PUM2 Lowers HDAC9 mRNA Stability to Improve Contrast-Induced Acute Kidney Injury by Attenuating Oxidative Stress and Promoting Autophagy
Article information
Abstract
Background
Contrast-induced acute kidney injury (CIAKI) is the third leading cause of hospital-acquired acute kidney injury and diabetes mellitus (DM) has been identified as a risk factor for CIAKI. However, the molecular mechanism underlying DM-CIAKI remains unclear and requires further investigation.
Methods
Mouse and cell models of DM-CIAKI were established. Kidney function was evaluated by measuring biochemical indicators and using hematoxylin and eosin staining. Gene and protein abundance was assessed using real-time quantitative reverse transcription polymerase chain reaction, immunohistochemistry, immunofluorescence, and Western blotting. Glutathione peroxidase, superoxide dismutase, and malondialdehyde were measured using commercial kits, and reactive oxygen species were detected using a dihydroethidium (DHE) probe and the 2′,7′-dichlorofluorescein diacetate (DCFH-DA) method. Apoptosis in tissues and cells was evaluated by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). Cell viability and proliferation were measured using Cell Counting Kit-8 and 5-ethynyl-2′-deoxyuridine (EdU) assays. The interaction between pumilio RNA binding family member 2 (PUM2) and histone deacetylase 9 (HDAC9) was validated using RNA immunoprecipitation (RIP) and RNA pull-down assays.
Results
PUM2 expression was markedly reduced in DM-CIAKI models, whereas HDAC9 expression was notably increased. Subsequently, PUM2 silencing aggravated kidney injury in DM-CIAKI mice by enhancing oxidative stress and suppressing autophagy, whereas HDAC9 inhibition or HDAC9 silencing had the opposite effects. Mechanistically, PUM2 could suppressed the stability of HDAC9 mRNA, thereby attenuating HDAC9 expression. Furthermore, HDAC9 overexpression abolished PUM2 overexpression-mediated inhibition of oxidative stress and promotion of autophagy in high glucose- and contrast media-treated human kidney-2 (HK-2) cells.
Conclusion
PUM2 overexpression suppressed oxidative stress and promoted autophagy to alleviate renal injury in DM-CIAKI by interacting with HDAC9 mRNA, which mediated HDAC9 and mRNA degradation and inhibited HDAC9 expression.
Highlights
• PUM2 expression decreased and HDAC9 increased in DM-CIAKI.
• PUM2 knockdown or HDAC9 upregulation aggravated kidney injury in DM-CIAKI.
• PUM2 inhibited the stability of HDAC9 mRNA and reduced HDAC9 expression.
• PUM2 regulated oxidative stress and autophagy in DM-CIAKI by reducing HDAC9.
INTRODUCTION
With widespread use of contrast media (CM)-related clinical examination techniques, such as enhanced computed tomography and vascular diagnosis, the incidence of contrast-induced acute kidney injury (CIAKI) has increased and has gradually attracted clinical attention [1,2]. Currently, CIAKI is reported to be the third leading cause of hospital-acquired renal failure [3]. After receiving iodinated contrast agents, about 30% of patients develop acute kidney injury (AKI) [4]. Furthermore, evidence supports diabetes mellitus (DM) as a risk factor for CIAKI [5]. It has been reported that patients with kidney injury, especially those with DM-associated nephropathy, have a significantly increased incidence of CIAKI [6]. However, the pathogenesis of CIAKI is relatively complex, and current interventions have very limited effects [7]. Thus, the underlying molecular mechanism of CIAKI with DM requires further exploration.
The gene encoding pumilio RNA binding family member 2 (PUM2) produces a protein that belongs to the RNA-binding protein (RBP) family [8]. PUM2 has been reported to be implicated in various diseases, such as osteoporosis, glioma and acute ischemic kidney injury [9–11]. Wang et al. [10] suggested that PUM2 expression was evidently decreased in acute ischemic kidney injury and that PUM2 overexpression exerted protective effects on renal tubules by improving mitochondrial quality. These results indicate that PUM2 may play a crucial role in CIAKI with DM. At present, the role of PUM2 in AKI, especially CIAKI with DM, remains unclear, and further elucidation of the role of PUM2 and the underlying molecular regulatory mechanism in CIAKI with DM is needed.
RBPs, which have been widely investigated in multiple diseases, can mediate various post-transcriptional processes, such as affecting mRNA stability, function, and cellular localization to participate in the disease progression [12]. For instance, PUM1, an RBP, negatively regulated the stability of p21 mRNA and p21 expression in colorectal cancer [13]. As previously documented, PUM2 decreased mitochondrial fission factor (Mff) expression by interacting with the 3’ untranslated region (3’ UTR) of Mff mRNA in acute ischemic kidney injury [10]. In addition, PUM2 negatively regulated distal-less homeobox 5 (DLX5) expression by binding to DLX5 mRNA, thereby mediating osteoporosis progression [11]. Therefore, we sought to identify the target gene that binds to RBP PUM2 and affects gene expression, thereby participating in the progression of CIAKI with DM.
