ORAI channels are critical for receptor-mediated endocytosis of albumin
Impaired albumin reabsorption by proximal tubular epithelial cells (PTECs) has been highlighted in diabetic nephropathy (DN), but little is known about the underlying molecular mechanisms. Here we find that ORAI1-3, are preferentially expressed in PTECs and downregulated in patients with DN. Hyperglycemia or blockade of insulin signaling reduces the expression of ORAI1-3. Inhibition of ORAI channels by BTP2 and diethylstilbestrol or silencing of ORAI expression impairs albumin uptake. Transgenic mice expressing a dominant-negative Orai1 mutant (E108Q) increases albuminuria, and in vivo injection of BTP2 exacerbates albuminuria in streptozotocin-induced and Akita diabetic mice. The albumin endocytosis is Ca2+-dependent and accompanied by ORAI1 internalization. Amnionless (AMN) associates with ORAIs and forms STIM/ORAI/AMN complexes after Ca2+ store depletion. STIM1/ORAI1 colocalizes with clathrin, but not with caveolin, at the apical membrane of PTECs, which determines clathrin-mediated endocytosis. These findings provide insights into the mechanisms of protein reabsorption and potential targets for treating diabetic proteinuria.ORAI1-3 channels were detected in human kidney samples both at the mRNA and protein levels (Supplementary Fig. 1a, b). They were preferentially localized to kidney tubules, with stronger staining in the proximal tubules than in the distal convoluted tubules (Supplementary Fig. 1c). This is in agreement with rat RNAseq data showing higher tubular than glomerular expression of all ORAI1-3 genes10 and also with available human RNAseq data (Nephroseq) for ORAI2 expression in human kidney samples from healthy living donors, confirming higher tubular than glomerular expression of this gene (Supplementary Fig. 2a). ORAI1-3 immunostaining in both proximal tubules and distal tubules was weaker in kidney tissue sections from type 1 diabetic patients with DN when compared to non-diabetic controls (Fig. 1a–c; Supplementary Table 1 for patient characteristics). Correspondingly, human RNAseq data from DN patients with estimated glomerular filtration rate (eGFR) ranging between 12 and 60 showed that expression of ORAI2 in the tubulointerstitium was lower in patients with DN than that in controls, and positively correlated to eGFR (Supplementary Fig. 2b, c), suggesting the expression of ORAI2 in tubules is related to the severity of DN.Rodents are in general quite resilient to develop diabetic complications as seen in humans and often replicate only early features of DN. We examined ORAI1-3 and STIM1-2 mRNA expression in whole kidney homogenates from Akita mice and STZ-diabetic mice and found increased expression of all targets in diabetic mice (Supplementary Fig. 2d, e). In the Pima Indian study cohort, patients with DN at early stage (normal GFR or hyperfiltration stage), a small increasing trend but not statistically significant was observed for ORAI2 expression in the tubulointerstitium (Supplementary Fig. 2f), suggesting that downregulation of ORAIs is associated with fast kidney function decline during the late stage of DN, but not the early stage of DN. Taken together, the rodent expression and human RNAseq data demonstrate plasticity of ORAI and STIM genes during the development of DN.To investigate whether changes in ORAI expression are driven by glucose or insulin, in vitro experiments were performed using primary cultured human PTECs. Cells were characterized by positive lectin staining in their apical membrane (Fig. 1d). Both mRNA and protein levels of ORAI1-3 were significantly downregulated by high glucose (Fig. 1e, f). Incubation of PTECs with insulin increased mRNA levels of ORAI1-3, while incubation with tyrphostin A23, a protein tyrosine kinase inhibitor that prevents insulin receptor activation, reduced the expression (Fig. 1g, h). Insulin also increased the expression of STIM1 and STIM2, while tyrphostin A23 decreased them (Fig. 1i). STIM1 and STIM2 were also detected in both glomerulus and tubulointerstitium of human kidney (Supplementary Fig. 3a–d). In the Nephroseq data set, STIM1 mRNA intensity was higher in glomeruli and positively correlated to eGFR (Supplementary Fig. 3c, d). Downregulation of STIM1 was also observed in kidney sections from STZ-induced diabetic mice after long-term hyperglycemia (Supplementary Fig. 3e, f). These data demonstrate the expression of ORAIs and STIMs in kidney tubules and downregulation under diabetic conditions.Akita type 1 diabetic mice (C57BL/6-Ins2Akita/J) were obtained from the Jackson laboratories and bred at Lund University. For the streptozotocin (STZ)-induced type 1 diabetic model, adult C57BL/6 J mice were administered STZ (Sigma-Aldrich; 55 mg kg−1 body weight, pH 4.5) or citrate buffer (vehicle) by intraperitoneal (i.p.) injections once a day for 5 days. The animal studies were approved by the Malmö/Lund Animal Care and Use Committee, and abided by the Guide for the Care and Use of Laboratory Animals published by the Directive 2010/63/EU of the European Parliament. Diabetic and control mice were treated twice daily with intraperitoneal (i.p.) injections of BTP2 (Tocris Bioscience, UK; 0.29 mg kg−1 body weight per day) for 3 days. Treatment with BTP2 was performed after at least 4 weeks of diabetes for STZ-induced mice, and 10 weeks of age for male Akita mice, which had been diabetic for at least 6 weeks according to the onset age of diabetes at 4 weeks old. Urine was collected during a 2-h period on day 5 after the first injection of BTP2. Albumin concentration was determined by ELISA (Albuwell M kit, Exocell Inc., USA) and creatinine using a picric acid-based assay (Creatinine Companion; Exocell Inc., USA) and an enzymatic colorimetric assay (COBAS, Roche; USA) according to the manufacturer´s instructions.We thank G. Cooksey, S. Krauss and S. Hetherington (Castle Hill Hospital, Hull and East Yorkshire Hospitals NHS Trust) for kindly providing human kidney samples; N. Watson and A. Green for technical help. This work was supported in part by the Hull York Medical School Pump Priming Award, British Heart Foundation and Leverhulme Trust (to S.-Z.X.); National Natural Science Foundation of China (31300965 to G.-L.C. and 31300949 to B.Z.); China Scholarship Council (to B.Z.); University PhD studentship (to N.D.); National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (P30 DK081943, to M.K.); Intramural Research Program of the National Institute of Diabetes and Digestive and Kidney Diseases (to R.G.N.); and by the Swedish Research Council (2011-3900 and 2013-0700), Albert Påhlsson and Diabetes foundations, and the Swedish Foundation for Strategic Research (to M.F.G.). This project has received funding from Innovative Medicines Initiative Joint Undertaking under grant agreement [No. 115006 (SUMMIT) and No. 115974 (BEAt-DKD)], comprising funds from the European Union’s Seventh Framework Program [FP7/2007-2013] and Horizon 2020 research and innovation programme and EFPIA (to M.F.G., S.-Z.X., M.K.).The authors declare that they have no competing financial interests.Bo Zeng, Gui-Lan Chen and Eliana Garcia-Vaz contributed equally to this work.Electronic supplementary materialSupplementary Information accompanies this paper at 10.1038/s41467-017-02094-y.Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.