Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Diabetes Aggravates Post-ischaemic Renal Fibrosis through Persistent Activation of TGF-β1 and Shh Signalling

Diabetes Aggravates Post-ischaemic Renal Fibrosis through Persistent Activation of TGF-β1 and Shh... www.nature.com/scientificreports OPEN Diabetes Aggravates Post- ischaemic Renal Fibrosis through Persistent Activation of TGF-β and Received: 25 May 2017 Shh Signalling Accepted: 20 November 2017 Published: xx xx xxxx 2 3 1 1 1 Dong-Jin Kim , Jun Mo Kang , Seon Hwa Park , Hyuk-Kwon Kwon , Seok-Jong Song , 1 1 1 1 1 1 Haena Moon , Su-Mi Kim , Jung-Woo Seo , Yu Ho Lee , Yang Gyun Kim , Ju-Young Moon , 3 2 1 So-Young Lee , Y oungsook Son & Sang-Ho Lee Diabetes is a risk factor for acute kidney injury (AKI) and chronic kidney disease (CKD). Diabetic patients are easy to progress to CKD after AKI. Currently, activation of fibrotic signalling including transforming growth factor-β (TGF-β ) is recognized as a key mechanism in CKD. Here, we investigated the influence 1 1 of diabetes on CKD progression after AKI by using a unilateral renal ischaemia–reperfusion injury (IRI) model in diabetic mice. IRI induced extensive tubular injury, fibrosis and lymphocyte recruitment at 3 weeks after IRI, irrespective of diabetes. However, diabetes showed sustained tubular injury and markedly increased fibrosis and lymphocyte recruitment compared with non-diabetes at 5 week after IRI. The mRNAs and proteins related to TGF-β and sonic hedgehog (Shh) signalling were significantly higher in diabetic versus non-diabetic IRI kidneys. During the in vitro study, the hyperglycaemia induced the activation of TGF-β and Shh signalling and also increased profibrogenic phenotype change. However, hyperglycaemic control with insulin did not improve the progression of renal fibrosis and the activation of TGF-β and Shh signalling. In conclusion, diabetes promotes CKD progression of AKI via activation of the TGF-β and Shh signalling pathways, but insulin treatment was not enough for preventing the progression of renal fibrosis. Aer ac ft ute kidney injury (AKI), incomplete tubular recovery leads to renal fibrosis and decreased renal function, the common components of chronic kidney disease (CKD) . Generally, in patients with no underlying diseases, recovery from acute kidney injury occurs without significant renal fibrosis. However, transient kidney damage may eventually lead to renal fibrosis in the presence of underlying diseases such as diabetes and CKD . This phe- nomenon is not only limited to the kidney but can also occur aer s ft kin damage or hind limb ischaemia in animal 3,4 models with diabetes . Additionally, the presence of diabetes or underlying CKD are independent risk factors 5,6 for acute kidney injury aer c ft ardiac surgery and coronary/vascular interventions using contrast . Therefore, the transition of AKI to CKD is a clinically serious problem for diabetic patients. Over the past several decades, many studies have been conducted to identify the pathophysiology involved in the development of AKI. However, much remains unknown about the mechanism of transition from AKI to CKD. Recent studies have focused on the role of damaged tubules and a subpopulation of incompletely recov- ered tubules ae ft r AKI, which lead to abnormal growth arrest, failure to redifferentiate into normal tubules, and finally atrophy, as the result of abnormal wound healing . If abnormal wound healing persists or when metabolic derangements impair normal wound healing, atrophic tubules produce persistent and progressively increas- 8,9 ing levels of profibrotic signalling molecules such as TGF-β and Shh . These paracrine factors intrinsically 10,11 play a role in mediating normal wound repair . However, persistent activation of these signalling pathways and abnormal cross talk between unhealed tubular cells and interstitial cells such as infiltrating immune cells Division of Nephrology, Department of Internal Medicine, Kyung Hee University Hospital at Gangdong, College of Medicine, Kyung Hee University, Seoul, Korea. Department of Genetic Engineering, College of Life Science and Graduate School of Biotechnology, Kyung Hee University Global Campus, Yongin, Korea. Division of Nephrology, Department of Internal Medicine, CHA Bundang Medical Center, CHA University, Seongnam, Korea. Correspondence and requests for materials should be addressed to Y.S. (email: ysson@khu.ac.kr) or S.-H.L. (email: lshkidney@khu.ac.kr) SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 1 www.nature.com/scientificreports/ or activated fibroblasts eventually leads to myofibroblast transformation of pericyte-like fibroblast or bone 12,13 marrow-derived precursor cells, the final and common pathological feature of renal fibrosis . 9,14 Both of TGF-β and Shh pathways are known as typical signalling mediators which lead to renal fibrosis . Neutralization of TGF-β prevents blood vessel loss and development of tubulointerstitial fibrosis after IRI. 15,16 Additionally, blockage of Shh signalling also reduces renal fibrosis . Hyperglycaemia induces high expression 17,18 levels of TGF-β and increases levels of Smad 2/3 and CTGF induced by TGF-β . However, the relationship 1 1 between hyperglycaemia and the activation of the Shh pathway is currently unclear. It is also well known that the mechanism of the injury and repair process is abnormally controlled in the diabetic condition . In addition, aberrant inflammatory cell recruitment and activation of prob fi rotic signalling pathways are already among the major pathologic mechanisms of diabetic nephropathy . e uni Th lateral ischaemia reperfusion injury model is suitable for observing the progression of CKD because the characteristics of CKD such as renal mass reduction and tubulointerstitial fibrosis increase with the severity of 21,22 ischaemic-reperfusion injury . Therefore, we hypothesize that enhanced and persistent activation of prob fi rotic signalling molecules such as TGF-β and Shh under diabetic conditions induces abnormal b fi rotic repair rather than normal wound healing aer AKI, w ft hich finally accelerates the progression of CKD. Results Diabetes impaired the improvement of tubular injury and aggravated renal fibrosis after IRI. To investigate the effect of diabetes on the progression of post-ischaemic renal fibrosis, we used the unilateral renal ischaemia-reperfusion injury (IRI) model in non-diabetic and diabetic mice (Fig. 1a). Compared with sham treatment, IRI induced extensive tubular injury and increased the fibrotic area at 3 weeks aer IRI, ir ft respective of diabetes. While the degree of tubular injury between 3 and 5 weeks after IRI was significantly improved in non-diabetic mice, it was maintained in diabetic mice (Fig. 1b,d). Furthermore, continuous renal mass reduction was seen only in diabetic IRI (Table 1). The degree of renal fibrosis at 5 weeks after IRI was increased in both non-diabetic and diabetic mice. However, its progression was significantly higher in diabetic than non-diabetic mice (Fig. 1c,e). These results indicate that apoptotic tubular damage aer IRI l ft asts longer and is accompanied by the progression of renal fibrosis under diabetic conditions. Diabetes also aggravated aberrant lymphocyte recruitment after IRI. Because aberrant lympho- 23 + cyte recruitment of is one of the key features of diabetic nephropathy , we assessed the infiltration of CD4 , + + CD8 and CD20 lymphocytes during the progression of kidney fibrosis aer AKI. A ft t 3 weeks aer IRI, t ft he num- ber of inflammatory immune cells was increased in IRI kidneys, but showed no difference between non-diabetic + + + and diabetic mice. At 5 weeks aer IRI, sig ft nificantly increased infiltration of CD4 , CD8 and CD20 lympho- cytes was observed in diabetic mice compared to non-diabetic mice (Fig. 2a–f ). The mRNA expression levels of inflammatory cytokines including TNF-α , IFN-γ and CCL2 were significantly higher in IRI kidneys than sham kidneys at 3 weeks after IRI (Fig.  2g–i). The expression levels of TNF-α and CCL2 were also higher in diabetic kidneys than in non-diabetic kidneys. From 3 to 5 weeks after IRI, TNF- α and IFN-γ levels were significantly decreased in non-diabetic IRI kidneys, but not in diabetic IRI kidneys (Fig. 2g,h). Although CCL-2 levels were decreased at 5 weeks in both diabetic and non-diabetic kidneys, they were still higher in diabetic compared to non-diabetic IRI kidneys (Fig. 2i). These results indicate that diabetes augmented lymphocyte recruitment to scar kidneys aer ft IRI and maintained the activation of intra-renal inflammation and inflammatory mediators, which could play pivotal roles in the fibrotic cascade. Diabetes induced persistent activation of TGF-β and Shh signalling after IRI. First, we per- formed immunohistochemistry for TGF-β , Shh and Smo to identify their expression in kidneys after IRI (Fig. 3a–c). In diabetic kidneys, the slightly increased expression of TGF-β was shown in remaining tubules and strongly stained the cytoplasm of interstitial cells, especially at 5 weeks aer IRI (Fig ft .  3a). Expression of Shh and Smo was increased in injured tubules rather than the fibrotic area at 3 and 5 weeks aer IRI (Fig ft .  3b,c). The Shh-positive area was slightly reduced in diabetic kidney at 5 weeks because of the replacement of inflamma - tion and expansion of the extracellular matrix. However, Shh expression in remaining and atrophic tubules was increased and more prominent (Fig. 3b). Smo expression was also increased in IRI kidneys, but the increase was higher in diabetic mice at 3 and 5 weeks aer IRI (Fig ft .  3c). Next, we evaluated the protein expression of TGF-β and Shh signalling pathways in kidneys at 5 weeks ae ft r IRI (Fig. 3d–f ). TGF-β expression was not significantly increased and Smad2 expression was even decreased at 5 weeks aer IRI in n ft on-diabetic mice. However, their expression levels were significantly higher in diabetic IRI kidneys compared with sham or non-diabetic kidneys (Fig. 3d). Shh signalling also showed a pattern similar to TGF-β signalling. Shh expression was increased only in diabetic IRI kidneys at 5 weeks aer IRI. S ft mo expression was also significantly increased only in diabetic kidneys (Fig.  3e). Subsequently, we assessed the expression of α-SMA and fibronectin, markers of fibrosis and extracellular matrix accumulation respectively. α -SMA expres- sion was increased in IRI kidneys compared to sham kidneys, and their tendencies were significantly enhanced in diabetic IRI kidneys. The expression of fibronectin was significantly increased only in diabetic IRI kidneys (Fig. 3f ). Finally, we confirmed the mRNA expression of fibrosis related proteins in kidneys after IRI to demonstrate the effect of diabetes on TGF-β and Shh signalling pathways. mRNA expression of TGF-β was induced at 3 1 1 weeks after IRI but was more significantly increased in diabetic IRI kidneys. Although TGF-β mRNA levels were decreased from 3 to 5 weeks aer IRI in b ft oth groups, they were still significantly higher in diabetic than in non-diabetic IRI kidneys (Fig. 3g). The downstream factors of TGF-β signalling such as CTGF and their target collagen Ia-1 were increased at 3 weeks after IRI but were significantly higher in diabetic than in non-diabetic IRI kidneys. At 5 weeks aer IRI, t ft he mRNA expression levels of CTGF and collagen Ia-1 were slightly decreased SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 2 www.nature.com/scientificreports/ Figure 1. Diabetes impaired the improvement of tubular injury and aggravated renal fibrosis aer IRI. ( ft a) Experimental design. (b) Periodic acid–Schiff (PAS) staining was used to detect interstitial tubular injury at 3 and 5 weeks aer IRI. ( ft c) Masson’s trichrome (MT) staining was used to detect kidney fibrosis at 3 and 5 weeks aer IRI. ( ft d) Quantification of interstitial tubular injury based on a grading score at 3 and 5 weeks aer IRI. ( ft e) Quantification of renal fibrosis at 3 and 5 weeks aer IRI. V ft alues are expressed as the mean ± S.E.M. p < 0.05 # † versus sham, p < 0.05 versus non-diabetic, p < 0.05 versus values at 3 weeks. Scale bar = 80 µm. in non-diabetic IRI kidneys. However, their expression levels were still maintained or increased in diabetic IRI kidneys (Fig. 3h,i). Although mRNA expression of Shh was not significantly increased in diabetic IRI kidneys, its downstream signal, Gli-1, showed similar expression to TGF-β (Fig. 3j,k). Its expression was significantly higher in both non-diabetic and diabetic IRI kidneys at 3 weeks. Even during its decrease between 3 and 5 weeks, Gli-1 expression remained significantly higher in diabetic IRI kidneys at 5 weeks (Fig.  3k). Furthermore, the mRNA expression of Snail1, which was known to be a downstream signal of Gli-1, was strongly induced only in diabetic kidneys at 5 weeks aer IRI (Fig ft .  