High glucose induces trafficking of prorenin receptor and stimulates profibrotic factors in the collecting duct

Growing evidence indicates that prorenin receptor (PRR) is upregulated in collecting duct (CD) of diabetic kidney. Prorenin is secreted by the principal CD cells, and is the natural ligand of the PRR. PRR activation stimulates fibrotic factors, including fibronectin, collagen, and transforming growth factor-β (TGF-β) contributing to tubular fibrosis. However, whether high glucose (HG) contributes to this effect is unknown. We tested the hypothesis that HG increases the abundance of PRR at the plasma membrane of the CD cells, thus contributing to the stimulation of downstream fibrotic factors, including TGF-β, collagen I, and fibronectin. We used streptozotocin (STZ) male Sprague–Dawley rats to induce hyperglycemia for 7 days. At the end of the study, STZ-induced rats showed increased prorenin, renin, and angiotensin (Ang) II in the renal inner medulla and urine, along with augmented downstream fibrotic factors TGF-β, collagen I, and fibronectin. STZ rats showed upregulation of PRR in the renal medulla and preferential distribution of PRR on the apical aspect of the CD cells. Cultured CD M-1 cells treated with HG (25 mM for 1 h) showed increased PRR in plasma membrane fractions compared to cells treated with normal glucose (5 mM). Increased apical PRR was accompanied by upregulation of TGF-β, collagen I, and fibronectin, while PRR knockdown prevented these effects. Fluorescence resonance energy transfer experiments in M-1 cells demonstrated augmented prorenin activity during HG conditions. The data indicate HG stimulates profibrotic factors by inducing PRR translocation to the plasma membrane in CD cells, which in perspective, might be a novel mechanism underlying the development of tubulointerstitial fibrosis in diabetes mellitus.

I RIA kit (DIASORIN, Stillwater, MN, USA)]. Urine samples were spiked with 1 μM synthetic renin substrate tetradecapeptide (RST, Sigma). The generated Ang I was determined by RIA. To exclude the effect of peptidases, identical urine samples-RST with the specific renin inhibitor WFML peptide (ANASPEC; Fremont, CA) were used as controls.
Immunoblotting analyses. Protein samples were electrophoretically separated on a precast NuPAGE 10% Bis-Tris gel (NOVEX) at 200 v for 45 min followed by semi-dry transference to a nitrocellulose membrane using iBlot (INVITROGEN, Carlsbad, CA, USA). Blots were blocked with Odyssey blocking buffer (LI-COR BIOSCIENCES, Lincoln, NE, USA) at RT for 3 h, incubated overnight with specific primary antibody at 4 °C, followed by the incubation with the corresponding secondary antibodies donkey anti-rabbit or antimouse (1:15,000 dilutions), at room temperature for 45 min and analyses by normalization against β-actin, used as a loading control. Renin and prorenin protein levels were quantified using the rabbit anti-renin/prorenin antibody (Cat. # sc-22752; SANTA CRUZ, CA). PRR protein levels were detected using a polyclonal rabbit anti-PRR (Atp6ap2, 1:200, SIGMA-ALDRICH) that recognizes the intracellular segment and the ectodomain 27  Trafficking assay of the prorenin receptor (PRR). To determine whether HG induces the translocation of PRR from the cytosol to the cell surface we used trafficking assay in M-1 cells treated with either normal glucose (NG) or HG. Briefly, cultured cells were split into 10 cm dishes and incubated with culture media containing normal glucose (NG, 5.5 mM glucose + 19.5 mM mannitol) and high glucose (HG, 25 mM glucose) in supplement serum free media for 1, 6 h specified in each protocol. Abcam Plasma Membrane Isolation Kit (Ab65400; ABCAM, Cambridge) was used to extract the membrane proteins. Western blotting was performed using antibody anti-PRR (Atp6ap2, 1:200, SIGMA-ALDRICH) overnight at 4 °C, followed by the incubation with the secondary antibody donkey anti-rabbit (1:10,000), at room temperature for 45 min. Detection was accomplished using the Odyssey detection system (LI-COR BIOSCIENCES, Lincoln, NE) and protein expression was normalized against E-cadherin.