Histone deacetylases (HDACs) deacetylate histones, alter the static properties of chromatin, and facilitate gene suppression [14]. Previous research suggested that HDACs play multiple roles in kidney development and the pathogenesis of kidney disease [15]. Currently, HDACs are grouped into types I, II, III, and IV. Histone deacetylase 9 (HDAC9), a class II HDAC, has been implicated in tumors, inflammation, atherosclerosis, and diabetic nephropathy by affecting target genes [16–18]. Some HDACs, such as HDAC2 and HDAC5, have been reported to play harmful roles in septic AKI [19]. In addition, evidence supports the suppressive role of HDAC9 inhibition in diabetic nephropathy [20], implying that HDAC9 is probably implicated in CIAKI with DM. Thus, we aimed to provide more evidence about the role and underlying mechanism of HDAC9 in CIAKI with DM.
Oxidative stress and autophagy are reported closely related to CIAKI [21]. Some studies have supported that PUM2 and HDAC9 can regulate oxidative stress and autophagy in diseases [22–25]. On the basis of this background, we hypothesized that PUM2 attenuates oxidative stress and stimulates autophagy to alleviate renal injury in DM-CIAKI by interacting with HDAC9 mRNA and reducing HDAC9 mRNA stability and HDAC9 expression. Our findings identify additional target genes for CIAKI with DM.
METHODS
Mouse models of diabetes
C57BL/6J male mice aged 4 weeks were acquired from Hunan Slack King Experimental Animal Company (Changsha, China). To establish a DM model, mice were fed a high-fat diet for 4 weeks. Afterwards, 50 mg/kg streptozotocin (STZ) was intraperitoneally injected into mice for 5 consecutive days. After 3 days, mice with blood glucose levels above 200 mg/dL were diagnosed with diabetes. STZ-induced mice were maintained for an additional 4 to 5 weeks before contrast medium administration. Animal experiments in this study were approved by the ethics committee of the Second Xiangya Hospital (approval no.: 20240280).
CIAKI model
To achieve different experimental purposes, the mice were divided into several groups and treated differently. (1) Mice were divided into four groups, with six mice in each group: control, DM, CIAKI, and DM-CIAKI. (2) PUM2 knockout (PUM-KO) mice were purchased from Cyagen (catalog number: S-KO-18016; Suzhou, China, https://www.cyagen.cn/mice-bank/S-KO-18016). PUM2 wild-type (WT) and PUM2-KO mice were divided into the PUM2-WT, PUM2-KO, DM-CIAKI+PUM2-WT, and DM-CIAKI+PUM2-KO groups, with six mice in each group. (3) Mice were divided into five groups, with six mice in each group: control, DM, CIAKI, DM-CIAKI, and DM-CIAKI+HDCAi. (4) Mice were divided into three groups, with six mice in each group: control, DM-CIAKI+small interfering negative control (si-NC), and DM-CIAKI+si-HDAC9. As described previously [26], a CIAKI model was established in mice. Briefly, after 16 hours without water, NG-nitro-L-arginine methyl ester (L-NAME, 10 mg/kg intraperitoneally) was injected into mice to inhibit nitric oxide synthase. After 15 and 30 minutes, the prostaglandin synthesis inhibitor indomethacin (10 mg/kg intraperitoneally) and the low-osmolar monoiodinated contrast agent Omnipaque (iodine 3.0 g/kg; GE Healthcare Inc., Marlborough, MA, USA) were injected into the mice through the tail vein, respectively. Mice in the control group were not given Omnipaque (3.0 g iodine/kg) but were instead given saline. Other treatments were consistent with with those used in the experimental group. After iohexol or saline injection, mice were allowed free access to food and water for 24 hours. To investigate the influence of BRD4354, an HDAC9 inhibitor (HDACi), BRD4354 was intraperitoneally injected into mice at a dose of 10 mg/kg 20 minutes after the intraperitoneal injection of L-NAME and indomethacin. In addition, to silence HDAC9, a total of 100 μL lentivirus (1×105 transduction units [TU]/μL) carrying si-HDAC9 was injected into the renal tissues of mice. Finally, the contrast agent Omnipaque was injected into the mice. After 48 hours, the required samples were collected.
Immunofluorescence assay
Tissue samples and human kidney-2 (HK-2) cells were fixed with 4% paraformaldehyde and permeabilized with 0.3% Triton X-100. After blocking with 5% bovine serum albumin, the samples were incubated with primary antibodies against PUM2 (PA5-56436, Thermo Fisher Scientific, Waltham, MA, USA) and light chain 3B (LC3B; PA5-32254, Thermo Fisher Scientific) overnight at 4°C. Secondary antibodies were added, and 4′,6-diamidino-2-phenylindole (DAPI) was used to stain cell nuclei. Images were captured using a fluorescence microscope and fluorescence intensity was analyzed using ImageJ (National Institutes of Health, Bethesda, MD, USA).
Intracellular reactive oxygen species detection
To evaluate reactive oxygen species (ROS) levels in mouse renal tissues, the fluorescent probe dihydroethidium (DHE) was used. Briefly, renal tissues from mice were sectioned into 2-μm-thick frozen slices. These sections were then incubated with a 10 μM solution of the fluorescent probe DHE at 37°C for 1 hour in the dark. Following incubation, the stained renal tissues were examined and visualized using a fluorescent microscope (excitation: 515 nm; emission: 585 nm).
To assess ROS levels in HK-2 cells, a 2′,7′-dichlorofluorescein diacetate (DCFH-DA) (Biosharp, Seattle, WA, USA) was used for ROS detection. HK-2 cells were cultured in 6-well plates overnight. The cells were collected and stained with 10 μM DCFH-DA in the dark for 30 minutes. Fluorescence intensity (excitation: 488 nm; emission: 525 nm) was measured using a fluorescent microscope.