3l). These results showed that TGF-β and Shh signalling were persistently acti- vated in diabetic IRI kidneys until 5 weeks aer IRI, a ft nd were associated with increasing renal fibrosis. SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 3 www.nature.com/scientificreports/ Non-diabetic Diabetic Sham IRI Sham IRI # # 3w 136.0 ± 4.7 131.5 ± 6.4 504.5 ± 42.4 568.7 ± 10.2 Blood glucose level (mg/dL) # # 5w 147.3 ± 4.1 157.0 ± 7.3 515.6 ± 49.8 442.4 ± 42.6 # # 3w 26.67 ± 1.17 25.08 ± 0.64 22.27 ± 0.87 20.62 ± 0.33 Body weight (g) † # # 5w 28.55 ± 0.86 28.30 ± 0.96 24.44 ± 0.81 22.20 ± 1.25 3w 0.163 ± 0.008 0.107 ± 0.004* 0.168 ± 0.005 0.097 ± 0.006 Kidney weight (g) *#† 5w 0.140 ± 0.007 0.105 ± 0.013 0.162 ± 0.008 0.062 ± 0.005 # * 3w 0.613 ± 0.011 0.425 ± 0.014* 0.756 ± 0.028 0.470 ± 0.029 Kidney to body weight ratio (%) † # *† 5w 0.490 ± 0.011 0.380 ± 0.060 0.669 ± 0.051 0.282 ± 0.026 3w 1.073 ± 0.021 0.901 ± 0.021* 1.076 ± 0.010 0.844 ± 0.020 Kidney length (cm) *#† 5w 1.073 ± 0.015 0.902 ± 0.041* 1.052 ± 0.011 0.743 ± 0.024 * # † Table 1. Values are the mean ± S.E.M. p < 0.05 versus Sham, p < 0.05 versus non-diabetic, p < 0.05 versus values at 3 weeks. Treatment of insulin was showed no improvement of renal fibrosis after IRI. To explore whether the regulation of hyperglycaemia by insulin treatment ae ff ct the progression of renal fibrosis after IRI, we con- ducted follow-up experiments with some modic fi ation as shown Fig.  4a. IRI was induced at 8 weeks ae ft r strepto - zotocin injection and insulin was administrated from 2 weeks aer t ft he induction of diabetes until the end of the experiment. Insulin administration significantly mitigated the increase of HgA1c levels during the study period (Fig. 4b). From 5 to 8 weeks aer IRI, t ft he tubular injury was significantly decreased in all groups. Renal fibrosis was not significantly increased in non-diabetic IRI group. As expected, renal fibrosis in diabetic IRI group was significantly increased at the same time period. However, treatment of insulin to diabetic mice did not ae ff ct the tubular injury and renal fibrosis induced by IRI (Fig.  4c–f ). Regardless of the treatment of insulin, the protein expressions of TGF-β and Shh were significantly higher in diabetic IRI groups, not mitigated by insulin. Both TGF-β and Shh expression were significantly activated in IRI kidneys as compared with those of sham. Shh expression was significantly higher at 5 weeks and TGF-β expression was significantly higher at 8 weeks in dia- betic IRI group as compared with those of non-diabetic IRI group (Fig. 4g,h). Hyperglycaemia augmented the effect of Shh and TGF-β on profibrogenic phenotype change in renal tubular cells. To demonstrate the effect of hyperglycaemia on TGF-β and Shh signalling, HKC-8 was cultured under hyperglycaemic conditions. Compared to normoglycaemic conditions (5 mM D-Glucose, NG), the hyperglycaemic state (30 mM D-Glucose, HG) induced TGF-β and Shh expression at 6 hours. Although Shh expression was significantly reduced at 24 hours, TGF-β was maintained until 24 hours (Fig. 5a–c). Smo expression was significantly increased in HG at 18 and 24 hours (Fig. 5d). However, Gli-1 was induced at 6 hours and returned to its basal level at 18 hours (Fig. 5e). Fibronectin and α-SMA expression were induced at 6 hours and were maintained through 24 hours (Fig. 5f,g). These results suggest that hyperglycaemia activates the profi- brotic signalling pathway through the activation of Shh as well as TGF-β . Next, we exposed recombinant human Shh or TGF-β to NG or HG for 24 hours. Shh treatment increased the expression of TGF-β in both NG and HG (Fig. 6a). Expression levels of α-SMA and fibronectin were increased in both NG and HG by Shh treatment aer ft 24 hours. However, the induction of α-SMA and b fi ronectin by Shh was significantly greater in HG than in NG (Fig.  6a,b). In the case of TGF-β treatment, Shh expression was increased in both NG and HG (Fig. 6c). Fibronectin expression was increased in both NG and HG at 24 hours aer T ft GF-β treatment. However, there was no difference between NG and HG (Fig.  6c). α-SMA expression of was increased by TGF-β after 24 hours in HG, but not in NG (Fig. 6d). These results indicate that hyperglycaemia increased the expression of Shh or TGF-β in terms of prob fi rogenic phenotype change in HKC-8 cells. Although Shh and TGF-β interacted with each other, hyperglycaemia did not ae ff cted the crosstalk between Shh and TGF-β . 1 1 Insulin increased the expressions of Shh and TGF-β without affecting the interaction between Shh and TGF-β1 in hyperglycaemic conditions. To investigate the mechanisms by which insulin has no effect on renal fibrosis aer IRI, HK ft C-8 was cultured in LG and HG with insulin stimulation. The expressions of TGF-β and Shh were increased by insulin stimulation as well as in HG. In HG, the expression of Shh was not enhanced by insulin, but the expression of TGF-β was significantly increased by insulin (Fig.  7a,b). The increased mutual stimulation between Shh and TGF-β under hyperglycaemic condition was not altered by insu- lin (Fig. 7c,d). Insulin also did not mitigate the activated expression of α-SMA by Shh or TGF-β under the HG, but significantly increased the Shh-induced expression of α-SMA in HG (Fig.  7c). Discussion Among diverse causes of chronic kidney disease, diabetes accounts for the largest portion of kidney failure, account- ing for approximately 50% of cases in the developed world . High blood pressure, poor glycaemic control and albu- minuria are well-known risk factors for the development or progression of diabetic nephropathy. However, these factors cannot explain all of the interindividual variability in the rate of progression to end-stage of CKD . Recently, AKI episodes have been known to be associated with a cumulative risk for developing advanced CKD in diabetes mellitus, independent of other major risk factors of progression . Additional loss of parenchyma caused by failed SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 4 www.nature.com/scientificreports/ Figure 2. Diabetes aggravated inflammation aer IRI. ( ft a–c) The infiltrated (a) CD4-positive, (b) CD8-positive T cells and (c) CD20-positive B cells measured by immunohistochemistry. (d–f) Quantification of the number of infiltrated (d) CD4-positive, (e) CD8-positive T cells and (f) CD20-positive B cells in whole kidneys. (g–i) mRNA expression of inflammatory cytokine markers, (g) TNF-α, (h) IFN-γ, and (i) CCL-2, determined * # via qRT-PCR. Values are expressed as the mean ± S.E.M. p < 0.05 versus sham, p 0.05 versus non-diabetic, p < 0.05 versus at 3 weeks. Scale bar = 50 µm. repair of AKI has been implicated in the development and progression of experimental nephropathy in diabetic rodents using an acute-on-chronic injury model . However, the underlying mechanism of the progression of CKD aer AKI un ft der diabetic conditions is not currently well known. Here, we demonstrated why diabetes is particularly prone to progression to CKD aer AKI t ft hrough three major findings in our study. SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 5 www.nature.com/scientificreports/ Figure 3. Diabetes induced persistent activation of TGF-β and Shh signalling. (a–c) Immunohistochemistry for kidney levels of (a) TGF-β , (b) Shh and (c) Smo at 3 and 5 weeks aer IRI. S ft cale bar = 80 µm. (d) The kidney levels of expressed TGF-β and Smad2 were measured by western blot at 5 weeks aer IRI. ( ft e) The kidney levels of expressed Shh and Smo were measured by western blot at 5 weeks aer IRI. ( ft f) The kidney levels of expressed α-SMA and fibronectin were measured by western blot at 5 weeks aer IRI. E ft ach protein expression was normalized by GAPDH. The fold-change of each protein was calculated as the ratio of averages versus non- diabetic sham. (g–l) mRNA expression of the major TGF-β and Shh signalling pathway factors in the kidney were determined via qRT-PCR at 3 and 5 weeks aer IRI. Th ft e targets of mRNA were (g) TGF-β , (h) CTGF, (i) collagen I-a1, (j) Shh, (k) Gli-1 and (l) Snail1. Each value of the target mRNA was normalized by 18 S. Values are * # † expressed as the mean ± S.E.M. p < 0.05 versus sham, p < 0.05 versus non-diabetic p < 0.05 versus at 3 weeks. First, diabetes exacerbated abnormal lymphocyte infiltration aer IRI. I ft n this study, the recruitment of CD4 , + + CD8 and CD20 lymphocytes in the kidney was increased at 3 weeks aer IRI. H ft owever, there was no difference between non-diabetic and diabetic kidneys at 3 weeks ae ft r IRI. At 5 weeks ae ft r IRI, levels of infiltrating CD4 , + + CD8 and CD20 lymphocytes were sharply increased, but were significantly higher in diabetic IRI kidneys than in non-diabetic IRI kidneys. Aberrant recruitment of T and B lymphocytes is one of the major findings in diabetic 27,28 nephropathy . At the early phase of AKI, the major infiltrating immune cells in the kidney are neutrophils and SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 6 www.nature.com/scientificreports/ Figure 4. Treatment of insulin was showed no improvement of renal fibrosis aer IRI. ( ft a) Experimental design. (b) The value of HbA1c (c) Periodic acid–Schiff (PAS) staining was used to detect kidney tubular injuries and to view necrosis, atrophy and normal tubules at 5 and 8 weeks aer IRI. ( ft d) Masson’s trichrome (MT) staining was used to detect kidney fibrosis at 5 and 8 weeks aer IRI. ( ft e) The quantification of total interstitial tubular injury based on a grading score at 5 and 8 weeks aer IRI. ( ft f) The quantification of renal fibrosis at 5 and 8 weeks aer ft IRI. (g) The expression of TGF-b1 and Shh in the kidney at 5 weeks aer IRI. ( ft h) The expression of TGF-b1 and Shh in the kidney at 8 weeks aer IRI. V ft alues are expressed as the mean ± S.E.M. p < 0.05 versus sham, # † ‡ p < 0.05 versus non-diabetic, p < 0.05 versus at 5 weeks, p < 0.05 versus between diabetic and diabetic insulin group. Scale bar = 80 µm. macrophages. In contrast, the recruitment of lymphocytes into the injured kidney was reported to be prominent 29,30 during the recovery phase . e Th exact role of these lately infiltrating lymphocytes has not been studied exten- sively. However, some studies suggested that B cell infiltration aer AKI mig ft ht hinder normal wound healing . Interestingly, recent studies also suggested that infiltrated lymphocytes directly ae ff cted renal fibrosis by increas- 32,33 ing levels of TGF-β . e Th major finding of this study was that diabetes induced persistent activation of TGF- β and Shh signalling. In this study, persistent activation of the TGF-β and Shh signalling pathways was maintained only in the dia- betic group at 5 weeks after IRI. Furthermore, the downstream signalling molecules Smo and Gli-1 were also persistently enhanced in diabetic IRI kidneys. These results showed that TGF-β and Shh signalling were per- sistently activated in diabetic IRI kidneys until 5 weeks aer IRI a ft s well as were associated with increasing renal b fi rosis. e Th TGF-β signalling pathway reduces tissue injury and promotes tubule restoration through epithelial 34,35 de-differentiation at the early stage of acute injury . However, under maladaptive wound healing conditions, SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 7 www.nature.com/scientificreports/ Figure 5. Hyperglycaemia induced EMT via activation of TGF-β and Shh signalling. HKC-8 cells were cultured in 30 mM D-glucose to induce hyperglycaemic conditions in the cells. (a) The expressions of TGF-β , Shh, Smo, Gli-1, fibronectin and α-SMA were measured using western blotting in a time-dependent manner. e f Th old-changes of (b) TGF-β , (c) Shh, (d) Smo, (e) Gli-1, (f) fibronectin and (g) α-SMA were calculated as the ratios of averages versus a 5 mM D-glucose control. Expression of each protein was normalized by GAPDH. Values are expressed as the mean ± S.E.M. p < 0.05 versus 5 mM D-glucose control. Figure 6. Hyperglycaemia augmented the effect of Shh and TGF-β on prob fi rogenic phenotype change in renal tubular cells. HKC-8 cells were cultured in 5 mM D-glucose or 30 mM D-glucose with 10 ng/ml Shh or 10 ng/ ml TGF-β . (a) Expression levels of TGF-β and fibronectin at 24 hours aer S ft hh treatment. (b) Representative 1 1 confocal fluorescence images of α-SMA were examined at 24 hours aer S ft hh treatment. (c) Expression levels of Shh and fibronectin at 24 hours aer T ft GF-β treatment. (d) Representative confocal fluorescence images of α-SMA were examined at 24 hours aer T ft GF-β treatment. Expression of each protein was normalized by GAPDH. The fold-change of each protein was calculated as the ratio of averages versus a 5 mM D-glucose control. Values are expressed as the mean ± S.E.M. p < 0.05 versus 5 mM D-glucose or 30 mM D-glucose