Abstract

Hyperglycemia-induced renal epithelial-to-mesenchymal transition (EMT) is a key pathological factor in diabetic renal tubulointerstitial fibrosis (RIF). Our previous studies have shown that connexin 43 (Cx43) activation attenuated the development of diabetic renal fibrosis. However, whether Cx43 regulates the EMT of renal tubular epithelial cells (TECs) and the pathological process of RIF under the diabetic conditions remains to be elucidated. In the present study, we identified that Cx43 protein expression was down-regulated in the kidney tissues of db/db mice as well as in high glucose (HG)-induced NRK-52E cells. Overexpression of Cx43 improved renal function in db/db spontaneous diabetic model mice, increased SIRT1 levels, decreased hypoxia-inducible factor (HIF)-1α expression, and reduced production of EMT markers and extracellular matrix (ECM) components. Additionally, Cx43 overexpression inhibited the EMT process and reduced the expression of ECM components such as fibronectin (FN), Collagen I, and Collagen IV in HG-induced NRK-52E cells, whereas Cx43 deficiency had the opposite effects. Mechanistically, Cx43 in a carboxyl-terminal signal transduction-dependent manner could up-regulate SIRT1 expression and enhance SIRT1-dependent deacetylation of HIF-1α to reduce HIF-1α activity, which eventually ameliorated renal EMT and diabetic RIF. Our study indicates the essential role of Cx43 in regulating renal EMT and diabetic RIF via regulating the SIRT1-HIF-1α signaling pathway and provides an experimental basis for Cx43 as a potential target for diabetic nephropathy (DN).

Introduction

The epidemic of diabetes mellitus and its complications are considered a major global health threat [1]. Diabetic nephropathy (DN), one of the major chronic microvascular complications of diabetes, is a critical cause of end-stage renal disease [1]. Renal fibrosis is the main pathological change of DN, which is characterized by glomerular sclerosis and renal tubulointerstitial fibrosis (RIF) [2,3]. The histological feature of DN is excessive accumulation of extracellular matrix (ECM) in the mesangial and tubular interstitium, manifested as excessive renal scarring and reduced excretory function [4]. Renal tubular epithelial cells (TECs) are important intrinsic cells of the renal tubule [5]. The renal epithelial-to-mesenchymal transition (EMT) that occurs in diabetes mellitus is characterized by phenotypic transformation that involves the loss of epithelial markers such as E-cadherin and zonula occludens-1 (ZO-1) and increased mesenchymal markers including α-smooth muscle actin (α-SMA) and vimentin, which lead to the accumulation of fibronectin (FN), one of the ECM components in tubular interstitium, contributes to the development of the RIF [6–8]. Therefore, inhibiting the process of EMT can reduce the accumulation of ECM components, delay the progression of renal fibrosis, and thus effectively prevent the occurrence and development of DN.

Gap junction membrane channels are composed of connexins (Cx), which are assembled as a hexameric connexon that provides a conduit for paracrine signaling of small molecules and ions to regulate the function and activity of adjacent cells [9–11]. The biological functions of Cx are accomplished by its gap junctional intercellular communication role or by interacting with numerous cytokines and proteins through its carboxyl tail [12,13]. Recently, considerable attention has been focused on the regulation of the Cx proteins, especially connexin 43 (Cx43), in DN and renal fibrosis [14]. Cx43 is the first Cx protein to be discovered and has the most abundant expression and widest distribution in all tissues among the Cx proteins [14]. Previous studies reported that a significant reduction in Cx43 expression and function was observed in the kidney tissues of DN patients and diabetic model animals [14,15]. The same phenomenon was also observed in proximal TECs under the high glucose (HG) conditions [16,17]. Our previous studies have found that Cx43 activation can inhibit c-Src and NF-κB, increase the nuclear aggregation, DNA-binding activity, and transcriptional activity of nuclear factor E2-related factor 2 (Nrf2), reduce the expression of fibrotic components such as transforming growth factors (TGF)-β1 and FN in HG-induced glomerular mesangial cells (GMCs), and thus prevent the development of diabetic renal fibrosis [18–20]. Since Cx43 is a central regulator of anti-renal fibrosis in HG-induced GMCs, it is plausible that Cx43 can affect the EMT of TECs and the pathological process of RIF under the diabetic conditions.

SIRT1, a nicotinamide-adenine-dinucleotide (NAD+)-dependent deacetylase, known to be a member of the mammalian Sirtuin family, is a principal regulator of gene transcription, chromosome stability, and target protein activity and therefore, participates in a series of physiological functions including metabolism, cancer, and so on [21]. Studies have shown that SIRT1 plays an important protective role in the kidney through its antioxidative function [22–24]. Specific knockdown of SIRT1 can aggravate renal injury in both db/db diabetic model mice and STZ-induced diabetic mice [25,26]. Our previous studies also revealed that SIRT1 inhibits the expression of FN and TGF-β1 by activating the Nrf2/antioxidant response element antioxidative stress pathway in advanced glycation end products (AGEs)-induced GMCs, and thereby ameliorates the progression of DN [27]. Accumulating evidence has highlighted the critical role of SIRT1 in EMT [28,29]. It was documented that up-regulating SIRT1 attenuates EMT by improving mitochondrial function to resist oxidative stress [30].

Hypoxia-inducible factor (HIF) is a heterodimeric transcription factor composed of an O2-regulated α subunit and a constitutively expressed β subunit [31]. HIF-1α is a key mediator in cell metabolism, tumorigenesis, and inflammation under the hypoxic conditions [32–34]. Under the DN conditions, the antioxidative capacity of the organism was decreased [35]. One study showed that HIF-1α expression was increased in a blood glucose-dependent manner in the kidney tissues of mice with STZ-induced type 1 diabetes and the db/db spontaneous diabetic model mice as well as in HG-treated GMCs [35]. HIF-1α can directly regulate the occurrence and development of EMT [36–39]. In addition, studies have shown that SIRT1 can bind to HIF-1α and deacetylate its lysine residues [40,41]. Based on the facts that both Cx43 and SIRT1 possess antioxidative stress capacity, we aimed to investigate whether Cx43 could exert anti-EMT effects and protect against RIF by targeting the SIRT1-HIF-1α signaling pathway.

Materials and methods

Reagents and antibodies

Dulbecco’s Modified Eagle’s Medium (DMEM), fetal bovine serum, RNAiMAX and Lipofectamine® LTX & Plus reagents, and Lucifer Yellow were acquired from Life Technologies (Grand Island, NY, U.S.A.). d-Glucose was obtained from AMRESCO (Solon, OH, U.S.A.). Polyvinylidene difluoride (PVDF) membrane was purchased from Immobilon®-PSQ (Millipore, CA, U.S.A.). Nylon membrane was purchased from Amersham (Pittsburgh, PA, U.S.A.). Enhanced chemiluminescence substrate for the detection of horseradish peroxidase (HRP) and LightShift® chemiluminescent electrophoretic mobility shift assay (EMSA) kit were purchased from Thermo Fisher Scientific (Rockford, IL, U.S.A.).

Antibodies against α-Tubulin (catalog: 66031-1-lg), FN (catalog: 15613-1-AP), and ZO-1 (catalog: 21773-1-AP) were purchased from Proteintech Group (Chicago, IL, U.S.A.). Antibodies against Cx43 (catalog: sc-59949) were purchased from Santa Cruz Biotechnology (CA, U.S.A.). Antibodies against Cx43 (catalog: ab11370), Collagen IV (catalog: ab6586), and Lamin B (catalog: ab133741) were purchased from Abcam (Cambridge, MA, U.S.A.). Antibodies against HIF-1α (catalog: NB100-105) were purchased from Novus Biologicals (Littleton, CO, U.S.A.). Antibodies against E-cadherin (catalog: #14472), vimentin (catalog: #5741), SIRT1 (catalog: #8469), α-SMA (catalog: #19245), Collagen I (catalog: #39952), and acetylated-Lysine (catalog: #9441S) were purchased from Cell Signaling Technology (Danvers, MA, U.S.A.). HRP-conjugated secondary antibodies were obtained from Promega Corporation (Madison, WI, U.S.A.). Goat anti-rabbit IgG labeled with Alexa Fluor® 488 (catalog: A11008) and goat anti-mouse IgG labeled with Alexa Fluor® 594 (catalog: A32742) were purchased from Thermo Fisher Scientific (Rockford, IL, U.S.A.).

Cell culture

Rat NRK-52E cells were maintained in DMEM with the presence of 10% fetal bovine serum incubated at 37°C in 5% CO2 in air. At appropriate subconfluence, the NRK-52E cells were rendered quiescent by incubation for 12–16 h in serum-free medium before being divided into different groups and treated with glucose (5.6 mM as normal glucose (NG) and 30 mM as HG. All experiments were performed in triplicate.

Transfections of plasmids, short hairpin RNAs, and small interfering RNAs

The following plasmids were obtained from GeneChem (Shanghai, China): CMV-Cx43-HA (1–382 aa), CMV-Cx43ΔCT-HA (1–234 aa) (a plasmid that lacked the Cx43 C-terminus), and CMV-Cx43CT-HA (235–382 aa) (a plasmid that is only expressed at the Cx43 C-terminus). pcDNA3-myc-SIRT1 plasmids was provided by Dr. Hueng-Sik Choijia (School of Biological Sciences and Technology, Chonnam National University, Korea) [42]. The NRK-52E cells were plated in 35-mm plates 24 h prior to transfection with 2 μg of the indicated plasmids according to the instructions using the transfection reagent LTX & Plus. After 24 h, the transfection medium was removed, and the cells were treated with corresponding stimuli (5.6 mM glucose as NG and 30 mM glucose as HG). After that, the cells were harvested and further detection was performed.

