Inhibitor of myogenic differentiation family isoform a, a new positive regulator of fibronectin production by glomerular mesangial cells

Abstract

Overproduction of extracellular matrix proteins, including fibronectin by mesangial cells (MCs), contributes to diabetic nephropathy. Inhibitor of myogenic differentiation family isoform a (I-mfa) is a multifunctional cytosolic protein functioning as a transcriptional modulator or plasma channel protein regulator. However, its renal effects are unknown. The present study was conducted to determine whether I-mfa regulated fibronectin production by glomerular MCs. In human MCs, overexpression of I-mfa significantly increased fibronectin abundance. Silencing I-mfa significantly reduced the level of fibronectin mRNA and blunted transforming growth factor-β1-stimulated production of fibronectin. We further found that high glucose increased I-mfa protein content in a time course (≥48 h) and concentration (≥25 mM)-dependent manner. Although high glucose exposure increased I-mfa at the protein level, it did not significantly alter transcripts of I-mfa in MCs. Furthermore, the abundance of I-mfa protein was significantly increased in the renal cortex of rats with diabetic nephropathy. The I-mfa protein level was also elevated in the glomerulus of mice with diabetic kidney disease. However, there was no significant difference in glomerular I-mfa mRNA levels between mice with and without diabetic nephropathy. Moreover, H₂O₂ significantly increased I-mfa protein abundance in a dose-dependent manner in cultured human MCs. The antioxidants polyethylene glycol-catalase, ammonium pyrrolidithiocarbamate, and N-acetylcysteine significantly blocked the high glucose-induced increase of I-mfa protein. Taken together, our results suggest that I-mfa, increased by high glucose/diabetes through the production of reactive oxygen species, stimulates fibronectin production by MCs.

INTRODUCTION

Diabetic nephropathy (DN) is one of the most common complications of diabetes and the major cause of end-stage renal disease (24, 46). This kidney disease is characterized by mesangial extracellular matrix (ECM) accumulation, podocyte injury, and basement membrane thickening in the early stage and glomerulosclerosis and renal insufficiency in the late stage (23, 24, 36, 46). Despite the use of antihypertensive drugs and renin-angiotensin system inhibitors in patients with DN, the improvement of renal function is modest (40). New and effective treatments would, therefore, be a significant advance. Glomerular mesangial cells (MCs) and their matrix form the central stalk of the glomerulus. Dysfunction of the cells plays a pivotal role in the development of DN (45). MCs are the major source of mesangial matrix, and overproduction of ECM proteins, such as fibronectin by MCs, contributes to glomerular damage in DN (1). Thus, exploration of the molecular pathways inhibiting ECM production by MCs would help find therapeutic strategies for patients with DN.

Inhibitor of myogenic differentiation family isoform a (I-mfa) is a small cytosolic protein with a unique cysteine-rich domain, first identified as an interacting protein interacting with MyoD (8) and subsequently with the Wnt/β-catenin pathway (26, 39, 47, 48), JNK signaling (26), human immunodeficiency virus 1 TAT-dependent transcription (28), and SERTA domain proteins (27). Recently, I-mfa was found to be an endogenous inhibitor of store-operated Ca²⁺ channels in several types of cells and tissues (31, 37). Among those, I-mfa-regulated pathways, Wnt/β-catenin, JNK, and store-operated Ca²⁺ channel signaling in MCs play an important role in the pathogenesis and progression of DN (5, 7, 29, 44, 55, 57–61). However, whether I-mfa is involved in MC injury in the diabetic environment is not known. The present study was conducted to determine 1) if I-mfa stimulated production of fibronectin, a major ECM protein produced by MCs, and 2) if I-mfa was a downstream target of high glucose/hyperglycemia.

MATERIALS AND METHODS

Materials

Expression plasmids.

The expression plasmids of flag-tagged I-mfa (F-I-mfa) and hemagglutinin (HA)-tagged inhibitor of myogenic differentiation family isoform b (I-mfb) were generated by Dr. Leonidas Tsiokas (University of Oklahoma Health Sciences Center) and have been previously described by Ma et al. (31) and Ong et al. (37).

Primary antibodies.