controls, p < 0.05 versus 5 mM D-glucose with treatment. TGF-β signalling could promote abnormal organ fibrosis through myob fi roblast differentiation from fibroblasts 36,37 or bone marrow-derived precursor cells . The Shh signalling pathway could also modulate the activation of 38–40 myob fi roblast and kidney b fi rosis aer ft tubular injury . Shh was secreted from the damaged renal tubular epi- 41,42 thelium and induced fibroblasts to differentiate into myob fi roblasts . Furthermore, the overexpression of Shh was reported to increase extracellular matrix (ECM) synthesis . Contrary to the TGF-β signalling pathway, the effect of diabetes on Shh signalling pathway has not yet been clarified. However, the results of this study indicate that sustained tissue injury, activation of inflammation as well as hyperglycaemia itself could be suggested as mechanisms underlying the activation of Shh signalling during CKD progression aer di ft abetic AKI. We also demonstrated using in vitro experiments that hyperglycaemia induced the activation of Shh signalling as well as TGF-β and augmented the effect of both Shh and TGF-β on prob fi rogenic phenotype change in renal 1 1 tubular cells. Although cell culture models in diabetic kidney injuries such as our study, have several limitations in mimicking in vivo tissue environment, these results could be explained by the finding that hyperglycaemia increased the activation of Shh and TGF-β . Recent studies indicated that Shh signalling could be activated by TGF-β . Furthermore, TGF-β and its receptor were also reported to be increased by the activation of Shh sig- 1 1 nalling . The crosstalk between Shh and TGF-β was also recently reported to enhance kidney fibrosis in several SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 8 www.nature.com/scientificreports/ Figure 7. Insulin increased the expressions of Shh and TGF-β without ae ff cting the interaction between Shh and TGF-β in hyperglycaemic condition. (a) Expression level of TGF-β at 24 hours aer in ft sulin treatment. 1 1 (b) Expression level of Shh at 24 hours aer in ft sulin treatment. (c) Expression levels of TGF-β and α-SMA at 24 hours aer in ft sulin and Shh treatment. (d) Expression levels of Shh and α-SMA at 24 hours aer in ft sulin and TGF-β treatment. Expression of each protein was normalized by GAPDH. The fold-change of each protein was calculated as the ratio of averages versus a 5 mM D-glucose control. Values are expressed as the mean ± S.E.M. * # † p < 0.05 versus 5 mM D-glucose, p < 0.05 versus 30 mM D-glucose. p < 0.05 versus between 5 mM D-glucose with insulin treatment and 30 mM D-glucose with insulin treatment. p < 0.05 versus 30 mM D-glucose with Shh or TGF-β treatment. 44–46 models of fibrotic kidney diseases . Our results also showed the mutual influence of Shh and TGF-β However, hyperglycaemia did not have a significant effect on the interaction between Shh and TGF-β Since insulin treatment is one way to control the hyperglycemia, which may be the target to rescue the kidney b fi rosis following IRI, insulin treatment could be expected to reduce the expression of TGF-β and Shh signaling and finally ameliorate the progression of renal fibrosis aer IRI. ft To answer this question, we conducted the additional experiment with some modifications as in Fig. 4 . As expected, longer-lasting diabetes before IRI induction and hyperglycemia extended to 8 weeks significantly aggra- vated the post-ischaemic fibrosis. However, treatment of insulin did not improved renal fibrosis in our animal model showing progressive kidney fibrosis aer IRI, e ft ven though insulin showed significantly improved control of blood glucose level. Our results also showed that treatment of insulin had no effect on the expression of TGF-β and Shh at both 5 and 8 weeks aer IRI. N ft egative result with insulin treatment could be anticipated, because the correction of hyperglycemia might be suboptimal in our experiment (mean HgA1c 6.5%) or ischaemic kidney damage in diabetic group was too severe to overcome with the blood glucose control. However, we had noted the results in the previous reports, which showed that insulin induced the expression of TGF-β in renal proximal 47 48 tubular cells , and insulin stimulated SLT2-mediated tubular glucose absorption via oxidative stress generation . In our in vitro experiment, we also confirmed that insulin itself could induce the expression of TGF-β and also ae ff ct the expression of Shh in HKC-8 cells. However, insulin was not altered the mutual interaction between Shh and TGF-β under high glucose condition. Based on these results, control of hyperglycemia by treatment of SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 9 www.nature.com/scientificreports/ insulin may not be enough for preventing renal fibrosis aer IRI, a ft nd further studies on the effect of other hypo- glycemic agents with different action mechanism on tubular cells, are necessary. In conclusion, we demonstrated that diabetes promoted CKD progression following AKI through recruitment of lymphocytes and persistent activation of TGF-β and Shh signalling pathways. The control of hyperglycaemia with insulin administration was not enough for preventing the progression of renal fibrosis. Future study of the mechanism of renal fibrosis progression aer di ft abetic AKI will contribute to solving the unresolved mechanisms underlying kidney fibrosis and to locating potential therapeutic targets for treatment of diabetic kidney failure. Materials and Methods Animals and experimental design. Effect of hyperglycaemia on the progression of renal fibrosis after IRI. Diabetes was induced by streptozotocin (Sigma-Aldrich, St. Louis, MO, USA) injected once a day at a concentration of 50 mg/kg for five consecutive days into 8-week-old male C57BL/6 mice. After 2 weeks, a uni- lateral renal ischaemia–reperfusion injury (IRI) model was established in non-diabetic mice and diabetic mice with blood glucose levels above 250 mg/dL. We induced ischaemia in the left kidneys of the mice using a clamp to obstruct blood circulation for 25 minutes. During the ischaemic period, mice were maintained within a body temperature range of 36–37 °C using a heating pad and a rectal temperature probe. Mice were sacrificed at 3 or 5 weeks after IRI and divided into four groups: (1) non-diabetic controls (Nor Sham), (2) non-diabetic IRI (Nor & IRI), (3) diabetic controls (DM sham), and (4) diabetic IRI (DM & IRI). Mice were monitored for body weights and blood glucose levels every 2 weeks and HbA1c levels at 2 and 6 weeks. All animal experiments were performed in compliance with the guidelines of the Animal Research Ethics Committee of Kyung Hee University and Institutional Animal Care and Use committee Kyung Hee University Hospital at Gangdong, Seoul, Korea (approval number: KHNMC AP 2015-001). Ee ff ct of insulin treatment on the progression of post-ischaemic renal fibrosis. At 2 weeks aer di ft abetic induction, Insulin (10 unit/kg, insulin glargine, LANTUS by Sanofi-Aventis was administrated once a day until the end of the experiment. At 8 weeks aer di ft abetic induction, the unilateral IRI was established in non-diabetic, diabetic and insulin-treated diabetic mice. Mice were sacrificed at 5 and 8 weeks after IRI and divided into six groups: (1) non-diabetic controls (Nor Sham), (2) non-diabetic IRI (Nor & IRI), (3) diabetic controls (DM Sham), (4) diabetic IRI (DM & IRI), (5) insulin-treated diabetic controls (Ins Sham) and (6) insulin-treated diabetic IRI (Ins & IRI). Blood glucose level was checked every 2 weeks and HbA1c was checked at 2, 6, 9 and 12 weeks aer ft diabetic induction. All animal experiments were performed in compliance with the guidelines of the Animal Research Ethics Committee of Kyung Hee University and Institutional Animal Care and Use committee Kyung Hee University Hospital at Gangdong, Seoul, Korea (approval number: KHNMC AP 2016-007). Histology. Sections were cut at 4 µ m thickness. For the histological assessment of tubulointerstitial injury (tubular necrosis, tubular atrophy, and glomerular cysts), the sections were stained with periodic acid-Schiff reagent. Ten corticomedullary fields were examined in each section at 200x magnification, and a semiquanti- tative analysis of tubulointerstitial injury was performed. Total tubular injury was graded on a scale of 0 to 5 based on the percentage of normal tubules and the amount of tubular necrosis and tubular atrophy as follows: 0, absent; 1, 1–25%; 2, 26–50%; 3, 51–75%; 4, 76–99%; and 5, 100%. Fibrosis was quantified using Masson’s tri- chrome staining and computer-assisted image analysis. For immunohistochemistry, we used the Bond Polymer Refine Detection system (Vision BioSystems, Hingham, MA, USA) with antibodies against CD4, CD8 and CD20. Four-micron-thick kidney sections were deparaffinized using Bond Dewax solution, and an antigen retrieval pro- cedure was then performed using Bond ER solution for 30 minutes at 100 °C. Endogenous peroxidase activity was terminated by incubating the tissues with hydrogen peroxide for 5 minutes. The sections were then incubated with CD4 (1:100, Abcam, Cambridge, MA, USA), CD8, and CD20 (1:100; Abcam, Cambridge, MA, USA) antibod- ies using a biotin-free polymeric horseradish peroxidase-linked antibody conjugate system (Vision BioSystems, + + + Hingham, MA, USA). To assess T and B lymphocyte infiltration, CD4 /CD8 /CD20 cells were counted in 20 randomly selected fields from each section at 400x magnification under a light microscope. In vitro cell culture. e h Th uman renal proximal tubular epithelial cell line HKC8 was obtained from Dr. L. Rausen (Johns Hopkins University, Baltimore, MD, USA) and was maintained in Dulbecco’s Modified Eagle’s Medium supplemented with Ham’s F12 medium (DMEM/F12; Invitrogen, Carlsbad, CA, USA) DMEM/F12 was supplemented with 5% foetal bovine serum and 1% penicillin/streptomycin (WelGENE, Daegu, Korea). Before experiments, cells were maintained in a medium containing 5 mM D-glucose. 3 × 10 cells were seeded in a 60 mm dish with 5 mM D-glucose and cultured for 24 hours . To confirm the ee ff ct of hyperglycaemia on the prob fi ro - genic signalling pathway in a time-dependent manner, we changed the media containing 30 mM D-glucose and collected cells at 3, 6, 18 and 24 hours aer t ft reatment. To demonstrate the effects of TGF-β and Shh on HKC-8 cells, media were changed following these conditions: (1) 5 mM D-glucose, (2) 5 mM D-glucose + TGF-β 10 ng/ ml, (3) 5 mM D-glucose + Shh 10 ng/ml, (4) 30 mM D-glucose, (5) 30 mM D-glucose + TGF-β 10 ng/ml, and (6) 30 mM D-glucose + Shh 10 ng/ml. To demonstrate the effect of insulin on the expressions of TGF-β and Shh in HKC-8 cells, media were changed following these conditions: (1) 5 mM D-glucose, (2) 5 mM D-glucose + insu- lin 50 ng/ml, (3) 5 mM D-glucose + insulin 5000 ng/ml, (4) 30 mM D-glucose, (5) 30 mM D-glucose + insulin 50 ng/ml, and (6) 30 mM D-glucose + insulin 5000 ng/ml. To identify the effect of insulin on the sensitivity to TGF-β and Shh, media were changed following these conditions: (1) 5 mM D-glucose, (2) 30 mM D-glucose, (3) 30 mM D-glucose + TGF-β 10 ng/ml, (4) 30 mM D-glucose + TGF-β 10 ng/ml + insulin 50 ng/ml, (5) 30 mM 1 1 D-glucose + Shh 10 ng/ml, and (6) 30 mM D-glucose + Shh 10 ng/ml + insulin 50 ng/ml. After 24 hours, cells were collected for protein analysis. SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 10 www.nature.com/scientificreports/ Isolation of total RNA and real-time PCR. Total RNA was extracted from kidney tissue by using the Nucleospin RNA kit (Macherey-Nagel, Düren, W. Germany) according to the manufacturer’s instructions and was quantified using a NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific Inc., San Jose, CA, USA). Complementary DNA was synthesized using random primers (Promega, Madison, WI, USA), DNTP mixture (TaKaRa Bio Inc., Otsu-shi, Shiga, Japan) and M-MLV reverse transcriptase (Mbiotech Inc., Hanam, Korea). Real-time PCR was performed in reactions with a final volume of 20 μl containing 1 μl of cDNA, 10 pmol of each sense and antisense primer, and 17 μl of Power SYBR Green PCR Master mix (Applied Biosystems, Beverly, MA, USA) and was detected using the Applied Biosystems StepOnePlus Real-Time PCR System. The primers for ® ™ TGF-β , CTGF, collagen IV-a1, collagen I-a1, Snail1, Shh, Gli-1, IFN-γ, TNF-α, CCl-2 and 18 S were purchased from Mbiotech. Each sample was run in duplicate in separate tubes to quantify target gene expression, and the results were normalized to 18 S expression. Western blot analysis. Cells and kidney tissues were washed with PBS and lysed in the M-PER mam- malian protein extraction reagent with protease inhibitor cocktail (Thermo Fisher Scientific Inc., San Jose, CA, USA). Proteins were separated with 8–15% SDS-PAGE and then were