Variations in MicroRNA-25 Expression Influence the Severity of Diabetic Kidney Disease

Abstract

Diabetic nephropathy is characterized by persistent albuminuria, progressive decline in GFR, and secondary hypertension. MicroRNAs are dysregulated in diabetic nephropathy, but identification of the specific microRNAs involved remains incomplete. Here, we show that the peripheral blood from patients with diabetes and the kidneys of animals with type 1 or 2 diabetes have low levels of microRNA-25 (miR-25) compared with those of their nondiabetic counterparts. Furthermore, treatment with high glucose decreased the expression of miR-25 in cultured kidney cells. In db/db mice, systemic administration of an miR-25 agomir repressed glomerular fibrosis and reduced high BP. Notably, knockdown of miR-25 in normal mice by systemic administration of an miR-25 antagomir resulted in increased proteinuria, extracellular matrix accumulation, podocyte foot process effacement, and hypertension with renin-angiotensin system activation. However, excessive miR-25 did not cause kidney dysfunction in wild-type mice. RNA sequencing showed the alteration of miR-25 target genes in antagomir-treated mice, including the Ras-related gene CDC42. In vitro, cotransfection with the miR-25 antagomir repressed luciferase activity from a reporter construct containing the CDC42 3′ untranslated region. In conclusion, these results reveal a role for miR-25 in diabetic nephropathy and indicate a potential novel therapeutic target for this disease.

Diabetic nephropathy (DN) is a progressive kidney disease that develops secondary to diabetes, and it is the single most common disorder leading to renal failure in developing countries.¹ The percentage of individuals with DN in China also exceeds the percentage with CKD related to GN.² DN is characterized by persistent albuminuria (>300 mg/d), a progressive decline in the GFR, and elevated arterial BP.³ BP control and efficient blockade of the renin-angiotensin system (RAS) to reduce microalbuminuria together with tight glycemic control are important therapeutic strategies for DN.⁴ However, this strategy is insufficient to prevent progression to renal failure in a substantial proportion of patients.⁵ Thus, novel effective therapeutic approaches are critical for the treatment and prevention of the disease. Intriguingly, recent studies have shown a link between ACE2/Apelin signaling and microRNAs (miRNAs) in the pathogenesis of hypertension, which serves as a reminder that the deregulation of miRNA expression plays a key role in hypertension, especially originating from DN.⁶ miRNAs are short noncoding RNAs of approximately 22 nucleotides in length that exert their canonical function by binding the 3′ untranslated region (3′UTR) of mRNAs and regulate the expression of 50% of genes.^7,8 It is well known that one miRNA can target multiple mRNAs, whereas one mRNA can be bound by a combination of numerous miRNAs. Therefore, miRNAs and their target mRNAs form an intricate regulatory network that can be expressed as a reciprocal “one-to-more” relationship, which has been hypothesized as a “fail-safe” mechanism to control gene expression.⁹ In mice, a podocyte-specific knockout of Dicer, an enzyme critical to the miRNA biogenesis pathway, leads to progressive renal glomerular and tubular damage.^10–12 A similar phenotype is also observed when Drosha is inducibly deleted in podocytes.¹³ Thus, we conclude that excessive deficiency of miRNAs due to deletion of Dicer and Drosha contributes to renal dysfunction and suggest that dysregulation of miRNAs can promote kidney disease. Emerging evidence confirms that individual miRNAs, including microRNA-23b (miR-23b),¹⁴ miR-192/216a/217,^15,16 and miR-377,¹⁷ are involved in DN. Recent studies have also shown that miRNAs, including miR-143/145, are associated with the pathogenesis of hypertension.^18–20 In this study, we have shown that miR-25 is commonly downregulated in the serum and kidneys of patients with DN, the kidneys of mouse models of type 1 and 2 diabetes, and specific kidney cells exposed to high glucose. We also observed that downregulation of miR-25 in normal mice causes severe renal disease with hypertension, and upregulation of miR-25 in db/db mice can reverse DN alterations and reduce high BP.

