- Email: [email protected]
Changes in snail and SRF expression in the kidneys of diabetic rats during ageing
Acta Histochemica xxx (xxxx) xxxx
Contents lists available at ScienceDirect
Acta Histochemica journal homepage: www.elsevier.com/locate/acthis
Changes in snail and SRF expression in the kidneys of diabetic rats during ageing ⁎
Sandra Kostica, , Brandon Williamsb,1, Samy Ksouric,1, Leon Hardungc,1, Natalija Filipovicd, Lejla Ferhatovic Hamzice, Livia Puljake,2, Nasrollah Ghahramanib, Katarina Vukojevicc a
Laboratory for Microscopy, Department of Anatomy, Histology and Embryology, University of Split School of Medicine, Soltanska 2, 21000 Split, Croatia Penn State College of Medicine Division of Nephrology, Hershey, PA 17033, United States c Laboratory for Early Human Development, Department of Anatomy, Histology and Embryology, University of Split School of Medicine, Soltanska 2, 21000 Split, Croatia d Laboratory for Neurocardiology, Department of Anatomy, Histology and Embryology, University of Split School of Medicine, Soltanska 2, 21000 Split, Croatia e Laboratory for Pain Research, Department of Anatomy, Histology and Embryology, University of Split School of Medicine, Soltanska 2, 21000 Split, Croatia b
A R T I C LE I N FO
A B S T R A C T
Keywords: Snail SRF Diabetic nephropathy Renal fibrosis
Background: Diabetic nephropathy is a progressive condition which develops for many years. We analyzed expression of Snail and serum response factor (SRF), epithelial-mesenchymal transition (EMT) regulatory transcription factors with a key role in renal fibrosis, in different renal areas of diabetic rats during ageing. Methods: Male Sprague–Dawley rats were treated with 55 mg/kg streptozotocin (model of type 1 diabetes mellitus; DM group) or citrate buffer (control). DM group received insulin weekly to prevent ketoacidosis. After 2 weeks, 2, 6 and 12 months kidney samples were collected and analysed in different renal areas. Results: Snail expression was located within cortex in proximal convoluted tubules, in control and DM groups, in the cytoplasm. Percentage of Snail-positive cells in control groups was high and decreased with time, whereas in DM groups the highest percentage was after 2 weeks. In all time points, smaller percentage of Snail expression was seen in DM groups compared to controls. SRF expression was mostly located in the proximal convoluted tubules, always in the cytoplasm. In control groups SRF was expressed in all time periods in proximal convoluted tubules, with decrement after 12 months. Percentage of SRF-positive cells was higher in control groups compared to DM in all time points, with the exception of 12 months. To a smaller degree, SRF expression was seen in the glomeruli and distal convoluted tubules, with more SRF positive cells in DM compared to their control groups. Conclusions: While Snail expression remained lower in diabetic tissues, compared to controls, expression of SRF increased in diabetic tissues in the second part of the year. These changes may need long time to develop, and, in line with earlier reports, it is possible that insulin treatment of DM rats once a week reduces possibility of EMT and development of renal fibrosis even in the long term.
1. Introduction According to the World Health Organization (WHO) projections, diabetes (DM) will be the seventh leading cause of death in 2030 (Mathers and Loncar, 2006). Among the common diabetic complications is diabetic nephropathy (DN) (Decleves and Sharma, 2010), a slowly progressive condition which develops over time for many years. Studies have implicated genetic components of the DN, as well as oxidative stress as an important mechanism relating to major pathways responsible for diabetic damage (Makuc and Petrovic, 2011). DN is a
complex syndrome that includes the loss of podocytes, hypertrophy of mesangial cells, glomerular basement membrane thickening, and renal fibrosis (Brosius and Banes-Berceli, 2010; Steffes et al., 1989; Ziyadeh, 1993). Loss of glomerular filtration function occurs relatively late from the onset of disease, culminating in the end-stage renal disease. It is possible that in early DN proximal convoluted tubule dysfunction may precede glomerular injury (Petrica et al., 2011a, b). The patients with early DN may display more abnormal tubular injury markers as signs of proximal convoluted tubule dysfunction (Zhang et al., 2015). Advanced
⁎
Corresponding author. E-mail address: [email protected] (S. Kostic). 1 These authors contributed equally. 2 Present address: Center for Evidence-Based Medicine and Health Care, Catholic University of Croatia, Ilica 10000 Zagreb, Croatia. https://doi.org/10.1016/j.acthis.2019.151460 Received 4 June 2019; Received in revised form 10 October 2019; Accepted 10 October 2019 0065-1281/ © 2019 Elsevier GmbH. All rights reserved.
Please cite this article as: Sandra Kostic, et al., Acta Histochemica, https://doi.org/10.1016/j.acthis.2019.151460
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
renal fibrosis is considered to be the result of the epithelial-mesenchymal transition (EMT) process (Badid et al., 2002; Valcourt et al., 2005), however, the molecular mechanisms by which this chronic fibrosis develops is yet to be fully understood. Serum response factor (SRF), a type of transcription factor, has been shown to promote EMT in various cell types including tubular epithelial cells (Miyata et al., 2008). It was reported that SRF has an essential role in tumour progression, especially in the EMT and metastasis of epithelial tumour cells (Miano, 2010; Zhao et al., 2014). The SRF is transferred from the cytoplasm to the nucleus to upregulate Snail expression. This subcellular localization of Snail and SRF influences the phenotype and function of the affected cells (He et al., 2014; Fang et al., 2008). Snail has been marked as an important EMT regulatory transcription factor with a key role in renal fibrosis (Harney et al., 2012). The primary mechanism by which Snail mediates podocyte EMT is by supressing the expression of epithelial proteins, P-cadherin, nephrin and synaptopodin (Kwon et al., 2016; Li and Siragy, 2014). Synaptopodin is an actin-associated podocyte protein which determines the specific micro-filaments’ arrangement in the foot processes, and is considered a podocyte marker (Kwon et al., 2016; Petermann et al., 2004). Additionally, alpha smooth muscle actin (aSMA) was reported in increased amounts in the renal structures of the diabetic animals (Sohn et al., 2018). Since the transition of epithelial cells to cells with a mesenchymal phenotype is considered to be one of the crucial causes of chronic renal fibrosis in DN, we aimed to investigate the changes of EMT marker Snail and SRF, together with synaptopodin and aSMA, in the cytoplasm and nuclei of the renal cortex structures of ageing diabetic rats, which could help elucidate the mechanisms underlying the progression of type I DM. To the best of our knowledge, this is the first long term study that quantifies the expression of Snail and SRF markers in the kidney structures of diabetic rats during ageing.
same time period. Six animals were assigned to each of the four control (C-2 w, C-2 m, C-6 m, C-12 m) and four experimental groups (DM-2 w, DM-2 m, DM-6 m, DM-12 m). In order to verify diabetes within the rat models, body mass and plasma glucose were measured regularly. One drop of blood was collected from the tail vein, and a blood glucose level was determined by single touch glucometer (OneTouchVITa, LifeScan, High Wycombe, UK). Rats were considered diabetic with a glucose level above 16.5 mmol/L, and were therefore included in further experiments. The rats that developed DM received 1 unit of long-acting insulin glargine (Lantus Solostar, Sanofi-Aventis Deutschland GmbH, Frankfurt am Main, Germany) weekly in order to prevent ketoacidosis. Glucose level was measured just before the administration of the insulin and values were between 16 −30 mmol/L. 2.3. Tissue collection and immunohistochemistry Isoflurane (Forane, Abbott Laboratories, Queenborough, UK) was used to anesthetize the rats. The rats were perfused with 300 ml of saline immediately followed by 300 mL of Zamboni’s fixative (4% paraformaldehyde and 15% picric acid in 0.1 M phosphate-buffered saline). Kidneys were collected and post-fixed in the same fixative. After washing, transverse cuts were made. The kidney tissue was embedded in paraffin blocks and cut into 7 μm thick sections, as described previously (Delic Jukic et al., 2018; Racetin et al., 2019). Briefly, following deparaffinization, ethanol and water were used to rehydrate the samples. Rinsing of the tissue with distilled water was followed by heating in sodium citrate buffer (pH 6.0) on 95 °C for 12 min in a microwave oven. After cooling to the room temperature, the sections were incubated with primary antibodies: goat anti-Snail (ab53519) (1:500), rabbit anti-SRF (ab53147) (1: 500), rabbit anti-synaptopodin (ab117702) (1:500), mouse anti-aSMA (M0851, Dako, Denmark) (1:150). All antibodies were diluted in Dako REAL antibody diluent. Sections were then kept overnight in a humidified chamber at room temperature. DAPI was used as a nuclear marker. Sections were rinsed with PBS and incubated with secondary antibodies for one hour in a humidified chamber: donkey anti-goat IgG H&L (Alexa Fluor® 594) (ab150132), donkey anti-rabbit IgG H&L (Alexa Fluor® 488) (ab150073) and goat anti-mouse Rhodamine (1:300, AP124R, Jackson Immuno Research Lab., PA, USA). The sections were examined under a BX51 microscope (Olympus, Tokyo, Japan) equipped with a DP71digital camera (Olympus, Tokyo, Japan). Images were taken using CellA Imaging Software for Life Sciences Microscopy (Olympus Tokyo, Japan). Kidney sections were analyzed focusing on the cytoplasms and nuclei of the cells in three different areas as described in our previous work (Delic Jukic et al., 2018): glomerulus, proximal convoluted tubule and distal convoluted tubule. For each of the areas, 5 non overlapping fields were captured for analysis using 40× objective magnification, each field representing one image. Microphotographs were examined using ImageJ software (National Institutes of Health, Bethesda, MD, USA). The number of positive cells was counted, as determined by the intensity of fluorescence of the renal tissue. The percentage of positive cells within each area (glomerulus, proximal convoluted tubule and distal convoluted tubule) were compared between the experimental (diabetic) and control groups. The cell counting was done in the scattered non-overlapping fields. In order to calculate the percentage of fluorescent positive cells the number of calculated cells differ between N = 112–134, as previously described (Rancic et al., 2017). Distinct analyses were conducted for the three sections at each time point, and then the data were aggregated for all areas of the control and diabetic rats and afterwards evaluated.