Primary antibodies against fibronectin (rabbit polyclonal, catalog no. F3648, 1:1,000 dilution) and flag (mouse monoclonal, catalog no. F1804, 1:1,000 dilution) were purchased from Sigma-Aldrich. Anti-α-tubulin primary antibody (sc-5286) was purchased from Santa Cruz Biotechnology (Dallas, TX). Anti-HA antibody was purchased from Biolegend (mouse monoclonal, catalog no. 901501, 1:1,000 dilution). Anti-β-actin antibody was purchased from Calbiochem (mouse monoclonal, catalog no. CP01, 1:200 dilution). Anti-I-mfa polyclonal antibody was generated by Dr. Tsiokas (University of Oklahoma Health Sciences Center), as previously described by Ma et al. (31). However, the I-mfa primary antibody shown in Fig. 7 was purchased from Life Technologies (rabbit polyclonal, catalog no. PAS36599, 1:500 dilution). All commercial primary antibodies were validated by the manufacturers. However, the anti-I-mfa primary antibody made by Dr. Tsiokas was validated using I-mfa-specific siRNA, which significantly reduced (but not scrambled control siRNA) the antibody-reactive immunoblot.

Fig. 7.

Antioxidants blunted the high glucose (HG)-induced increase of inhibitor of myogenic differentiation family isoform a (I-mfa) protein abundance in human mesangial cells (MCs). A: representative immunoblots showing I-mfa protein abundance in human MCs exposed to either 5.6 mM d-glucose + 20 mM α-mannitol [normal glucose (NG)], 25 mM d-glucose (HG) alone, 25 mM d-glucose + 300 U/ml Polyethylene glycol-catalase (HG + Cat), 25 mM d-glucose + 75 µM ammonium pyrrolidithiocarbamate (PDTC), and 25 mM d-glucose + 4 mM N-acetylcysteine (NAC) for 2 days. MCs were growth arrested by 0.5% FBS overnight before various treatments were applied. Tubulin served as a loading control. B: summary data from the experiments shown in A. *P < 0.05 compared with all other groups. n = 5 independent experiments.

siRNA oligonucleotides.

siRNAs against human I-mfa were obtained from Integrated DNA Technologies (Coralville, IA) (sense: 5′-GCGAGUUCCUGACGCUGUGCAACAT-3′ and antisense: 5′-GUUGCACAGCGUCAGGAACUCGC-3′). Universal scrambled siRNAs were purchased from Dharmacon (catalog no. D-001810-01-20).

PCR primers.

All primers were synthesized by Integrated DNA Technologies. The sequences of all primers used in the present study are shown in Table 1.

Table 1.

Primers used for real-time PCR

Gene Name	Forward Primer	Reverse Primer
Human I-mfa	5′-GGGCAGCAAGAAGAGTAAGA-3′	5′-GTCAGGAACTCGCAGAACA-3′
Mouse I-mfa	5′-CGCAGTCCAGGAGGATGTTACAGA-3′	5′-GCCCACCGGAAGTTGCAGACG-3′
Human fibronectin	5′-TGGACCAAGTTGATGACACC-3′	5′-CACCAGGTTGCAAGTCACTG-3′
Human β-actin	5′-GAGCTACGAGCTGCCTGAC-3′	5′-GACTCCATGCCCAGGAAG-3′
Mouse β-actin	5′-CGGTTCCGATGCCCTGAGGCTCTT-3′	5′-CGTCACACTTCATGATGGAATTGA-3′

I-mfa, inhibitor of myogenic differentiation family isoform a.

Glomerular RNA extracts.

Mouse glomerular RNA extracts from wild-type (WT) and MCK-KR-hIGF-IR (MKR) mice were donated by Dr. Sandeep Mallipattu (Department of Medicine, Mount Sinai School of Medicine). The approaches for isolating glomeruli and extracting RNAs have been previously described by Fu et al. (13).

Chemicals.

Human recombinant transforming growth factor (TGF)-β1 (catalog no. 240-B-002) was purchased from R&D Systems. Polyethylene glycol-catalase (catalog no. C4963), ammonium pyrrolidithiocarbamate (catalog no. P8765), N-acetylcysteine (catalog no. A7250), and urea H₂O₂ (catalog no. 289132) were purchased from Sigma-Aldrich.

