(Pro)renin receptor is involved in mesangial fibrosis and matrix expansion

(Pro)renin receptor [(P)RR] is expressed in the kidney and is involved in renal injury. Although (P)RR is activated by indoxyl sulfate (IS) and may be related to renal injury, the details remain unclear. We used mouse mesangial cell line SV40 MES13 to investigate the association of (P)RR with mesangial fibrosis or expansion. Furthermore, we examined the correlation between serum soluble (P)RR [s(P)RR] and various laboratory data including serum IS, a uremic toxin that induces renal fibrosis through (P)RR, and pathological indices in chronic kidney disease and particularly in IgA nephropathy patients. In vitro study using SV40 MES13 cells revealed that (P)RR expression significantly increased in the presence of IS. IS stimulated the fibrotic factors’ expression, which was significantly suppressed by (P)RR knockdown. Moreover, it significantly increased the expression of matrix metalloproteinase 9 and tissue inhibitor of metalloproteinase 1 via the ERK1/2 pathway. In addition, the s(P)RR level significantly correlated with serum IS and mesangial injury markers in our patients. Our results suggest that (P)RR is associated with mesangial fibrosis and matrix expansion through the IS-(P)RR-ERK1/2 pathway. Clinically, s(P)RR may be a biomarker of mesangial fibrosis and matrix expansion.

(Pro)renin receptor [(P)RR] was initially identified as a member of the renin angiotensin system (RAS) in 2002 1 . There are three forms of (P)RR. Full length (P)RR is cleaved by furin to a single transmembrane form and a soluble form [s(P)RR], which exists in the blood 2 and urine 3 . (P)RR is ubiquitasly expressed in the kidney including mesangial cells, podocytes and tubular cells 1,4,5 . Binding of renin or prorenin to (P)RR leads to activation of the intracellular signalling of extracellular signal-regulated kinase (ERK) 1/2. This activation is involved in renal damage by releasing transforming growth factor (TGF)-β1 and cytokines. We have demonstrated that (P)RR is expressed in human peripheral blood mononuclear cells (PBMCs), and that renin stimulates ERK1/2 phosphorylation and interleukin-6 release from PBMCs isolated from healthy subjects. We have also shown that (P)RR is present on infiltrating lymphocytes and macrophages around glomeruli in anti-neutrophil cytoplasmic antibody (ANCA)-associated GN 6 . Previous studies have revealed that (P)RR blockade improved glomerulosclerosis in human (P)RR-transgenic rats 7 and that the serum levels of s(P)RR correlated with the degree of renal dysfunction in chronic kidney disease (CKD) patients 8 . These results suggest that (P)RR is associated with the progress of renal injury.
CKD is defined as a decrease in glomerular filtration rate and persistent proteinuria. Several uremic toxins accumulate in the blood with CKD progress. Indoxyl sulfate (IS), which is a representative uremic toxin and a metabolite of tryptophan derived from dietary protein, also accumulates in the circulation of CKD patients, and induces glomerulosclerosis 9,10 , mesangial cell proliferation 11 and expression of fibrotic genes such as TGF-β1 and type-I collagen 12 in the kidney. In addition, IS promotes cell proliferation and tissue factor expression in vascular smooth muscle cells 13 , and induces TGF-β1 and α-smooth muscle actin expression in proximal tubular cells through (P)RR 14 .
Based on these results, we hypothesized that IS accumulated by renal failure activates (P)RR and contributes to mesangial injury, and that (P)RR could be a marker for evaluating pathological change of CKD. To verify this hypothesis, we examined the effect of IS and other uremic toxins on the proliferation of mesangial cells via activated (P)RR. Additionally, we investigated the relationship between (P)RR and IS in CKD patients. IS enhances the expression of fibrotic genes through (P)RR. As CKD progresses, accumulated IS induces the expression of fibrotic factors. The protein expression levels of type IV collagen, TGF-β and fibronectin significantly increased in SV40 MES13 cells after IS stimulation ( Fig. 3A-C). To examine the role of (P)RR in IS-stimulated expression of these fibrotic factors, we knocked down (P)RR expression using small interfering RNA (siRNA). The greatest reduction in (P)RR expression (almost 50%) was observed 48 h after (P)RR siRNA transfection (Fig. 3D). Furthermore, (P)RR knockdown in SV40 MES13 cells did not affect cell growth or viability (Fig. 3E). Figure 3A-C also shows that IS did not increase the expression of the abovementioned fibrotic factors after (P)RR knockdown. These findings suggest that IS induced the expression of the fibrotic factors, collagen IV, TGF-β and fibronectin, through (P)RR in SV40 MES13 cells.
Mesangial matrix remodeling by IS through (P)RR-mediated ERK1/2 pathway. IS stimulation did not affect the cell proliferation of SV40 MES13 cells (Fig. 1E). We speculated that IS affects mesangial matrix expansion. To examine this speculation, we checked the expression of MMP9 and TIMP1 in SV40 MES13 cells. MMP9 and TIMP1 expression significantly increased after IS stimulation at both the mRNA (Fig. 4A, P = 0.005; 4B, P = 0.03) and protein levels ( Fig. 4C, P = 0.008; 4D, P = 0.005), and the increases were inhibited by (P)RR knockdown ( Fig. 4C and D).
IS induced ERK1/2 phosphorylation in SV40 MES13 cells (Fig. 5A). Next, we used the specific ERK1/2 inhibitor, U0126, to examine the effect of the ERK1/2 pathway on the induction of MMP9 and TIMP1. U0126 surely inhibited the ERK1/2 pathway (Fig. 5B). The increase in these proteins' expression by IS stimulation was suppressed in the presence of U0126 ( Fig. 5C and D). These results suggest that the expression of MMP9 and TIMP1 is induced by a (P)RR-mediated ERK1/2 pathway. Subject characteristics. The pathological diagnosis and characteristics of the subjects for CKD and IgA nephropathy(IgAN) patients at the time of biopsy are shown in Tables 1, 2  In CKD patients, the mean age was 50.4 ± 19.1 years, the estimated glomerular filtration rate (eGFR) was 71.5 ± 30.8 mL/min/1.73 m 2 and the urinary protein was 2.3 ± 3.5 g/gCr. On the other hands, in IgAN patients, the mean age was 47.0 ± 19.2 years, the eGFR was 77.3 ± 27.6 mL/min/1.73 m 2 , the urinary protein was 1.1 ± 2.1 g/gCr, IGL was 1.6 ± 0.4 and the interstitial fibrosis area was 22.4 ± 17.2%. Mesangial hypercellularity (M1) was present in 14.5%, segmental glomerulosclerosis (S1) was present in 60.0%, tubular atrophy/interstitial fibrosis (T1 + T2) was present in 21.8%, and endocapillary hypercellularity (E1) was present in 5.5% of the subjects.

