Associations Between Prenatal Vitamin D and Placental Gene Expression

Abstract

Background: Vitamin D is a hormone regulating gene transcription. Prenatal vitamin D has been linked to immune and vascular function in the placenta, a key organ of pregnancy. To date, studies of vitamin D and placental gene expression have focused on a limited number of candidate genes. Transcriptome-wide RNA sequencing can provide a more complete representation of the placental effects of vitamin D.

Objective: We investigated the association between prenatal vitamin D levels and placental gene expression in a large, prospective pregnancy cohort.

Methods: Participants were recruited in Shelby County, Tennessee in the Conditions Affecting Neurocognitive Development and Learning in Early childhood (CANDLE) study. Vitamin D level (plasma total 25-hydroxyvitatmin D, [25(OH)D]) was measured at mid-pregnancy (16-28 weeks' gestation) and delivery. Placenta samples were collected at birth. RNA was isolated and sequenced. We identified differentially expressed genes (DEGs) using adjusted linear regression models. We also conducted weighted gene co-expression network analysis (WGCNA).

Results: The median 25(OH)D of participants was 21.8 ng/mL at mid-pregnancy (N=774, IQR: 15.4-26.5 ng/mL) and 23.6 ng/mL at delivery (N=753, IQR: 16.8-29.1 ng/mL). Placental expression of 25 DEGs was associated with 25(OH)D at mid-pregnancy, but no DEG was associated with 25(OH)D at delivery. DEGs were related to energy metabolism, cytoskeletal function, and RNA transcription. Using WGCNA, we identified 2 gene modules whose expression was associated with 25(OH)D at mid-pregnancy and 1 module associated with 25(OH)D at delivery. These modules were enriched for genes related to mitochondrial and cytoskeletal function, and were regulated by transcription factors including ARNT2, BHLHE40, FOSL2, JUND, and NFKB1.

Conclusions: Our results indicate that 25(OH)D during mid-pregnancy, but not at delivery, is associated with placental gene expression at birth. Future research is needed to investigate a potential role of vitamin D in programming placental mitochondrial metabolism, intracellular transport, and transcriptional regulation during pregnancy.

Figure 1.. Vitamin D levels at mid-pregnancy and delivery.

Density plots of vitamin D levels measured as plasma 25(OH)D levels at (A) enrollment during mid-pregnancy (N=774) and (B) at delivery (N=753). Tertile cutoffs are indicated by dotted vertical lines. (C) A scatter plot of Vitamin D levels at mid-pregnancy and delivery (N=752) shows high correlation.

Figure 2.

Upset plot depicting shared and distinct changes in placental gene expression. Maternal plasma 25(OH)D levels at mid-pregnancy were analyzed as continuous variable and based on tertile membership. The horizontal bars indicate the total number of upregulated (red) DEGs and downregulated (blue) DEGs in each analysis. The vertical bars indicate the number of unique or overlapping DEGs. The DEGs that are significantly associated with vitamin D levels in more than one analysis are annotated in color-coded boxes.

Figure 3:

Weighted Gene Co-expression Network Analysis (WGCNA) modules associated with maternal 25(OH)D at mid-pregnancy and delivery. The point size corresponds to −log(p) in each panel. (A) Associations between modules and maternal 25(OH)D concentrations at mid-pregnancy and delivery were assessed with multiple linear regression adjusted for covariates (p < 0.05). Negative associations indicated with downward-pointing triangles and positive associations indicated with upward-pointing triangles. (B) Transcription factors (TFs) that were differentially expressed with 25(OH)D and whose gene targets were enriched in these modules were identified using over-representation analysis (FDR < 0.05). (C) KEGG pathways that were enriched for module genes were identified through an over-representation analysis (FDR < 0.05).