Animals

WT and MKR mice were used to extract glomerular RNA (Dr. Sandeep Mallipattu, Mount Sinai School of Medicine). The derivation and characterization of MKR mice, which bear a dominant negative insulin-like growth factor-I receptor specifically targeted in skeletal muscle, have been previously described by Dr. Mallipattu (12, 35). These mice develop type 2 diabetes mellitus due to functional inactivation of the insulin-like growth factor-I receptor in skeletal muscle (12). MKR mice under stress, such as when fed a high-fat diet (HFD), develop progressive diabetic nephropathy secondary to type 2 diabetes mellitus (35). All mice used in these experiments were male mice on the FVB/n background (35).

Male Sprague-Dawley rats fed with a low-fat diet (LFD) and HFD plus streptozotocin (STZ) treatment were used to extract cortical proteins. The protocols of feeding the LFD and HFD as well as STZ treatment have been described in our previous article (7).

All animal procedures were approved by Institutional Animal Care and Use Committee of either the University of North Texas Health Science Center or Mount Sinai School of Medicine. The National Institutes of Health Guide for the Care and Use of Laboratory Animals was followed strictly.

Cell Culture

Human MCs were purchased from Sciencell Research Laboratories (catalog no. 4200, Carlsbad, CA). MCs were cultured in 5.6 mM d-glucose-containing DMEM (GIBCO, Carlsbad, CA) supplemented with 25 mM HEPES, 4 mM glutamine, 1.0 mM sodium pyruvate, 0.1 mM nonessential amino acids, 100 U/ml penicillin, 100 µg/ml streptomycin, and 15% FBS. Subconfluent cells (~80% confluence) were growth arrested with 0.5% serum overnight and were undertaken with various treatments as specified in the figures. Culture media were replaced with fresh media every 2 days. Only subpassages 4–9 of MCs were used in the present study.

Transient Transfection

siRNAs against human I-mfa and scrambled control siRNAs (both 50 nM) were transfected into human MCs using Dharmafect 2 transfection reagent (Thermo Scientific, Rockford, IL) in serum-free DMEM following the protocol provided by the manufacturer. Media were changed to 15% FBS-containing DMEM after 6 h. Cells were harvested for Western blot analysis 48 h after transfection. Flag-I-mfa and HA-I-mfb expression plasmids were transfected into MCs at 0.5 µg/ml using Lipofectamine and PLUS reagent (Invitrogen-BRL, Carlsbad, CA) following the protocols provided by the manufacturer. Cells were harvested 48 h after transfection for immunoblot analysis.

Quantitative Real-Time RT-PCR

Total RNA was isolated from cultured human MCs using the PerfectPure RNA cultured cell kit (5 Prime, Hamburg, Germany) following the manufacturer’s protocol. A total of 1.0 μg RNA from MCs or mouse glomerular extracts at a final volume of 20 μl was used for reverse transcription reactions using the iScript cDNA synthesis kit (Bio-Rad) following the manufacturer’s reaction protocol. A total of 0.2 μg reverse transcription product and 100 nM primer was used for real-time PCR, which was performed using iQ SYBR Green supermix (Bio-Rad) at a final volume of 20 μl. The PCR mix was denatured at 95°C for 10 min followed by 45 cycles of melting at 95°C for 15 s, annealing at 57°C for 10 s, and elongation at 72°C for 15 s. After amplification, a melting curve analysis from 65 to 95°C with a heating rate of 0.02°C/s with a continuous fluorescence acquisition was made. The assay was run on a C1000 Thermal Cycler (Bio-Rad). The average threshold cycle (C_t) of the fluorescence unit was used to analyze mRNA levels. Levels of interest mRNAs were normalized by their corresponding β-actin mRNA levels. Quantification was calculated as follows: interest mRNA levels = $2^{\Delta {\text{C}}_{\text{t}}}$, where ΔC_t = C_t,interest – C_t,actin.

