Vitamin D receptor polymorphisms in spontaneous preterm birth: a case-control study

Aim To evaluate the association between the FokI (rs2228570), ApaI (rs7975232), Bsml (rs1544410), TaqI (rs 731236), and Cdx2 (rs11568820) single nucleotide polymorphisms (SNPs) in the vitamin D receptor (VDR) gene and spontaneous preterm birth (SPTB), as well as their effect on clinical characteristics of women with SPTB and their newborns. Methods This case-control study enrolled women who gave birth at the Department of Obstetrics and Gynecology, University Medical Center Ljubljana between 2010 to 2019. Cases were 118 women with spontaneous initiation of PTB after natural conception and 119 controls with a term singleton delivery after an uncomplicated pregnancy. The molecular analysis of VDR SNPs employed polymerase chain reaction and restriction fragment length polymorphism. Results Patients and controls did not significantly differ in the distribution of genotype or allele SNP frequencies. However, the FokI polymorphism had a significant effect on newborn birth weight in women with SPTB but not in controls (F = 5.17, P = 0.007, one-way ANOVA with post-hoc Scheffe test), with newborns of FokI TT carriers having the lowest birth weight (P = 0.011). No other VDR SNP was associated with any other clinical characteristic of women with SPTB and their newborns. Conclusion The TT genotype of the VDR FokI polymorphism is associated with newborn birth weight in women of European origin with SPTB.

Preterm birth (PTB), defined as a birth at less than 37 weeks of gestational age (1), represents the leading determinant of neonatal morbidity and mortality, as well as the main cause of long-term neurologic disabilities in children (2). In spite of the progression in obstetrics and clinical management, PTB incidence in the last decades has not been reduced (3).
PTB can be either spontaneous (SPTB) or medically induced. The heterogeneity of SPTB may be attributed to multiple factors, including socioeconomic status, environmental factors, and genetics (4). The etiology of PTB is still not completely understood, although numerous risk factors have been recognized: previous PTB, uterus and cervix anomalies, repeated miscarriages in the second trimester, in-vitro fertilization, multiple pregnancy, vaginal bleeding, abnormal placental factors, low socioeconomic status, smoking, etc (5). Considering that personal and family history of PTB are strongest risk factors for a subsequent PTB, research is focused on identifying variability in potential gene candidates (6,7).
The function of vitamin D metabolism in pregnancy, especially PTB, is currently being intensively investigated (8,9). Vitamin D is a fat-soluble vitamin and hormone precursor, with one of its two forms, cholecalciferol (or vitamin D3), being synthesized in the skin by the exposure to UV radiation. The role of vitamin D in calcium and phosphorus metabolism is well known, but in recent years it has been studied in many other biological processes, particularly in immunomodulation at the feto-maternal interface.
Vitamin D deficiency is related to numerous pregnancy complications (10,11). In addition, several studies clearly showed a causal association between vitamin D status in pregnant women and PTB (12)(13)(14). Likewise, preterm infants had lower circulating levels of vitamin D (15).
The association between vitamin D levels and PTB may be mediated by single nucleotide polymorphisms (SNPs) in the vitamin D receptor gene (VDR) (16,17). Sixty-three polymorphisms of VDR gene have been found, and five of them have been intensively investigated: Cdx2 in the promotion region, FokI in the second coding exon, and BsmI, ApaI, and TaqI in the 3' untranslated region (18). Available evidence indicates a possible causal connection between certain SNPs and PTB, although results are contradictory (16,17,(19)(20)(21).
Therefore, the aim of our study was to evaluate the association between the FokI (rs2228570), ApaI (rs7975232), Bsml (rs1544410), TaqI (rs 731236), and Cdx2 (rs11568820) SNPs of the VDR gene and SPTB, as well as their effect on clinical characteristics of women with SPTB and their newborns.

