.

Cytokines, hormones, and angiogenic mediators play critical roles in reproductive events. Although they are necessary for normal pregnancy development, depending on the conditions such as quantity, stage of gestation, and locality of expression, these factors may affect trophoblast–endometrial interaction leading to RSA or other pregnancy complication.[8-10] A broad spectrum of single nucleotide polymorphisms (SNP) located in genes coding for these mediators have been recognized.[4, 6, 7, 11] Although the contribution of these genetic markers to RSA has been extensively investigated, their exact role remains unclear.

The aim of this study is to review previous reports on the relationship between RSA and genetic polymorphism related to cytokines, hormones, and angiogenic mediators. This review does not intend to map in detail all the studies on this topic, but to highlight and synthesize the main findings up to date and indicate areas that could benefit from further or improved investigation.

Method

A Medline search was performed for studies published in the last 10 years using the following general keywords ‘genetic polymorphism’ or synonyms, and ‘habitual abortion’ or synonyms. The retrieved references were selected based on their relevance. We searched for primary studies or systematic reviews that focused on the relationship between RSA and gene polymorphism related to mediators involved in pregnancy development (cytokines, angiogenic factors, and hormones). We assessed the value of these genetic variants for the screening of women with unexplained RSA and the quality of the existing evidence.

Cytokine gene polymorphisms and RSA

A complex network of factors and mechanisms is involved in maternal–fetal immune interaction to insure the adequate development of pregnancy. Cytokines, secreted by leukocytes and non-leukocytes cells, influence all steps of reproduction, playing a critical role in pregnancy outcome. The pattern of maternal immune response to trophoblast cells depends on the specific type of mediator that is predominant, on its quantity and on the period of pregnancy. A pro-inflammatory immune response characterized by the expression of interleukin (IL)-1, tumor necrosis factor-alpha (TNF-a), IL-6 and IL-10, along with reduced levels of IL-2, and interferon-gamma (IFN-g) has been detected at the maternal–fetal interface immediately after implantation of normal pregnancies.[12] Other cytokines, such as IL-4, IL-11, IL-12, IL-13, IL-15, IL-16, IL-17, and IL-18, have also been identified in trophoblast during the first trimester of healthy pregnancies.[9, 10]

An anti-inflammatory (T helper (Th) 2-type) immunity profile is apparently essential for a successful pregnancy outcome, whereas a dominant inflammatory (Th1-type) pattern has been associated with poor gestational outcomes. Although necessary for normal embryo implantation, IFN-g and TNF-a have also been linked to negative gestational outcomes. This abortive effect has been attributed to deficient trophoblast invasion and placentation, and also to excessive activation of uterine natural killer (NK) cells and macrophages. Interleukin-4 and IL-10, important anti-inflammatory cytokines during the period of embryonic development, apparently act by inhibiting Th-1 cells and macrophage function.[8, 13]

In addition to unbalanced Th1/Th2 ratio, RSA patients also have decreased number and/or function of T-reg cells, or even expansion of Th-17 leading to increased secretion of inflammatory mediators.[14] Interleukin-6, a Th2 cytokine, induces the differentiation of Th-17 and, thus, may contribute to pregnancy loss. Indeed, patients with RSA usually have increased IL-6 and soluble IL-6 receptor serum levels.[10, 11, 15]

A number of functional cytokine gene polymorphisms have been identified,[11, 16] and their relationship with RSA has been extensively investigated. However, because of the controversial findings in many of these studies, this association needs to be further explored.

Polymorphisms in TNF-a gene have been extensively investigated in cases of unexplained RSA. While the majority of these studies evaluated SNP at the −308 position, some assessed other functional TNF-a polymorphisms, such as −238 and −863.[17-19] Besides genotype, some investigators also evaluated the expression of TNF-a at the maternal–fetal unit and/or in peripheral blood in patients with RSA. Despite some conflicting results, most reports did not detect a significant association between any of these TNF-a polymorphisms and RSA. Moreover, a meta-analysis suggests that if there is a relationship between RSA and TNF-a (−308), it seems to be very weak.[16] A recent study involving seven TNF-a SNPs identified an association between the TNF-a (−1031T/C) variant and increased risk for RSA.[19] However, because this is the first study to investigate this particular SNP in women with RSA, these results need to be confirmed before being accepted as conclusive.

