Role of miR-379 in high-fat diet-induced kidney injury and dysfunction

Abstract

Obesity is associated with increased risk for diabetes and damage to the kidneys. Evidence suggests that miR-379 plays a role in the pathogenesis of diabetic kidney disease. However, its involvement in obesity-induced kidney injury is not known and was therefore investigated in this study by comparing renal phenotypes of high-fat diet (HFD)-fed wild-type (WT) and miR-379 knockout (KO) mice. Male and female WT mice on the HFD for 10 or 24 wk developed obesity, hyperinsulinemia, and kidney dysfunction manifested by albuminuria and glomerular injuries. However, these adverse alterations in HFD-fed WT mice were significantly ameliorated in HFD-fed miR-379 KO mice. HFD feeding increased glomerular expression of miR-379 and decreased its target gene, endoplasmic reticulum (ER) degradation enhancing α-mannosidase-like protein 3 (Edem3), a negative regulator of ER stress. Relative to the standard chow diet-fed controls, expression of profibrotic transforming growth factor-β1 (Tgf-β1) was significantly increased, whereas Zeb2, which encodes ZEB2, a negative regulator of Tgf-β1, was decreased in the glomeruli in HFD-fed WT mice. Notably, these changes as well as HFD-induced increased expression of other profibrotic genes, glomerular hypertrophy, and interstitial fibrosis in HFD-fed WT mice were attenuated in HFD-fed miR-379 KO mice. In cultured primary glomerular mouse mesangial cells (MMCs) isolated from WT mice, treatment with high insulin (mimicking hyperinsulinemia) increased miR-379 expression and decreased its target, Edem3. Moreover, insulin also upregulated Tgf-β1 and downregulated Zeb2 in WT MMCs, but these changes were significantly attenuated in MMCs from miR-379 KO mice. Together, these experiments revealed that miR-379 deletion protects mice from HFD- and hyperinsulinemia-induced kidney injury at least in part through reduced ER stress.

NEW & NOTEWORTHY miR-379 knockout mice are protected from high-fat diet (HFD)-induced kidney damage through key miR-379 targets associated with ER stress (Edem3). Mechanistically, treatment of mesangial cells with insulin (mimicking hyperinsulinemia) increased expression of miR-379, Tgf-β1, miR-200, and Chop and decreases Edem3. Furthermore, TGF-β1-induced fibrotic genes are attenuated by a GapmeR targeting miR-379. The results implicate a miR-379/EDEM3/ER stress/miR-200c/Zeb2 signaling pathway in HFD/obesity/insulin resistance-induced renal dysfunction. Targeting miR-379 with GapmeRs can aid in the treatment of obesity-induced kidney disease.

INTRODUCTION

Obesity and the associated chronic inflammation increase the risk of type 2 diabetes, which, in turn, are risk factors for the development of chronic kidney disease (CKD) and cardiovascular complications (1, 2). Obesity and related insulin resistance/hyperinsulinemia lead to various adverse structural and functional changes in the kidney due to intensive metabolic demands of weight gain. This includes enlargement of Bowman’s space, increased glomerular hypertrophy, mesangial extracellular matrix (ECM) accumulation, and albuminuria. These alterations are closely linked to adipose tissue dysfunction (3), which increases insulin resistance/hyperinsulinemia (4) and contributes to the progression of glomerulopathy (5, 6), impaired kidney function, and CKD through excessive production of inflammatory cytokines and fibrotic factors and reduction in anti-inflammatory adipokines (6–9).

Noncoding RNAs (ncRNAs), including long ncRNAs (lncRNAs) and microRNAs (miRNAs), and their specific gene targets are under intense investigation because they are emerging as critical players in various diseases including obesity, diabetes, and associated complications (10, 11). miRNAs (∼22 nt) are small ncRNAs that regulate gene expression by posttranscriptional mechanisms involving translation repression or can function to degrade the target mRNAs, leading to altered cellular function and, consequently, disease (12).

We and others have adopted in vitro cell models, mechanistic approaches, and in vivo mouse models to demonstrate the functional involvement of several key miRNAs and lncRNAs in the pathogenesis of diabetic kidney disease (DKD) and inflammation (13–19). However, much less is known about the role of miRNAs in obesity-induced kidney dysfunction. The host transcript lncRNA of miR-379, called lnc-megacluster (lncMGC), which encompasses several miRNAs (20, 21), was shown to be upregulated in mesangial cells via CCAAT-enhancer-binding protein (C/EBP) homologous protein (CHOP), an endoplasmic reticulum (ER) stress-associated transcription factor, and plays a key role in promoting renal fibrosis and dysfunction in a mouse model of type 1 diabetes (20). Moreover, genetic deletion of miR-379 in mice protected them from diabetic kidney mice through miR-379 targets ER degradation enhancing α-mannosidase-like protein 3 (Edem3; which protects from ER stress) and mitochondrial fission 1 protein (FIS1) related to adaptive mitophagy (21). We also very recently showed that miR-379 knockout (KO) mice are protected from high fat-diet (HFD)-induced obesity, insulin resistance, and adipose dysfunction (22). However, the specific role of miR-379 in HFD-induced kidney damage requires more investigation.

In the present study, we used miR-379^em1COH KO mice (miR-379 KO, male and female), generated by CRISPR-Cas9 editing (21), and examined their response to obesity-induced kidney damage during HFD feeding. We observed that HFD increases miR-379 expression in the glomeruli; interestingly, miR-379 KO mice were protected from HFD-induced weight gain, hyperinsulinemia, and renal dysfunction. This protection was associated with altered expression of key miR-379 target genes with functions related to ER stress, fibrosis, and renal dysfunction in kidney glomeruli. Furthermore, the positive correlation observed between plasma insulin and glomerular expression of transforming growth factor-β1 (TGF-β1) in HFD-fed wild-type (WT-HFD) mice was lost in HFD-fed miR-379 KO (miR-379 KO-HFD) mice (male and female).