Cell treatment
HK-2 cells, a human kidney cell line, were purchased from the American Type Culture Collection (Manassas, VA, USA). All cells were cultured in DMEM/F12 medium (Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific), and 100 U/mL penicillin at 37°C in a humidified atmosphere containing 5% CO2.
To induce a cell model of DM, HK-2 cells were treated with different concentrations of D-glucose (5.5, 30, and 50 mM) for 48 hours. To induce a cell model of AKI with iohexol, HK-2 cells were treated with 150 mg/mL iohexol [27] for 0, 2, 4, and 6 hours. To inhibit the effect of HDAC9 on HK-2 cells, 2 μmol/L BRD4354 [28] was used to treat HK-2 cells for 2 hours. To suppress autophagy, 5 mM 3-methyladenine (3-MA) was added to HK-2 cells for 24 hours.
Cell transfection
Small interfering RNA targeting HDAC9 (si-HDAC9), overexpression plasmids for PUM2 (ov-PUM2) and HDAC9 (ov-HDAC9), and short hairpin RNA targeting PUM2 (sh-PUM2), along with corresponding controls (si-NC, ov-NC, and sh-NC), were purchased from GenePharma (Shanghai, China). HK-2 cells were seeded onto 6-well plates and incubated overnight. HK-2 cells were transfected with plasmids for 48 hours using Lipofectamine 3000 (Invitrogen, Waltham, MA, USA) according to the manufacturer’s instructions. Of note, the sequences of sh-PUM2, si-HDAC9, sh-NC, and si-NC are presented in Supplementary Table 1.
RNA immunoprecipitation assay
A Magna RNA immunoprecipitation (RIP) RNA-Binding Protein Immunoprecipitation Kit (Millipore, Burlington, MA, USA) was used to conduct the RIP assay. HK-2 cells (approximately 1×107 cells) were lysed with RIP lysis buffer for 5 minutes. The supernatants were obtained after centrifugation. Magnetic beads conjugated with anti-PUM2 antibody (ab92390, Abcam, Cambridge, UK) or immunoglobulin G (IgG; ab172 730, Abcam) were added to the supernatants and incubated overnight at 4°C. IgG served as the control. Immune-precipitated HDAC9 mRNA was determined using real-time quantitative reverse transcription polymerase chain reaction, and the pimer sequences are shown in Supplementary Table 2.
RNA pull-down assay
Approximately 4×107 HK-2 cells were collected and lysed in 1 mL of radio-immunoprecipitation assay (RIPA) buffer. The lysates were centrifuged at 13,000 rpm for 10 minutes at 4°C. Biotinylated RNA probes specific for HDAC9 were incubated with the lysates for 4 hours at 37°C. Subsequently, the lysate was incubated with streptavidin-coated magnetic beads (Invitrogen) at 4°C overnight. After the beads were washed five times with wash buffer, PUM2 enrichment by the biotinylated RNA probe was examined by Western blotting.
Statistical analysis
All data are expressed as the mean±standard error of the mean and were analyzed using SPSS version 16.0 statistical software (SPSS Inc., Chicago, IL, USA). Student’s t-test was used for comparisons between two groups. Comparisons among three or more groups were analyzed by one-way analysis of variance. P<0.05 was considered statistically significant. Each experiment was repeated at least three times. Some methods are described in the Supplementary Methods.
RESULTS
PUM2 expression was abnormally downregulated in DM-CIAKI mice, while HDAC9 expression was abnormally elevated
CIAKI has been reported to have a worse prognosis in patients with DM [29]. To investigate crucial molecules in DM-CIAKI, we established a model of CIAKI in STZ-induced mice and in normal mice. The detailed groups were control, DM, CIAKI, and DM-CIAKI. As presented in Supplementary Fig. 1A, compared with control mice, the ratio of kidney weight/body weight (KW/BW) and the levels of serum creatinine (SCr), 24-hour urinary protein, blood urea nitrogen (BUN), fasting blood glucose (FBG), and glycated serum protein (GSP) were markedly higher in DM, CIAKI, and DM-CIAKI mice, especially in DM-CIAKI mice. In addition, hematoxylin and eosin (HE) staining revealed obvious vacuolar degeneration, fragmented cells, and luminal dilatation in kidney tissue sections of DM, CIAKI, and DM-CIAKI mice, especially in DM-CIAKI mice (Supplementary Fig. 1B). PUM2 expression was evidently reduced, whereas HDAC9 expression was apparently enhanced in DM, CIAKI, and DM-CIAKI mice. The reduction in PUM2 expression and the elevation in HDAC9 expression were most obvious in DM-CIAKI mice (Supplementary Fig. 1C and D). Taken together, PUM2 and HDAC9 may play crucial roles in DM-CIAKI.