Results

Overexpression of miR-25 Reverses Albuminuria and Fibrosis in Diabetes

miR-25 levels in the sera of patients with diabetes both with and without nephropathy were downregulated compared with those in healthy controls (Figure 1A). A similar downregulation of miR-25 occurred as detected by quantitative PCR (qPCR) in the kidney tissues of streptozotocin-induced diabetic and db/db mice (Figure 1, B and C) and cultured cells exposed to high glucose (Supplemental Figure 1, A–C). We also found that miR-25 in the serum of db/db mice was decreased significantly compared with in the wild type (Supplemental Figure 1D). To regulate their target mRNAs, the miRNA minimum threshold of expression must be reached, and the abundance of miRNAs in the miRNAome of a specific cell or tissue may be more important for their function at the organismal level.²¹ We examined miR-25 levels in multiple tissues, including the kidneys, and found that miR-25 was highly expressed in mouse heart, muscle, adipose tissue, and kidneys compared with other tissues (Figure 1D). To establish the functional consequence of miR-25 upregulation in the kidneys, we performed intravenous miR-25 agomir (2.5 mg/kg) injection in db/db mice. Delivery efficiencies showed an approximately fourfold increase in kidney miR-25 expression (Supplemental Figure 1E). Importantly, miR-25 agomir injection in db/db mice decreased their proteinuria (Figure 1E, Supplemental Figure 1F) and reversed their glomerular fibrosis as shown by H&E, PAS, Sirius Red, and α-SMA staining (Figure 1, F and G). Furthermore, miR-25 agomir treatment reversed the upregulation of a series of fibrosis gene markers; however, miRNEG treatment had no such effect (Supplemental Figure 1G). miR-25 overexpression also alleviated the amount of foot process fusion and glomerular basement membrane (GBM) thickening observed in db/db mice (Figure 1, F, H, and I).

Figure 1.

Excessive miR-25 reverse albuminuria and fibrosis in diabetic nephropathy. (A) miR-25 levels in the sera of patients with type 1 diabetes (T1DM), type 2 diabetes (T2DM), or DN were reduced compared with the healthy controls (Con). (B and C) Similarly, miR-25 was reduced in the kidneys of type 1/2 diabetic (Dia [n=5] and db/db [n=5]) animals compared with controls (Con; n=5). (D) miR-25 expression levels in various tissues of normal C57BL/6 mice. (E) Quantification of 24-hour albumin excretion in miR-25 agomir–treated mice. (F) Representative images of H&E (row 1), PAS (row 2), Sirius Red (row 3), α-SMA (row 4), and TEM (row 5) staining from kidney sections of miR-25 agomir–injected mice. Green arrows indicate areas of fibrosis, an asterisk indicates the GBM, and a red arrow represents a foot process. (G) Quantification of the α-SMA–positive area within glomeruli from miR-25 agomir–treated kidney sections. (H) Quantification of GBM thickening after miR-25 agomir injection. (I) Quantification of foot process fusion from miR-25 agomir–treated db/db mice. (J) Measurement of BP from miR-25 agomir–injected db/db mice. Data are shown as the mean±SEM. **P<0.01; ***P<0.001.

Overexpression of miR-25 Alleviates Secondary Hypertension in Diabetes

One feature of DN is hypertension. To examine if miR-25 could reverse high BP in db/db mice, we first showed that miR-25 agomir treatment significantly alleviated hypertension compared with that in untreated db/db mice; however, miRNEG treatment did not (Figure 1J). Because kidney artery resistance (KRI) is an important feature of hypertension,²² we performed a color Doppler ultrasound to determine if overexpression of miR-25 was related to the decreased KRI. This imaging showed that miR-25 agomir significantly reduced KRI compared with in db/db and miRNEG-treated mice (Figure 2A). Alteration of the RAS is a major cause of hypertension. To identify if miR-25 agomir treatment decreased hypertension by inactivating the RAS, using immunochemistry and ELISA to detect renin, angiotensin I/II, aldosterone, and angiotensin 1, we subsequently determined that the RAS was inhibited as shown by their decreased levels (Figure 2, A and B). We also confirmed using vWf immunochemistry that overexpression of miR-25 in db/db mice could rescue their blood vessel structural abnormalities; however, the miRNEG had no effect (Figure 2A).

