Rosiglitazone treatment of type 2 diabetic db/db mice attenuates urinary albumin and angiotensin converting enzyme 2 excretion

Abstract

Alterations within the renal renin angiotensin system play a pivotal role in the development and progression of cardiovascular and renal disease. Angiotensin converting enzyme 2 (ACE2) is highly expressed in renal tubules and has been shown to be renoprotective in diabetes. The protease, a disintegrin and metalloprotease (ADAM) 17, is involved in the ectodomain shedding of several transmembrane proteins including ACE2. Renal ACE2 and ADAM17 were significantly increased in db/db mice compared to controls. We investigated the effect of the insulin sensitizer, rosiglitazone, on albuminuria, renal ADAM17 protein expression and ACE2 shedding in db/db diabetic mice. Rosiglitazone treatment of db/db mice normalized hyperglycemia, attenuated renal injury and decreased urinary ACE2 and renal ADAM17 protein expression. Urinary excreted ACE2 is enzymatically active. Western blot analysis of urinary ACE2 demonstrated two prominent immunoreactive bands at approximately 70 & 90 kDa. The predominant immunoreactive band is approximately 20 kDa shorter than the one demonstrated for kidney lysate, indicating possible ectodomain shedding of active renal ACE2 in the urine. Therefore, it is tempting to speculate that renoprotection of rosiglitazone could be partially mediated via downregulation of renal ADAM17 and ACE2 shedding. In addition, there was a positive correlation between blood glucose, urinary albumin, plasma glucagon, and triglyceride levels with urinary ACE2 excretion. In conclusion, urinary ACE2 could be used as a sensitive biomarker of diabetic nephropathy and for monitoring the effectiveness of renoprotective medication.

Figure 1. Chronic treatment with rosiglitazone (10–20 mg/kg/day) normalized hyperglycemia and improved glucose tolerance in db/db mice.

(A) Non fasted blood glucose levels in control, control+rosiglitazone, db/db and db/db+rosiglitazone mice. Repeated measures two-way ANOVA using a Bonferroni’s posthoc test showed that treatment with rosiglitazone caused a significant decrease in blood glucose levels of db/db mice [F (3, 54) = 176.04], p<0.0001. Similarly, duration of treatment showed a significant decrease in blood glucose levels of db/db mice [F (21, 54) = 16.34], p<0.0001. Data are represented as mean ± SEM of group size (n = 6–8). (B) Glucose tolerance test in rosiglitazone treated and untreated lean control and db/db mice. After eight weeks of treatment with rosiglitazone, mice were fasted for 16 hours and dosed with glucose (1.5 g/kg I.P). Blood glucose levels were measured by tail tip bleed at 0, 15, 30, 60, 90 and 120 minutes post administration. *p<0.001 Vs age-matched lean control and lean control+rosiglitazone mice. ^# p<0.001, ^$ p<0.01 Vs untreated db/db mice. Data are represented as mean ± SEM of group size (n = 6–8). (C) One-way ANOVA of area under curve showed that rosiglitazone significantly improved the glucose tolerance in db/db+rosiglitazone mice compared to untreated db/db mice. *p<0.001 Vs age-matched lean control mice. ^# p<0.05 Vs untreated db/db mice. Each bar represents mean ± SEM of group size (n = 6–8).

Figure 2. Chronic treatment with rosiglitazone attenuated glucose excretion, albuminuria and plasma creatinine levels in db/db mice.

(A) Urinary glucose excretion in lean control, db/db and db/db+rosiglitazone mice. One-way ANOVA showed that urinary glucose excretion increased in db/db mice compared to lean controls (*p<0.001). Eight weeks after treatment commenced there was a significant decrease in urinary glucose excretion of db/db+rosiglitazone mice compared to untreated db/db mice. #p<0.001 Vs untreated db/db mice. Each bar represents mean ± SEM of group size (n = 6–8). (B) Urinary albumin excretion in control, rosiglitazone treated and untreated db/db mice 2 wks, 4 wks, 6 wks and 8 wks after the commencement of treatment. Repeated measures two-way ANOVA using a Bonferroni’s posthoc test showed that treatment resulted in a significant decrease in urinary albumin excretion of db/db+rosiglitazone mice [F (1, 20) = 36.004], p<0.0001. Similarly, duration of treatment showed a significant decrease in urinary albumin excretion of db/db+rosiglitazone mice after 2 wks, 4 wks, 6 wks and 8 wks of treatment [F (2, 20) = 7.70], p<0.001. *p<0.05 Vs age-matched lean control mice. ^# p<0.001 Vs untreated db/db mice. ^$ p<0.05 Vs 2 wks untreated db/db mice. ^γ p<0.05 4 wks Vs 8 wks untreated db/db mice. Each bar represents mean ± SEM of group size (n = 6–7).

