MBD2 mediates renal cell apoptosis via activation of Tox4 during rhabdomyolysis‐induced acute kidney injury

Abstract Our study investigated the role of Methyl‐CpG–binding domain protein 2 (MBD2) in RM‐induced acute kidney injury (AKI) both in vitro and in vivo. MBD2 was induced by myoglobin in BUMPT cells and by glycerol in mice. MBD2 inhibition via MBD2 small interfering RNA and MBD2‐knockout (KO) attenuated RM‐induced AKI and renal cell apoptosis. The expression of TOX high mobility group box family member 4 (Tox4) induced by myoglobin was markedly reduced in MBD2‐KO mice. Chromatin immunoprecipitation analysis indicated that MBD2 directly bound to CpG islands in the Tox4 promoter region, thus preventing promoter methylation. Furthermore, siRNA inhibition of Tox4 attenuated myoglobin‐induced apoptosis in BUMPT cells. Finally, MBD2‐KO mice exhibited glycerol‐induced renal cell apoptosis by inactivation of Tox4. Altogether, our results suggested that MBD2 plays a role in RM‐induced AKI via the activation of Tox4 and represents a potential target for treatment of RM‐associated AKI.


| INTRODUC TI ON
Rhabdomyolysis (RM) is a life-threatening condition, characterized by skeletal muscle tissue damage and leakage of intracellular components into the bloodstream. Traumatic RM occurs in the case of crush syndrome, 1,2 which is a medical condition that occurs due to crushing skeletal injury in victims of natural disasters (eg earthquakes), accidents and wars. 3,4 One of the most common complications of RM is acute kidney injury (AKI), and 10% of all AKI patients present with RM. 5 AKI is a clinical disorder with high mortality rates. 6 It is known that the renal toxicity of myoglobin plays a crucial role in RM-induced AKI via increased oxidative stress, cast formation, inflammation and apoptosis. 7 Although numerous studies have focused on RM-induced AKI, 7-10 the underlying mechanisms remain largely unknown, thus resulting in non-specific therapy.
Previous research has shown that DNA methylation occurs in regions containing a high density of CpG dinucleotides, which is typically thought to result in epigenetic modification related to transcriptional repression; however, the role and traditional interpretation of DNA methylation has been recently challenged. 11 Based on different methylation patterns, it has been reported that DNA methylation may play a role in both transcriptional activation and silencing. Furthermore, accumulating evidence suggests that DNA methylation is associated with AKI. 5,12,13 However, the role of DNA methylation in RM-induced AKI may vary based on differential regulation of methylation and the type of experimental model used. In mice that are injected with glycerol in their bilateral hindlimb muscles, 14 RM-induced AKI is characterized by renal cortical necrosis, cast formation and inflammatory infiltration. 15,16 Apoptosis is a key phenomenon in glycerol-induced AKI, which may, in turn, be regulated by other cellular pathways. [16][17][18] Therefore, it is important to gain a holistic understanding of the mechanisms underlying RM-induced AKI to better design targeted treatment therapies.
Here, we aimed to study the role of DNA methylation in AKI by analysing the function of Methyl-CpG-binding domain protein 2 (MBD2), a key methylated protein reader, in RM-induced AKI, both in vitro using a mouse proximal tubule-derived cell line (BUMPT) and in vivo using C57BL/6 mice.

| Reagents and antibodies
Antibodies were purchased from the following sources: MBD2 from

| Animal model of RM-induced AKI
MBD2 knockout (MBD2-KO) mice were obtained from Cyagen Biosciences Co., Ltd (Guangzhou, China). Male MBD2-KO mice aged 10-12 weeks were injected in the bilateral hindlimb skeletal muscles with a dose of 8.0 mL/kg glycerol (50% v/v in sterile saline); the control group was injected with an equal volume of normal saline. Littermate male MBD2-wild-type (WT) mice were treated using the same method mentioned above. Animal experimental protocols were approved by the Care and Use of Laboratory Animals Institutional Committee from Second Xiangya Hospital, China. The mice were housed at stable room temperature in a 12hour light/dark cycle and provided adequate supplies of standard rodent chow and water.

