ABSTRACT: Perilipin 5 (PLIN5) is a lipid droplet (LD) protein highly expressed in tissues that are active in fatty acid oxidation such as the heart, oxidative muscle, the liver, and brown adipocytes. PLIN5 shares homology with PLIN1: it prevents lipolysis in basal state but promotes lipolysis upon the activation of protein kinase A (PKA). In vitro studies expressing exogenous PLIN5 have shown that PLIN5 is phosphorylated at serine 155 (S155) by PKA. In addition to the regulation of lipolysis, phosphorylation of PLIN5 at S155 has been shown to regulate expression of genes in lipid metabolism, mitochondria, inflammation, and autophagy by traveling to the nucleus and activating peroxisome proliferater-activated-receptor-gamma-coactivator-1-α (PGC1a). However, a role of PLIN5 phosphorylation at S155 has been primarily studied in vitro with exception of a few studies expressing exogenous wild type (WT) and phosphorylation resistant PLIN5 (S155A) in the heart or the liver of mice. Here, we aim to determine the extent of endogenous PLIN5 phosphorylation in vivo and its role in the regulation of lipid metabolism and gene expression in the liver using PLIN5 S155A knock in mice (S155A). First, we enriched LDs from the liver of fed and fasted C57BL6 WT mice to measure non-phosphorylated and S155 phosphorylated PLIN5 by LC-MS/MS. While peak area of non-phosphorylated PLIN5 was similar, S155 phosphorylation was increased in the liver from fasted mice compared with fed mice confirming that endogenous PLIN5 is phosphorylated at S155 when PKA is activated during fasting (3.0-fold increase, p<0.05). Then, we utilized Phos-tag gel as a simple method to assess the extent of PLIN5 phosphorylation. When the heart and the liver from WT and S155A mice were run on Phos-tag gel and immunoblotted for PLIN5, a phosphorylated band was significantly reduced in S155A tissues confirming that S155 is the dominant phosphorylation site in vivo. Next, we assessed phenotypes of S155A mice to gauge the necessity of S155 phosphorylation for metabolism and gene expression in the fasted liver. For the glucose tolerance test, 4-month-old male and female S155A mice showed no differences, but one year-old S115A males had impaired glucose tolerance. There was not a significant difference in weight or blood glucose in fasted young S155A females compared to wild type, but aged male S155A mice had a higher fasted blood glucose (p<0.05) without difference in weight. There was not a significant difference in serum insulin, triglycerides (TG), beta-hydroxybutyrate, or non-esterified fatty acids levels between WT and S155A mice in either males or females. The liver TG contents did not differ between WT and S155A mice of both genders. When qPCR tested the liver of fasted WT and S155A mice, Pgc1a target genes upregulated by fasting and proposed to be regulated by PLIN5 phosphorylation did not show striking difference in expression. Pgc1a, Ppara, Acot1, Pdk4, and Ascl1 expression was similar between WT and S155A liver of fasted mice. Cpt1a and G6pc showed trend of reduction in S155A liver but did not reach statistical significance (n=5 to 7). Inflammation markers such as IL1b and Ccl2 did not differ between WT and S155A liver of fasted mice either. We performed unbiased RNA sequencing of WT and S155A liver from fasted female mice to identify genes differentially regulated by phosphorylation of PLIN5 at S155. There were limited number of genes differentially expressed in S155A female livers with 3 genes < 0.05 for adjusted p value. Interestingly, RNA seq identified insulin receptor substrate 1 (Irs1) as a gene with significantly decreased expression in the liver of female S155A mice (Log2 fold change -0.74, adjusted p value 0.003). The difference was confirmed in qPCR of fasted liver for the female mice (p<0.05), but not for males. Interestingly, insulin receptor substrate 2 (Irs2) measured by qPCR was not different in the liver of female S155A mice but was reduced in male S155A mice (p<0.05). The reduction of IRS2 protein in the liver of S155A male mice was confirmed by Western blot. In conclusion, endogenous PLIN5 in the liver increases phosphorylation at S155 upon fasting in mice. However, the loss of S155 phosphorylation does not affect serum lipids or liver TG. Also, changes in the expression of PGC1a target genes were subtle in the liver of fasted S155A mice compared with WT mice indicating that the loss of PLIN5 S155 phosphorylation can be readily compensated by other mechanisms. Unexpectedly, IRS2 showed decreased expression in the liver of S155A male mice that may explain glucose intolerance in aged male mice. Thus, PLIN5 phosphorylation may impact the insulin signaling in the liver by supporting IRS2 expression in male mice.