Figure 2.

Overexpression of miR-25 inhibit RAS activation. (A) Representative images of KRI (row 1), angiotensin 1 (AT1; row 2), renin (row 3), and vWf (row 4) staining from miR-25 agomir–treated db/db mouse kidney sections (right panel represents a quantification of the images). (B) Quantification of an ELISA for serum renin, angiotensin I/II, and aldosterone (ALD) from miR-25 agomir– and antagomir–treated mice. Data are shown as the mean±SEM. Con, control. ***P<0.001.

Inhibition of miR-25 Causes Proteinuria and Kidney Fibrosis

Because miR-25 is downregulated in DN, we explored whether inhibition of miR-25 could accelerate kidney fibrosis and proteinuria in normal mice. Therefore, wild-type (nondiabetic) mice were treated intravenously with 80 mg/kg miR-25 antagomir or miRNEG antagomir for 3 consecutive days and euthanized at the end of 6 weeks. Delivery efficiencies, assessed by measuring miR-25 expression in the kidneys, showed an approximately sevenfold decrease in kidney miR-25 expression (Supplemental Figure 1H). We observed that inhibition of miR-25 also increased the level of proteinuria and the urinary albumin-to-creatinine ratio, but miRNEG had no such effect (Figure 3, A and B). H&E, PAS, and collagen staining of the kidneys from miR-25 antagomir–treated mice revealed increased deposition of fibrillar collagens, mainly in the glomeruli, compared with in the control and miRNEG-treated mice (Figure 3, C–F). We also analyzed fibrosis-related transcripts by qPCR. In the miR-25 antagomir–treated mice, mRNA expression levels of collagen 1α (I), collagen 4α (IV), Tgf-β1, and Fn in the kidneys were significantly increased compared with those of controls and miRNEG-treated mice (Supplemental Figure 1I). In keeping with these findings, mesangial expansion, GBM thickening, and podocyte foot collapse were observed under electron microscopy in miR-25 antagomir–treated mice but were not observed in the mice receiving miRNEG injections (Figure 3, D and G). Because accumulated renal fibrosis and the collapse of the foot processes are the main characteristics of glomerulosclerosis, we, therefore, measured immunostaining for glomerulosclerosis marker PODXL and found that its expression was significantly decreased (Figure 3, D and H). A series of genes related to glomerulosclerosis displayed a similar decrease compared with in control and miRNEG-treated mice (Supplemental Figure 2A).

Figure 3.

Inhibition of miR-25 resulted in proteinuria and kidney fibrosis in normal mice. (A and B) Quantification of 24-hour albumin excretion and urine albumin-to-creatinine ratio (ACR) in miR-25 antagomir mice. (C and D) Representative images of H&E, PAS, collagen I/IV, TEM, and PODXL staining from kidney sections of miR-25 antagomir–injected mice. Asterisks indicates mesangial expansion, and arrows indicate foot processes in the TEM images. (E and F) Quantification of the collagen I/IV–positive area within the glomeruli. (G) Quantification of GBM thickening after miR-25 antagomir injection. (H) Quantification of the PODXL-positive staining area within the glomeruli. Data are shown as the mean±SEM. Con, control. ***P<0.001.

Inhibition of miR-25 Activates the Renin-Angiotensin-Aldosterone System

Additionally, we explored whether inhibition of miR-25 could accelerate hypertension in wild-type mice. We performed a color Doppler ultrasound to determine if miR-25 was inhibited relative to the abnormal KRI. The ultrasound showed that miR-25 antagomir treatment significantly increased KRI (Figure 4A). Antagomir-25–treated mice displayed high BP compared with control and miRNEG mice (Figure 4B). The increased BP corresponded to RAS activation and blood vessel dysfunction detected by immunochemistry (Figure 4, C–E), and levels were not changed by miRNEG injection. Similar upregulation of the RAS was detected by ELISA in the sera from miR-25 antagomir–injected mice compared with control and miRNEG mice (Figure 4, F–I). We also detected significantly elevated expression changes in a series of genes related to the RAS using qPCR (Supplemental Figure 2B). Oxidant stress is a strong contributor to RAS activation. To rule out whether hypertension caused by miR-25 inhibition is related to oxidant stress, we analyzed several markers of oxidant stress and found significant alterations in these markers in miR-25 antagomir–injected mice (Supplemental Figure 2, C–E). However, no such change occurred in miR-25 agomir–treated mice (Supplemental Figure 2, C–E).