Renin secretion to extracellular media. To examine renin secretion into culture media, M-1 cells were starved overnight without glucose, then treated with NG and HG at different time intervals. Supernatants were collected and 50-fold concentrated by using Amicon Ultra-4 (MILLIPORE, Carrigtwohill, IE). Briefly, 4 ml of culture supernatant was added to the ultra-filter device, and centrifugated to concentrate at 4,000 g for 15 min. Concentrated cell culture media was recovered, and immunoblotting analysis was completed as mentioned above by loading an equal volume of protein. Equal loading of proteins was confirmed by Ponceau red staining.

Statistical analyses.
Results are expressed as mean ± SEM. Grubb's test was used to detect outliers in univariate data assumed to come from a normally distributed population and using a significance level of alpha = 0.05. These exclusion criteria were applied in experiments using more than 6 samples. Comparisons between groups were performed using One-Way ANOVA and Tukey's post-test, when appropriate. P ≤ 0.05 values were considered statistically significant, with P = NS demonstrating no significance. Table 1, the bodyweight of STZ rats decreased at 7d after STZ administration and was significantly less than that of the control rats. Overnight fasting blood glucose levels were significantly higher and plasma insulin levels were significantly lower in STZ-induced hyperglycemic rats than in control rats. Urine volume and protein excretion were higher in STZ rats compared with the control group, whereas urinary creatinine levels were lower in STZ rats than in controls.

STZ rats showed a marked increase of renin and prorenin peptides in renal tissues and augmented urinary renin activity.
After 7d of STZ administration, plasma renin activity in hyperglycemic rats averaged 42.32 ± 11.87 versus 21.28 ± 8.50 ng Ang I/mL/h (Fig. 1A). Ang II levels were markedly lower in plasma of STZ rats compared with control rats (STZ: 60.1 ± 5.6 vs. C: 358.2 ± 1.7 fmol/mL, P < 0.001) (Fig. 1B). Renin and prorenin proteins were both greater expressed in the renal medulla of STZ than in control rats (Fig. 2 PRR expression is augmented in the renal medulla of STZ rats. Figure 4A-C shows the renal PRR mRNA and protein levels. PRR mRNA was markedly increased in the renal medulla of STZ rats (fold change of control: STZ: 1.50 ± 0.09 vs. C: 1.0 ± 0.07, P < 0.001) (Fig. 4A). The protein expression levels of the full-length form of the PRR were greater in STZ rats than in control rats (PRR/β-actin protein ratio: STZ: 0.55 ± 0.03 vs. C: 0.44 ± 0.02, P < 0.01, Fig. 4B). In the cortex, PRR expression was not altered in STZ rats (Fig. 4C). Because soluble form of the PRR (sPRR) also interacts with prorenin 8,27 we evaluated sPRR levels in urine and plasma. Although diabetic rats demonstrated significantly lower levels of sPRR in plasma (STZ: 10,814 ± 138 vs. C: 14,146 ± 426 pg/ Table 1. Physiological parameters of STZ rats on Day 7, body weight, blood glucose, plasma insulin, urine volume, urine protein concentration, urinary creatinine, urine protein/creatinine ratio. Mean ± SEM, *P < 0.05, **P < 0.01, ***P < 0.00.

STZ rats showed increased apical membrane distribution of PRR in the collecting duct
Immunohistochemistry was performed to determine whether HG conditions cause changes in the specific localization of the PRR in the CD. In STZ rats, PRR was detected at the luminal side of the plasma membrane in renal medullary CD cells, while in control rats, PRR was homogeneously distributed within each cell (Fig. 6A, arrows). Enriched plasma membrane fractions from control and STZ rats were extracted from renal medullary tissues. We found a higher abundance of PRR at the plasma membrane of renal medullas from STZ rats compared to control rats (STZ: 0.94 ± 0.11 vs. C: 0.57 ± 0.08, P < 0.05) (Fig. 6B).