PUM2-KO enhanced oxidative stress in DM-CIAKI mice
To further probe the influence of PUM2 on DM-CIAKI, PUM2-KO mice were used. First, immunofluorescence assay revealed that PUM2 expression in PUM2-KO mice was greatly decreased compared with WT mice (Fig. 1A), indicating successful PUM2 knockdown in mice. Then, mice with or without PUM2-KO were divided into the PUM2-WT, PUM2-KO, DM-CIAKI+PUM2-WT, and DM-CIAKI+PUM2-KO groups. Fig. 1B indicated that PUM2 expression was decreased in the PUM2-KO, DM-CIAKI+ PUM2-WT, and DM-CIAKI+PUM2-KO groups relative to the PUM2-WT and was lowest in DM-CIAKI+PUM2-KO group among groups. Compared to DM-CIAKI+PUM2-WT group, the ratio of KW/BW and the levels of SCr, 24-hour urinary protein, BUN, FBG, and GSP in mice were dramatically increased by PUM2-KO (Fig. 1C). Moreover, PUM2-KO further aggravated kidney tissue injury in the DM-CIAKI+PUM2-WT group (Fig. 1D). In addition, DM-CIAKI induction promoted the levels of malondialdehyde (MDA) and ROS and decreased the levels of glutathione peroxidase (GSH-PX) and superoxide dismutase (SOD) in PUM2-WT mice; these effects were further strengthened by PUM2-KO (Fig. 1E and F). Moreover, terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) showed more apoptotic cells in the DM-CIAKI+ PUM2-WT group than in the PUM2-WT group, and this effect was further increased by PUM2-KO (Fig. 1G). Notably, as presented in Fig. 1C–G, PUM2-KO failed to change the ratio of KW/BW, the levels of SCr, 24-hour urinary protein, BUN, FBG, GSP, MDA, ROS, GSH-PX, and SOD, renal morphology, or cell apoptosis in mice in the PUM2-WT group. Collectively, PUM2-KO further deteriorated injury and oxidative stress in DM-CIAKI mice.
Pumilio RNA binding family member 2 (PUM2) knockout enhanced oxidative stress in diabetes mellitus (DM)-contrast-induced acute kidney injury (CIAKI) mice. (A) PUM2 expression in kidney tissues was measured in PUM2-knockout (KO) and PUM2-wild-type (WT) mice using immunofluorescence assay. (B–G) PUM2-KO and PUM2-WT mice were treated with streptozotocin and iohexol. The detailed groups were as follows: PUM2-WT, PUM2-KO, DM-CIAKI+PUM2-WT, and DM-CIAKI+PUM2-KO. (B) PUM2 expression in kidney tissues was examined using real-time quantitative reverse transcription polymerase chain reaction. (C) The kidney weight/body weight (KW/BW) ratio and the levels of serum creatinine, 24-hour urinary protein, blood urea nitrogen (BUN), fasting blood glucose (FBG), and glycated serum protein (GSP) were detected. (D) Kidney tissue morphology was examined by hematoxylin and eosin staining. (E) Glutathione peroxidase (GSH-PX), superoxide dismutase (SOD), and malondialdehyde (MDA) levels in kidney tissues were detected using commercial kits. (F) Reactive oxygen species production was evaluated using a dihydroethidium (DHE) probe. (G) Cell apoptosis in kidney tissues was investigated using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). aP<0.05, bP<0.01, cP<0.001.
HDAC9 inhibitor or HDAC9 knockdown suppressed oxidative stress and alleviated the injury in DM-CIAKI mice
To explore the impact of HDAC9 on DM-CIAKI. DM-CIAKI mice were prepared and then administrated BRD-4354, an inhibitor of HDAC9, or lentivirus carrying si-HDAC9. As indicated in Fig. 2A and B, and Supplementary Fig. 2A, HDACi treatment or HDAC9 knockdown partially reversed the DM-CIAKI-induced elevation of HDAC9 expression in mice. Moreover, the DM-CIAKI-induced increase in the KW/BW ratio and the levels of SCr, 24-hour urinary protein, BUN, FBG, and GSP in mice was attenuated by HDACi treatment or HDAC9 knockdown (Fig. 2C, Supplementary Fig. 2B). As expected, HDACi treatment or HDAC9 knockdown improved morphological injury in kidney tissues of DM-CIAKI mice (Fig. 2D, Supplementary Fig. 2C). Moreover, elevated levels of MDA and ROS and decreased levels of GSH-PX and SOD in DM-CIAKI mice were reversed by HDACi treatment or HDAC9 knockdown (Fig. 2E and F, Supplementary Fig. 2D and E). Furthermore, HDACi treatment or HDAC9 knockdown attenuated cell apoptosis in DM-CIAKI mice (Fig. 2G, Supplementary Fig. 2F). Overall, HDAC9 inhibition or HDAC9 knockdown alleviated kidney injury and inhibited oxidative stress and tissue apoptosis in DM-CIAKI mice.
Histone deacetylase 9 (HDAC9) knockdown suppressed oxidative stress and alleviated injury in diabetes mellitus (DM)-contrast-induced acute kidney injury (CIAKI) mice. Mice were treated with streptozotocin and iohexol and then received lentivirus carrying small interfering RNA targeting HDAC9 (si-HDAC9). The detailed groups were control, DM-CIAKI, and DM-CIAKI+si-HDAC9. (A, B) HDAC9 expression was detected using real-time quantitative reverse transcription polymerase chain reaction and Western blotting. (C) The kidney weight/body weight (KW/BW) ratio and the levels of serum creatinine, 24-hour urinary protein, blood urea nitrogen (BUN), fasting blood glucose (FBG), and glycated serum protein (GSP) were detected. (D) Kidney tissue morphology was examined by hematoxylin and eosin staining. (E) Glutathione peroxidase (GSH-PX), superoxide dismutase (SOD), and malondialdehyde (MDA) levels in kidney tissues were detected using commercial kits. (F) Reactive oxygen species production was evaluated using a dihydroethidium (DHE) probe. (G) Cell apoptosis in kidney tissues was investigated using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). NC, negative control. aP<0.05, bP<0.01, cP<0.001.
