Maternal Vitamin D Status During Pregnancy Is Associated With Fetal Intra-uterine Growth Restriction

Intrauterine growth restriction (IUGR), which manifests as small for gestational age (SGA), not only increases infant mortality and morbidity but also has implications from a life-course perspective. Almost25years ago, Barker and coworkers described LBW as highly correlated with increased risk for the development of cardiovascular diseases during adulthood. Since then, numerous epidemiologic studies have collected a convincing link between IUGR and an increased risk for adult onset of metabolic as well as non-metabolic diseases. Thus, the underlying mechanism for IUGR is of great concern. Vitamin D is essential for the maintenance of calcium homoeostasis. Maternal vitamin D deficiency during pregnancy is prevalent and is increasingly recognized as a public health problem. Recently, several reports demonstrate that maternal vitamin D status during pregnancy is associated with adverse pregnancy outcomes. In the present study, we will elucidate the association between maternal vitamin D status during pregnancy and IUGR in a population-based birth cohort study. Next, we will establish mouse model of vitamin D deficiency during pregnancy, to investigate the effects of vitamin D deficiency during pregnancy on fetal intrauterine growth restriction and to assess the potential role of inflammation in vitamin D deficiency during pregnancy induced intrauterine growth restriction. The IUGR mouse model was established by administration of LPS. We will further investigate the effects of vitamin D deficiency during pregnancy on LPS-induced IUGR in mice, In addation, we will be to clarify the molecular mechanisms of placental inflammation mediated the role of vitamin D deficiency during pregnancy induced IUGR. Finally, by using a nest case-control design, we will further prove the role of placental inflammation on vitamin D deficiency during pregnancy induced IUGR. These studies are significant for us to understand the etiology and pathogenesis of fetal developmental toxicity, and to effectively prevent the incidence of fetal developmental toxicity.1. Maternal vitamin D status during pregnancy is associated with birth weight:a population-based birth cohort studyVitamin D status was assessed among3658pregnant women. For serum25-(OH)D levels, only964pregnant women were sufficient (25-(OH)D>30ng/ml,26.35%),1289(35.24%) insufficient,1359(37.15%) deficient, and46(1.26%) with severely deficient. There results suggest that maternal vitamin D deficiency during pregnancy is prevalent in Hefei. In addition, there was a linear correlation between birth weight and maternal serum25-(OH)D level (r=0.477, P<0.01).Association between vitamin D status and low birth weight (LBW) was analyzed. Interestingly,52.17%(24/46) neonates were with LBW among women with vitamin D extreme deficiency (RR:125.74;95%CI:41.90-377.38),3.38%(46/1359) among women with vitamin D deficiency (RR:8.12;95%CI:2.93-22.74), and1.32(17/1289) among women with vitamin D insufficiency (RR:3.18;95%CI:1.07,9.48). Logistic regression analysis, adjusted for BMI, maternal age, and gestational week of blood sample, showed that adjusted OR for LBW was275.33(95%CI:87.55-865.84) among pregnant women with vitamin D extreme deficiency,8.06(95%CI:2.89-22.49) among pregnant women with vitamin D deficiency, and3.12(95%CI:1.05-9.31) among pregnant women with vitamin D insufficiency. Next, association between vitamin D status and small-for-gestational age (SGA) are analyzed.56.52%(26/46) neonates were with SGA among pregnant women with extreme vitamin D deficiency (RR:20.18;95%CI:10.92-37.31),14.64%(199/1359) among pregnant women with vitamin D deficiency (RR:5.23;95%CI:3.47-7.88),5.59%(72/1289) among pregnant women with vitamin D insufficiency (RR:1.99;95%CI:1.27-3.13). Logistic regression showed that adjusted OR for SGA was47.11(95%CI:23.36-95.00) among pregnant women with extreme vitamin D deficiency,5.82(95%CI:3.85-8.78) among pregnant women with vitamin D deficiency, and2.03(95%CI:1.29-3.18) among pregnant women with vitamin D insufficiency. These results suggest that maternal vitamin D deficiency or insufficiency during pregnancy elevated the risk of not only LBW but also SGA births.2. The role of Vitamin D deficiency during pregnancy on placental and fetal developmentTo further demonstrate association of vitamin D status during pregnancy with