MiR-542-5p regulates the progression of diabetic retinopathy by targeting CARM1

Diabetic retinopathy (DR), as the most frequent microvascular complication of diabetes mellitus (DM), causes vision loss and blindness in adults worldwide with increasing incidence. MicroRNAs (miRNAs) are involved in the regulation of DR. However, the role of miR-542-5p is still unknown. Here, we demonstrate that miR-542-5p is down-regulated in patients with DR and in high-glucose (HG)-treated retinal pigment epithelial cells. Moreover, miR-542-5p overexpression inhibits apoptosis in retinal pigment epithelial cells exposed to HG. The interaction between miR-542-5p and co-activator-associated arginine methyltransferase 1 (CARM1) is confirmed. MiR-5425p mimics decrease the CARM1 level and miR-542-5p inhibitor increases the CARM1 level. Additionally, CARM1 overexpression promotes the miR-542-5p-mediated apoptosis in HG-treated retinal pigment epithelial cells. In summary, the data suggest that miR-542-5p may suppress apoptosis in retinal pigment epithelial cells via targeting CARM1, which provides a new therapeutic target for the treatment of patients with DR.


INTRODUCTION
Diabetes is considered the most common metabolic disease in the world with increasing mortality (Ravindran et al., 2015;Wild et al., 2004). Diabetic retinopathy (DR), as the most frequent microvascular complication of diabetes mellitus (DM), is characterized by hemorrhages, lipid exudates and neovascularization (Shao & Yao, 2016;Barot et al., 2013). DR mainly causes vision loss and blindness in adults worldwide (Chong et al., 2017;Klein, 2007;Cheung et al., 2010). The incidence of DR has been increasing annually worldwide (Wang et al., 2012;Klein et al., 2017). Laser photocoagulation treatment shows no significant improvement in visual acuity in the long term . Therefore, it is necessary to further investigate more efficient therapeutic strategies to improve the treatment of patients with DR.
MicroRNAs (miRNAs) represent a group of highly conserved short (19-25 nucleotides) non-coding RNAs that bind to the 3'-untranslated region (UTR) of mRNAs to suppress gene expression in various cell processes (Li & Liu, 2011;Bartel, 2004;Hwang & Mendell, 2007). Accumulating evidence has shown that miRNAs are involved in the regulation of DR (Hagiwara et al., 2013, Wu et al., 2014. For example, miR-590-3p has been found to elevate cell viability and repress pyroptosis in DR (Gu et al., 2019). Moreover, miR-203-3p represses oxygen-induced retinopathy-induced retinal angiogenesis and improves proliferative diabetic retinopathy (Han et al., 2020). A previous study also shows that miR-542-5p plays a suppressive role in neuroblastoma (Bray et al., 2011). Zhu and others (Zhu et al., 2020) found that miR-542-5p mediates osteosarcoma tumorigenesis and enhances proliferation in osteosarcoma. These findings indicate that miR-542-5p may take part in the development of various diseases. Interestingly, Wu and others (Wu et al., 2012) observed that miR-542-5p is abnormally expressed in DR. Therefore, we hypothesize that miR-542-5p may play an important role in the development of DR.
Co-activator-associated arginine methyltransferase 1 (CARM1) is a coactivator of transcriptional activation and forms homodimers, which is necessary for catalytic activity (Chen et al. 1999;Higashimoto et al., 2007). Recently, CARM1 was found to be increased in patients with diabetes (Porta et al., 2019). Additionally, CARM1 is elevated and participates in the regulation of high-glucose (HG) treated human retinal pigment epithelial cells (Kim et al., 2014). These results suggest that CARM1 may be closely associated with the development of DR.
Here, we explore the role of miR-542-5p and CARM1 in HG-treated human retinal pigment epithelial cells.
Our study suggests that miR-542-5p may be involved in the development of DR by targeting CARM1. in DMEM containing 10% fetal bovine serum at 37°C under 5% CO 2 . The cells were treated with 5 mM glucose (Control) or 25 mM glucose (HG) for 24 h. Additionally, the cells were transfected with miR-542-5p mimics or negative control mimics (NC mimics) for 24 h and then exposed to HG for another 24 h. Quantitative real-time PCR (qRT-PCR). Total RNA was extracted using TRIzol reagent (Takara, Dalian, China). Then, cDNAs were synthesized through Taqman MicroRNA assays (Beyotime, Shanghai, China). The qRT-PCR was performed, and miR-542-5p and CARM1 levels were determined via SYBR Green (Takara, Dalian, China). The relative expression of miR-542-5p and CARM1 was calculated via the 2 −ΔΔCT method. The primers were as follows:
Statistical analysis. Data were presented as mean ± S.D. GraphPad Prism 6.0 was used to carry out statistical analysis. Statistical significance was assessed using Student's t-test and one-way ANOVA followed by the Bonferroni test. The correlation between the two groups was confirmed through Pearson's correlation analysis. P<0.05 was considered statistically significant.