MATERIALS AND METHODS

Mouse Models of Obesity

All animal experiments were conducted according to protocols approved by the Institutional Animal Care and Use Committee of the Beckman Research Institute of City of Hope. To determine the in vivo role of miR-379 in HFD-induced CKD, we used the C57BL/6J (RRID: IMSR_JAX:000664) WT and miR-379 KO model as previously described (21). We randomly divided age-matched 8-wk-old WT and miR-379 KO mice (male and female) into groups and fed them with the laboratory control chow diet or HFD consisting of 26% calories from carbohydrate, 60% fat, and 14% protein (D12492I, Research Diets) for 10 wk (n = 5–8 mice/group, male or female) and 24 wk [n = 17–18 mice/group (male) and n = 6–12 mice/group (female)] with free access to food and water. Body weights and blood glucose levels were monitored and measured weekly. Plasma samples were harvested for measuring insulin levels. Urine samples (24-h urine collection) were collected to measure urine creatinine and albumin. Kidneys were harvested for glomerular and mouse mesangial cell (MMC) isolation. Kidney cortexes were used for histology examinations. Tissue histology analyses were performed in a blinded manner (Supplemental Fig. S1).

Physiological and Metabolic Analysis

Plasma insulin levels were measured using an ELISA kit (ultrasensitive mouse insulin ELISA kit, No. 90080, Crystal Chem). Albumin was measured as a biomarker for kidney damage and dysfunction in urine samples collected in a 24-h period at 24 wk of HFD feeding, using an ELISA kit (Mouse Albumin ELISA kit, No. 80630, Crystal Chem). Urine creatinine was measured using a high-quality enzymatic assay for mouse creatinine (No. 80350, Crystal Chem; 21).

Isolation of Mouse Glomeruli from Freshly Harvested Kidneys

Glomeruli were isolated from freshly harvested mouse kidneys using an established method (20). Briefly, chopped kidney cortical tissues were gently strained through different pore size stainless sieves (200, 150, and 75 μm, respectively). Pooled glomeruli were then collected to prepare MMCs or for RNA extractions.

Primary Mesangial Cells

Primary mouse MMCs were obtained from freshly harvested kidneys/glomeruli from WT and miR-379 KO male mice and cultured in vitro as previously described (20, 21). MMCs (1 × 10⁶ cells) were cultured in six-well plates and treated with 200 nM insulin (No. I0516, Millipore Sigma, St. Louis, MO) and incubated in a humidified incubator maintained at 37°C with 5% CO₂ for 6 h. Noninsulin-treated cells were used as a control. Cultured MMCs from WT mice were treated with recombinant human TGF-β1 (10 ng/mL, R&D Systems, Minneapolis, MN) or locked nucleic acid (LNA)-modified miR-379 GapmeR antisense oligonucleotides (oligos) [379F (full 21mer5′-CCTacgttccatagtctaCCA-3′) LNA (uppercase), DNA (lowercase), phosphorothioate (backbone)] (1 μmol/L, Integrated DNA Technologies) for 24 h (20). MMCs treated with control GapmeR oligos [5′-ATTttattcggaGCT-3′ LNA (uppercase), DNA (lowercase), phosphorothioate (backbone)] (1 μmol/L, Integrated DNA Technologies) were used as a negative control (NC). Cells were then harvested, and RNA was extracted to measure the expression of target genes.

Quantitative Reverse Transcriptase Polymerase Chain Reaction

Quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) analysis was performed to analyze gene expression. cDNA was prepared from glomerular cells and MMCs using a GeneAmp RNA PCR kit (Applied Biosystems, Carlsbad, CA) and POWER SYBR Green mix (Applied Biosystems) as previously described (20). miRNA was extracted using the qScript miRNA cDNA synthesis kit (Quanta Biosciences, Gaithersburg, MD) and amplified using PerfeCTa SYBR Green Supermix (Quanta Biosciences). For miRNAs, specific mature miRNA sequences were used as forward primers, and the universal primer provided in the kit as the reverse primer. Cypa was used as an internal control for mRNA and U6 as an internal control for miRNAs. PCR primer sequences are shown in Supplemental Table S1.

Histology and Immunohistochemical Staining

Kidney cortex samples were fixed in 10% formalin and embedded in paraffin for histological analysis according to routine histological techniques (Pathology: Solid Tumor Core, City of Hope). Deparaffinized tissue slides were stained with hematoxylin and eosin stain to evaluate morphological structure and with periodic acid-Schiff (PAS) to identify ECM accumulation in glomerular mesangial areas. Masson’s trichrome staining was used for the detection of fibrosis in kidney cortex sections. Under light microscopy examination, glomeruli, mesangial PAS-positive areas, and glomeruli and tubulointerstitial fibrotic areas were quantified in each group using ImageJ software (RRID:SCR_003070).

The cellular localization of EDEM3 (1:100, Cat. No. NBP1-88342, Novus, RRID: AB_11039514) and TGF-β1 (1:150, Cat. No. ab92486, Abcam, RRID: AB_10562492) was examined in glomerular sections using immunohistochemical staining. Biotinylated goat anti-rabbit (1:200, Cat. No. BA-1000, Vector Laboratories, RRID: AB_2313606) was used as the secondary antibody. Immunohistochemical images were taken at ×20 magnification using KEYENCE-BZ-X800 Series (Osaka, Japan; image intensity and contrast were kept consistent throughout). Fluorescence intensity and immune-positive reaction areas were quantified in glomeruli in each group using ImageJ software (RRID:SCR_003070).