TGF-β mRNA is augmented in the renal medulla, but not cortex, in STZ rats. Figure 7 shows that expression of TGF-β mRNA in the renal medulla was significantly higher than in the cortex and was markedly greater in renal medullae of STZ rats than in control rats (STZ: 1.22 ± 0.06 vs. C: 0.97 ± 0.03 TGF-β/β-actin mRNA ratio, P < 0.01).  (Fig. 8A). No differences were found at 1 h (data not shown). Prorenin and renin secretion to the cell culture media was induced as early as 1 h of HG incubation (HG: 107 ± 12 vs. 35 ± 5, pixel intensity/region of prorenin plus renin band, P < 0.05, (Fig. 8B). Under normal glucose conditions, PRR immunofluorescence showed perinuclear area localization (Fig. 8C). After incubation with HG (1 h), the immunoreactivity was spread diffusely inside the cell, with increased localization at the apical surface of the cells (Fig. 8C right insert). To further examine whether HG enhances the physical interaction between PRR and prorenin, we used FRET assay using plasma membrane fractions from M-1 cells in NG or HG conditions. Figure 8D shows the AGT-FRET experiment methodology and the fluorescence intensity measured after the addition of substrate AGT (Fig. 8E). The intensities were normalized by the same amounts of protein amount extracted. Quantification of FRET data indicated that even in the presence of similar amounts of extracted plasma membrane fractions from NG and HG treated M-1 cells, there was an increase in renin activity during HG conditions (Fig. 8F).

Discussion
In the present study, we report that SD rats with STZ-induced hyperglycemia for 7 days exhibit augmentation of PRR, and prorenin and renin proteins expression in the CD of the kidney. These changes are accompanied by augmented urinary excretion of AGT renin, Ang II, sPRR. We further showed evidence that HG induces PRR translocation to the plasma membrane of M-1 cells, which increases the physical interaction between PRR and prorenin, PRR-dependent activation of ERK pathway, and upregulation of downstream targets as TGF-β, fibronectin and collagen I. www.nature.com/scientificreports/ The PRR is considered a multifunctional receptor 34 . As an accessory protein of the multi-subunit complex, vacuolar H + -ATPase (v-ATPase), PRR plays a key role in intracellular acidification 35,36 , autophagy 37 , and kidney development 38 . In the kidney, the expression of v-ATPase on the membranes of intracellular organelles and A-type intercalated cells 39 , emphasizes the relevance of PRR in the regulation of intracellular pH 36 . In addition, in vitro evidence indicates that PRR increases renin activity and fully activates prorenin 27,40 . These findings are further supported by in vivo data from different models of experimental hypertension demonstrating that PRR in the CD is required for the local formation of Ang II 41,42 . However, whether PRR binding to prorenin increases renin activity and fully activates prorenin in vivo, still remains controversial 43 .
Augmented PRR expression may play important pathophysiological roles in the development and progression of renal fibrosis during DM. Likewise, PRR mRNA and protein abundance are increased in the hearts of TGR[m(Ren2)-27] diabetic rats in association with diastolic dysfunction, myocyte hypertrophy, and interstitial fibrosis 44 . Furthermore, PRR protein levels are augmented in mesangial cells during HG conditions 45 and in CD during diabetes 46 . In the present study, PRR protein levels were augmented in the renal inner medulla of STZinduced hyperglycemic SD rats (Fig. 4), which are devoid of glomeruli and primarily contain CD. We detected a band of 35 kDa, which corresponds with the full-length PRR molecular form (Fig. 4). Along with the increased www.nature.com/scientificreports/ expression of PRR protein, the preferential apical distribution of PRR in the CD of STZ-induced hyperglycemic rats as compared to control rats suggests that HG also influences PRR cell distribution in the distal nephron segments. We also demonstrated that M-1 cells incubated with HG showed increased PRR protein expression in the plasma membrane fractions after 1 h and 6 h treatment compared with NG-treated cells (Fig. 8A,C). Kang et al. demonstrated that CD is the main source of prorenin in STZ-type 1 diabetic rats 3 . Therefore, our evidence demonstrating that in M-1 cells, HG increases mainly prorenin in the extracellular media (Fig. 8B), and stimulates PRR trafficking to the PM further supports our hypothesis that HG increases the physical interaction between prorenin and PRR in the CD. Our data is also supported by evidence