Interleukin-1 is an inflammatory mediator involved in implantation, trophoblastic invasion, and also in the process of maternal–fetal immune tolerance.[20] However, after this early stage of pregnancy, increased expression of IL-1 and its receptor (IL-1R) seems to have a deleterious effect and has been associated with RSA.[10] Different polymorphisms in genes coding for IL-1 and IL-1R have also been evaluated in RSA patients, and almost all studies failed to detect any relationship between these genetic markers and pregnancy loss.[16] Only two studies reported a positive relation, but each one with a different SNP.[21, 22]

Other inflammatory cytokines that may be harmful for pregnancy development are IFN-g, IL-12, and IL-18.[10] On the basis of this observation, different investigators have evaluated allelic variants on genes encoding for these mediators in patients with RSA.[11, 16] One study reported an association between RSA and IL-18 rs360717,[23] and another one suggested a relationship with IFN-g (+874T).[18] Additionally, this specific SNP has been associated with increased production of IFN-g and RSA patients seem to have increased levels of this cytokine.[18, 24] However, there are very few studies on polymorphisms encoding for these cytokines (IFN-g, IL-12, and IL-18), and their results are contradictory.[11, 18, 23, 25] Thus, it is too early to reach any definitive conclusion on this subject.

The most representative Th2 cytokine is IL-4, and it has been extensively investigated in various pregnancy complications because of its intense anti-inflammatory effect.[10, 24] An unbalanced ratio between TNF-a or IFN-g and IL-4 has been observed in T cells of RSA patients. The two IL-4 polymorphisms that have been evaluated in RSA were not associated with increased risk of abortion.[26, 27]

The role of IL-6 in pregnancy outcome remains unclear. Nevertheless, the anti-inflammatory properties of this Th2-type cytokine are well recognized. Some women with RSA may have impaired IL-6 expression, both locally as well as in peripheral blood.[10, 28] In contrast, increased levels of this mediator have been associated with pregnancy complications such as pre-term birth.[29] Almost all of the studies that evaluated IL-6 polymorphism in RSA patients assessed the SNP at position −174C, and all of them failed to demonstrate an association between this specific polymorphism and RSA.[17, 18, 30] A single study evaluated the IL-6 variant −634 in Japanese patients and found that the frequency of the G allele was significantly lower in RSA patients.[31] This result needs to be confirmed, especially because the genotype frequency of these IL-6 polymorphisms varies in different ethnic populations.[32, 33]

Interleukin-10 interacts with different factors and cells playing a central role in pregnancy. Produced both by Th2 and Treg cells, IL-10 exerts an inhibitory effect on Th1 cells and modulates maternal immune response.[10] The production of IL-10 varies according to specific alleles at position −1082, −592, or −819.[34, 35] On the basis of these findings, it has been proposed that certain IL-10 genotypes might be associated with increased risk for RSA or, the opposite, could represent a protective factor for miscarriage. Among these variants, the most investigated is located at position (−1082A). Although most of the individual studies did not find an association between any of these IL-10 polymorphisms and RSA,[18, 26] a meta-analysis suggested a possible relationship between IL10 GG (−1082) and RSA patients in comparison with the general pregnant population.[17] Besides genotype controversies, the role of IL-10 in abortion is still unclear. Although it is considered as an important modulator of maternal inflammatory response, increased levels of IL-10 have been identified in some cases that end in spontaneous abortion than in those who have a healthy pregnancy.[11, 12]

Few investigators evaluated TNF-b, TGF-b, and IL-21 polymorphisms in women with RSA, and none of the existing studies detected an association between genetic variants and pregnancy loss.[18, 26, 36, 37]

