NF-κB System Is Chronically Activated and Promotes Glomerular Injury in Experimental Type 1 Diabetic Kidney Disease

Abstract

High glucose concentration can activate TLR4 and NF-κB, triggering the production of proinflammatory mediators. We investigated whether the NF-κB pathway is involved in the pathogenesis and progression of experimental diabetic kidney disease (DKD) in a model of long-term type 1 diabetes mellitus (DM). Adult male Munich-Wistar rats underwent DM by a single streptozotocin injection, and were kept moderately hyperglycemic by daily insulin injections. After 12 months, two subgroups - progressors and non-progressors - could be formed based on the degree of glomerulosclerosis. Only progressors exhibited renal TLR4, NF-κB and IL-6 activation. This scenario was already present in rats with short-term DM (2 months), at a time when no overt glomerulosclerosis can be detected. Chronic treatment with the NF-κB inhibitor, pyrrolidine dithiocarbamate (PDTC), prevented activation of renal TLR4, NF-κB or IL-6, without interfering with blood glucose. PDTC prevented the development of glomerular injury/inflammation and oxidative stress in DM rats. In addition, the NF-κB p65 component was detected in sclerotic glomeruli and inflamed interstitial areas in biopsy material from patients with type 1 DM. These observations indicate that the renal NF-κB pathway plays a key role in the development and progression of experimental DKD, and can become an important therapeutic target in the quest to prevent the progression of human DKD.

Keywords: NF-κB; diabetic kidney disease; glomerulosclerosis; innate immunity; pyrrolidine dithiocarbamate.

FIGURE 1

Time course of (A) blood glucose concentration (BG, mg/dL) in diabetic (DM, n = 32) and control (C, n = 29) rats. (B) shows the frequency of glomeruli with sclerotic lesions (GS,%) in C and DM at 12 months. Results expressed as means ± SE. * p < 0.05 vs. C.

FIGURE 2

Twelve months after STZ injection, the 10 DM rats with the highest values for percent glomerulosclerosis (%GS) were used to create the DKD+ group, whereas the 10 DM rats with the lowest %GS values constituted the DKD– group. Twelve non-diabetic age-matched rats were taken as controls (C). The time course of blood glucose concentration (BG, mg/dL) is shown in panel (A), whereas the %GS and glomerular macrophage density (ED-1, cells/mm²) at 12 months of DM appear at panels (B,C), respectively. Representative microphotographs of renal tissue stained with PAS (for %GS) or immunohistochemistry (for macrophage-specific ED-1 antigen) are shown in panel (D) (x400). Results expressed as means ± SE. * p < 0.05 vs. C, ^#p < 0.05 vs. DKD–.

FIGURE 3

Diabetic rats with (DKD+ group, n = 10) or without (DKD– group, n = 10) significant glomerular injury were analyzed 12 months after streptozotocin injection. Twelve non-diabetic age-matched rats were taken as controls (C group). The renal cortical contents of (A) Toll-like receptor 4 (TLR4), (B) nuclear fraction of phosphorylated NF-κB component (p65), and (C) interleukin 6 (IL-6), were quantified using (D) western blot analysis. Panel (E) shows the percent glomerular area staining for NLRP3 (%) quantified by immunohistochemistry. (F) Representative microphotographs show the presence of p65-positive or NLRP3-positive immunostaining (brown) in glomeruli (×400). No difference in NLRP3 positivity was observed between groups DKD– (non-progressors) and DKD+ (progressors). Results expressed as means ± SE. * p < 0.05 vs. C, ^#p < 0.05 vs. DKD–.

FIGURE 4

Blood glucose concentration (BG, mg/dL) and frequency of glomeruli with sclerotic lesions (GS,%), observed in Control (C_2m group, n = 5) and DM (DM_2m group, n = 7) rats 2 months after STZ injection, are shown in panels (A,B), respectively. The renal cortical contents of Toll-like receptor 4 (TLR4) (C), nuclear fraction of phosphorylated p65 (D) and interleukin-6 (IL-6) (E) were quantified using Western blot analysis (F). In panel (G), representative microphotographs show the presence of p65-positive immunostaining (brown) in glomeruli (×400). Results expressed as means ± SE. * p < 0.05 vs. C_2m.

FIGURE 5

DM rats treated with the NF-κB inhibitor pyrrolidine dithiocarbamate (PDTC) (DM+PDTC group, n = 11) or vehicle (DM+V, n = 16), were followed for 12 months. At the end of this period, the renal cortical content of (A) Toll-like receptor 4 (TLR4), (B) nuclear fraction of phosphorylated p65, (C) interleukin 6 (IL-6), and (D) high mobility group box protein-1 (HMGB1) were quantified using (E) Western blot analysis. Values obtained from groups DM+V and DM+PDTC were factored by age-matched controls rats (C, horizontal dotted lines). Results expressed as means ± SE. * p < 0.05 vs. DM+V.

FIGURE 6

Time course of (A) body weight (BW, g) and (B) blood glucose concentration (BG, mg/dL) in rats treated with the NF-κB inhibitor pyrrolidine dithiocarbamate (Group DM+PDTC, n = 11) or vehicle (Group DM+V, n = 16) for 12 months. In panel (C), illustrative microphotographs (×400) of glomerulosclerosis in PAS-stained renal tissue and detection, by immunostaining, of zonula occludens-1 (ZO-1) (brown), and macrophage (ED-1) (red) in renal tissue at 12 months of DM. Quantification of (D) frequency of glomeruli with sclerotic lesions (GS,%), (E) percent area of ZO-1 and (F) macrophage density (ED-1, cells/mm²) in the glomerular area at this time. Results expressed as means ± SE. * p < 0.05 vs. DM+V.

FIGURE 7

Renal cortical abundance of (A) heme oxygenase-1 (HO-1) and (B) superoxide dismutase 2 (SOD2), quantified by Western blot analysis (C) in rats treated with the NF-κB inhibitor pyrrolidine dithiocarbamate (Group DM+PDTC, n = 11) or vehicle (Group DM+V, n = 16) at 12 months of DM. The values obtained from the groups DM+V and DM+PDTC were factored by age-matched controls rats (C, horizontal dotted lines). Results expressed as means ± SE. * p < 0.05 vs. DM+V.

FIGURE 8

Microphotographs from Type 1 diabetic patients with advanced diabetic kidney disease, immunostained for p65 (brown). The p65 protein was detected in a relatively preserved glomerulus (A) and around Kimmelstiel-Wilson nodules in a more severely affected tuft (B), as well as in inflamed interstitial areas (A,B), whereas in normal renal tissue (C) only scattered tubular profiles were stained. Magnification: x200 (B,C) and x400 (A and insets).