Patients
This case-control study enrolled women who gave birth at the Division of Perinatology, Department of Obstetrics and Gynaecology, University Medical Center in Ljubljana, Slovenia between 2010 and 2019. All participants gave written informed consent for study participation. The study was approved by the Slovenian National Medical Ethics Committee and the Ethics Committee for Biomedical Research of the Faculty of Medicine, University of Rijeka.
The patient group comprised 118 women with SPTB. Demographic and clinical data of women with SPTB and their newborns, including family history of SPTB, smoking, maternal age, gestational week of birth, and birth weight of newborns, were collected as described in genetic epidemiology studies on PTB (22) by means of a self-developed interviewer-administered questionnaire. The inclusion criteria for the SPTB group were singleton pregnancy following natural conception and spontaneous initiation of PTB before the 37th week of gestation. Gestational age was estimated from the date of last menstrual period and confirmed by ultrasound in the first trimester. If the measurements differed for more than seven days, gestational age was modified according to the ultrasound measurement. The control group consisted of 119 women who had a term birth of a singleton baby after an uncomplicated pregnancy and were of the same age and parity as patients. The exclusion criteria for both patients and controls were diabetes, hypertension, kidney disease, autoimmune conditions, allergic diseases, birth canal infections, in vitro fertilization, and complications of pregnancy and congenital anomalies or evidence of infection in newborns. Cases' and controls' maternal and newborn characteristics are shown in Table 1.

DNA extraction and genotyping
Genomic DNA was extracted from peripheral blood leukocytes by standard procedure with Qiagen FlexiGene DNA kit (Qiagen GmbH, Hilden, Germany). All samples were stored at -20 °C. The molecular analysis of VDR SNPs employed polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP). Genotyping conditions, including primer sequences, PCR protocol, and RFLP fragment sizes are shown in Table 2 (23). PCR reactions were carried out in thermal cyclers (Mastercycle Per-sonal, Eppendorf, Hamburg, Germany and 2720 Thermal Cycler, Applied Biosystems, Carlsbad, CA, USA). Following restriction enzyme digestion, the fragments were separat-  The normality of distribution of the numerical variables was tested with the Kolmogorov-Smirnov test. The Pearson chi square test was used to assess the significance of differences in genotype, allele, and haplotype frequencies between the groups. To assess the genetic association with SPTB we calculated odds ratios (OR) and 95% confidence intervals. The t test was used to assess the significance of differences in age and newborn birth weight between patients and controls. One-way and two-way analysis of variance (ANOVA) with factorial design and post-hoc Scheffe test was used to compare maternal and/or gestational age and fetal birth weight between genotypes of VDR SNPs (one-way ANOVA for individual SNPs, two-way ANOVA for the interaction between SNPs).

Association of clinical characteristics of women and their newborns with VDR SNPs
One-way ANOVA showed a significant effect of the FokI and ApaI SNP on newborn birth weight in women with SPTB (FokI F = 5.17, P = 0.007; ApaI F = 3.48, P = 0.034) but not in the control group (FokI F = 0.57, P = 0.569; ApaI F = 0.37, P = 0.689). However, two-way ANOVA with factorial design showed no significant interaction between the FokI and ApaI SNPs and found only a significant effect of FokI had on newborn birth weight in women with SPTB (

