Wnt signaling, which encompasses multiple biochemical pathways that regulate neural development downstream of extracellular Wnt glycoprotein ligands, has been suggested to contribute to major psychiatric disorders including autism spectrum disorders (ASD). We used next-generation sequencing and Sequenom genotyping technologies to resequence 10 Wnt signaling pathway genes in 198 ASD patients and 240 matched controls. Results for single-nucleotide polymorphisms (SNPs) of interest were confirmed in a second set of 91 ASD and 144 control samples. We found a significantly increased burden of extremely rare missense variants predicted to be deleterious by PolyPhen-2, distributed across seven genes in the ASD sample (3.5% in ASD vs 0.8% in controls; Fisher's exact test, odds ratio (OR)=4.37, P=0.04). We also found a missense variant in WNT1 (S88R) that was overrepresented in the ASD sample (8 A/T in 267 ASD (minor allele frequency (MAF)=1.69%) vs 1 A/T in 377 controls (MAF=0.13%), OR=13.0, Fisher's exact test, P=0.0048; OR=8.2 and P=0.053 after correction for population stratification). Functional analysis revealed that WNT1-S88R is more active than wild-type WNT1 in assays for the Wnt/β-catenin signaling pathway. Our findings of a higher burden in ASD of rare missense variants distributed across 7 of 10 Wnt signaling pathway genes tested, and of a functional variant at the WNT1 locus associated with ASD, support that dysfunction of this pathway contributes to ASD susceptibility. Given recent findings of common molecular mechanisms in ASD, schizophrenia and affective disorders, these loci merit scrutiny in other psychiatric conditions as well.

Prostaglandin-endoperoxide synthase (PTGS) (also known as cyclooxygenase) converts arachidonic acid into several prostaglandins, many of which have roles in vasodilation and vasoconstriction under normal and pathological conditions. There are two isoforms of PTGS: PTGS-1 and PTGS-2; PTGS-1 is constitutively expressed in many tissues and is believed to be involved in the homeostatic maintenance of the body. In contrast, PTGS-2 is believed to have a "differentiative" role in the cells and is highly inducible during inflammation and in response to lipopolysaccharide (LPS). Endothelial cells as well as vascular smooth muscle cells can be a source of PTGS within the artery. The objective of this study was to determine the cell population(s) in uterine arteries that respond to LPS with an increase in PTGS-2 protein expression. Uterine arteries collected from ewes during the follicular (Day 0, Day 0 = estrus, n = 4) or luteal (Day 10, n = 4) phase were treated in vitro with LPS as intact artery segments, cut-open artery segments, or cut-open and denuded (endothelial cells absent) artery segments. After 24 h of LPS treatment, intact, cut-open, and denuded uterine artery segments were collected into homogenization buffer for determination of PTGS-2 protein levels by Western blot analysis. The culture medium was collected and used for detection of 6-keto-prostaglandin F(1alpha) (6-keto-PGF(1alpha)), the stable metabolite of prostacyclin, using an enzyme immunoassay. In addition, the location of PTGS-2 after LPS treatment was analyzed by immunohistochemistry in intact artery segments. Denuded arteries (endothelium absent) did not show increases in PTGS-2 protein in the homogenates or 6-keto-PGF(1alpha) in the culture medium after LPS exposure. In contrast, cut uterine arteries responded to LPS stimulation with a significant increase in PTGS-2 protein in homogenates and 6-keto-PGF(1alpha) in culture medium. Immunohistochemical staining for PTGS-2 was associated with both endothelial cells and vascular smooth muscle cells. These results suggest that while both endothelial cells and vascular smooth muscle cells are associated with PTGS-2, after LPS exposure it is the endothelial cells that are essential in uterine artery increases in PTGS-2 and prostacyclin in response to LPS stimulation.