Three short hairpin RNAs (shRNAs) that targeted Cx43 were synthesized by GeneChem (Shanghai, China). The sequences were as follows: 85-2: sense: AGGAA GAGAAGCTAAA CAA, antisense: TTGTTTAGCTTCTCTTCCTTC; 86-1: sense: G CTGGTTACTGGTGACA GA, antisense: TCTGTCACCAGTAACCAGCTT; 87-2: sense: AGAGCACGGCAAGGTG AAA, antisense: TTTCACCTTGCCGTGCTCTTC. Small interfering RNA (siRNA) that targeted SIRT1 was synthesized by GenePharma (Shanghai, China). The sequences were as follows: sense: CCAGUAGCACUAAUUC CAATT, antisense: UUGGAAUUAGUGCUACU GGTT. The NRK-52E cells were grown in 35-mm plates, and were transfected with 5 μl of siRNA (50 nM) using RNAiMAX according to the manufacturer’s protocol. After further treatment, the cells were harvested, and Western blot assay was performed to determine the expression of the indicated proteins.

Western blot assay

Western blot assay was performed as previously described [43]. Briefly, kidney cortex fragments or NRK-52E cells were lysed using RIPA lysis buffer (Beyotime, Haimen, China) supplemented with a protease inhibitor cocktail and phosphatase inhibitors A and B (Bimake, Houston, U.S.A.). After centrifuging at 12000×g for 15 min at 4°C, the protein concentration was determined by using a BCA™ Protein Assay Kit (Pierce, Rockford, IL, U.S.A.). Then, equal amounts of protein samples were separated by 8–12% sodium dodecyl sulfate (SDS)/polyacrylamide gel electrophoresis and then transferred on to PVDF membranes. The blots were visualized using a GE ImageQuant LAS4000mini (GE Healthcare, Waukesha, WI, U.S.A.) and analyzed with Quantity One Protein Analysis Software (Bio-Rad, Hercules, CA, U.S.A.).

Immunofluorescence staining

Briefly, the NRK-52E cells were seeded on glass coverslips in 24-well plates at 60% confluence, and they were treated with various stimuli. After the designated treatments, the samples were washed with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde for 15 min, and permeabilized with 0.02% Triton X-100 in PBS for 10 min at room temperature. The samples were then blocked with 10% goat serum for 50 min at room temperature and then incubated with the corresponding primary antibodies overnight at 4°C. The coverslips were washed three times with PBS and incubated with a secondary antibody (Alexa Fluor® 488 or Alexa Fluor® 594) in the dark at room temperature for 1 h. The nuclei were co-labeled with DAPI dihydrochloride (5 mg/ml in PBS, Sigma; St. Louis, MO, U.S.A.) for 10 min in the dark at room temperature. Finally, the coverslips were mounted on slides using anti-fade mounting medium (Beyotime, Haimen, Jiangsu, China). The images were collected with a Zeiss LSM 510 laser confocal fluorescence microscope (Carl Zeiss, Oberkochen, Germany).

Immunoprecipitation assay

The NRK-52E cells were lysed on ice for 30 min with immunoprecipitation (IP) lysate buffer (Beyotime, Haimen, Jiangsu, China) that contained phosphatase inhibitors A and B and a protease inhibitor cocktail and then centrifuged at 12000×g for 15 min at 4°C. The whole-cell lysates (400 μg) were incubated with 20 μl of protein agarose A/G beads (Pierce, Rockford, IL, U.S.A.) before IP to reduce non-specific binding, followed by transient centrifugation to collect the supernatants. The supernatants were then incubated with 2 μg of SIRT1 antibodies or IgG overnight at 4°C with shaking. Protein agarose A/G beads (20 μl) were added to the supernatants for 4 additional hours at 4°C with shaking. SDS loading buffer (Fdbio, Hangzhou, Zhejiang, China) (20 μl) was added to the beads after extensively washing three times with IP buffers 1, 2, and 3. Finally, the mixtures were boiled for 5 min and then analyzed by Western blot assay with the respective antibodies.

Assessment of gap junctional intercellular communication

The scrape loading/dye transfer (SL/DT) assay has been widely used to elucidate the gap junctional intercellular communication status of many cell types in various biological circumstances [44]. Briefly, the NRK-52E cells were grown in six-well plates and were treated with corresponding stimuli at 60% confluence. After various treatments, the cells were carefully washed with PBS. The cell layer was then scraped with a scalpel blade in PBS containing Lucifer Yellow (1 mg/ml), which can pass through the gap junctions of the loaded cells to their neighbors. After incubating for 10 min at 37°C, the Lucifer yellow solution was removed and the cells were gently washed twice. Fluorescence photomicrographs were captured with an EVOS FL Auto fluorescence microscope (Life Technologies, Grand Island, NE, U.S.A.).

EMSA assay

The DNA-binding activity of HIF-1α was determined by EMSA using the nuclear extracts prepared with a nuclear extraction kit (Active Motif, Carlsbad, CA, U.S.A.) according to manufacturer’s instructions. The sequence of the biotin-labeled oligonucleotide probe for HIF-1α was 5′-TCTGTACGTGACCACACTCACCTC-3′, 3′-AGACATGCACTGGTGTGAGTG GAG-5′. The procedures were performed following the instructions of the manufacturer (Light Shift Chemiluminescent EMSA Kit, Pierce, Rockford, IL, U.S.A.). For each sample, 7 μg of the nuclear proteins were incubated with the mixture in the presence of 50 ng/ml poly (dIdC), 0.05% Nonidet P-40, 5 mM of MgCl2, and 2.5% glycerol for 10 min, and subsequently incubated with the HIF-1α probe for another 20 min at room temperature. The reaction mixtures were separated using 6% non-denaturing PAGE and transferred to nylon membranes for DNA-protein cross-links. After blocking for 1.5 h at room temperature, the membrane was incubated with HRP-conjugated streptavidin antibody (1:300) for 15 min at room temperature, then visualized with enhanced chemiluminescence using an ImageQuant LAS4000mini instrument.

Dual luciferase reporter assay

The NRK-52E cells were grown to 60% confluence in 96-well plates and co-transfected with 0.1 μg of pGMHIF-1α-Luc (Yeasen, Shanghai, China) and 0.02 μg of pRL-TK (Promega, Madison, WI, U.S.A.) in the presence or absence of 0.05 μg of Cx43-HA or SIRT1. After the designated treatments, the cells were harvested and the luciferase activity was assessed using the dual-luciferase reporter assay system kit (Promega, Madison, WI, U.S.A.) to analyze the effect of Cx43 and SIRT1 on the transcriptional activity of HIF-1α. The luciferase activity was normalized to the Renilla luciferase activity.

Animals and treatment

All animal care and experimental procedures complied with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1996), in-line with the China Animal Welfare Legislation, and were approved by the Ethics Committee on the Care and Use of Laboratory Animals of Sun Yat-sen University, Guangzhou, Guangdong, China.

Briefly, 22 healthy specific pathogen-free (SPF) male C57BL/KS-db leptin receptor-deficient type 2 diabetic mice (abbreviated as db/db) aged 5 weeks and a control group of 11 male C57BL/KS wild-type (WT) mice aged 5 weeks were supplied by the Model Animal Research of Nanjing University (Nanjing, Jiangsu, China; animal quality certification number: 32002100006714). All animals were housed under SPF conditions and in a temperature-controlled (20–25°C) and humidity-controlled (40–70%) barrier system with a 12-h:12-h light and dark cycle in Sun Yat-sen University Laboratory Animal Center (Sun Yat-sen University, 132 WaiHuan East Road, Guangzhou Higher Education Mega Center). The mice were adapted to the environment for 1 week and then randomly divided into three groups as follows: WT control group (n=11), db/db model group (n=11), and Cx43-overexpressing adenovirus-injected group (n=11). The mice in both the WT control group and the db/db model group were injected with Cx43 negative control (NC) adenovirus. In addition, the mice in the Cx43-overexpressing adenovirus-injected group were injected with the Cx43-overexpressing adenovirus once a week for 9 weeks. Fasting blood glucose (FBG) was evaluated every other week using a one-touch glucometer (Johnson, Milpitas, CA, U.S.A.). At the termination of the experiment, the animals were placed into metabolic cages overnight to collect urine to detect the levels of 24-h urinary protein (UP). Finally, the mice were anesthetized with ether inhalation and blood as well as kidney samples were collected to determine the levels of glycosylated serum protein (GSP), blood urea nitrogen (BUN), and serum creatinine (Cr). The expression of the EMT marker protein was observed in one kidney of the mice. The pathological changes of the kidney were observed after the other kidney tissue was fixed, embedded, and stained.

Statistical analysis

All data analyses were performed by using GraphPad Prism 5.0. The data are expressed as the means ± SDs. Unpaired Student’s t test was used to compare two groups and by one-way ANOVA with post hoc multiple comparisons for multiple comparisons. Post hoc tests were further conducted only if the F was significant. P<0.05 was considered to be statistically significant.

Results

Cx43 protein expression was significantly decreased in both the kidney tissues of db/db mice and the NRK-52E cells cultured in HG

To demonstrate that Cx43 might be associated with diabetic RIF, we initially detected the expression of Cx43 under the diabetic conditions. The Western blot data showed that Cx43 levels was significantly decreased in the kidney tissues of db/db mice compared with those in C57BL/6J mice as normal controls (Figure 1A), which is consistent with the results of immunohistochemical analysis (Figure 1B). In NRK-52E cells cultured in HG, the Western blot data showed that the protein expression of Cx43 was down-regulated in a time-dependent manner (Figure 1C). The IF data revealed that the Cx43 levels were significantly decreased in the NRK-52E cells cultured in HG for 36 h (Figure 1D). These data showed a likely link between Cx43 and DN.