Angiogenic gene polymorphisms and RSA

Angiogenesis and vasculogenesis play essential roles in embryonic, fetal, and placental development. One of the most important mediators involved in these processes is the vascular endothelial growth factor (VEGF). It has an important effect on angiogenesis as well as on vascular permeability, and it also acts as an antiapoptotic molecule.[38, 39] Owing to these multiple roles, VEGF has a critical influence in all stages of pregnancy and may therefore interfere with its development.[40] Deficient VEGF expression in early pregnancy has been associated with spontaneous abortion.[38, 40, 41] This finding has led to several studies on VEGF gene polymorphisms in cases of RSA.[38, 39, 42, 43] VEGF gene variants at positions −2578, −634, and +936 have been studied in women with RSA, but each of these variants has been analyzed in only one or two studies with discrepant results.[38, 43-46] In contrast, different reports suggested an association between (−1154G/A) and RSA, and also with VEGF serum levels.[41, 42, 47, 48] Up to the moment, only two published study did not confirm this finding.[45, 46]

Vascular endothelial growth factor is a well-known mediator of nitric oxide (NO) release and endothelial nitric oxide synthetase (eNOS) activity. NO is a critical modulator of vascular tone and blood flow. As eNOS is required for NO production, low eNOS levels may impair placental perfusion leading to pregnancy loss. A 27 bp of the variable nucleotide tandem repeat (VNTR) polymorphism in intron 4 and the Gly298Asp polymorphism in exon 7 seem to affect NO production and have been associated with different disorders of pregnancy.[39]

A recent study suggested that 4 VNTR polymorphism is associated with RSA,[49] and other investigators have found similar results.[50] However, most existing studies and a recent systematic review did not confirm this association.[39, 51, 52] On the other hand, another systematic review indicates that the Gly298Asp polymorphism seems to be correlated with RSA.[39] Besides these variants, a limited number of studies have investigated the possible association between RSA and other polymorphisms of the eNOS gene.[49, 53] Thus, a larger number of patients need to be assessed for a conclusive result.

The tumor suppressor protein 53 (p53) is another pleiotropic factor involved in the vasculogenesis process and cell apoptosis and is also a potential mediator of pregnancy with estrogen and progesterone activities.[39, 54] A common p53 polymorphism at codon 72, encoding either proline or arginine was shown to interfere with p53 activity and affect reproductive events.[54] This has led to extensive investigations on the relationship between this p53 polymorphism and RSA.[54-56] A meta-analysis of these studies suggests that the risk of RSA was higher in the carriers of the p53 (codon 72) polymorphism.[39]

Hormone polymorphisms and RSA

The role of hormones in reproductive events has been extensively investigated. Their effects are mediated through specific receptors. Progesterone is the hormone most clearly associated with maternal adaptation to pregnancy. It is particularly involved in early stages of gestation, and it is well known that reduced levels of progesterone may lead to RSA.[2, 57, 58] Different functional polymorphisms in gene encoding progesterone receptor (PROGINS) have been described and investigated in pregnant women. Although two studies suggest a relationship between RSA and PROGINS variants,[59, 60] other reports including a meta-analyses did not confirm this finding.[43, 61, 62]

Estrogen is also important for reproductive success because it influences endometrial development as well as placental function, maturation, and perfusion.[63, 64] These effects depend on the interaction of estrogen with its receptors, which are expressed throughout the female reproductive tract.[62] Although several polymorphisms encoding estrogen receptors have been evaluated in RSA patients, each of them has been analyzed in only two or three studies.[62, 65, 66] Thus, this relationship needs to be further investigated to analyze the predictive value of these polymorphisms for RSA.

