Thrombomodulin Ameliorates Transforming Growth Factor-b1–mediated Chronic Kidney Disease Via The G-protein Coupled Receptor 15/Akt Signal Pathway

Contact: joanna.jia@wecistanche.com / WhatsApp: 008618081934791

Atsuro Takeshita1,2,8 , Taro Yasuma1,2,8 , Kota Nishihama1 , Corina N. D’Alessandro-Gabazza2, Masaaki Toda2 , Toshiaki Totoki4 , Yuko Okano1,2 , Akihiro Uchida1 , Ryo Inoue6 , Liqiang Qin7 , Shujie Wang5, Valeria Fridman D’Alessandro2 , Tetsu Kobayashi3 , Yoshiyuki Takei4 , Akira Mizoguchi5 , Yutaka Yano1,9 and Esteban C. Gabazza2,9

1 Department of Diabetes, Metabolism, and Endocrinology, Mie University Graduate School of Medicine, Tsu-city, Mie, Japan; 2 Department of Immunology, Mie University Graduate School of Medicine, Tsu-city, Mie, Japan; 3 Department of Pulmonary and Critical Care Medicine, Mie University Graduate School of Medicine, Tsu-city, Mie, Japan; 4 Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu-city, Mie, Japan; 5 Department of Neural Regeneration and Cell Communication, Mie University Graduate School of Medicine, Tsu-city, Mie, Japan; 6 Central Institute for Experimental Animals, Kawasaki-ku, Kawasaki, Kanawaga, Japan; and 7 Department of Nephrology, Taizhou Hospital, Wenzhou Medical University, Lihai, Zhejiang Province, People’s Republic of China.

Kidney International (2020) 98, 1179–1192; https://doi.org/10.1016/ j.kint.2020.05.041

Copyright ª 2020, International Society of Nephrology. Published by Elsevier Inc. This is an open-access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Correspondence: Esteban C. Gabazza, Department of Immunology, Mie University School of Medicine, Edobashi 2-174, Tsu-city, Mie 514-8507, Japan. E-mail: gabazza@doc.medic.mie-u.ac.jp; or Yutaka Yano, Diabetes, Metabolism and Endocrinology, Mie University Graduate School of Medicine, Edobashi 2-174, Tsu-city, Mie 514-8507, Japan. E-mail: yanoyuta@clin.medic.mie-u.ac.jp 8 AT and TY contributed equally to this work. 9 YY and ECG are co-senior authors. Received 1 September 2019; revised 24 April 2020; accepted 7 May 2020

KEYWORDS: apoptosis; chronic kidney disease; G-protein coupled receptor; recombinant human thrombomodulin; transforming growth factor-b1

Kidney fifibrosis is the common consequence of chronic kidney diseases that inexorably progresses to end-stage kidney disease with organ failure treatable only with replacement therapy. Since transforming growth factor-b1 is the main player in the pathogenesis of kidney fifibrosis, we posed the hypothesis that recombinant thrombomodulin can ameliorate transforming growth factor-b1–mediated progressive kidney fifibrosis and failure. To interrogate our hypothesis, we generated a novel glomerulus-specific human transforming growth factor-b1 transgenic mouse to evaluate the therapeutic effect of recombinant thrombomodulin. This transgenic mouse developed progressive glomerular sclerosis and tubulointerstitial fifibrosis with kidney failure. Therapy with recombinant thrombomodulin for four weeks significantly inhibited kidney fifibrosis and improved organ function compared to untreated transgenic mice. Treatment with recombinant thrombomodulin significantly inhibited apoptosis and mesenchymal differentiation of podocytes by interacting with the G-protein coupled receptor 15 to activate the Akt signaling pathway and to upregulate the expression of anti-apoptotic proteins including survivin. Thus, our study strongly suggests the potential therapeutic efficacy of recombinant thrombomodulin for the treatment of chronic kidney disease and subsequent organ failure.

cistanche can treat kidney disease improve renal function

Translational Statement

Fibrosis and dysfunction of the kidneys are currently major health problems worldwide. There is currently no antifibrotic drug approved for the treatment of renal fifibrosis. Transforming growth factor-b1 is the main and common driver of renal fibrogenesis in chronic kidney disease caused by many disorders. We found here that recombinant thrombomodulin, a drug approved in Japan to treat disseminated intravascular coagulation, suppresses the progression of glomerular sclerosis, tubulointerstitial fifibrosis, and renal failure caused by overexpression of human transforming growth factor-b1, suggesting its potential therapeutic value for the treatment of chronic kidney diseases.

Chronic renal disease is a major public health problem associated with high morbidity and mortality that affects about 13% of the adult population in developed countries.1 The World Health Organization reported that chronic kidney disease (CKD) was the cause of 1.5% of deaths worldwide in 2012.1 Furthermore, recent epidemiological data indicate a global steady rise in the number of patients with CKD.2,3 In most cases, the pathological process progressively evolves, leading ultimately to end-stage kidney failure, which is treatable solely with lifelong dialysis or renal transplantation.4,5 Diabetes mellitus (DM) and arterial hypertension are the most frequent causes of CKD followed by ischemia, glomerulosclerosis of unknown etiology, urologic obstructions, and chronic infections.6 Regardless of the underlying disorder, the final and common pathological consequence of CKD is the fifibrosis of the kidneys.7 Renal fifibrosis is the aberrant healing and remodeling of parenchymal structures of the kidneys subjected to prolonged or repetitive injury characterized by the presence of tubulointerstitial fifibrosis, glomerulosclerosis, and tubular atrophy.8 The main driver of renal fifibrogenesis is transforming growth factor (TGF)-b1.8,9 TGFb1 may promote fifibrosis by stimulating the secretion of extracellular matrix proteins and chemotactic factors or proliferating factors of fibroblasts, by inhibiting metalloproteinases, and by promoting epithelial-mesenchymal transition.10 Apart from treating the underlying disease, an approved drug that targets specifically renal fifibrosis is currently unavailable.6

Thrombomodulin (TM) is a transmembrane glycoprotein with multiple biological functions including modulation of the clotting system, the immune response, inflflammatory reactions, and cell survival.11 The molecular structure of TM contains a lectin-like domain, 6 epidermal growth factor-like domains, a serine/threonine-rich domain, a transmembrane portion, and a cytoplasmic tail.12 Thrombin, a procoagulant factor generated during coagulation system activation, becomes an anticoagulant and antifibrinolytic factor after binding to TM.13 The TM-thrombin complex increases the generation of activated protein C (APC), an anticoagulant with anti-inflflammatory and cytoprotective activity, by activating protein C. In addition, TM alone can directly downregulate the inflflammatory response by inhibiting the activity of high-mobility group protein B-1 (HMGB1), by suppressing immunogenic dendritic cells, eosinophils, mast cells, and the complement system.13–18

There are published results showing preventive effects of TM in diabetic renopathy and ischemia-reperfusion renal injury.19-21 However, no study has assessed the effect of recombinant TM on progressive kidney fifibrosis induced by TGFb1, a common driver of renal fifibrosis in several diseases causing CKD. We posed the hypothesis that TM can ameliorate TGFb1-mediated kidney fifibrosis and renal failure. To interrogate this hypothesis, we evaluated the therapeutic effect of recombinant TM in a newly developed glomerulus-specifific TGFb1 transgenic (TG) mouse that develops progressive kidney fifibrosis and renal failure.

cistanche for kidney

RESULTS

Increased circulating TM fragments correlate with renal dysfunction

TM, an endothelial cell membrane-bound glycoprotein, is cleaved in fragments, losing its protective functions during endothelial injury.22 Diabetic nephropathy is associated with endothelial injury.23,24 Patients with DM with nephropathy, compared with patients without nephropathy, showed signifificantly high circulating levels of TM (4.4 vs. 3.3 pg/ml) and active TGFb1 (0.28 vs. 0.24 ng/ml) (Supplementary Table S1, Supplementary Figure S1A). TM is significantly correlated with creatinine, active TGFb1, and soluble podocin (Supplementary Figure S1B). These observations suggest that a loss of functional membrane-bound TM is associated with increased release of active TGFb1 and kidney dysfunction.