Immunofluorescence staining was performed to demonstrate the cellular localization of antibodies to kidney injury molecule (KIM)-1 (1:150, Cat. No. AF1817, R&D Systems, RRID: AB_2116446) and p57 (1:250, Cat. No. ab75974, Abcam, RRID: AB_1310535) in kidney cortex sections. The following secondary antibodies were used: Alexa Fluor 555 donkey anti-goat (1:400, Cat. No. A-32816,Thermo Fisher Scientific, RRID: AB_2762839) and Alexa Fluor 488 goat anti-rabbit (1:400, Cat. No. A-11008, Thermo Fisher Scientific, RRID: AB_143165), respectively. Immunofluorescence images were taken using an original magnification of ×20. The number of p57-positive cells and KIM-1-positive area were measured from each group using ImageJ software.

Statistics

In vitro experiments were performed at least three times. In vivo experiments were performed twice for the 24-wk study and once for the 10-wk study. Animal cohorts consisted of 5–18 mice/group. Power analysis showed that a sample size of 5 animals/group would provide us 79% power to detect a mean difference of 2 SD (effect size 2) using two-tailed t tests at a significance level of P < 0.05. Results are expressed as means ± SE. One-way ANOVA followed by a post hoc Tukey test or two-way repeated-measures ANOVA for multiple groups was used to determine the significance of differences among four independent groups. Comparisons between two groups were performed using Student’s t tests. Significance was set at P < 0.05.

RESULTS

miR-379 KO-HFD Mice Exhibit Lower Body Weights and Plasma Insulin Levels

Supplemental Fig. S1 provides the experimental scheme for the animal experiments. As shown in Table 1, HFD feeding (60 kcal from fat) increased body weights in both WT and miR-379 KO male and female mice. Although there were significant increases in weight gain in HFD-fed mice in both groups and sexes, miR-379 KO-HFD mice showed a lower rate of weight gain compared with WT-HFD mice (Supplemental Fig. S2). Thus, miR-379 KO male and female mice exhibited lower body weight compared with WT-HFD (male and female) groups both at 10 and 24 wk of HFD feeding (Table 1 and Supplemental Fig. S2). Since obesity is accompanied by hyperinsulinemia/insulin resistance, which, in turn, can affect other organs like the kidney directly or indirectly (23), we next assessed plasma insulin levels. Both WT-HFD male and female mice showed hyperinsulinemia, which was, however, significantly lower in miR-379 KO-HFD (male and female) mice relative to WT-HFD mice (Table 1). There was no significant change in blood glucose levels between WT and miR-379 KO male or female HFD mice (Table 1).

Table 1.

HFD-induced increased body weight and plasma insulin levels were ameliorated in miR-379 KO-HFD mice

Parameters	WT-Con		WT-HFD	miR-379 KO-Con	miR-379 KO-HFD
Body weight, g
10 wk	Male	27.60 ± 0.74	42.00 ± 1.00b	29.88 ± 0.548	38.75 ± 0.629d
	Female	21.80 ± 0.58	28.80 ± 1.02b	20.86 ± 0.404	24.60 ± 1.69d
24 wk	Male	32.32 ± 2.82	53.32 ± 3.04c	31.32 ± 1.74	49.25 ± 3.40f
	Female	24.54 ± 1.17	46.13 ± 4.42c	23.88 ± 2.15	38.58 ± 3.40f
Blood glucose levels (24 wk), mg/dL	Male	142.0 ± 6.31	149.6 ± 5.99	158.2 ± 8.82	157.8 ± 6.57
	Female	149.5 ± 4.93	150.4 ± 7.03	172.3 ± 11.80	175.3 ± 7.13
Plasma insulin levels (24 wk), ng/mL	Male	1.142 ± 0.07	6.219 ± 0.46c	1.033 ± 0.10	4.781 ± 0.60e
	Female	0.671 ± 0.06	2.117 ± 0.39a	0.6380 ± 0.12	0.8510 ± 0.25d

All data are presented as means ± SE. Average body weights in male and female mice at 10 wk (n = 5–8, male and female) and 24 wk (n = 17 or 18 mice/group for male mice and n = 6–8 mice/group for female mice) of high-fat diet (HFD) are shown. Also shown are average blood glucose levels at 24 wk of HFD in male and female mice (n = 12–14 mice/group) and average plasma insulin levels at 24 wk of HFD in male and female mice (n = 7–11 mice/group for male mice and n = 5 mice/group for female mice). Statistical analysis was by one-way ANOVA with a post hoc Tukey’s test for multiple comparisons. ^aP < 0.01, ^bP < 0.001, and ^cP < 0.0001, wild-type (WT)-HFD vs. WT-control (Con) groups. ^dP < 0.05, ^eP < 0.01, and ^fP < 0.001, miR-379 knockout (KO)-HFD vs. WT-HFD groups.

HFD-Induced Renal Dysfunction, Hypertrophy, and Glomerular ECM Accumulation Are Attenuated in miR-379 KO-HFD Mice

We measured urine albumin and creatinine levels as markers of kidney function (24) in control and HFD-fed mice at 24 wk of HFD (Fig. 1, A–D). In male mice, average urine albumin and creatinine levels were significantly increased in WT-HFD mice compared with WT-control mice, which were significantly attenuated in miR-379 KO-HFD male mice (Fig. 1, A and B); a similar trend was observed in females (Fig. 1, C and D). We observed significant increases in kidney weights in WT-HFD male and female mice but not in miR-379 KO-HFD male and female mice at 24 wk of HFD feeding (Fig. 1, E and F). Compared with the respective standard diet-fed controls, we observed significant increases in glomerular size in WT-HFD male and female mice, which were significantly attenuated in miR-379 KO-HFD (male and female) mice at 24 wk of HFD (Fig. 1, G and H). WT-HFD male and female mice also showed significant increases in mesangial matrix expansion and ECM accumulation as detected by PAS staining (Fig. 1, I and J). By quantitative analysis, we found that ECM accumulation was significantly attenuated in miR-379 KO-HFD male and female mice at 10 and 24 wk of HFD (Fig. 1, K–N). We also observed moderate glomerular and tubulointerstitial fibrosis, as indicated by Masson’s trichrome staining (blue color), in WT-HFD male and female mice (Fig. 2, A and B) but not in miR-379 KO-HFD (male and female) mice at 24 wk of HFD (Fig. 2, A and B). These results show that miR-379 KO mice are protected from HFD-induced kidney hypertrophy, fibrosis, and glomerular damage.