Androgen is also recognized as an essential factor for successful reproduction.[67] At the molecular level, androgenic effects are mediated through the activation of androgen receptor (AR). This receptor is encoded by the AR gene located on chromosome X. A polymorphic CAG repeat sequence (CAG) in the AR gene has been associated with several clinical disorders.[68, 69] Most of the RSA genetic studies in this field focus on the investigation of X chromosome inactivation (XCI) determined by CAG polymorphism. Inactivation of either the maternal or paternal X chromosome leads to skewed XCI. Increases in skewed XCI have been investigated in patients with RSA with controversial results.[68, 70, 71] Although most of the studies suggest that patients with higher percentages of skewed XCI also have a higher risk for RSA, a small number of publications did not confirm this association.[72, 73]

Main implications for practice

This review presented an overview of the published studies of polymorphisms in genes coding for different mediators involved in the pathogenesis of RSA. Although most of these studies investigated functional polymorphisms, relatively few of them analyzed the relationship between genotype and phenotype. The vast majority on genetic studies involving patients with RSA focus on cytokine gene polymorphisms. Both inflammatory and anti-inflammatory mediators gene variants have been investigated with most studies focusing especially on TNF-a, IL10, and IL-6 SNPs. However, the results of these studies are controversial. The large number on TNF-a gene polymorphisms in women with RSA indicates that there is no association between these SNPs and this disorder. Despite a few positive associations in some studies, these have not been confirmed in others. Additionally, when the correction factor for multiple testing is applied, these associations frequently disappear.

Several previous reviews and meta-analyses failed to identify any significant associations between RSA and cytokine gene polymorphisms. On the basis of the aforementioned findings, there is no evidence to support the inclusion of exams to test for these polymorphisms in all women with unexplained RSA.

Angiogenic mediators gene polymorphisms seem to be more promising in terms of practical implications. Despite significant controversies, a recent metanalysis detected a positive association between RSA and gene polymorphisms coding for the most relevant mediators of angiogenesis: VEGF (−1154 in genes), P53 (codon 72), and eNOS (Glu298Asp). However, the risk for RSA was only slightly higher in women carrying these SNPs.

Among the hormones gene polymorphisms, only skewed XCI was significantly associated with RSA. The relationship between estrogen or progesterone polymorphism and RSA needs to be further investigated to reach definite conclusions.

In accordance with our findings, the most recent guidelines produced by reputed panels of experts[74-77] state that, up to the present, there is insufficient evidence to support the inclusion of the evaluation of cytokines, angiogenic mediators, and hormones cytokines gene polymorphisms in the routine workup of all women with unexplained RSA.

Implications for research

Taking together, the relationship between RSA and genetic polymorphism remains unclear, and therefore, there is a need for more studies in this area. However, the multiple causes implicated in RSA makes the recognition of its essential mediators very difficult. This obstacle impairs the selection of gene polymorphisms to be tested. Moreover, RSA seems to be a syndrome and not a disease. This implies that it is probably caused by multiple factors, and therefore, RSA might be related to various polymorphisms. The problem is that, up to the present, most of the studies have evaluated the effect of single SNPs, and not of different SNPs combinations.

In addition, genetic studies are complex, and their results need to be confirmed in different settings. However, some studies are so different that they cannot be repeated or compared. Ethnic difference may interfere with genotype frequencies and influence the results. An example is the different distribution of IL-6 (−174) genotypes in different ethnic groups. Therefore, one SNP may be associated with RSA in a specific population, but not in a different population.

A major problem of most studies published up to the present is their small sample size. As RSA occurs in approximately 0.5–1% of all women, it is difficult for a single center to recruit a representative number of RSA patients. The limited number of cases included in most studies reduces their statistical power often leading to non-significant results. On the other hand, meta-analyses can increase the precision of these estimates by combining similar studies and pooling their results. But to do this, authors of systematic reviews need to have access to complete information from primary studies, which unfortunately is often unavailable in the existing publications.

A critical issue in genetic polymorphism studies is the detailed characterization of the study group. Many of the existing studies provide no clear definition of RSA, very limited information of participant characteristics, and even fewer details on the controls. All these points could and should be improved in future studies focused on assessing the association between RSA and genetic polymorphisms.

Ancillary