demonstrates that glucose stimulates polarized translocation of v-ATPase to the apical plasma membrane in proximal tubular HK-2 cells 47 . This  www.nature.com/scientificreports/ concept was further confirmed using FRET, a tool that allows examining the energy transfer between a donor and acceptor pair of fluorophores, thus quantifying molecular dynamics and the interactions between proteins. When the donor and acceptor fluorophores are in close proximity to each other, excitation of the donor results in detectable emission only from the acceptor. Using a fluorogenic renin substrate, AGT, containing the renin cleavage site at the Leu-Val bond 48 , we demonstrated FRET-based assay or renin enzymatic activity in M-1 cells treated with HG. The activation of recombinant prorenin as shown by the increased renin activity in plasma membranes fractions as compared to those extracted from NG-treated cells (Fig. 8D,E,F) reflect the physiological significance of the effect of HG on the enhancement of physical interaction between PRR and prorenin, which might be of high relevance in diabetes. The underlying mechanisms involved in the regulation of PRR by HG have been studied in mesangial cells 45,49,50 , and might be related to glucose transport. The CD express GLUT transporters, particularly GLUT-1 and GLUT-12 51 These glucose transporters may modulate HG-dependent PRR trafficking through mechanisms such as glycolysis and metabolic intermediaries such as succinate and alpha-ketoglutarate (Fig. 10) as suggested previously 52 , indeed it has been shown that in the collecting duct the GLUT transporters are upregulated during HG conditions 51 . Additional studies are necessary to elucidate the intrinsic mechanism involved in this process. Activation of PRR promotes the induction of profibrotic genes 8,11 . However, elevations in Ang II, blood pressure, inflammation, and oxidative stress may also stimulate fibrotic factors [53][54][55][56] . The fact that binding of prorenin to PRR triggers intracellular signals linked to tissue fibrosis raises the possibility whether prorenin could be directly responsible for tissue fibrosis during the early phase of diabetes, this aspect should be further determined in additional studies by using collecting duct specific AT1 receptor knockout 57,58 . The PRR/MAPK/ ERK1/2 pathway mediates the stimulation of TGF-β, fibronectin, and collagen leading to tissue fibrosis and inflammation 59,60 . In the present study, SD rats with STZ-induced hyperglycemia exhibited increased expression of medullary TGF-β along with augmented mRNA and protein levels of PRR in the renal inner medulla (Figs. 6,  7). Furthermore, HG treatment of M-1 cells not only increased PRR in the plasma membrane fractions, but also stimulated the expressions of TGF-β and fibrotic factors, including fibronectin and collagen (Fig. 9C). This effect was prevented by PRR knockdown. It should be noted that changes in the PRR and prorenin/renin observed during the very early phase of STZ-induced diabetes may let us evaluate in future studies whether these changes continue in later stages of the diabetic disease or in the presence of hyperglycemia.
In the kidney, hyperglycemia augments AGT in the proximal tubule, changes that are associated with high blood pressure and DN via stimulation of oxidative stress 61 . Inhibition of glucose transport via SGLT2 in the proximal tubule prevents intrarenal AGT upregulation. Although we did not measure blood pressure in our model, it is likely that during the late phase of the STZ-diabetic rat model, high levels of circulating prorenin 62 augmented expression of renin in the collecting duct 3 and AGT in the proximal tubule 2 along with the increased abundance of PRR in the plasma membrane of the collecting duct cells may promote high blood pressure, renal www.nature.com/scientificreports/ tubular fibrosis 63 . Future studies are necessary to determine the progression of the blood pressure along the early and late stages of STZ-induced diabetic disease. Prorenin levels are increased in the plasma and kidney in diabetes [64][65][66] . In the present study, in rats with STZ-induced hyperglycemia, renin content was augmented in the plasma and prorenin and renin amounts were increased in the renal medullary tissues. These findings further support the concept that in this rat model, concomitant augmented excretion of renin and Ang II contents in the urine further supports the concomitant activation of the intratubular RAS. Further studies are ongoing to examine the relative contribution of other RAS components to this mechanism.