Figure 4.

Decreased miR-25 cause high blood pressure and activate RAS. (A) Representative image of KRI by color Doppler ultrasound from miR-25 antagomir–treated mice, with quantification of the KRI in lower right panel. (B) Quantification of the BP from miR-25 antagomir–injected mice. (C–E) Representative images of renin, angiotensin 1 (AT1), and vWf staining of kidney sections from miR-25 antagomir–treated mice; lower right panels represent the quantification of renin-, AT1-, and vWf-positive areas. (F–I) Quantification of the ELISA for serum renin, angiotensin І/П, and aldosterone (ALD) from miR-25 antagomir–injected mice. Data are shown as the mean±SEM. Con, control. *P<0.05; **P<0.01; ***P<0.001.

Overexpression of miR-25 Does Not Cause Kidney Injury

We also injected the agomir-25 into normal mice to evaluate if this treatment could disturb kidney function. miR-25 expression was increased significantly by the treatment, but no significant increase in proteinuria was detected (Figure 5, A and B). Renal histology of miR-25 agomir–treated mice did not display any obvious changes in arteriolar thickening, tubular dilation and atrophy, GBM thickening, or mesangial expansion (Figure 5C). Only under electron microscopy was a weak partial flattening of podocyte foot processes relative to controls visible after miR-25 treatment, but the effect was very weak and was not significantly different from that of the miR-25 antagomir–treated group (Figure 5C). Furthermore, on the ultrasound, the KRI and BP were not increased in agomir-25–injected mice compared with controls (Figure 5D). This finding is consistent with the expression level of RAS-related genes and serum renin, angiotensin І/II, and aldosterone (Figure 5, E and F, Supplemental Figure 2B).

Figure 5.

Overexpression of miR-25 does not cause kidney injury in normal mice. (A) miR-25 agomir transfection efficiency in the kidney. (B) Quantification of the 24-hour albumin excretion in miR-25 agomir–treated normal mice. (C) Representative images of H&E, PAS, collagen I, and TEM staining from kidney sections of miR-25 agomir–treated normal mice. (D) Representative images of KRI from an miR-25 agomir–treated mouse kidney. The right panel represents quantification of the KRI. (E) Representative images of vWf, renin, and angiotensin 1 (AT1) staining from kidney sections of miR-25 agomir–injected wild-type mice. (F) Quantification of the ELISA for serum renin, angiotensin І/П, and aldosterone (ALD) from miR-25 agomir–injected mice. Data are shown as the mean±SEM. Con, control. ***P<0.001.