2. Materials and methods 2.1. Ethics The experimental protocol was approved by the University of Split School of Medicine Ethics Committee, no. 16/4-09. 2.2. Experimental animals In this study 48 male Sprague–Dawley rats weighing 160–200 g were used. All rats were held under controlled conditions (temperature: 22.1 °C and 12/12 h of light schedule) at Animal Facility of University of Split. The sample size was calculated using G*Power software (HeinrichHeine-University, Düsseldorf, Germany) and the effect size obtained was d’ = 2.3. As a result, four animals per group were needed in order to obtain a (1-β) 75% power at the (α) 5% significance. However, we used 6 animals per group in order to obtain a (1-β) 80% power at the (α) 5% significance. For the induction of model of type I DM (DM group), rats received intraperitoneal injection of 55 mg/kg of the streptozotocin (STZ), a compound toxic to the insulin-producing pancreatic β cells (Weir et al., 1981), dissolved in citrate buffer (pH 4.5) after overnight fasting (Ferhatovic et al., 2013). The control (C) group received vehicle citrate buffer injection. The rats were fed ad libitum with normal pellet diet, made up of 27% proteins, 9% fat, and 64% of carbohydrates (4RF21 GLP, Mucedola srl, Settimo Milanese, Italy). DM rats were assigned into four groups based on the duration of diabetes, as measured from the point of injection to the termination of the experiment (2 weeks, 2 months, 6 months, 12 months). Each experimental group was matched with a control group and kept over the
2.4. Statistics Results were expressed as means ± SD. Statistical comparisons 2
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
Fig. 1. Transversal section through the renal cortex at 2 weeks in control (A) and diabetes (B); at 2 months in control (C) and diabetes (D). Serum response factor (SRF) and Snail positive cells (arrows) were seen as green or red staining of cytoplasm within different areas of the cortex of kidneys. Synaptopodin positive cells (arrows) were seen as green staining of cytoplasm within kidney cortex. Alpha smooth muscle actin (aSMA) positive cells (arrows) were seen as red staining in the blood vessels (bv). Co-localization of Snail and SRF were shown by arrowheads in the far right column (merge); dt- distal convoluted tubule; g- glomerulus; ptproximal convoluted tubule; C- control; DM- diabetes mellitus type 1. Scale bar 25 μm.
Strong Snail expression intensity was also seen in the proximal convoluted tubules in all control groups, in decreasing pattern with each subsequent time point. In the C-2 w, C-2 m, C-6 m groups the expression intensity was strong, while in C-12 m group the intensity of staining was moderate. In DM groups, Snail staining intensity was mostly mild, except in 6 months’ group when it was moderate (data not shown). SRF expression was mostly located within the cortex, in the proximal convoluted tubules and always in the cytoplasm in both DM and control animals. In the control groups, SRF was expressed in all time periods in proximal convoluted tubules, however, significantly lower percentage of positive cells was seen after 12 months. A significant difference was seen in the total number of SRF positive cells between all the DM groups during time except DM-2 w, DM-6 m, with DM-2 m expression being the lowest of the DM groups (p < 0.001). There was a significant difference in the total percentage of SRF positive cells between DM and control groups in all time periods, with more positive cells in the control groups with the exception of 12 months’ groups, where there were more SRF positive cells in DM group (Fig. 3B). The SRF expression was also seen in the glomeruli and distal convoluted tubules in both control and DM groups, although to a smaller degree. In the control groups, SRF was expressed only in the 6 months group within the glomeruli. A significant difference was seen in the total percentage of SRF positive cells between all DM groups during
were analysed by Student’s t-test for individual control vs. diabetic group. One-way ANOVA test for multiple comparisons with Bonferroni correction were used for analysis of different time periods. Program GraphPad Prism version 5 was used (GraphPad Software, San Diego, CA). P < 0.001 was considered significant.
3. Results Snail and SRF staining shown in Figs. 1 and 2 was analysed between the control and diabetic groups during time periods of 2 weeks, 2, 6 and 12 months, in the kidney’s cortical structures of glomeruli, proximal and distal convoluted tubules. Snail expression was located within cortex in the proximal convoluted tubules, in both the control and diabetic groups, and always in the cytoplasm. Snail positive cells were not present in glomeruli or distal convoluted tubules. The percentage of Snail positive cells in the control groups was very high, around 90%, after two weeks, and it decreased with time in the cytoplasm of the proximal convoluted tubules cells (Figs. 1 and 2). A significant difference was seen in the total percentage of Snail positive cells in the diabetic groups during time, with highest percentage in the DM-2 w group, 55% (p < 0.001). There was a significant difference in the total percentage of Snail positive cells between the DM and control groups, in all time periods, with lower percentage of Snail positive cells in the DM groups compared to their respective controls (Fig. 3A). 3
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
Fig. 2. Transversal section through the renal cortex at 6 months in control (A) and diabetes (B); at 12 months in control (C) and diabetes (D). Snail positive cells (arrows) were seen as red staining of cytoplasm within different areas of the renal cortex. Serum response factor (SRF) and synaptopodin positive cells (arrows) were seen as green staining of cytoplasm within the renal cortex. Alpha smooth muscle actin (aSMA) positive cells (arrows) were seen as red staining in the blood vessels (bv). Co-localization of markers was not observed in the far right column (merge); bv- blood vessel; dt- distal convoluted tubule; g- glomerulus; pt- proximal convoluted tubule; C- control; DM- diabetes mellitus type 1. Scale bar 25 μm.
Renal interstitial fibrosis or tubulointerstitial fibrosis and glomerulosclerosis, with signs such as collagen deposition and reduction in the number of capillaries, are the characteristic representation of ongoing diabetic nephropathy. The signs of renal fibrosis were not observed in both groups of animals in our Mallory stained sections (data not shown). aSMA positive cells were seen only in the blood vessels in both groups (Figs. 1 and 2). The total number of synaptopodin positive cells in the glomeruli was analysed between the control and diabetic groups during all time periods (Figs. 1 and 2). In the control groups, synaptopodin was expressed in all time periods in a large percentage (data not shown). Significantly lower number of synaptopodin-positive cells was seen in diabetic groups in a decreasing pattern with highest value in 2 weeks’ group of 52%. A significant difference was seen in the total number of synaptopodin positive cells between all diabetic and control groups in all time periods (Fig. 4). Strong synaptopodin staining intensity was observed in all time periods in control glomeruli, while diabetic groups’ glomeruli showed moderate staining intensity.
time (p < 0.001) with peak expression after 2 months. There were no SRF positive cells in the 12 months’ groups for both DM and control. There was a significant difference in the total percentage of SRF positive cells between the 6 months DM and control groups with more positive cells in the DM groups compared to their respective controls (Fig. 3C). In the distal convoluted tubules, in the control groups SRF was expressed in all time periods except after 2 weeks. A significant difference was seen in the total percentage of SRF positive cells between all the diabetic groups during time except DM-2 w and DM-2 m; DM-6 m and DM-12 m (p < 0.001). There was a significant difference in the total percentage of SRF positive cells between DM and control groups in all time periods, with more positive cells in the DM group (Fig. 3D). Moderate SRF expression intensity can be seen in the proximal convoluted tubules in the control in 2 weeks and 6 months while mild intensity was observed in the 2 and 12 months groups. In diabetic groups, SRF staining intensity was mostly mild except in 6 months’ group when it was moderate (data not shown). In glomeruli, only 6 months control group had mild staining intensity. In the diabetic groups, mostly mild intensity was observed, except for moderate intensity at the 2 months group. In distal convoluted tubules, the intensity was predominantly mild for both diabetic and control groups, with the exception of the 2 week’s control where no staining intensity was observed.
4. Discussion In our study, Snail expression was located in the proximal convoluted tubules of both control and diabetic groups, whereas no expression was seen in the glomeruli or distal convoluted tubules. 4
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
Fig. 3. Percentage of Snail (A) and serum response factor (SRF) (B, C and D) positive cells in 2 weeks (2 w), 2 months (2 m), 6 months (6 m) and 12 months (12 m) of control (C) and diabetic (DM) rat groups within different areas of the renal cortex (glomeruli, proximal and distal convoluted tubules). The number of animals for each group was six (n = 6). Data presented as M ± SD, t-test. Asterisk denotes significant difference: p < 0.001.