HDAC9 silencing improved high glucose and CM treatments-induced oxidative stress and apoptosis of HK-2 cells
Subsequently, HK-2 cells were subjected to a series of high glucose (HG) concentrations (5.5, 30, and 50 mM). As presented in Fig. 3A, HG treatment promoted HDAC9 expression in a concentration-dependent manner. In addition, under the 50 mM glucose condition, iohexol was used to treat HK-2 cells for 2, 4, and 6 hours. We found that iohexol (represented by CM in images) also elevated HDAC9 expression in HG-induced HK-2 cells in a time-dependent manner (Fig. 3B). Furthermore, HK-2 cells underwent CM (iohexol), HG+CM, and HG+CM+HDACi (BRD4354) treatments. As shown in Fig. 3C, HG or CM apparently decreased cell viality, and the combination of HG and CM produced the strongest inhibory effect. However, HDACi partially reversed the suppressive effect caused by HG and CM treatment. Moreover, HG or CM evidently enhanced ROS level in HK-2 cells, especially the combination of HG and CM, but HDACi attenuated the elevated ROS level caused by the combination of HG and CM treatments (Fig. 3D). Afterwards, si-HDAC9-1, si-HDAC9-2, and si-HDAC9-3 were used to knock down HDAC9 expression in HK-2 cells. si-HDAC9-3 transfection resulted in the most obvious downregulation of HDAC9 expression (Fig. 3E), and it was selected for subsequent experiments. si-HDAC9-3-transfected HK-2 cells were treated with the combination of HG and CM. Fig. 3F shows that HDAC9 knockdown abolished the HG+CM-mediated enhancement of HDAC9 expression. As expected, decreased cell viability and increased ROS production and cell apoptosis caused by the combination of HG and CM were attenuated by HDAC9 knockdown (Fig. 3G–I). Collectively, HDAC9 knockdown alleviated oxidative stress and decreased cell apoptosis in HG and CM-treated HK-2 cells.
Histone deacetylase 9 (HDAC9) silencing improved high glucose (HG)- and contrast media (CM)-induced oxidative stress and apoptosis in human kidney-2 (HK-2) cells. (A) HDAC9 expression was measured in HK-2 cells treated with 5.5, 30, and 50 mM glucose using Western blotting. Iohexol was used to treat HK-2 cells for 2, 4, or 6 hours in the presence of 50 mM glucose. (B) HDAC9 expression was measured using real-time quantitative reverse transcription polymerase chain reaction (RT-qPCR). HK-2 cells were subjected to 50 mM glucose and/or 150 mg/mL iohexol and then received HDAC9 inhibitor (HDACi) treatment. (C) Cell viability was examined using Cell Counting Kit-8 (CCK-8). (D) Reactive oxygen species (ROS) production was evaluated using 2′,7′-dichlorofluorescein diacetate (DCFH-DA) method. (E) HDAC9 expression was measured using Western blotting in HK-2 cells transfected with small interfering RNA targeting HDAC9 (si-HDAC9)-1, -2, or -3. HK-2 cells were transfected with si-HDAC9-3 and then treated with the combination of HG and CM. (F) HDAC9 expression was examined using RT-qPCR. (G) Cell viability was evaluated using CCK-8. (H) ROS production was evaluated using the DCFH-DA method. (I) Cell apoptosis was investigated using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). NC, negative control. aP<0.05, bP<0.01, cP<0.001.
PUM2 attenuated stability of HDAC9 mRNA and reduced HDAC9 expression
Based on the above results, PUM2 and HDAC9 were implicated in DM-CIAKI. Therefore, we suspected that there may be an interaction between the two. PUM2 is an RBP that has been reported to regulate the mRNA stability of target genes [30,31]. Further experiments including RIP and RNA-pull-down assays, validated the interaction between PUM2 and HDAC9 (Fig. 4A and B). Specifically, the PUM2 antibody successfully enriched HDAC9 mRNA, and the HDAC9 probe successfully pulled down PUM2 protein (Fig. 4A and B). In addition, PUM2 knockdown led to increased HDAC9 expression, whereas PUM2 overexpression had the opposite effect (Fig. 4C and D). Furthermore, PUM2 knockdown promoted the stability of HDAC9 mRNA, whereas PUM2 overexpression attenuated the stability of HDAC9 mRNA (Fig. 4E). Taken together, PUM2 overexpression suppressed HDAC9 mRNA stability and reduced HDAC9 expression through this interaction.
Pumilio RNA binding family member 2 (PUM2) attenuated the stability of histone deacetylase 9 (HDAC9) mRNA and reduced HDAC9 expression. (A, B) The interaction between PUM2 and HDAC9 was validated in human kidney-2 (HK-2) cells by RNA immunoprecipitation and RNA pull-down assays. HK-2 cells were transfected with an overexpression plasmid for PUM2 (ov-PUM2) or short hairpin RNA targeting PUM2 (sh-PUM2). (C, D) HDAC9 expression was evaluated using real-time quantitative reverse transcription polymerase chain reaction and Western blotting. (E) HDAC9 mRNA stability was detected in HK-2 cells after actinomycin D treatment using real-time quantitative reverse transcription polymerase chain reaction. IgG, immunoglobulin G; NC, negative control. aP<0.05, bP<0.01, cP<0.001.