Cx43 protein expression was significantly decreased in both the kidney tissues of db/db mice and the NRK-52E cells cultured in HG

Figure 1
Cx43 protein expression was significantly decreased in both the kidney tissues of db/db mice and the NRK-52E cells cultured in HG

(A,B) The expression of Cx43 in the kidney tissues of db/db mice (n=6) assessed by Western blot assay and IHC (400× magnification). **P<0.01 vs. C57BL/6J. (C) After treatment of 30 mM of HG for various times (0, 12, 24, 36, 48 h), the Cx43 levels in the NRK-52E cells were determined using the Western blot assay. *P<0.05 vs. 0 h. (D) The IF results showed the expression of Cx43 in the NRK-52E cells under the HG (30 mM) conditions for 36 h (600× magnification). Independent experiments were performed three times with similar results. Abbreviation: IHC, immunohistochemistry.

Figure 1
Cx43 protein expression was significantly decreased in both the kidney tissues of db/db mice and the NRK-52E cells cultured in HG

(A,B) The expression of Cx43 in the kidney tissues of db/db mice (n=6) assessed by Western blot assay and IHC (400× magnification). **P<0.01 vs. C57BL/6J. (C) After treatment of 30 mM of HG for various times (0, 12, 24, 36, 48 h), the Cx43 levels in the NRK-52E cells were determined using the Western blot assay. *P<0.05 vs. 0 h. (D) The IF results showed the expression of Cx43 in the NRK-52E cells under the HG (30 mM) conditions for 36 h (600× magnification). Independent experiments were performed three times with similar results. Abbreviation: IHC, immunohistochemistry.

HG stimulation induced EMT in NRK-52E cells and promoted the expression of ECM components

The NRK-52E cells were treated with 30 mM of HG for 0, 12, 24, 36, and 48 h in a simulated in vitro EMT model. The Western blot data showed that the expression of epithelial markers E-cadherin and ZO-1 was down-regulated (Figure 2A), which was followed by up-regulation of the mesenchymal markers α-SMA and vimentin (Figure 2B), as well as ECM components FN, Collagen I, and Collagen IV (Figure 2C). The results in Figure 2A–C showed that HG (30 mM) treatment for 36 h significantly induced the EMT process in the NRK-52E cells. Therefore, HG (30 mM) treatment for 36 h was used for the subsequent experiments. An IF experiment was performed to define the changes in the EMT markers and ECM components in the above EMT models. The IF data revealed that the expression of E-cadherin was clearly decreased (Figure 2D), and the expression of α-SMA (Figure 2E) and FN (Figure 2F) was significantly increased in the NRK-52E cells that were treated with 30 mM of HG for 36 h.

HG stimulation induced EMT in NRK-52E cells and promoted the expression of ECM components

Figure 2
HG stimulation induced EMT in NRK-52E cells and promoted the expression of ECM components

(A–C) The expression of epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) were assessed using Western blot assay under the HG (30 mM) conditions at various times (0, 12, 24, 36, 48 h). * P<0.05 vs. 0 h, **P<0.01 vs. 0 h. (D–F) IF was performed to evaluate the protein levels of E-cadherin, α-SMA, and FN under either the NG or HG conditions for 36 h (600× magnification). Independent experiments were performed three times with similar results.

Figure 2
HG stimulation induced EMT in NRK-52E cells and promoted the expression of ECM components

(A–C) The expression of epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) were assessed using Western blot assay under the HG (30 mM) conditions at various times (0, 12, 24, 36, 48 h). * P<0.05 vs. 0 h, **P<0.01 vs. 0 h. (D–F) IF was performed to evaluate the protein levels of E-cadherin, α-SMA, and FN under either the NG or HG conditions for 36 h (600× magnification). Independent experiments were performed three times with similar results.

Overexpression of Cx43 suppressed the EMT process and expression of ECM components in NRK-52E cells

To explore whether Cx43 affected the development of DN, NRK-52E cells that were cultured under the NG and HG conditions were transfected with plasmids that overexpress Cx43-HA. The Western blot data of the tagged HA protein showed the success of Cx43 overexpression (Figure 3A). In the NRK-52E cells under the NG conditions, Cx43 overexpression had no obvious influence on the expression of E-cadherin, ZO-1, α-SMA, vimentin, FN, Collagen I, and Collagen IV (Figure 3B–D). However, in the HG-treated NRK-52E cells, Cx43 overexpression significantly increased the Cx43 levels (Figure 3A), reversed the down-regulated expression of E-cadherin and ZO-1 (Figure 3B), and reduced the α-SMA and vimentin levels (Figure 3C), which indicated that Cx43 suppressed the EMT process. In addition to the suppressed EMT, Cx43 overexpression also decreased the expression of the ECM components FN, Collagen I, and Collagen IV in the NRK-52E cells that had been challenged with HG (Figure 3D). These findings indicated that Cx43 plays a positive role in diabetic RIF.

Overexpression of Cx43 suppressed the EMT process and ECM components expression in NRK-52E cells

Figure 3
Overexpression of Cx43 suppressed the EMT process and ECM components expression in NRK-52E cells

(A-D) The protein expression of HA and Cx43, epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) in NRK-52E were determined using Western blot assay after transfected with 2 μg of the Cx43 under either the NG or HG conditions for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG, #P<0.05, ##P<0.01, ###P<0.001. vs. HG. Ctrl: control. Independent experiments were performed three times with similar results.

Figure 3
Overexpression of Cx43 suppressed the EMT process and ECM components expression in NRK-52E cells

(A-D) The protein expression of HA and Cx43, epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) in NRK-52E were determined using Western blot assay after transfected with 2 μg of the Cx43 under either the NG or HG conditions for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG, #P<0.05, ##P<0.01, ###P<0.001. vs. HG. Ctrl: control. Independent experiments were performed three times with similar results.

Cx43 depletion further aggravated the EMT process and promoted the expression of ECM components

To further explore the influence of Cx43 on the renal EMT process, the NRK-52E cells were transfected with the NC and three pairs of shRNA that targeted Cx43. The Western blot results showed that compared with the NC, shRNA-85-2, 86-1, and shRNA-87-2 significantly decreased the protein levels of Cx43 to 81.5, 37.9, and 33.9%, respectively (Figure 4A). Therefore, shRNA-87-2, which targeted Cx43, was selected for the subsequent experiments. Next, NRK-52E cells were transfected with the shRNA-87-2 under the NG and HG conditions. In the NRK-52E cells under the NG conditions, transfection of shRNA-87-2 that targeted Cx43 decreased the expression of E-cadherin and ZO-1 as well as increased the expression of vimentin and Collagen I (Figure 4B–D). Whereas Cx43 depletion further decreased the levels of E-cadherin and ZO-1 under the HG condition (Figure 4B), promoted the expression of α-SMA and vimentin (Figure 4C), and ultimately increased the expression of the ECM components FN, Collagen I, and Collagen IV (Figure 4D) in the NRK-52E cells under the HG conditions.

Cx43 depletion further aggravated the EMT process and promoted the expression of ECM components

Figure 4
Cx43 depletion further aggravated the EMT process and promoted the expression of ECM components

(A) Western blot assay was performed to identify the effects of three shRNAs that targeted Cx43 on the protein levels of Cx43. ***P<0.001 vs. NC. The NRK-52E cells were transfected with 2 μg of Cx43-shRNA under either the NG or HG conditions for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. (B–D) The protein expression of epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) were evaluated using Western blot assay. *P<0.05, **P<0.01 vs. NG, #P<0.05, ##P<0.01 vs. HG. Ctrl: control. Independent experiments were performed three times with similar results.

Figure 4
Cx43 depletion further aggravated the EMT process and promoted the expression of ECM components

(A) Western blot assay was performed to identify the effects of three shRNAs that targeted Cx43 on the protein levels of Cx43. ***P<0.001 vs. NC. The NRK-52E cells were transfected with 2 μg of Cx43-shRNA under either the NG or HG conditions for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. (B–D) The protein expression of epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) were evaluated using Western blot assay. *P<0.05, **P<0.01 vs. NG, #P<0.05, ##P<0.01 vs. HG. Ctrl: control. Independent experiments were performed three times with similar results.

The above-mentioned results provided a preliminary demonstration that Cx43 plays an important role in the development of diabetic RIF by resisting the HG-induced EMT process. However, the mechanisms by which Cx43 suppressed the expression of EMT markers in DN remain to be explored.

Cx43 regulated the EMT process in NRK-52E cells independent of its regulation of gap junctional intercellular communication

Cx43 mainly exists in the membrane [45]. This protein consists of two extracellular subdomains and three cytosolic subdomains with both the C- and N-terminal ends facing the cytosol [45]. Considering that the cytosolic C-terminal tail of Cx43 (Cx43) is an important site for protein–protein interactions and plays a vital role in regulating various cell functions [45], the model cells were transfected with the WT Cx43-HA plasmid (Cx43), a plasmid that lacked the C-terminus Cx43ΔCT-HA (Cx43ΔCT), and a plasmid that only expressed the Cx43 C-terminus Cx43CT-HA (Cx43CT) to investigate whether Cx43 inhibits the EMT process and renal fibrosis development by interacting with other proteins through its C-terminus rather than completely depending on its gap junction communication function.