miR-25 Target Gene Alteration Is Responsible for Disrupted Kidney Function

miRNAs function as post-transcriptional regulators, mainly by affecting the stability of their target mRNAs.²³ To reveal mRNA transcriptomic changes responsible for the kidney impairment observed in mice with miR-25 inhibition and/or overexpression, we performed RNA sequencing (RNA-seq) analyses using kidney tissues from wild-type and miR-25 agomir– or antagomir–treated mice. According to the TargetScan program, the miR-25 family is predicted to target 674 mRNAs in the kidney. A total of 411 mRNAs were significantly dysregulated in miR-25 antagomir–injected mouse kidneys (negative binomial regression analysis, P<0.05), including 105 upregulated and 306 downregulated mRNAs (Figure 6A, Supplemental Figures 3 and 4, Supplemental Tables 1–3, Supplemental Material). By contrast, in miR-25 agomir–injected mice, 228 miR-25 targeted mRNAs were significantly altered, including 31 upregulated and 197 downregulated mRNAs (Figure 6B). We hypothesize that miR-25 caused these alterations by regulating these targeted 674 genes directly or indirectly. To evaluate whether these targeted genes, as a whole, were directly affected by miR-25 antagomir as primary effects, we randomly selected 674 nontarget genes expressed in the kidney and compared the P values between the target and nontarget genes in the RNA-seq data (Figure 6C). Interestingly, the average P value of the 674 target genes was significantly lower than that of the control nontarget genes, suggesting that the expression of these direct target genes is indeed altered in miR-25 antagomir–injected mice, whereas miR-25 agomir–treated mice did not show this effect (Figure 6D). One gene can be targeted by multiple miRNAs; thus, the miRNA number is also an important reason for regulating gene expression. We also found that the amounts of miR-25 target gene alterations were negatively correlated with the number of miRNA-targeted genes in miR-25 antagomir– but not agomir–injected mouse kidneys (Figure 6, E and F). To detect whether miR-25 could positively regulate its target genes, we performed a luciferase assay, which confirmed that inhibition of miR-25 could significantly repress CDC42 3′-UTR luciferase activity and could cause a mild decrease in RAP1B activity (Figure 6G), whereas we saw no such repression when we performed similar experiments using overexpression of miR-25, which elevated RAP1B activity (Figure 6H). Therefore, these results support the hypothesis that the kidney failure that results from miR-25 inhibition mainly originates from direct alteration of miR-25 targets. To determine the exact signaling pathway involved in the kidney failure caused by inhibition of miR-25, we used a bioinformatics approach to construct a protein network (string-db.org) of miR-25 targets. We discovered that genes in the proto-oncogene Ras signaling pathway (Mapk8, Pik3ca/b, Pik3r3, Cdc42, and Rap1a/b) were at the center of this network (www.kegg.jp) (Supplemental Figure 5A). Most interestingly, we analyzed all of the altered miR-25 targets using the KEGG pathway database and discovered that the Ras signaling pathway was the pathway most significantly altered in miR-25 antagomir–injected mouse kidneys (Supplemental Figure 5B). However, miR-25 agomir–treated mouse kidneys do not display altered Ras signaling (Supplemental Figure 5C). The mRNA levels of these Ras-related genes as measured by real-time PCR were significantly downregulated in miR-25 antagomir–treated mice but were not altered in miR-25 agomir–injected mice (Supplemental Figure 6, A and B). Furthermore, these same genes were also downregulated in mice with DN, but the effect could be reversed by miR-25 overexpression (Supplemental Figure 6C). miR-25 has also been reported to alleviate oxidant stress in DN.²⁴ Thus, we assessed some markers of oxidant stress to elucidate whether they compose a part of the mechanism of miR-25–induced kidney failure. Interestingly, we found that oxidant stress was significantly activated in miR-25 antagomir–injected mice but not in miR-25 agomir–treated mice (Supplemental Figure 2, C–E). Most importantly, the overexpression of miR-25 in mice with DN could reverse this oxidant stress (Supplemental Figure 2, F–H).

Figure 6.

miR-25 positive regulate target gene alteration is responsible for kidney dysfunction. (A and B) Scatterplots representing the alteration of miR-25 target genes from miR-25 antagomir– or agomir–treated mice compared with controls. (C and D) Densitometry of the P values of miRNA target and control genes in the RNA-seq data from miR-25 antagomir– and agomir–injected mice, respectively. (E and F) The correlation between the number of miRNA-targeted mRNAs or agomir-injected mice. (G and H) Luciferase activity in HEK293A cells that were transfected with the indicated 3′-UTR reporter constructs showing binding of miR-25 to the 3′-UTR of CDC42. Data are shown as the mean±SEM. **P<0.01.