transferred onto a nitrocellulose membrane (Millipore, Madrid, Spain) by electroblotting. The membrane was blocked for 1 hour at room temperature and then was incubated overnight at 4 °C with anti-Shh, anti-E-cadherin, anti-Smad2, anti-Smo, anti-Gli-1 (1:1000, Santa Cruz biotechnology, Santa Cruz, CA, USA), anti-fibronectin (R&D system Inc. Minneapolis, MN, USA), anti-Bax, anti-Bcl-2, anti-TGF-β (1:1000, Cell Signaling Technology, Beverly, MA, USA), and α-SMA (1:1000 Abcam Inc. Cambridge, MA, USA) primary antibodies. Subsequently, the membranes were stained with horse- radish peroxidase-conjugated goat anti-rabbit or mouse immunoglobulin G (1:2,000, Santa Cruz biotechnology, Santa Cruz, CA, USA). The immunoreactive bands were detected by chemiluminescence (enhanced chemilumi- nescence; BioFX Laboratories Inc., Owings Mills, Maryland, USA). GAPDH (1:2,000, Santa Cruz biotechnology, Santa Cruz, CA, USA) was used as an internal control. Confocal microscopy. Cells were fixed with 4% paraformaldehyde, permeabilized with 0.2% Triton X-100 solution, blocked with bovine serum albumin (BSA), and incubated with primary antibodies for 2 hours. Aer ft washing with PBS, the samples were re-incubated with secondary antibodies conjugated with Alexa Fluor 488 or Texas Red (Life Technologies, Gaithersburg, MD, USA) for 1 hour. Cells were counterstained with DAPI to delineate the nuclei, and the sections were examined using confocal microscopy (LSM-700; Carl Zeiss, Jena, ur Th ingia, Germany). Statistical analyses. All values are expressed as the mean ± SE. The results were analysed using the Kruskal-Wallis nonparametric test for multiple comparisons. Significant differences detected using the Kruskal-Wallis test was confirmed using the Wilcoxon rank sum and Mann-Whitney tests (to compare mean differences). p-values < 0.05 were considered to indicate statistical significance. References 1. Ferenbach, D. A. & Bonventre, J. V. Mechanisms of maladaptive repair aer AKI le ft ading to accelerated kidney ageing and CKD. Nat Rev Nephrol 11, 264–276, https://doi.org/10.1038/nrneph.2015.3 (2015). 2. McCullough, P. A. Contrast-induced acute kidney injury. J Am Coll Cardiol 51, 1419–1428, https://doi.org/10.1016/j. jacc.2007.12.035 (2008). 3. Huang, P. H. et al. Pioglitazone ameliorates endothelial dysfunction and restores ischemia-induced angiogenesis in diabetic mice. Biomed Pharmacother 62, 46–52, https://doi.org/10.1016/j.biopha.2007.06.014 (2008). 4. Falanga, V. Wound healing and its impairment in the diabetic foot. Lancet 366, 1736–1743, https://doi.org/10.1016/S0140- 6736(05)67700-8 (2005). 5. Rosner, M. H. & Okusa, M. D. Acute kidney injury associated with cardiac surgery. Clinical journal of the American Society of Nephrology 1, 19–32 (2006). 6. Weisberg, L. S., Kurnik, P. B. & Kurnik, B. R. Risk of radiocontrast nephropathy in patients with and without diabetes mellitus. Kidney international 45, 259–265 (1994). 7. Venkatachalam, M. A., Weinberg, J. M., Kriz, W. & Bidani, A. K. Failed Tubule Recovery, AKI-CKD Transition, and Kidney Disease Progression. J Am Soc Nephrol 26, 1765–1776, https://doi.org/10.1681/ASN.2015010006 (2015). 8. Zhou, D. et al. Sonic hedgehog is a novel tubule-derived growth factor for interstitial fibroblasts aer k ft idney injury. Journal of the American Society of Nephrology, ASN. 2013080893 (2014). 9. Pohlers, D. et al. TGF-beta and fibrosis in different organs - molecular pathway imprints. Biochim Biophys Acta 1792, 746–756, https://doi.org/10.1016/j.bbadis.2009.06.004 (2009). 10. Asai, J. et al. Topical sonic hedgehog gene therapy accelerates wound healing in diabetes by enhancing endothelial progenitor cell- mediated microvascular remodeling. Circulation 113, 2413–2424, https://doi.org/10.1161/CIRCULATIONAHA.105.603167 (2006). 11. Bielefeld, K. A., Amini-Nik, S. & Alman, B. A. Cutaneous wound healing: recruiting developmental pathways for regeneration. Cell Mol Life Sci 70, 2059–2081, https://doi.org/10.1007/s00018-012-1152-9 (2013). 12. LeBleu, V. S. et al. Origin and function of myofibroblasts in kidney fibrosis. Nat Med 19, 1047–1053, https://doi.org/10.1038/ nm.3218 (2013). 13. Liu, Y. Cellular and molecular mechanisms of renal fibrosis. Nat Rev Nephrol 7, 684–696, https://doi.org/10.1038/nrneph.2011.149 (2011). 14. Fernandez, I. E. & Eickelberg, O. New cellular and molecular mechanisms of lung injury and fibrosis in idiopathic pulmonary b fi rosis. Lancet 380, 680–688, https://doi.org/10.1016/S0140-6736(12)61144-1 (2012). 15. Zhou, D. et al. Sonic hedgehog is a novel tubule-derived growth factor for interstitial fibroblasts aer k ft idney injury. J Am Soc Nephrol 25, 2187–2200, https://doi.org/10.1681/ASN.2013080893 (2014). 16. Spurgeon, K. R., Donohoe, D. L. & Basile, D. P. Transforming growth factor-beta in acute renal failure: receptor expression, effects on proliferation, cellularity, and vascularization aer r ft ecovery from injury. Am J Physiol Renal Physiol 288, F568–577, https://doi. org/10.1152/ajprenal.00330.2004 (2005). 17. Connolly, S. B., Sadlier, D., Kieran, N. E., Doran, P. & Brady, H. R. Transcriptome profiling and the pathogenesis of diabetic complications. J Am Soc Nephrol 14, S279–283 (2003). SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 11 www.nature.com/scientificreports/ 18. Reeves, W. B. & Andreoli, T. E. Transforming growth factor beta contributes to progressive diabetic nephropathy. Proc Natl Acad Sci USA 97, 7667–7669 (2000). 19. Brem, H. & Tomic-Canic, M. Cellular and molecular basis of wound healing in diabetes. J Clin Invest 117, 1219–1222, https://doi. org/10.1172/JCI32169 (2007). 20. Brosius, F. C. III. New insights into the mechanisms of fibrosis and sclerosis in diabetic nephropathy. Rev Endocr Metab Disord 9, 245–254, https://doi.org/10.1007/s11154-008-9100-6 (2008). 21. Le Clef, N., Verhulst, A., D’Haese, P. C. & Vervaet, B. A. Unilateral renal ischemia-reperfusion as a robust model for acute to chronic kidney injury in mice. PloS one 11, e0152153 (2016). 22. Jang, H. S., Kim, J. I., Han, S. J. & Park, K. M. Recruitment and subsequent proliferation of bone marrow-derived cells in the postischemic kidney are important to the progression of fibrosis. Am J Physiol Renal Physiol 306, F1451–1461, https://doi. org/10.1152/ajprenal.00017.2014 (2014). 23. Moon, J. Y. et al. Aberrant recruitment and activation of T cells in diabetic nephropathy. Am J Nephrol 35, 164–174, https://doi. org/10.1159/000334928 (2012). 24. Jha, V. et al. Chronic kidney disease: global dimension and perspectives. Lancet 382, 260–272, https://doi.org/10.1016/S0140- 6736(13)60687-X (2013). 25. Ritz, E. & Orth, S. R. Nephropathy in patients with type 2 diabetes mellitus. N Engl J Med 341, 1127–1133, https://doi.org/10.1056/ NEJM199910073411506 (1999). 26. Thakar, C. V., Christianson, A., Himmelfarb, J. & Leonard, A. C. Acute kidney injury episodes and chronic kidney disease risk in diabetes mellitus. Clin J Am Soc Nephrol 6, 2567–2572, https://doi.org/10.2215/CJN.01120211 (2011). 27. Galkina, E. & Ley, K. Leukocyte recruitment and vascular injury in diabetic nephropathy. J Am Soc Nephrol 17, 368–377, https://doi. org/10.1681/ASN.2005080859 (2006). 28. Hills, C. E. & Squires, P. E. TGF-beta1-induced epithelial-to-mesenchymal transition and therapeutic intervention in diabetic nephropathy. Am J Nephrol 31, 68–74, https://doi.org/10.1159/000256659 (2010). 29. Devarajan, P. Update on mechanisms of ischemic acute kidney injury. J Am Soc Nephrol 17, 1503–1520, https://doi.org/10.1681/ ASN.2006010017 (2006). 30. Kinsey, G. R., Li, L. & Okusa, M. D. Inflammation in acute kidney injury. Nephron Exp Nephrol 109, e102–107, https://doi. org/10.1159/000142934 (2008). 31. Jang, H. R. et al. B cells limit repair after ischemic acute kidney injury. J Am Soc Nephrol 21, 654–665, https://doi.org/10.1681/ ASN.2009020182 (2010). 32. Tapmeier, T. T. et al. Pivotal role of CD4+ T cells in renal fibrosis following ureteric obstruction. Kidney Int 78, 351–362, https://doi. org/10.1038/ki.2010.177 (2010). 33. Hui Han, J. Z. et al. and Ruiyan Zhang. Renal recruitment of B lymphocytes exacerbates tubulointerstitial fibrosis by promoting monocyte mobilization and infiltration aer uni ft lateral ureteral obstruction. Journal of Pathology 241, 80–90 (2017). 34. Chen, H., Li, D., Saldeen, T. & Mehta, J. L. TGF-beta 1 attenuates myocardial ischemia-reperfusion injury via inhibition of upregulation of MMP-1. Am J Physiol Heart Circ Physiol 284, H1612–1617, https://doi.org/10.1152/ajpheart.00992.2002 (2003). 35. Ishibe, S. & Cantley, L. G. Epithelial–mesenchymal–epithelial cycling in kidney repair. Current opinion in nephrology and hypertension 17, 379–385 (2008). 36. Fan, J. M. et al. Transforming growth factor-beta regulates tubular epithelial-myob fi roblast transdie ff rentiation in vitro . Kidney Int 56, 1455–1467, https://doi.org/10.1046/j.1523-1755.1999.00656.x (1999). 37. Meran, S. & Steadman, R. Fibroblasts and myofibroblasts in renal fibrosis. Int J Exp Pathol 92, 158–167, https://doi.org/10.1111/ j.1365-2613.2011.00764.x (2011). 38. Amankulor, N. M. et al. Sonic hedgehog pathway activation is induced by acute brain injury and regulated by injury-related inflammation. J Neurosci 29, 10299–10308, https://doi.org/10.1523/JNEUROSCI.2500-09.2009 (2009). 39. Stewart, G. A. et al. Expression of the developmental Sonic hedgehog (Shh) signalling pathway is up-regulated in chronic lung fibrosis and the Shh receptor patched 1 is present in circulating T lymphocytes. J Pathol 199, 488–495, https://doi.org/10.1002/ path.1295 (2003). 40. Horn, A. et al. Hedgehog signaling controls fibroblast activation and tissue fibrosis in systemic sclerosis. Arthritis & Rheumatism 64, 2724–2733 (2012). 41. Ding, H. et al. Sonic hedgehog signaling mediates epithelial-mesenchymal communication and promotes renal fibrosis. J Am Soc Nephrol 23, 801–813, https://doi.org/10.1681/ASN.2011060614 (2012). 42. Fabian, S. L. et al. Hedgehog-Gli pathway activation during kidney fibrosis. Am J Pathol 180, 1441–1453, https://doi.org/10.1016/j. ajpath.2011.12.039 (2012). 43. Bai, Y. et al. Sonic hedgehog-mediated epithelial-mesenchymal transition in renal tubulointerstitial fibrosis. International journal of molecular medicine 37, 1317–1327 (2016). 44. Chung, Y. & Fu, E. Crosstalk between Shh and TGF-beta signaling in cyclosporine-enhanced cell proliferation in human gingival b fi roblasts. PLoS One 8, e70128, https://doi.org/10.1371/journal.pone.0070128 (2013). 45. Kramann, R., DiRocco, D. P. & Humphreys, B. D. Understanding the origin, activation and regulation of matrix-producing myob fi roblasts for treatment of fibrotic disease. J Pathol 231, 273–289, https://doi.org/10.1002/path.4253 (2013). 46. Lu, H., Chen, B., Hong, W., Liang, Y. & Bai, Y. Transforming growth factor-beta1 stimulates hedgehog signaling to promote epithelial-mesenchymal transition aer k ft idney injury. FEBS J 283, 3771–3790, https://doi.org/10.1111/febs.13842 (2016). 47. Morrisey, K., Evans, R. A., Wakefield, L. & Phillips, A. O. Translational regulation of renal proximal tubular epithelial cell transforming growth factor-β1 generation by insulin. e A Th merican journal of pathology 159, 1905–1915 (2001). 48. Nakamura, N., Matsui, T., Ishibashi, Y. & Yamagishi, S.-i Insulin stimulates SGLT2-mediated tubular glucose absorption via oxidative stress generation. Diabetology & metabolic syndrome 7, 48 (2015). 49. Lee, S. Y. et al. PGC1alpha Activators Mitigate Diabetic Tubulopathy by Improving Mitochondrial Dynamics and Quality Control. J Diabetes Res 2017, 6483572, https://doi.org/10.1155/2017/6483572 (2017). Acknowledgements This research was supported by a grant from the National Research Foundation of Korea in 2012 (NRF- 2012M3A9C6050511). Author Contributions D.J.K. designed and performed all of the experiments. J.M.K. designed in vitro cell culture and performed western blot analysis of in vivo and in vitro experiments. S.H.P. and H.M. performed all of the histological analysis. H.K.K. and S.J.S. performed confocal microscopy. S.M.K. and J.W.S. performed diabetes experiments. Y.H.L., Y.G.K. and J.Y.M. performed IRI model. All authors participated in interpreting the results. D.J.K., Y.G.K., S.Y.L. and S.H.L. wrote the manuscript. S.H.L. and Y.S. coordinated the project. SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 12 www.nature.com/scientificreports/ Additional Information Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-017-16977-z. Competing Interests: The authors declare that they have no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2017 SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 13 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Scientific Reports Springer Journals

Diabetes Aggravates Post-ischaemic Renal Fibrosis through Persistent Activation of TGF-β1 and Shh Signalling

Loading next page...