After using SL/DT to confirm that the transfected Cx43CT could not form complete cell gap junctions (Figure 5A), the NRK-52E cells were transfected with the Cx43, Cx43ΔCT, or Cx43CT under the HG conditions. The Western blot data of the tagged HA protein showed the success of Cx43, Cx43ΔCT, or Cx43CT plasmid transfected (Figure 5B). As shown in Figure 5C–E, the transfection of either Cx43 or Cx43CT could reverse the reduction in E-cadherin and ZO-1 as well as the elevation of α-SMA, vimentin, and ECM components: FN, Collagen I, and Collagen IV in the NRK-52E cells that were challenged with HG. However, Cx43ΔCT overexpression had no significant effect on those changes induced by HG, which suggested that Cx43 prevented the EMT process mainly through its C-terminal signal transduction role rather than its gap junction communication function.

Cx43 regulated the EMT process in NRK-52E cells independent of its gap junctional intercellular communication

Figure 5
Cx43 regulated the EMT process in NRK-52E cells independent of its gap junctional intercellular communication

(A) The SL/DT results showed that cells transfected with Cx43CT could not form complete cell gap junctions. (B) The Western blot data of the tagged HA protein showed the success of Cx43, Cx43ΔCT, or Cx43CT plasmid transfected. (C–E) The protein expression of epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) were determined using Western blot assay after transfection with the Cx43, Cx43CT, or Cx43ΔCT, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG; #P<0.05, ##P<0.01 vs. HG; $P<0.05, $$P<0.01 vs. Cx43. Ctrl: control. Independent experiments were performed three times with similar results.

Figure 5
Cx43 regulated the EMT process in NRK-52E cells independent of its gap junctional intercellular communication

(A) The SL/DT results showed that cells transfected with Cx43CT could not form complete cell gap junctions. (B) The Western blot data of the tagged HA protein showed the success of Cx43, Cx43ΔCT, or Cx43CT plasmid transfected. (C–E) The protein expression of epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) were determined using Western blot assay after transfection with the Cx43, Cx43CT, or Cx43ΔCT, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG; #P<0.05, ##P<0.01 vs. HG; $P<0.05, $$P<0.01 vs. Cx43. Ctrl: control. Independent experiments were performed three times with similar results.

Cx43 inhibited the EMT process in the NRK-52E cells by up-regulating SIRT1 expression

To determine whether SIRT1 mediates the regulation of Cx43 on the EMT process in the NRK-52E cells, we investigated the role of SIRT1 in renal fibrosis. The Western blot results revealed that the protein level of SIRT1 was time-dependently down-regulated in the NRK-52E cells that were treated with HG (Figure 6A). The NRK-52E cells were then transfected with the Cx43, Cx43ΔCT, or Cx43CT under the HG conditions. As illustrated in Figure 6B, the transfection with either Cx43 or Cx43CT could reverse the HG-induced reduction in SIRT1, rather than Cx43ΔCT, which indicated that the effect of Cx43 on SIRT1 was mainly related to its C-terminal signal transduction role. The immunofluorescence (IF) results showed that Cx43 and SIRT1 co-localized in the cytoplasm of the NRK-52E cells (Figure 6C). Furthermore, the co-IP results found that Cx43 directly binds to SIRT1 (Figure 6D), which provided a spatial possibility for their interaction, but the specific mechanism needs to be further studied.

Cx43 inhibited the EMT process in the NRK-52E cells by up-regulating SIRT1 expression

Figure 6
Cx43 inhibited the EMT process in the NRK-52E cells by up-regulating SIRT1 expression

(A) The Western blot results revealed the protein level of SIRT1 in the NRK-52E cells which was under the HG (30 mM) conditions at various times (0, 3, 6, 12, 24, 36 h). (B) The expression of SIRT1 in the NRK-52E cells under HG for 36 h was determined by Western blot assay after transfection with Cx43, Cx43CT, or Cx43ΔCT, and the whole-cell lysates were extracted after 48 h of transfection. (C) The IF results showed that Cx43 and SIRT1 co-localized in the cytoplasm of the NRK-52E cells (600× magnification). (D) The co-IP results showed that Cx43 directly binds to SIRT1. (E–H) The levels of the protein expression of Cx43 and SIRT1, epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) were determined using Western blot assay after transfection with the Cx43, si-SIRT1, or co-transfection with the Cx43and si-SIRT1 at the same time under HG for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG; #P<0.05, ##P<0.01 vs. HG; $$P<0.01 vs. Cx43. Ctrl: control. Independent experiments were performed three times with similar results.

Figure 6
Cx43 inhibited the EMT process in the NRK-52E cells by up-regulating SIRT1 expression

(A) The Western blot results revealed the protein level of SIRT1 in the NRK-52E cells which was under the HG (30 mM) conditions at various times (0, 3, 6, 12, 24, 36 h). (B) The expression of SIRT1 in the NRK-52E cells under HG for 36 h was determined by Western blot assay after transfection with Cx43, Cx43CT, or Cx43ΔCT, and the whole-cell lysates were extracted after 48 h of transfection. (C) The IF results showed that Cx43 and SIRT1 co-localized in the cytoplasm of the NRK-52E cells (600× magnification). (D) The co-IP results showed that Cx43 directly binds to SIRT1. (E–H) The levels of the protein expression of Cx43 and SIRT1, epithelial markers (E-cadherin and ZO-1), mesenchymal markers (α-SMA and vimentin), and ECM components (FN, Collagen I, and Collagen IV) were determined using Western blot assay after transfection with the Cx43, si-SIRT1, or co-transfection with the Cx43and si-SIRT1 at the same time under HG for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG; #P<0.05, ##P<0.01 vs. HG; $$P<0.01 vs. Cx43. Ctrl: control. Independent experiments were performed three times with similar results.

To further explore the influence of SIRT1 on the EMT process and the relationship between SIRT1 and Cx43, si-RNA that targeted SIRT1 (si-SIRT1) was introduced in the subsequent experiments. The NRK-52E cells were transfected with si-SIRT1 and subsequently transfected with Cx43. As the western blot results in Figure 6E show, si-SIRT1 markedly reduced the SIRT1 protein levels. As for the EMT markers and ECM components, si-SIRT1 clearly decreased E-cadherin and ZO-1 levels and increased α-SMA, vimentin, FN, Collagen I, and Collagen IV levels in the NRK-52E cells that were treated with HG (Figure 6F–H). Although Cx43 overexpression inhibited the EMT process and renal fibrosis, pretreatment with si-SIRT1 markedly blunted its effects as demonstrated by reduced E-cadherin and ZO-1 expression, increased α-SMA and vimentin levels, and up-regulated FN, Collagen I, and Collagen IV levels in the NRK-52E cells that were treated with HG (Figure 6F–H). These results suggested that SIRT1 mediates the regulation of Cx43 on the EMT process in the NRK-52E cells.

Cx43 reduced HIF-1α activity by up-regulating SIRT1 to deacetylate HIF-1α

Numerous results have revealed that HIF-1α has a close relationship with EMT [35–38]. Thus, we further investigated whether Cx43 reduced HIF-1α activity by SIRT1 deacetylation, thereby inhibiting the EMT process and ECM components expression in model cells. The IF results revealed that SIRT1 and HIF-1α were co-localized in the cytoplasm of the NRK-52E cells, which provided a spatial possibility for their interaction (Figure 7A). In the NRK-52E cells, HG treatment increased the protein levels, DNA-binding activity, transcriptional activity and acetylation level of HIF-1α (Figure 7B–E). Cx43 overexpression could increase SIRT1 expression, and decrease the protein levels, DNA-binding activity, transcriptional activity and the acetylation level of HIF-1α in HG-induced NRK-52E cells. (Figure 7B–E). However, si-SIRT1 pretreatment before Cx43 transfection abolished those effects of Cx43 (Figure 7B–E).

Cx43 reduced HIF-1α activity by up-regulating SIRT1to deacetylate HIF-1α

Figure 7
Cx43 reduced HIF-1α activity by up-regulating SIRT1to deacetylate HIF-1α

(A) The IF results showed that SIRT1 and HIF-1α were co-localized. (B–E) The protein expression, DNA-binding activity, transcriptional activity, and acetylation levels of HIF-1α were determined by Western blot assay after transfection with Cx43 in the presence and absence of the si-SIRT1 under HG for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG; #P<0.05, ##P<0.01 vs. HG. (F,G) The Western blot and IF results showed the expression of HIF-1α after transfection with the SIRT1 overexpression plasmid under the HG conditions for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG; #P<0.05, ##P<0.01 vs. HG. (H–J) The DNA-binding activity, transcriptional activity, and acetylation levels of HIF-1α were determined by Western blot assay after transfection with the SIRT1 overexpression plasmid under HG for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. Ctrl: control. Independent experiments were performed three times with similar results.

Figure 7
Cx43 reduced HIF-1α activity by up-regulating SIRT1to deacetylate HIF-1α

(A) The IF results showed that SIRT1 and HIF-1α were co-localized. (B–E) The protein expression, DNA-binding activity, transcriptional activity, and acetylation levels of HIF-1α were determined by Western blot assay after transfection with Cx43 in the presence and absence of the si-SIRT1 under HG for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG; #P<0.05, ##P<0.01 vs. HG. (F,G) The Western blot and IF results showed the expression of HIF-1α after transfection with the SIRT1 overexpression plasmid under the HG conditions for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. *P<0.05, **P<0.01 vs. NG; #P<0.05, ##P<0.01 vs. HG. (H–J) The DNA-binding activity, transcriptional activity, and acetylation levels of HIF-1α were determined by Western blot assay after transfection with the SIRT1 overexpression plasmid under HG for 36 h, and the whole-cell lysates were extracted after 48 h of transfection. Ctrl: control. Independent experiments were performed three times with similar results.