Discussion

In this study, we found that miR-25 was decreased in patients with DN and diabetic animals or cultured cells exposed to elevated glucose. Additional study showed a protective effect of miR-25 against DN and DN-induced secondary hypertension. Diabetic kidneys showed increased extracellular matrix protein production and RAS activation. However, miR-25 agomir injection inhibited such processes in DN and also reduced proteinuria in vivo, possibly by inhibiting podocyte dysfunction and oxidative stress. Notably, inhibition of miR-25 in normal mice resulted in increased kidney fibrosis, podocyte collapse, proteinuria, and hypertension. By contrast, overexpression of miR-25 in wild-type mice failed to induce obvious kidney dysfunction or abnormal BP. Thus, increasing miR-25 would be advantageous from the therapeutic perspective of diabetic kidney disease. Additionally, we have shown that miR-25 target gene alteration induced by reduction in the levels of the miRNA may contribute to disrupted kidney function and that this process may be regulated by the signaling pathway involving the proto-oncogene Ras.

Oxidative stress contributes to the pathogenesis of DN, and thus, antioxidants are used as an adjunctive therapeutic target for diabetic complications.²⁵ In this study, we discovered that attenuation of miR-25 in normal mice resulted in increased production of oxidative stress; however, overexpression of miR-25 could reverse this process in db/db mice. In turn, augmentation of miR-25 did not noticeably increase oxidative stress biosynthesis in normal mice. miR-25 has been reported to negatively regulate NADPH oxidase 4 (NOX4) in a rat DN model,²⁴ whereas in this study, we observed decreased NOX4 levels in miR-25 antagomir–treated mice. A debate that centers around NOX4, which is responsible for the production of superoxide, is whether the protein might function as a “double-edged sword” to confer both protective and detrimental effects.²⁶

Oxidative stress can also promote Ras signaling pathway dysfunction. In this project, we revealed that the kidney failure induced by miR-25 inhibition was related to the Ras signaling pathway. Ras is a member of a family of related G proteins that are ubiquitously expressed in all cell lineages and organs, and a series of studies have confirmed that Ras signaling plays a key role in renal fibrosis.^27–29 Herein, we found that some Ras signaling–related genes (e.g., Cdc42 and Rap1a/b) and other likely miR-25 targets were significantly decreased in miR-25 antagomir–treated mice. However, miR-25 agomir–injected mice do not show this effect. Overexpression of miR-25 in diabetic mice could also reverse these gene alterations. Recent studies have indicated that in vivo podocyte–specific deletion of CDC42, a downstream effector of Ras, leads to congenital nephrotic syndrome and glomerulosclerosis.³⁰ In mouse models, podocyte-specific inactivation of Rap1a and Rap1b induced massive glomerulosclerosis and premature death.³¹ Other studies have also reported a protective role for RAP1 against tubular damage in DN, primarily acting to modulate mitochondria-derived oxidative stress.³² Hypertension is a secondary complication from DN; here, we observed that miR-25 inhibition led to high BP in normal mice and caused severe kidney dysfunction, whereas overexpression of miR-25 could reverse this phenotype. Oxidative stress and Ras signaling contribute to diabetic secondary hypertension,³² and decreased expression of MAPK8 and Ras-related gene RAP1B are associated with hypertension. In this study, we observed that MAPK8 and RAP1B are reduced in miR-25 antagomir–injected mice with hypertension, and therefore, we speculate that the reversal of hypertension by miR-25 excess may occur through Ras signaling. On the basis of the RNA-seq results, most of miR-25’s potential targets in kidney cells in normal mice appear surprisingly to be downregulated not to be upregulated by the miRNA. Recently, a new perspective has emerged that the result of miRNA:mRNA targeting is not limited to mRNA destabilization or inhibition but also, can stabilize and activate transcription or translation.^33–35 Proof of miR-25’s positive regulation of one potential target, CDC42, was shown by changes in a luciferase reporter induced by inhibition of miR-25 but not by its overexpression. Thus, we speculate that inhibition of miR-25 may destabilize its target mRNAs, which in turn, may contribute to kidney dysfunction in normal mice, but the exact mechanism by which miR-25 reduction promotes kidney dysfunction will need to be resolved in the future.

Collectively, our findings reveal an essential role of miR-25 in normal kidney function. Further characterization may reveal more subtle defects, and detailed molecular analyses will define the gene networks responsible for the critical developmental processes controlled by this miRNA.