HDAC9 overexpression abolished PUM2 upregulation-mediated alleviation of cell injury and suppression of oxidative stress in HG and CM treatments-induced HK-2 cells
To probe the role of the PUM2/HDAC9 axis in DM-CIAKI, HK-2 cells were subjected to ov-PUM2 and/or ov-HDAC9 under HG and CM treatments. The detailed groups were control, HG+CM+ov-NC, HG+CM+ov-PUM2, HG+CM+ov-HDAC9, and HG+CM+ov-PUM2+ov-HDAC9. Regarding PUM2 and HDAC9 expression, PUM2 overexpression promoted PUM2 expression and inhibited HDAC9 expression in HG- and CM-treated HK-2 cells. However, HDAC9 overexpression only increased HDAC9 expression and partially reversed the inhibitory effect of PUM2 overexpression on HDAC9 expression (Fig. 5A and B). In addition, PUM2 overexpression increased cell viability and proliferation, whereas HDAC9 overexpression had the opposite effects in HG- and CM-treated HK-2 cells. Notably, HDAC9 overexpression abolished the promoting effects of PUM2 overexpression on cell viability and proliferation (Fig. 5C and D). Furthermore, ROS production and cell apoptosis were inhibited by PUM2 overexpression in HG- and CM-treated HK-2 cells. However, HDAC9 overexpression produced the opposite effects and attenuated the suppressive effects of PUM2 overexpression on ROS production and cell apoptosis (Fig. 5E and F). Overall, PUM2 upregulation improved cell injury and suppressed oxidative stress in HK-2 cells under HG and CM treatments by decreasing HDAC9 expression.
Histone deacetylase 9 (HDAC9) overexpression abolished pumilio RNA binding family member 2 (PUM2) upregulation-mediated alleviation of cell injury and suppression of oxidative stress in high glucose (HG) and contrast media (CM)-treated human kidney-2 (HK-2) cells. HK-2 cells were transfected with an overexpression plasmid for PUM2 (ov-PUM2) and/or ov-HDAC9 and then treated with HG and CM. (A, B) PUM2 and HDAC9 expression was evaluated using real-time quantitative reverse transcription polymerase chain reaction and Western blotting. (C) Cell viability was evaluated using Cell Counting Kit-8 (CCK-8). (D) Cell proliferation was detected using 5-ethynyl-2′-deoxyuridine (EdU). (E) Reactive oxygen species (ROS) production was evaluated using the 2′,7′-dichlorofluorescein diacetate (DCFH-DA) method. (F) Cell apoptosis was investigated using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). NC, negative control. aP<0.05, bP<0.01, cP<0. 001.
The dysregulation of autophagy was observed in DM-CIAKI mice and was regulated by HDAC9 and PUM2
Growing evidence has demonstrated that autophagy is closely related to AKI including contrast-induced AKI [32]. Therefore, we detected the expression of LC3BII/I and p62, which were identified as indicators of autophagy, in various groups including control, DM, CIAKI, DM-CIAKI, and DM-CIAKI+HDACi. Our results showed that in the DM, CIAKI, and DM-CIAKI groups, the LC3BII/I ratio was evidently decreased, whereas p62 expression was apparently increased, especially in the DM-CIAKI group. However, HDACi treatment abolished the DM-CIAKI-mediated reduction in the LC3BII/I ratio and elevation of p62 expression in mice (Fig. 6A). In addition, lentivirus carrying si-HDAC9 was injected into DM-CIAKI mice, and this increased the LC3BII/I ratio and decreased p62 expression (Fig. 6B). Subsequently, we investigated how PUM2 influences the indicators of autophagy. PUM2-WT and PUM2-KO mice received DM-CIAKI treatments. PUM2 knockdown further strengthened the DM-CIAKI-mediated reduction in the LC3BII/I ratio and elevation of p62 expression in mice (Fig. 6C). Overall, autophagy was suppressed in DM-CIAKI mice and was regulated by HDAC9 and PUM2 in vivo.
Histone deacetylase 9 (HDAC9) downregulation inhibited oxidative stress and apoptosis caused by high glucose (HG) and contrast media (CM) induction in human kidney-2 (HK-2) cells by promoting autophagy. Mice were grouped into control, diabetes mellitus (DM), contrast-induced acute kidney injury (CIAKI), DM-CIAKI, and DM-CIAKI+HDAC9 inhibitor (HDACi) groups. (A) Light chain 3B (LC3B) II/I and p62 expression in kidney tissues was evaluated using Western blotting. DM-CIAKI mice were injected with lentivirus carrying small interfering RNA targeting HDAC9 (si-HDAC9). The detailed groups were control, DM-CIAKI+si-negative control (NC), and DM-CIAKI+si-HDAC9. (B) LC3BII/I and p62 expression in kidney tissues was evaluated using Western blotting. Mice were divided into PUM2-wild-type (WT), PUM2-knockout (KO), DM-CIAKI+PUM2-WT, and DM-CIAKI+PUM2-KO groups. (C) LC3BII/I and p62 expression in kidney tissues was measured using Western blotting. (D–G) HK-2 cells received the indicated treatments and the detailed groups were control, HG+CM+si-NC, HG+CM+si-HDAC9, and HG+CM+si-HDAC9+3-methyladenine (3-MA). (D) LC3BII/I and p62 expression was measured using Western blotting. (E) LC3B expression was investigated using immunofluorescence assay. (F) Cell viability was evaluated using Cell Counting Kit-8 (CCK-8). (G) Reactive oxygen species (ROS) production was evaluated using a dihydroethidium (DHE) probe. (H) Cell apoptosis was investigated using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). aP<0.05, bP<0.01, cP<0.001.