Figure 3. Histological analysis using PAS and Masson’s Trichrome stain after 8 weeks of treatment with rosiglitazone.

(A) Representative photomicrographs depicting PAS staining of kidney sections in control, untreated and rosiglitazone treated db/db mice. Mesangial matrix and glomerular surface areas were significantly increased in db/db mice compared to control mice. Eight weeks after treatment with rosiglitazone there was a significant decrease in relative mesangial matrix area as well as glomerular surface area compared to untreated db/db mice. *p<0.001 Vs control mice. ^# p<0.001 Vs untreated db/db mice. Each bar represents mean ± SEM of group size (n = 6–12). (B) Representative photomicrographs depicting Masson’s Trichrome staining of kidney sections in control, untreated and rosiglitazone treated db/db mice. Glomerular fibrotic areas were significantly increased in db/db mice compared to control mice. Eight weeks after treatment with rosiglitazone there was a significant decrease in glomerular fibrotic area compared to untreated db/db mice. **p<0.0001 Vs control mice. Each bar represents mean ± SEM of group size (n = 10–11).

Figure 4. Immunofluorescence of nephrin, ACE2 and ADAM17 after 8 weeks of treatment with rosiglitazone.

(A) Immunofluorescence staining for nephrin in the glomeruli of control, untreated and rosiglitazone treated db/db mice at 20× magnification. Nephrin expression was significantly decreased in db/db mice. After eight weeks of treatment with rosiglitazone there was a significant increase in nephrin expression compared to untreated db/db mice. *p<0.01 Vs control mice. ^# p<0.05 Vs untreated db/db mice. Each bar represents mean ± SEM of group size (n = 11–18). (B) Immunofluorescence staining for ACE2 in cortical tubules and glomeruli of control, untreated and rosiglitazone treated db/db mice at 20× magnification. White arrows indicate glomeruli. While tubular ACE2 expression was increased, glomerular ACE2 expression was significantly decreased in db/db mice. After eight weeks of treatment with rosiglitazone there was a significant increase in glomerular ACE2 expression while tubular ACE2 expression was unchanged compared to untreated db/db mice. *p<0.001 Vs control mice. ^# p<0.01 Vs untreated db/db mice. Each bar represents mean ± SEM of group size (n = 11–18). (C) Immunofluorescence staining for ADAM17 in cortical tubules of control, untreated and rosiglitazone treated db/db mice at 20× magnification. (D) Immunofluorescence double staining for ACE2 and ADAM17 in cortical tubules of db/db mice at 60× magnification.

Figure 5. Western blot analysis for ACE2 protein expression in urine and ACE2, ADAM17 and TIMP3 protein expression in the kidney of db/db, db/db+rosiglitazone mice.

(A) Western blot analysis for ACE2 protein expression in urine. Lane 1 represents mouse kidney lysate (positive control), lane 2 and 3 represent urine samples from 6 wks old control mice, lanes 4 and 5 represent urine samples from 6 wks old db/db mice, lanes 6 and 7 represent urine samples from 7 wks old db/db+rosiglitazone mice. Immunoreactive band for ACE2 was observed at 90 kDa in mouse kidney lysate (lane 1), but was seen at ∼70 kDa in the urine (lanes 4 and 5). One-way ANOVA showed that urinary ACE2 protein excretion significantly increased in db/db mice compared to their age-matched lean control mice. Rosiglitazone treatment decreased urinary ACE2 protein excretion compared to untreated db/db mice. *p<0.01 Vs age-matched lean control mice. ^# p<0.05 Vs untreated db/db mice. Each bar represents mean ± SEM of group size (n = 6–8). (B) Western blot analysis for ACE2 protein expression in kidney. One-way ANOVA showed that ACE2 protein expression was significantly increased in db/db mice compared to their age-matched lean control mice. There was no significant difference in the ACE2 protein expression in the treated db/db mice compared to untreated db/db mice after eight weeks of treatment. Each bar represents mean ± SEM of group size (n = 4–5). (C) Western blot analysis for renal ADAM17 protein expression. One-way ANOVA showed that renal ADAM17 significantly increased in db/db mice compared to their age-matched lean control mice. ADAM17 protein expression was significantly decreased in treated db/db mice compared to untreated db/db mice after eight weeks of treatment. *p<0.05 Vs age-matched lean control mice. Each bar represents mean ± SEM of group size (n = 3–5). (D) Western blot analysis for renal TIMP3 protein expression. One-way ANOVA showed that renal TIMP3 significantly decreased in untreated and treated db/db mice compared to 30 wks old lean control mice. *p<0.01 Vs lean control mice. Each bar represents mean ± SEM of group size (n = 6).