Concise Methods

Serum and Tissue Collection from Human Subjects with Diabetes

This study was approved by the Ethics Review Board of Mudanjiang Medical University, Heilongjiang, China. Informed consents were obtained from all patients. All blood samples and kidney tissues were collected from patients at the Affiliated Hong Qi Hospital, Mudanjiang Medical University from August of 2014 to February of 2017. All samples were transported to the laboratory within 30 minutes after collection. Clinical information from the patients is presented in Supplemental Table 1.

Animal Treatment

The Mudanjiang Medical University Animal Care and Veterinary Services approved all protocols. The investigations conform to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health. Male C57BL/6 mice, weighing 24–26 g (9 weeks), were obtained from the SLAC Laboratory (Shanghai, China). The animals were fed on a standard rodent diet and water ad libitum. One group of animals (n=5) received weekly intravenous injections by tail vein of 2.5 mg/kg miR-25 agomir (Rabio Co., Guangzhou, China) for 6 weeks, and another group received the same dose of a scrambled control miRNA in the tail vein (n=5). Normal male C57BL/6 mice were intravenously injected by tail vein with 80 mg/kg miR-25 antagomir and miRNEG for 3 consecutive days, and two additional injections were performed for 6 weeks (n=4).³⁶ The animals were euthanized at the end of 6 weeks after the injections. Male db/db mice (10 weeks; n=5) were obtained from the SLAC Laboratory (Shanghai, China), and db/db animals received weekly intravenous injections by tail vein of 2.5 mg/kg miR-25 agomir and miRNEG for 1 month. Type 1 diabetes was induced by intraperitoneal injections of streptozotocin (in citrate buffer [pH 5.6] by three 70-mg/kg consecutive injections on alternate days; n=5). Diabetes was defined as a blood glucose level >17 mmol/L on 2 consecutive days, and the animals were euthanized after 3 months of diabetes. BP measurements (RM6240 Biologic Signal Collection and Processing System; Chengdu Instrument Factory) and echocardiography (VEVO 2100; STTARR, Toronto, Canada) were performed on all of the mice receiving miR-25 agomir, miR-25 antagomir, or scrambled miRNA before euthanasia of the animals. Kidneys, hearts, and serum were collected, and the majority of the tissues were stored at −80°C.

Cell Culture

Human kidney 2 epithelial cells and human renal glomerular endothelial cells were obtained from ATCC (Shanghai, China). Mouse podocytes were purchased from Peking Union Medical College Basic Medical Sciences Cell Resource Center. The cells were serum starved overnight before all experiments, and the cells were treated with normal glucose (5 mmol/L d-glucose), high glucose (25 mmol/L d-glucose), or osmotic control (25 mmol/L l-glucose) for 24 hours.

miRNA Extraction and Analysis

For serum miRNA extraction, samples were centrifuged at 8000×g for 20 minutes, and the supernatants were collected. After adding 750 μl TRIzol LS (Invitrogen Life Technologies, Carlsbad, CA) into 250 μl effusion samples, another 25 μl ath-miR-156a (20 nM; synthesized by Rabio Co.) was supplemented into each sample tube. Total RNA was extracted using TRIzol LS reagent (Invitrogen Life Technologies) according to the manufacturer’s instructions. The miRNA expression levels were quantified by real-time PCR. Briefly, miRNAs were reverse transcribed to cDNA with the MMLV reverse transcription kit (Promega Corp., Madison, WI). Subsequently, real-time PCR was performed on an AB7300 (Stratagene, Valencia, CA) using ath-miR-156a as a reference and a mixture of RNA extracted from the patient with diabetes as the calibrator. For cell and tissue extraction, miRNAs were extracted from the cells and tissues using the OMEGA RNA isolation kit (Omega Bio-Tek Inc., Norcross, GA) according to the manufacturer’s instructions. Real-time PCR was used with a final reaction volume of 20 μl containing 9 μl Fast Start Universal SYBR Green Master Mix (Roche, Beijing, China), 7.4 μl nuclease-free water, 0.8 μl miRNA primers (Rabio Co.), and a 2-μl RT product. The data were normalized to ath-miR-156a or RNU6B small nuclear RNA by a standard curve method to account for differences in reverse transcription efficiencies and the amount of template in the reaction mixtures.

mRNA Extraction and RT-PCR

Total RNA was extracted from cells and tissues using TRIzol (Invitrogen, Shanghai, China) according to the manufacturer’s instructions. First-strand cDNA was synthesized from 2 μg total RNA using Moloney murine leukemia virus reverse transcription and a Random Primers kit (Promega Corp.). The ribosomal protein S16 mRNA level served as the internal control. The primer sequences used are listed in Supplemental Table 2.