To further verify that SIRT1 affects HIF-1α, we overexpressed SIRT1 and measured the expression and activity changes of HIF-1α. The Western blot data and IF results showed that the overexpression of SIRT1, but not that of the vector, could reduce the protein expression of HIF-1α, down-regulate the DNA-binding activity, transcriptional activity and acetylation level of HIF-1α in HG-treated NRK-52E cells (Figure 7F-J), which demonstrated that SIRT1 deacetylates HIF-1α to decrease its activity.

Taken together, these results indicated that Cx43 up-regulated SIRT1 levels to deacetylate its target, thus inhibiting the occurrence of renal EMT and improving RIF.

Caudal vein injection of Cx43 adenovirus ameliorated renal fibrosis in the db/db spontaneous diabetic model mice

All the above results have confirmed that Cx43 can inhibit the occurrence of EMT and expression of ECM components in NRK-52E cells treated with HG. To validate these in vitro findings, we further explored the role of Cx43 in the db/db spontaneous diabetic model mice. Overexpression of Cx43 significantly reduced the FBG GSP, BUN, Cr, and 24-h UP in the db/db spontaneous diabetic model mice (Figure 8A–E). The Hematoxylin- Eosin (HE) staining results revealed that Cx43 adenovirus injection could improve renal tubular ECM deposition and inflammatory cell infiltration (Figure 8F). The Periodic acid–Schiff (PAS) staining results showed that Cx43 attenuated the renal tubular hypertrophy and mesangial expansion (Figure 8G). Masson staining results revealed that Cx43 reduced the deposition of collagen fibers in the renal tubular of diabetic mice (Figure 8H). These findings demonstrated that Cx43 activation ameliorated renal injury in db/db spontaneous diabetic model mice.

Caudal vein injection of Cx43 adenovirus ameliorated renal fibrosis in the db/db spontaneous diabetes model mice

Figure 8
Caudal vein injection of Cx43 adenovirus ameliorated renal fibrosis in the db/db spontaneous diabetes model mice

(A–E) The FBG, GSP, BUN, Cr, and 24-h UP were evaluated to assess renal injury (n=11). *P<0.05, **P<0.01, ***P<0.001 vs. Control + Ad-V; #P<0.05, ###P<0.001 vs. db/db + Ad-V. (F–H) Renal tubular histopathological analysis in the kidneys of the mice was performed using HE staining, Masson staining, and PAS staining (400× magnification). (I) The expression levels of Cx43, E-cadherin, α-SMA, and FN, in the renal tubules were evaluated using IHC staining (400× magnification). (J) The expression levels of Cx43, E-cadherin, ZO-1, α-SMA, vimentin, FN, Collagen I, and Collagen IV in the kidney tissues of mice were evaluated using Western blot assay (n=11). *P<0.05, **P<0.01 vs. Control + Ad-V; #P<0.05 vs. db/db + Ad-V. ( K,L) The expression levels of SIRT1 and HIF-1α in the renal tubules are shown by immunohistochemical staining (400× magnification) and Western blot assay (n=11). *P<0.05, ***P<0.001 vs. Control + Ad-V; #P<0.05, ###P<0.001, vs. db/db + Ad-V. Independent experiments were performed three times with similar results. Abbreviations: Ad-C, adenovirus-Cx43; Ad-V, adenovirus-vector; IHC, immunohistochemistry.

Figure 8
Caudal vein injection of Cx43 adenovirus ameliorated renal fibrosis in the db/db spontaneous diabetes model mice

(A–E) The FBG, GSP, BUN, Cr, and 24-h UP were evaluated to assess renal injury (n=11). *P<0.05, **P<0.01, ***P<0.001 vs. Control + Ad-V; #P<0.05, ###P<0.001 vs. db/db + Ad-V. (F–H) Renal tubular histopathological analysis in the kidneys of the mice was performed using HE staining, Masson staining, and PAS staining (400× magnification). (I) The expression levels of Cx43, E-cadherin, α-SMA, and FN, in the renal tubules were evaluated using IHC staining (400× magnification). (J) The expression levels of Cx43, E-cadherin, ZO-1, α-SMA, vimentin, FN, Collagen I, and Collagen IV in the kidney tissues of mice were evaluated using Western blot assay (n=11). *P<0.05, **P<0.01 vs. Control + Ad-V; #P<0.05 vs. db/db + Ad-V. ( K,L) The expression levels of SIRT1 and HIF-1α in the renal tubules are shown by immunohistochemical staining (400× magnification) and Western blot assay (n=11). *P<0.05, ***P<0.001 vs. Control + Ad-V; #P<0.05, ###P<0.001, vs. db/db + Ad-V. Independent experiments were performed three times with similar results. Abbreviations: Ad-C, adenovirus-Cx43; Ad-V, adenovirus-vector; IHC, immunohistochemistry.

Additionally, concomitant with the up-regulation of Cx43 expression, Cx43 adenovirus injection increased the E-cadherin protein levels but decreased the α-SMA and FN protein levels of diabetic mice as shown by the immunohistochemistry (IHC) staining results (Figure 8I). Furthermore, the Western blot data in Figure 8J revealed that Cx43 overexpression increased the E-cadherin and ZO-1 protein levels but decreased the α-SMA, vimentin, FN, Collagen I, and Collagen IV protein levels in the kidney tissues of db/db diabetic mice, which indicated that Cx43 prevented the EMT process and RIF. Additionally, we observed that Cx43 restored the expression of SIRT1 and down-regulated the expression of HIF-1α in the kidneys of db/db diabetic mice as shown by the western blot assay and IHC staining results (Figure 8K,L), which confirmed that Cx43 prevents the progression of diabetic RIF by regulating the SIRT1-HIF-1α signaling pathway.

Discussion

Diabetic renal fibrosis is a combination of glomerular sclerosis and RIF [6]. The tubulointerstitium, which has many vital functions, occupies more than 90% of the renal parenchyma [46]. It is generally believed that the primary factor of chronic kidney disease including DN is glomerular injury, and the tubular injury is secondary to glomerular injury [47]. Recent studies have implied that the tubular injury might be the originating factor of renal disease [47], and suggested that RIF, characterized by EMT of the TECs, plays an important role in the pathological process of diabetic renal fibrosis [48].

In 1994, Barajas et al. found that Cx43 localizes in renal vasculature, mesangial cells as well as collecting ducts [49]. Guo et al. reported the expression of Cx43 in tubular segments in 1998 [50]. After that, a number of studies suggested a significance of Cx in the renal circulation [51]. Cx43 is considered as the most widely expressed Cx and was identified as an important player in the renal disease [52]. Our previous studies mainly focused on the regulation of Cx43 in GMCs in which Cx43 inhibits the pathological development of diabetic renal fibrosis by preventing GMCs hypertrophy, increasing the nuclear aggregation, DNA-binding activity, and transcriptional activity of Nrf2, and thus reducing the expression of inflammatory and fibrotic components, such as FN, TGF-β1, and intercellular adhesion molecule-1 [18–20]. Based on the vital role of RIF as characterized by TECs EMT in the development of diabetic renal fibrosis, the current study was undertaken to explore whether Cx43 could regulate the EMT of TECs to improve RIF and to explore the underlying mechanism, thus providing more sufficient experimental evidence for Cx43 as a potential target of DN.

In the present study, the effects of Cx43 on EMT inhibition and RIF improvement were verified in both an in vivo animal model and an in vitro cell model of diabetic renal fibrosis. In the kidney tissues of the db/db diabetic model mice and HG-induced TECs, Cx43 expression was down-regulated, concomitant with decreased expression of the epithelial markers E-cadherin and ZO-1, increased expression of the mesenchymal markers α-SMA and vimentin, and elevated levels of ECM components including FN, Collagen I, and Collagen IV. These results indicated that Cx43 may regulate the pathological process of HG-induced EMT of TECs. Further study revealed that Cx43 overexpression in the above models prevent renal EMT, reduced ECM components, and improve in RIF. In addition, Cx43 depletion further aggravated the EMT process and promoted the expression of ECM components. The Cx43 C-terminus can interact with multiple signal proteins to regulate cell adhesion, migration, proliferation, and other processes, which is independent of its well-established gap junction communication function on the cell membrane [53]. Here, we found that transfection with either Cx43 or Cx43CT, but not with Cx43ΔCT that lacks the C-terminus, could reverse the EMT process and reduce the ECM components levels in the NRK-52E cells that were challenged with HG. These data demonstrated that the effects of Cx43 on EMT are mainly related to its C-terminal signal transduction rather than its gap junction communication function. These channel-independent effects of Cx43 might be mediated via interactions with other proteins.