TG mouse with a glomerulus-specific expression of full-length human TGFb1 gene

We developed a TG mouse overexpressing the full-length human TGFb1 gene in podocytes. The mouse was generated by placing the human TGFb1 under the control of the podocin promoter on a bacterial artificial chromosome (BAC) construct (Supplementary Figure S2; Supplementary Figure S3). We obtained 5 founder mice expressing 3 copies and 3 founder mice expressing 1 copy of the human TGFb1 transgene. The expression of the transgene was kidney-specific, and the offspring of the founders were viable (Supplementary Figure S3). Mice had full-term pregnancies and litter were of normal size. Mice were born in the expected Mendelian ratio.

To characterize the TG mice, we measured several renal parameters every 4 weeks for a period of 16 weeks (Supplementary Figure S4). The plasma and urine concentrations of TGFb1 were signifificantly increased in the TGFb1- TG mice compared with wild-type (WT) mice from early weeks after birth and then remained stable at a high level (Figure 1a). TGFb1-TG mice showed a significant increase in the renal tissue content of hydroxyproline at the 20th week (190.5 vs. 96.0 mg/kidney), mesangial expansion from the 4th week (1.9 vs. 1.1 scores), and in collagen deposition from the 12th week (0.9% vs. 0.1%) after birth compared to their WT counterparts (Figure 1a–f). Conventional optical microscopy showed widening of the basement membrane of the Bowman capsule, thickening of the glomerular basement membrane, increased collagen deposition in mesangial and interstitial spaces, and tubular atrophy (Figure 1bd). Transmission electron microscopy showed glomerulosclerosis including microvillous transformation and foot process effacement of podocytes, decreased glomerular capillary endothelial fenestration, thickening of the glomerular basement membrane, and increased mesangial matrix deposition (Figure 2a–i). The urinary concentrations of the early markers of nephropathy Fatty acid-binding protein (185.1 vs. 87.0 pg/ml) and kidney injury molecule 1 (441.9 vs. 256.9 pg/ml) were signifificantly enhanced in TGFb1-TG mice from the 4th and 8th weeks of age, respectively, compared with their age-matched WT mice (Supplementary Figure S5). Total proteinuria and the total protein-creatinine ratio in urine were signifificantly increased at 4th, 8th, 12th, 16th, and 20th weeks in TGFb1-TG mice compared with their WT counterparts (Figure 3a). The plasma level of blood urea nitrogen was signifificantly elevated from the 4th week (5.1 vs. 4.1 mg/dl) and the concentration of creatinine from the 8th week (1.1 vs. 0.4 mg/dl) in TGFb1-TG mice compared with their WT counterparts (Figure 3b).

rhTM inhibits glomerulosclerosis and tubulointerstitial fifibrosis

When compared with TGFb1-TG mice treated with saline (SAL) and WT mice treated with recombinant human (RH) TM or SAL, TGFb1-TG mice treated with rhythm showed signifificantly reduced mesangial expansion/cellularity (1.3 vs. 3.0 score) and signifificantly low tubulointerstitial collagen deposition and glomerulosclerosis (121.3% vs. 139.7%) (Figure 4a–e). Consistent with these observations, the hydroxyproline content (11.7 vs. 19.9 mg/g), the kidney tissue concentrations of collagen I (101.3 vs. 196.7 ng/mg protein) and periostin (21.7 vs. 40.8 ng/mg protein), and the relative mRNA expression of collagen I were signifificantly decreased in renal tissues from TGFb1-TG mice treated with rhTM compared with renal tissues from control mice (Figure 4f, Supplementary Tables S2, and S3). The transmission electron microscopy showed a significant reduction of foot process effacement of podocytes and thickening of glomerular basement membrane in mice treated with rhTM compared with mice treated with SAL alone (Supplementary Figure S6AB and B). In addition, the kidney tissue concentrations of the profibrotic cytokines monocyte chemoattractant protein-1 (45.0 vs. 76.1 pg/mg protein), interleukin-13 (612.1 vs. 1002.0 pg/mg protein), and active TGFb1 (131.0 vs. 151.5 pg/ mg protein) were signifificantly reduced in TGFb1-TG mice treated with rhTM compared with control mice (Supplementary Figure S7). The kidney tissue concentration of HMGB1 was also signifificantly decreased in renal tissues from TGFb1-TG mice treated with rhTM compared with renal tissues from control mice (Supplementary Figure S7). The plasma concentration of thrombin antithrombin complex was signifificantly increased in the TGFb1-TG/SAL group compared to the WT/SAL group but no difference was found between the TGFb1-TG/rhTM and WT/rhTM groups (Supplementary Figure S8A). As expected, there was a high concentration of TM in the plasma of TGFb1-TG and WT mice treated with rhTM. The plasma concentration of the APC/antitrypsin complex was signifificantly decreased in the TGFb1-TG/SAL group compared with the WT/SAL and TGFb1-TG/rhTM groups (280.5 vs. 384.6 pg/ml), and there was no significant difference in the level of plasminogen activator inhibitor-1 (Supplementary Figure S8A). The levels of C5a in plasma, urine, and kidney tissue (630.1 vs. 1075.0 pg/mg protein) and the plasma soluble podocin were significantly decreased in TGFb1-TG mice treated with rhTM compared with TGFb1-TG mice treated with SAL (Supplementary Figure S8B). The level of podocin in urine was higher in the TGFb1-TG/SAL group than in the WT/SAL and TGFb1-TG/rhTM groups (Supplementary Figure S9A–C).

Figure 1 | The human transforming growth factor b1 (TGFb1) transgenic (TG) mouse develops progressive kidney fifibrosis. (a) The concentrations of TGFb1 protein in plasma and urine were measured by enzyme immune assay and the tissue hydroxyproline content by a colorimetric assay. (b–d) Sections of renal tissue were stained with (b) periodic acid–Schiff (bars ¼ 20 mm) and with (c,d) Masson trichrome (bars ¼ 100 mm) and then (e,f) quantified using a scoring system or the WinROOF imaging software (Mitani Corporation, Tokyo, Japan). The number of mice for kidney tissue evaluation: for wild-type (WT) mice, n ¼ 4 at 4, 12, and 20 weeks; for TG mice, n ¼ 7 at 4 weeks, n ¼ 8 at 16 weeks, and n ¼ 9 at 20 week-olds. The number of mice for plasma and urine evaluation: for WT mice, n ¼ 12 at 4 weeks, n ¼ 8 at 8 weeks, n ¼ 7 at 12 weeks, and n ¼ 4 at 16 and 20 weeks; for TG mice, n ¼ 24 at 4 weeks, n ¼ 17 at 8 and 12 weeks, and n ¼ 9 at 16 and 20 weeks. Data are expressed as median interquartile range. Statistical analysis by Mann-Whitney U test. *P < 0.05, **P < 0.01, ****P < 0.0001. NS, not significant. To optimize viewing of this image, please see the online version of this article at www.kidney-international.org.

