Temporal candidate gene expression patterns in the sow placenta during early gestation and the effect of maternal L-arginine supplementation.

Abstract

The trend towards high ovulation rates in mature commercial sows has resulted in intra-uterine crowding in the immediate post-implantation period, with negative impacts on placental development and later fetal development (Town et al. 2004). Factors that improve placental angiogenesis could offset the effects of intra-uterine crowding by supporting placental development at critical times in gestation. Feeding of L-arginine has been shown to have beneficial effects on placental vascularization in gilts (Hazeleger et al. 2007) and on litter size born in gilts (Mateo et al. 2007) and sows (Ramaekers et a/. 2006). In the present study, we investigated the effects of L-arginine supplementation to commercial sows on global placental gene expression, and on temporal changes in the expression of a panel of eight candidate genes known to be involved in angiogenesis, in early pregnancy. Multiparous sows (n —48) were either non-supplemented Controls or were fed an L-arginine supplement (20g.d) from d 15 through 29 of gestation. A representative number of sows were euthanized on days 16 through 49 of gestation and embryonic and placental tissues were collected from two average-sized conceptuses from each uterine horn and placed in RNAlater for later analysis. To obtain temporal expression profiles for specific genes involved in placental angiogenesis, total placental RNA was extracted from all Control samples collected, reverse transcribed and real-time PCR used to determine the transcript abundance of: vascular endothelial growth factor (VEGF) -A; the two VEGF receptors, fms-related tyrosine kinase 1 (FLT1) and fetal liver kinase-1(flk-1/KDR); hypoxia-inducible factor (HIF)1A; the Angiopoietins (ANGPT) -1 and -2 and their receptor, TEK tyrosine kinase; and finally Angiogenin (ANG) -1. The delta ACt values were calculated using 185 as an internal control, and data were analyzed using regression analysis (SAS Institute Inc., Cary, NC). To determine the cumulative effect of L-arginine treatment, real-time PCR for these same candidate genes was also performed on the d 30 placental samples from both Control and L-arginine sows. The relative ACt values for d 30 samples were again calculated using 185 as an internal control and data were analyzed using MIXED models (SAS Institute Inc., Cary, NC). Effects of L-arginine on global placental gene expression (n =4 representative sows per treatment) were also analyzed using PigOligoArray slides and placental tissues collected at d 30 of gestation. Total RNA was extracted, purified using mRNA mini kits (lnvitrogen), amplified with aminoallyl mRNA amplification kit (Ambion), and labeled with Cy3 or Cy5 in a random block dye-swap design. The hybridized slide images were captured with Genepix software and an Axon Scanner set for optimized PMT for each dye. Median spot intensities underwent Loess and quantile normalization and were analyzed using linear models, all in limma (Smyth, 2004).