Induction of Autophagy by Pterostilbene Contributes to the Prevention of Renal Fibrosis via Attenuating NLRP3 Inflammasome Activation and Epithelial-Mesenchymal Transition

Abstract

Chronic kidney disease (CKD) is recognized as a global public health problem. NLRP3 inflammasome activation has been characterized to mediate diverse aspect mechanisms of CKD through regulation of proinflammatory cytokines, tubulointerstitial injury, glomerular diseases, renal inflammation, and fibrosis pathways. Autophagy is a characterized negative regulation mechanism in the regulation of the NLRP3 inflammasome, which is now recognized as the key regulator in the pathogenesis of inflammation and fibrosis in CKD. Thus, autophagy is undoubtedly an attractive target for developing new renal protective treatments of kidney disease via its potential effects in regulation of inflammasome. However, there is no clinical useful agent targeting the autophagy pathway for patients with renal diseases. Pterostilbene (PT, trans-3,5-dimethoxy-4-hydroxystilbene) is a natural analog of resveratrol that has various health benefits including autophagy inducing effects. Accordingly, we aim to investigate underlying mechanisms of preventive and therapeutic effects of PT by reducing NLRP3 inflammasome activation and fibrosis through autophagy-inducing effects. The renal protective effects of PT were evaluated by potassium oxonate (PO)-induced hyperuricemia and high adenine diet-induced CKD models. The autophagy induction mechanisms and anti-fibrosis effects of PT by down-regulation of NLRP3 inflammasome are investigated by using immortalized rat kidney proximal tubular epithelial NRK-52E cells. To determine the role of autophagy induction in the alleviating of NLRP3 inflammasome activation and epithelial-mesenchymal transition (EMT), NRK-52E with Atg5 knockdown [NRK-Atg5-(2)] cells were applied in the study. The results indicated that PT significantly reduces serum uric acid levels, liver xanthine oxidase activity, collagen accumulation, macrophage recruitment, and renal fibrosis in CKD models. At the molecular levels, pretreatment with PT downregulating TGF-β-triggered NLRP3 inflammasome activation, and subsequent EMT in NRK-52E cells. After blockage of autophagy by treatment of Atg5 shRNA, PT loss of its ability to prevent NLRP3 inflammasome activation and EMT. Taken together, we suggested the renal protective effects of PT in urate nephropathy and proved that PT induces autophagy leading to restraining TGF-β-mediated NLRP3 inflammasome activation and EMT. This study is also the first one to provide a clinical potential application of PT for a better management of CKD through its autophagy inducing effects.

Keywords: NLRP3 inflammasome; autophagy; epithelial-mesenchymal transition; pterostilbene; renal fibrosis.

FIGURE 1

Urate lowering effects of PT in a hyperuricemia animal model. (A) Scheme of PO-induced hyperuricemia model as described in the Materials and Methods. (B) UA concentration and (C) liver XOD activities of mice treated with Control, PO (400 mg/kg), PO + AP (10 mg/kg), or PO + PT (200 mg/kg) for 7 days (n = 3 in each group). Data represent the mean ± SD. *p < 0.05 compared with the Control groups. ^#p < 0.05 compared with PO groups. PO: potassium oxonate, PT: pterostilbene, AP: allopurinol, and XOD: xanthine oxidase.

FIGURE 2

Renal protective effects of PT in a high adenine-induced CKD model. (A) The high adenine-induced CKD model was performed as described in Materials and Method. (B) Changes of body weight, (C) daily food intake, and (D) daily water intake are demonstrated for ICR mice fed with 0.9% saline (Control), PO combined with 0.175% adenine (PO + AD), AP (10 mg/kg) combined with PO + AD (PO + AD + AP), or PT (200 mg/kg) combined with PO + AD (PO + AD + PT). Effects of PT on (E) 24-h urine output and (F) kidney relative weight (kidney weight/final body weight × 100) in mice. Each column and vertical bar are mean ± SD of 5 mice/group. *p < 0.05 compared with the control groups.

FIGURE 3

Improvement of renal function by PT in the CKD model. Serum and urine levels of UA (A,B), creatinine (CRE; C,D), blood urea nitrogen (BUN; F,G), and clearance rate of CRE (E), and BUN (H) in the CKD model. Data were expressed as mean ± SD of 5 mice per group. *p < 0.05 compared with the control groups. #p < 0.05 compared with PO + AD groups.

FIGURE 4

PT alleviates inflammation and interstitial fibrosis in kidney tissues in the CKD model. (A) The size of kidney was increased in PO + AD groups. The appearance of kidney in PO + AD groups was rough and pale. (B) Renal tubular injury was assessed by H&E staining. The deposition of fibrosis in renal tissues was determined by Masson Trichrome staining. Blue color represents collagen fibers, red color represents muscle fibers. Infiltration of macrophages was detected by CD68 staining (brown color, arrow indicated). (C) Tubulointerstitial inflammation and EMT was assessed by staining of TGF-β, Vimentin, and E-cadherin. The results of immunohistochemistry were quantified by ImagJ (n = 3). Bar = 100 μm. *p < 0.05 compared with the control groups. #p < 0.05 compared with PO + AD groups. (D) Compared with PO + AD groups, PT increased the expression of E-cadherin, and decreased the expression of fibronectin, α-SMA, and Vimentin in renal tissues detected by Western blotting analysis. GAPDH was used as an internal control. Representative data from one of three independent experiments are shown. The number below each line indicates the relative intensity of protein expression compared to the control (defined as 1; Figures 4–7).

