The SERPINA4 rs2070777 AA Genotype is Associated with an Increased Risk of Recurrent Miscarriage in a Southern Chinese Population.

BACKGROUND: Many inflammation-related gene polymorphisms are associated with susceptibility to recurrent miscarriage. SERPINA4 is involved in inflammation and is associated with susceptibility to a variety of diseases, but its relevance in recurrent miscarriage is unclear. Therefore, this study aimed to investigate the relationship between SERPINA4 gene polymorphisms and susceptibility to recurrent spontaneous abortion.
METHODS: Two SERPINA4 polymorphisms were genotyped in 631 patients with recurrent miscarriage and 771 controls by TaqMan real-time polymerase chain reaction, and the strength of each association was evaluated through 95% confidence intervals (CIs) and odds ratios (ORs).
RESULTS: The results showed that SERPINA4 rs2070777 AA genotypes were associated with an increased risk of recurrent miscarriage (AA vs AT/TT adjusted OR=1.409, 95% CI=1.032-1.924, P=0.0309), and we also found a significant association between the rs910352 T allele in the SERPINA4 gene and susceptibility to recurrent miscarriage (CT vs CC adjusted OR=1.579, 95% CI=1.252-1.992, P=0.0001; TT vs CC adjusted OR=1.524, 95% CI=1.134-2.049, P=0.0052). The combined analysis of two SNPs of the SERPINA4 gene revealed that carriers with one to two unfavorable genotypes were associated with a higher risk for recurrent miscarriage compared with individuals with no unfavorable genotypes (adjusted OR=1.257, 95% CI=1.019-1.550). Moreover, our study indicates that having one to two unfavorable genotypes is associated with an increased risk of recurrent miscarriage in women 35-40 years of age.
CONCLUSION: Our study suggests that SERPINA4 rs2070777AA genotypes might contribute to an increased risk of recurrent miscarriage in a southern Chinese population.

Introduction

Recurrent miscarriage is a common complication of pregnancy that is defined as two or more consecutive pregnancy losses before 20 weeks of gestation.1 Epidemiological studies have shown that approximately 1–2% of women are affected by recurrent miscarriage, and this complication places considerable mental and psychological burdens on the patients and their family members.2 Many factors contribute to the occurrence of recurrent miscarriage, and these include autoimmune, infectious, endocrine, anatomic and genetic factors.1,2 However, the causes of recurrent miscarriages in approximately 50% of patients are unclear.3 The etiopathogenesis of recurrent miscarriage is uncertain, and increasing numbers of recent studies have suggested that genetic variants might participate in the onset of recurrent miscarriage.4 Infective and immunogenetic factors are thought to contribute to the etiopathology of recurrent miscarriage.5,6 Many inflammation-related gene polymorphisms, such as FOXP3, CLOCK, IL-10, TNF-α and IL-17F, have been associated with recurrent miscarriage.5,7–9 CRP variants are associated with recurrent miscarriage and influence the circulating C-reactive protein levels in chronic inflammatory conditions.10 VEGFA polymorphisms are associated with susceptibility to recurrent miscarriage and are independently associated with the VEGF serum levels, and VEGFA single-nucleotide polymorphisms (SNPs) might contribute to the pathogenesis of recurrent miscarriage.11

Kallistatin is a member of the serine proteinase inhibitor (SERPIN) superfamily that is encoded by the SERPINA4 gene in humans.12 Recent studies have revealed that kallistatin participates in a variety of pathophysiological processes, including the regulation of cancer development, angiogenesis, apoptosis, oxidative stress and inflammation.13–16 For example, Lin WC et al reported that kallistatin might inhibit inflammation and apoptosis to protect against sepsis-related acute lung injury.17 Moreover, kallistatin blocks the VEGF signaling pathway to inhibit angiogenesis.18 Li P et al reported that kallistatin regulates miR-203, miR-21 and miR-34a synthesis in breast cancer cells to result in autophagy and apoptosis.19 Li D et al suggested that miR-34a is associated with unexplained recurrent spontaneous abortion,20 and Magdoud K et al found that plasminogen activator inhibitor type 1 (belonging to the SERPIN family) is associated with an increased risk of recurrent miscarriage.21

The above-mentioned studies suggest that SERPINA4 gene polymorphisms might be associated with abortion. However, few studies have investigated the association between SERPINA4 gene polymorphisms and susceptibility to recurrent miscarriage. In this study, we examined the association between susceptibility to recurrent miscarriage and SERPINA4 gene polymorphisms in a southern Chinese population comprising 631 women with recurrent miscarriage and 771 healthy controls with no history of miscarriages.