Figure 2 | Transmission electron microscopic findings in the transforming growth factor b1–induced kidney fifibrosis model. Fixation, handling, and removal of the kidneys from mice were performed as described in the Methods. (a,b) Microvillous transformation and (a,b,d,e,g) foot process effacement (white arrowheads) of podocytes, (c–e) decreased glomerular capillary endothelial fenestration (yellow arrowheads), (f) thickening of the glomerular basement membrane (asterisks), and (h,i) increased mesangial matrix deposition (white arrows) are present. CL, capillary lumen. To optimize viewing of this image, please see the online version of this article at www.kidney-international.org.

rhTM ameliorates renal function

The levels of L-fatty acid-binding protein (197.0 vs. 313.4 pg/ ml), kidney injury molecule 1 (299.9 vs. 596.2 pg/ml), blood urea nitrogen (12.9 vs 34.8 mg/dl), creatinine (0.5 vs. 1.4 mg/ dl), and the albumin-creatinine ratio were signifificantly decreased in TGFb1-TG mice with renal fifibrosis treated with rhTM compared with their untreated TG counterparts (Figure 5). The urine total protein and the total protein creatinine ratio were also decreased in TGFb1-TG mice treated with rhTM compared with their untreated counterparts (Figure 5).

rhTM reduces apoptosis of glomerular cells

Terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling staining showed a signifificantly decreased number of apoptotic cells in glomeruli from TGFb1-TG mice treated with rhTM compared with glomeruli from TGFb1- TG mice treated with SAL (Supplementary Figure S10A and B). Cleavage of caspase-3 was also signifificantly reduced in kidney tissues from TGFb1-TG mice treated with rhTM compared with kidney tissues from TGFb1-TG mice treated with SAL (Supplementary Figure S10C). The kidney tissues from TGFb1-TG mice treated with rhTM, compared with those from TGFb1-TG treated with SAL, showed significant elevation in mRNA levels of B-cell lymphoma 2 (Bcl-2), B-cell lymphoma-extra large (Bcl-XL), a baculoviral inhibitor of apoptosis repeat-containing 5 (BIRC5, also known as survivin), and BIRC6 (Apollon) with increased Bcl-2–Bax ratio (Supplementary Figure S11).

Figure 3 | The human transforming growth factor b1 (TGFb1) transgenic (TG) mouse has renal dysfunction. (a) Total protein and (b) blood urea nitrogen (BUN) were measured by colorimetric methods and creatinine by an enzymatic method. The number of mice for plasma and urine evaluation: for wild-type (WT) mice, n ¼ 12 at 4 weeks, n = 7 at 8 and 12 weeks, and n = 4 at 16 and 20 weeks; for TG mice, n =24 at 4 weeks, n = 17 at 8 and 12 weeks, and n = 9 at 16 and 20 weeks. Data are expressed as median ± interquartile range. Statistical analysis by Mann-Whitney U test. *P < 0.05, **P < 0.01, **** P < 0.0001.

rhTM inhibits apoptosis of podocytes

Pretreatment of podocytes with rhTM signifificantly decreased apoptosis of podocytes cultured in the presence of TGFb1 as assessed by the number of cells in the subG1 phase (3.2% vs. 5.2%), terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling–positive cells (1.0 vs. 5.4 cells/Fifield) and the degree of caspase-3 cleavage (0.9 vs. 1.1 ratios) (Figure 6a–e). Screening of antiapoptotic factors in cultured podocytes showed that rhTM signifificantly increases the mRNA expression of the antiapoptotic factor Bcl-2 compared with expression in untreated cells (Supplementary Figure S12). The mRNA expression of the antiapoptotic factor BIRC5 also increased in cells treated with rhTM compared with expression in untreated cells (Supplementary Figure S12). The mRNA expression of the proapoptotic factor Bax was signifificantly reduced by rhTM treatment compared with no treatment (Supplementary Figure S12). Treatment with rhTM also signifificantly inhibited the expression of annexin V and terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling staining in podocytes cultured in the presence of hydrogen peroxide (Supplementary Figure S13A–E) and under high glucose condition (Supplementary Figure S14A–E) further confirming the antiapoptotic property of rhTM on podocytes. Exploration of the antiapoptotic protein kinase B (Akt) pathway25 showed that rhTM enhanced the phosphorylation of Akt in human primary podocytes cultured in the presence of hydrogen peroxide or TGFb1 (Supplementary Figure S15A and B). We then isolated podocytes from each group of mice and evaluated Akt phosphorylation by Western blotting. There was signifificantly enhanced phosphorylation of Akt in podocytes isolated from the TGFb1-TG/rhTM group compared with podocytes from the untreated group (1.1 vs. 0.7 ratios) (Supplementary Figure S16A and B).

GPR15 mediation

Previous studies reported that TM activates intracellular pathways by interacting with fibroblast growth factor receptor 1 (FGFR1) and G-protein coupled receptor 15 (GPR15).26,27 Podocytes express FGFR128 but whether they express GPR15 is unclear. Here we isolated podocytes from each group of mice and showed that podocytes also express GPR15 (Supplementary Figure S17A–E). We found that podocytes from healthy control and a patient with glomerulosclerosis also express GPR15 (Supplementary Figure S18). To clarify whether FGFR1 or GPR15 mediates the rhTM inhibitory activity on apoptosis of podocytes, we evaluated the antiapoptotic activity of rhTM in TGFb1-treated podocytes in the presence of FGFR1 inhibitor or after transfecting the cells with small interfering RNA (siRNA) against FGFR1 or GPR15. Pretreatment of podocytes with FGFR1 inhibitor (Supplementary Figure S19A and B) or FGFR1 siRNA (13.2% vs. 7.9%) (Figure 7a and b) was unable to abolish the inhibitory activity of rhTM on apoptosis of podocytes. However, transfection of cells with GPR15 siRNA completely abolished the inhibitory activity of rhTM (14.6% vs. 13.8%) on apoptosis of podocytes (Figure 7a–c).