 
/lp/springer_journal/diabetes-aggravates-post-ischaemic-renal-fibrosis-through-persistent-ZBtmTuZuq8

References (55)

Publisher
Springer Journals
Copyright
Copyright © 2017 by The Author(s)
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2045-2322
DOI
10.1038/s41598-017-16977-z
Publisher site
See Article on Publisher Site

Abstract

www.nature.com/scientificreports OPEN Diabetes Aggravates Post- ischaemic Renal Fibrosis through Persistent Activation of TGF-β and Received: 25 May 2017 Shh Signalling Accepted: 20 November 2017 Published: xx xx xxxx 2 3 1 1 1 Dong-Jin Kim , Jun Mo Kang , Seon Hwa Park , Hyuk-Kwon Kwon , Seok-Jong Song , 1 1 1 1 1 1 Haena Moon , Su-Mi Kim , Jung-Woo Seo , Yu Ho Lee , Yang Gyun Kim , Ju-Young Moon , 3 2 1 So-Young Lee , Y oungsook Son & Sang-Ho Lee Diabetes is a risk factor for acute kidney injury (AKI) and chronic kidney disease (CKD). Diabetic patients are easy to progress to CKD after AKI. Currently, activation of fibrotic signalling including transforming growth factor-β (TGF-β ) is recognized as a key mechanism in CKD. Here, we investigated the influence 1 1 of diabetes on CKD progression after AKI by using a unilateral renal ischaemia–reperfusion injury (IRI) model in diabetic mice. IRI induced extensive tubular injury, fibrosis and lymphocyte recruitment at 3 weeks after IRI, irrespective of diabetes. However, diabetes showed sustained tubular injury and markedly increased fibrosis and lymphocyte recruitment compared with non-diabetes at 5 week after IRI. The mRNAs and proteins related to TGF-β and sonic hedgehog (Shh) signalling were significantly higher in diabetic versus non-diabetic IRI kidneys. During the in vitro study, the hyperglycaemia induced the activation of TGF-β and Shh signalling and also increased profibrogenic phenotype change. However, hyperglycaemic control with insulin did not improve the progression of renal fibrosis and the activation of TGF-β and Shh signalling. In conclusion, diabetes promotes CKD progression of AKI via activation of the TGF-β and Shh signalling pathways, but insulin treatment was not enough for preventing the progression of renal fibrosis. Aer ac ft ute kidney injury (AKI), incomplete tubular recovery leads to renal fibrosis and decreased renal function, the common components of chronic kidney disease (CKD) . Generally, in patients with no underlying diseases, recovery from acute kidney injury occurs without significant renal fibrosis. However, transient kidney damage may eventually lead to renal fibrosis in the presence of underlying diseases such as diabetes and CKD . This phe- nomenon is not only limited to the kidney but can also occur aer s ft kin damage or hind limb ischaemia in animal 3,4 models with diabetes . Additionally, the presence of diabetes or underlying CKD are independent risk factors 5,6 for acute kidney injury aer c ft ardiac surgery and coronary/vascular interventions using contrast . Therefore, the transition of AKI to CKD is a clinically serious problem for diabetic patients. Over the past several decades, many studies have been conducted to identify the pathophysiology involved in the development of AKI. However, much remains unknown about the mechanism of transition from AKI to CKD. Recent studies have focused on the role of damaged tubules and a subpopulation of incompletely recov- ered tubules ae ft r AKI, which lead to abnormal growth arrest, failure to redifferentiate into normal tubules, and finally atrophy, as the result of abnormal wound healing . If abnormal wound healing persists or when metabolic derangements impair normal wound healing, atrophic tubules produce persistent and progressively increas- 8,9 ing levels of profibrotic signalling molecules such as TGF-β and Shh . These paracrine factors intrinsically 10,11 play a role in mediating normal wound repair . However, persistent activation of these signalling pathways and abnormal cross talk between unhealed tubular cells and interstitial cells such as infiltrating immune cells Division of Nephrology, Department of Internal Medicine, Kyung Hee University Hospital at Gangdong, College of Medicine, Kyung Hee University, Seoul, Korea. Department of Genetic Engineering, College of Life Science and Graduate School of Biotechnology, Kyung Hee University Global Campus, Yongin, Korea. Division of Nephrology, Department of Internal Medicine, CHA Bundang Medical Center, CHA University, Seongnam, Korea. Correspondence and requests for materials should be addressed to Y.S. (email: ysson@khu.ac.kr) or S.-H.L. (email: lshkidney@khu.ac.kr) SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 1 www.nature.com/scientificreports/ or activated fibroblasts eventually leads to myofibroblast transformation of pericyte-like fibroblast or bone 12,13 marrow-derived precursor cells, the final and common pathological feature of renal fibrosis . 9,14 Both of TGF-β and Shh pathways are known as typical signalling mediators which lead to renal fibrosis . Neutralization of TGF-β prevents blood vessel loss and development of tubulointerstitial fibrosis after IRI. 15,16 Additionally, blockage of Shh signalling also reduces renal fibrosis . Hyperglycaemia induces high expression 17,18 levels of TGF-β and increases levels of Smad 2/3 and CTGF induced by TGF-β . However, the relationship 1 1 between hyperglycaemia and the activation of the Shh pathway is currently unclear. It is also well known that the mechanism of the injury and repair process is abnormally controlled in the diabetic condition . In addition, aberrant inflammatory cell recruitment and activation of prob fi rotic signalling pathways are already among the major pathologic mechanisms of diabetic nephropathy . e uni Th lateral ischaemia reperfusion injury model is suitable for observing the progression of CKD because the characteristics of CKD such as renal mass reduction and tubulointerstitial fibrosis increase with the severity of 21,22 ischaemic-reperfusion injury . Therefore, we hypothesize that enhanced and persistent activation of prob fi rotic signalling molecules such as TGF-β and Shh under diabetic conditions induces abnormal b fi rotic repair rather than normal wound healing aer AKI, w ft hich finally accelerates the progression of CKD. Results Diabetes impaired the improvement of tubular injury and aggravated renal fibrosis after IRI. To investigate the effect of diabetes on the progression of post-ischaemic renal fibrosis, we used the unilateral renal ischaemia-reperfusion injury (IRI) model in non-diabetic and diabetic mice (Fig. 1a). Compared with sham treatment, IRI induced extensive tubular injury and increased the fibrotic area at 3 weeks aer IRI, ir ft respective of diabetes. While the degree of tubular injury between 3 and 5 weeks after IRI was significantly improved in non-diabetic mice, it was maintained in diabetic mice (Fig. 1b,d). Furthermore, continuous renal mass reduction was seen only in diabetic IRI (Table 1). The degree of renal fibrosis at 5 weeks after IRI was increased in both non-diabetic and diabetic mice. However, its progression was significantly higher in diabetic than non-diabetic mice (Fig. 1c,e). These results indicate that apoptotic tubular damage aer IRI l ft asts longer and is accompanied by the progression of renal fibrosis under diabetic conditions. Diabetes also aggravated aberrant lymphocyte recruitment after IRI. Because aberrant lympho- 23 + cyte recruitment of is one of the key features of diabetic nephropathy , we assessed the infiltration of CD4 , + + CD8 and CD20 lymphocytes during the progression of kidney fibrosis aer AKI. A ft t 3 weeks aer IRI, t ft he num- ber of inflammatory immune cells was increased in IRI kidneys, but showed no difference between non-diabetic + + + and diabetic mice. At 5 weeks aer IRI, sig ft nificantly increased infiltration of CD4 , CD8 and CD20 lympho- cytes was observed in diabetic mice compared to non-diabetic mice (Fig. 2a–f ). The mRNA expression levels of inflammatory cytokines including TNF-α , IFN-γ and CCL2 were significantly higher in IRI kidneys than sham kidneys at 3 weeks after IRI (Fig.  2g–i). The expression levels of TNF-α and CCL2 were also higher in diabetic kidneys than in non-diabetic kidneys. From 3 to 5 weeks after IRI, TNF- α and IFN-γ levels were significantly decreased in non-diabetic IRI kidneys, but not in diabetic IRI kidneys (Fig. 2g,h). Although CCL-2 levels were decreased at 5 weeks in both diabetic and non-diabetic kidneys, they were still higher in diabetic compared to non-diabetic IRI kidneys (Fig. 2i). These results indicate that diabetes augmented lymphocyte recruitment to scar kidneys aer ft IRI and maintained the activation of intra-renal inflammation and inflammatory mediators, which could play pivotal roles in the fibrotic cascade. Diabetes induced persistent activation of TGF-β and Shh signalling after IRI. First, we per- formed immunohistochemistry for TGF-β , Shh and Smo to identify their expression in kidneys after IRI (Fig. 3a–c). In diabetic kidneys, the slightly increased expression of TGF-β was shown in remaining tubules and strongly stained the cytoplasm of interstitial cells, especially at 5 weeks aer IRI (Fig ft .  3a). Expression of Shh and Smo was increased in injured tubules rather than the fibrotic area at 3 and 5 weeks aer IRI (Fig ft .  3b,c). The Shh-positive area was slightly reduced in diabetic kidney at 5 weeks because of the replacement of inflamma - tion and expansion of the extracellular matrix. However, Shh expression in remaining and atrophic tubules was increased and more prominent (Fig. 3b). Smo expression was also increased in IRI kidneys, but the increase was higher in diabetic mice at 3 and 5 weeks aer IRI (Fig ft .  3c). Next, we evaluated the protein expression of TGF-β and Shh signalling pathways in kidneys at 5 weeks ae ft r IRI (Fig. 3d–f ). TGF-β expression was not significantly increased and Smad2 expression was even decreased at 5 weeks aer IRI in n ft on-diabetic mice. However, their expression levels were significantly higher in diabetic IRI kidneys compared with sham or non-diabetic kidneys (Fig. 3d). Shh signalling also showed a pattern similar to TGF-β signalling. Shh expression was increased only in diabetic IRI kidneys at 5 weeks aer IRI. S ft mo expression was also significantly increased only in diabetic kidneys (Fig.  3e). Subsequently, we assessed the expression of α-SMA and fibronectin, markers of fibrosis and extracellular matrix accumulation respectively. α -SMA expres- sion was increased in IRI kidneys compared to sham kidneys, and their tendencies were significantly enhanced in diabetic IRI kidneys. The expression of fibronectin was significantly increased only in diabetic IRI kidneys (Fig. 3f ). Finally, we confirmed the mRNA expression of fibrosis related proteins in kidneys after IRI to demonstrate the effect of diabetes on TGF-β and Shh signalling pathways. mRNA expression of TGF-β was induced at 3 1 1 weeks after IRI but was more significantly increased in diabetic IRI kidneys. Although TGF-β mRNA levels were decreased from 3 to 5 weeks aer IRI in b ft oth groups, they were still significantly higher in diabetic than in non-diabetic IRI kidneys (Fig. 3g). The downstream factors of TGF-β signalling such as CTGF and their target collagen Ia-1 were increased at 3 weeks after IRI but were significantly higher in diabetic than in non-diabetic IRI kidneys. At 5 weeks aer IRI, t ft he mRNA expression levels of CTGF and collagen Ia-1 were slightly decreased SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 2 www.nature.com/scientificreports/ Figure 1. Diabetes impaired the improvement of tubular injury and aggravated renal fibrosis aer IRI. ( ft a) Experimental design. (b) Periodic acid–Schiff (PAS) staining was used to detect interstitial tubular injury at 3 and 5 weeks aer IRI. ( ft c) Masson’s trichrome (MT) staining was used to detect kidney fibrosis at 3 and 5 weeks aer IRI. ( ft d) Quantification of interstitial tubular injury based on a grading score at 3 and 5 weeks aer IRI. ( ft e) Quantification of renal fibrosis at 3 and 5 weeks aer IRI. V ft alues are expressed as the mean ± S.E.M. p < 0.05 # † versus sham, p < 0.05 versus non-diabetic, p < 0.05 versus values at 3 weeks. Scale bar = 80 µm. in non-diabetic IRI kidneys. However, their expression levels were still maintained or increased in diabetic IRI kidneys (Fig. 3h,i). Although mRNA expression of Shh was not significantly increased in diabetic IRI kidneys, its downstream signal, Gli-1, showed similar expression to TGF-β (Fig. 3j,k). Its expression was significantly higher in both non-diabetic and diabetic IRI kidneys at 3 weeks. Even during its decrease between 3 and 5 weeks, Gli-1 expression remained significantly higher in diabetic IRI kidneys at 5 weeks (Fig.  3k). Furthermore, the mRNA expression of Snail1, which was known to be a downstream signal of Gli-1, was strongly induced only in diabetic kidneys at 5 weeks aer IRI (Fig ft .  