IUGR, in VDD (vitamin D deficiency) group, female mice were fed with low vitamin D diets beginning at2weeks before mating and throughout the pregnancy. In control group, female mice were fed with standard feed. All pregnant mice were sacrificed on gd18. The number of live fetuses, dead fetuses and resorption sites was counted. Live fetuses in each litter were weighed. Crown-rump and tail lengths were examined and skeletal development was evaluated. As expected,25-(OH)D was significantly decreased in VDD mice. Although no abortion was observed throughout the pregnancy, the number of dead fetuses was significantly increased in VDD mice. Fetal weight and crown-rump length were significantly reduced in VDD mice. Vitamin D deficiency during pregnancy retards fetal skeletal ossification in caudal vertebrae and ribs. Interestingly, placenta weight was also reduced in VDD mice. These results suggest that maternal vitamin D deficiency during pregnancy retards fetal development in mice, and causes IUGR in mice.Next, we examine whether vitamin D deficiency impairs placental development and placental function. The labyrinth layer was severely disrupted in placentas of VDD mice, with a reduction in the internal space of fetal and maternal blood vessels, suggesting that there would be restricted nutrient and/or oxygen exchange between the mother and the developing embryos. We then measured whether vitamin D deficiency regulates the expressions of placental Glutl, a glucose transporter, Fatp4, a fatty acid transport protein, and snat2, a neutral amino acid transporter. Placental glutl,fatp4and snat2were obviously down-regulated in VDD mice. These results indicate that vitamin D deficiency during pregnancy impairs placental development and placenta function in mice.Unexpectedly,1α,25-(OH)2D3, an active form of vitamin D, did not affect expression of GLUT1, FATP4and SNAT2in human trophoblast-derived JEG-3cells, indicating that GLUT], FATP4and SNAT2are not direct targets of placental VDR signaling. Since tnf-a and il-6, two cytokines, and kc, mcp-1and mip-2, three chemokines, were up-regulated in placentas of VDD mice, we thus test whether la,25-(OH)2D3blocks inflammation-induced down-regulation of GLUT!, FATP4and SNAT2in human JEG-3cells. As expected, LPS markedly up-regulated TNF-a and IL-6in human JEG-3cells, which was blocked in1α,25-(OH)2D3-pretreated JEG-3cells. LPS significantly down-regulated GLUT1, FATP4and SNAT2in JEG-3cells, which was also reversed in1a,25-(OH)2D3-pretreated JEG-3cells. These results suggest that inflammation is, at least in part, mediated in vitamin D deficiency during pregnancy impairs placental and fetal development in mice.Since placenta weight is decreased in VDD mice, we therefore test whether vitamin D deficiency impairs cell proliferation in placentas. The number of PCNA-positive cells was obviously reduced in placentas of VDD mice. Placental cyclin Dl was down-regulated in VDD mice. We then test whether1α,25-(OH)2D3regulates the proliferation of JEG-3cells.1α,25-(OH)2D3alone had no effect on the expression of cyclin Dl in JEG-3cells. Correspondingly, la,25-(OH)2D3alone did not affect the proliferation of JEG-3cells. Since several growth factors, such as IGF2, PGF and VEGF-A, are essential for the development of the labyrinth in the placenta, we analyze whether vitamin D status affects the expression of ig/2, igf2r, pgf and vegf-a in mouse placenta. Placental igf2, igf2r, pgf and vegf-a were down-regulated in VDD-fed mice. Although1α,25-(OH)2D3alone had no effect on the expression of IGF2, IGF2R, PGF and VEGF-a in human JEG-3cells, la,25-(OH)2D3blocked LPS-induced down-regulation of IGF2, IGF2R, PGF and VEGF-a in JEG-3cells. Moreover, la,25-(OH)2D3significantly attenuated LPS-induced repression of JEG-3cell proliferation. In addition,1α,25-(OH)2D3inhibited LPS-induced down-regulation of cyclin D1in JEG-3cells. These results suggest that vitamin D deficiency during pregnancy impairs proliferation in trophoblasts.Since vitamin D deficiency causes inflammation-associated dysfunction of the placenta, we tested whether vitamin D deficiency during pregnancy exacerbates inflammation-associated IUGR in mice. As expected, the number of dead fetuses was significantly increased in LPS-injected mice. Fetal weight and crown-rump length were significantly decreased in LPS-injected mice. Of interest, vitamin D deficiency exacerbated LPS-induced fetal death and IUGR in mice.These