Astragalus polysaccharide improve the proliferation and insulin secretion of mouse pancreatic β cells induced by high glucose and palmitic acid partially through promoting miR-136-5p and miR-149-5p expression

ABSTRACT The current drugs for the treatment of type 2 diabetes mellitus (T2DM) can cause side effects after long-time use. Hence, the novel drugs were urgent need to developed for T2DM patients. In this study, the effect of astragalus polysaccharide on dysfunctional insulin cells was investigated to clarify whether astragalus polysaccharide could be a novel drug for T2DM treatment. MIN6 cells (mouse pancreatic β-cell line) were treated with high glucose (HG)+ palmitic acid (PA) and then treated with astragalus polysaccharide. The proliferation, apoptosis, and insulin secretion were measured using CCK8, flow cytometry, and ELISA, respectively. Pancreatic and duodenal homeobox 1 (PDX1), miR-136-5p, and miR-149-5p expression levels were measured by RT-qPCR. The combination of EF-hand domain family member 2 (EFHD2) and miR-136-5p or miR-149-5p was analyzed by luciferase reporter assay. EFHD2 protein level was measured by western blot. We found that HG+PA treatment reduced MIN6 cell viability, insulin secretion, and PDX1 expression and promoted MIN6 cell apoptosis. Astragalus polysaccharide treatment reversed the effect of HG+PA on MIN6 cells. Additionally, astragalus polysaccharide treatment promoted miR-136-5p and miR-149-5p expression. Silencing of miR-136-5p and miR-149-5p expression partially reversed the therapeutic effects of astragalus polysaccharide. Furthermore, EFHD2 was the target of miR-136-5p and miR-149-5p. Meanwhile, astragalus polysaccharide treatment inhibited EFHD2 protein level in HG+PA treated MIN6 cell. Finally, EFHD2 overexpression partially reversed the therapeutic effects of astragalus polysaccharide. In conclusion, astragalus polysaccharide treatment improved proliferation and insulin secretion in HG+PA-treated MIN6 cells partially by promoting miR-136-5p and miR-149-5p expression to inhibit EFHD2 expression.


Introduction
Type 2 diabetes mellitus (T2DM) is a multifactorial chronic disorder and accounts for the most prevalent type of patients with diabetes, approximately 90-95% [1,2]. Long-term hyperglycemia and highsaturated free fatty acids are the key contributors which inhibited insulin secretion, promoted pancreatic β-cell apoptosis, and decrease the number of normal functioning pancreatic β cells [3][4][5]. Therefore, maintaining pancreatic β-cell survival and improving functional pancreatic β-cells are the main therapeutic strategies for the clinical management of T2DM [6]. Various therapeutic agents, including sulfonylureas, α-glucosidase inhibitors, and biguanides have been used to treat T2DM, which can induce many side effects after long-term use [7]. Therefore, the novel drugs that are effective, nontoxic, and have a lighter burden for T2DM patients were urgent need to developed.
Previous studies have shown that biological macromolecules derived from natural products, mainly polysaccharides, are less toxic than chemically synthesized drugs and possess prominent efficacies on DM [8]. The dry root of Astragalus membranaceus (Chinese: Huang Qi) has been used as traditional Chinese medicine and health-care food [9]. Astragalus polysaccharide is the central active component which extracted from the stems or dried roots of Astragalus membranaceus. Previous studies have demonstrated that astragalus polysaccharide has anti-oxidative, anti-aging, anti-hypertensive, anti-tumor etc. pharmacological effects [10]. Moreover, it reduces blood sugar levels. Majorly, astragalus polysaccharide can effectively alleviate diabetes and diabetic complications such as diabetic foot, diabetic neuropathy, and diabetic cardiomyopathy [11]. However, the effect of astragalus polysaccharide on dysfunctional pancreatic β-cells and the underlying molecular mechanisms remain unclear.
In this study, we hypothesize that astragalus polysaccharide can improve insulin secretion in dysfunctional pancreatic β cells through miRNAs. First, MIN6 cells were induced by high glucose (HG) and palmitic acid (PA) to mimic glucolipotoxicity and establish the T2DM cell model [15]. Additionally, the effect of astragalus polysaccharide on HG+PA induced cells was investigated. Finally, the differential expression of miRNAs in T2DM and its role in astragalus polysaccharide treatment of MIN6 cells were studied. This study clarified the effect and mechanism of astragalus polysaccharide on glucolipotoxicity-induced mouse β-cell, which provides a theoretical basis for the clinical application of astragalus polysaccharide.