3l). These results showed that TGF-β and Shh signalling were persistently acti- vated in diabetic IRI kidneys until 5 weeks aer IRI, a ft nd were associated with increasing renal fibrosis. SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 3 www.nature.com/scientificreports/ Non-diabetic Diabetic Sham IRI Sham IRI # # 3w 136.0 ± 4.7 131.5 ± 6.4 504.5 ± 42.4 568.7 ± 10.2 Blood glucose level (mg/dL) # # 5w 147.3 ± 4.1 157.0 ± 7.3 515.6 ± 49.8 442.4 ± 42.6 # # 3w 26.67 ± 1.17 25.08 ± 0.64 22.27 ± 0.87 20.62 ± 0.33 Body weight (g) † # # 5w 28.55 ± 0.86 28.30 ± 0.96 24.44 ± 0.81 22.20 ± 1.25 3w 0.163 ± 0.008 0.107 ± 0.004* 0.168 ± 0.005 0.097 ± 0.006 Kidney weight (g) *#† 5w 0.140 ± 0.007 0.105 ± 0.013 0.162 ± 0.008 0.062 ± 0.005 # * 3w 0.613 ± 0.011 0.425 ± 0.014* 0.756 ± 0.028 0.470 ± 0.029 Kidney to body weight ratio (%) † # *† 5w 0.490 ± 0.011 0.380 ± 0.060 0.669 ± 0.051 0.282 ± 0.026 3w 1.073 ± 0.021 0.901 ± 0.021* 1.076 ± 0.010 0.844 ± 0.020 Kidney length (cm) *#† 5w 1.073 ± 0.015 0.902 ± 0.041* 1.052 ± 0.011 0.743 ± 0.024 * # † Table 1. Values are the mean ± S.E.M. p < 0.05 versus Sham, p < 0.05 versus non-diabetic, p < 0.05 versus values at 3 weeks. Treatment of insulin was showed no improvement of renal fibrosis after IRI. To explore whether the regulation of hyperglycaemia by insulin treatment ae ff ct the progression of renal fibrosis after IRI, we con- ducted follow-up experiments with some modic fi ation as shown Fig.  4a. IRI was induced at 8 weeks ae ft r strepto - zotocin injection and insulin was administrated from 2 weeks aer t ft he induction of diabetes until the end of the experiment. Insulin administration significantly mitigated the increase of HgA1c levels during the study period (Fig. 4b). From 5 to 8 weeks aer IRI, t ft he tubular injury was significantly decreased in all groups. Renal fibrosis was not significantly increased in non-diabetic IRI group. As expected, renal fibrosis in diabetic IRI group was significantly increased at the same time period. However, treatment of insulin to diabetic mice did not ae ff ct the tubular injury and renal fibrosis induced by IRI (Fig.  4c–f ). Regardless of the treatment of insulin, the protein expressions of TGF-β and Shh were significantly higher in diabetic IRI groups, not mitigated by insulin. Both TGF-β and Shh expression were significantly activated in IRI kidneys as compared with those of sham. Shh expression was significantly higher at 5 weeks and TGF-β expression was significantly higher at 8 weeks in dia- betic IRI group as compared with those of non-diabetic IRI group (Fig. 4g,h). Hyperglycaemia augmented the effect of Shh and TGF-β on profibrogenic phenotype change in renal tubular cells. To demonstrate the effect of hyperglycaemia on TGF-β and Shh signalling, HKC-8 was cultured under hyperglycaemic conditions. Compared to normoglycaemic conditions (5 mM D-Glucose, NG), the hyperglycaemic state (30 mM D-Glucose, HG) induced TGF-β and Shh expression at 6 hours. Although Shh expression was significantly reduced at 24 hours, TGF-β was maintained until 24 hours (Fig. 5a–c). Smo expression was significantly increased in HG at 18 and 24 hours (Fig. 5d). However, Gli-1 was induced at 6 hours and returned to its basal level at 18 hours (Fig. 5e). Fibronectin and α-SMA expression were induced at 6 hours and were maintained through 24 hours (Fig. 5f,g). These results suggest that hyperglycaemia activates the profi- brotic signalling pathway through the activation of Shh as well as TGF-β . Next, we exposed recombinant human Shh or TGF-β to NG or HG for 24 hours. Shh treatment increased the expression of TGF-β in both NG and HG (Fig. 6a). Expression levels of α-SMA and fibronectin were increased in both NG and HG by Shh treatment aer ft 24 hours. However, the induction of α-SMA and b fi ronectin by Shh was significantly greater in HG than in NG (Fig.  6a,b). In the case of TGF-β treatment, Shh expression was increased in both NG and HG (Fig. 6c). Fibronectin expression was increased in both NG and HG at 24 hours aer T ft GF-β treatment. However, there was no difference between NG and HG (Fig.  6c). α-SMA expression of was increased by TGF-β after 24 hours in HG, but not in NG (Fig. 6d). These results indicate that hyperglycaemia increased the expression of Shh or TGF-β in terms of prob fi rogenic phenotype change in HKC-8 cells. Although Shh and TGF-β interacted with each other, hyperglycaemia did not ae ff cted the crosstalk between Shh and TGF-β . 1 1 Insulin increased the expressions of Shh and TGF-β without affecting the interaction between Shh and TGF-β1 in hyperglycaemic conditions. To investigate the mechanisms by which insulin has no effect on renal fibrosis aer IRI, HK ft C-8 was cultured in LG and HG with insulin stimulation. The expressions of TGF-β and Shh were increased by insulin stimulation as well as in HG. In HG, the expression of Shh was not enhanced by insulin, but the expression of TGF-β was significantly increased by insulin (Fig.  7a,b). The increased mutual stimulation between Shh and TGF-β under hyperglycaemic condition was not altered by insu- lin (Fig. 7c,d). Insulin also did not mitigate the activated expression of α-SMA by Shh or TGF-β under the HG, but significantly increased the Shh-induced expression of α-SMA in HG (Fig.  7c). Discussion Among diverse causes of chronic kidney disease, diabetes accounts for the largest portion of kidney failure, account- ing for approximately 50% of cases in the developed world . High blood pressure, poor glycaemic control and albu- minuria are well-known risk factors for the development or progression of diabetic nephropathy. However, these factors cannot explain all of the interindividual variability in the rate of progression to end-stage of CKD . Recently, AKI episodes have been known to be associated with a cumulative risk for developing advanced CKD in diabetes mellitus, independent of other major risk factors of progression . Additional loss of parenchyma caused by failed SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 4 www.nature.com/scientificreports/ Figure 2. Diabetes aggravated inflammation aer IRI. ( ft a–c) The infiltrated (a) CD4-positive, (b) CD8-positive T cells and (c) CD20-positive B cells measured by immunohistochemistry. (d–f) Quantification of the number of infiltrated (d) CD4-positive, (e) CD8-positive T cells and (f) CD20-positive B cells in whole kidneys. (g–i) mRNA expression of inflammatory cytokine markers, (g) TNF-α, (h) IFN-γ, and (i) CCL-2, determined * # via qRT-PCR. Values are expressed as the mean ± S.E.M. p < 0.05 versus sham, p 0.05 versus non-diabetic, p < 0.05 versus at 3 weeks. Scale bar = 50 µm. repair of AKI has been implicated in the development and progression of experimental nephropathy in diabetic rodents using an acute-on-chronic injury model . However, the underlying mechanism of the progression of CKD aer AKI un ft der diabetic conditions is not currently well known. Here, we demonstrated why diabetes is particularly prone to progression to CKD aer AKI t ft hrough three major findings in our study. SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 5 www.nature.com/scientificreports/ Figure 3. Diabetes induced persistent activation of TGF-β and Shh signalling. (a–c) Immunohistochemistry for kidney levels of (a) TGF-β , (b) Shh and (c) Smo at 3 and 5 weeks aer IRI. S ft cale bar = 80 µm. (d) The kidney levels of expressed TGF-β and Smad2 were measured by western blot at 5 weeks aer IRI. ( ft e) The kidney levels of expressed Shh and Smo were measured by western blot at 5 weeks aer IRI. ( ft f) The kidney levels of expressed α-SMA and fibronectin were measured by western blot at 5 weeks aer IRI. E ft ach protein expression was normalized by GAPDH. The fold-change of each protein was calculated as the ratio of averages versus non- diabetic sham. (g–l) mRNA expression of the major TGF-β and Shh signalling pathway factors in the kidney were determined via qRT-PCR at 3 and 5 weeks aer IRI. Th ft e targets of mRNA were (g) TGF-β , (h) CTGF, (i) collagen I-a1, (j) Shh, (k) Gli-1 and (l) Snail1. Each value of the target mRNA was normalized by 18 S. Values are * # † expressed as the mean ± S.E.M. p < 0.05 versus sham, p < 0.05 versus non-diabetic p < 0.05 versus at 3 weeks. First, diabetes exacerbated abnormal lymphocyte infiltration aer IRI. I ft n this study, the recruitment of CD4 , + + CD8 and CD20 lymphocytes in the kidney was increased at 3 weeks aer IRI. H ft owever, there was no difference between non-diabetic and diabetic kidneys at 3 weeks ae ft r IRI. At 5 weeks ae ft r IRI, levels of infiltrating CD4 , + + CD8 and CD20 lymphocytes were sharply increased, but were significantly higher in diabetic IRI kidneys than in non-diabetic IRI kidneys. Aberrant recruitment of T and B lymphocytes is one of the major findings in diabetic 27,28 nephropathy . At the early phase of AKI, the major infiltrating immune cells in the kidney are neutrophils and SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 6 www.nature.com/scientificreports/ Figure 4. Treatment of insulin was showed no improvement of renal fibrosis aer IRI. ( ft a) Experimental design. (b) The value of HbA1c (c) Periodic acid–Schiff (PAS) staining was used to detect kidney tubular injuries and to view necrosis, atrophy and normal tubules at 5 and 8 weeks aer IRI. ( ft d) Masson’s trichrome (MT) staining was used to detect kidney fibrosis at 5 and 8 weeks aer IRI. ( ft e) The quantification of total interstitial tubular injury based on a grading score at 5 and 8 weeks aer IRI. ( ft f) The quantification of renal fibrosis at 5 and 8 weeks aer ft IRI. (g) The expression of TGF-b1 and Shh in the kidney at 5 weeks aer IRI. ( ft h) The expression of TGF-b1 and Shh in the kidney at 8 weeks aer IRI. V ft alues are expressed as the mean ± S.E.M. p < 0.05 versus sham, # † ‡ p < 0.05 versus non-diabetic, p < 0.05 versus at 5 weeks, p < 0.05 versus between diabetic and diabetic insulin group. Scale bar = 80 µm. macrophages. In contrast, the recruitment of lymphocytes into the injured kidney was reported to be prominent 29,30 during the recovery phase . e Th exact role of these lately infiltrating lymphocytes has not been studied exten- sively. However, some studies suggested that B cell infiltration aer AKI mig ft ht hinder normal wound healing . Interestingly, recent studies also suggested that infiltrated lymphocytes directly ae ff cted renal fibrosis by increas- 32,33 ing levels of TGF-β . e Th major finding of this study was that diabetes induced persistent activation of TGF- β and Shh signalling. In this study, persistent activation of the TGF-β and Shh signalling pathways was maintained only in the dia- betic group at 5 weeks after IRI. Furthermore, the downstream signalling molecules Smo and Gli-1 were also persistently enhanced in diabetic IRI kidneys. These results showed that TGF-β and Shh signalling were per- sistently activated in diabetic IRI kidneys until 5 weeks aer IRI a ft s well as were associated with increasing renal b fi rosis. e Th TGF-β signalling pathway reduces tissue injury and promotes tubule restoration through epithelial 34,35 de-differentiation at the early stage of acute injury . However, under maladaptive wound healing conditions, SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 7 www.nature.com/scientificreports/ Figure 5. Hyperglycaemia induced EMT via activation of TGF-β and Shh signalling. HKC-8 cells were cultured in 30 mM D-glucose to induce hyperglycaemic conditions in the cells. (a) The expressions of TGF-β , Shh, Smo, Gli-1, fibronectin and α-SMA were measured using western blotting in a time-dependent manner. e f Th old-changes of (b) TGF-β , (c) Shh, (d) Smo, (e) Gli-1, (f) fibronectin and (g) α-SMA were calculated as the ratios of averages versus a 5 mM D-glucose control. Expression of each protein was normalized by GAPDH. Values are expressed as the mean ± S.E.M. p < 0.05 versus 5 mM D-glucose control. Figure 6. Hyperglycaemia augmented the effect of Shh and TGF-β on prob fi rogenic phenotype change in renal tubular cells. HKC-8 cells were cultured in 5 mM D-glucose or 30 mM D-glucose with 10 ng/ml Shh or 10 ng/ ml TGF-β . (a) Expression levels of TGF-β and fibronectin at 24 hours aer S ft hh treatment. (b) Representative 1 1 confocal fluorescence images of α-SMA were examined at 24 hours aer S ft hh treatment. (c) Expression levels of Shh and fibronectin at 24 hours aer T ft GF-β treatment. (d) Representative confocal fluorescence images of α-SMA were examined at 24 hours aer T ft GF-β treatment. Expression of each protein was normalized by GAPDH. The fold-change of each protein was calculated as the ratio of averages versus a 5 mM D-glucose control. Values