1,25-Dihydroxyvitamin D3 suppresses high glucose-induced angiotensinogen expression in kidney cells by blocking the NF-{kappa}B pathway

Abstract

The renin-angiotensin system (RAS) is a major mediator of renal injury in diabetic nephropathy. Our previous studies demonstrated that 1,25-dihydroxyvitamin D(3) [1,25(OH)(2)D(3)] plays a renoprotective role by suppressing the RAS, with renin and angiotensinogen (AGT) as the main targets. The mechanism whereby 1,25(OH)(2)D(3) transcriptionally suppresses renin gene expression has been elucidated; however, how vitamin D regulates AGT remains unknown. Exposure of mesangial cells or podocytes to high glucose (HG; 30 mM) markedly stimulated AGT expression. In mesangial cells, the stimulation was inhibited by 1,25(OH)(2)D(3) (20 nM) or NF-kappaB inhibitor BAY 11-7082, suggesting the involvement of NF- kappaB in HG-induced AGT expression and the interaction between 1,25(OH)(2)D(3) and NF-kappaB in the regulation. Plasmid pNF-kappaB-Luc luciferase reporter assays showed that 1,25(OH)(2)D(3) blocked HG-induced NF-kappaB activity. EMSA and ChIP assays demonstrated increased p65/p50 binding to a NF-kappaB binding site at -1734 in the AGT gene promoter upon high glucose stimulation, and the binding was disrupted by 1,25(OH)(2)D(3) treatment. Overexpression of p65/p50 overcame 1,25(OH)(2)D(3) suppression, and mutation of this NF-kappaB binding site blunted 1,25(OH)(2)D(3) suppression of the promoter activity. In mice lacking the vitamin D receptor, AGT mRNA expression in the kidney was markedly increased compared with wild-type mice, and AGT induction in diabetic mice was suppressed by treatment with a vitamin D analog. These data indicate that 1,25(OH)(2)D(3) suppresses hyperglycemia-induced AGT expression by blocking NF-kappaB-mediated pathway.

Fig. 1.

1,25-Dihydroxyvitamin D₃ [1,25(OH)₂D₃] suppresses high glucose (HG)-induced angiotensinogen (AGT) mRNA expression. Mesangial cells (MCs) or podocytes cultured in low-glucose (LG; 5 mM) or HG (30 mM) media were exposed to 20 nM 1,25(OH)₂D₃ (1,25D or +D) as indicated. Total RNAs were extracted 24 h later. A: RT-PCR determination of AGT mRNA levels in MC or podocytes as indicated. B: quantitation of AGT mRNA levels based on the RT-PCR results. #P < 0.05 vs. LG. *P < 0.05 vs. corresponding untreated samples. C: immunostaining of MCs with ANG I/II-specific antibody. MCs were cultured in LG or HG media in the presence of absence of 20 nM 1,25D (+D) for 24 h. D: MCs were exposed to LG, HG, or high mannitol (5 mM LG + 25 mM mannitol) as indicated for 24 h, and AGT mRNA expression was determined by RT-PCR. E: quantitation of AGT mRNA levels based on the RT-PCR results in D. *P < 0.05, **P < 0.01 vs. LG.

Fig. 2.

HG induction of AGT is blocked by NF-κB inhibitor. A: MCs were cultured in LG (5 mM) or HG (30 mM) and were treated with NF-κB-specific inhibitor Bay 11–7082 (Bay) for 24 h, and AGT mRNA levels were determined by RT-PCR. B: luciferase reporter assays. MCs were transfected with pNF-κB-Luc reporter plasmid in serum-free LG media. After 6 h, the cells were exposed to LG or HG media containing 10% FBS in the presence of 20 nM 1,25(OH)₂D₃ (+D) or 2 μM Bay. Luciferase activity was determined after 24 h. #P < 0.01 vs. LG. *P < 0.05 vs. corresponding untreated sample.

Fig. 4.

1,25(OH)₂D₃ inhibits HG-induced p65 and p50 binding to the NF-κB site in the AGT promoter. A: schematic illustration of the NF-κB site and the location of the primers used in the ChIP assay. B: ChIP assays. MCs were pretreated with or without 20 nM 1,25(OH)₂D₃ (1,25D) for 24 h, and the cells were then exposed to LG (5 mM) or HG (30 mM) media for 30 min before the ChIP assays. Sonicated chromatins were precipitated with anti-p65 or anti-p50 antibody. Input (Inp) represents 5% total chromatin isolated from the cells. IgG, chromatin precipitated with nonimmune IgG.

Fig. 3.

1,25(OH)₂D₃ suppresses HG-induced binding of NF-κB in the AGT gene promoter. A: EMSA shows specific binding of nuclear proteins to the AGT NF-κB site through cold competition. In the EMSAs nuclear extracts isolated from MCs were incubated with ³²P-labeled NF-κB probes in the presence or absence of an excess amount (25-, 50-, or 100-fold) of unlabeled probe of AGT NF-κB sequence (NF-κB), mutant NF-κB (NF-κBmut), or canonical NF-κBc (NF-κBc) as shown at bottom of gel. C, no competition control. B: blocking of HG-induced NF-κB p65/p50 DNA binding by 1,25(OH)₂D₃. MCs cultured in LG (5 mM) or HG (30 mM) media in the presence or absence of 20 nM 1,25(OH)₂D₃ (1,25D) for 24 h. Nuclear extracts were prepared. EMSA was performed with ³²P-labeled NF-κB probe. Some reactions were incubated in the presence of antibody against p65 or p50 as indicated. Arrows indicate p65/p50-DNA complex and supershift bands as indicated.

Fig. 5.

1,25(OH)₂D₃ suppresses HG-induced AGT promoter activity. A: pAGT-Luc and pAGTmut-Luc constructs illustrating the NF-κB site and its mutation. B: luciferase reporter assays. MCs were transfected with wild-type (WT) pAGT-Luc or pAGTmut-Luc (2 mutant clones). The transfected cells were cultured in LG or HG media in the presence or absence of 20 nM 1,25(OH)₂D₃ (+D). Luciferase activity was determined after 24 h. C: overcome of vitamin D suppression by p65/p50 overexpression. MCs were cotransfected with pAGT-Luc and plasmids expressing p65 and p50, pcDNA-p65 and pcDNA-p50, respectively. The cells were grown in LG or HG media in the presence or absence of 20 nM 1,25(OH)₂D₃ (+D) before luciferase activity was assayed. #P < 0.05 vs. LG. *P < 0.05 vs. corresponding untreated samples.

Fig. 6.

Regulation of renal AGT expression in vivo. A: effect of vitamin D receptor (VDR) inactivation on AGT expression in the kidney. Renal AGT mRNA levels in 4-mo-old WT and VDR(−/−) mice at baseline were measured by RT-PCR. Total RNAs were extracted from whole kidneys. *P < 0.05 vs. WT; n = 4 in each genotype. B: renal AGT mRNA expression in diabetic WT mice with or without paricalcitol (Pari; 0.4 μg/kg) treatment for 10 wk starting from week 3. The mice were induced to diabetes with streptozotocin (STZ) for 13 wk before death. Total RNAs were extracted from whole kidneys. *P < 0.05 vs. C. #P = 0.05 vs. STZ; n = 4 in each genotype.