Materials and Methods

Study Population

For this study, 771 healthy controls and 631 women diagnosed with recurrent miscarriage were recruited from Guangzhou Women and Children’s Medical Center (Department of Gynecology) between June 2017 and June 2019. The control women had undergone at least two normal pregnancies and had no history of miscarriage. Recurrent miscarriage was defined as two or more unknown etiologies of spontaneous miscarriages (5–20 weeks of gestation). Specimens were collected from the patients and controls during routine blood tests performed at the hospital. None of the patients with recurrent miscarriage or the control women had a history of uterine anatomic abnormalities or embryo chromosomal abnormalities. Accordance with the Declaration of Helsinki, this study was approved by the Ethics Committee of Guangzhou Women and Children’s Medical Center (2018022202). Each participant (patients with recurrent miscarriage and controls) provided written informed consent to participate in the study.

Genotyping and DNA Extraction

Total genomic DNA was extracted from peripheral blood samples collected from participants using the Blood DNA Kit (Tiangen, Beijing, China) following the manufacturer’s specifications. The typing probes (rs910352, [C_9596926_10] and rs2070777 [C_15867824_20]) were purchased from Applied Biosystems (Applied Biosystems TaqMan, Foster City, CA, USA). Genotyping for the two SNPs (rs910352 and rs2070777) was performed in a 384-well plate with an ABI Q6 instrument according to the TaqMan real-time polymerase chain reaction protocol (Applied Biosystems TaqMan, Foster City, CA, USA). PCR amplification was performed in a final volume of 5 μL, which consisted of 0.04 µL of the primers, 2.5 μL of 2× Mix (Tiangen, Beijing, China), 1.46 μL of ddH2O and 2.5 ng of DNA.

Statistical Analysis

The genotypic and demographic differences between the control subjects and the patients with recurrent miscarriage were tested using the two-sided chi-square test. Hardy-Weinberg equilibrium (HWE) was tested by the goodness-of-fit χ2 test. The association between the SERPINA4 gene polymorphism and susceptibility to recurrent miscarriage was described by 95% confidence intervals (CIs) and odds ratios (ORs). A stratified analysis was performed with respect to age and the number of abortions. All statistical analyses were performed using SAS software (version 9.4; SAS Institute, Cary, NC, USA). P < 0.05 was considered to indicate statistical significance.

Results

Population Characteristics

A total of 631 patients with recurrent miscarriage and 771 controls were included in this study (Table 1). No significant difference in age was found between the patients with recurrent miscarriage and the controls (32.39±5.38 vs 32.50±5.20 years old, P=0.6978). Moreover, among the patients with recurrent miscarriage, approximately 59.59% had suffered two to three spontaneous abortions, and 40.41% suffered at least four spontaneous abortions.

Table 1

Frequency Distribution of Selected Characteristics in Patients with Recurrent Miscarriage and Controls

Variables	Cases (n =631)		Controls (n = 771)		Pa
	No.	%	No.	%
Age range, years	20–46		20–49
Mean ± SD	32.39±5.38		32.50±5.20		0.6978
<35	419	66.4	516	66.75
35–40	155	24.56	187	24.19
>40	57	9.03	70	9.06
No. and % of abortions
2–3	376	59.59
≥4	255	40.41

Notes: aThe P value was obtained from a two-sided χ2 test of the distributions between the patients with recurrent miscarriage and the controls.

Relationship Between SERPINA4 Polymorphisms and Susceptibility to Recurrent Miscarriage

We detected the genotype frequency distributions of the controls and patients with recurrent miscarriage to study the association between SERPINA4 polymorphisms and susceptibility to recurrent miscarriage (as shown in Table 2). No significant deviations from HWE were detected in the control group (P=0.173 for rs910352, P=0.226 for rs2070777). After adjustments for age, the rs2070777 AA genotype was associated with a significantly increased risk of recurrent miscarriage compared with the TT/AT genotypes (TA vs TT: adjusted OR=1.437, 95% CI=1.151–1.794, P=0.0014; AA vs TT: adjusted OR=1.909, 95% CI=1.369–2.661, P=0.0001; AA vs TT/AT adjusted OR=1.409, 95% CI=1.032–1.924, P=0.0309). In addition, we found a significant association between the rs910352 T allele in the SERPINA4 gene and susceptibility to recurrent miscarriage (CT vs CC: adjusted OR=1.579, 95% CI=1.252–1.992, P=0.0001; TT vs CC: adjusted OR=1.524, 95% CI=1.134–2.049, P=0.0052). Consequently, the risk genotypes used for the calculation were SERPINA4 rs910352 CT/TT and 2070777 AA. Compared with individuals with no unfavorable genotype, those who carried one or two unfavorable genotypes exhibited an increased risk for recurrent miscarriage (adjusted OR=1.257, 95% CI=1.019–1.550).