are expressed as the mean ± S.E.M. p < 0.05 versus 5 mM D-glucose or 30 mM D-glucose controls, p < 0.05 versus 5 mM D-glucose with treatment. TGF-β signalling could promote abnormal organ fibrosis through myob fi roblast differentiation from fibroblasts 36,37 or bone marrow-derived precursor cells . The Shh signalling pathway could also modulate the activation of 38–40 myob fi roblast and kidney b fi rosis aer ft tubular injury . Shh was secreted from the damaged renal tubular epi- 41,42 thelium and induced fibroblasts to differentiate into myob fi roblasts . Furthermore, the overexpression of Shh was reported to increase extracellular matrix (ECM) synthesis . Contrary to the TGF-β signalling pathway, the effect of diabetes on Shh signalling pathway has not yet been clarified. However, the results of this study indicate that sustained tissue injury, activation of inflammation as well as hyperglycaemia itself could be suggested as mechanisms underlying the activation of Shh signalling during CKD progression aer di ft abetic AKI. We also demonstrated using in vitro experiments that hyperglycaemia induced the activation of Shh signalling as well as TGF-β and augmented the effect of both Shh and TGF-β on prob fi rogenic phenotype change in renal 1 1 tubular cells. Although cell culture models in diabetic kidney injuries such as our study, have several limitations in mimicking in vivo tissue environment, these results could be explained by the finding that hyperglycaemia increased the activation of Shh and TGF-β . Recent studies indicated that Shh signalling could be activated by TGF-β . Furthermore, TGF-β and its receptor were also reported to be increased by the activation of Shh sig- 1 1 nalling . The crosstalk between Shh and TGF-β was also recently reported to enhance kidney fibrosis in several SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 8 www.nature.com/scientificreports/ Figure 7. Insulin increased the expressions of Shh and TGF-β without ae ff cting the interaction between Shh and TGF-β in hyperglycaemic condition. (a) Expression level of TGF-β at 24 hours aer in ft sulin treatment. 1 1 (b) Expression level of Shh at 24 hours aer in ft sulin treatment. (c) Expression levels of TGF-β and α-SMA at 24 hours aer in ft sulin and Shh treatment. (d) Expression levels of Shh and α-SMA at 24 hours aer in ft sulin and TGF-β treatment. Expression of each protein was normalized by GAPDH. The fold-change of each protein was calculated as the ratio of averages versus a 5 mM D-glucose control. Values are expressed as the mean ± S.E.M. * # † p < 0.05 versus 5 mM D-glucose, p < 0.05 versus 30 mM D-glucose. p < 0.05 versus between 5 mM D-glucose with insulin treatment and 30 mM D-glucose with insulin treatment. p < 0.05 versus 30 mM D-glucose with Shh or TGF-β treatment. 44–46 models of fibrotic kidney diseases . Our results also showed the mutual influence of Shh and TGF-β However, hyperglycaemia did not have a significant effect on the interaction between Shh and TGF-β Since insulin treatment is one way to control the hyperglycemia, which may be the target to rescue the kidney b fi rosis following IRI, insulin treatment could be expected to reduce the expression of TGF-β and Shh signaling and finally ameliorate the progression of renal fibrosis aer IRI. ft To answer this question, we conducted the additional experiment with some modifications as in Fig. 4 . As expected, longer-lasting diabetes before IRI induction and hyperglycemia extended to 8 weeks significantly aggra- vated the post-ischaemic fibrosis. However, treatment of insulin did not improved renal fibrosis in our animal model showing progressive kidney fibrosis aer IRI, e ft ven though insulin showed significantly improved control of blood glucose level. Our results also showed that treatment of insulin had no effect on the expression of TGF-β and Shh at both 5 and 8 weeks aer IRI. N ft egative result with insulin treatment could be anticipated, because the correction of hyperglycemia might be suboptimal in our experiment (mean HgA1c 6.5%) or ischaemic kidney damage in diabetic group was too severe to overcome with the blood glucose control. However, we had noted the results in the previous reports, which showed that insulin induced the expression of TGF-β in renal proximal 47 48 tubular cells , and insulin stimulated SLT2-mediated tubular glucose absorption via oxidative stress generation . In our in vitro experiment, we also confirmed that insulin itself could induce the expression of TGF-β and also ae ff ct the expression of Shh in HKC-8 cells. However, insulin was not altered the mutual interaction between Shh and TGF-β under high glucose condition. Based on these results, control of hyperglycemia by treatment of SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 9 www.nature.com/scientificreports/ insulin may not be enough for preventing renal fibrosis aer IRI, a ft nd further studies on the effect of other hypo- glycemic agents with different action mechanism on tubular cells, are necessary. In conclusion, we demonstrated that diabetes promoted CKD progression following AKI through recruitment of lymphocytes and persistent activation of TGF-β and Shh signalling pathways. The control of hyperglycaemia with insulin administration was not enough for preventing the progression of renal fibrosis. Future study of the mechanism of renal fibrosis progression aer di ft abetic AKI will contribute to solving the unresolved mechanisms underlying kidney fibrosis and to locating potential therapeutic targets for treatment of diabetic kidney failure. Materials and Methods Animals and experimental design. Effect of hyperglycaemia on the progression of renal fibrosis after IRI. Diabetes was induced by streptozotocin (Sigma-Aldrich, St. Louis, MO, USA) injected once a day at a concentration of 50 mg/kg for five consecutive days into 8-week-old male C57BL/6 mice. After 2 weeks, a uni- lateral renal ischaemia–reperfusion injury (IRI) model was established in non-diabetic mice and diabetic mice with blood glucose levels above 250 mg/dL. We induced ischaemia in the left kidneys of the mice using a clamp to obstruct blood circulation for 25 minutes. During the ischaemic period, mice were maintained within a body temperature range of 36–37 °C using a heating pad and a rectal temperature probe. Mice were sacrificed at 3 or 5 weeks after IRI and divided into four groups: (1) non-diabetic controls (Nor Sham), (2) non-diabetic IRI (Nor & IRI), (3) diabetic controls (DM sham), and (4) diabetic IRI (DM & IRI). Mice were monitored for body weights and blood glucose levels every 2 weeks and HbA1c levels at 2 and 6 weeks. All animal experiments were performed in compliance with the guidelines of the Animal Research Ethics Committee of Kyung Hee University and Institutional Animal Care and Use committee Kyung Hee University Hospital at Gangdong, Seoul, Korea (approval number: KHNMC AP 2015-001). Ee ff ct of insulin treatment on the progression of post-ischaemic renal fibrosis. At 2 weeks aer di ft abetic induction, Insulin (10 unit/kg, insulin glargine, LANTUS by Sanofi-Aventis was administrated once a day until the end of the experiment. At 8 weeks aer di ft abetic induction, the unilateral IRI was established in non-diabetic, diabetic and insulin-treated diabetic mice. Mice were sacrificed at 5 and 8 weeks after IRI and divided into six groups: (1) non-diabetic controls (Nor Sham), (2) non-diabetic IRI (Nor & IRI), (3) diabetic controls (DM Sham), (4) diabetic IRI (DM & IRI), (5) insulin-treated diabetic controls (Ins Sham) and (6) insulin-treated diabetic IRI (Ins & IRI). Blood glucose level was checked every 2 weeks and HbA1c was checked at 2, 6, 9 and 12 weeks aer ft diabetic induction. All animal experiments were performed in compliance with the guidelines of the Animal Research Ethics Committee of Kyung Hee University and Institutional Animal Care and Use committee Kyung Hee University Hospital at Gangdong, Seoul, Korea (approval number: KHNMC AP 2016-007). Histology. Sections were cut at 4 µ m thickness. For the histological assessment of tubulointerstitial injury (tubular necrosis, tubular atrophy, and glomerular cysts), the sections were stained with periodic acid-Schiff reagent. Ten corticomedullary fields were examined in each section at 200x magnification, and a semiquanti- tative analysis of tubulointerstitial injury was performed. Total tubular injury was graded on a scale of 0 to 5 based on the percentage of normal tubules and the amount of tubular necrosis and tubular atrophy as follows: 0, absent; 1, 1–25%; 2, 26–50%; 3, 51–75%; 4, 76–99%; and 5, 100%. Fibrosis was quantified using Masson’s tri- chrome staining and computer-assisted image analysis. For immunohistochemistry, we used the Bond Polymer Refine Detection system (Vision BioSystems, Hingham, MA, USA) with antibodies against CD4, CD8 and CD20. Four-micron-thick kidney sections were deparaffinized using Bond Dewax solution, and an antigen retrieval pro- cedure was then performed using Bond ER solution for 30 minutes at 100 °C. Endogenous peroxidase activity was terminated by incubating the tissues with hydrogen peroxide for 5 minutes. The sections were then incubated with CD4 (1:100, Abcam, Cambridge, MA, USA), CD8, and CD20 (1:100; Abcam, Cambridge, MA, USA) antibod- ies using a biotin-free polymeric horseradish peroxidase-linked antibody conjugate system (Vision BioSystems, + + + Hingham, MA, USA). To assess T and B lymphocyte infiltration, CD4 /CD8 /CD20 cells were counted in 20 randomly selected fields from each section at 400x magnification under a light microscope. In vitro cell culture. e h Th uman renal proximal tubular epithelial cell line HKC8 was obtained from Dr. L. Rausen (Johns Hopkins University, Baltimore, MD, USA) and was maintained in Dulbecco’s Modified Eagle’s Medium supplemented with Ham’s F12 medium (DMEM/F12; Invitrogen, Carlsbad, CA, USA) DMEM/F12 was supplemented with 5% foetal bovine serum and 1% penicillin/streptomycin (WelGENE, Daegu, Korea). Before experiments, cells were maintained in a medium containing 5 mM D-glucose. 3 × 10 cells were seeded in a 60 mm dish with 5 mM D-glucose and cultured for 24 hours . To confirm the ee ff ct of hyperglycaemia on the prob fi ro - genic signalling pathway in a time-dependent manner, we changed the media containing 30 mM D-glucose and collected cells at 3, 6, 18 and 24 hours aer t ft reatment. To demonstrate the effects of TGF-β and Shh on HKC-8 cells, media were changed following these conditions: (1) 5 mM D-glucose, (2) 5 mM D-glucose + TGF-β 10 ng/ ml, (3) 5 mM D-glucose + Shh 10 ng/ml, (4) 30 mM D-glucose, (5) 30 mM D-glucose + TGF-β 10 ng/ml, and (6) 30 mM D-glucose + Shh 10 ng/ml. To demonstrate the effect of insulin on the expressions of TGF-β and Shh in HKC-8 cells, media were changed following these conditions: (1) 5 mM D-glucose, (2) 5 mM D-glucose + insu- lin 50 ng/ml, (3) 5 mM D-glucose + insulin 5000 ng/ml, (4) 30 mM D-glucose, (5) 30 mM D-glucose + insulin 50 ng/ml, and (6) 30 mM D-glucose + insulin 5000 ng/ml. To identify the effect of insulin on the sensitivity to TGF-β and Shh, media were changed following these conditions: (1) 5 mM D-glucose, (2) 30 mM D-glucose, (3) 30 mM D-glucose + TGF-β 10 ng/ml, (4) 30 mM D-glucose + TGF-β 10 ng/ml + insulin 50 ng/ml, (5) 30 mM 1 1 D-glucose + Shh 10 ng/ml, and (6) 30 mM D-glucose + Shh 10 ng/ml + insulin 50 ng/ml. After 24 hours, cells were collected for protein analysis. SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 10 www.nature.com/scientificreports/ Isolation of total RNA and real-time PCR. Total RNA was extracted from kidney tissue by using the Nucleospin RNA kit (Macherey-Nagel, Düren, W. Germany) according to the manufacturer’s instructions and was quantified using a NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific Inc., San Jose, CA, USA). Complementary DNA was synthesized using random primers (Promega, Madison, WI, USA), DNTP mixture (TaKaRa Bio Inc., Otsu-shi, Shiga, Japan) and M-MLV reverse transcriptase (Mbiotech Inc., Hanam, Korea). Real-time PCR was performed in reactions with a final volume of 20 μl containing 1 μl of cDNA, 10 pmol of each sense and antisense primer, and 17 μl of Power SYBR Green PCR Master mix (Applied Biosystems, Beverly, MA, USA) and was detected using the Applied Biosystems StepOnePlus Real-Time PCR System. The primers for ® ™ TGF-β , CTGF, collagen IV-a1, collagen I-a1, Snail1, Shh, Gli-1, IFN-γ, TNF-α, CCl-2 and 18 S were purchased from Mbiotech. Each sample was run in duplicate in separate tubes to quantify target gene expression, and the results were normalized to 18 S expression. Western blot analysis. Cells and kidney tissues were washed with PBS and lysed in the M-PER mam- malian protein extraction reagent with protease inhibitor cocktail (Thermo Fisher