Snail being either completely absent in the control or found in small quantity. This could be partially explained due to different methodology, i.e. Western blotting and RT-PCR, where all kidney structures were included or the research focused on glomeruli only. Snail needs to translocate to the nucleus in order to exert its function (Wu and Zhou, 2010). However, in our study, although present in cytoplasm, no Snail positivity was seen in nucleus of the control groups. Snail has an important role in proliferation and survival, especially during embryonic development, while knockdown of Snail expression increases apoptosis in colon cancer in animal models (Roy et al., 2004). The Snail expression enhances resistance to cell death provoked by DNA damage. Also, the role of Snail as an important factor in the preservation of stem cell function has been identified (Wu and Zhou, 2010). This could explain the existence of Snail in cytoplasm of proximal convoluted tubules, ready to translocate to nucleus to exert its roles in various functions carried out by proximal convoluted tubules. Snail-positive cells were observed in DM animals in all tested time points, with the highest percentage after 2 weeks, however with a smaller percentage as compared to their respective controls. In DM animals, as in controls, Snail positive cells were found only in the cytoplasm and not in nucleus. Additionally, the signs of renal fibrosis were not observed in our Mallory stained sections. Contrary to our results, others reported increment of Snail expression in the nucleus, specifically in the glomeruli (Gagliardini et al., 2013). Due to longevity of our experiment, the diabetic rats were given insulin once a week, which was not the case in most of previously reported literature. In line with our results, Fang et al. (Fang et al., 2008) showed a decrease of Snail expression in the nuclei in insulin treated group of diabetic animals in period of 6 months after onset of diabetes. Similarly, they recorded the Snail expression in the proximal convoluted tubules and not in glomeruli. Taking these results into the consideration, it could be implied that
Fig. 4. Percentage of synaptopodin positive cells in 2 weeks (2 w), 2 months (2 m), 6 months (6 m) and 12 months (12 m) of control (C) and diabetic (DM) rat groups within glomeruli. The number of animals for each group was six (n = 6). Data presented as M ± SD, t-test. Asterisk denotes significant difference: p < 0.001.
Surprisingly, the number of Snail positive cells in the control groups was very high, around 90%, after two weeks, and it decreased with time. However, no positivity was found in the nuclei, only in the cytoplasm. In the cytoplasm, Snail has a rather short half-life and it is a target for ubiquitin-mediated proteasome degradation (Wu and Zhou, 2010). In other similar studies, Snail has been regarded as an important EMT regulatory transcription factor responsible for renal fibrosis (Fang et al., 2008; Harney et al., 2012). However, the results varied, with 5
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
glucose and blood pressure control. Reducing microalbuminuria to normal urine albumin excretion is more likely at lower levels of microalbuminuria together with improved blood pressure, glucose and lipid control (Marshall, 2014a). In our study we did not use groups of DM animals that were not treated with insulin, and this could be considered a study limitation. For long-term experiments we administered slow-release insulin subcutaneously four days after induction of diabetes (1–2 U insulin/d). The purpose of insulin treatment was not to normalize glucose, but to moderately lower glucose levels so as to reduce insulin resistance. Studying these rats for a long time period without insulin treatment would be ethically questionable, so when designing this study, we had taken into the consideration long-term duration of diabetes in our study, and general health of animals. Furthermore, also by giving insulin we mimicked real-life conditions, as patients with type 1 diabetes mellitus receive insulin in usual clinical practice. Furthermore, it should be emphasized that we have used STZ model of DM, and STZ has been associated with proximal tubular toxicity. Therefore, altered expressions of Snail and SRF in proximal tubules could be attributable to hyperglycemia and STZ toxicity. However, in our study, the percentage of Snail-positive cells was higher in control groups, compared to DM animals in all time points, and the percentage of SRF-positive cells was higher in control than in DM animals in the first three time points, but not after 12 months. If only STZ was contributing to a higher SRF expression in DM animals, compared to controls, one year after diabetes induction, then the same effect would be expected in earlier time points as well. Nevertheless, it would be good to continue these studies in other diabetic models, such as Zucker fatty rats. In conclusion, although the heterogeneous histological changes can be seen throughout the literature, giving the insulin even once a week to the animals with experimental diabetes retains the almost normal structure, with minimal or no glomerular changes. This shows that early start of the treatment with insulin can have critical influence on the development of these changes, one of which is the activation of EMT and expression of Snail, a regulatory molecule with a wide range of biological functions and probably many more processes yet to be discovered.
insulin treatment of DM rats once a week reduces possibility of EMT and consequent development of renal fibrosis even in the long term. However, we can only make such inferences by comparison with results of other researchers, since in this study we did not have DM animals that did not receive insulin. Additionally, aSMA positive cells were seen only in the blood vessels in both groups. Sohn et al. found aSMA and fibrosis in glomeruli of 6 months old diabetic rats. However, the rats did not receive insulin (Sohn et al., 2018). On the contrary, Fang et al. (Fang et al., 2008) found aSMA only in the blood vessels of controls and insulin treated diabetic rats. Lu et al. reported increased aSMA in diabetic Wistar rats, but it was not clear in which kidney structure since Western blot was used. In their study, animals were also not treated with insulin (Lu et al., 2015). In our study, the podocyte marker synaptopodin was found in the glomeruli of both control and DM animals during all time points. In the control groups, it was expressed in large percentage during the whole experiment. Significantly lower number of positive cells was seen in diabetic groups in a decreasing pattern with highest value in the 2 weeks groups showing depletion of podocytes during time. This implies that physiological function is altered and impaired, even though histological changes were not observed up to one year of experiment. This is in accordance with the study on diabetic patients, where authors argued that it was possible that in early DM proximal tubule dysfunction may precede glomerular injury. (Milas et al., 2018). Zhou et al. stated that unlike epithelial tubular cells, podocytes undergo ‘physiologic EMT’ under normal conditions, through expression of vimentin and showing a pericyte-like morphology (Zhou and Liu, 2015). In our study, there was always Snail expression in the cytoplasm of the proximal convoluted tubules, but not in the glomeruli, and it never translocated into the nuclei. However, we did observe the decrement of synaptopodin positive cells, without any visible signs of fibrosis in the diabetic groups, which could imply that podocytes go through apoptosis, rather than EMT. This remains to be further investigated. In our study, the expression of transcription factor SRF was observed only in the cytoplasm in all three cortical structures, but mostly located within the proximal convoluted tubules of both the controls and diabetic groups. In the control groups, SRF was expressed in all time periods in proximal convoluted tubules in significant number, with decrement after 12 months. More SRF-positive cells were seen in the control than DM groups with the exception of the 12 months groups, where there were more SRF positive cells in DM group. These results imply a possible broader regulatory role of SRF within proximal convoluted tubules which is impaired during ageing, but also in diabetes. The SRF expression was also seen in the glomeruli and distal convoluted tubule, although in smaller numbers and only in the cytoplasm. In the control groups, SRF was expressed only in 6 months group in the glomeruli. The diabetic groups peak expression was seen after 2 months. There were no SRF positive cells in 12 months’ groups, for both DM and control. Zhao et al. found that SRF is activated and translocated into the nuclei of the cultured cells during EMT of podocytes stimulated by high glucose (Zhao et al., 2016). In their study, SRF was upregulated in DM group in a time-dependent manner, implying that SRF also participates in the podocyte injury induced by hyperglycaemia in vivo. Additionally, they found that SRF overexpression in podocytes activates Snail and facilitates dedifferentiation, migration and dysfunction of podocytes (Zhao et al., 2016). In our study, in distal convoluted tubules of the control groups, SRF was expressed in all time periods except after 2 weeks. More SRF positive cells were seen in the DM group in all time periods but still only in the cytoplasm. Although in our study we did not analyse biochemical markers of microalbuminuria or clinical proteinuria, in patients these events appear 5–10 years after the diagnosis of diabetes, and if untreated, it gradually increases over the next 10–15 years (Marshall, 2014b). The rate of progression of diabetic glomerulosclerosis has slowed in the past decades secondary to improving
CRediT authorship contribution statement Sandra Kostic: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing. Brandon Williams: Data curation, Formal analysis, Writing - review & editing. Samy Ksouri: Data curation, Formal analysis, Writing - review & editing. Leon Hardung: Data curation, Formal analysis, Writing - review & editing. Natalija Filipovic: Data curation, Formal analysis, Writing review & editing. Lejla Ferhatovic Hamzic: Data curation, Writing review & editing. Livia Puljak: Data curation, Writing - review & editing. Nasrollah Ghahramani: Data curation, Formal analysis, Writing - review & editing. Katarina Vukojevic: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing. Declaration of Competing Interest None. Acknowledgements The study was supported by University of Split School of Medicine Excellence Grants for prof. Sandra Kostic, Katarina Vukojevic and 6
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
Natalija Filipovic and by Penn State – University of Split Collaboration Development Fund 2017 for prof. Vukojevic. The animal kidney samples of 6 and 12 months were funded by the Croatian Foundation for Science (HRZZ) grant no. 02.05./28.