Figure 1.

High-fat diet (HFD)-induced renal dysfunction, hypertrophy, and glomerular extracellular matrix (ECM) accumulation are attenuated in HFD-fed miR-379 knockout (KO) mice. Urine albumin (A and C) and creatinine (B and D) levels in male (M; A and B) and female (F; C and D) mice at 24 wk of HFD. n = 7–10 mice/group for male mice and n = 5–7 mice/group for female mice. Each symbol represents 1 mouse/group. Kidney weights in male (E) and female (F) mice at 24 wk of HFD. Glomerular size in male (G) and female (H) mice at 24 wk of HFD. n = 30 glomeruli/group. Periodic acid-Schiff (PAS) staining showing increased glomerular mesangial area and ECM accumulation in male (I) and female (J) mice at 10 and 24 wk of HFD. Quantitative analysis of PAS-positive glomerular areas at 10 wk (K and M) and 24 wk (L and N) of HFD in male (K and L; n = 20–50 glomeruli/group) and female (M and N; n = 20–30 glomeruli/group) mice. Scale bars = 50 µm. Magnification: ×40. Statistical analysis was by one-way ANOVA with a post hoc Tukey’s test for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. All data are presented as means ± SE. Con, control; WT, wild type.

Figure 2.

High-fat diet (HFD)-induced renal fibrosis is attenuated in HFD-fed miR-379 knockout (KO) mice. A: Masson’s trichrome staining to detect fibrosis (blue color) showed increased glomerular and tubulointerstitial fibrosis in HFD-fed wild-type (WT) mice (male and female) at 24 wk of HFD. B: quantitative analysis of Masson’s trichrome staining-positive areas in male (n = 20 area/group) and female (n = 20–30 area/group) mice. Scale bars = 100 µm. Magnification: ×20. Statistical analysis was by one-way ANOVA with a post hoc Tukey’s test for multiple comparisons. *P < 0.05 and ****P < 0.0001. All data are presented as means ± SE. Con, control; WT, wild type.

Increased Glomerular Expression of miR-379 and Chop in WT-HFD Mice Is Reversed in miR-379 KO-HFD Mice

We have recently shown that expression of miR-379 and Chop (the ER stress response transcription factor) were increased in glomeruli isolated from streptozotocin-injected diabetic WT mice, which were attenuated in diabetic miR-379 KO mice (21). In the present study, we also found significant increases in glomerular expression of miR-379 in WT-HFD male and female mice compared with their respective standard diet-fed controls (Fig. 3, A and B). Chop expression was significantly increased at 10 wk of HFD in WT (male and female) mice but not in miR-379 KO-HFD mice. HFD did not increase Chop expression at 24 wk in both WT and miR-379 KO mice (male and female), although Chop expression levels were lower in miR-379 KO-HFD relative to WT-HFD mice (Fig. 3, C and D). X box-binding protein 1 (XBP1) is spliced and activated by ER stress and, in turn, can activate CHOP (25, 26). We have previously shown that Xbp1 splicing is involved in upregulation of Chop in response to HG or TGF-β1 in MMCs (20). Therefore, we next measured the glomerular expression of total Xbp1 at 24 wk of HFD. As shown in Fig. 3, E and F, Xbp1 expression was significantly increased in WT-HFD male and female mice relative to control but not in miR-379 KO-HFD (male and female) mice (Fig. 3, E and F), supporting decreases in ER stress in miR-379 KO-HFD mice.

Figure 3.

Increased glomerular expression of miR-379 and C/EBP homologous protein (Chop) and decreased expression of key miR-379 target genes in high-fat diet (HFD) wild-type (WT) mice are reversed in HFD-fed miR-379 knockout (KO) mice. Expression of miR-379 in glomerular in male (M; A) and female (F; B) mice at 24 wk of HFD. Chop expression in male (C) and female (D) mice at 10 and 24 wk of HFD. X box-binding protein 1 (Xbp1) expression in male (E) and female (F) mice at 10 and 24 wk of HFD. Expression of miR-379 target genes [endoplasmic reticulum degradation enhancing α-mannosidase-like protein 3 (Edem3), mitochondrial fission 1 protein (Fis1), and thioredoxin 1 (Txn1)] in male (G and H) and female (I and J) mice at 10 wk (G and I) and 24 wk (H and J) of HFD. Each symbol represents 1 mouse/group. K: immunohistochemistry staining of EDEM3 in male and female kidney cortex sections at 24 wk of HFD. Scale bars = 50 µm. Quantitative analysis of EDEM3 in male (L) and female (M) mice. n = 30–40 glomeruli/group. Statistical analysis was by one-way ANOVA with a post hoc Tukey’s test for multiple comparisons. All data are presented as means ± SE. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Con, control.