Figure 4 | Recombinant human thrombomodulin (rhTM) inhibits glomerulosclerosis and tubulointerstitial fifibrosis. Sections of renal tissue were stained (a,b) with periodic acid–Schiff and (c,d) with Masson trichrome and (e) then quantified using a scoring system or the WinROOF imaging software. (e) The mean value of the wild-type (WT)/saline (SAL) group was taken as 100%. Statistical analysis by Mann-Whitney U test. (f) The tissue hydroxyproline content was measured by a colorimetric method, concentration of collagen I-a1 (Col1a1) and periostin by enzyme immunoassay, and mRNA expression by reverse transcriptase-polymerase chain reaction. Statistical analysis by Kruskal Wallis analysis of variance and corrected Dunn test. n = 8 in each group. Bars = (a,c) 50 mm and (d) 20 mm. Data are expressed as median ± interquartile range. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. TG, transgenic; TGFb1, transforming growth factor b1. To optimize viewing of this image, please see the online version of this article at www.kidney-international.org.

rhTM inhibits EMT of podocytes

There was a signifificantly increased area with positive staining for podocin and a-smooth muscle actin (a-SMA) (1.4% vs. 11.2%) in TGFb1-TG/SAL mice compared with TGFb1-TG/ rhTM mice (Figure 8a and b). We then cultured in vitro primary human podocytes, pretreated with rhTM before adding TGFb1 protein to the culture medium. The fibroblast-like morphology and enhanced expression of a-SMA were suppressed in podocytes treated with rhTM compared with untreated cells (Supplementary Figure S20A). In addition, rhTM inhibited the mRNA expression of fibronectin and vimentin, although it enhanced the mRNA expression of E-cadherin in podocytes compared with expression in untreated cells (Supplementary Figure S20B). SMAD family members 2 (Smad2) and Smad3 play a critical role in TGFb1-mediated epithelial-mesenchymal transition (EMT).29 Treatment with rhTM signifificantly suppressed the activation of Smad2 and Smad3 in TGFb1-TG mice compared with untreated TG mice (Figure 8c) and in human podocytes cultured in the presence of TGFb1 (Supplementary Figure S20C).29 TG mice treated with rhTM (TGFb1-TG/rhTM) also show less expression of a-SMA (3.7% vs. 17.2%) in tubular epithelial cells compared with their untreated counterparts (Supplementary Figure S21A and B).

Figure 5 | Recombinant human thrombomodulin (rhTM) ameliorates kidney injury and renal dysfunction. Creatinine was measured by an enzymatic method; total protein by colorimetric method; and blood urea nitrogen (BUN) and albumin, kidney injury molecule 1 (KIM-1), L-fatty acid-binding protein (L-FABP), and total transforming growth factor b1 (TGFb1) by enzyme immune assay. n = 8 in each group. Data are expressed as median±interquartile range. Statistical analysis by Kruskal-Wallis analysis of variance and uncorrected Dunn test. * P < 0.05, **P < 0.01, ***P < 0.001, **** P < 0.0001, #P = 0.06. NS, not significant; SAL, saline; TG, transgenic; WT, wild type.

GPR15 mediates the inhibitory activity of rhTM on EMT

Transfection of podocytes with FGFR1 siRNA was unable to abolish the inhibitory activity of rhTM on the relative mRNA expressions of both collagen I-a1 and a-SMA in TGFb1- treated podocytes (Supplementary Figure S22). However, transfection of cells with GPR15 siRNA signifificantly abolished the inhibitory activity of rhTM on the relative mRNA expressions of both collagen I-a1 and a-SMA in TGFb1-treated podocytes (Supplementary Figure S22).

DISCUSSION

TGFb1 and injury of glomerular cells

The common consequence of disorders causing CKD is renal fifibrosis.8,30,31 TGFb1 is the common driver of fibrogenesis in the kidneys associated with CKD caused by diseases including DM, arterial hypertension, and autoimmune disorders.10 Cells from the glomerulus and tubulointerstitial spaces can secrete the latent forms of TGFb1 that, when excessively activated during tissue injury, may lead to renal scarring.32 As TGFb1 can stimulate its own secretion, the fibrotic process generally becomes a vicious cycle.8 An early event in the pathogenic process of TGFb1- mediated kidney fifibrosis is an injury of podocytes and glomerular endothelial cells.33–35 The plasma level of soluble TM is a marker of endothelial injury. Consistent with the role of TGFb1 in renal cell injury, here we found a significant correlation of active TGFb1 with soluble TM, soluble podocin, and creatinine in plasma from patients with DM. Crosstalk between glomerular endothelial cells and podocytes during a renal injury leads to a local expression of proteases causing glomerular basement membrane degradation.34–36 This may explain the detection of soluble podocin and its significant correlation with soluble TM in our patients with DM.

Figure 6 | Recombinant human thrombomodulin (rhTM) suppresses apoptosis of podocytes induced by transforming growth factor b1 (TGFb1). (a) rhTM was added to the culture medium of podocytes 1 hour before inducing apoptosis with 10 ng/ml TGFb1 for 48 hours. (b) The percentage of cells in the subG1 phase was detected by fellow cytometry. (a,b) n = 3 in each group. (c,d) The number of cells with DNA fragmentation was measured by terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) analysis (n = 3 in saline [SAL]/SAL and rhTM/SAL groups; n = 6 in SAL/TGFb1 and rhTM/TGFb1 groups), and (e) the degree of caspase-3 cleavage was measured by Western blotting (n = 4 in each group). Bars = 100 mm. Data are expressed as median ± interquartile range. Statistical analysis by Mann-Whitney U test. *P < 0.05. DAPI, 40 ,6-diamidino-2-phenylindole; HPF, high-powered Field; NS, not significant. Histograms are displayed as %Max (% of maximum value), scaling each curve to mode = 100%. To optimize viewing of this image, please see the online version of this article at www.kidney-international.org.

Podocyte-specific human TGFb1 overexpression-associated kidney fifibrosis

A drug that can counteract the effect of TGFb1 would be ideal to block renal fifibrosis. Here we have generated a TG mouse overexpressing the human TGFb1 gene in the glomerulus that develops spontaneous and progressive glomerular sclerosis and tubulointerstitial fifibrosis with renal failure as early as 4 weeks after birth. The model shows advanced glomerulopathy with injury of podocytes and glomerular endothelial cells; glomerular basement membrane thickening and mesangial expansion with interstitial scarring; increased markers of kidney tissue injury and renal dysfunction; and increased TGFb1 in plasma, renal tissue, and urine. The increase in urinary excretion of proteins, TGFb1, and activation of the complement system may explain the concurrent development of interstitial scarring in our present model.37–40 Further experiments revealed increased apoptotic cells and activation of Smad proteins in kidney tissue from untreated TGFb1-TG mice compared with WT mice. Overall, these findings point to this novel TGFb1-TG mouse as a suitable model for drug discovery in kidney fifibrosis.

Figure 7 | G-protein coupled receptor (GPR15) mediates inhibition of apoptosis recombinant human thrombomodulin (rhTM) in podocytes. Human primary podocyte cells were transfected with fibroblast growth factor receptor 1 (FGFR1) small interfering RNA (siRNA), GPR15 siRNA, or scrambled siRNA for 48 hours, and then rhTM was added to the cell culture 1 hour before treating with transforming growth factor b1 (TGFb1). Apoptotic cells were (a) assessed by fellow cytometry and (b) then quantified. (c) Cell lysates were prepared for Western blotting. n=3 in each group. Data are expressed as median ± interquartile range. Statistical analysis by Mann-Whitney U test. *P < 0.05, #P =0.1. SAL, saline.