A., Ursoniu, S., Vlad, D., Petrica, L., 2018. Deregulated profiles of urinary microRNAs may explain podocyte injury and proximal tubule dysfunction in normoalbuminuric patients with type 2 diabetes mellitus. J. Investig. Med. 66 (4), 747–754. Miyata, K., Ohashi, N., Suzaki, Y., Katsurada, A., Kobori, H., 2008. Sequential activation of the reactive oxygen species/angiotensinogen/renin-angiotensin system axis in renal injury of type 2 diabetic rats. Clin. Exp. Pharmacol. Physiol. 35 (8), 922–927. Petermann, A.T., Pippin, J., Krofft, R., Blonski, M., Griffin, S., Durvasula, R., Shankland, S.J., 2004. Viable podocytes detach in experimental diabetic nephropathy: potential mechanism underlying glomerulosclerosis. Nephron Exp. Nephrol. 98 (4), e114–123. Petrica, L., Petrica, M., Vlad, A., Jianu, D.C., Gluhovschi, G., Ianculescu, C., Firescu, C., Dumitrascu, V., Giju, S., Gluhovschi, C., Bob, F., Gadalean, F., Ursoniu, S., Velciov, S., Bozdog, G., Milas, O., 2011a. Proximal tubule dysfunction is dissociated from endothelial dysfunction in normoalbuminuric patients with type 2 diabetes mellitus: a cross-sectional study. Nephron Clin. Pract. 118 (2), c155–164. Petrica, L., Vlad, A., Petrica, M., Jianu, C.D., Gluhovschi, G., Gadalean, F., Dumitrascu, V., Ianculescu, C., Firescu, C., Giju, S., Gluhovschi, C., Bob, F., Velciov, S., Bozdog, G., Milas, O., Marian, R., Ursoniu, S., 2011b. Pioglitazone delays proximal tubule dysfunction and improves cerebral vessel endothelial dysfunction in normoalbuminuric people with type 2 diabetes mellitus. Diabetes Res. Clin. Pract. 94 (1), 22–32. Racetin, A., Raguz, F., Durdov, M.G., Kunac, N., Saraga, M., Sanna-Cherchi, S., Soljic, V., Martinovic, V., Petricevic, J., Kostic, S., Mardesic, S., Tomas, S.Z., Kablar, B., Restovic, I., Lozic, M., Filipovic, N., Saraga-Babic, M., Vukojevic, K., 2019. Immunohistochemical expression pattern of RIP5, FGFR1, FGFR2 and HIP2 in the normal human kidney development. Acta Histochem. 121 (5), 531–538. Rancic, A., Filipovic, N., Marin Lovric, J., Mardesic, S., Saraga-Babic, M., Vukojevic, K., 2017. Neuronal differentiation in the early human retinogenesis. Acta Histochem. 119 (3), 264–272. Roy, H.K., Iversen, P., Hart, J., Liu, Y., Koetsier, J.L., Kim, Y., Kunte, D.P., Madugula, M., Backman, V., Wali, R.K., 2004. Down-regulation of SNAIL suppresses MIN mouse tumorigenesis: modulation of apoptosis, proliferation, and fractal dimension. Mol. Cancer Ther. 3 (9), 1159–1165. Sohn, E., Kim, J., Kim, C.S., Jo, K., Kim, J.S., 2018. Osteomeles schwerinae extract prevents diabetes-induced renal injury in spontaneously diabetic torii rats. Evid. Complement. Alternat. Med. 2018, 6824215. Steffes, M.W., Osterby, R., Chavers, B., Mauer, S.M., 1989. Mesangial expansion as a central mechanism for loss of kidney function in diabetic patients. Diabetes 38 (9), 1077–1081. Valcourt, U., Kowanetz, M., Niimi, H., Heldin, C.H., Moustakas, A., 2005. TGF-beta and the Smad signaling pathway support transcriptomic reprogramming during epithelial-mesenchymal cell transition. Mol. Biol. Cell 16 (4), 1987–2002. Weir, G.C., Clore, E.T., Zmachinski, C.J., Bonner-Weir, S., 1981. Islet secretion in a new experimental model for non-insulin-dependent diabetes. Diabetes 30 (7), 590–595. Wu, Y., Zhou, B.P., 2010. Snail: more than EMT. Cell Adh. Migr. 4 (2), 199–203. Zhang, Y., Yang, J., Zheng, M., Wang, Y., Ren, H., Xu, Y., Yang, Y., Cheng, J., Han, F., Yang, X., Chen, L., Shan, C., Chang, B., 2015. Clinical characteristics and predictive factors of subclinical diabetic nephropathy. Exp. Clin. Endocrinol. Diabetes 123 (2), 132–138. Zhao, L., Wang, X., Sun, L., Nie, H., Liu, X., Chen, Z., Guan, G., 2016. Critical role of serum response factor in podocyte epithelial-mesenchymal transition of diabetic nephropathy. Diab. Vasc. Dis. Res. 13 (1), 81–92. Zhao, X., He, L., Li, T., Lu, Y., Miao, Y., Liang, S., Guo, H., Bai, M., Xie, H., Luo, G., Zhou, L., Shen, G., Guo, C., Bai, F., Sun, S., Wu, K., Nie, Y., Fan, D., 2014. SRF expedites metastasis and modulates the epithelial to mesenchymal transition by regulating miR-199a-5p expression in human gastric cancer. Cell Death Differ. 21 (12), 1900–1913. Zhou, L., Liu, Y., 2015. Wnt/beta-catenin signalling and podocyte dysfunction in proteinuric kidney disease. Nat. Rev. Nephrol. 11 (9), 535–545. Ziyadeh, F.N., 1993. Renal tubular basement membrane and collagen type IV in diabetes mellitus. Kidney Int. 43 (1), 114–120.
References Badid, C., Desmouliere, A., Babici, D., Hadj-Aissa, A., McGregor, B., Lefrancois, N., Touraine, J.L., Laville, M., 2002. Interstitial expression of alpha-SMA: an early marker of chronic renal allograft dysfunction. Nephrol. Dial. Transplant 17 (11), 1993–1998. Brosius 3rd, F.C., Banes-Berceli, A., 2010. A new pair of SOCS for diabetic nephropathy. J. Am. Soc. Nephrol. 21 (5), 723–724. Decleves, A.E., Sharma, K., 2010. New pharmacological treatments for improving renal outcomes in diabetes. Nat. Rev. Nephrol. 6 (6), 371–380. Delic Jukic, I.K., Kostic, S., Filipovic, N., Gudelj Ensor, L., Ivandic, M., Dukic, J.J., Vitlov Uljevic, M., Ferhatovic Hamzic, L., Puljak, L., Vukojevic, K., 2018. Changes in expression of special AT-rich sequence binding protein 1 and phosphatase and tensin homologue in kidneys of diabetic rats during ageing. Nephrol. Dial. Transplant 33 (10), 1734–1741. Fang, K.Y., Lou, J.L., Xiao, Y., Shi, M.J., Gui, H.Z., Guo, B., Zhang, G.Z., 2008. [Transforming growth factor-beta1 and Snail1 mediate tubular epithelial-mesenchymal transition in diabetic rats]. Sheng Li Xue Bao 60 (1), 125–134. Ferhatovic, L., Banozic, A., Kostic, S., Kurir, T.T., Novak, A., Vrdoljak, L., Heffer, M., Sapunar, D., Puljak, L., 2013. Expression of calcium/calmodulin-dependent protein kinase II and pain-related behavior in rat models of type 1 and type 2 diabetes. Anesth. Analg. 116 (3), 712–721. Gagliardini, E., Perico, N., Rizzo, P., Buelli, S., Longaretti, L., Perico, L., Tomasoni, S., Zoja, C., Macconi, D., Morigi, M., Remuzzi, G., Benigni, A., 2013. Angiotensin II contributes to diabetic renal dysfunction in rodents and humans via Notch1/Snail pathway. Am. J. Pathol. 183 (1), 119–130. Harney, A.S., Meade, T.J., LaBonne, C., 2012. Targeted inactivation of Snail family EMT regulatory factors by a Co(III)-Ebox conjugate. PLoS One 7 (2), e32318. He, L., Lou, W., Ji, L., Liang, W., Zhou, M., Xu, G., Zhao, L., Huang, C., Li, R., Wang, H., Chen, X., Sun, S., 2014. Serum response factor accelerates the high glucose-induced Epithelial-to-Mesenchymal Transition (EMT) via snail signaling in human peritoneal mesothelial cells. PLoS One 9 (10), e108593. Kwon, S.K., Kim, S.J., Kim, H.Y., 2016. Urine synaptopodin excretion is an important marker of glomerular disease progression. Korean J. Intern. Med. 31 (5), 938–943. Li, C.X., Siragy, H.M., 2014. High glucose induces podocyte injury via enhanced (Pro) renin receptor-wnt-beta-Catenin-Snail signaling pathway. PLoS One 9 (2). Lu, Q., Zuo, W.Z., Ji, X.J., Zhou, Y.X., Liu, Y.Q., Yao, X.Q., Zhou, X.Y., Liu, Y.W., Zhang, F., Yin, X.X., 2015. Ethanolic Ginkgo biloba leaf extract prevents renal fibrosis through Akt/mTOR signaling in diabetic nephropathy. Phytomedicine 22 (12), 1071–1078. Makuc, J., Petrovic, D., 2011. A review of oxidative stress related genes and new antioxidant therapy in diabetic nephropathy. Cardiovasc. Hematol. Agents Med. Chem. 9 (4), 253–261. Marshall, S.M., 2014a. Natural history and clinical characteristics of CKD in type 1 and type 2 diabetes mellitus. Adv. Chronic Kidney Dis. 21 (3), 267–272. Marshall, S.M., 2014b. Preventing kidney failure in people with diabetes. Diabet. Med. 31 (11), 1280–1283. Mathers, C.D., Loncar, D., 2006. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med. 3 (11), e442. Miano, J.M., 2010. Role of serum response factor in the pathogenesis of disease. Lab. Invest. 90 (9), 1274–1284. Milas, O., Gadalean, F., Vlad, A., Dumitrascu, V., Gluhovschi, C., Gluhovschi, G., Velciov, S., Popescu, R., Bob, F., Matusz, P., Pusztai, A.M., Cretu, O.M., Secara, A., Simulescu,
7
Contents lists available at ScienceDirect
Acta Histochemica journal homepage: www.elsevier.com/locate/acthis
Changes in snail and SRF expression in the kidneys of diabetic rats during ageing ⁎
Sandra Kostica, , Brandon Williamsb,1, Samy Ksouric,1, Leon Hardungc,1, Natalija Filipovicd, Lejla Ferhatovic Hamzice, Livia Puljake,2, Nasrollah Ghahramanib, Katarina Vukojevicc a
Laboratory for Microscopy, Department of Anatomy, Histology and Embryology, University of Split School of Medicine, Soltanska 2, 21000 Split, Croatia Penn State College of Medicine Division of Nephrology, Hershey, PA 17033, United States c Laboratory for Early Human Development, Department of Anatomy, Histology and Embryology, University of Split School of Medicine, Soltanska 2, 21000 Split, Croatia d Laboratory for Neurocardiology, Department of Anatomy, Histology and Embryology, University of Split School of Medicine, Soltanska 2, 21000 Split, Croatia e Laboratory for Pain Research, Department of Anatomy, Histology and Embryology, University of Split School of Medicine, Soltanska 2, 21000 Split, Croatia b
A R T I C LE I N FO
A B S T R A C T
Keywords: Snail SRF Diabetic nephropathy Renal fibrosis