FIGURE 5

PT inhibits TGF-β-triggered EMT in NRK-52E cells. NRK-52E cells were treated with (A) 0, 0.5, 1, or 2.5 μM PT, or (B) 0, 2.5, 5, or 10 ng/ml TGF-β for 24 or 48 h. Cell viability were detected by MTT assay. Mean ± SD; n = 3. (C) NRK-52E cells were treated with DMSO (Control), PT 2 μM, TGF-β 2.5 ng/ml, or PT combined with TGF-β for 72 h. The morphological changes of NRK-52E cells were recorded under a phase-contrast microscopy. F-actin/ZO-1 co-staining, or Vimentin staining were determined by immunofluorescence staining. Bar = 100 μm (D) The expression of E-cadherin, fibronectin, α-SMA, and Vimentin were determined by Western blotting analysis in NRK-52E cells treated with PT, TGF-β, or PT + TGF-β. The membrane was probed with anti-GAPDH to confirm equal loading of proteins. Immunoblots are representative of at least three independent experiments.

FIGURE 6

PT and MCC950 attenuates TGF-β-induced NLRP3 inflammasome activation and EMT in NRK-52E cells. (A) Immunoblotting for NLRP3 inflammasome components NLRP3, ASC, caspase-1, degraded form of caspase-I, IL-1β, and degraded form of IL-1β in NRK-52E cells following treatment with DMSO (Control), PT 2 μM, TGF-β 2.5 ng/ml, or PT combined with TGF-β for 48 h (PT + TGF-β). (B) Immunoblotting for NLRP3, ASC, caspase-1, degraded form of caspase-1, and EMT markers fibronectin and α-SMA at 48 h following treatment with DMSO (Control), TGF-β 2.5 ng/ml, TGF-β combined with NLRP3 inhibitor MCC950 (MCC 10 nM), or MCC in NRK-52E cells. The membrane was probed with anti-GAPDH to confirm equal loading of proteins. Immunoblots are representative of at least three independent experiments.

FIGURE 7

Autophagy inducing effects of PT in NRK-52E cells. (A) PT 2 μM induced activation of AKT followed by inhibition of mTOR pathways in a time-dependent manner. (B) Autophagic markers p62 and LC3-II were determined using Western blotting analysis in NRK-52E cells treated with PT for 24, 48, and 72 h. (C) Immunofluorescence staining showed the autolysosomes co-stained with LC3-II and lysosomal marker LAMP-1 in PT treated groups (arrow indicated). Bars = 100 mm. (D) The induction of autophagy was measured by acidic vesicular organelles (AVOs) using flow cytometry. Quantification of AVOs in NRK-52E cells treated with DMSO (Control), PT 2 μM, TGF-β 2.5 ng/ml, or PT combined with TGF-β for 48 h. The data represent the means ± SD of three independent experiments; *P < 0.05 compared with Control groups; and ^#P < 0.05 compared with TGF-β groups. (E) Immunoblotting for autophagy markers p62, and LC3-II in NRK-52E cells followed by the treatment of DMSO (Control), PT 2 μM, TGF-β 2.5 ng/ml, or PT + TGF-β for 48 h. The membrane was probed with anti-GAPDH to confirm equal loading of proteins. Immunoblots are representative of at least three independent experiments.

FIGURE 8

Role of autophagy in attenuating NLRP3 activation and EMT in response to PT treatment. (A) Atg5 expression in NRK-52E cells stably transfected with Atg5 knockdown shRNA [Atg5-(1) and Atg5-(2)]. (B) The induction of autophagy by PT was measured by flow cytometry in NRK-52E and Atg5 knockdown NRK-Atg5-(2) cells. The data represent the means ± SD of three independent experiments; *P < 0.05 compared with Control groups; ^#P < 0.05 compared with NRK-52E cells. (C) The percentage of acridine orange positive staining cells in NRK and NRK-Atg5-(2) treated with PT (2.5 μM), TGF-β, or PT + TGF-β for 48 h. (D) The protein expression of autophagy markers (p62 and LC3-II), NLRP3 inflammasome components (NLRP3, caspase-1 and degraded form of caspase-1), and EMT markers (fibronectin and α-SMA) were detected by Western blotting in NRK and NRK-Atg5-(2) cells following 48 h treatment with TGF-β (TG) or PT combined with TGF-β (P + T). The membrane was probed with anti-GAPDH to confirm equal loading of proteins. Immunoblots are representative of at least three independent experiments. (E) Proposed model for renal protective effects of PT. PT significantly reduces UA production and XOD activities, prevents renal dysfunction, and ameliorates renal fibrosis in animal models. The mechanistic studies implicate that PT induces autophagy through AMPK activation to restrain TGF–triggered NLRP3 inflammasome activation and EMT, subsequently contributing to the protection of renal fibrosis.