Western Blot Analysis

Whole cell lysates or renal cortical extracts were fractionated by 10% SDS-PAGE, transferred to PVDF membranes, and probed with primary antibodies. Bound antibodies were visualized with Super Signal West Femto or Pico Luminol/Enhancer Solution (Thermo Scientific). The specific protein bands were visualized and captured using an AlphaEase FC Imaging System (Alpha Innotech, San Leandro, CA). With respect to fibronectin protein, in rare cases, in addition to the predicted band at ~220 kDa, a lower band at ~100 kDa was also present. Under this circumstance, we only used the upper band (220 kDa) for quantification. To quantify abundance of a protein, the integrated density value of the target band of that was measured by drawing a rectangle outlining the band using AlphaEase FC software with auto background subtraction. Expression levels of target proteins were quantified by normalization of the integrated density values of those protein bands to that of actin or tubulin bands on the same blot.

Immunofluorescence Histochemistry

Paraffin-embedded kidney sections of WT and MKR mice were donated by Dr. Sandeep Mallipattu (Department of Medicine, Mount Sinai School of Medicine). After being deparaffinized and washed with PBS, sections were incubated with ice-cold acetone at −20°C for 10 min. After 30 min of incubation with blocking buffer, sections were incubated with anti-I-mfa primary antibody (359J rabbit polyclonal) at 1:100 dilution in PBS plus 10% donkey serum and 0.2% Triton X-100 at 4°C overnight. After three washes with PBS, sections were then incubated with donkey anti-rabbit Alexa Fluor 488 (Invitrogen) at a concentration of 1:500 for 1 h at room temperature. 4′,6-Diamidino-2-phenylindole (DAPI; catalog no. H-1200, Invitrogen) was used for staining nuclei. Sections were examined using an Olympus microscope (BX41) equipped for epifluorescence and an Olympus DP70 digital camera with DP manager software (version 2.2.1). Images were uniformly adjusted for brightness and contrast using Adobe Photoshop CS5. The images shown in Fig. 5C were representative images from five or six kidney sections from two mice of each group (2 or 3 kidney sections/mouse and 2 mice/group).

Fig. 5.

Diabetic influence on inhibitor of myogenic differentiation family isoform a (I-mfa) protein and/or mRNA expression in the kidney in rats and mice. A and B: Western blots showing I-mfa protein content in extracts of the renal cortex from rats fed a low-fat diet (LFD; control) and rats fed a high-fat diet (HFD) plus streptozotocin (STZ) treatment. A: representative immunoblots. Actin was used as a loading control. B: summary data. **P < 0.01 compared with LFD-fed rats. n = 3 rats/group. C: immunofluorescence staining showing I-mfa expression (green signals) in kidneys from a wild-type (WT) mouse fed the LFD (nondiabetic control) and a MCK-KR-hIGF-IR (MKR) mouse fed a HFD (diabetic nephropathy). The white dashlined rectangles on the left images indicate glomeruli. The right images show enlarged images of the glomeruli with nucleus staining (DAPI). Original magnification: ×200. Images are representative images from 5−6 kidney sections of 2 mice/group (2 or 3 kidney sections/mouse and 2 mice/group). D: quantitative real-time RT-PCR showing mRNA expression levels in glomerular extracts from WT mice fed with a LFD (n = 3) and MKR mice fed a LFD (n = 4) or HFD (n = 4). NS, no significant difference. N, number of mice/group.

Statistical Analysis

Data are reported as means ± SE. One-way ANOVA plus Student-Newman-Keuls post hoc analysis and Student unpaired t test were used to analyze differences among multiple groups and between two groups, respectively. P < 0.05 was considered statistically significant. Statistical analysis was performed using SigmaStat (Jandel Scientific, San Rafael, CA).

RESULTS

Overexpression of I-mfa Increased Fibronectin Protein Abundance in Human MCs

To determine whether I-mfa regulated fibronectin protein production by MCs, we carried out Western blot analysis and examined the effects of overexpression of I-mfa on ECM protein abundance. As shown in Fig. 1, transfection of MCs with F-I-mfa significantly increased content of fibronectin compared with cells without transfection. However, overexpression of HA-I-mfb, a nonfunctional isoform of inhibitor of the MyoD family, did not change the abundance of fibronectin. Overexpression of I-mfa and I-mfb was confirmed by immunoblots detected by anti-flag and -HA primary antibodies, respectively (Fig. 1A).