Figure 8 | Inhibition of epithelial-mesenchymal transition by recombinant human thrombomodulin (rhTM). (a) Podocin and a-smooth muscle actin (a-SMA) were stained as described in the Methods. (b) The area positive for a-SMA staining was quantified by the WinROOF image processing software. n = 3 in wild-type (WT)/saline (SAL) and WT/rhTM groups and n = 5 in transforming growth factor-b1–transgenic (TGFb1-TG)/SAL and TGFb1-TG/rhTM groups. (c) Total (t) and phosphorylated (p) SMAD family member (Smad) proteins were evaluated by Western blotting. n = 8 in each group. Data are expressed as median ± interquartile range. Statistical analysis by Kruskal-Wallis analysis of variance and corrected Dunn test. *P < 0.05, **P < 0.01, ***P < 0.001, #P = 0.08. To optimize viewing of this image, please see the online version of this article at www.kidney-international.org.

rhTM ameliorates kidney fifibrosis

We have shown that rhTM ameliorates pulmonary fifibrosis developed in mice overexpressing human TGFb1 in the lungs by suppressing alveolar epithelial cell apoptosis.41 Clinical trials have also demonstrated amelioration of idiopathic pulmonary fifibrosis after rhTM treatment.42,43 These previous observations suggest the potential of rhTM for organ fifibrosis therapy. We hypothesized that rhTM would be effective in TGFb1- associated kidney fifibrosis. To test this hypothesis, we treated kidney-specific TGFb1-TG mice with rhTM.44 Administration of rhTM for 4 weeks signifificantly attenuated injury, dysfunction, and fifibrosis of the kidneys. Mice treated with rhTM showed low urinary levels of total protein, albumin, TGFb1, C5a, and reduced renal levels of profibrotic cytokines, C5a and HMGB1.45 Therapy with rhTM also inhibited both apoptosis and EMT of podocytes. However, rhTM exerted no effect on thrombin antithrombin complex, a coagulation activation marker, although it increased the generation of APC, an anticoagulant factor with anti-inflflammatory and anti-apoptotic activity.23 It is worth noting that thrombin, the procoagulant enzyme, may paradoxically promote anti-coagulation under low-grade prothrombotic states by forming the TM/thrombin complex, which increases the generation of the anticoagulant APC. However, thrombin works mainly as a procoagulant under excessive prothrombotic states (e.g., sepsis).46 This dual and paradoxical effect of thrombin may explain the apparent inefficacy of rhTM to inhibit coagulation activation in our model, which is in a low-grade prothrombotic state. Overall, these observations suggest that rhTM amelio-rates progressive fifibrosis and dysfunction of the kidneys by prolonging the survival of or preventing EMT of glomerular cells and by suppressing inflflammation, complement activation, and growth factor activity directly or indirectly via activation of the protein C pathway and reduction of HMGB1 expression. Improvement of diabetic nephropathy in mice with increased circulating TM lectin-like domain and disease deterioration in mice lacking the lectin-like domain support the beneficial effect of rhTM on renal fifibrosis.20,21

Inhibition of podocyte apoptosis

Podocytes play critical roles in the maintenance of the glomerular filtration barrier and the formation of the slit diaphragm to prevent the loss of essential circulating proteins.47 Renal injury caused by reactive oxygen species, high glucose, or inflflammatory mediators including TGFb1 induces apoptosis of podocytes leading to podocyte depletion that may eventually end up causing renal dysfunction.47 After binding and activating the serine/threonine kinase transmembrane heteromeric type I and type II receptor complex, TGFb1 emits intracellular signals through the Smad family of transcription factors or through Smad-independent signaling pathways.48 Activation of the Smad-dependent pathway occurs when activated TGFb1 receptor phosphorylates Smad2 and Smad3, which with Smad4 translocate to the nucleus.49 The Smad2/Smad3/Smad4 complex stimulates the transcription of proapoptotic factors and reduces that of antiapoptotic factors leading to cell apoptosis.49 Consistent with this, we found the high renal level of the proapoptotic factor Bax and a low level of the antiapoptotic factors Bcl2 and Bcl-XL in TGFb1- TG mice. TM can inhibit apoptosis of different cell types.21,41,49 Consistent with this, here we found that rhTM inhibits apoptosis of podocytes induced by hydrogen peroxide, high glucose, or TGFb1, and this finding may explain the beneficial effect of rhTM in CKD. In addition, rhTM tilted the balance toward inhibition of apoptosis by reducing the expression of Bax, by increasing the expression of Bcl2, Bcl-XL, BIRC5, and BIRC6, and by increasing the activation of the Akt pathway in kidney tissues. Overall, these observations suggest that rhTM stimulates podocyte survival by favoring Akt activation and antiapoptotic factor expression.

cistanche for chronic kidney disease

Inhibition of EMT

EMT of podocytes also contributes to renal fifibrosis. Podocytes undergoing EMT release extracellular matrix proteins that accumulate and deposit during TGFb1-associated renal fifibrogenesis.50 TGFb1 promotes EMT through receptor-mediated activation of the Smad2/Smad3/Smad4 complex. Phosphorylated Smad3 promotes EMT by stimulating the transcription of matrix proteins and by reducing the expression of epithelial markers.51 Consistent with this, we found increased EMT of podocytes in TGFb1-TG mice and podocytes cultured in the presence of TGFb1 or under oxidant or high glucose conditions. Previous reports suggested that rhTM suppresses EMT.52,53 Here we found that rhTM inhibits EMT of podocytes and tubulointerstitial epithelial cells in TGFb1-TG mice. Inhibition of Smad protein activation appears to be the mechanism of the beneficial effect of rhTM on EMT because TGFb1-TG mice and primary podocytes treated with rhTM depicted signifificantly reduced phosphorylation of Smad2 and Smad3 compared with untreated conditions. These findings support the inhibitory activity of TM on EMT.

Receptor mediation in rhTM protective activity

Previous studies have shown that GPR15 or FGFR1 mediates the cytoprotective activity of TM.26,27,54 We tested whether these receptors mediate the protective activity of rhTM against apoptosis and EMT. While downregulation of GPR15 protein by siRNA completely abolished the suppressive activity of rhTM on TGFb1-mediated apoptosis of podocytes, neither FGFR1 siRNA nor its inhibitor abolished it, suggesting that GPR15 mediates the rhTM protective activity. There was enhanced phosphorylation of Akt in cultured podocytes and podocytes isolated from TGFb1-TG mice after treatment with rhTM, suggesting the involvement of the intracellular Akt pathway. A work showing that rhTM activates the Akt pathway in endothelial cells supports this finding.55 Activation of the Akt signaling pathway may further potentiate the rhTM effect by increasing the GPR15 surface expression.56 These observations indicate that rhTM protects podocytes from apoptosis in our TGFb1-TG mice by activating the GPR15/Akt axis leading to enhanced expression of anti-apoptotic factors and decreased expression of proapoptotic factors.57 On the other hand, previous studies demonstrated that the anaphylatoxins C3a and C5a through their GPRs may contribute to CKD by injuring podocytes and that APC may inhibit podocyte apoptosis and kidney fifibrosis via its endothelial protein C receptor and protease-activated receptor 1.23,58–61 Here, we found that rhTM inhibits the complement system and increases the generation of APC. Therefore, apart from the activation of the GPR15/Akt pathway, inhibition of complement factors and increased activation of protein C and its receptors may also explain the beneficial effects of rhTM in our TGFb1-associated kidney fifibrosis model.