DISCUSSION
In this study, TT genotype of the VDR FokI polymorphism was associated with newborn birth weight women of European origin with SPTB.
Complex molecular and biological interactions between the mother and fetus are not entirely elucidated. Although newborn birth weight is affected by numerous sociological and environmental factors, the strongest effect is exerted by genetics (24). In addition, a causal relationship between pregnancy outcome and vitamin D has been reported (14,25). Although the role of vitamin D in human growth and bone maturation is well-known, vitamin D plays many other roles during gestation, such as influence on implantation, maintenance of normal pregnancy, support of fetal growth through calcium delivery, secretion of numerous placental hormones, and reduction of proinflammatory cytokine production (25). Given the numerous functions of vitamin D, this study focused on the VDR gene as a major etiological factor in SPTB. VDR is a mediator of the pleiotropic actions of vitamin D, and its reduced placental expression is implicated in the pathology of complicated pregnancies (26). VDR expression in the human placenta indicates that vitamin D could locally influence the fetalplacental development, growth, and cell differentiation as well as signaling at the maternal-fetal interface.
Although the influence of VDR SNPs on VDR protein function and signaling is still largely unclear, some polymorphisms affect the circulating levels of vitamin D and its bio-  logical activity (27). Research suggests that VDR genotype affects the optimal vitamin D concentration and reduces the incidence of adverse perinatal outcome. Although the present study showed an effect of the FokI TT genotype on newborn birth weight, we did not find any difference in the frequencies of FokI genotypes and alleles between patients and controls. However, Manzon et al (20) and Javorski et al (17) showed that the maternal FokI T allele increased the risk for SPTB, with an OR of 3.32 and 7.49, respectively. The only previous study that investigated the association between VDR SNPs and newborn birth weight found, contrary to our results, that the FokI A allele was associated with lower birth weight (19). We assume that the differences between these studies might be attributable to patient and control selection, sample size, ethnicity, sun exposure, and maternal life-style.
The FokI SNP is considered an independent marker in the VDR gene since it does not show linkage disequilibrium with other polymorphisms within the gene. It is a T-C SNP with two protein variants: a long (T-allele) and a short form of VDR protein (C-allele). Variations in the protein sequence significantly affect the protein function. Using a cell line of human fibroblasts, it has been shown that the FokI "f" isoform (T allele) exhibits a less active form of endogenous VDR (28). Since in this study the TT genotype was associated with lower newborn birth weight, we assume that this is the genetic variant responsible for the lower birth weight of prematurely born children. Furthermore, other authors noticed an interesting effect of fetal VDR genotype in pregnant women with a lower vitamin D concentration on anthropometric measures at birth. Morley et al (29) observed that infant sequence variants in the VDR gene modified the effect of maternal vitamin D deficiency on infant size at birth. Low vitamin D concentrations were associated with lower newborn birth weight only among infants homozygous for the FokI major allele or among heterozygotes.
The function of other VDR SNPs -Bsml, ApaI, TaqI, and Cdx2 is less clear. Although we did not demonstrate a significant association with SPTB for any of these SNPs, they represent good research candidates due to the multiple functions of the VDR gene, particularly in immunomodulation (cell proliferation, differentiation, invasion, and apoptosis). Baczynska-Strzecha et al (21) showed that individual maternal TaqI, Bsml, and ApaI SNPs did not increase the risk of PTB, but some of the genotype combinations (BsmI-ApaI-TaqI) were significantly more frequent in women with PTB. Rosenfield et al (16) showed the association between both maternal and neonatal VDR variants and PTB. The effect of these vari-ants on newborn birth weight was also shown in several studies, which is contrary to our results. Rosenfield et al (16) showed that ApaI AA homozygotes had a 2-fold increased risk for PTB compared with heterozygotes (OR = 1.97). Furthermore, Javorski et al (17) found a significant difference in Cdx2 allele frequencies between the SPTB and control maternal group, and observed an association between this SNP and newborn birth weight. Children of ApaI heterozygotes also had lower birth weight (30). Swamy et al (31) demonstrated the correlation between some other SNPs in the mother's VDR gene and lower newborn birth weight in non-Hispanic black women (31). The differences between our and other studies for all five investigated SNPs might be attributed to population differences, participant selection criteria, and sample size.
Future studies should consider environmental factors for low birth-weight, including smoking. The present study showed no association between VDR SNP genotypes and smoking before pregnancy. Nevertheless, previous research indicated that cigarette smoke decreased vitamin D production, which could affect VDR expression levels (32,33).
A possible limitation of our study is the analysis of only the mother's genotype. However, previous studies have shown that maternal genotype is the most important factor affecting subsequent PTB (34). Additional limitations could be multiple comparisons and possible false positive findings, which is why the robustness of our conclusions should be confirmed by an independent replication study. Our study's advantages are the strict screening of patients with SPTB and exclusion of all PTBs that did not meet the standard clinical definition of SPTB, as well as a sufficient statistical power of the research.
In conclusion, our study showed that the VDR FokI SNP was associated with SPTB in women of European origin and that it could affect the lower birth weight of their newborns. Nevertheless, other genetic associations and expression studies in different populations are needed to investigate the role of VDR SNPs in SPTB, especially studies assessing maternal vitamin D status.