Fig. 1.

Western blots showing increased abundance of fibronectin (FN) by inhibitor of myogenic differentiation family isoform a (I-mfa) in human mesangial cells (MCs). A: representative blots showing expression of FN, flag, and hemagglutinin (HA) in human MCs without transfection [untransfected (UT)] or transfected with HA-tagged inhibitor of myogenic differentiation family isoform b (HA-I-mfb) or flag-tagged I-mfa (F-I-mfa). TB, α-tubulin (loading control). B: summary data from the experiments shown in A. **P < 0.01. n = 6 independent experiments.

Knockdown of I-mfa Decreased Fibronectin Expression at Both Messenger and Protein Levels in Human MCs

To further determine the positive regulation of I-mfa on fibronectin production, we silenced I-mfa using the siRNA approach and evaluated its effects on fibronectin expression at both mRNA and protein levels. siRNA against human I-mfa markedly reduced transcripts of I-mfa (Fig. 2A). Correspondingly, the fibronectin mRNA expression level was also significantly decreased in MCs treated with I-mfa siRNA (Fig. 2B). However, scrambled siRNAs did not alter the levels of both I-mfa and fibronectin (Fig. 2, A and B).

Fig. 2.

Inhibition of inhibitor of myogenic differentiation family isoform a (I-mfa) on fibronectin (FN) mRNA expression and protein production in human mesangial cells (MCs). A and B: quantitative real-time RT-PCR showing I-mfa (A) and FN (B) mRNA expression in human MCs transfected with scrambled siRNA (Scram) or I-mfa siRNA (siRNA) or without transfection [untransfected (UT)]. **P < 0.01; ***P < 0.001. n = 6 independent experiments in A and n = 5 independent experiments in B. C and D: Western blot showing FN protein abundance in human MCs with different treatments. The following groups are shown: cells treated with transforming growth factor (TGF)-β1 at 5 ng/mL for 15 h (TGF-β1), cells transfected with scrambled siRNA and treated with TGF-β1 at 5 ng/mL for 15 h (TGF-β1 + Scram), and cells transfected with I-mfa siRNA and treated with TGF-β1 at 5 ng/mL for 15 h (TGF-β1 + siRNA). TB, α-tubulin (loading control). D: summary data from the experiments shown in C. *P < 0.05 and **P < 0.01 vs. UT; †P < 0.05 vs. both TGF-β1 and TGF-β1 + Scram. n = 9 independent experiments.

We next examined whether knockdown of I-mfa could decrease fibronectin expression at the protein level. It is known that TGF-β signaling is a potent stimulator for fibronectin production by MCs and matrix accumulation in the glomerulus (24, 56, 58). We stimulated MCs with TGF-β1 to produce fibronectin and evaluated the TGF-β1 response with and without downregulation of I-mfa. As expected, TGF-β1 (5 ng/mL) treatment for 15 h dramatically increased the abundance of fibronectin protein. Treatment with I-mfa siRNA (knockdown of I-mfa), but not scrambled siRNA, significantly attenuated the TGF-β1 response (Fig. 2, C and D). These data suggest that I-mfa stimulated fibronectin production by MCs, probably at the transcriptional level.

High Glucose Increased I-mfa Protein Production by MCs

DN is characterized by the accumulation of ECM proteins in glomerular mesangium (23, 62). MCs are the major source of the ECM in the kidney affected by diabetes (1). Fibronectin is a major component of MC-derived ECM proteins in response to diabetic stimulation (15, 17, 34). High glucose is a pathogenic stimulator for ECM production by MCs during the progress of DN (3, 17, 22, 24, 52, 56). To explore the pathological relevance of I-mfa-stimulated fibronectin production by MCs, we examined high glucose effects on I-mfa protein production in human MCs. As shown in Fig. 3, A and B, exposure of MCs to high glucose (25 mM) increased the abundance of I-mfa in a time-dependent manner. A significant increase occurred at 48-h treatment and thereafter. The I-mfa response was specific to high glucose because the increase of I-mfa showed a trend of dose dependence under the osmotic control (Fig. 3, C and D).

Fig. 3.