Decreased Expression of Key miR-379 Target Genes in Glomeruli of WT-HFD Mice Is Restored in miR-379 KO-HFD Mice

We measured the glomerular expression of key previously identified miR-379 target genes (21) (Edem3, Fis1, and thioredoxin 1 (Txn1; antioxidant)] in control and HFD-fed mice at 10 and 24 wk of HFD. In male mice, we observed a significant reduction of Edem3 and Fis1 in WT-HFD mice versus WT-control mice only at 24 wk of HFD, which was significantly restored in miR-379 KO male mice (Fig. 3, G and H). In female mice, expression of Edem3 and Txn1 were significantly higher in miR-379 KO-control and HFD-fed mice versus WT-control mice at 10 wk of HFD (Fig. 3I). At 24 wk of HFD, we observed reductions of Edem3 and Txn1 in WT-HFD female mice, which were restored in miR-379 KO-HFD female mice (Fig. 3J).

We also used immunohistochemical staining to detect the protein expression of EDEM3. Kidney cortex sections were stained to examine EDEM3 expression in the glomerular compartments at 24 wk of HFD. We observed a significant decrease of EDEM3 expression in glomerular cells in WT-HFD male (Fig. 3, K and L) and WT-HFD female (Fig. 3, K and M) mice, which were significantly reversed in miR-379 KO-HFD male and female mice (Fig. 3, K–M).

Increased Expression of Profibrotic Genes in the Glomeruli of WT-HFD Mice Is Reversed in miR-379 KO-HFD Mice

To determine the role of miR-379 in the known association between HFD-induced fibrosis, ECM accumulation, and profibrotic genes, we compared expression of Tgf-β1, collagens [collagen type I-α₂ (Col1α2) and collagen type IV-α₁ (Col4α1)], and connective tissue growth factor (Ctgf) in renal glomeruli isolated from WT and miR-379 controls and HFD-fed mice (Fig. 4, A–D). In male mice, expression of Tgf-β1, Col4α1, Col1α2, and Ctgf was significantly decreased in miR-379 KO-HFD mice compared with WT-HFD mice at 10 wk of HFD feeding (Fig. 4A). At 24 wk, the increased expression of Tgf-β1 and Col1α2 in WT-HFD male mice was significantly attenuated in miR-379 KO-HFD male mice (Fig. 4B). In female mice, compared with the corresponding WT-control mice, we observed a significant increase in the expression of Tgf-β1, Col4α1, Col1α2, and Ctgf in WT-HFD mice, which was significantly reduced or ameliorated in miR-379 KO-HFD mice at 10 wk of HFD (Fig. 4C). At 24 wk of HFD feeding, the upregulation of Tgf-β1, Col4α1, and Ctgf observed in WT-HFD female mice was significantly attenuated in miR-379 KO-HFD female mice (Fig. 4D).

Figure 4.

Increased expression of profibrotic genes in the glomeruli of high-fat diet (HFD)-fed wild-type (WT) mice is reversed in HFD-fed miR-379 knockout (KO) mice. Glomerular expression of profibrotic genes [transforming growth factor-β1 (Tgf-β1), collagen type IV-α₁ (Col4α1), collagen type I-α₂ (Col1α2), and connective tissue growth factor (Ctgf)] at 10 wk (A and C) and 24 wk (B and D) of HFD in male (M; A and B) and female (F; C and D) mice. E: immunohistochemistry staining of TGF-β1 in male and female kidney cortex sections at 24 wk of HFD. Scale bars = 50 µm. Quantitative analysis of TGF-β1 in male (F) and female (G) mice. n = 30 glomeruli/group. Glomerular expression of miR-200c in male (H) and female (I) mice and Zeb2 in male (J) and female (K) mice at 24 wk of HFD. Each symbol represents 1 mouse/group. Statistical analysis was by one-way ANOVA with a post hoc Tukey’s test for multiple comparisons. All data are presented as means ± SE. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Con, control.

We next examined TGF-β1 expression in the glomerular compartments by immunohistochemical staining. The average TGF-β1-positive glomerular area was significantly increased in WT-HFD male and female mice relative to their respective controls, as expected. Notably, this increase was significantly attenuated in miR-379 KO-HFD (male and female) mice (Fig. 4, E–G).

We next examined potential mechanisms by which miR-379 regulates TGF-β1 and whether it involves cross talk with other miRNAs. We have previously shown that miR-192 and miR-200b/c can upregulate TGF-β1 in mesangial cells and glomeruli of diabetic mice by targeting Zeb1/2 E-box repressors (27). Therefore, herein, we compared the expression of miR-200c and its known target Zeb2 in mouse glomeruli at 24 wk of HFD. Relative to WT-control mice, miR-200c expression was increased in WT-HFD male and female mice. This was significantly reduced in miR-379 KO-HFD male and female mice versus their respective controls (Fig. 4, H and I). In parallel, expression of Zeb2 was significantly decreased in glomeruli from WT-HFD (male and female) mice compared with control mice, and this reduction was significantly reversed in miR-379 KO-HFD male and female mice (Fig. 4, J and K). We also observed a close correlation between the expression of miR-379 and miR-200 in WT-HFD (male and female) mouse glomeruli (Supplemental Fig. S3).

Together, the results from gene expression and histopathology suggest that the genetic deletion of miR-379 attenuates glomerular fibrosis and ECM accumulation due to downregulation of key profibrotic genes, including TGF-β1, in obese mice and that this can involve cross talk with miR-200c.

Markers of Podocyte and Tubular Damage Are Reduced in miR-379 KO-HFD Mice

We and others have shown that TGF-β1 can promote ER stress and induce podocyte damage in glomeruli from diabetic (20, 28–30) and HFD-fed (31) mice. In this study, the number of podocytes was detected by p57-positive cells (nuclear green fluorescent staining) and examined by immunofluorescence staining at 24 wk of HFD (Fig. 5A). Quantification of the immunofluorescence data showed that the number of podocytes was significantly lower in WT-HFD male and female mice versus control mice, and this reduction was attenuated in miR-379 KO-HFD male and female mice (Fig. 5, C and D).

Figure 5.