results suggest that vitamin D deficiency during pregnancy activated placental NF-κB and its target genes, and impairs proliferation in trophoblasts. Moreover, vitamin D deficiency during pregnancy causes placental dysfunction and fetal IUGR.3. The role of vitamin D supplementation on inflammation-associated impairment of fetal and placental development in miceTo investigated the role of maternal vitamin D supplementation on inflammation-associated impairment on placental development and IUGR in mice. In LPS-treated group, the pregnant mice were injected with LPS (100μg/kg, ip) daily on gestational late stage. In the VitD+LPS group, the pregnant mice were injected with25μg/kg of vitamin D (VitD, p.o.) at24h and1h before LPS (100μg/kg, ip) treatment. The saline-and VitD-treated pregnant mice served as controls. Some pregnant mice were sacrificed on the day of injection, maternal serum and placenta were dissected for measurement of inflammatory cytokines/chemokines concentration and so on. The left were injected the drug continuously and killed on gd18, For each litter, the number of live fetuses, dead fetuses and resorption sites were counted. Live fetuses in each litter were weighted. Crown-rump and tail lengths were measured. And the skeleton of all live fetuses in each litter were evaluated. LPS-induced elevation of TNF-a and IL-6was significantly attenuated in maternal sera of D3-pretreated mice. LPS-induced up-regulation of placental tnf-α, il-1β, il-6, il-12, mip-2, kc and mcp-1was also blocked by vitamin D. Interestingly, LPS-induced placental iNOS was significantly repressed in D-pretreated mice. These results indicate that vitamin D plays a pivotal role in controlling placental inflammatory response.We next tested whether vitamin D alleviates inflammation-associated impairment on placental development. As expected, placental glutl, fatp4, snat2, igf2and igf2r were down-regulated in LPS-treated mice. Placenta weight was reduced in LPS-treated mice. Interestingly, vitamin D significantly alleviated LPS-induced reduction of placenta weight. Moreover, vitamin D significantly attenuated LPS-induced down-regulation of placental glutl,fatp4, snat2, igf2and igf2r. These results indicate that maternal vitamin D supplementation during pregnancy significantly attenuated LPS impairs placental development in mice. Since vitamin D supplementation during pregnancy prevents from inflammation-associated impairment on placental development, we analyzed whether vitamin D protects against inflammation-associated IUGR in mice. As expected, no dams died and no abortion was observed throughout the pregnancy. Vitamin D alone did not induce fetal death. In addition, vitamin D alone did not affect fetal weight and crown-rump length in mice. Of interest, vitamin D protected against LPS-induced fetal death. Moreover, vitamin D supplementation significantly attenuated LPS-induced IUGR. These results indicate that vitamin D supplementation during pregnancy significantly alleviated LPS-induced IUGR in mice.4. The mechanism that vitamin D exerts its anti-inflammatory activity Since vitamin D inhibits LPS-induced placental cytokines, we tested whether vitamin D represses signaling cascades that regulate LPS-induced cytokines. In LPS-treated group, the pregnant mice were injected with LPS (100μg/kg, ip) daily on gestational late stage. In the VitD+LPS group, the pregnant mice were injected with25μg/kg of vitamin D (VitD, p.o.) at24h and1h before LPS (100μg/kg, ip) treatment. The saline-and VitD-treated pregnant mice served as controls. All pregnant mice were sacrificed2h after LPS injection. Placenta were dissected. Vitamin D did not affect the expressions of placental tlr4and MyD88. Moreover, vitamin D had no effect on LPS-evoked activation of placental MAPK p38. In addition, vitamin D did not repress LPS-induced placental Akt phosphorylation. Interestingly, vitamin D blocked LPS-evoked translocation of NF-κB p65to the nuclei. Correspondingly, vitamin D markedly inhibited LPS-evoked placental NF-κB binding activity. To further demonstrate the association between vitamin D status and placental NF-κB signaling, we measured placental NF-κB activity in low vitamin D diets-fed mice. Surprisingly, placental nuclear NF-κB p65was significantly elevated in low vitamin D diets-fed mice, indicating that placental NF-κB signaling is activated. Generally, NF-κB is retained in the cytoplasm by binding to the