Volume 30 Issue 2 - February 5, 2016 PDF
Albumin stimulates renal tubular inflammation through an HSP70-TLR4 axis in mice with early diabetic nephropathy
Huei-Fen Jheng1,2, Pei-Jane Tsai1,3, Yi-Lun Chuang4, Yi-Ting Sheng5, Ting-An Tai4, Wen-Chung Chen6,Chuan-Kai Chou7, Li-Chun Ho8, Ming-Jer Tang4, Kuei-Tai A. Lai9, Junne-Ming Sung5, and Yau-Sheng Tsai1,2,10,*
1Institute of Basic Medical Sciences, 2Institute of Clinical Medicine, 3Department of Medical Laboratory Science and Biotechnology, and 4Department of Physiology, National Cheng Kung University, Tainan, Taiwan. 5Division of Nephrology, Department of Internal Medicine, 6Department of Pathology, National Cheng and 10Research Center of Clinical Medicine, Kung University Hospital, Tainan, Taiwan. 7National Laboratory Animal Center, National Applied Research Laboratories, Taipei, Taiwan. 8Division of Nephrology, Department of Internal Medicine, E-DA Hospital/I-Shou University, Kaohsiung, Taiwan. 9NovoTaiwan Biotech, Taipei, Taiwan, Republic of China.
Font Enlarge
Taiwan has the highest incidence and prevalence rates of end-stage renal disease (ESRD), with one out of every 650 people suffering in kidney dialysis. The National Health Insurance program needs to cover huge budgets, as NT$37 billion in each year. Diabetic nephropathy (DN), a complication of diabetes, is the most common cause of ESRD. Whereas increased urinary albumin excretion is not simply an aftermath of glomerular injury, it is also involved in the progression of DN. Toll-like receptors (TLRs), which could be activated by pathogen-associated or damage-associated molecular pattern molecules (DAMPs), are incriminated in the renal inflammation of DN. The aim of this study is to clarify whether and how albumin is involved in the TLR-related renal inflammatory response. First, the kidneys of diabetic mice exhibited significant increases of heat shock protein 70 (HSP70), one of TLR4 endogenous ligands, and nuclear factor-κB promoter activity (Fig. 1). Base on the result of urine analysis and renal pathological examination, we found albuminuria, tubulointerstitial fibrosis, and tubular injuries were alleviated in TLR4-deficient diabetic mice (Fig.2). In vitro studies revealed that albumin stimulated proximal tubular cells for abundant HSP70 release, which triggered the production of inflammatory mediators in a TLR4-dependent manner. Moreover, HSP70 blockade ameliorated diabetes-induced albuminuria, inflammatory response and tubular injury. Finally, we found that individuals with DN had higher levels of TLR4 and HSP70 in the dilated tubules than non-diabetic controls. (Fig.3) This study highlights the HSP70-TLR4 axis as a key mediator of tubular inflammation and emphasizes the potential contribution of albuminuria to tubular injury in DN. The inhibition of tubular inflammation with agents that target the albumin-HSP70-TLR4 axis might represent a new therapeutic strategy to delay progression of DN in humans.
Fig.1 Expression and localization of inflammatory mediators in the diabetic kidney. (A) Imaging and photon counting in diabetic and control male transgenic (NF-κB-RE-luciferase) mice that were diabetic for 0.5, 3 or 5 months. (B) Immunofluorescence staining for HSP60, HSP70, biglycan and HMGB1 (red) in the kidney of 1-month-diabetic and control mice. The DAPI nuclear counterstain appears blue. Scale bars, 50 μm. G, glomerulus.

Fig.2 Reduction of albuminuria, fibrosis and renal tubular injury Tlr4−/− diabetic mice. (A) UAE and (B) urinary albumin-creatinine ratio (UACR) of Tlr4−/− and WT mice 1 to 3 months after induction of diabetes. (C) Masson trichrome stain in the kidney of Tlr4−/− and WT 3-month diabetic mice. Representative tubular morphology (D), quantification of the tubular nucleus-to-cytoplasm ratio (E) and distribution of tubular epithelial thickness (F) in Tlr4−/− and WT 1-month-diabetic mice. Scale bars, 50 μm. *P<0.05 and **P<0.01.

Fig.3 Renal expression of TLR4 and HSP70 in human biopsies. Representative photomicrographs (A) and scoring (B) of TLR4 and HSP70 immunohistochemical staining in the renal tissue from DN patients (n=11) and non-diabetic controls (Non-DM, n=10). Scale bars, 200 μm. **P<0.01 and ***P<0.001.
< Previous
Next >
Copyright National Cheng Kung University