In addition, downregulation of GPR15 but not that of FGFR1 blocked the inhibitory effect of rhTM on EMT, suggesting that GPR15 also mediates this rhTM protective activity. Inhibition of Smad proteins is involved in EMT suppression because treatment with rhTM inhibited the phosphorylation of both Smad2 and Smad3 in TGFb1-TG mice and cultured podocytes. However, the precise mechanism of Smad protein inhibition via GPR15 is unclear. Some evidence shows that the Akt pathway may crosstalk with and regulate the Smad signaling pathway,62–64 and that Akt can prevent phosphorylation of Smad3 by directly interacting with unphosphorylated Smad3 to sequester it outside the nucleus leading to inhibition of transcription and EMT.62–64 Based on these reports, it is conceivable that Akt activated after rhTM binding to GPR15 sequesters unphosphorylated Smad3 leading to podocyte EMT inhibition. It is worth noting that this Akt-mediated beneficial effect is observed only in nonmalignant cells.65–67 In malignant cells, the interaction of TGFb1 with its receptors may directly activate the phosphoinositide 3-kinase/Akt/Snail pathway and cause EMT.65–67 Overall, the results of our study support the role of GPR15 as the receptor mediating the beneficial effects of rhTM on podocytes.

Conclusion

In summary, here we report for the first time a novel transgenic TG mouse overexpressing the full-length of human TGFb1 gene specifically in glomeruli that develop spontaneous and progressive glomerular sclerosis, tubulointerstitial fifibrosis and renal failure, and amelioration of established kidney fifibrosis/renal failure by rhTM interaction with GPR15 that inhibits apoptosis and mesenchymal transition of podocytes.

METHODS

Generation of the TGFb1 BAC TG mouse

TGFb1-BAC-TG mouse expressing the full-length human TGFb1 gene under the mouse podocin promoter control was generated by pronuclear injection into 392 C57BL/6J mouse embryos (CLEA Japan, Inc., Tokyo, Japan). We assessed the TG founders and germline transmission of the BAC TG construct by Southern blotting (Supplementary Materials and Methods).

Experimental animals

The TGFb1-TG mice were bred for more than 10 generations under C57BL/6 background before using in the experiments. WT littermates were used as the control animals. All animals were maintained in a specific pathogen-free environment and subjected to a 12-hour light-dark cycle at an ambient temperature and humidity ranging between 22℃ and 26 ℃ and 40% and 70%, and with ad libitum access to food and water in the animal house of Mie University (Supplementary Materials and Methods).

Ethical statement

The Recombinant DNA Experiment Safety Committee (approval no. I-629; date: September 19, 2013) and the Committee on Animal Investigation of Mie University (approval no. 27-4; date: August 19, 2015) approved the protocols of the study. All animal procedures were performed in accordance with the institutional guidelines of Mie University and following the internationally approved principles of laboratory animal care published by the National Institute of Health (https://olaw.nih.gov/).

For the clinical investigation, written informed consent was given by all patients and healthy subjects, and the study protocol was approved by the Ethics Committee for Clinical Investigation of Mie University (approval nos. 1043 and 2194).

Experimental design

For characterization of the kidney fifibrosis model, we allocated male TGFb1-TG mice (n = 24) and male WT mice (n = 12) 4 weeks of age and weighing 20 to 23 g into 3 groups with 8 TGF b1-TG mice and 4 WT mice in each group. Mice from each WT and TGFb1-TG group were euthanized on weeks 0, 8, or 16 to collect samples of urine, blood, and kidney for assessing changes in fifibrosis and renal function parameters over time.

To evaluate the therapeutic effectiveness of rhTM (rhTM was kindly provided by Asahi Kasei Pharma Corporation, Tokyo, Japan) in renal fifibrosis, TGFb1-TG mice (n= 8) or WT littermates (n = 8) were treated with rhTM (3 mg/kg) by i.p. injection, 3 times a week during the 4 weeks before the mice were killed. TGFb1-TG mice (n =8) or WT littermates (n= 8) receiving an equal amount of physiological SAL by i.p. injection were used as negative control mice.

The protocol of the present study followed the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines for animal investigation. Mice were randomized, and researchers who measured parameters were blinded to treatment groups.

Biochemical analysis

The concentrations of total protein (BCA protein assay kit; Pierce, Rockford, IL), TGFb1 (R&D System, Minneapolis, MN), monocyte chemoattractant protein-1 (BD Biosciences Pharmingen, San Diego, CA), thrombin antithrombin complex (Cedarlane Laboratories, Hornby, Ontario, Canada) were measured using commercial enzyme immunoassay kits following the manufacturer’s instructions (Supplementary Materials and Methods).

Cell culture

The human podocyte primary cells were purchased from CELPROGEN (Torrance, CA). Human podocyte primary cells were cultured in Dulbecco’s modified Eagle medium in a humidified, 5% CO2 atmosphere at 37℃. The medium was supplemented with 10% heat-inactivated fetal bovine serum (Bio Whittaker, Walkersville, MD), 100 IU/ml penicillin, 100 mg/ml streptomycin, and L-glutamine (Supplementary Materials and Methods).

Statistical analysis

Data are expressed as median±interquartile range. The statistical difference between variables was calculated by Kruskal-Wallis analysis of variance with post hoc analysis using the Dunn test. Mann-Whitney U test was used to evaluate differences between 2 groups. Statistical analyses were done using the GraphPad Prism version 8.0.1 (GraphPad Software, San Diego, CA). Statistical significance was considered as P < 0.05.

DISCLOSURE

ECG, CND-G, and YY have a patent on the TGFb1-TG mouse with renal fifibrosis used in the present study. CND-G and YY received a grant from the Ministry of Education, Culture, Sports, Science, and Technology of Japan for the present study. ECG, TY, CND-G, and MT received a grant from Shionogi Pharmaceuticals. All the other authors declared no competing interests.

ACKNOWLEDGMENTS

This research was supported in part by grants from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (Kakenhi no. 17K09824 for YY; Kakenhi no. 17K08442 for CND-G), and in part by a grant from Shionogi & Co, Ltd., Japan. The funders had no role in study design, data analysis, decision to publish, or preparation of the manuscript.

Part of this work was published in abstract form.

AUTHOR CONTRIBUTIONS

AT prepared the disease model and wrote the first draft of the manuscript. TY, KN, TT, RI, and CND-G prepared the disease models and measured parameters. AM and SW performed the transmission microscopic study. MT, YO, and AU measured parameters and performed the in vitro experiments. YT, LQ, TK, and VFD provided intellectual contributions. YY and ECG corrected the manuscript draft and designed the study.

cistanche for kidney failure symptoms

SUPPLEMENTARY MATERIAL

Supplementary File (PDF) Supplementary Materials and Methods.

Table S1. Characteristics of the subjects.

Table S2. Primers for RT-PCR of mouse tissues.

Table S3. Primers for RT-PCR of human podocytes.

Figure S1. Soluble thrombomodulin fragments correlate with TGFb1 and creatinine in patients with DM.

Figure S2. Human TGFb1–bacterial artificial chromosome (BAC) construct. Figure S3. Founder mice expressing the full-length of human TGFb1 gene.

Figure S4. Characterization of the glomerulus-specific transforming growth factor b1 transgenic mouse.

Figure S5. The human TGFb1 transgenic mouse has increased markers of kidney injury.

Figure S6. Therapy with recombinant human thrombomodulin (rhTM) reduces foot process effacement of podocytes and thickening of the glomerular basement membrane.

Figure S7. TGFb1 transgenic mice treated with recombinant human thrombomodulin (rhTM) have a low concentration of profibrotic factors and HMGB1 in renal tissue. Figure S8. Therapy with recombinant human thrombomodulin (rhTM) increases the generation of activated protein C, inhibits the complement system, decreased circulating soluble podocin, although exerts no effect on the coagulation system in TGFb1 transgenic mice.