Figure 6. ACE2 and ACE activity in urine, plasma and kidney of control, db/db and db/db+rosiglitazone mice using a fluorometric enzyme assay.

(A) Urinary ACE2 activity in control, db/db and db/db+rosiglitazone mice before and after the commencement of treatment. Two-way ANOVA showed an increase in urinary ACE2 activity of db/db mice compared to control mice. Four and eight weeks after treatment commenced there was a significant decrease in urinary ACE2 activity of the db/db+rosiglitazone mice compared to untreated db/db mice. *p<0.001 Vs control mice. ^# p<0.001 Vs untreated db/db mice. Each bar represents mean ± SEM of group size (n = 6–7). (B) Plasma and renal ACE2 activity in control, db/db and db/db+rosiglitazone mice. There was no plasma ACE2 activity in control and db/db mice but a significant increase in renal ACE2 activity of db/db mice compared to control mice was observed. Treatment with rosiglitazone had no significant effect on renal ACE2 activity of treated db/db mice compared to untreated db/db mice.*p<0.05 Vs control kidney. Each bar represents mean ± SEM of group size (n = 5–8). (C) Plasma ACE activity in control, db/db and db/db+rosiglitazone mice 8 wks after the commencement of treatment. One-way ANOVA showed an increase in plasma ACE activity of db/db mice compared to control mice. Eight weeks after treatment commenced there was a significant decrease in plasma ACE activity of the db/db+rosiglitazone mice compared to untreated db/db mice. *p<0.05, **p<0.001 Vs control mice. ^# p<0.05 Vs untreated db/db mice. Each bar represents mean ± SEM of group size (n = 6–7).

Figure 7. Mass spectrometric analysis of ACE2 activity in urine from control, db/db and db/db+rosiglitazone mice.

Urine (2 µl) was incubated for 1.5 h at 37°C in 50 mM MES buffer pH 6.75 containing 0.5 µM Ang II, 2 mM PMSF and 20 µM bestatin. Shown is the conversion of Ang II (m/z 1046) to Ang-(1–7) (m/z 899). (A) Urinary ACE2 activity in control mice. (B) Urinary ACE2 activity in db/db mice. (C) Urinary ACE2 activity in db/db mice treated with rosiglitazone. (D) Urinary ACE2 activity in db/db mice in incubations with the ACE2 inhibitor, MLN-4760. (E) MS/MS of enzymatically generated Ang-(1–7) (upper panel) and synthetic Ang-(1–7) (lower panel).

Figure 8. Linear regression analysis between urinary ACE2 activity and urinary albumin, plasma glucagon, plasma triglycerides and blood glucose levels.

(A) Association of urinary ACE2 activity and urinary albumin excretion in control, db/db and db/db+rosiglitazone mice before and 4 or 8 weeks after commencement of rosiglitazone treatment. (B) Correlation between plasma glucagon and urinary ACE2 activity in control, db/db and db/db+rosiglitazone mice 8 weeks after commencement of rosiglitazone treatment. (C) Correlation between plasma triglycerides and urinary ACE2 activity in control, db/db and db/db+rosiglitazone mice 8 weeks after commencement of rosiglitazone treatment. (D) Correlation between urinary ACE2 activity and non fasted blood glucose levels in control, db/db and db/db+rosiglitazone mice 8 weeks after commencement of rosiglitazone treatment.