Luciferase Reporter Assay

RAS-related gene 3′-UTR luciferase vector containing the miR-25 response elements was amplified by PCR from mice cDNA. Plasmid DNA (wt-Luc or β-galactosidase control vector) and miR-25 agomir and antagomir were cotransfected into HEK293A cells for 48 hours. Luciferase activity was measured using SpectraMax M5 (Molecular Devices, Sunnyvale, CA) and normalized by measuring β-galactosidase activity. The primers used to generate specific fragments for the mouse RAS-related gene 3′-UTR are listed in Supplemental Table 3.

Histology, Immunofluorescence, and Immunohistochemistry

For immunofluorescence and immunochemistry, mouse mAb against PODXL, actin, collagen I and IV (Abcam, Shanghai, China), renin, angiotensin 1 (Santa Cruz Biotechnologies Inc., Santa Cruz, CA), vWf (EMD Millipore, Billerica, MA), and goat polyclonal secondary antibody to mouse and rabbit HRP (Abcam) were used. For quantitative morphometry, cells stained in ten randomly selected micrographs were counted using Image ProPlus software (Image-Pro Plus; Media Cybernetics, Rockville, MD). For H&E, PAS, and Sirius Red staining, kidney paraffin sections (5 μm) were stained using PAS (ScyTek Laboratories Inc., Logan, UT) and Sirius Red (Abcam) kits according to the manufacturer’s protocols.

ELISA

The mouse blood samples were centrifuged 5000 rpm for 10 minutes, and the serum was collected. An ELISA for renin, angiotensin 1/2, and aldosterone was performed using a commercially available kit for mouse (Abcam; ENZO Life Sciences, Farmingdale, NY) according to the manufacturer’s instructions.

Ultrastructural Analyses

Kidney tissues were fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) at 4°C for 24 hours. The samples were then washed with phosphate buffer (0.1 M, pH 7.4) for 12 hours and postfixed for 20 minutes in 1% OsO₄ in 0.1 M phosphate buffer (pH 7.4). The samples were then washed with phosphate buffer (0.1 M, pH 7.4) for 30 minutes, dehydrated, and embedded in Epon. Thin sections (50 nm) were placed on copper grids and stained with 2% uranyl acetate solution and a 1% solution of lead citrate for 30 minutes. A JEM-1010 transmission electron microscope was used to visualize the ultrastructure. From each specimen, ten randomly selected areas were photographed and analyzed using Image ProPlus software (Image-Pro Plus; Media Cybernetics, Shanghai, China).

RNA-Seq

A total amount of 1.5 μg RNA per sample was used as the input material for the RNA sample preparations. Sequencing libraries were generated using the NEBNext RNA Library Prep Kit for Illumina following the manufacturer’s recommendations (New England BioLabs, Ipswich, MA), and index codes were added to attribute sequences to each sample. The clustering of the index-coded samples was performed on a cBot Cluster Generation System using the HiSEquation 4000 PE Cluster Kit (Illumina, San Diego, CA) according to the manufacturer’s instructions. After cluster generation, the libraries were sequenced at the Novogene Bioinformatics Institute (Beijing, China) on an Illumina HisEquation 4000 platform, and 150-bp paired end reads were generated. RNA-seq data are available at the Gene Expression Omnibus under accession number GSE83144.

Statistical Analyses

Data are expressed as the mean±SEM and analyzed using the one-way ANOVA and a t test with Dunnett comparison using Prism (version 4; GraphPad Software Inc., San Diego, CA). The values were considered significantly different if the P value was <0.05.

Disclosures

None.