Figure S9. Therapy with recombinant human thrombomodulin (rhTM) reduces the urine concentration of podocin.

Figure S10. Therapy with recombinant human thrombomodulin (rhTM) reduces apoptosis of glomerular cells.

Figure S11. Treatment of TGFb1-overexpression associated kidney fifibrosis with recombinant human thrombomodulin (rhTM) inhibits apoptosis in renal tissue.

Figure S12. Recombinant human thrombomodulin (rhTM) increases the expression of antiapoptotic factors in podocytes.

Figure S13. Recombinant human thrombomodulin (rhTM) suppresses apoptosis of podocytes induced by hydrogen peroxide.

Figure S14. Recombinant human thrombomodulin (rhTM) suppresses apoptosis of podocytes induced by high glucose levels.

Figure S15. Recombinant human thrombomodulin (rhTM) increases the activation of the Akt pathways in podocytes.

Figure S16. Therapy with recombinant human thrombomodulin (rhTM) increases Akt phosphorylation in podocytes from TGFb1-TG mice.

Figure S17. Podocytes from each treatment group of mice express GPR15 mRNA.

Figure S18. Staining of GPR15 in podocytes from a healthy individual and a patient with focal segmental glomerulosclerosis.

Figure S19. Fibroblast growth factor receptor-1 is not involved in the inhibitory activity of recombinant human thrombomodulin (rhTM) in podocytes.

Figure S20. Recombinant human thrombomodulin (rhTM) inhibits the epithelial-mesenchymal transition of podocytes.

Figure S21. Therapy with recombinant human thrombomodulin (rhTM) inhibits the expression of a-smooth muscle actin in renal tubules.

Figure S22. G-protein coupled receptor (GPR15) mediates the inhibitory activity of recombinant human thrombomodulin (rhTM) on the epithelial-mesenchymal transition of podocytes.

Supplementary References.

REFERENCES

1. Webster AC, Nagler EV, Morton RL, et al. Chronic kidney disease. Lancet. 2017;389:1238–1252.

2. Bello AK, Levin A, Tonelli M, et al. Assessment of global kidney health care status. JAMA. 2017;317:1864–1881.

3. Levin A, Tonelli M, Bonventre J, et al. Global kidney health 2017 and beyond: a roadmap for closing gaps in care, research, and policy. Lancet. 2017;390:1888–1917.

4. Ackland P. Prevalence, detection, evaluation, and management of chronic kidney disease. BMJ. 2014;348:f7688.

5. Turner JM, Bauer C, Abramowitz MK, et al. Treatment of chronic kidney disease. Kidney Int. 2012;81:351–362.

6. Breyer MD, Susztak K. The next generation of therapeutics for chronic kidney disease. Nat Rev Drug Discov. 2016;15:568–588.

7. Liu Y. Renal fifibrosis: new insights into the pathogenesis and therapeutics. Kidney Int. 2006;69:213–217.

8. Liu Y. Cellular and molecular mechanisms of renal fifibrosis. Nat Rev Nephrol. 2011;7:684–696.

9. Xavier S, Vasko R, Matsumoto K, et al. Curtailing endothelial TGF-beta signaling is sufficient to reduce endothelial-mesenchymal transition and fifibrosis in CKD. J Am Soc Nephrol. 2015;26:817–829.

10. Higgins SP, Tang Y, Higgins CE, et al. TGF-beta1/p53 signaling in renal fifibrogenesis. Cell Signal. 2018;43:1–10.

11. Conway EM. Thrombomodulin and its role in inflflammation. Semin Immunopathol. 2012;34:107–125.

12. Martin FA, Murphy RP, Cummins PM. Thrombomodulin and the vascular endothelium: insights into functional, regulatory, and therapeutic aspects. Am J Physiol Heart Circ Physiol. 2013;304:H1585–H1597.

13. Morser J. Thrombomodulin links coagulation to inflflammation and immunity. Curr Drug Targets. 2012;13:421–431.

14. Roeen Z, Toda M, D’Alessandro-Gabazza CN, et al. Thrombomodulin inhibits the activation of eosinophils and mast cells. Cell Immunol. 2015;293:34–40.

15. Takagi T, Taguchi O, Toda M, et al. Inhibition of allergic bronchial asthma by thrombomodulin is mediated by dendritic cells. Am J Respir Crit Care Med. 2011;183:31–42.

16. Tateishi K, Imaoka M, Matsushita M. Dual modulating functions of thrombomodulin in the alternative complement pathway. Biosci Trends. 2016;10:231–234.

17. Toda M, D’Alessandro-Gabazza CN, Takagi T, et al. Thrombomodulin modulates dendritic cells via both antagonisms of high mobility group protein B1 and an independent mechanism. Allergol Int. 2014;63:57–66.

18. Van de Wouwer M, Plaisance S, De Vriese A, et al. The lectin-like domain of thrombomodulin interferes with complement activation and protects against arthritis. J Thromb Haemost. 2006;4:1813–1824.

19. Sharfuddin AA, Sandoval RM, Berg DT, et al. Soluble thrombomodulin protects ischemic kidneys. J Am Soc Nephrol. 2009;20:524–534.

20. Wang H, Vinnikov I, Shahzad K, et al. The lectin-like domain of thrombomodulin ameliorates diabetic glomerulopathy via complement inhibition. Thromb Haemost. 2012;108:1141–1153.

21. Yang SM, Ka SM, Wu HL, et al. Thrombomodulin domain 1 ameliorates diabetic nephropathy in mice via anti-NF-kappaB/NLRP3 inflammasome-mediated inflflammation, enhancement of NRF2 antioxidant activity, and inhibition of apoptosis. Diabetologia. 2014;57:424–434.

22. Ohlin AK, Larsson K, Hansson M. Soluble thrombomodulin activity and soluble thrombomodulin antigen in plasma. J Thromb Haemost. 2005;3: 976–982.

23. Gil-Bernabe P, D’Alessandro-Gabazza CN, Toda M, et al. Exogenous activated protein C inhibits the progression of diabetic nephropathy. J Thromb Haemost. 2012;10:337–346.

24. Yasuma T, Yano Y, D’Alessandro-Gabazza CN, et al. Amelioration of diabetes by protein S. Diabetes. 2016;65:1940–1951.

25. Havasi A, Borkan SC. Apoptosis and acute kidney injury. Kidney Int. 2011;80:29–40.

26. Kuo CH, Sung MC, Chen PK, et al. FGFR1 mediates recombinant thrombomodulin domain-induced angiogenesis. Cardiovasc Res. 2015;105:107–117.

27. Pan B, Wang X, Kojima S, et al. The fifth epidermal growth factor-like region of thrombomodulin alleviates LPS-induced sepsis through interacting with GPR15. Thromb Haemost. 2017;117:570– 579.

28. Lu Y, Ye Y, Bao W, et al. Genome-wide identification of genes essential for podocyte cytoskeletons based on single-cell RNA sequencing. Kidney Int. 2017;92:1119–1129.

29. Vigolo E, Marko L, Hinze C, et al. Canonical BMP signaling in tubular cells mediates recovery after acute kidney injury. Kidney Int. 2019;95:108–122.

