The heme oxygenase system suppresses perirenal visceral adiposity, abates renal inflammation and ameliorates diabetic nephropathy in Zucker diabetic fatty rats

Abstract

The growing incidence of chronic kidney disease remains a global health problem. Obesity is a major risk factor for type-2 diabetes and renal impairment. Perirenal adiposity, by virtue of its anatomical proximity to the kidneys may cause kidney disease through paracrine mechanisms that include increased production of inflammatory cytokines. Although heme-oxygenase (HO) is cytoprotective, its effects on perirenal adiposity and diabetic nephropathy in Zucker-diabetic fatty rats (ZDFs) remains largely unclear. Upregulating the HO-system with hemin normalised glycemia, reduced perirenal adiposity and suppressed several pro-inflammatory/oxidative mediators in perirenal fat including macrophage-inflammatory-protein-1α (MIP-1α), endothelin (ET-1), 8-isoprostane, TNF-α, IL-6 and IL-1β. Furthermore, hemin reduced ED1, a marker of pro-inflammatory macrophage-M1-phenotype, but interestingly, enhanced markers associated with anti-inflammatory M2-phenotype such as ED2, CD206 and IL-10, suggesting that hemin selectively modulates macrophage polarization towards the anti-inflammatory M2-phenotype. These effects were accompanied by increased adiponectin, HO-1, HO-activity, atrial-natriuretic peptide (ANP), and its surrogate marker, urinary-cGMP. Furthermore, hemin reduced renal histological lesions and abated pro-fibrotic/extracellular-matrix proteins like collagen and fibronectin that deplete nephrin, an important transmembrane protein which forms the scaffolding of the podocyte slit-diaphragm allowing ions to filter but not massive excretion of proteins, hence proteinuria. Correspondingly, hemin increased nephrin expression in ZDFs, reduced markers of renal damage including, albuminuria/proteinuria, but increased creatinine-clearance, suggesting improved renal function. Conversely, the HO-blocker, stannous-mesoporphyrin nullified the hemin effects, aggravating glucose metabolism, and exacerbating renal injury and function. The hemin effects were less-pronounced in Zucker-lean controls with healthy status, suggesting greater selectivity of HO in ZDFs with disease. We conclude that the concomitant reduction of pro-inflammatory/oxidative mediators, macrophage infiltration and profibrotic/extracellular-matrix proteins, coupled to increased nephrin, adiponectin, ANP, cGMP and creatinine clearance may account for improved renal function in hemin-treated ZDFs. These findings suggest that HO-inducers like hemin may be explored against the co-morbidity of perirenal adiposity and diabetic nephropathy.

Figure 1. Effects of hemin, the HO inducer and SnMP, the HO inhibitor on HO-1 and HO activity of parirenal adipose tissue from ZDF and ZL rats.

(A) The basal HO-1 levels in ZDF rats were lower than in age/sex-matched ZL-control rats, but were increased by hemin, whereas SnMP nullified the hemin effect. (B) The basal HO activity in ZDF rats was depressed as compared to ZL-control rats. Treatment with hemin markedly enhanced HO activity, whereas SnMP annulled the hemin effect. Hemin also enhanced HO-1 and HO activity in ZL rats, though less effectively as compared to ZDF rats. Bars represent means ± SEM; n = 6 rats per group (*p<0.05 vs all groups, ^†p<0.05, ^††p<0.01 vs all groups; ^§p<0.05 vs all groups).

Figure 2. Effects of hemin, the HO inducer and SnMP, the HO inhibitor on 8-isoprostane and ET-1 of the parirenal adipose tissue from ZDF and ZL rats.

(A) The basal 8-isoprostane levels in ZDF rats were markedly elevated as compared to ZL-control rats, but were significantly reduced by hemin, whereas SnMP nullified the hemin effect. (B) The basal ET-1 levels in ZDF rats were significantly elevated as compared to ZL-control rats, but were reduced by hemin, whereas SnMP nullified the hemin effect. Hemin also reduced 8-isoprostane and ET-1 in ZL rats, but less effectively as compared to ZDF rats. Bars represent means ± SEM; n = 6 rats per group (*p<0.05 vs all groups, ^†p<0.05, ^††p<0.01 vs all groups; ^§p<0.05 vs all groups).

Figure 3. Effects of hemin, the HO inducer and SnMP, the HO inhibitor on TNF-α, IL-6 and IL-1β, of the parirenal adipose tissue from ZDF and ZL rats.