Correlation between s(P)RR and uremic toxins in CKD patients.
There was a significant association between s(P)RR and IS (P = 0.02). However, there were not the correlations between s(P)RR and other uremic toxins (methylguanidine; P = 0.6, indole-3-acetic acid (IAA); P = 0.14) after adjustment of sex, age, BMI and diabetes.

Correlation among s(P)RR and uremic toxins and renal histopathology in IgAN patients.
We evaluated the correlation between s(P)RR and the clinical and pathological parameters.
The mean serum s(P)RR concentration was significantly higher in the presence of mesangial hypercellularity (P = 0.03) or segmental glomerulosclerosis (P = 0.004) in the MEST score (Table 4). Moreover, high IGL and IS levels were significantly associated with high concentrations of serum s(P)RR (Fig. 6). After adjustments for age, gender, body mass index and diabetes, the IGL (β = 0.41, P = 0.02) and IS levels (β = 0.12, P = 0.04) remained significantly associated with serum s(P)RR. In addition, we evaluate the correlation among s(P)RR, other uremic toxins and histopathology. Multivariate analysis (adjustment of sex, age, BMI and diabetes) revealed that there were the significant associations between s(P)RR and each uremic toxins (methylguanidine; P = 0.01, IAA; P = 0.001). However, there were not the correlation between IGL and them (methylguanidine; P = 0.68, IAA; P = 0.57).