Background: Diabetic nephropathy is a progressive condition which develops for many years. We analyzed expression of Snail and serum response factor (SRF), epithelial-mesenchymal transition (EMT) regulatory transcription factors with a key role in renal fibrosis, in different renal areas of diabetic rats during ageing. Methods: Male Sprague–Dawley rats were treated with 55 mg/kg streptozotocin (model of type 1 diabetes mellitus; DM group) or citrate buffer (control). DM group received insulin weekly to prevent ketoacidosis. After 2 weeks, 2, 6 and 12 months kidney samples were collected and analysed in different renal areas. Results: Snail expression was located within cortex in proximal convoluted tubules, in control and DM groups, in the cytoplasm. Percentage of Snail-positive cells in control groups was high and decreased with time, whereas in DM groups the highest percentage was after 2 weeks. In all time points, smaller percentage of Snail expression was seen in DM groups compared to controls. SRF expression was mostly located in the proximal convoluted tubules, always in the cytoplasm. In control groups SRF was expressed in all time periods in proximal convoluted tubules, with decrement after 12 months. Percentage of SRF-positive cells was higher in control groups compared to DM in all time points, with the exception of 12 months. To a smaller degree, SRF expression was seen in the glomeruli and distal convoluted tubules, with more SRF positive cells in DM compared to their control groups. Conclusions: While Snail expression remained lower in diabetic tissues, compared to controls, expression of SRF increased in diabetic tissues in the second part of the year. These changes may need long time to develop, and, in line with earlier reports, it is possible that insulin treatment of DM rats once a week reduces possibility of EMT and development of renal fibrosis even in the long term.
1. Introduction According to the World Health Organization (WHO) projections, diabetes (DM) will be the seventh leading cause of death in 2030 (Mathers and Loncar, 2006). Among the common diabetic complications is diabetic nephropathy (DN) (Decleves and Sharma, 2010), a slowly progressive condition which develops over time for many years. Studies have implicated genetic components of the DN, as well as oxidative stress as an important mechanism relating to major pathways responsible for diabetic damage (Makuc and Petrovic, 2011). DN is a
complex syndrome that includes the loss of podocytes, hypertrophy of mesangial cells, glomerular basement membrane thickening, and renal fibrosis (Brosius and Banes-Berceli, 2010; Steffes et al., 1989; Ziyadeh, 1993). Loss of glomerular filtration function occurs relatively late from the onset of disease, culminating in the end-stage renal disease. It is possible that in early DN proximal convoluted tubule dysfunction may precede glomerular injury (Petrica et al., 2011a, b). The patients with early DN may display more abnormal tubular injury markers as signs of proximal convoluted tubule dysfunction (Zhang et al., 2015). Advanced
⁎
Corresponding author. E-mail address: [email protected] (S. Kostic). 1 These authors contributed equally. 2 Present address: Center for Evidence-Based Medicine and Health Care, Catholic University of Croatia, Ilica 10000 Zagreb, Croatia. https://doi.org/10.1016/j.acthis.2019.151460 Received 4 June 2019; Received in revised form 10 October 2019; Accepted 10 October 2019 0065-1281/ © 2019 Elsevier GmbH. All rights reserved.
Please cite this article as: Sandra Kostic, et al., Acta Histochemica, https://doi.org/10.1016/j.acthis.2019.151460
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
renal fibrosis is considered to be the result of the epithelial-mesenchymal transition (EMT) process (Badid et al., 2002; Valcourt et al., 2005), however, the molecular mechanisms by which this chronic fibrosis develops is yet to be fully understood. Serum response factor (SRF), a type of transcription factor, has been shown to promote EMT in various cell types including tubular epithelial cells (Miyata et al., 2008). It was reported that SRF has an essential role in tumour progression, especially in the EMT and metastasis of epithelial tumour cells (Miano, 2010; Zhao et al., 2014). The SRF is transferred from the cytoplasm to the nucleus to upregulate Snail expression. This subcellular localization of Snail and SRF influences the phenotype and function of the affected cells (He et al., 2014; Fang et al., 2008). Snail has been marked as an important EMT regulatory transcription factor with a key role in renal fibrosis (Harney et al., 2012). The primary mechanism by which Snail mediates podocyte EMT is by supressing the expression of epithelial proteins, P-cadherin, nephrin and synaptopodin (Kwon et al., 2016; Li and Siragy, 2014). Synaptopodin is an actin-associated podocyte protein which determines the specific micro-filaments’ arrangement in the foot processes, and is considered a podocyte marker (Kwon et al., 2016; Petermann et al., 2004). Additionally, alpha smooth muscle actin (aSMA) was reported in increased amounts in the renal structures of the diabetic animals (Sohn et al., 2018). Since the transition of epithelial cells to cells with a mesenchymal phenotype is considered to be one of the crucial causes of chronic renal fibrosis in DN, we aimed to investigate the changes of EMT marker Snail and SRF, together with synaptopodin and aSMA, in the cytoplasm and nuclei of the renal cortex structures of ageing diabetic rats, which could help elucidate the mechanisms underlying the progression of type I DM. To the best of our knowledge, this is the first long term study that quantifies the expression of Snail and SRF markers in the kidney structures of diabetic rats during ageing.
same time period. Six animals were assigned to each of the four control (C-2 w, C-2 m, C-6 m, C-12 m) and four experimental groups (DM-2 w, DM-2 m, DM-6 m, DM-12 m). In order to verify diabetes within the rat models, body mass and plasma glucose were measured regularly. One drop of blood was collected from the tail vein, and a blood glucose level was determined by single touch glucometer (OneTouchVITa, LifeScan, High Wycombe, UK). Rats were considered diabetic with a glucose level above 16.5 mmol/L, and were therefore included in further experiments. The rats that developed DM received 1 unit of long-acting insulin glargine (Lantus Solostar, Sanofi-Aventis Deutschland GmbH, Frankfurt am Main, Germany) weekly in order to prevent ketoacidosis. Glucose level was measured just before the administration of the insulin and values were between 16 −30 mmol/L. 2.3. Tissue collection and immunohistochemistry Isoflurane (Forane, Abbott Laboratories, Queenborough, UK) was used to anesthetize the rats. The rats were perfused with 300 ml of saline immediately followed by 300 mL of Zamboni’s fixative (4% paraformaldehyde and 15% picric acid in 0.1 M phosphate-buffered saline). Kidneys were collected and post-fixed in the same fixative. After washing, transverse cuts were made. The kidney tissue was embedded in paraffin blocks and cut into 7 μm thick sections, as described previously (Delic Jukic et al., 2018; Racetin et al., 2019). Briefly, following deparaffinization, ethanol and water were used to rehydrate the samples. Rinsing of the tissue with distilled water was followed by heating in sodium citrate buffer (pH 6.0) on 95 °C for 12 min in a microwave oven. After cooling to the room temperature, the sections were incubated with primary antibodies: goat anti-Snail (ab53519) (1:500), rabbit anti-SRF (ab53147) (1: 500), rabbit anti-synaptopodin (ab117702) (1:500), mouse anti-aSMA (M0851, Dako, Denmark) (1:150). All antibodies were diluted in Dako REAL antibody diluent. Sections were then kept overnight in a humidified chamber at room temperature. DAPI was used as a nuclear marker. Sections were rinsed with PBS and incubated with secondary antibodies for one hour in a humidified chamber: donkey anti-goat IgG H&L (Alexa Fluor® 594) (ab150132), donkey anti-rabbit IgG H&L (Alexa Fluor® 488) (ab150073) and goat anti-mouse Rhodamine (1:300, AP124R, Jackson Immuno Research Lab., PA, USA). The sections were examined under a BX51 microscope (Olympus, Tokyo, Japan) equipped with a DP71digital camera (Olympus, Tokyo, Japan). Images were taken using CellA Imaging Software for Life Sciences Microscopy (Olympus Tokyo, Japan). Kidney sections were analyzed focusing on the cytoplasms and nuclei of the cells in three different areas as described in our previous work (Delic Jukic et al., 2018): glomerulus, proximal convoluted tubule and distal convoluted tubule. For each of the areas, 5 non overlapping fields were captured for analysis using 40× objective magnification, each field representing one image. Microphotographs were examined using ImageJ software (National Institutes of Health, Bethesda, MD, USA). The number of positive cells was counted, as determined by the intensity of fluorescence of the renal tissue. The percentage of positive cells within each area (glomerulus, proximal convoluted tubule and distal convoluted tubule) were compared between the experimental (diabetic) and control groups. The cell counting was done in the scattered non-overlapping fields. In order to calculate the percentage of fluorescent positive cells the number of calculated cells differ between N = 112–134, as previously described (Rancic et al., 2017). Distinct analyses were conducted for the three sections at each time point, and then the data were aggregated for all areas of the control and diabetic rats and afterwards evaluated.