Table 2

Genotype and Allele Frequencies of SERPINA4 in Patients with Recurrent Miscarriage and Controls

Genotype/Allele	RM (N =631)	Controls (N=771)	P-value	OR (95% CI)	P-value	Adjusted OR (95% CI)	P-valueb
SERPINA4/rs910352 C > T (HWE = 0.173)
CC	194(30.74)	266(34.41)	0.3388	1.00	/	1.00	/
CT	312(49.45)	359(46.44)	/	1.577(1.251–1.988)	0.0001	1.579(1.252–1.992)	0.0001
TT	125(19.81)	148(19.15)	/	1.532(1.140–2.060)	0.0047	1.524(1.134–2.049)	0.0052
Dominant (TT/TC vs CC)	437(69.26)	507(65.59)	0.1449	1.182(0.944–1.480)	0.1455	1.186(0.947–1.485)	0.1382
Recessive (TT vs CT/AA)	506(80.19)	625(80.85)	0.7547	1.043(0.800–1.360)	0.7543	1.039(0.797–1.355)	0.777
SERPINA4//rs2070777 T>A (HWE = 0.226)
TT	233(36.93)	319(41.27)	0.0541	1.00	/	1.00	/
TA	302(47.86)	367(47.48)	/	1.430(1.146–1.785)	0.0016	1.437(1.151–1.794)	0.0014
AA	96(15.21)	87(11.25)	/	1.918(1.376–2.673)	0.0001	1.909(1.369–2.661)	0.0001
Dominant (AA/TA vs TT)	398(63.07)	454(58.73)	0.0972	1.200(0.967–1.490)	0.0977	1.206(0.971–1.497)	0.0896
Recessive (AA vs TA/TT)	535(84.79)	686(88.75)	0.0288	1.415(1.036–1.932)	0.0289	1.409(1.032–1.924)	0.0309
Combined risk-effect of genotypes*
0	346(54.83)	285(45.17)		1	/	1	/
1–2	517(60.19)	342(39.81)	0.0388	1.264(1.025–1.558)	0.0283	1.257(1.019–1.550)	0.0326

Notes: *The risk genotypes used for the calculation were rs910352 CT/TT + rs2070777 AA. Dominant: homozygous rare+heterozygous vs homozygous frequent allele. Recessive: homozygous rare vs heterozygous+homozygous frequent allele. Statistically significant values are shown in bold (P<0.05). bAdjusted for age.

Abbreviations: OR, odds ratio; HWE, Hardy–Weinberg equation; RM, recurrent miscarriage.

Stratification Analysis

We further explored the associations between the SERPINA4 gene variant genotypes of the two selected SNPs (rs910352 and rs2070777) and the risk of recurrent miscarriage through a stratified analysis based on the number of abortions and age (Table 3). Compared with the rs2070777 AA genotype, the AT/TT genotypes were associated with a higher risk for women with two to three miscarriages (adjusted OR=1.548, 95% CI=1.088–2.201, P=0.0151) due to the cumulative effect of gene mutations (microeffect gene accumulators can exert a significant phenotypic effect). We combined the analysis of rs910352 CT/TT and rs2070777 AA with respect to risk for recurrent abortion. The combined analysis for all risk genotypes among individuals aged 35–40 years revealed that individuals carrying one or two risk genotypes had a higher risk than individuals with no risk genotypes (adjusted OR=1.571, 95% CI=1.135–2.175). However, no significant associations were observed in the other stratified analyses, such as those conducted with the group of individuals with four miscarriages.

Table 3

Stratification Analysis of the Associations Between SERPINA4 Polymorphisms and Recurrent Miscarriage Risk in a Southern Chinese Population

Variable	rs2070777 (Cases/Controls)		P	OR (95% CI)	P	Adjust OR (95% CI)	P	Combined (Cases/Controls)*		P	OR (95% CI)	P	Adjusted OR (95% CI)	P
	AT/TT	AA						0	1–2
Age
<35	176/210	31/32	0.5944	1.156 (0.678–1.970)	0.5942	/	/	116/128	91/115	0.4752	0.873 (0.602–1.267)	0.4754	/	/
35–40	213/280	40/37	0.1523	1.421(0.878–2.300)	0.1523	/	/	134/237	119/134	0.0065	1.571(1.135–2.175)	0.0065	/	/
>40	146/196	25/18	0.0555	1.865 (0.981–3.545)	0.0574	/	/	96/152	75/93	0.228	1.277 (0.858–1.900)	0.2279	/	/
No. of abortions
2–3	313/686	63/87	0.0107	1.587(1.118–2.254)	0.0098	1.548 (1.088–2.201)	0.0151	210/517	166/342	0.155	1.195(0.935–1.527)	0.1544	1.161 (0.907–1.486)	0.2356
≥4	222/686	33/87	0.4715	1.172(0.764–1.799)	0.4674	1.177(0.766–1.810)	0.4575	136/517	119/342	0.0519	1.323 (0.998–1.753)	0.0514	1.314 (0.991–1.742)	0.058

Notes: *The combination of risk genotypes used for the calculation was SERPINA4 rs910352 CT/TT + rs2070777 TT. Statistically significant values are shown in bold (P<0.05).