Markers of podocytes and tubular damage are reduced in high-fat diet (HFD)-fed miR-379 knockout (KO) mice. A: immunofluorescence staining of p57 (podocyte marker) in kidney cortex sections of male and female mice at 24 wk of HFD. Green signals: p57; blue: nucleus (DAPI). Scale bars = 50 µm. Magnification: ×40. B: immunofluorescence staining of kidney injury molecule-1 (KIM-1; proximal tubular injury marker) in kidney cortex sections of male and female mice at 24 wk of HFD. Red signals: KIM-1; blue: nucleus (DAPI). Scale bars = 50 µm. Magnification: ×20. Quantitative analysis of p57 in male (C) and female (D) mice. n = 70 glomeruli/group for male mice; n = 30 glomeruli/group for female mice. Quantitative analysis of KIM-1 in male (E) and female (F) mice. n = 50 area/group. Statistical analysis was by one-way ANOVA with a post hoc Tukey’s test for multiple comparisons. All data are presented as means ± SE. *P < 0.05, **P < 0.01, and ****P < 0.0001. Con, control; WT, wild type.

Next, we measured the level of KIM-1 as a marker of tubular injury by immunofluorescence staining (24, 32). We observed that KIM-1 was either nondetectable or restricted to very few positive areas in control mice (Fig. 5B). Instead, proximal tubules showed distinctly positive staining for KIM-1 that was significantly higher in WT-HFD (male and female) compared with their respective WT-control mice. On the other hand, this KIM-1 staining was significantly reduced in miR-379 KO-HFD male and female mice relative to their controls (Fig. 5, B, E, and F).

Insulin Treatment Increases miR-379 and Tgf-β1 Expression in Mouse MMCs

We found a significant positive correlation between plasma insulin levels (hyperinsulinemia) and (increased) expression of glomerular Tgf-β in WT-HFD male and female mice (Fig. 6, A and B). However, this correlation was not observed in miR-379 KO-HFD male and female mice (Fig. 6, C and D). Therefore, to determine the regulatory effect of insulin on the expression of miR-379 and consequent glomerular cell damage, we isolated MMCs from WT and miR-379 KO mice and treated them with insulin (200 nM, to mimic hyperinsulinemia) for 6 h in vitro. miR-379 expression was significantly increased in insulin-treated WT MMCs (Fig. 7A). Furthermore, we observed a significant decrease in Edem3 (Fig. 7B) and increases in Chop and total Xbp1 expression in insulin-treated WT MMCs (Fig. 7, C and D), suggesting insulin-induced increases in parameters of ER stress. Expression of Tgf-β1 and profibrotic genes, Col4α1 and fibronectin 1 (Fn1), were also significantly increased in insulin-treated WT MMCs (Fig. 7, E–G). In parallel, we also observed a significant increase in the expression of miR-200c and decrease in the expression of its target, Zeb2 (Fig. 7, H and I). Notably, insulin-induced changes in the expression of these candidate downstream genes seen in WT MMCs were significantly reversed in miR-379 KO MMCs (Fig. 7, J–Q).

Figure 6.

Positive correlation between hyperinsulinemia and glomerular expression of transforming growth factor-β1 (Tgf-β1). Positive correlation of plasma insulin levels and glomerular Tgf-β1 expression in high-fat diet (HFD)-fed wild-type (WT) male (M; A) and female (F; B) mice at 24 wk of HFD. On the other hand, there was no correlation between plasma insulin levels and glomerular Tgf-β1 expression in HFD-fed miR-379 knockout (KO) male (C) and female (D) mice at 24 wk of HFD. n = 5–8 mice/group.

Figure 7.

Insulin treatment increases miR-379 and transforming growth factor-β1 (Tgf-β1) expression in mouse mesangial cells (MMCs). A: miR-379 expression in MMCs isolated from the glomeruli of wild-type (WT) mice. Endoplasmic reticulum degradation enhancing α-mannosidase-like protein 3 (Edem3; B), C/EBP homologous protein (Chop; C), X box-binding protein-1 (Xbp1; D), Tgf-β1 (E), collagen type IV-α₁ (Col4α1; F), fibronectin-1 (Fn1; G), miR-200c (H), and Zeb2 (I) expression in MMCs isolated from the glomeruli of wild-type (WT) mice. n = 3–9/group. Comparison of the fold change expression of Edem3 (J), Chop (K), Xbp1 (L), Tgf-β1 (M), Col4α1 (N), Fn1 (O), miR-200c (P), and Zeb2 (Q) expression in insulin-treated WT and miR-379 knockout (KO) MMCs. n = 5–9/group. MMCs were treated with 200 nM insulin (MMC-insulin). Noninsulin-treated MMCs were used as a control (Con). Targeting miR-379 with GapmeRs has protective effects in MMCs. R–V: gene expression of target genes in WT MMCs treated with recombinant human TGF-β1 (10 ng/mL) or miR-379 GapmeR (1 μmol/L) or GapmeR negative control (NC; 1 μmol/L) for 24 h. R: miR-379; S: Edem3; T: collagen type I-α₂ (Col1α2); U: Col4α1; V: Fn1. n = 3–9/group. Statistical analysis was by one-way ANOVA with a post hoc Tukey’s test for multiple comparisons. Comparisons between two groups were performed using Student’s t tests. All data are presented as means ± SE. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Targeting miR-379 With GapmeRs Has Protective Effects in MMCs

We next examined the effects of targeting miR-379 with an LNA-modified antisense oligonucleotide (GapmeR). MMCs from WT mice were treated with TGF-β1 (10 ng/mL) and either NC GapmeR or miR-379 targeting GapmeR. Untreated cells served as controls (Con). TGF-β1 treatment increased the expression of miR-379 (Fig. 7R) and decreased the expression of miR-379 target Edem3 (Fig. 7S). Expression of profibrotic genes (Col1α2, Col4α1, and Fn1) was significantly increased by TGF-β1 treatment (Fig. 7, T–V). We found that GapmeR-mediated inhibition of miR-379 in TGF-β1-treated cells was accompanied by reversal of TGF-β1 effects as shown by upregulation of Edem3 and downregulation of Col1α2, Col4α1, and Fn1 gene expression relative to their respective NCs (Fig. 7, R–V).