High glucose (HG) treatment increased inhibitor of myogenic differentiation family isoform a (I-mfa) protein abundance in mesangial cells (MCs) but not in podocytes. Representative Western blots from human MC lysates are shown. A and C: MCs were incubated for different time periods in 0.5% FBS with normal glucose (NG; 5.6 mM glucose + 20 mM mannitol) or HG (25 mM) medium (A) or for 48 h in 0.5% FBS medium containing different concentrations of glucose with the appropriate amounts of mannitol for osmotic control (C). TB, α-tubulin (loading control). B and D: summary data for the experiments shown in A and C, respectively. In B, **P < 0.01, HG vs. NG at the same time period; in D, **P < 0.01 and *P < 0.05, both vs. 5 mM glucose. n = 5 independent experiments in B and n = 6 independent experiments in D.

High Glucose Did Not Increase I-mfa Messenger Levels in MCs

To determine whether high glucose also increased the expression level of I-mfa messengers in MCs, we analyzed I-mfa mRNA levels in human MCs with and without high glucose treatment for various time periods. As shown in Fig. 4, high glucose exposure for the timeframe from 1 to 48 h did not significantly alter the levels of the transcripts of I-mfa, although high glucose treatment for 48 h significantly increased I-mfa protein abundance (Fig. 3). These results suggest that high glucose upregulated I-mfa protein levels, at least in part, through posttranscriptional mechanisms.

Fig. 4.

High glucose (HG) did not increase inhibitor of myogenic differentiation family isoform a (I-mfa) mRNA levels in human mesangial cells (MCs). Shown are the results of quantitative real-time RT-PCR showing mRNA expression levels in human MCs treated with 5.6 mM glucose plus 20 mM α-mannitol [normal glucose (NG)] or 25 mM glucose (HG) for different time periods as indicated. n = 5 or 10 independent experiments.

The I-mfa Protein Level Was Increased in the Renal Cortex of Rats With DN and in the Glomerulus of Mice With DN

The in vitro effects of high glucose on I-mfa protein abundance in cultured MCs were further studied in animals. Two animal models of DN were used to detect diabetic effect on I-mfa protein expression levels in the glomerulus/cortex, where MCs are located. Feeding rats a HFD followed by STZ treatment is a well-established nongenetic type II diabetes model (9, 14, 42, 51). We have successfully established this rat model, which manifests overt type II diabetic phenotypes characterized by albuminuria, hyperlipidemia, hyperglycemia, and hyperinsulinemia (7). Western blot analysis showed that the abundance of I-mfa protein in the renal cortex was significantly greater in diabetic rats compared with control nondiabetic rats (fed a LFD) (Fig. 5, A and B).

We also conducted an immunohistochemistry assay and examined the diabetic effect on I-mfa protein expression in the glomerulus in a recently established mouse model of DN in MKR mice fed a HFD (35). These mice develop progressive DN and meet most of the criteria defined by the Animal Models of Diabetic Complications Consortium (2, 4). As shown in Fig. 5C, I-mfa-specific fluorescence staining in the glomerulus was much stronger in MKR mice on a HFD (DN) compared with WT mice on a LFD (nondiabetic control). These data suggest that protein levels of glomerular I-mfa are increased in the setting of DN.

The I-mfa mRNA Expression Level Was Not Altered in Glomeruli of Mice With DN

We then examined the diabetic effect on I-mfa messenger expression levels in the glomerulus in this mouse model of DN (MKR mice on a HFD). Glomerular extracts from MKR mice fed a HFD and LFD as well as WT mice fed a LFD (nondiabetic controls) were used. MKR mice, either fed a LFD or a HFD, show diabetic phenotypes (35). MKR mice on a LFD had modest renal injury, and those on a HFD showed advanced DN (35). However, there was no significant difference in I-mfa mRNA expression levels among the three group of mice (Fig. 5D).

Taken together, the data from animal kidney tissues are consistent with the results from cultured cells and further suggest that the diabetic environment upregulated the I-mfa protein level by posttranscriptional mechanisms.