Discussion
The present study revealed new aspects of (P)RR in the progression of mesangial fibrosis and matrix expansion. In this study, we showed that IS stimulated the expression of (P)RR. The expression of the fibrotic factors, TGF-β, type IV collagen and fibronectin, increased after IS stimulation, and these increases were significantly suppressed by (P)RR knockdown in SV40 MES13 cells. Although IS did not affect the number of mesangial cells, it significantly increased the expression of MMP9 and TIMP1 via (P)RR and the ERK1/2 pathway. In addition, the serum s(P)RR level correlated with IS in CKD patients. There was the association between s(P)RR and part of the MEST score in the Oxford Classification particularly in IgAN patients. Furthermore, higher IGL and serum IS levels were also associated with higher serum s(P)RR levels. These findings indicate that s(P)RR may be a biomarker of the degree of mesangial expansion and fibrosis, which would be helpful because it is difficult to predict chronic changes in glomeruli from laboratory data.
To confirm and extend our original observations that (P)RR activated by IS is involved in mesangial expansion, we performed experiments using the SV40 MES13 mouse mesangial cell line. We revealed that (P)RR expression was enhanced in SV40 MES13 cells in the presence of IS. In addition, IS induced the expression of the fibrotic factors, type IV collagen, TGF-β and fibronectin, and this induction was suppressed by (P)RR knockdown. Moreover, we demonstrated that IS affected mesangial matrix accumulation rather than cell proliferation. The deposition of extracellular matrix (ECM) is most commonly due to failures in the coordinated expression of proteins regulating collagen metabolism. In particular, an imbalance between MMPs and TIMPs is thought to be the main cause of increased matrix deposition 18 , consequently resulting in tissue fibrosis. MMP2 and MMP9 cleave collagen IV produced mainly by mesangial cells 19 , and TIMP1 and TIMP2 are strong inhibitors of MMPs. The expression of TIMP1 and MMP9 significantly increased by IS stimulation, which was suppressed by (P)RR knockdown in this study. In contrast, the mRNA expression of TIMP2 and MMP2 did not change after IS stimulation (data not shown). TIMPs were traditionally thought to control ECM proteolysis through direct inhibition of MMP-dependent ECM proteolysis. Thus, the upregulation of TIMP1 mRNA may be a compensatory mechanism for the increase of MMP9 expression by IS stimulation. Furthermore, although TGF-β induces MMP9 secretion to digest excess ECM, chronic stimulation by TGF-β may become a maladaptation 20 . In CKD, chronic IS stimulation could induce maladaptive MMP9 through TGF-β, which may cause the matrix expansion/remodelling.
To evaluate active MMP9, we performed MMP9 activity assay. Although we used 10-fold concentrated culture medium of SV40 MES13, we could not detect both total and active MMP9. Thus, we tried the same experiment using human sera just before the introduction of hemodialysis. The result which total MMP9 was decreased was different from that of the present study. This difference is due to the exposure period. The patients are chronically exposed to uremic toxins, but the exposure time of IS was for 48 hours in our experiments. Indeed, it is reported that the production of MMP9 of patients on hemodialysis is decreased 21,22 .
It has been reported that RAS activation plays a role in mesangial expansion in IgAN model mice because administration of an angiotensin II type1 receptor blocker ameliorated the extent of mesangial matrix expansion 23 . The inhibition of (P)RR or ERK1/2 signalling attenuated mesangial expansion and decreased the expression of fibrotic factors 24,25 . Indeed, we found that the (P)RR and ERK1/2 pathway was involved in the induction of MMP9 and TIMP1. Furthermore, the binding of renin or prorenin to (P)RR activates the ERK1/2 signalling 1 . Thus, the (P)RR-mediated ERK1/2 pathway may be associated with mesangial expansion based on the increase in the expression of TIMP1 and MMP9.
We identified that there was a significant association between s(P)RR and IS in CKD patient. Moreover, we evaluated the correlation among s(P)RR, uremic toxins and histopathological change in IgAN patients because IgAN is characterized by mesangial expansion. Although s(P)RR correlated with IGL and IS significantly, there was not significant association between IGL and other uremic toxins. Actually, western blot analysis revealed that (P)RR was not increased by methylguanidine and hippuric acid in SV40 MES13. Therefore, uremic toxins which increased (P)RR expression or have the significant association with IGL could be limited. As there are as yet no reliable biomarkers for mesangial expansion, s(P)RR may be a promising marker which we could predict the level of mesangial expansion at diagnosis or follow up period. This is the first study reporting an association of s(P)RR with mesangial hypercellularity (M1) and segmental glomerulosclerosis (S1) in the MEST score. In contrast, tubular atrophy/interstitial fibrosis (T1) and extracapillary hypercellularity (E1) showed no significant association. Mesangial matrix accumulation has been reported to result in glomerulosclerosis 26 . Furthermore, we demonstrated that the serum s(P)RR concentration was independently associated with IGL. IGL is a predictor of mesangial proliferation and sclerotic change [27][28][29] . These results suggest that (P)RR is involved in mesangial injury. In addition, the serum IS concentration was independently related to the s(P)RR concentration in IgAN patients. Previous studies have indicated that IS is involved in glomerulosclerosis 9,10 , and that it upregulates prorenin expression in proximal tubular cells 30 . Taken together, our pathological analysis of IgAN patients suggests that (P)RR contributes to mesangial injury in the presence of IS.
As a limitation, we used mouse mesangial cell line in vitro study. However, only humans and higher primates have IgA1 subclass, of which -glycosylation is linked to the pathogenesis of IgAN. Further investigation in human mesangial cells will help to elucidate the role of (P)RR in IgAN.
In conclusion, (P)RR is involved in mesangial fibrosis and matrix expansion through ERK1/2 phosphorylation in the presence of IS. Clinically, s(P)RR may be used as a biomarker of mesangial fibrosis and matrix expansion.