SIRT1, the most extensively studied Sirtuin, plays an important role in the DNA damage response, carcinogenesis, and lifespan regulation through its NAD+-dependent deacetylase activity [54]. It was documented that resveratrol, an agonist of SIRT1, can improve DN by fighting oxidative stress [55]. Studies have shown that SIRT1 reduces EMT in fibrosis and the loss of SIRT1 in kidney tubular epithelial cells exacerbates kidney fibrosis, suggesting that SIRT1 plays an important protective role in the kidney [28]. During the past decade, a number of SIRT1-activating compounds have been synthesized and analyzed and many protein modulators of SIRT1 have also been identified, such as the necdin and the AROS (active regulator of SIRT1) [56–58]. The subcellular localization of mammalian sirtuins, which are found in numerous compartments within the cell, probably depends on cell type, molecular interactions as well as stress status [59]. For example, SIRT1 were found to shuttle between the nucleus and cytoplasm to interact with both nuclear and cytosolic proteins [60]. Cx43, as the component of gap junctions, provides direct links between adjacent cells under many physiological processes, which consists of four membrane-spanning domains with two conserved extracellular subdomains that are the essential elements of the docking process, one cytoplasmic loop, and N- and C-terminal ends facing the cytosol [61]. In addition to the canonical role as a component of gap junctions, the C-terminal tail of Cx43 contains multiple protein-protein interaction motifs and target sites including Src, ZO-1, microtubules, CCN3, and CIP85 [9]. Besides, study has shown that the C-terminal fragment of Cx43 that localizes at the nucleus can play a direct role as a transcriptional regulator of gene expression, such as N-cadherin in vivo [62], which suggests that the C-terminal fragment of Cx43 may regulate the expression of SIRT1 in the nucleus. Based on these foundations, we speculate that the C-terminal tail of Cx43 probably regulate SIRT1 through its target sites or transcriptional regulator role. Here, we found that Cx43 and SIRT1 co-localized in the cytoplasm of the NRK-52E cells and overexpression of Cx43 up-regulates the expression of SIRT1 through its C-terminus. Additionally, SIRT1 interference further aggravated the occurrence of EMT and promoted the expression of ECM components, which suggested that SIRT1 plays a protective role in TECs under the HG conditions. However, the inhibitory effects of Cx43 in the EMT process and the expression of ECM components were blunted by the simultaneous transfection of Cx43 and si-SIRT1in the NRK-52E cells that were treated with HG, which suggested that Cx43 prevented EMT by up-regulating the expression of SIRT1. However, the underlying molecular mechanism driving this regulation remains unknown.

Magnetic resonance imaging of diabetic animals and patients with chronic kidney disease has been reported to show hypoxia in the kidneys, therefore hypoxia may be an important factor that leads to DN development [63]. HIF-1α, a nuclear factor with transcriptional activity, can translocate to the nucleus upon hypoxia stimuli [64]. Furthermore, HIF-1α functions as a key mediator in cell metabolism, inflammation, and tumors under the low oxygen environment [65]. The mechanism for oxygen regulation by HIF-1α has been well established [64]. Under the DN conditions, the glomerular microvasculature in the peripheral tissues gradually becomes sclerotic, which is associated with an insufficient blood supply [66]. In addition, the expression of HIF-1α in functional cells was increased under the conditions of ischemia and hypoxia in tissues [67]. Studies suggested that HIF-1α activation promotes EMT and renal fibrogenesis [68]. It was reported that HIF-1α regulates Snail activation by binding directly to the hypoxia-response element in the proximal promoter of Snail, which triggers the EMT by repressing the expression of tight junction proteins, such as E-cadherin [69]. Here, we found that the HIF-1α expression was significantly up-regulated in both the kidney tissues of db/db diabetic mice and NRK-52E cells treated with HG, which was consistent with the literature reports [31]. Furthermore, the protein expression, DNA-binding activity, transcriptional activity, and acetylation levels of HIF-1α were found to be down-regulated by either Cx43 or SIRT1 overexpression in the NRK-52E cells that were challenged with HG. However, the effects of Cx43 were cancelled out when the cells were simultaneously transfected with Cx43 and si-SIRT1, which suggested that Cx43 up-regulates SIRT1 to deacetylate HIF-1α, decrease the activity of HIF1-α, and ultimately inhibit the occurrence of renal EMT and improve RIF.

Conclusion

In summary, the in vitro and in vivo experiments demonstrated that Cx43 ameliorated the pathological progress of diabetic RIF through its C-terminus by increasing SIRT1 expression and promoted SIRT1-mediated deacetylation of HIF-1α, thereby decreasing HIF-1α activity. We further demonstrated that Cx43 prevented the progression of diabetic RIF by regulating the SIRT1-HIF-1α signaling pathway, which provided new experimental evidence for the potential clinical application of Cx43 as an anti-DN drug.

Clinical perspectives

  • Hyperglycemia-induced renal EMT is a key pathological factor in diabetic RIF. However, whether Cx43, an important player in the renal disease, regulates the EMT of renal TECs and the pathological process of RIF under the diabetic conditions remains unknown.

  • In the present study, Cx43 expression was detected in db/db spontaneous diabetic model mice and NRK-52E cells to evaluate its potential role in renal tubular EMT. Overexpression of Cx43 reverses the down-regulation of SIRT1 induced by HG), which promoted SIRT1-mediated deacetylation of HIF-1α, thereby decreasing HIF-1α activity and eventually effectively suppressed the EMT process and expression of ECM components.

  • These results have expanded our knowledge of Cx43 and provided new experimental evidence for the potential clinical application of Cx43 as an anti-DN drug.

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Funding

This work was supported by the National Natural Foundation of China [grant numbers 81603168, 81573477, 81770816, 81973375]; the Key Projects of Guangdong Natural Science Foundation [grant number 2017A030311036]; the Guangdong Provincial Key Laboratory of Construction Foundation [grant number 2017B030314030]; and the Natural Science Foundation of Guangdong Province, China [grant number 2017A030313678].

Author Contribution

Xiaohong Sun and Kaipeng Huang analyzed the data. Xiaohong Sun, Kaipeng Huang, Haiming, Xiao, Zeyuan Lin, Yan Yang and Meng Zhang performed the research. Xiaohong Sun, Kaipeng Huang and Heqing Huang designed the research and wrote the manuscript. Kaipeng Huang, Peiqing Liu and Heqing Huang conceived and supervised the study. All authors have reviewed and approved the final version of the manuscript.

Abbreviations

     
  • aa

    amino acid

  •  
  • BUN

    blood urea nitrogen

  •  
  • Cr

    creatinine

  •  
  • Cx

    connexin

  •  
  • Cx43

    connexin 43

  •  
  • c-Src

    non-receptor tyrosine kinase-c-src

  •  
  • DAPI

    diamidino-2-phenylindole

  •  
  • dIdC

    poly (dI-dC)