HDAC9 downregulation inhibited oxidative stress and apoptosis caused by HG and CM induction in HK-2 cells through promoting autophagy
Here, experiments were designed to assess the effects of autophagy on HDAC9 downregulation-mediated cell viability, oxidative stress and apoptosis in HG- and CM-treated HK-2 cells. HK-2 cells were transfected with si-HDAC9 and then treated with HG, CM and 3-MA, an inhibitor of autophagy. The detailed groups were control, HG+CM+si-HDAC9, and HG+CM+si-HDAC9+3-MA. Our results indicated that HG and CM treatments suppressed the LC3BII/I ratio and enhanced p62 expression, whereas HDAC9 knockdown reversed these alterations. However, 3-MA abolished the HDAC9 knockdown-mediated elevation of the LC3BII/I ratio and reduction in p62 expression in HG- and CM-treated HK-2 cells (Fig. 6D). As expected, 3-MA offset the HDAC9 knockdown-mediated promotion of LC3B expression (Fig. 6E). Furthermore, HDAC9 knockdown promoted cell viability and inhibited ROS production and cell apoptosis in HG- and CM-treated HK-2 cells, and these effects were offset by 3-MA treatment (Fig. 6F–H). Taken together, HDAC9 silencing suppressed oxidative stress and apoptosis in HG- and CM-treated HK-2 cells by promoting autophagy.
DISCUSSION
Various mechanisms have been linked to the renal toxicity of CM, including renal medullary hypoxia, oxidative stress, direct toxicity of contrast agents, inflammation, apoptosis, immunity and autophagy [33–35]. At present, mounting evidence has linked oxidative stress and autophagy to AKI including CIAKI in the context of DM [36,37]. Rapamycin, an autophagy inducer, can suppress inflammatory responses and oxidative stress and alleviate renal injury in iodixanol-induced DM rats [37]. In this study, our findings suggested that PUM2 inhibited oxidative stress and promoted autophagy in mice with DM-CIAKI and in HG- and CM-treated HK-2 cells by weakening HDAC9 mRNA stability and reducing HDAC9 expression, thereby alleviating AKI.
PUM2 is a member of the FBF/PUF family, which has been extensively investigated in multiple biological activities including cell viability, apoptosis, oxidative stress and autophagy [22,23,38]. For example, PUM2 participates in cell viability and apoptosis in various types of cancers [38,39]. It was found that PUM2 overexpression improved behavioral and cognitive changes in subarachnoid hemorrhage (SAH) mice by inhibiting SAH-induced oxidative stress and neuronal apoptosis [23]. D’Amico et al. [22] indicated that PUM2 mediated mitophagy in aging. More importantly, PUM2 was reported to affect mitochondrial quality in acute ischemic kidney injury by inhibiting Mff expression [10]. In addition, many studies have demonstrated that promoting autophagy exerts a protective role in contrast-induced AKI [32,40]. For instance, αKlotho protein alleviated contrast-induced AKI by suppressing NLR family pyrin domain containing 3 (NLRP3) inflammasome-mediated pyroptosis and promoting autophagy [32]. However, how PUM2 affects oxidative stress and autophagy in DM-CIAKI has not been elucidated. In this study, we firstly established cell and mouse models of DM-CIAKI and then observed that PUM2 expression was evidently inhibited in DM-CIAKI models. In addition, DM-CIAKI induced oxidative stress and inhibited autophagy and PUM2-KO further strengthened these DM-CIAKI-mediated effects in mice. Moreover, HG and CM treatments elevated oxidative stress and attenuated cell autophagy in HK-2 cells; however, PUM2 overexpression attenuated HG and CM treatments-generated influences the effects of HG and CM treatments on oxidative stress and autophagy. Our findings are the first to elucidate the effects of PUM2 on oxidative stress and autophagy in DM-CIAKI.
It is worth noting that PUM2, an RBP, can bind to the 3′ UTR of a specific target mRNA to block formation of the translation initiation complex, thereby inhibiting the expression of target genes [30,31]. Therefore, in this study, PUM2 was considered a post-transcriptional suppressor. To date, more than 1,000 mRNAs have been identified as harboring PUM2-binding moieties, including mRNAs associated with autophagy, suggesting that PUM2 is a post-transcriptional regulator of these genes [22,41]. As previously documented, Janus kinase 2 (JAK2), Runt-related transcription factor 2 (RUNX2), insulinoma-associated protein 1 (INSM1), and peroxiredoxin 6 (PRDX6) has been identified as target genes of PUM2 [42–44]. Tao et al. [43] revealed that PUM2 facilitated INSM1 mRNA degradation to reduce INSM1 expression, thereby participating in the malignant features of breast cancer. In the current study, we predicted that PUM2 had the binding site on the 3′ UTR of HDAC9 and further validated the interaction between PUM2 and HDAC9. Further experiments showed that PUM2 overexpression accelerated HDAC9 mRNA degradation and reduced HDAC9 expression, whereas PUM2 silencing had the opposite effects. Thus, our results indicated that HDAC9 was a target gene of PUM2.