2. Materials and methods 2.1. Ethics The experimental protocol was approved by the University of Split School of Medicine Ethics Committee, no. 16/4-09. 2.2. Experimental animals In this study 48 male Sprague–Dawley rats weighing 160–200 g were used. All rats were held under controlled conditions (temperature: 22.1 °C and 12/12 h of light schedule) at Animal Facility of University of Split. The sample size was calculated using G*Power software (HeinrichHeine-University, Düsseldorf, Germany) and the effect size obtained was d’ = 2.3. As a result, four animals per group were needed in order to obtain a (1-β) 75% power at the (α) 5% significance. However, we used 6 animals per group in order to obtain a (1-β) 80% power at the (α) 5% significance. For the induction of model of type I DM (DM group), rats received intraperitoneal injection of 55 mg/kg of the streptozotocin (STZ), a compound toxic to the insulin-producing pancreatic β cells (Weir et al., 1981), dissolved in citrate buffer (pH 4.5) after overnight fasting (Ferhatovic et al., 2013). The control (C) group received vehicle citrate buffer injection. The rats were fed ad libitum with normal pellet diet, made up of 27% proteins, 9% fat, and 64% of carbohydrates (4RF21 GLP, Mucedola srl, Settimo Milanese, Italy). DM rats were assigned into four groups based on the duration of diabetes, as measured from the point of injection to the termination of the experiment (2 weeks, 2 months, 6 months, 12 months). Each experimental group was matched with a control group and kept over the
2.4. Statistics Results were expressed as means ± SD. Statistical comparisons 2
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
Fig. 1. Transversal section through the renal cortex at 2 weeks in control (A) and diabetes (B); at 2 months in control (C) and diabetes (D). Serum response factor (SRF) and Snail positive cells (arrows) were seen as green or red staining of cytoplasm within different areas of the cortex of kidneys. Synaptopodin positive cells (arrows) were seen as green staining of cytoplasm within kidney cortex. Alpha smooth muscle actin (aSMA) positive cells (arrows) were seen as red staining in the blood vessels (bv). Co-localization of Snail and SRF were shown by arrowheads in the far right column (merge); dt- distal convoluted tubule; g- glomerulus; ptproximal convoluted tubule; C- control; DM- diabetes mellitus type 1. Scale bar 25 μm.
Strong Snail expression intensity was also seen in the proximal convoluted tubules in all control groups, in decreasing pattern with each subsequent time point. In the C-2 w, C-2 m, C-6 m groups the expression intensity was strong, while in C-12 m group the intensity of staining was moderate. In DM groups, Snail staining intensity was mostly mild, except in 6 months’ group when it was moderate (data not shown). SRF expression was mostly located within the cortex, in the proximal convoluted tubules and always in the cytoplasm in both DM and control animals. In the control groups, SRF was expressed in all time periods in proximal convoluted tubules, however, significantly lower percentage of positive cells was seen after 12 months. A significant difference was seen in the total number of SRF positive cells between all the DM groups during time except DM-2 w, DM-6 m, with DM-2 m expression being the lowest of the DM groups (p < 0.001). There was a significant difference in the total percentage of SRF positive cells between DM and control groups in all time periods, with more positive cells in the control groups with the exception of 12 months’ groups, where there were more SRF positive cells in DM group (Fig. 3B). The SRF expression was also seen in the glomeruli and distal convoluted tubules in both control and DM groups, although to a smaller degree. In the control groups, SRF was expressed only in the 6 months group within the glomeruli. A significant difference was seen in the total percentage of SRF positive cells between all DM groups during
were analysed by Student’s t-test for individual control vs. diabetic group. One-way ANOVA test for multiple comparisons with Bonferroni correction were used for analysis of different time periods. Program GraphPad Prism version 5 was used (GraphPad Software, San Diego, CA). P < 0.001 was considered significant.
3. Results Snail and SRF staining shown in Figs. 1 and 2 was analysed between the control and diabetic groups during time periods of 2 weeks, 2, 6 and 12 months, in the kidney’s cortical structures of glomeruli, proximal and distal convoluted tubules. Snail expression was located within cortex in the proximal convoluted tubules, in both the control and diabetic groups, and always in the cytoplasm. Snail positive cells were not present in glomeruli or distal convoluted tubules. The percentage of Snail positive cells in the control groups was very high, around 90%, after two weeks, and it decreased with time in the cytoplasm of the proximal convoluted tubules cells (Figs. 1 and 2). A significant difference was seen in the total percentage of Snail positive cells in the diabetic groups during time, with highest percentage in the DM-2 w group, 55% (p < 0.001). There was a significant difference in the total percentage of Snail positive cells between the DM and control groups, in all time periods, with lower percentage of Snail positive cells in the DM groups compared to their respective controls (Fig. 3A). 3
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
Fig. 2. Transversal section through the renal cortex at 6 months in control (A) and diabetes (B); at 12 months in control (C) and diabetes (D). Snail positive cells (arrows) were seen as red staining of cytoplasm within different areas of the renal cortex. Serum response factor (SRF) and synaptopodin positive cells (arrows) were seen as green staining of cytoplasm within the renal cortex. Alpha smooth muscle actin (aSMA) positive cells (arrows) were seen as red staining in the blood vessels (bv). Co-localization of markers was not observed in the far right column (merge); bv- blood vessel; dt- distal convoluted tubule; g- glomerulus; pt- proximal convoluted tubule; C- control; DM- diabetes mellitus type 1. Scale bar 25 μm.
Renal interstitial fibrosis or tubulointerstitial fibrosis and glomerulosclerosis, with signs such as collagen deposition and reduction in the number of capillaries, are the characteristic representation of ongoing diabetic nephropathy. The signs of renal fibrosis were not observed in both groups of animals in our Mallory stained sections (data not shown). aSMA positive cells were seen only in the blood vessels in both groups (Figs. 1 and 2). The total number of synaptopodin positive cells in the glomeruli was analysed between the control and diabetic groups during all time periods (Figs. 1 and 2). In the control groups, synaptopodin was expressed in all time periods in a large percentage (data not shown). Significantly lower number of synaptopodin-positive cells was seen in diabetic groups in a decreasing pattern with highest value in 2 weeks’ group of 52%. A significant difference was seen in the total number of synaptopodin positive cells between all diabetic and control groups in all time periods (Fig. 4). Strong synaptopodin staining intensity was observed in all time periods in control glomeruli, while diabetic groups’ glomeruli showed moderate staining intensity.
time (p < 0.001) with peak expression after 2 months. There were no SRF positive cells in the 12 months’ groups for both DM and control. There was a significant difference in the total percentage of SRF positive cells between the 6 months DM and control groups with more positive cells in the DM groups compared to their respective controls (Fig. 3C). In the distal convoluted tubules, in the control groups SRF was expressed in all time periods except after 2 weeks. A significant difference was seen in the total percentage of SRF positive cells between all the diabetic groups during time except DM-2 w and DM-2 m; DM-6 m and DM-12 m (p < 0.001). There was a significant difference in the total percentage of SRF positive cells between DM and control groups in all time periods, with more positive cells in the DM group (Fig. 3D). Moderate SRF expression intensity can be seen in the proximal convoluted tubules in the control in 2 weeks and 6 months while mild intensity was observed in the 2 and 12 months groups. In diabetic groups, SRF staining intensity was mostly mild except in 6 months’ group when it was moderate (data not shown). In glomeruli, only 6 months control group had mild staining intensity. In the diabetic groups, mostly mild intensity was observed, except for moderate intensity at the 2 months group. In distal convoluted tubules, the intensity was predominantly mild for both diabetic and control groups, with the exception of the 2 week’s control where no staining intensity was observed.
4. Discussion In our study, Snail expression was located in the proximal convoluted tubules of both control and diabetic groups, whereas no expression was seen in the glomeruli or distal convoluted tubules. 4
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
Fig. 3. Percentage of Snail (A) and serum response factor (SRF) (B, C and D) positive cells in 2 weeks (2 w), 2 months (2 m), 6 months (6 m) and 12 months (12 m) of control (C) and diabetic (DM) rat groups within different areas of the renal cortex (glomeruli, proximal and distal convoluted tubules). The number of animals for each group was six (n = 6). Data presented as M ± SD, t-test. Asterisk denotes significant difference: p < 0.001.