The effect of HG treatment on apoptosis of retinal pigment epithelial cells
We first measured the level of miR-542-5p in patients with DR and HG-treated retinal pigment epithelial cells. qRT-PCR showed that the miR-542-5p level is lower in diabetes mellitus (DM)+non-diabetic retinopathy (NDR) group than in the Healthy group, and miR-542-5p is reduced in patients with DM and DR compared with patients with DM. Moreover, the miR-542-5p in DM+NDR group is reduced compared with that in the Healthy group (Fig. 1A). The results from qPCR assay also show that miR-542-5p is reduced in HG-treated ARPE-19 cells compared with control (Fig. 1B). Cell apoptosis analysis reveals that apoptosis is enhanced in HG-treated ARPE-19 cells (Fig. 1C). Western blot analysis indicates that HG treatment causes the elevation of Cleaved caspase 3 protein level and the reduction of Bcl-2 protein level in ARPE-19 cells (Fig. 1D). These findings imply that miR-542-5p might be related with apoptosis in HG-treated retinal pigment epithelial cells.

DISCUSSION
Here, we demonstrate that miR-542-5p is reduced in patients with DR and in HG-treated retinal pigment epithelial cells. Functional analysis indicated that miR-542-5p represses apoptosis in retinal pigment epithelial cells in the presence of HG. Moreover, the interaction between miR-542-5p and CARM1 was identified. Further, CARM1 overexpression promotes miR-542-5p-mediated apoptosis in retinal pigment epithelial cells exposed to HG. The data imply that miR-542-5p may function in the apoptosis of retinal pigment epithelial cells, which contributes to the treatment of patients with DR.
Increasing evidence has revealed that abnormal expression of miRNAs has a close association with DR. Some miRNAs are upregulated. For example, miR-218 expression is increased, and it represses proliferation as well as stimulates apoptosis in human retinal pigment epithelium cells (Yao et al., 2019). MiR-29b-3p is elevated in patients with DR, and enhances the ratio of Bax/Bcl-2 in human retinal microvascular endothelial cells (Zeng et al., 2019). Conversely, some miRNAs have the opposite role. For instance, miR-7 has been demonstrated to inhibit apop- tosis and alleviate HG-induced damage in retinal pigment epithelial cells (Yang et al., 2019). MiR-142-5p is decreased in retinal tissues of DR rats and HG-treated human retinal endothelial cells, and improves pathological effects in retinal tissues (Liu et al., 2020). Like the later studies, our research verifies that miR-542-5p is suppressed in patients with DR and in HG exposed ARPE-19 cells. Moreover, miR-542-5p represses HGinduced apoptosis in ARPE-19 cells, accompanied by the reduction of the Cleaved caspase 3 protein level and elevation of the Bcl-2 protein level. These findings imply that miR-542-5p may be closely connected with the development of DR.
Previously, miRNAs were found to regulate the development of DR through targeting the 3'-UTR of mRNAs. For example, Fu and Ou (Fu & Ou, 2020) found that miR-152 is reduced and represses HG-induced angiogenesis in human retinal endothelial cells through targeting LIN28B. Chen and others (Chen et al., 2019 discovered that miR-126 is decreased, and stimulates proliferation as well as suppresses apoptosis in HG-induced human retinal endothelial cells via targeting interleukin-17A). Interestingly, our study shows that miR-542-5p could target CARM1. Additionally, miR-542-5p down-regulates the CARM1 expression level. More importantly, accumulating evidence has shown that the function of miR-542-5p may vary in different diseases. For example, miR-542-5p has been found to alleviate fibroblast activation, and suppress proliferation and migration through targeting integrin α6 (Yuan et al., 2018). Conversely, Cheng and others (Cheng et al., 2015) have reported that miR-542-5p stimulates proliferation in osteosarcoma through targeting HUWE1. These findings suggest that miR-542-5p may regulate development of some diseases via targeting the target gene. Further analysis in this study proves that CARM1 overexpression increases the miR-542-5pmediated apoptosis in retinal pigment epithelial cells treated with HG. Notably, CARM1 was reported to be enhanced in HG-treated retinal pigment epithelial cells, and to promote apoptosis in retinal pigment epithelial cells (Kim et al., 2014). Moreover, Porta and others (Porta et al., 2019) demonstrated that the CARM1 expression level is high in type 2 diabetes. Our study is consistent with previous research. The data suggest that miR-542-5p may take part in the inhibition of apoptosis in retinal pigment epithelial cells through targeting CARM1.
However, other miRNAs may also participate in regulating the progression of DR by targeting CARM1. Moreover, the clinical application of miR-542-5p was not evaluated, and the effect of miR-542-5p on apoptosis in patients with DR by targeting CARM1 was not explored in our study. Therefore, to further investigate the mechanism underlying miR-542-5p in patients with DR, more experiments will be carried out in the future.
In summary, the current study shows that miR-542-5p is down-regulated and CARM1 is up-regulated in HGtreated retinal pigment epithelial cells and in patients with DR. MiR-542-5p participates in the progression of DR via targeting CARM1 (Fig. 5), which could be useful for the treatment of patients with DR.