Materials and Methods
Cell culture. SV40 MES13, a mouse mesangial cell line, was obtained from American Type Culture Collection (Manassas, VA) and was cultured in DMEM/F12 medium (Wako, Tokyo, Japan) with 14 mM HEPES (Nacalai tesque, Kyoto, Japan), 5% fetal bovine serum (Life Technologies, Grand Island, NY), 100 U/mL penicillin and 100 µg/mL streptomycin (Sigma-Aldrich, Munich, Germany). Cells were incubated in a humidified incubator with 5% CO 2 at 37 °C. Cells were serum-starved for 24 h before IS stimulation experiments. We used 50 mM Tris-HCl buffer as a vehicle control because of the IS solvent.
Patients and sample collection. Data and serum samples were collected from 100 patients performed renal biopsy between 2014 and 2016 at the Tohoku University Hospital. Sera were obtained from the peripheral blood on the day of the renal biopsy and stored at −80 °C until use. The study was performed under the declaration of Helsinki principles. The study protocol was approved by the Institutional Review Board of the Tohoku University School of Medicine (registration number: 2010-108). All participants provided written informed consent after a full explanation of the purpose of the study and the potential risk involved.
Clinical evaluation. The clinical and laboratory data of these patients on the day of the renal biopsy were retrospectively analysed. Diabetes was defined as random blood glucose ≥11.1 mmol/L (≥200 mg/dL), HbA1c ≥6.5% according to the Japan Diabetes Society, using medication to control diabetes, and/or a history of diabetes. The GFR was calculated using the new Japanese equation 33 . Uremic toxins including IS, methylguanidine and IAA were measured by liquid chromatography-mass spectrometry (LC-MS)/MS as previously reported 34 . Serum s(P)RR levels were measured using an enzyme-linked immunosorbent assay (ELISA) kit (Immuno-Biological Laboratories Co., Tokyo, Japan), consisting of a solid-phase sandwich ELISA with two kinds of highly specific antibodies 35 .

Statistical analysis.
Values are expressed as the mean ± standard error (SE). Data were analysed using JMP software version 12 (SAS Institute Inc., Cary, NC). Multiple groups were compared using ANOVA with the Tukey-Kramer test for parametric values. Otherwise, the Kruscal-Wallis test, followed by the Steel-Dwass test was used for non-parametric values. P < 0.05 was considered statistically significant.
Data availability. All data generated or analysed during this study are included in this published article and its Supplementary Information files.