Glucose-stimulated insulin secretion (GSIS)
The assessment of GSIS was carried out as mentioned in previous studies [17]. Briefly, MIN6 cells were incubated in HEPES-balanced Krebs-Ringer bicarbonate buffer with 0.2% BSA (pH 7.4) for 2 h, then stimulated with low glucose (2.5 mmol/L) or high glucose (20 mmol/L) for 1 h. Insulin in the media was then quantified using a mouse insulin enzymelinked immunosorbent assay (ELISA) Detection Kit (Sinobestbio, Shanghai, China). The insulin levels were then normalized based on the cell number.

Bioinformatics analysis and luciferase reporter assay
The potential target genes of miR-136-5p and miR-149-5p were analyzed using Starbase 3.0 [19]. Additionally, GSE20966 were chosen on Gene Expression Omnibus repository (GEO, https://www. ncbi.nlm.nih.gov/geo/) to analyzed the differentexpressed mRNA in β cell between non-diabetic condition and diabetic condition [20]. GSE20966 dataset included 10 β cells from the pancreatic tissue sections of type 2 diabetic subjects (T2DM group) and 10 β cells from the pancreatic tissue sections of type 2 diabetic subjects (Normal group). Then, the differentially expressed mRNAs (DEmRNAs) were analyzed by GEO2R. The selection condition of DEmRNA is Padj<0.05 (T2DM group vs control group). In addition, wild-type EFHD2 (WT EFHD2) and mutated EFHD2 (Mut EFHD2) were inserted into psi-check2, which was co-transfected with miR-136-5p, miR-149-5p mimic, or NC mimic into 293 T using Lipofectamine 3000. At 24 h after co-transfection, luciferase assays were performed using the dualluciferase reporter assay system (Promega), and the renilla/firefly luciferase activity ratio was calculated.

Statistical analysis
All experiment was performed in triplicate. Normally distributed data is expressed as mean ± standard deviation and analyzed by the SPSS software (version 19.0; IBM, Chicago, IL, USA). The differences between the three groups were compared using one-way analysis of variance followed by Tukey's post-hoc test. Statistical significance was set at P < 0.05.

ASP improves survival and insulin secretion in HG+PA-treated MIN6 cells
To study the effect of astragalus polysaccharide on HG+PA-treated MIN6 cells, astragalus polysaccharide was added to treat HG+PA-treated MIN6 cells. The molecular structure of astragalus polysaccharide is shown in Figure 1(a). CCK8 results revealed that astragalus polysaccharide did not significantly inhibit cell viability when the astragalus polysaccharide concentration was less than 200 μg/mL (Figure 1(b-d)). When the astragalus polysaccharide concentration was more than 400 μg/mL, MIN6 cell viability was remarkably reduced after astragalus polysaccharide treatment for 24, 48, and 72 h (Figure 1(b-d)). Hence, 50, 100, and 200 μg/mL astragalus polysaccharide were selected as the low-, medium-, and highdose groups for further study. MIN6 cell viability, insulin secretion, and PDX1 mRNA expression in the model group (HG+PA-treated) were remarkably decreased compared with the blank group (Figure 2(a-c)). After treatment, MIN6 cell viability, insulin secretion, and PDX1 mRNA expression in the 50, 100, and 200 μg/ mL astragalus polysaccharide and metformin treatment groups were remarkably higher than those in the model group (Figure 2(a-c)). Compared with the 50 μg/mL astragalus polysaccharide treatment group, MIN6 cell viability, insulin secretion, and PDX1 mRNA expression in the 100 μg/mL and 200 μg/mL astragalus polysaccharide and metformin treatment groups significantly enhanced ( Figure  2(a-c)). Additionally, MIN6 cell viability, insulin secretion, and PDX1 mRNA expression were not significantly different between the 100 μg/mL and 200 μg/mL astragalus polysaccharide and metformin treatment groups (Figure 2(a-c)). MIN6 cell apoptosis in the model group (HG+PAtreated) was remarkably more than that in the blank group, whereas MIN6 cell apoptosis was s remarkably reversed after 200 μg/mL astragalus polysaccharide treatment (Figure 2(d-e)).