  •  
  • DN

    diabetic nephropathy

  •  
  • ECM

    extracellular matrix

  •  
  • EMSA

    electrophoretic mobility shift assay

  •  
  • EMT

    epithelial-to-mesenchymal transition

  •  
  • FBG

    fasting blood glucose

  •  
  • FN

    fibronectin

  •  
  • GMC

    glomerular mesangial cell

  •  
  • GSP

    glycosylated serum protein

  •  
  • HG

    high glucose

  •  
  • HIF

    hypoxia-inducible factor

  •  
  • HRP

    horseradish peroxidase

  •  
  • IF

    immunofluorescence

  •  
  • IHC

    immunohistochemistry

  •  
  • NAD+

    nicotinamide-adenine-dinucleotide

  •  
  • NC

    negative control

  •  
  • NG

    normal glucose

  •  
  • Nrf2

    nuclear factor E2-related factor 2

  •  
  • PBS

    phosphate-buffered saline

  •  
  • PVDF

    polyvinylidene difluoride

  •  
  • RIF

    renal tubulointerstitial fibrosis

  •  
  • SDS

    sodium dodecyl sulfate

  •  
  • shRNA

    short hairpin RNA

  •  
  • siRNA

    small interfering RNA

  •  
  • SIRT1

    sirtuin 1

  •  
  • SL/DT

    scrape loading/dye transfer

  •  
  • STZ

    streptozotocin

  •  
  • TEC

    renal tubular epithelial cell

  •  
  • TGF

    transforming growth factor

  •  
  • UP

    urinary protein

  •  
  • WT

    wild-type

  •  
  • ZO-1

    zonula occludens-1

  •  
  • α-SMA

    α-smooth muscle actin

References

References
1.
Zheng
Y.
,
Ley
S.H.
and
Hu
F.B.
(
2018
)
Global aetiology and epidemiology of type 2 diabetes mellitus and its complications
.
Nat. Rev. Endocrinol.
14
,
88
98
[PubMed]
2.
Azushima
K.
,
Gurley
S.B.
and
Coffman
T.M.
(
2018
)
Modelling diabetic nephropathy in mice
.
Nat. Rev. Nephrol.
14
,
48
56
[PubMed]
3.
Qian
Y.
,
Feldman
E.
,
Pennathur
S.
,
Kretzler
M.
and
Brosius
F.C.
III
(
2008
)
From fibrosis to sclerosis: mechanisms of glomerulosclerosis in diabetic nephropathy
.
Diabetes
57
,
1439
1445
[PubMed]
4.
Wada
J.
and
Makino
H.
(
2013
)
Inflammation and the pathogenesis of diabetic nephropathy
.
Clin. Sci.
124
,
139
152
[PubMed]
5.
Humphreys
B.D.
,
Valerius
M.T.
,
Kobayashi
A.
,
Mugford
J.W.
,
Soeung
S.
,
Duffield
J.S.
et al.
(
2008
)
Intrinsic epithelial cells repair the kidney after injury
.
Cell Stem Cell
2
,
284
291
[PubMed]
6.
Kanwar
Y.S.
,
Sun
L.
,
Xie
P.
,
Liu
F.Y.
and
Chen
S.
(
2011
)
A glimpse of various pathogenetic mechanisms of diabetic nephropathy
.
Annu. Rev. Pathol. Mech.
6
,
395
423
7.
Simonson
M.S.
(
2007
)
Phenotypic transitions and fibrosis in diabetic nephropathy
.
Kidney Int.
71
,
846
854
[PubMed]
8.
Bai
X.Y.
,
Geng
J.
,
Zhou
Z.M.
,
Tian
J.W.
and
Li
X.
(
2016
)
MicroRNA-130b improves renal tubulointerstitial fibrosis via repression of Snail-induced epithelial-mesenchymal transition in diabetic nephropathy
.
Sci. Rep.
6
,
1
16
9.
Giepmans
B.N.G.
(
2006
)
Role of connexin43-interacting proteins at gap junctions
.
Adv. Cardiol.
42
,
41
56
[PubMed]
10.
Cotrina
M.L.
,
Lin
J.H.
and
Nedergaard
M.
(
2008
)
Adhesive properties of connexin hemichannels
.
Glia
56
,
1791
1798
[PubMed]
11.
Luo
C.
,
Yuan
D.
,
Li
X.
,
Yao
W.
,
Luo
G.
,
Chi
X.
et al.
(
2015
)
Propofol attenuated acute kidney injury after orthotopic liver transplantation via inhibiting gap junction composed of connexin 32
.
Anesthesiology
122
,
72
86
[PubMed]
12.
Harris
A.L.
(
2007
)
Connexin channel permeability to cytoplasmic molecules
.
Prog. Biophys. Mol. Biol.
94
,
120
143
[PubMed]
13.
Herve
J.C.
,
Bourmeyster
N.
and
Sarrouilhe
D.
(
2004
)
Diversity in protein-protein interactions of connexins: emerging roles
.
Biochim. Biophys. Acta
1662
,
22
41
[PubMed]
14.
Takenaka
T.
,
Inoue
T.
,
Okada
H.
,
Ohno
Y.
,
Miyazaki
T.
,
Chaston
D.J.
et al.
(
2011
)
Altered gap junctional communication and renal haemodynamics in Zucker fatty rat model of type 2 diabetes
.
Diabetologia
54
,
2192
2201
[PubMed]
15.
Zeng
O.
,
Li
F.
,
Li
Y.
,
Li
L.
,
Xiao
T.
,
Chu
C.
et al.
(
2016
)
Effect of Novel Gasotransmitter hydrogen sulfide on renal fibrosis and connexins expression in diabetic rats
.
Bioengineered
7
,
314
320
[PubMed]
16.
Haefliger
J.A.
,
Krattinger
N.
,
Martin
D.
,
Pedrazzini
T.
,
Capponi
A.
,
Doring
B.
et al.
(
2006
)
Connexin43-dependent mechanism modulates renin secretion and hypertension
.
J. Clin. Invest.
116
,
405
413
[PubMed]
17.
Li
A.F.
and
Roy
S.
(
2009
)
High glucose-induced downregulation of Connexin 43 expression promotes apoptosis in microvascular endothelial cells
.
Invest. Ophthalmol. Vis. Sci.
50
,
1400
1407
18.
Xie
X.
,
Lan
T.
,
Chang
X.T.
,
Huang
K.P.
,
Huang
J.
,
Wang
S.G.
et al.
(
2013
)
Connexin43 mediates NF-kappa B signalling activation induced by high glucose in GMCs: involvement of c-Src
.
Cell Commun. Signal.
11
,
1
17
[PubMed]
19.
Xie
X.
,
Chen
C.
,
Huang
K.P.
,
Wang
S.G.
,
Hao
J.
,
Huang
J.Y.
et al.
(
2014
)
RhoA/rho kinase signaling reduces connexin43 expression in high glucose-treated glomerular mesangial cells with zonula occludens-1 involvement
.
Exp. Cell Res.
327
,
276
286
[PubMed]
20.
Chen
Z.
,
Xie
X.
,
Huang
J.
,
Gong
W.
,
Zhu
X.
,
Chen
Q.
et al.
(
2017
)
Connexin43 regulates high glucose-induced expression of fibronectin, ICAM-1 and TGF-beta1 via Nrf2/ARE pathway in glomerular mesangial cells
.
Free Radic. Biol. Med.
102
,
77
86
21.
Rahman
S.
and
Islam
R.
(
2011
)
Mammalian Sirt1: insights on its biological functions
.
Cell Commun. Signal.
9
,
1
8
[PubMed]
22.
Hasegawa
K.
,
Wakino
S.
,
Yoshioka
K.
,
Tatematsu
S.
,
Hara
Y.
,
Minakuchi
H.
et al.
(
2010
)
Kidney-specific overexpression of Sirt1 protects against acute kidney injury by retaining peroxisome function
.
J. Biol. Chem.
285
,
13045
13056
[PubMed]
23.
He
W.
,
Wang
Y.
,
Zhang
M.Z.
,
You
L.
,
Davis
L.S.
,
Fan
H.
et al.
(
2010
)
Sirt1 activation protects the mouse renal medulla from oxidative injury
.
J. Clin. Invest.
120
,
1056
1068
[PubMed]
24.
Kume
S.
,
Uzu
T.
,
Horiike
K.
,
Chin-Kanasaki
M.
,
Isshiki
K.
,
Araki
S.
et al.
(
2010
)
Calorie restriction enhances cell adaptation to hypoxia through Sirt1-dependent mitochondrial autophagy in mouse aged kidney
.
J. Clin. Invest.
120
,
1043
1055
[PubMed]
25.
Liu
R.J.
,
Zhong
Y.F.
,
Li
X.Z.
,
Chen
H.B.
,
Jim
B.
,
Zhou
M.M.
et al.
(
2014
)
Role of transcription factor acetylation in diabetic kidney disease
.
Diabetes
63
,
2440
2453
[PubMed]
26.
Chuang
P.Y.
,
Xu
J.
,
Dai
Y.
,
Jia
F.
,
Mallipattu
S.K.
,
Yacoub
R.
et al.
(
2014
)
In vivo RNA interference models of inducible and reversible Sirt1 knockdown in kidney cells
.
Am. J. Pathol.
184
,
1940
1956
[PubMed]
27.
Huang
K.P.
,
Chen
C.
,
Hao
J.
,
Huang
J.Y.
,
Liu
P.Q.
and
Huang
H.Q.
(
2015
)
AGEs-RAGE system down-regulates Sirt1 through the ubiquitin-proteasome pathway to promote FN and TGF-beta 1 expression in male rat glomerular mesangial cells
.
Endocrinology
156
,
268
279
[PubMed]
28.
Simic
P.
,
Williams
E.O.
,
Bell
E.L.
,
Gong
J.J.
,
Bonkowski
M.
and
Guarente
L.
(
2013
)
SIRT1 suppresses the epithelial-to-mesenchymal transition in cancer metastasis and organ fibrosis
.
Cell Rep.
3
,
1175
1186
[PubMed]
29.
Sun
Z.Q.
,
Ma
Y.L.
,
Chen
F.
,
Wang
S.Y.
,
Chen
B.P.
and
Shi
J.
(
2018
)
miR-133b and miR-199b knockdown attenuate TGF-beta 1-induced epithelial to mesenchymal transition and renal fibrosis by targeting SIRT1 in diabetic nephropathy
.
Eur. J. Pharmacol.
837
,
96
104
[PubMed]
30.
Yuan
Y.G.
,
Chen
Y.
,
Zhang
P.
,
Huang
S.M.
,
Zhu
C.H.
,
Ding
G.X.
et al.
(
2012
)
Mitochondrial dysfunction accounts for aldosterone-induced epithelial-to-mesenchymal transition of renal proximal tubular epithelial cells
.
Free Radic. Biol. Med.
53
,
30
43
31.
Higgins
D.F.
,
Kimura
K.
,
Bernhardt
W.M.
,
Shrimanker
N.
,
Akai
Y.
,
Hohenstein
B.
et al.
(
2007
)
Hypoxia promotes fibrogenesis in vivo via HIF-1 stimulation of epithelial-to-mesenchymal transition