The PRR exhibits a full-length molecular form that is bound to the PM and a soluble form that is secreted into the extracellular space, including plasma 33 . urine 27 , and extracellular media 67,68 . In the STZ-induced hyperglycemic rats of the present study, sPRR levels were decreased in plasma, but in contrast, it was augmented in the urine compared with those levels quantified in controls, suggesting that sPRR in the urine does not come from the systemic circulation. We previously demonstrated that augmented sPRR in the urine of chronic Ang IIinfused rats is associated with increased renin activity in the urine despite the characteristic suppression of PRA in these hypertensive rats 27 . Nonetheless, it remains unclear if renal intratubular sPRR could actually contribute to the activation of prorenin locally secreted by the CD. The sPRR is generated by the cleavage action of furin, ADAM 19, and S1P 32,33,69 . Furin expression is augmented in glomerular cells in HG conditions 70,71 ; however, in the present study, intrarenal furin mRNA levels did not differ between two groups of rats (Fig. 5C). Protein levels were also examined; however, we did not detect differences among the groups (Fig. 5D). It is possible that this protease is not the primary source of sPRR augmentation in the urine of STZ hyperglycemic rats, therefore, the involvement of other intracellular serine proteases should be explored.   65 . These effects were accompanied by upregulation of PRR mRNA, but downregulation of its protein levels. Pre-treatment with proteasome inhibitor MG-132 for 4 h blunted these effects and increased PRR protein levels after 3 d of cell exposure to HG 65 . Discrepancies between the two studies could be due to the differences between the two cell models used. We performed the experiments using M-1 cells, an established cortical CD cell line from mouse origin; while Siragy and associates employed primary cell cultures from rat renal inner medulla. Nevertheless, it is important to carefully address the deleterious effects that HG may exert on cells of the distal nephron segment, which usually are not exposed to luminal glucose. Increased PRR abundance in the CD might be twice harmful. On one side, it results in increases in renin activity and intratubular Ang II formation 9,28,29 , if renal AGT and ACE activity, are present. On the other side, it triggers intracellular signals that upregulate pro-fibrotic factors 12,48,52 . Therefore, during HG conditions as occurs in diabetes, PRR in the CD may contribute to increasing not only sodium reabsorption by Ang II-dependent mechanisms, but also to tubulointerstitial fibrosis independent of Ang II generation. Further studies are encouraged to examine the underlying mechanisms involved in the HG-dependent regulation of PRR.
In conclusion, our data suggest that the presence of high glucose in the distal nephron segments may induce PRR translocation to the apical cell side in the CD cells. During these conditions, high glucose facilitates the physical interaction between PRR and prorenin, thereby increasing PRR-dependent upregulation of downstream fibrotic factors, tubulointerstitial fibrosis, and likely, other inflammatory cytokines as TNF-α and IL-1β, independent of renal Ang II 58 . Our data may also serve as a trigger for future investigations considering PRR blockade as a novel target for the prevention and progression of kidney disease in patients with diabetes.
Received: 23 April 2020; Accepted: 8 June 2021 Figure 10. Working hypothesis representing the potential effects of high glucose on PRR cellular preferential distribution to the plasma membrane. In the collecting duct, the principal cell is the main source of prorenin in diabetes. Prorenin released to the lumen, binds PRR, thus leading to its activation and stimulation of MAPK/ ERK1/2 intracellular pathway which is known to induce profibrotic factors in these cells. Upregulation of PRR is mediated by high glucose, but not osmolality. In this scenario, it is likely that GLUT-1 and GLUT-12 transporters contribute to the HG-dependent PRR trafficking through mechanisms including glycolysis and metabolic intermediaries such as succinate and alpha-ketoglutarate through their receptors. www.nature.com/scientificreports/