| Cell culture and treatments
BUMPT cells were cultured in Dulbecco's modified Eagle's medium with 10% foetal bovine serum, 0.5% penicillin (Thermo Fisher Scientific), and streptomycin, and then maintained in a 5%-CO2 incubator at 37°C. Both myoglobin and ascorbic acid were then added to the medium at a final concentration of 200 mM (3.6 mg/mL) and 2 mM, respectively. Ascorbic acid reduces myoglobin to a ferrous status, 29

| Methylated CpG-DNA immunoprecipitation
Immunoprecipitation of methylated CpG-DNA was performed, as described previously (Zymo Research, Irvine, CA, USA). 32 Briefly, sheared DNA was used for methylated CpG immunoprecipitation, and the methylated DNA was analysed by PCR analysis using an ABI Onestepplus real-time PCR system.

| Histology, immunohistochemistry and immunoblot analyses
Haematoxylin and eosin (H&E) staining was used for histological analysis. The tubular damage scores were assessed according to the percentage of damaged tubules, as described previously. 26 TUNEL assay was used for detecting renal cell apoptosis, and the percentage of the total number of TUNEL-positive cells were calculated in 10-20 microscopic fields that were randomly selected per tissue section. 33 Immunohistochemical staining and image analysis of MBD2 were performed according to previously described methods. 34 Lysates of kidney tissues and BUMPT cells were subjected to sodium dodecyl sulphate-polyacrylamide gel electrophoresis and immunoblotting was performed using MBD2, Tox4, Caspase-3, cleaved caspase-3 and β-Tubulin antibodies following standard procedures.

| Statistical analysis
Data were expressed as mean ± standard deviation (SD). Two-group comparisons were made using 2-tailed Student t tests. Multiple group data were evaluated using one-way analysis of variance. P < .05 was considered statistically significant.

| MBD2 expression was induced by RM in BUMPT cells and C57BL/6 mice
Using Western blotting, we investigated the expression of MBD2

| Renal dysfunction, tubular damage and cell apoptosis are attenuated in MBD2-KO mice during RM-induced AKI
We injected the MBD2-KO and WT littermate mice with glycerol and analysed the effects after 24 hours. Glycerol injections induced renal failure in WT mice, as indicated by the increased levels of blood urea nitrogen (BUN) and creatinine compared with control. However, in MBD2-KO mice, the increase in BUN and creatinine levels was lower compared with that in MBD2-WT mice (Figure 2A,B). H&E staining also confirmed that the deletion of MBD2 markedly reduced tubular damage in the cortex of the kidney in glycerol-injected mice ( Figure 2C). Additionally, glycerolinjected MBD2-KO mice showed a lower tubular damage score of 1.9 compared with glycerol-injected WT mice that showed a score of 3.5 ( Figure 2D). Previous results have indicated that apoptosis plays a key role in the progression of RM-induced AKI. 16,17,[19][20][21] To investigate whether MBD2 promoted renal cell apoptosis and, therefore, tubular cell damage, we performed TUNEL staining to assess apoptosis in WT and MBD2-KO renal tissues. We ob-  Figure 2G), which was also supported by densitometric quantitation ( Figure 2H).

| MBD2 mediated myoglobin-induced apoptosis in BUMPT cells
As MBD2 expression was up-regulated as a result of myoglobin treatment in BUMPT cells ( Figure 1A,B), we investigated whether MBD2 played a role in renal cell apoptosis in vitro. BUMPT cells were first transfected with MBD2 siRNA or MBD2-containing plasmid, followed by myoglobin or saline treatment. Flow cytometry analysis demonstrated that the transfection of MBD2 siRNA reduced apoptosis in BUMPT cells induced by myoglobin ( Figure 3A).
Immunoblotting analysis showed that MBD2 siRNA markedly down-regulated the expression levels of MBD2 and cleaved caspase-3 ( Figure 3B), which was confirmed by densitometric analysis ( Figure 3C). In contrast, myoglobin-treated BUMPT cells containing the MBD2 plasmid showed increased levels of cleaved caspase-3 compared with those cells that lacked the plasmid ( Figure 3D), indicating that apoptosis was enhanced in myoglobin-treated cells that ectopically expressed MBD2. This result was further validated by immunoblotting and densitometric analyses ( Figure 3E,F).