.
J. Clin. Invest.
117
,
3810
3820
[PubMed]
32.
Rezvani
H.R.
,
Ali
N.
,
Nissen
L.J.
,
Harfouche
G.
,
de Verneuil
H.
,
Taieb
A.
et al.
(
2011
)
HIF-1alpha in epidermis: oxygen sensing, cutaneous angiogenesis, cancer, and non-cancer disorders
.
J. Invest. Dermatol.
131
,
1793
1805
[PubMed]
33.
Konisti
S.
,
Kiriakidis
S.
and
Paleolog
E.M.
(
2012
)
Hypoxia–a key regulator of angiogenesis and inflammation in rheumatoid arthritis
.
Nat. Rev. Rheumatol.
8
,
153
162
[PubMed]
34.
Rohwer
N.
,
Zasada
C.
,
Kempa
S.
and
Cramer
T.
(
2013
)
The growing complexity of HIF-1alpha’s role in tumorigenesis: DNA repair and beyond
.
Oncogene
32
,
3569
3576
[PubMed]
35.
Isoe
T.
,
Makino
Y.
,
Mizumoto
K.
,
Sakagami
H.
,
Fujita
Y.
,
Honjo
J.
et al.
(
2010
)
High glucose activates HIF-1-mediated signal transduction in glomerular mesangial cells through a carbohydrate response element binding protein
.
Kidney Int.
78
,
48
59
[PubMed]
36.
Ryu
D.R.
,
Yu
M.R.
,
Kong
K.H.
,
Kim
H.
,
Kwon
S.H.
,
Jeon
J.S.
et al.
(
2019
)
Sirt1-hypoxia-inducible factor-1alpha interaction is a key mediator of tubulointerstitial damage in the aged kidney
.
Aging Cell
18
,
e12904
[PubMed]
37.
Ohkubo
T.
and
Ozawa
M.
(
2004
)
The transcription factor Snail downregulates the tight junction components independently of E-cadherin downregulation
.
J. Cell Sci.
117
,
1675
1685
[PubMed]
38.
Kim
B.J.
,
Hancock
B.M.
,
Bermudez
A.
,
Del Cid
N.
,
Reyes
E.
,
van Sorge
N.M.
et al.
(
2015
)
Bacterial induction of Snail1 contributes to blood-brain barrier disruption
.
J. Clin. Invest.
125
,
2473
2483
[PubMed]
39.
Zhu
Y.
,
Tan
J.
,
Xie
H.
,
Wang
J.
,
Meng
X.
and
Wang
R.
(
2016
)
HIF-1alpha regulates EMT via the Snail and beta-catenin pathways in paraquat poisoning-induced early pulmonary fibrosis
.
J. Cell. Mol. Med.
20
,
688
697
[PubMed]
40.
Joo
H.Y.
,
Yun
M.
,
Jeong
J.
,
Park
E.R.
,
Shin
H.J.
,
Woo
S.R.
et al.
(
2015
)
SIRT1 deacetylates and stabilizes hypoxia-inducible factor-1alpha (HIF-1alpha) via direct interactions during hypoxia
.
Biochem. Biophys. Res. Commun.
462
,
294
300
[PubMed]
41.
Lim
J.H.
,
Lee
Y.M.
,
Chun
Y.S.
,
Chen
J.
,
Kim
J.E.
and
Park
J.W.
(
2010
)
Sirtuin 1 modulates cellular responses to hypoxia by deacetylating hypoxia-inducible factor 1alpha
.
Mol. Cell
38
,
864
878
[PubMed]
42.
Xie
Y.B.
,
Park
J.H.
,
Kim
D.K.
,
Hwang
J.H.
,
Oh
S.
,
Park
S.B.
et al.
(
2009
)
Transcriptional corepressor SMILE recruits SIRT1 to inhibit nuclear receptor estrogen receptor-related receptor gamma transactivation
.
J. Biol. Chem.
284
,
28762
28774
[PubMed]
43.
Jiang
Q.
,
Liu
P.Q.
,
Wu
X.Q.
,
Liu
W.H.
,
Shen
X.Y.
,
Lan
T.A.
et al.
(
2011
)
Berberine attenuates lipopolysaccharide-induced extracelluar matrix accumulation and inflammation in rat mesangial cells: involvement of NF-kappa B signaling pathway
.
Mol. Cell. Endocrinol.
331
,
34
40
[PubMed]
44.
Babica
P.
,
Sovadinova
I.
and
Upham
B.L.
(
2016
)
Scrape loading/dye transfer assay
.
Methods Mol. Biol.
1437
,
133
144
[PubMed]
45.
Basheer
W.
and
Shaw
R.
(
2016
)
The “tail” of Connexin43: an unexpected journey from alternative translation to trafficking
.
Biochim. Biophys. Acta
1863
,
1848
1856
[PubMed]
46.
Gilbert
R.E.
and
Cooper
M.E.
(
1999
)
The tubulointerstitium in progressive diabetic kidney disease: More than an aftermath of glomerular injury?
Kidney Int.
56
,
1627
1637
[PubMed]
47.
Lim
B.J.
,
Yang
J.W.
,
Zou
J.
,
Zhong
J.Y.
,
Matsusaka
T.
,
Pastan
I.
et al.
(
2017
)
Tubulointerstitial fibrosis can sensitize the kidney to subsequent glomerular injury
.
Kidney Int.
92
,
1395
1403
[PubMed]
48.
Venkatachalam
M.A.
,
Weinberg
J.M.
,
Kriz
W.
and
Bidani
A.K.
(
2015
)
Failed tubule recovery, AKI-CKD transition, and kidney disease progression
.
J. Am. Soc. Nephrol.
26
,
1765
1776
[PubMed]
49.
Barajas
L.
,
Liu
L.
and
Tucker
M.
(
1994
)
Localization of connexin43 in rat kidney
.
Kidney Int.
46
,
621
626
[PubMed]
50.
Guo
R.
,
Liu
L.
and
Barajas
L.
(
1998
)
RT-PCR study of the distribution of connexin 43 mRNA in the glomerulus and renal tubular segments
.
Am. J. Physiol.
275
,
R439
R447
[PubMed]
51.
Wagner
C.
(
2008
)
Function of connexins in the renal circulation
.
Kidney Int.
73
,
547
555
[PubMed]
52.
Prakoura
N.
,
Kavvadas
P.
and
Chadjichristos
C.E.
(
2018
)
Connexin 43: a new therapeutic target against chronic kidney disease
.
Cell. Physiol. Biochem.
49
,
998
1009
53.
Leithe
E.
(
2016
)
Regulation of connexins by the ubiquitin system: Implications for intercellular communication and cancer
.
Biochim. Biophys. Acta Rev. Cancer
1865
,
133
146
54.
Lo
C.S.
,
Shi
Y.X.
,
Chenier
I.
,
Ghosh
A.
,
Wu
C.H.
,
Cailhier
J.F.
et al.
(
2017
)
Heterogeneous nuclear ribonucleoprotein F stimulates Sirtuin-1 gene expression and attenuates nephropathy progression in diabetic mice
.
Diabetes
66
,
1964
1978
[PubMed]
55.
Cannito
S.
,
Novo
E.
,
di Bonzo
L.V.
,
Busletta
C.
,
Colombatto
S.
and
Parola
M.
(
2010
)
Epithelial-mesenchymal transition: from molecular mechanisms, redox regulation to implications in human health and disease
.
Antioxid. Redox Signal.
12
,
1383
1430
56.
Lavu
S.
,
Boss
O.
,
Elliott
P.J.
and
Lambert
P.D.
(
2008
)
Sirtuins–novel therapeutic targets to treat age-associated diseases
.
Nat. Rev. Drug Discov.
7
,
841
853
[PubMed]
57.
Hasegawa
K.
and
Yoshikawa
K.
(
2008
)
Necdin regulates p53 acetylation via sirtuin1 to modulate DNA damage response in cortical neurons
.
J. Neurosci.
28
,
8772
8784
[PubMed]
58.
Kim
E.J.
,
Kho
J.H.
,
Kang
M.R.
and
Um
S.J.
(
2007
)
Active regulator of SIRT1 cooperates with SIRT1 and facilitates suppression of p53 activity
.
Mol. Cell
28
,
277
290
[PubMed]
59.
Haigis
M.C.
and
Sinclair
D.A.
(
2010
)
Mammalian sirtuins: biological insights and disease relevance
.
Annu. Rev. Pathol. Mech.
5
,
253
295
60.
Tanno
M.
,
Sakamoto
J.
,
Miura
T.
,
Shimamoto
K.
and
Horio
Y.
(
2007
)
Nucleocytoplasmic shuttling of the NAD+-dependent histone deacetylase SIRT1
.
J. Biol. Chem.
282
,
6823
6832
[PubMed]
61.
Maria
K.
,
Elias
H.B.
,
Jonathan
L.
,
Marc
G.
,
Verena
R.
,
Alexandra
S.
et al.
(
2018
)
Gap junction protein Connexin-43 is a direct transcriptional regulator of N-cadherin in vivo
.
Nat. Commun.
9
,
1
17
[PubMed]
62.
Sohl
G.
,
Maxeiner
S.
and
Willecke
K.
(
2005
)
Expression and functions of neuronal gap junctions
.
Nat. Rev. Neurosci.
6
,
191
200
[PubMed]
63.
Inoue
T.
,
Kozawa
E.
,
Okada
H.
,
Inukai
K.
,
Watanabe
S.
,
Kikuta
T.
et al.
(
2011
)
Noninvasive evaluation of kidney hypoxia and fibrosis using magnetic resonance imaging
.
J. Am. Soc. Nephrol.
22
,
1429
1434
[PubMed]
64.
Schofield
C.J.
and
Ratcliffe
P.J.
(
2004
)
Oxygen sensing by HIF hydroxylases
.
Nat. Rev. Mol. Cell Biol.
5
,
343
354
65.
Lefere
S.
,
Van Steenkiste
C.
,
Verhelst
X.
,
Van Vlierberghe
H.
,
Devisscher
L.
and
Geerts
A.
(
2016
)
Hypoxia-regulated mechanisms in the pathogenesis of obesity and non-alcoholic fatty liver disease
.
Cell. Mol. Life Sci.
73
,
3419
3431
[PubMed]
66.
Yang
M.H.
and
Wu
K.J.
(
2008
)
TWIST activation by hypoxia inducible factor-1 (HIF-1): implications in metastasis and development
.
Cell Cycle
7
,
2090
2096
[PubMed]
67.
Rosenberger
C.
,
Heyman
S.N.
,
Rosen
S.
,
Shina
A.
,
Goldfarb
M.
,
Griethe
W.
et al.
(
2005
)
Up-regulation of HIF in experimental acute renal failure: evidence for a protective transcriptional response to hypoxia
.
Kidney Int.
67
,
531
542
[PubMed]
68.
Higgins
D.F.
,
Kimura
K.
,
Iwano
M.
and
Haase
V.H.
(
2008
)
Hypoxia-inducible factor signaling in the development of tissue fibrosis
.
Cell Cycle
7
,
1128
1132
[PubMed]
69.
Luo
D.
,
Wang
J.
,
Li
J.
and
Post
M.
(
2011
)
Mouse snail is a target gene for HIF
.
Mol. Cancer Res.
9
,
234
245
[PubMed]

Author notes

*

These authors contributed equally to this work.