PAN induced nephrosis was accompanied by loss of body
PAN-induced nephrosis was accompanied by loss of body weight, decrease of food intake, increase in urine volume and increases in triglycerides and cholesterol levels, which have been reported previously (Natori et al., 1996). Except for urine volume, each one of these measures was improved by A-306989 with either prophylactic or therapeutic treatment. These actions of AK inhibition were likely a consequence of prevention or attenuation of the renal pathology, but it cannot be ruled out that an ancillary upstream action of AK inhibition may have contributed. ADO has been associated with insulin homeostasis and lipid metabolism (Johnston-Cox et al., 2012) and, in particular, the A2BR has been implicated in regulating hyperlipidemia (Koupenova et al., 2012), and inactivation of A2A receptor results in an elevation of plasma cholesterol (Wang et al., 2009). Thus, inhibition of AK appeared to improve animal quality of life in PAN injected animals. ADO and AK inhibitors have been associated with anti-inflammatory activity in a variety of tissues including renal, and the mechanism of action may include the modulation of infiltrating leukoocytes and macrophages, as well as the associated cytokines (Awad et al., 2006). In the current PAN studies, the mRNA Ethyl 3-Aminobenzoate methanesulfonate of key inflammatory and/or fibrotic markers, including Tnfα, CD80/B7.1, collagen I and IV, serpine, Timp1, Tgfβ1, Ccl2, Ccl12, and osteopontin was upregulated in the nephrotic animals. The enhanced levels of each of these markers were significantly attenuated by prophylactic A-306989 indicating that early and sustained exposure to A-306989 protects against PAN related injury. The prevention of mRNA increase for these diverse genes following PAN injection may be due to a direct ADO action on the respective signalling or a downstream consequence of an action on a more limited set of genes that prevent further damage and upregulation of associated genes. Therapeutic dosing of A-306989 did not attenuate the mRNA expression of any gene except osetopontin, yet still significantly decreased proteinuria. Osteopontin is a proinflammatory/profibrotic cytokine that is linked to the development of proteinuria and DN (O'Regan and Berman, 2000, Susztak et al., 2004). Consistent with the current data, osteopontin has been found to be upregulated in the PAN model as well as in podocytes themselves, and osteopontin deleted animals do not develop albuminuria in response to a lipopolysaccharide challenge (Lorenzen et al., 2008). Osteopontin upregulation is also associated with glomerular and interstitial accumulation of macrophages and ensuing glomerulosclerosis (Merszei et al., 2010). Administration of an A2A receptor agonist also depressed osteopontin expression in a model of glomerulonephritis (Garcia et al., 2011). Thus, reducing osteopontin may be an important component to the observed efficacy of A-306989. A-306989 was also efficacious in the anti-GBM model of glomerulonephritis, and reduced proteinuria similar to that observed in the PAN model. Proteinuria was dramatically increased on day 4 after injection of the anti-GBM serum, and declined to lower, but still significantly high, levels on days 7 and 14. These later levels are comparable to previous reports of proteinuria in this model at these time points (Garcia et al., 2011, Suzuki et al., 2013). At each time point, both doses of A-306989 significantly decreased proteinuria to levels that were comparable to the non-specific immunosuppressant drug cyclosporine. Additionally, macrophage infiltration into glomeruli is a key pathophysiological event in this disease (Garcia et al., 2008), and depletion of renal macrophages reduces glomerular injury and proteinuria (Garcia et al., 2008, Garcia et al., 2011). Renal macrophages were elevated in anti-GBM treated rats compared to sham controls, and these enhanced levels were decreased in animals administered with A-306989. In line with our findings, another AK inhibitor ABT-702 also reduced renal infiltration of macrophages in the STZ model of DN (Pye et al., 2014).