| Tox4 is down-regulated in MBD2-KO mice and is involved in renal cell apoptosis
Tox4 is novel gene that is reportedly involved in DNA reprogramming,

| MBD2 up-regulates Tox4 expression by inhibiting methylation of the Tox4 promoter
We next investigated whether MBD2 regulated Tox4 expression via DNA methylation. Tox4 has been predicted by MethPrimer to contain a CpG island in its promoter region (http://www.uroge ne.org/cgi-bin/ methp rimer/ MethP rimer.cgi), and five pairs of PCR primers for this region have been designed by this website ( Figure 5A). ChIP assays showed two binding sites for MBD2 in the Tox4 promoter region-a 160-bp fragment (mBS1) and a 161-bp fragment (mBS2) ( Figure 5B), which validated the prediction that MBD2 could directly bind to the Tox4 promoter region. The Tox4 promoter region containing the DNA methylation target sequences was cloned into a CpG-free pCpGI luciferase reporter plasmid and co-transfected with plasmids containing MBD2 or mutated MBD2 that lacks the DNA-binding domain (mtMBD2) into BUMPT cells. We found that the MBD2 plasmid markedly activated the transcription level of Tox4 compared with the plasmid containing mtMBD2 and the control plasmid ( Figure 5C).

Methylation analysis indicated that endogenous MBD2-bound DNA
markedly suppressed the methylated Tox4 pCpGI; meanwhile, this suppression impact was significantly increased by ectopic MBD2 ( Figure 5D). Immunoblot analysis showed that Tox4 expression was up-regulated by the addition of myoglobin, which was further en-

| D ISCUSS I ON
RM is often accompanied by AKI caused by crushing trauma. 5,25 However, due to a lack of knowledge of the underlying mechanisms, no specific therapies for RM-induced AKI currently exist, thus making it a life-threatening medical condition. We believe that the present study is the first to show that MBD2 induces renal cell apoptosis in RM-induced AKI by up-regulating Tox4 expression via promoter demethylation.
We have previously shown that MBD2 mediates apoptosis during vancomycin-associated AKI. 26 In this study, we observed that MBD2 is activated by glycerol in mice and by myoglobin in BUMPT cells, and its expression is largely localized in the nucleus of injured tubular cells ( Figure 1). Our results also showed that the deletion of MBD2 could attenuate renal injury and dysfunction in RM-induced AKI mice. Previous results have indicated that apoptosis plays a key role in the progression of RM-induced AKI. 16,17,[19][20][21]27 However, few researches have revealed the connection and the underlying mechanism between apoptosis and DNA methylation. 28  Thus, inhibition of Tox4 ameliorates myoglobin-induced renal cell apoptosis suggesting that Tox4 may be a potential therapeutic target for RM-associated AKI. We also found that the expression of Tox4 was up-regulated following transfection of the MBD2 plasmid, whereas it was decreased in response to the MBD2 siRNA treatment ( Figure 5E-H). The ChIP assays demonstrated that MBD2 could directly interact with a binding site within the Tox4 promoter domain it ( Figure 5B). Moreover, observed that MBD2 activated Tox4 via demethylation ( Figure 5C,D). These results confirmed our hypothesis that Tox4 is a direct downstream target gene of MBD2. In recent studies, Tox4 was identified as a novel transcriptional modulator, 22 but the impact of Tox4 on apoptosis in this experimental model remained poorly understood. In order to explore the regulatory mechanism of Tox4 in RM-associated AKI, we transfected Tox4 siRNA synthesized by the Ruibo RUIBO Biology company Co. in BUMPT cells. As shown in Figure 6, our results verified that Tox4 promoter had a direct interaction with MBD2 in this experimental model, which up-regulated TOX4 transcription level and its protein expression. However, we have not found the interactions between Tox4 protein and other apoptotic proteins or DNA fragments, which need to be further explored in the future work.
In conclusion, our results indicated that MBD2 mediates apoptosis during RM-associated AKI. In addition, we found that MBD2

ACK N OWLED G EM ENTS
This study was supported by National Natural Science Foundation of China (No. 81472058), the Hunan Science and Technology Innovation Plan (2018SK2105).

CO N FLI C T O F I NTE R E S T
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

O RCI D
Youwen Deng https://orcid.org/0000-0002-9117-4078 F I G U R E 6 The molecular mechanism of MBD2 in rhabdomyolysis (RM)associated acute kidney injury (AKI). When RM occurs, myoglobin is released into the blood, thus inducing nephrotoxicity and, often, AKI. Subsequently, MBD2 activates the expression of Tox4 via demethylation of its promoter, which mediates renal cell apoptosis [Colour figure can be viewed at wileyonlinelibrary.com]