30. Nangaku M. Chronic hypoxia and tubulointerstitial injury: a final common pathway to end-stage renal failure. J Am Soc Nephrol. 2006;17: 17–25.

31. Thomas R, Kanso A, Sedor JR. Chronic kidney disease and its complications. Prim Care. 2008;35:329–344, vii.

32. Mozes MM, Bottinger EP, Jacot TA, et al. Renal expression of fibrotic matrix proteins and of transforming growth factor-beta (TGF-beta) isoforms in TGF-beta transgenic mice. J Am Soc Nephrol. 1999;10:271– 280.

33. Arif E, Solanki AK, Srivastava P, et al. The motor protein Myo1c regulates transforming growth factor-beta-signaling and fifibrosis in podocytes. Kidney Int. 2019;96:139–158.

34. Ebefors K, Wiener RJ, Yu L, et al. Endothelin receptor-A mediates degradation of the glomerular endothelial surface layer via pathologic crosstalk between activated podocytes and glomerular endothelial cells. Kidney Int. 2019;96:957–970.

35. Fu J, Lee K, Chuang PY, et al. Glomerular endothelial cell injury and cross talk in diabetic kidney disease. Am J Physiol Renal Physiol. 2015;308:F287– F297.

36. Masum MA, Ichii O, Elewa YHA, et al. Modifified scanning electron microscopy reveals pathological crosstalk between endothelial cells and podocytes in a murine model of membranoproliferative glomerulonephritis. Sci Rep. 2018;8:10276.

37. Abbate M, Zoja C, Rottoli D, et al. Proximal tubular cells promote fifibrogenesis by TGF-beta1-mediated induction of peritubular myofifibroblasts. Kidney Int. 2002;61:2066–2077.

38. Liu BC, Tang TT, Lv LL, et al. Renal tubule injury: a driving force toward chronic kidney disease. Kidney Int. 2018;93:568–579.

39. Loefflfler I, Wolf G. Transforming growth factor-beta and the progression of renal disease. Nephrol Dial Transplant. 2014;29(suppl 1):i37–i45.

40. Murakami K, Takemura T, Hino S, et al. Urinary transforming growth factor-beta in patients with glomerular diseases. Pediatr Nephrol. 1997;11:334–336.

41. Fujiwara K, Kobayashi T, Fujimoto H, et al. Inhibition of cell apoptosis and amelioration of pulmonary fifibrosis by thrombomodulin. Am J Pathol. 2017;187:2312–2322.

42. Kataoka K, Taniguchi H, Kondoh Y, et al. Recombinant human thrombomodulin in acute exacerbation of idiopathic pulmonary fifibrosis. Chest. 2015;148:436–443.

43. Tsushima K, Yamaguchi K, Yokoyama T, et al. Thrombomodulin for acute exacerbations of idiopathic pulmonary fifibrosis: a proof of concept study. Pulm Pharmacol Ther. 2014;29:233–240.

44. Umemura Y, Yamakawa K. Optimal patient selection for anticoagulant therapy in sepsis: an evidence-based proposal from Japan. J Thromb Haemost. 2018;16:462–464.

45. Chen Q, Guan X, Zuo X, et al. The role of high mobility group box 1 (HMGB1) in the pathogenesis of kidney diseases. Acta Pharm Sin B. 2016;6:183–188.

46. Miyake Y, D’Alessandro-Gabazza CN, Takagi T, et al. Dose-dependent differential effects of thrombin in allergic bronchial asthma. J Thromb Haemost. 2013;11:1903–1915.

47. Assady S, Wanner N, Skorecki KL, et al. New insights into podocyte biology in glomerular health and disease. J Am Soc Nephrol. 2017;28: 1707–1715.

48. Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways in TGF-beta family signaling. Nature. 2003;425:577–584.

49. Schuster N, Krieglstein K. Mechanisms of TGF-beta-mediated apoptosis. Cell Tissue Res. 2002;307:1–14.

50. Greka A, Mundel P. Cell biology and pathology of podocytes. Annu Rev Physiol. 2012;74:299–323.

51. Isaka Y. Targeting TGF-beta signaling in kidney fifibrosis. Int J Mol Sci. 2018;19:2532.

52. Chang YJ, Cheng YW, Lin RK, et al. Thrombomodulin influences the survival of patients with non-metastatic colorectal cancer through epithelial-to-mesenchymal transition (EMT). PLoS One. 2016;11:e0160550.

53. Zheng N, Huo Z, Zhang B, et al. Thrombomodulin reduces tumorigenic and metastatic potential of lung cancer cells by up-regulation of E-cadherin and down-regulation of N-cadherin expression. Biochem Biophys Res Commun. 2016;476:252–259.

54. Pan B, Wang X, Nishioka C, et al. G-protein coupled receptor 15 mediates angiogenesis and cytoprotective function of thrombomodulin. Sci Rep. 2017;7:692.

55. Chen PS, Wang KC, Chao TH, et al. Recombinant thrombomodulin exerts anti-autophagic action in endothelial cells and provides anti-atherosclerosis effect in apolipoprotein E deficient mice. Sci Rep. 2017;7: 3284.

56. Chung JJ, Okamoto Y, Coblitz B, et al. PI3K/Akt signaling-mediated protein surface expression sensed by 14-3-3 interacting motif. FEB J. 2009;276:5547–5558.

57. Sanchez-Capelo A. Dual role for TGF-beta1 in apoptosis. Cytokine Growth Factor Rev. 2005;16:15–34.

58. Griffifin JH, Zlokovic BV, Mosnier LO. Activated protein C, protease-activated receptor 1, and neuroprotection. Blood. 2018;132:159–169.

59. Isermann B, Vinnikov IA, Madhusudhan T, et al. Activated protein C protects against diabetic nephropathy by inhibiting endothelial and podocyte apoptosis. Nat Med. 2007;13:1349–1358.

60. Klos A, Tenner AJ, Johswich KO, et al. The role of the anaphylatoxins in health and disease. Mol Immunol. 2009;46:2753–2766.

61. Morigi M, Perico L, Corna D, et al. C3a receptor blockade protects podocytes from injury in diabetic nephropathy. JCI Insight. 2020;5: e131849.

62. Conery AR, Cao Y, Thompson EA, et al. Akt interacts directly with Smad3 to regulate the sensitivity to TGF-beta induced apoptosis. Nat Cell Biol. 2004;6:366–372.

63. Derynck R, Muthusamy BP, Saeteurn KY. Signaling pathway cooperation in TGF-beta-induced epithelial-mesenchymal transition. Curr Opin Cell Biol. 2014;31:56–66.

64. Remy I, Montmarquette A, Michnick SW. PKB/Akt modulates TGF-beta signalling through a direct interaction with Smad3. Nat Cell Biol. 2004;6: 358–365.

65. Hamidi A, Song J, Thakur N, et al. TGF-beta promotes PI3K-AKT signaling and prostate cancer cell migration through the TRAF6-mediated ubiquitylation of p85alpha. Sci Signal. 2017;10:eaal4186.

66. Peng Z, Weber JC, Han Z, et al. Dichotomy effects of Akt signaling in breast cancer. Mol Cancer. 2012;11:61.

67. Zhou F, Geng J, Xu S, et al. FAM83A signaling induces epithelialmesenchymal transition by the PI3K/AKT/Snail pathway in NSCLC. Aging (Albany NY). 2019;11:6069–6088.