Genetic and Functional Analysis of Congenital Portosystemic Shunts in Dogs

Abstract

The general aim of this thesis was to gain further insight into the pathogenesis of congenital portosystemic shunts (CPSS), elucidate mechanisms involved in the pathophysiology of CPSS, and explore predictors of recovery after surgical ligation of the shunt. For intrahepatic portosystemic shunts (IHPSS), the genetic background of the different anatomical locations is likely to be identical as they are observed within one litter of Irish Wolfhounds. In this thesis, the anatomical distribution of extrahepatic portosystemic shunt (EHPSS) subtypes was retrospectively analysed in 135 dogs with a single EHPSS. It was demonstrated that both EHPSS subtypes, portocaval and porto-azygous shunts, coexist in all studied dog breeds (apart from the Pug), suggesting that the two subtypes are variants of the same inherited disorder. These findings indicate that for studies into the genetic background of both IHPSS and EHPSS no distinction is required considering subtypes. Subsequently, a micro-array study was performed on liver tissue of 47 CPSS dogs to unravel genes involved in the pathogenesis of CPSS.By comparing expression levels in both shunt types with the levels observed in healthy liver tissue, a list of only 26 genes that were differentially expressed for one of the two types of shunt was found. Follow-up experiments including qPCR, immunohistochemistry and Western blots, resulted in two possible candidate genes, or their regulating pathways, causing CPSS in dogs; namely VCAM1 for EHPSS and WEE1 for IHPSS. Another aspect of this study was to gain further insight into the pathophysiology of one of the characteristics shared in hepatic biopsies of both IHPSS and EHPSS; hepatic lipid accumulation. A 12-fold increase of lipid content was observed in shunt biopsies when compared to normal liver tissue using Oil-red-O staining. Liver organoids from CPSS dogs and healthy controls were incubated with palmitic- and oleic-acid, and lipid accumulation was quantified. All organoids showed a similar increase in triacylglycerols after FFA enrichment indicating that lipid accumulation in the presence of portosystemic shunting is not caused by primary genetic defects. Taken together, lipid accumulation was observed in all shunt types and occurs secondary to portosystemic shunting. In the final part of the thesis, a predictive model was proposed, based on pre-operative plasma albumin and intra-operative mRNA expression levels of hepatic gene products (DHDH, ERLEC1, and LYSMD2) as predictors of recovery from portosystemic shunting after surgical ligation of the shunt. If surgery would be performed on dogs with a predicted probability >50% only, this model resulted in a sensitivity to correctly predicted recovery of 77% and a specificity of 90%. Furthermore, this model seems to have good discriminating abilities between recovered and non-recovered dogs. In conclusion, extensive genomic research based on the different aetiology and the shared physiology of EHPSS and IHPSS has resulted in novel insights in inherited congenital portosystemic shunts in dogs.