HDACs has been reported to be implicated in AKI [45]. A review revealed that inhibition of class IIa HDACs (4, 5, 7, and 9) alleviated AKI by inhibiting renal tubular cell apoptosis and promoting autophagy [46]. Wang et al. [47] proposed that HDAC2/4/5 expression was enhanced in the kidneys of STZ-induced rats, diabetic db/db mice and diabetic patients. In addition, HDAC9 upregulation contributed to podocyte injury and glomerulosclerosis in diabetic nephropathy mice [17]. Furthermore, Gene Expression Omnibus (GEO) data revealed that HDAC9 expression was enhanced in renal tissues of patients with diabetic nephropathy [20]. In our study, HDAC9 expression was apparently enhanced in mice with DM-CIAKI and in HG- and CM-treated HK-2 cells. In addition, HDAC9 inhibition and HDAC9 silencing obviously attenuated oxidative stress and promoted autophagy in mouse and cell models of DM-CIAKI. However, HDAC9 overexpression enhanced oxidative stress and cell apoptosis in HG- and CM-treated HK-2 cells. Furthermore, HDAC9 overexpression offset PUM2 overexpression-mediated inhibition of oxidative stress and promotion of autophagy in HG- and CM-treated HK-2 cells. Notably, HDAC9 downregulation-mediated inhibition of oxidative stress and promotion of autophagy were abolished by 3-MA, an inhibitor of autophagy, HG- and CM-treated HK-2 cells. Taken together, HDAC9 participates in DM-CIAKI development.
In conclusion, PUM2 expression was abnormally decreased and HDAC9 expression was abnormally enhanced in cell and mouse models of DM-CIAKI. Furthermore, PUM2 suppressed oxidative stress and promoted autophagy to improve kidney injury caused by DM-CIAKI by attenuating HDAC9 mRNA stability and reducing HDAC9 expression. Our findings may provide new targets for DM-CIAKI.
SUPPLEMENTARY MATERIALS
Supplementary materials related to this article can be found online https://doi.org/10.4093/dmj.2024.0396.
List of targeting sequences
Primer sequences for real-time quantitative reverse transcription polymerase chain reaction
Pumilio RNA binding family member 2 (PUM2) expression was abnormally downregulated in diabetes mellitus (DM)-contrast-induced acute kidney injury (CIAKI) mice while histone deacetylase 9 (HDAC9) expression was abnormally elevated. Mice were treated with streptozotocin or/and iohexol. The detailed groups were control, DM, CIAKI, and DMCIAKI. (A) The ratio of kidney weight/body weight (KW/BW) and the levels of serum creatinine, 24-hour urinary protein, blood urea nitrogen (BUN), fasting blood glucose (FBG), and glycated serum protein (GSP) were detected. (B) The morphology of kidney tissues was examined by hematoxylin and eosin (HE) staining. (C) PUM2 and HDAC9 expression in kidney tissues was measured using real-time quantitative reverse transcription polymerase chain reaction. (D) HDAC9 expression in kidney tissues was detected using immunohistochemistry. aP<0.05, bP<0.01, cP<0.001.
Histone deacetylase 9 (HDAC9) inhibitor suppressed oxidative stress and alleviated the injury in diabetes mellitus (DM)-contrast-induced acute kidney injury (CIAKI) mice. Mice were treated with streptozotocin or/and iohexol and then received HDAC9 inhibitor (HDACi) treatment. The detailed groups were control, DM, CIAKI, DM-CIAKI, and DMCIAKI+ HDACi. (A) HDAC9 expression was detected using real-time quantitative reverse transcription polymerase chain reaction. (B) The ratio of kidney weight/body weight (KW/BW) and the levels of serum creatinine, 24-hour urinary protein, blood urea nitrogen (BUN), fasting blood glucose (FBG), and glycated serum protein (GSP) were detected. (C) The morphology of kidney tissues was examined by hematoxylin and eosin staining. (D) Glutathione peroxidase (GSH-PX), superoxide dismutase (SOD), and malondialdehyde (MDA) in kidney tissues were detected using commercial kits. (E) Reactive oxygen species production was evaluated using 2ʹ,7ʹ-dichlorofluorescein diacetate (DCFH-DA) method. (F) Cell apoptosis in kidney tissues was investigated using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). DHE, dihydroethidium. aP<0.05, bP<0.01, cP<0.001.
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conception or design: W.C., H.L., L.H.
Acquisition, analysis, or interpretation of data: W.D., H.L., Y.C.
Drafting the work or revising: G.L.
Final approval of the manuscript: L.H.
FUNDING
This work was supported by the Natural Science Foundation of China (No.82470759 and 82000697), the Natural Science Foundation of Hunan Province (No.2024JJ3022), the Scientific Research Fund of Hunan Provincial Health Commission (B2023 03056777), the Hunan Provincial Natural Science Foundation for Outstanding Youth (No. 2022JJ10093).
ACKNOWLEDGMENTS
None