H₂O₂ Treatment Increased I-mfa Protein Abundance in MCs

It is well known that high glucose-induced pathological changes are associated with increased reactive oxygen species (ROS) in MCs (16–18, 20). We then sought to determine whether ROS were implicated in high glucose-induced I-mfa upregulation. Human MCs were cultured in the absence and presence of H₂O₂ at 1, 10, 100, or 500 µM for 24 h, and I-mfa protein abundance was assessed. As shown in Fig. 6, A and B, H₂O₂ recapitulated the high glucose response, resulting in an increase in I-mfa protein content in a dose-dependent manner. The H₂O₂ effect reached a significant level at 100 µM and a more profound increase at 500 µM.

Fig. 6.

H₂O₂ increased inhibitor of myogenic differentiation family isoform a (I-mfa) protein abundance in human mesangial cells (MCs). A: Western blot showing a representative dose-dependent response of I-mfa protein abundance to H₂O₂ treatment. Growth-arrested MCs were exposed to different concentrations of H₂O₂ for 24 h. TB, α-tubulin (loading control). B: summary data from the experiments shown in A. *P < 0.05 compared with 0 µM H₂O₂; †P < 0.05 compared with 0 and 1 µM H₂O₂. n = 5 independent experiments.

Antioxidants Attenuated the High Glucose-Induced Increase in I-mfa Protein in MCs

If ROS are considered a downstream mechanism for the high glucose-induced increase of I-mfa protein abundance in MCs, antioxidants should attenuate the effect. We then tested this hypothesis by culturing MCs with and without various antioxidants. Growth-arrested MCs were treated with high glucose in the presence of polyethylene glycol-catalase, ammonium pyrrolidithiocarbamate, or N-acetylcysteine for 48 h. These antioxidants reduce ROS levels through different mechanisms. In agreement with the data presented above, high glucose treatment significantly increased I-mfa protein levels. However, in the presence of any one of those antioxidants, the high glucose effect was not observed (Fig. 7). These data, in conjunction with the results from H₂O₂ treatment (Fig. 6), suggest that the production of ROS is a downstream mechanism for the high glucose-induced increase in I-mfa protein abundance in MCs.

DISCUSSION

I-mfa is a multifunctional cytosolic protein, first identified as a transcription modulator that binds to and suppresses the transcriptional activity of MyoD family members (8). Earlier studies have demonstrated that I-mfa plays an important role in the inhibition of myogenesis (8), skeletogenesis (25, 53), and osteoclastogenesis (37). I-mfa may also function as a tumor suppressor gene (27, 28, 43). The present study provided evidence, for the first time, that I-mfa is also present and functional in kidney cells. In contrast to the inhibitory effect in most cell types, I-mfa increased fibronectin protein abundance in glomerular MCs. Because overproduction of ECM proteins by MCs plays a key role in the development of DN and fibronectin is a major component of ECM (15, 17), the I-mfa-stimulated fibronectin production may be associated with renal injury in diabetes. In support of the detrimental role of I-mfa, high glucose treatment and diabetes significantly increased the amount of I-mfa protein in MCs and the renal cortex, respectively. Recently, in another line of study, we found that I-mfa was also expressed in podocytes and tubular epithelial cells (data not shown). Whether high glucose treatment results in a similar response of I-mfa in those cells is not known, and further study is warranted to answer this question.

Sustained or frequently recurring hyperglycemia has been shown to result in diabetic complications and organ dysfunction, including DN. The hyperglycemia-induced kidney damage involves multiple pathological changes of intracellular signaling pathways, such as alteration of gene expression, production of advanced glycation end products and ROS, activation of the TGF-β-Smad-MAPK and JAK/STAT pathways, mitochondrial dysfunction (23, 24), and Ca²⁺ signaling pathways (6, 7, 19, 20, 54). These cell signaling molecules/pathways downstream of high glucose treatment can interact with one another and, thus, form a complicated network in the setting of diabetes. The present study shows that I-mfa is also a downstream target of hyperglycemia/high glucose and the upregulation of I-mfa pathway by high glucose/hyperglycemia might be a contributor to mesangial expansion in the diabetic kidney.

