In addition to the liver
In addition to the liver, muscle is a key target organ of insulin action as it stimulates the uptake and disposal of glucose to maintain blood glucose levels within the normal (physiological) range. Relative gene expression of glucose transporter type 4 (Glut4; −29%; p < 0.01), muscle type 6-phosphofructokinase (Pfkm; −21%; p < 0.05) and hexokinase 2 (Hk2; −39%; p < 0.001), key genes in the utilization of glucose, was significantly lower in Prg4 KO muscle than in WT muscle (Fig. 4A). This reduction in insulin-regulated Glut4 expression translated in only a minor, non-significant reduction of uptake of radiolabeled 2-deoxyglucose into skeletal muscle (Fig. 4B). The Glut4 expression correlated significantly with the uptake of the radiolabeled 2-deoxyglucose (r = 0.43; p < 0.05; Fig. 4C). It thus appears that Prg4 KO mice also display a reduced muscular insulin action, possibly related to the lower insulin levels in the Prg4 KO mice.
In white adipose tissue, the actions of insulin are bilateral, as it both decreases intracellular lipolysis by inhibiting hormone sensitive lipase and it stimulates the storage of circulating glucose and fatty acids [, , ]. To facilitate the latter, the lipoprotein lipase-mediated uptake of triglyceride-derived fatty acids from the circulation is increased in response to insulin . Insulin-regulated gene expression of hormone sensitive lipase (Hsl), the rate limiting enzyme in adipocyte lipolysis, tended to be higher in Prg4 KO mice (+51%; p = 0.08) than in WT mice (Fig. 5A). Moreover, white adipose tissue of Prg4 KO mice took up significantly less VLDL-like particle-derived triglycerides labeled with glycerol tri[3H]oleate (−46%; p < 0.05) than WT adipose tissue (Fig. 5B). Insulin also stimulates the transcription of leptin , an adipokine that reduces food intake and increases Liproxstatin-1 expenditure . Leptin (Lep) gene expression was significantly lower (−51%; p < 0.01) in white adipose tissue of Prg4 KO mice than that of WT mice (Fig. 5A). Importantly, the gene expression of the macrophage marker Cd68 and monocyte chemoattractant protein 1 (Mcp1) in adipose tissue of Prg4 KO mice was significantly lower (−65% and −81%; p < 0.01) than that of WT controls (Fig. 5A). The decrease in Cd68 expression was paralleled by a decrease in the expression levels of both the M1 macrophage marker TNFα (−63%; p < 0.01) and the M2 macrophage marker Cd206 (−35%; p < 0.05) (Fig. 5A), suggesting that Prg4 deficiency was associated with a decrease in adipose tissue macrophage numbers in the context of an unchanged macrophage polarization state. White adipose tissue in Prg4 KO mice thus not only showed a more beneficial metabolic phenotype as compared to that of WT mice, but also exhibited a reduced inflammation extent.
From our aforementioned findings it appears that Prg4 KO mice have an improved metabolic profile as compared to WT mice, suggesting that the associations between Prg4 and metabolic disturbances previously found in humans are possibly causal. To provide further evidence for a role for Prg4 in the development of diabetes in humans, we analyzed Prg4 gene expression in a cohort of obese females with or without type 2 diabetes. The expression of Prg4 in the subcutaneous adipose tissue depot in female obese individuals with type 2 diabetes was higher as compared to glucose tolerant control individuals (+11%; p < 0.05). This effect seemed to be dependent on the adipose tissue depot, as omental adipose tissue Prg4 expression was not different between diabetics and controls (Fig. 6). Altogether, these data further suggest that Prg4 is likely to be clinically relevant in type 2 diabetes.
Discussion Importantly, various proteoglycans have previously been reported to be upregulated in murine models for obesity and type 2 diabetes and are therefore associated with the development of these pathologies [32,33]. Lower serum Prg4 levels specifically are associated with weight loss and improved plasma lipid profiles and ameliorated insulin resistance [, , ]. We here show a similar metabolic phenotype in mice, with HFD/fructose-challenged Prg4 KO mice. This study therefore points towards a causal role for Prg4 in these processes. Based on human association studies, we expected that Prg4 deficient mice would show a difference in body weight as compared to WT mice, which we did not observe. However, this could be a timing effect since the difference between the weights increased during the last weeks of the experiment. It would therefore be interesting to investigate the effect of Prg4 deficiency on body weight development under extended HFD feeding conditions.