Scientific Inc., San Jose, CA, USA). Proteins were separated with 8–15% SDS-PAGE and then were transferred onto a nitrocellulose membrane (Millipore, Madrid, Spain) by electroblotting. The membrane was blocked for 1 hour at room temperature and then was incubated overnight at 4 °C with anti-Shh, anti-E-cadherin, anti-Smad2, anti-Smo, anti-Gli-1 (1:1000, Santa Cruz biotechnology, Santa Cruz, CA, USA), anti-fibronectin (R&D system Inc. Minneapolis, MN, USA), anti-Bax, anti-Bcl-2, anti-TGF-β (1:1000, Cell Signaling Technology, Beverly, MA, USA), and α-SMA (1:1000 Abcam Inc. Cambridge, MA, USA) primary antibodies. Subsequently, the membranes were stained with horse- radish peroxidase-conjugated goat anti-rabbit or mouse immunoglobulin G (1:2,000, Santa Cruz biotechnology, Santa Cruz, CA, USA). The immunoreactive bands were detected by chemiluminescence (enhanced chemilumi- nescence; BioFX Laboratories Inc., Owings Mills, Maryland, USA). GAPDH (1:2,000, Santa Cruz biotechnology, Santa Cruz, CA, USA) was used as an internal control. Confocal microscopy. Cells were fixed with 4% paraformaldehyde, permeabilized with 0.2% Triton X-100 solution, blocked with bovine serum albumin (BSA), and incubated with primary antibodies for 2 hours. Aer ft washing with PBS, the samples were re-incubated with secondary antibodies conjugated with Alexa Fluor 488 or Texas Red (Life Technologies, Gaithersburg, MD, USA) for 1 hour. Cells were counterstained with DAPI to delineate the nuclei, and the sections were examined using confocal microscopy (LSM-700; Carl Zeiss, Jena, ur Th ingia, Germany). Statistical analyses. All values are expressed as the mean ± SE. The results were analysed using the Kruskal-Wallis nonparametric test for multiple comparisons. Significant differences detected using the Kruskal-Wallis test was confirmed using the Wilcoxon rank sum and Mann-Whitney tests (to compare mean differences). p-values < 0.05 were considered to indicate statistical significance. References 1. Ferenbach, D. A. & Bonventre, J. V. Mechanisms of maladaptive repair aer AKI le ft ading to accelerated kidney ageing and CKD. Nat Rev Nephrol 11, 264–276, https://doi.org/10.1038/nrneph.2015.3 (2015). 2. McCullough, P. A. Contrast-induced acute kidney injury. J Am Coll Cardiol 51, 1419–1428, https://doi.org/10.1016/j. jacc.2007.12.035 (2008). 3. Huang, P. H. et al. Pioglitazone ameliorates endothelial dysfunction and restores ischemia-induced angiogenesis in diabetic mice. Biomed Pharmacother 62, 46–52, https://doi.org/10.1016/j.biopha.2007.06.014 (2008). 4. Falanga, V. Wound healing and its impairment in the diabetic foot. Lancet 366, 1736–1743, https://doi.org/10.1016/S0140- 6736(05)67700-8 (2005). 5. Rosner, M. H. & Okusa, M. D. Acute kidney injury associated with cardiac surgery. Clinical journal of the American Society of Nephrology 1, 19–32 (2006). 6. Weisberg, L. S., Kurnik, P. B. & Kurnik, B. R. Risk of radiocontrast nephropathy in patients with and without diabetes mellitus. Kidney international 45, 259–265 (1994). 7. Venkatachalam, M. A., Weinberg, J. M., Kriz, W. & Bidani, A. K. Failed Tubule Recovery, AKI-CKD Transition, and Kidney Disease Progression. J Am Soc Nephrol 26, 1765–1776, https://doi.org/10.1681/ASN.2015010006 (2015). 8. Zhou, D. et al. Sonic hedgehog is a novel tubule-derived growth factor for interstitial fibroblasts aer k ft idney injury. Journal of the American Society of Nephrology, ASN. 2013080893 (2014). 9. Pohlers, D. et al. TGF-beta and fibrosis in different organs - molecular pathway imprints. Biochim Biophys Acta 1792, 746–756, https://doi.org/10.1016/j.bbadis.2009.06.004 (2009). 10. Asai, J. et al. Topical sonic hedgehog gene therapy accelerates wound healing in diabetes by enhancing endothelial progenitor cell- mediated microvascular remodeling. Circulation 113, 2413–2424, https://doi.org/10.1161/CIRCULATIONAHA.105.603167 (2006). 11. Bielefeld, K. A., Amini-Nik, S. & Alman, B. A. Cutaneous wound healing: recruiting developmental pathways for regeneration. Cell Mol Life Sci 70, 2059–2081, https://doi.org/10.1007/s00018-012-1152-9 (2013). 12. LeBleu, V. S. et al. Origin and function of myofibroblasts in kidney fibrosis. Nat Med 19, 1047–1053, https://doi.org/10.1038/ nm.3218 (2013). 13. Liu, Y. Cellular and molecular mechanisms of renal fibrosis. Nat Rev Nephrol 7, 684–696, https://doi.org/10.1038/nrneph.2011.149 (2011). 14. Fernandez, I. E. & Eickelberg, O. New cellular and molecular mechanisms of lung injury and fibrosis in idiopathic pulmonary b fi rosis. Lancet 380, 680–688, https://doi.org/10.1016/S0140-6736(12)61144-1 (2012). 15. Zhou, D. et al. Sonic hedgehog is a novel tubule-derived growth factor for interstitial fibroblasts aer k ft idney injury. J Am Soc Nephrol 25, 2187–2200, https://doi.org/10.1681/ASN.2013080893 (2014). 16. Spurgeon, K. R., Donohoe, D. L. & Basile, D. P. Transforming growth factor-beta in acute renal failure: receptor expression, effects on proliferation, cellularity, and vascularization aer r ft ecovery from injury. Am J Physiol Renal Physiol 288, F568–577, https://doi. org/10.1152/ajprenal.00330.2004 (2005). 17. Connolly, S. B., Sadlier, D., Kieran, N. E., Doran, P. & Brady, H. R. Transcriptome profiling and the pathogenesis of diabetic complications. J Am Soc Nephrol 14, S279–283 (2003). SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 11 www.nature.com/scientificreports/ 18. Reeves, W. B. & Andreoli, T. E. Transforming growth factor beta contributes to progressive diabetic nephropathy. Proc Natl Acad Sci USA 97, 7667–7669 (2000). 19. Brem, H. & Tomic-Canic, M. Cellular and molecular basis of wound healing in diabetes. J Clin Invest 117, 1219–1222, https://doi. org/10.1172/JCI32169 (2007). 20. Brosius, F. C. III. New insights into the mechanisms of fibrosis and sclerosis in diabetic nephropathy. Rev Endocr Metab Disord 9, 245–254, https://doi.org/10.1007/s11154-008-9100-6 (2008). 21. Le Clef, N., Verhulst, A., D’Haese, P. C. & Vervaet, B. A. Unilateral renal ischemia-reperfusion as a robust model for acute to chronic kidney injury in mice. PloS one 11, e0152153 (2016). 22. Jang, H. S., Kim, J. I., Han, S. J. & Park, K. M. Recruitment and subsequent proliferation of bone marrow-derived cells in the postischemic kidney are important to the progression of fibrosis. Am J Physiol Renal Physiol 306, F1451–1461, https://doi. org/10.1152/ajprenal.00017.2014 (2014). 23. Moon, J. Y. et al. Aberrant recruitment and activation of T cells in diabetic nephropathy. Am J Nephrol 35, 164–174, https://doi. org/10.1159/000334928 (2012). 24. Jha, V. et al. Chronic kidney disease: global dimension and perspectives. Lancet 382, 260–272, https://doi.org/10.1016/S0140- 6736(13)60687-X (2013). 25. Ritz, E. & Orth, S. R. Nephropathy in patients with type 2 diabetes mellitus. N Engl J Med 341, 1127–1133, https://doi.org/10.1056/ NEJM199910073411506 (1999). 26. Thakar, C. V., Christianson, A., Himmelfarb, J. & Leonard, A. C. Acute kidney injury episodes and chronic kidney disease risk in diabetes mellitus. Clin J Am Soc Nephrol 6, 2567–2572, https://doi.org/10.2215/CJN.01120211 (2011). 27. Galkina, E. & Ley, K. Leukocyte recruitment and vascular injury in diabetic nephropathy. J Am Soc Nephrol 17, 368–377, https://doi. org/10.1681/ASN.2005080859 (2006). 28. Hills, C. E. & Squires, P. E. TGF-beta1-induced epithelial-to-mesenchymal transition and therapeutic intervention in diabetic nephropathy. Am J Nephrol 31, 68–74, https://doi.org/10.1159/000256659 (2010). 29. Devarajan, P. Update on mechanisms of ischemic acute kidney injury. J Am Soc Nephrol 17, 1503–1520, https://doi.org/10.1681/ ASN.2006010017 (2006). 30. Kinsey, G. R., Li, L. & Okusa, M. D. Inflammation in acute kidney injury. Nephron Exp Nephrol 109, e102–107, https://doi. org/10.1159/000142934 (2008). 31. Jang, H. R. et al. B cells limit repair after ischemic acute kidney injury. J Am Soc Nephrol 21, 654–665, https://doi.org/10.1681/ ASN.2009020182 (2010). 32. Tapmeier, T. T. et al. Pivotal role of CD4+ T cells in renal fibrosis following ureteric obstruction. Kidney Int 78, 351–362, https://doi. org/10.1038/ki.2010.177 (2010). 33. Hui Han, J. Z. et al. and Ruiyan Zhang. Renal recruitment of B lymphocytes exacerbates tubulointerstitial fibrosis by promoting monocyte mobilization and infiltration aer uni ft lateral ureteral obstruction. Journal of Pathology 241, 80–90 (2017). 34. Chen, H., Li, D., Saldeen, T. & Mehta, J. L. TGF-beta 1 attenuates myocardial ischemia-reperfusion injury via inhibition of upregulation of MMP-1. Am J Physiol Heart Circ Physiol 284, H1612–1617, https://doi.org/10.1152/ajpheart.00992.2002 (2003). 35. Ishibe, S. & Cantley, L. G. Epithelial–mesenchymal–epithelial cycling in kidney repair. Current opinion in nephrology and hypertension 17, 379–385 (2008). 36. Fan, J. M. et al. Transforming growth factor-beta regulates tubular epithelial-myob fi roblast transdie ff rentiation in vitro . Kidney Int 56, 1455–1467, https://doi.org/10.1046/j.1523-1755.1999.00656.x (1999). 37. Meran, S. & Steadman, R. Fibroblasts and myofibroblasts in renal fibrosis. Int J Exp Pathol 92, 158–167, https://doi.org/10.1111/ j.1365-2613.2011.00764.x (2011). 38. Amankulor, N. M. et al. Sonic hedgehog pathway activation is induced by acute brain injury and regulated by injury-related inflammation. J Neurosci 29, 10299–10308, https://doi.org/10.1523/JNEUROSCI.2500-09.2009 (2009). 39. Stewart, G. A. et al. Expression of the developmental Sonic hedgehog (Shh) signalling pathway is up-regulated in chronic lung fibrosis and the Shh receptor patched 1 is present in circulating T lymphocytes. J Pathol 199, 488–495, https://doi.org/10.1002/ path.1295 (2003). 40. Horn, A. et al. Hedgehog signaling controls fibroblast activation and tissue fibrosis in systemic sclerosis. Arthritis & Rheumatism 64, 2724–2733 (2012). 41. Ding, H. et al. Sonic hedgehog signaling mediates epithelial-mesenchymal communication and promotes renal fibrosis. J Am Soc Nephrol 23, 801–813, https://doi.org/10.1681/ASN.2011060614 (2012). 42. Fabian, S. L. et al. Hedgehog-Gli pathway activation during kidney fibrosis. Am J Pathol 180, 1441–1453, https://doi.org/10.1016/j. ajpath.2011.12.039 (2012). 43. Bai, Y. et al. Sonic hedgehog-mediated epithelial-mesenchymal transition in renal tubulointerstitial fibrosis. International journal of molecular medicine 37, 1317–1327 (2016). 44. Chung, Y. & Fu, E. Crosstalk between Shh and TGF-beta signaling in cyclosporine-enhanced cell proliferation in human gingival b fi roblasts. PLoS One 8, e70128, https://doi.org/10.1371/journal.pone.0070128 (2013). 45. Kramann, R., DiRocco, D. P. & Humphreys, B. D. Understanding the origin, activation and regulation of matrix-producing myob fi roblasts for treatment of fibrotic disease. J Pathol 231, 273–289, https://doi.org/10.1002/path.4253 (2013). 46. Lu, H., Chen, B., Hong, W., Liang, Y. & Bai, Y. Transforming growth factor-beta1 stimulates hedgehog signaling to promote epithelial-mesenchymal transition aer k ft idney injury. FEBS J 283, 3771–3790, https://doi.org/10.1111/febs.13842 (2016). 47. Morrisey, K., Evans, R. A., Wakefield, L. & Phillips, A. O. Translational regulation of renal proximal tubular epithelial cell transforming growth factor-β1 generation by insulin. e A Th merican journal of pathology 159, 1905–1915 (2001). 48. Nakamura, N., Matsui, T., Ishibashi, Y. & Yamagishi, S.-i Insulin stimulates SGLT2-mediated tubular glucose absorption via oxidative stress generation. Diabetology & metabolic syndrome 7, 48 (2015). 49. Lee, S. Y. et al. PGC1alpha Activators Mitigate Diabetic Tubulopathy by Improving Mitochondrial Dynamics and Quality Control. J Diabetes Res 2017, 6483572, https://doi.org/10.1155/2017/6483572 (2017). Acknowledgements This research was supported by a grant from the National Research Foundation of Korea in 2012 (NRF- 2012M3A9C6050511). Author Contributions D.J.K. designed and performed all of the experiments. J.M.K. designed in vitro cell culture and performed western blot analysis of in vivo and in vitro experiments. S.H.P. and H.M. performed all of the histological analysis. H.K.K. and S.J.S. performed confocal microscopy. S.M.K. and J.W.S. performed diabetes experiments. Y.H.L., Y.G.K. and J.Y.M. performed IRI model. All authors participated in interpreting the results. D.J.K., Y.G.K., S.Y.L. and S.H.L. wrote the manuscript. S.H.L. and Y.S. coordinated the project. SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 12 www.nature.com/scientificreports/ Additional Information Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-017-16977-z. Competing Interests: The authors declare that they have no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2017 SCIENtIfIC REPO R TS | 7: 16782 | DOI:10.1038/s41598-017-16977-z 13

Journal

Scientific ReportsSpringer Journals

Published: Dec 1, 2017

There are no references for this article.