Discussion

This case-control study revealed that the SERPINA4 rs2070777 AA variant contributed to an increased risk of recurrent miscarriage in a southern Chinese population, and the increased risk in patients who carry one or two risk genotypes was more obvious in women aged 35–40 years. To our knowledge, this study constitutes the first investigation of the association between susceptibility to recurrent miscarriage and SERPINA4 polymorphisms in a southern Chinese population.

The human kallistatin gene (SERPINA4) is located on 14q32.13 and encodes kallistatin, a protein involved in the regulation of a variety of biological processes.22 The SNPs rs910352 and rs2070777 are located in an intronic region of the SERPINA4 gene. SERPINA4 is involved in regulating the expression of multiple genes and participates in the activation of multiple signaling pathways, including angiogenesis, vasodilation, anti-inflammation, apoptosis, antioxidant, and fibrosis.23–25 Recent studies have found that SERPINA4 gene polymorphisms are involved in the disease process. Vilander LM et al found that the SERPINA4 SNP rs2093266 contributes to the development of severe acute kidney injury.26 PAI-1 belongs to the SERPIN family, which contributes to an increased risk of recurrent miscarriage.21 Tamar Madar et al found that low levels of circulating alpha-1 antitrypsin, which belongs to the SERPIN family, are associated with spontaneous abortion.27 However, whether SERPINA4 polymorphisms are related to abortion has not been investigated. In our study, we evaluated the associations between SERPINA4 gene (rs910352 C>T and rs2070777 T>A) polymorphisms and susceptibility to recurrent miscarriage in 631 patients and 771 healthy controls. Our results showed that the rs2070777 AA variant is a risk factor for recurrent miscarriage susceptibility. To our knowledge, this study provides the first assessment of the association of SERPINA4 polymorphisms with recurrent miscarriage risk in a southern Chinese population. We propose that the rs28270177 AA variant might play an important role in the pathogenesis of patients with recurrent miscarriage.

Many studies have suggested that abortion is related to age and that advanced maternal age is related to an increasing incidence of recurrent miscarriage.28 Embryonic chromosomal abnormalities are the most common cause of early miscarriage, and the proportion of chromosomal abnormalities increases as the maternal age increases (≥35 years).29 Studies have also found that among pregnant women older than 30 years, the risk of miscarriage increases with maternal age.30 Nybo Andersen AM et al found that women ≤35 years have a higher risk of miscarriage in the range of 9 to 12% and that this risk increases to 75% in women older than 40 years.31 Although our study population excluded embryonic chromosomal abnormalities, the combined analysis for all risk genotypes among individuals aged 35-40 years revealed that individuals carrying one or two risk genotypes had a higher risk than individuals with no risk genotypes. Moreover, many studies have found that women with previous abortions have an increased risk of abortion.32,33 Interestingly, our research found that SERPINA4 rs2070777 AA genotypes are associated with a significantly increased risk among those with two to three miscarriages. However, the increased risk associated with the number of miscarriages was not observed in the group with more than four miscarriages, which might be related to the multiple biological functions of kallistatin. The reason for this finding is worth further exploration. Further studies with a larger sample size are needed to confirm the results.

The limitations of this study should be noted. First, only rs910352 C>T and rs2070777 T>A were included in our study, and other gene polymorphisms were not covered. Second, the sample size in the current study was relatively small, and larger sample sizes are needed to confirm the relationship between susceptibility to recurrent miscarriage and SERPINA4 gene polymorphisms. Third, due to the retrospective design of the study, only geographical factors and age were collected in this study. Other factors, such as eating habits and environmental exposure, were not collected. Moreover, the rs910352 heterozygous risk genotype CT in the SERPINA4 gene is associated with a higher risk of OR than homozygous TT. This finding might suggest that this rs910352 TT genotype is not as important for recurrent miscarriage susceptibility. Therefore, the study of the correlation between SERPINA4 rs2070777AA genotypes and recurrent miscarriage needs to be further expanded to verify the research findings.

In summary, our study suggests that the SERPINA4 rs2070777 AA variant is associated with increased susceptibility to recurrent miscarriage in a southern Chinese population. However, studies with a larger sample size and practical studies should be conducted to further explore the roles of the SERPINA4 gene in susceptibility to recurrent miscarriage.