We further show that the high glucose effect on I-mfa was mediated by ROS, one major product secondary to high glucose treatment (17, 18). Previous studies by us and others have demonstrated that high glucose-induced pathological changes in the kidney are associated with increased ROS levels in MCs (7, 16–20, 23, 24, 54). The present study provides evidence of the existence of a new cascade in MCs, i.e., high glucose/ROS/I-mfa, which may play a role in the development of DN. In addition, considering the multifactorial function of I-mfa and firm associations of ROS with cellular physiological and pathological processes, the findings from the present study may advance our understanding of the molecular mechanisms for ROS-associated cell function and dysfunction.

It is noted that a significant increase in I-mfa protein abundance by high glucose required a time period of 48 h, but occurred within 24 h with H₂O₂ treatment. Our previous study demonstrated that high glucose significantly elevated cellular H₂O₂ levels within 1 h (21). If H₂O₂ mediated the high glucose effect on I-mfa, why did high glucose treatment need an additional 24 h to increase the level of I-mfa protein? One possibility is that the H₂O₂ effect on I-mfa protein abundance is concentration dependent, as shown in Fig. 6. Although high glucose elevated the cellular H₂O₂ level in a short period of time, it might not reach the level that significantly increased I-mfa protein content. Therefore, additional time is needed to allow for accumulation of H₂O₂.

An interesting finding in the present study is that high glucose/diabetes did not alter I-mfa at the mRNA level, suggesting that there is a posttranscriptional mechanism involved. The regulatory site could be at the translational level and/or posttranslational level. In terms of posttranslational regulation, the ubiquitin-proteasome system and lysosomal pathway are two major mechanisms for protein degradation (30, 41). We recently reported that high glucose and H₂O₂ stimulated degradation of Orai1 protein, a Ca²⁺ channel protein in MCs (21). Whether this mechanism is involved in the I-mfa response to high glucose/H₂O₂ observed in the present study needs to be further investigated.

The mechanisms that mediate I-mfa-stimulated production of fibronectin remain unknown. I-mfa was first identified as a transcription modulator that binds and suppresses the transcriptional activity of MyoD family of myogenic transcription factors (8). Over the past decade, several signal pathways downstream of I-mfa have been reported. These include Wnt signaling (26, 38, 39, 48), JNK signaling (26), human immunodeficiency virus 1 TAT-dependent transcription (28), and SERTA domain proteins (27). In addition to functioning as a transcriptional regulator, I-mfa was recently found to inhibit store-operated Ca²⁺ channel function by directly interacting with transient receptor potential C1 protein in the plasma membrane (31, 37). Among these downstream pathways, Wnt and JNK signaling in MCs have been well known to be involved in the development of DN (10, 29, 44, 55, 59–61). Our previous studies have demonstrated that transient receptor potential C1 protein and Orai1-mediated store-operated Ca²⁺ channels are present and functional in MCs (7, 11, 32, 33, 49, 50). Recently, we found that store-operated Ca²⁺ channel signaling inhibits the production of ECM proteins, including fibronectin by MCs (5, 58). It is possible that I-mfa positively regulates fibronectin in MCs through one or more of those signaling pathways.

In summary, we defined a new signaling pathway downstream of high glucose, i.e., high glucose/ROS/I-mfa, which positively regulates fibronectin in MCs (Fig. 8). Because accumulation of ECM proteins in the glomerulus is one of the major early changes in DN, our findings suggest that I-mfa in MCs may be an alternative therapeutic option for intervening in diabetes-induced renal injury.

Fig. 8.

Diagram illustrating the pathway of inhibitor of myogenic differentiation family isoform a (I-mfa) in the positive regulation of fibronectin protein in mesangial cells (MCs).

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant RO1-DK-115424-01 (to R. Ma) and by American Heart Association Southwestern Affiliate Grant-in-Aid 16GRNT27780043 (to R. Ma).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

P.Y.S. and R.M. conceived and designed research; P.Y.S., S.C., and Y.T. performed experiments; P.Y.S., S.C., Y.T., and R.M. analyzed data; P.Y.S., S.C., and R.M. interpreted results of experiments; P.Y.S., S.C., and R.M. prepared figures; P.Y.S. and R.M. drafted manuscript; P.Y.S., S.C., Y.T., L.T., and R.M. edited and revised manuscript; P.Y.S., S.C., Y.T., L.T., and R.M. approved final version of manuscript.