These in vitro data support the notion that miR-379 and Tgf-β1 induced by hyperinsulinemia and ER stress (during HFD and insulin resistance) contribute to the increased expression of glomerular profibrotic genes, ECM accumulation, and abnormalities observed in the kidney/glomeruli of HFD mice and that miR-379 mediates the adverse effects of TGF-β, at least in part.

DISCUSSION

Evidence suggests that renal miR-379 is upregulated in diabetes and that miR-379 KO mice are protected from diabetes-induced kidney diseases due to improved mitophagy (21). In the present study, we found that miR-379 KO mice also exhibit renal protection during HFD-induced obesity. Our results support an important role for miR-379 upregulation in TGF-β1 and HFD-induced kidney damage, which may be mediated by hyperinsulinemia/insulin resistance and ER stress. We observed that key phenotypes observed during chronic HFD feeding, including significant increases in plasma insulin levels, renal hypertrophy, and fibrosis, in WT-HFD male and female mice were significantly attenuated in miR-379 KO-HFD mice. Moreover, we also found that the increased expression of Tgf-β1 and decreased expression of Zeb2 (a transcription factor that represses Tgf-β1 expression) observed in the glomeruli of WT-HFD (male and female) mice were reduced in miR-379 KO-HFD mice. In addition, in WT mice, HFD significantly increased KIM-1, a marker for tubular damage, and downregulated p57, a podocyte marker associated with declining renal function, and these aberrant changes were significantly ameliorated in miR-379 KO-HFD mice. The in vitro experiments provided mechanistic support for the regulatory effect of miR-379 on hyperinsulinemia/insulin resistance-induced glomerular damage. Insulin treatment increased miR-379 and ER stress (as suggested by increases in Chop and decreases in Edem3, a target of miR-379) and upregulated Tgf-β1, profibrotic genes, and miR-200c but downregulated Zeb2 (miR-200c target and negative regulator of Tgf-β1). These alterations were not observed in insulin-treated miR-379 KO MMCs. Notably, the increased expression of profibrotic genes observed in WT MMCs treated with TGF-β1 was attenuated by a GapmeR targeting miR-379.

Previously, we found that miR-379 and its host lncRNA were upregulated by the ER stress regulator CHOP (20). Here, we observed parameters of ER stress in WT-HFD (male and female) mice, as evidenced by the increase of Chop and Xbp1 and downregulation of EDEM3, a target of miR-379 and negative regulator of ER stress (20, 21). These factors were reversed in miR-379 KO-HFD mice. We also found that the downregulation of miR-379 targets related to oxidative stress (Txn1) and mitophagy (Fis1) observed in WT-HFD mice were also reversed in miR-379 KO-HFD mice, which may also contribute to the observed renal protection in the latter. It is well known that ER stress, oxidative stress, and mitochondrial dysfunction are increased in the kidneys of mouse models of DKD (20, 21, 33, 34) and HFD-induced kidney disease (35–37). Obesity is a significant and potentially preventable risk factor for nondiabetic kidney injury and contributes to oxidative stress, chronic inflammation, impaired secretion of adipokines, lipotoxicity (5), ER stress associated with intracellular lipid accumulation (38), and mitochondrial dysfunction (39). Further in-depth studies are needed to determine if functional indexes of ER stress, oxidative stress, and mitochondrial health as well as key adipokines are altered in the mouse models in our present study.

It is well known that miRNAs have multiple bonafide/true targets as well as other potential and putative targets based on seed sequence complementarities. In our previous study (21), we identified several true candidate targets of miR-379 relevant to diabetic nephropathy using the unbiased Ago2-CLASH technique to compare MMCs obtained from WT and miR-379 KO mice. Of those candidate miR-379 targets, in the present study, EDEM3 was the most relevant one related to ER stress regulation in HFD-induced kidney injury. This suggests some differences in functionally relevant targets of miR-379 between a model of diabetic nephropathy (streptozotocin-induced kidney injury) (where the mitochondrial mitophagy-related target FIS1 was important) and HFD-induced kidney injury. Thus, even though miRNAs have multiple targets, the functionally relevant critical targets in the kidney can be prioritized depending on the disease condition, such as obesity, diabetes, or hyperinsulinemia. miRNAs may also have organ- and cell-specific targets and functions. In this connection, in a recent study (22), we found that miR-379 and some of its key targets, including ER stress-related EDEM3 and angiogenesis-related VEGF, had functional roles related to adipose dysfunction in obesity.

Our in vitro data suggest that hyperinsulinemia induced during HFD can contribute to renal hypertrophy, albuminuria, podocyte loss, glomerular and tubular injury, fibrosis, and ECM accumulation through the actions of miR-379 and its cross talk with TGF-β1 (Fig. 8) because these adverse effects of HFD on renal function are attenuated in miR-379 KO mice. This is in line with evidence showing insulin receptor signaling at the level of podocytes and tubular cells can mediate renal function during metabolic disorders (24, 40). However, in vivo, other factors associated with insulin resistance, including dyslipidemia and free fatty acids, may also be involved.

Figure 8.