Snail being either completely absent in the control or found in small quantity. This could be partially explained due to different methodology, i.e. Western blotting and RT-PCR, where all kidney structures were included or the research focused on glomeruli only. Snail needs to translocate to the nucleus in order to exert its function (Wu and Zhou, 2010). However, in our study, although present in cytoplasm, no Snail positivity was seen in nucleus of the control groups. Snail has an important role in proliferation and survival, especially during embryonic development, while knockdown of Snail expression increases apoptosis in colon cancer in animal models (Roy et al., 2004). The Snail expression enhances resistance to cell death provoked by DNA damage. Also, the role of Snail as an important factor in the preservation of stem cell function has been identified (Wu and Zhou, 2010). This could explain the existence of Snail in cytoplasm of proximal convoluted tubules, ready to translocate to nucleus to exert its roles in various functions carried out by proximal convoluted tubules. Snail-positive cells were observed in DM animals in all tested time points, with the highest percentage after 2 weeks, however with a smaller percentage as compared to their respective controls. In DM animals, as in controls, Snail positive cells were found only in the cytoplasm and not in nucleus. Additionally, the signs of renal fibrosis were not observed in our Mallory stained sections. Contrary to our results, others reported increment of Snail expression in the nucleus, specifically in the glomeruli (Gagliardini et al., 2013). Due to longevity of our experiment, the diabetic rats were given insulin once a week, which was not the case in most of previously reported literature. In line with our results, Fang et al. (Fang et al., 2008) showed a decrease of Snail expression in the nuclei in insulin treated group of diabetic animals in period of 6 months after onset of diabetes. Similarly, they recorded the Snail expression in the proximal convoluted tubules and not in glomeruli. Taking these results into the consideration, it could be implied that
Fig. 4. Percentage of synaptopodin positive cells in 2 weeks (2 w), 2 months (2 m), 6 months (6 m) and 12 months (12 m) of control (C) and diabetic (DM) rat groups within glomeruli. The number of animals for each group was six (n = 6). Data presented as M ± SD, t-test. Asterisk denotes significant difference: p < 0.001.
Surprisingly, the number of Snail positive cells in the control groups was very high, around 90%, after two weeks, and it decreased with time. However, no positivity was found in the nuclei, only in the cytoplasm. In the cytoplasm, Snail has a rather short half-life and it is a target for ubiquitin-mediated proteasome degradation (Wu and Zhou, 2010). In other similar studies, Snail has been regarded as an important EMT regulatory transcription factor responsible for renal fibrosis (Fang et al., 2008; Harney et al., 2012). However, the results varied, with 5
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
glucose and blood pressure control. Reducing microalbuminuria to normal urine albumin excretion is more likely at lower levels of microalbuminuria together with improved blood pressure, glucose and lipid control (Marshall, 2014a). In our study we did not use groups of DM animals that were not treated with insulin, and this could be considered a study limitation. For long-term experiments we administered slow-release insulin subcutaneously four days after induction of diabetes (1–2 U insulin/d). The purpose of insulin treatment was not to normalize glucose, but to moderately lower glucose levels so as to reduce insulin resistance. Studying these rats for a long time period without insulin treatment would be ethically questionable, so when designing this study, we had taken into the consideration long-term duration of diabetes in our study, and general health of animals. Furthermore, also by giving insulin we mimicked real-life conditions, as patients with type 1 diabetes mellitus receive insulin in usual clinical practice. Furthermore, it should be emphasized that we have used STZ model of DM, and STZ has been associated with proximal tubular toxicity. Therefore, altered expressions of Snail and SRF in proximal tubules could be attributable to hyperglycemia and STZ toxicity. However, in our study, the percentage of Snail-positive cells was higher in control groups, compared to DM animals in all time points, and the percentage of SRF-positive cells was higher in control than in DM animals in the first three time points, but not after 12 months. If only STZ was contributing to a higher SRF expression in DM animals, compared to controls, one year after diabetes induction, then the same effect would be expected in earlier time points as well. Nevertheless, it would be good to continue these studies in other diabetic models, such as Zucker fatty rats. In conclusion, although the heterogeneous histological changes can be seen throughout the literature, giving the insulin even once a week to the animals with experimental diabetes retains the almost normal structure, with minimal or no glomerular changes. This shows that early start of the treatment with insulin can have critical influence on the development of these changes, one of which is the activation of EMT and expression of Snail, a regulatory molecule with a wide range of biological functions and probably many more processes yet to be discovered.
insulin treatment of DM rats once a week reduces possibility of EMT and consequent development of renal fibrosis even in the long term. However, we can only make such inferences by comparison with results of other researchers, since in this study we did not have DM animals that did not receive insulin. Additionally, aSMA positive cells were seen only in the blood vessels in both groups. Sohn et al. found aSMA and fibrosis in glomeruli of 6 months old diabetic rats. However, the rats did not receive insulin (Sohn et al., 2018). On the contrary, Fang et al. (Fang et al., 2008) found aSMA only in the blood vessels of controls and insulin treated diabetic rats. Lu et al. reported increased aSMA in diabetic Wistar rats, but it was not clear in which kidney structure since Western blot was used. In their study, animals were also not treated with insulin (Lu et al., 2015). In our study, the podocyte marker synaptopodin was found in the glomeruli of both control and DM animals during all time points. In the control groups, it was expressed in large percentage during the whole experiment. Significantly lower number of positive cells was seen in diabetic groups in a decreasing pattern with highest value in the 2 weeks groups showing depletion of podocytes during time. This implies that physiological function is altered and impaired, even though histological changes were not observed up to one year of experiment. This is in accordance with the study on diabetic patients, where authors argued that it was possible that in early DM proximal tubule dysfunction may precede glomerular injury. (Milas et al., 2018). Zhou et al. stated that unlike epithelial tubular cells, podocytes undergo ‘physiologic EMT’ under normal conditions, through expression of vimentin and showing a pericyte-like morphology (Zhou and Liu, 2015). In our study, there was always Snail expression in the cytoplasm of the proximal convoluted tubules, but not in the glomeruli, and it never translocated into the nuclei. However, we did observe the decrement of synaptopodin positive cells, without any visible signs of fibrosis in the diabetic groups, which could imply that podocytes go through apoptosis, rather than EMT. This remains to be further investigated. In our study, the expression of transcription factor SRF was observed only in the cytoplasm in all three cortical structures, but mostly located within the proximal convoluted tubules of both the controls and diabetic groups. In the control groups, SRF was expressed in all time periods in proximal convoluted tubules in significant number, with decrement after 12 months. More SRF-positive cells were seen in the control than DM groups with the exception of the 12 months groups, where there were more SRF positive cells in DM group. These results imply a possible broader regulatory role of SRF within proximal convoluted tubules which is impaired during ageing, but also in diabetes. The SRF expression was also seen in the glomeruli and distal convoluted tubule, although in smaller numbers and only in the cytoplasm. In the control groups, SRF was expressed only in 6 months group in the glomeruli. The diabetic groups peak expression was seen after 2 months. There were no SRF positive cells in 12 months’ groups, for both DM and control. Zhao et al. found that SRF is activated and translocated into the nuclei of the cultured cells during EMT of podocytes stimulated by high glucose (Zhao et al., 2016). In their study, SRF was upregulated in DM group in a time-dependent manner, implying that SRF also participates in the podocyte injury induced by hyperglycaemia in vivo. Additionally, they found that SRF overexpression in podocytes activates Snail and facilitates dedifferentiation, migration and dysfunction of podocytes (Zhao et al., 2016). In our study, in distal convoluted tubules of the control groups, SRF was expressed in all time periods except after 2 weeks. More SRF positive cells were seen in the DM group in all time periods but still only in the cytoplasm. Although in our study we did not analyse biochemical markers of microalbuminuria or clinical proteinuria, in patients these events appear 5–10 years after the diagnosis of diabetes, and if untreated, it gradually increases over the next 10–15 years (Marshall, 2014b). The rate of progression of diabetic glomerulosclerosis has slowed in the past decades secondary to improving
CRediT authorship contribution statement Sandra Kostic: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing. Brandon Williams: Data curation, Formal analysis, Writing - review & editing. Samy Ksouri: Data curation, Formal analysis, Writing - review & editing. Leon Hardung: Data curation, Formal analysis, Writing - review & editing. Natalija Filipovic: Data curation, Formal analysis, Writing review & editing. Lejla Ferhatovic Hamzic: Data curation, Writing review & editing. Livia Puljak: Data curation, Writing - review & editing. Nasrollah Ghahramani: Data curation, Formal analysis, Writing - review & editing. Katarina Vukojevic: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing. Declaration of Competing Interest None. Acknowledgements The study was supported by University of Split School of Medicine Excellence Grants for prof. Sandra Kostic, Katarina Vukojevic and 6
Acta Histochemica xxx (xxxx) xxxx
S. Kostic, et al.
Natalija Filipovic and by Penn State – University of Split Collaboration Development Fund 2017 for prof. Vukojevic. The animal kidney samples of 6 and 12 months were funded by the Croatian Foundation for Science (HRZZ) grant no. 02.05./28.