inhibitor of κB (IκB). IκB phosphorylation causes translocation of NF-κB p65to the nucleus. Unexpectedly, vitamin D had little effect on LPS-induced placental IκB phosphorylation.Since vitamin D, in the form of la,25-(OH)2D3, functions as a natural ligand of VDR, we thus test whether maternal vitamin D status influences placental VDR signaling. Although maternal vitamin D deficiency did not alter the expression of placental VDR, placental nuclear VDR level was significantly reduced in low vitamin D diets-fed mice. To further explore the functional role of VDR in modulating NF-κB activity, siRNA was used to inhibit the expression of VDR in JEG-3cells. VDR mRNA and protein were down-regulated by70%in VDR siRNA-transfected JEG-3cells. Correspondingly, CYP3A4, a downstream target of VDR, was also down-regulated in VDR siRNA-transfected JEG-3cells. We then analyzed the effects of1α,25-(OH)2D3on LPS-activated NF-κB signaling in JEG-3cells as well as in VDR siRNA-transfected JEG-3cells. As expected, pretreatment with10nM or higher concentrations of1α,25-(OH)2D3significantly attenuated LPS-induced translocation of NF-κB p65to the nuclei in JEG-3cells. By contrast, pretreatment with10nM1α,25-(OH)2D3had no effect on LPS-induced NF-κB activation in VDR siRNA-transfected JEG-3cells. Similarly,1α,25-(OH)2D3could not inhibit LPS-induced TNF-a and IL-6upregulation in VDR siRNA-transfected JEG-3cells.As expected, the level of nuclear VDR were elevated in the placentas of vitamin D-pretreated mice and1α,25-(OH)2D3-pretreated human JEG-3cells. Correspondingly, cyp24al, a target gene of VDR signaling, was up-regulated in mouse placenta. In addition, CYP3A4, a target gene of VDR signaling, were up-regulated in vitamin D-pretreated human JEG-3cells. Since vitamin D inhibits LPS-evoked NF-κB signaling in VDR-dependent manner, we test whether LPS represses vitamin D-activated VDR signaling. Although it did not affect placental VDR expression, LPS significantly repressed vitamin D-evoked nuclear VDR translocation and cyp24al up-regulation in mouse placentas. Of interest, LPS also inhibited1α,25-(OH)2D3-evoked nuclear VDR translocation and CYP3A4up-regulation in JEG-3cells. Co-Immunoprecipitation (CoIP) showed that pretreatment with vitamin D promoted the interaction between placental VDR and NF-κB p65in LPS-treated mice. Similarly, pretreatment with la,25-(OH)2D3promoted the interaction between VDR and NF-κB p65in LPS-treated human JEG-3cells.These results suggest that maternal vitamin D supplementation that activated placental VDR signaling inhibited placental NF-κB and its target genes. Moreover, maternal vitamin D supplementation protected mice from inflammation-associated placental dysfunction and fetal IUGR. 5. Expression of VDR and NF-κB p65in normal and SGA placentasWe compared placental VDR in SGA cases and controls. High levels of VDR were expressed in trophoblasts of control placentas. Of interest, placental nuclear VDR was markedly reduced in SGA cases as compared with controls. Correspondingly, the levels of25-(OH)D in maternal sera were lower in LBW and SGA cases than those of controls. Of interest,25-(OH)D in umbilical sera was also decreased in SGA cases as compared with controls. We then compared the expression of placental NF-κB p65in SGA cases and controls. Placental nuclear NF-κB p65was significantly elevated in SGA cases as compared with controls. The levels of TNF-a and IL-8in maternal sera were higher in LBW cases than those of controls. These results suggest an association between inflammatory cytokines and fetal IUGR. Moreover, placental VDR signaling may play an important role as a key regulator of inflammation and the immune response during pregnancy.In summary, the present results allow us to reach the following conclusions. First, Vitamin D deficiency and insufficiency during pregnancy elevated the risk of LBW and SGA infants. Second, vitamin D deficiency during pregnancy activated placental NF-κB signaling and up-regulated cytokine, and resulted in placental dysfunction and fetal IUGR in mice. Third, vitamin D supplementation protected against inflammation-associated placental dysfunction and IUGR in mice. Fourth, Vitamin D supplementation that activated placental VDR signaling inhibited placental NF-κB signaling. Fifth, overall, this study provides evidence for a role of placental VDR signaling as a key regulator of inflammation and the immune response during pregnancy.