Hemin therapy significantly reduced the elevated levels of (A) TNF-α, (B) IL-6 and (C) IL-1β in ZDF rats, but the hemin effects were reversed by co-treatment with the HO blocker SnMP, while treatment with SnMP alone further increased the levels. Hemin also reduced TNF-α, IL-6 and IL-1β in ZL rats, but less effectively as compared to ZDF rats. Bars represent means ± SEM; n = 6 rats per group (*p<0.05, **p<0.01 vs all groups; ^†p<0.05, ^††p<0.01 vs all groups; ^§p<0.05, ^§§p<0.01 vs all groups).

Figure 4. Effect of hemin, the HO inducer and SnMP, the HO inhibitor on plasma ANP, urinary cGMP and plasma adiponectin in ZDFs and ZLs.

Hemin therapy significantly increased the depressed basal levels of (A) ANP, (B) urinary cGMP and (C) plasma adiponectin in ZDF rats, but was reversed by co-treatment with the HO blocker SnMP, while treatment with SnMP alone further depleted the basal levels. Hemin also increased plasma ANP, urinary cGMP and plasma adiponectin in ZL rats. Bars represent means ± SEM; n = 6 rats per group (*p<0.05 vs all groups, ^†p<0.01 vs all groups, ^§p<0.05 vs all groups).

Figure 5. Effect of hemin on macrophage-inflammatory-protein-1 alpha (MIP-1α) in perirenal adipose tissue and the kidneys of ZDF.

Hemin therapy significantly reduced the elevated levels of MIP-1α in (A) perirenal adipose tissues and (B) kidney from ZDF, but the hemin-effect was annulled by co-treatment with the HO blocker SnMP, while treatment with SnMP alone further increased the levels Bars represent means ± SEM; n = 6 rats per group (*p<0.01 vs all groups; ^†p<0.05, vs all groups; ^§p<0.01 vs all groups).

Figure 6. Effect of hemin on the expression of ED-1, ED2, CD206 and IL10 in renal tissues of ZDF.

Representative Western immunoblots and relative densitometry indicates that hemin therapy significantly (A) reduced ED-1, but (B) enhanced ED2, (C) increased CD206, and (D) enhanced IL10 expression in ZDF. Bars represent means ± SEM; n = 4 rats per group (*p<0.05, **p<0.01 vs ZL-Control; ^§p<0.05, ^§§p<0.01 vs ZL-Control; ^#p<0.01 vs ZL-Control; ^@p<0.01 vs ZDF-Control).

Figure 7. Effect of hemin on the expression of collagen-IV, fibronectin and nephrin in renal tissues of ZDF.

Representative Western immunoblots and relative densitometry indicates that hemin therapy significantly (A) enhanced the expression of nephrin but, (B) abated collagen-IV expression, and (C) reduced the expression of fibronectin in ZDF. Bars represent means ± SEM; n = 4 rats per group (*p<0.01 vs all groups; ^#p<0.05, ^##p<0.01 vs ZL-Control; ^@p<0.01 vs ZDF-Control).

Figure 8. Masson’s trichrome staining of collagen deposition and fibrosis the in kidney.

(A) Representative images of kidney section from two different rats. Sections from untreated ZDF-controls (panels A-iii and A-iv) indicate severe fibrosis in tubulointerstitial, perivascular and glomerulus as compared with ZL-control rats (panels A-i and A-ii), which interestingly were attenuated by hemin (panels A-v and A-vi). (Magnification×200) (B) Semi-quantitative evaluation showed that hemin reduced collagen deposition. Bars represent means ± SEM; n = 4–6 rats per group (*p<0.05 vs all groups).

Figure 9. Effect of hemin therapy on kidney macrophage infiltration (A) Representative images of kidney section from different rats.

The images reveal that macrophage infiltration (ED1-positive cells stained dark brown in kidney sections were elevated in ZDF-controls (panels A-iii and A-iv) as compared to ZL-controls (panels A-i and A-ii), but interestingly were reduced by hemin (panels A-v and A-vi). (Magnification×200). (B) Quantitative analyses per field indicating that in ZDF-controls macrophage infiltration was significantly elevated as compared to ZL-control, but was significantly attenuated by hemin therapy. Bars represent means ± SEM; n = 4–6 rats per group (*p<0.01 vs all groups).

Figure 10. Immunolabelling of HO-1 in the kidney of ZDF-control and ZDF-treated with hemin therapy.

Representative images of kidney section from different rats reveal that HO-1 is more expressed in ZDF+hemin group (panels iv–vi) as compared to the ZDF-control group (panels i–iii). (Magnification×200).