Proposed model for the regulatory role of miR-379 in high-fat diet (HFD)-induced kidney damage. Metabolic dysfunction, insulin resistance, and hyperinsulinemia during HFD feeding are associated with increases in transforming growth factor-β1 (TGF-β1), markers of endoplasmic reticulum (ER) stress [X box-binding protein-1 (Xbp-1)-spliced and C/EBP homologous protein (Chop)], and miR-379. This leads to downregulation of the miR-379 target ER degradation enhancing α-mannosidase-like protein 3 (Edem3), which further promotes ER stress and increases in the expression of miR-379, and TGF-β1 via upregulation of miR-200c and downregulation of transcription factor Zeb2. TGF-β1, in turn, can upregulate miR-379 while also enhancing the expression of fibrotic genes associated with renal hypertrophy and glomerular, tubular, and podocyte damage, leading to kidney disease. HFD also downregulates the miR-379 target genes thioredoxin 1 (Txn1), a marker of oxidant stress, and mitochondrial fission 1 protein (Fis1), a mitophagy marker (not shown). Therefore, during HFD and obesity, cross talk between TGF-β1, ER stress, and miR-379 can have amplifying roles on downstream pathological features associated with chronic kidney disease. Col4α1, collagen type IV-α₁; Col1α2, collagen type I-α₂; Ctgf, connective tissue growth factor; ECM, extracellular matrix; Fn1, fibronectin-1.

The mechanisms by which miR-379 regulates Tgf-β1 are not fully understood. Several studies have shown the pathological role of increased TGF-β1 signaling and expression in kidney injuries in obesity (41, 42) as well as DKD in animal models (11, 27) and human subjects (43, 44). TGF-β1 is a master regulator of profibrotic genes in the pathogenesis of kidney disease (11, 45, 46), and blockade of TGF-β1 prevents profibrotic events (43). TGF-β1 can increase the risk of developing CKD via multiple pathways and mechanisms. These include the activation of various signaling kinases and pathways leading to activation of Smad transcription factors as well as via induction of key Smad3-dependent miRNAs/lncRNAs that are implicated in DKD (47). Previously, we demonstrated a mechanism for the autoregulation of TGF-β1 in renal MMCs involving the actions of miR-192 and miR200b/c and their targets, Zeb1/2 (27), and miR-200c has been shown to be induced by ER stress (48). Zeb1/2 are E box repressors that negatively regulate the expression of TGF-β1 as well as other collagens, such as Col1α2 and Col4α1, in kidney cells under diabetic conditions (27, 49, 50). In the present study, we showed that insulin treatment in MMCs significantly upregulated Tgf-β1 expression as well as miR-379 and miR-200c, but downregulated Zeb-2 expression. These alterations were not observed in miR-379 KO-HFD (male and female) mice or in insulin-treated miR-379 KO MMCs.

Together, these data support the notion that, under HFD, insulin resistance/hyperinsulinemia can increase TGF-β1, ER stress, and miR-379 (Fig. 8). miR-379 can downregulate its targets, like Edem3, which further promotes ER stress and miR-379 expression, which, in turn, can induce Tgf-β1 by inhibition of Zeb2 targeted by miR-200c. miR-200c can also be induced by ER stress (Fig. 8). TGF-β1, in turn, can induce miR-379 to create a feedforward amplifying loop while also enhancing the expression of key fibrotic factors associated with renal hypertrophy and kidney disease (Fig. 8). Tgf-β1 may be induced directly by hyperinsulinemia/HFD/insulin resistance due to increases in free fatty acids, calorie excess, and cross talk between adipose and the kidney. Therefore, during HFD and obesity, cross talk between TGF-β1, ER stress, and miR-379 can have amplifying roles on downstream pathological features associated with CKD.

It is possible that some of the renal protection observed in miR-379 KO-HFD mice relative to WT-HFD mice is due to the lower body weight gain in the former. In addition, our recent and unpublished data show that miR-379 KO-HFD mice display protection from adipose (22) and liver dysfunction, which may also contribute to the observed renal phenotypes. These data are consistent with studies showing that increased miR-379 is a biomarker of nonalcoholic fatty liver disease (51) and that deficiency of miR-379/miR-544 alleviates HFD-induced obesity and hepatic steatosis (52). Notwithstanding, our data showing the direct correlations between insulin levels, renal miR-379, and TGF-β1, strongly supporting a pathological role for miR-379 in the kidney during obesity.

Perspectives and Significance

Taken together, our study provides mechanistic understanding of the pathological role of miR-379 during renal injury and how miR-379 deficiency protects mice from HFD-induced kidney injuries. ER stress resulting from insulin resistance/hyperinsulinemia can increase expression of miR-379 via CHOP transcription factor and TGF-β1 via downregulation of Zeb2 transcription factor, which, in turn, can upregulate miR-379. Therefore, TGF-β1 can have an amplifying role on the expression of miR-379 and its downstream target genes related to obesity-induced kidney disease (Fig. 8). Importantly, our data showing efficacy of the miR-379 GapmeR in reducing the effects of TGF-β1 in MMCs suggest that targeting miR-379 (with modalities like GapmeRs) can be an effective therapeutic approach to limit renal injury and metabolic dysfunction in obesity.

SUPPLEMENTAL DATA

Supplemental Table S1 and Supplemental Figs. S1–S3: https://doi.org/10.6084/m9.figshare.20425440.v1.

GRANTS

This work was supported by National Institutes of Health Grants R01DK081705, R01DK065073, and R01HL106089 (to R.N.) and a Postdoctoral fellowship from the Larry L. Hillblom Foundation (to R.T.).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

M.A., M.K., and R.N. conceived and designed research; M.A., M.K., L.L., M.W., and R.T. performed experiments; M.A. analyzed data; M.A., M.K., and R.N. interpreted results of experiments; M.A. prepared figures; M.A. and R.N. drafted manuscript; M.A., M.K., and R.N. edited and revised manuscript; M.A., M.K., L.L., M.W., R.T., and R.N. approved final version of manuscript.