Proteoglycan is a member of
Proteoglycan 4 is a member of the proteoglycan family, a large group of macromolecules consisting of a protein core with negatively charged glycosaminoglycan side chains. Both the side chains as well as specific domains of the core protein can form local interactions with other molecules like extracellular matrix proteins, receptors or signaling molecules . Dependent on the tissue- and cellular localization, proteoglycans play a role in a diverse range of physiological and pathological mechanisms [8,9]. Contradicting roles have been found for proteoglycans in metabolic pathologies. For example, mice deficient for the chondroitin sulfate proteoglycan Neuron-glial antigen 2 develop adult-onset obesity, dyslipidemia and hepatic steatosis as a consequence of brown adipocyte dysfunction . In contrast, syndecan 3 deficient mice are less susceptible to diet-induced obesity due to an altered melanocortin signaling in the central nervous system .
Materials and methods
Results In human proteomic and transcriptomic studies, Prg4 is associated with obesity and related metabolic changes. To get an indication whether Prg4 in mice associated to similar processes, we measured the Prg4 mRNA expression in a subset of murine organs. Prg4 expression could be detected in most metabolically active organs, i.e. white adipose tissue and muscle, but was highest in the liver (Fig. 1A). Post-hoc analysis on previously generated liver cell fractions from chow diet-fed rats  indicated that hepatocytes primarily contribute to total liver Prg4 expression. More specifically, relative Prg4 expression levels were > 50 higher (Fig. 1B) in hepatocytes as compared to those in liver endothelial HG-9-91-01 and tissue macrophages (Kupffer cells). Importantly, as evident from Fig. 1C, challenging C57BL/6J mice with a lard-based high fat diet (HFD) significantly increased the expression of PRG4 in liver (+121%; p < 0.01), while leaving gonadal white adipose tissue PRG4 expression unaffected (p > 0.05). To study the potential relevance of the HFD-associated increase in hepatic Prg4 expression in the modulation of body metabolism, Prg4 KO mice and age- and sex-matched WT littermates were fed the same HFD for 16 weeks supplemented with fructose in the drinking water for the last 4 weeks. Prg4 KO mice showed slightly reduced, but not-significantly different, body weight gain (Fig. 2A) as compared to WT littermates (−26%; p = 0.16). The average weight at baseline for both groups was 31 g. This increased in WT mice to 44 g, while Prg4 KO mice only weighed 39 g on average at the end of the experiment. Food intake was monitored for 4 subsequent days, showing a similar food intake in Prg4 KO mice versus WT mice (WT: 3.0 ± 0.1 g/day vs Prg4 KO: 2.9 ± 0.1 g/day; p = 0.50). Since Prg4 and lipid levels in human plasma correlated in the clinical studies, we investigated the effect of Prg4 deficiency on plasma lipid levels in mice under high fat diet feeding conditions. Plasma free cholesterol levels (−20%; p < 0.001) and triglycerides (−16%; p < 0.05) were lower in Prg4 KO mice than in WT mice (Fig. 2B). Plasma cholesteryl esters were not significantly different between both groups (Fig. 2B). As evident from Fig. 2C, fasting glucose values in the circulation were not significantly different between Prg4 KO mice and WT mice. However, fasting insulin showed a clear trend towards lower levels (−43%, p = 0.07) in Prg4 KO as compared to WT (Fig. 2C). This coincided with a trend towards improved HOMA-IR score, a measure of the degree of insulin resistance, in the Prg4 KO mice as compared to WT mice (−49%; p = 0.06). To further investigate the role of Prg4 in glucose handling, mice of both genotypes were subjected to an oral glucose tolerance test (OGTT). Prg4 KO mice displayed improved glucose tolerance, with lower peak glucose values and an overall lower area-under-the-curve as compared to WT mice (AUC: −29%; p < 0.05; Fig. 2D). Importantly, Prg4 KO mice and their WT littermate controls did not already display a difference in glucose tolerance before the HFD challenge, i.e. under regular chow diet feeding conditions (OGTT AUC: 1705 ± 101 for WT (N = 7) and 1845 ± 121 for Prg4 KO (N = 7), respectively). It can therefore be suggested that the apparent difference in glucose tolerance is perhaps secondary to other Prg4 deficiency-associated effects on metabolism that are specifically induced by the HFD trigger.