A., Ursoniu, S., Vlad, D., Petrica, L., 2018. Deregulated profiles of urinary microRNAs may explain podocyte injury and proximal tubule dysfunction in normoalbuminuric patients with type 2 diabetes mellitus. J. Investig. Med. 66 (4), 747–754. Miyata, K., Ohashi, N., Suzaki, Y., Katsurada, A., Kobori, H., 2008. Sequential activation of the reactive oxygen species/angiotensinogen/renin-angiotensin system axis in renal injury of type 2 diabetic rats. Clin. Exp. Pharmacol. Physiol. 35 (8), 922–927. Petermann, A.T., Pippin, J., Krofft, R., Blonski, M., Griffin, S., Durvasula, R., Shankland, S.J., 2004. Viable podocytes detach in experimental diabetic nephropathy: potential mechanism underlying glomerulosclerosis. Nephron Exp. Nephrol. 98 (4), e114–123. Petrica, L., Petrica, M., Vlad, A., Jianu, D.C., Gluhovschi, G., Ianculescu, C., Firescu, C., Dumitrascu, V., Giju, S., Gluhovschi, C., Bob, F., Gadalean, F., Ursoniu, S., Velciov, S., Bozdog, G., Milas, O., 2011a. Proximal tubule dysfunction is dissociated from endothelial dysfunction in normoalbuminuric patients with type 2 diabetes mellitus: a cross-sectional study. Nephron Clin. Pract. 118 (2), c155–164. Petrica, L., Vlad, A., Petrica, M., Jianu, C.D., Gluhovschi, G., Gadalean, F., Dumitrascu, V., Ianculescu, C., Firescu, C., Giju, S., Gluhovschi, C., Bob, F., Velciov, S., Bozdog, G., Milas, O., Marian, R., Ursoniu, S., 2011b. Pioglitazone delays proximal tubule dysfunction and improves cerebral vessel endothelial dysfunction in normoalbuminuric people with type 2 diabetes mellitus. Diabetes Res. Clin. Pract. 94 (1), 22–32. Racetin, A., Raguz, F., Durdov, M.G., Kunac, N., Saraga, M., Sanna-Cherchi, S., Soljic, V., Martinovic, V., Petricevic, J., Kostic, S., Mardesic, S., Tomas, S.Z., Kablar, B., Restovic, I., Lozic, M., Filipovic, N., Saraga-Babic, M., Vukojevic, K., 2019. Immunohistochemical expression pattern of RIP5, FGFR1, FGFR2 and HIP2 in the normal human kidney development. Acta Histochem. 121 (5), 531–538. Rancic, A., Filipovic, N., Marin Lovric, J., Mardesic, S., Saraga-Babic, M., Vukojevic, K., 2017. Neuronal differentiation in the early human retinogenesis. Acta Histochem. 119 (3), 264–272. Roy, H.K., Iversen, P., Hart, J., Liu, Y., Koetsier, J.L., Kim, Y., Kunte, D.P., Madugula, M., Backman, V., Wali, R.K., 2004. Down-regulation of SNAIL suppresses MIN mouse tumorigenesis: modulation of apoptosis, proliferation, and fractal dimension. Mol. Cancer Ther. 3 (9), 1159–1165. Sohn, E., Kim, J., Kim, C.S., Jo, K., Kim, J.S., 2018. Osteomeles schwerinae extract prevents diabetes-induced renal injury in spontaneously diabetic torii rats. Evid. Complement. Alternat. Med. 2018, 6824215. Steffes, M.W., Osterby, R., Chavers, B., Mauer, S.M., 1989. Mesangial expansion as a central mechanism for loss of kidney function in diabetic patients. Diabetes 38 (9), 1077–1081. Valcourt, U., Kowanetz, M., Niimi, H., Heldin, C.H., Moustakas, A., 2005. TGF-beta and the Smad signaling pathway support transcriptomic reprogramming during epithelial-mesenchymal cell transition. Mol. Biol. Cell 16 (4), 1987–2002. Weir, G.C., Clore, E.T., Zmachinski, C.J., Bonner-Weir, S., 1981. Islet secretion in a new experimental model for non-insulin-dependent diabetes. Diabetes 30 (7), 590–595. Wu, Y., Zhou, B.P., 2010. Snail: more than EMT. Cell Adh. Migr. 4 (2), 199–203. Zhang, Y., Yang, J., Zheng, M., Wang, Y., Ren, H., Xu, Y., Yang, Y., Cheng, J., Han, F., Yang, X., Chen, L., Shan, C., Chang, B., 2015. Clinical characteristics and predictive factors of subclinical diabetic nephropathy. Exp. Clin. Endocrinol. Diabetes 123 (2), 132–138. Zhao, L., Wang, X., Sun, L., Nie, H., Liu, X., Chen, Z., Guan, G., 2016. Critical role of serum response factor in podocyte epithelial-mesenchymal transition of diabetic nephropathy. Diab. Vasc. Dis. Res. 13 (1), 81–92. Zhao, X., He, L., Li, T., Lu, Y., Miao, Y., Liang, S., Guo, H., Bai, M., Xie, H., Luo, G., Zhou, L., Shen, G., Guo, C., Bai, F., Sun, S., Wu, K., Nie, Y., Fan, D., 2014. SRF expedites metastasis and modulates the epithelial to mesenchymal transition by regulating miR-199a-5p expression in human gastric cancer. Cell Death Differ. 21 (12), 1900–1913. Zhou, L., Liu, Y., 2015. Wnt/beta-catenin signalling and podocyte dysfunction in proteinuric kidney disease. Nat. Rev. Nephrol. 11 (9), 535–545. Ziyadeh, F.N., 1993. Renal tubular basement membrane and collagen type IV in diabetes mellitus. Kidney Int. 43 (1), 114–120.
References Badid, C., Desmouliere, A., Babici, D., Hadj-Aissa, A., McGregor, B., Lefrancois, N., Touraine, J.L., Laville, M., 2002. Interstitial expression of alpha-SMA: an early marker of chronic renal allograft dysfunction. Nephrol. Dial. Transplant 17 (11), 1993–1998. Brosius 3rd, F.C., Banes-Berceli, A., 2010. A new pair of SOCS for diabetic nephropathy. J. Am. Soc. Nephrol. 21 (5), 723–724. Decleves, A.E., Sharma, K., 2010. New pharmacological treatments for improving renal outcomes in diabetes. Nat. Rev. Nephrol. 6 (6), 371–380. Delic Jukic, I.K., Kostic, S., Filipovic, N., Gudelj Ensor, L., Ivandic, M., Dukic, J.J., Vitlov Uljevic, M., Ferhatovic Hamzic, L., Puljak, L., Vukojevic, K., 2018. Changes in expression of special AT-rich sequence binding protein 1 and phosphatase and tensin homologue in kidneys of diabetic rats during ageing. Nephrol. Dial. Transplant 33 (10), 1734–1741. Fang, K.Y., Lou, J.L., Xiao, Y., Shi, M.J., Gui, H.Z., Guo, B., Zhang, G.Z., 2008. [Transforming growth factor-beta1 and Snail1 mediate tubular epithelial-mesenchymal transition in diabetic rats]. Sheng Li Xue Bao 60 (1), 125–134. Ferhatovic, L., Banozic, A., Kostic, S., Kurir, T.T., Novak, A., Vrdoljak, L., Heffer, M., Sapunar, D., Puljak, L., 2013. Expression of calcium/calmodulin-dependent protein kinase II and pain-related behavior in rat models of type 1 and type 2 diabetes. Anesth. Analg. 116 (3), 712–721. Gagliardini, E., Perico, N., Rizzo, P., Buelli, S., Longaretti, L., Perico, L., Tomasoni, S., Zoja, C., Macconi, D., Morigi, M., Remuzzi, G., Benigni, A., 2013. Angiotensin II contributes to diabetic renal dysfunction in rodents and humans via Notch1/Snail pathway. Am. J. Pathol. 183 (1), 119–130. Harney, A.S., Meade, T.J., LaBonne, C., 2012. Targeted inactivation of Snail family EMT regulatory factors by a Co(III)-Ebox conjugate. PLoS One 7 (2), e32318. He, L., Lou, W., Ji, L., Liang, W., Zhou, M., Xu, G., Zhao, L., Huang, C., Li, R., Wang, H., Chen, X., Sun, S., 2014. Serum response factor accelerates the high glucose-induced Epithelial-to-Mesenchymal Transition (EMT) via snail signaling in human peritoneal mesothelial cells. PLoS One 9 (10), e108593. Kwon, S.K., Kim, S.J., Kim, H.Y., 2016. Urine synaptopodin excretion is an important marker of glomerular disease progression. Korean J. Intern. Med. 31 (5), 938–943. Li, C.X., Siragy, H.M., 2014. High glucose induces podocyte injury via enhanced (Pro) renin receptor-wnt-beta-Catenin-Snail signaling pathway. PLoS One 9 (2). Lu, Q., Zuo, W.Z., Ji, X.J., Zhou, Y.X., Liu, Y.Q., Yao, X.Q., Zhou, X.Y., Liu, Y.W., Zhang, F., Yin, X.X., 2015. Ethanolic Ginkgo biloba leaf extract prevents renal fibrosis through Akt/mTOR signaling in diabetic nephropathy. Phytomedicine 22 (12), 1071–1078. Makuc, J., Petrovic, D., 2011. A review of oxidative stress related genes and new antioxidant therapy in diabetic nephropathy. Cardiovasc. Hematol. Agents Med. Chem. 9 (4), 253–261. Marshall, S.M., 2014a. Natural history and clinical characteristics of CKD in type 1 and type 2 diabetes mellitus. Adv. Chronic Kidney Dis. 21 (3), 267–272. Marshall, S.M., 2014b. Preventing kidney failure in people with diabetes. Diabet. Med. 31 (11), 1280–1283. Mathers, C.D., Loncar, D., 2006. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med. 3 (11), e442. Miano, J.M., 2010. Role of serum response factor in the pathogenesis of disease. Lab. Invest. 90 (9), 1274–1284. Milas, O., Gadalean, F., Vlad, A., Dumitrascu, V., Gluhovschi, C., Gluhovschi, G., Velciov, S., Popescu, R., Bob, F., Matusz, P., Pusztai, A.M., Cretu, O.M., Secara, A., Simulescu,
7