Analysis of potential target genes of miR-136-5p and miR-149-5p
To further understand the downstream genes regulated by mmu-miR-136-5p and mmu-miR -149-5p, this study used Starbase 3.0 to predict the target genes of mmu-miR-136-5p and mmu-miR-149-5p, and then the GSE20966 dataset was used to analyze the DEmRNAs in the pancreatic β cells of patients with T2DM, and the two groups were crossed to obtain the potential genes regulated by mmu-miR-136-5p and mmu-miR-149-5p. From the predictions of Starbase 3.0 and the results of GSE20966, it was only found one potential target gene of miR-136-5p and miR-149-5p ( Figure 5(a)). The heat map of GSE20966 was shown in Figure 5(b). GSE20966 dataset found that 19 DEmRNAs were upregulated and 19 DEmRNAs were downregulated. Considering that miR-136-5p and miR-149-5p expression are down-regulated in HG+PA-induced MIN6 cells, this study only selected up-regulated DEmRNAs for follow-up studies. The result of GSE20966 showed that EFHD2 expression in the β cells of patient with diabetes mellitus was significantly higher than that in on-diabetes mellitus ( Figure 5  (c)). In addition, the binding site between miR-136-5p, miR-149-5p and 3′-UTR was shown in Figure 5(d). Moreover, compared with NC mimic + WT EFHD2 group, the renilla/firefly ratio was significantly inhibited in miR-136-5p/miR-149-5p mimic + WT EFHD2 group, whereas the renilla/ firefly ratio had no significantly changed between NC mimic + Mut EFHD2 group and miR-136-5p/ miR-149-5p mimic + Mut EFHD2 group, suggesting that miR-136-5p and miR-149-5p can bind with the 3′-UTR of EFHD2 ( Figure 5(d)). Compared with blank group, EFHD2 expression was significantly promoted in model group. Whereas, its expression was significantly inhibited in HG+PA-treated MIN6 cells after astragalus polysaccharide and metformin treatment ( Figure 5(e)).

EFHD2 overexpression partially reverses the effect of astragalus polysaccharide on the HG +PA-treated MIN6 cells
To understand whether EFHD2 can participate in astragalus polysaccharide improving the insulin secretion of HG+PA-treated MIN6 cells, MIN6 cells were transfected with EFHD2-OV and then cotreated with HG+PA and 200 μg/mL astragalus polysaccharide. After transfection for 24 h, EFHD2 protein level in EFHD2-OV groups was remarkably more than that in the NC-OV group (Figure 6(a)). Additionally, compared with the NC-OV group, MIN6 cell viability significantly reduced in EFHD2-OV groups after transfection at 24, 48, and 72 h ( Figure 6(b)). And insulin secretion and PDX1 mRNA expression in MIN6 cells significantly reduced, whereas apoptosis increased in EFHD2-OV groups after transfection at 48 h ( Figure 6(c-e)).

Discussion
The prevalence of T2DM is steadily increasing worldwide. This problem is compounded by the fact that current treatment drugs can lead to side effects after long-time use. Therefore, the novel, effective, and nontoxic drugs were urgent need to developed. In this study, we found that HG+PA treatment reduced insulin secretion and promoted MIN6 cell apoptosis. In contrast, the 200 μg/mL astragalus polysaccharide treatment reversed the effect of HG+PA treatment on MIN6 cells, playing a protective role. Astragalus polysaccharide may be a potential drug for the treatment of T2DM. Astragalus polysaccharide is a monomer component extracted from the TCM of Huangqi (Radix astragali mongolici), which could improve plasma glucose, insulin resistance and secretion, and the complication of diabetes mellitus with promising effects in recent years [11]. However, the mechanism of action of astragalus polysaccharide on insulin cell apoptosis and secretion is still not well understood. With an increase in blood glucose levels and the continuous stimulation of hyperglycemia, the function and quality of pancreatic β cells reduced progressively, giving rise to abnormal insulin secretion in T2DM [23]. In this study, HG+PA was used to treat MIN6 cells to simulate the effect of the T2DM microenvironment on pancreatic β cells, and it was found that HG+PA treatment reduced MIN6 cell viability, insulin secretion, and PDX1 mRNA expression and promoted MIN6 cell apoptosis. PDX1 is a key transcriptional factor The heat map about the different-expressed mRNAs in β cell between non-diabetic condition and diabetic condition which analyzed by GEO2R in the GSE20966 dataset. Red: overexpression; Green: significant downexpression. (c) The EFHD2 expression was analyzed in GSE20966 dataset. (d) the bind site between miR-136-5p, miR-149-5p and 3′-UTR. And this bind was verified by luciferase reporter assay. The renilla/firefly ratio was significantly lower in miR-136-5p/miR-149-5p mimic + WT EFHD2 group than that in NC mimic + WT EFHD2 group, whereas the renilla/firefly ratio had no significantly changed between NC mimic + Mut EFHD2 group and miR-136-5p/miR-149-5p mimic + Mut EFHD2 group, suggesting that miR-136-5p and miR-149-5p can bind with the 3′-UTR of EFHD2. (e) EFHD2 expression was measured by western blot. that maintains pancreatic β-cell survival and function, which is important for maintaining the stability of cell function and inhibiting the occurrence and development of diabetes [24]. Additionally, astragalus polysaccharide treatment, especially the 200 μg/mL astragalus polysaccharide treatment, reversed the effect of HG+PA treatment on MIN6 cells. There was no significant difference between the therapeutic effect of 200 μg/mL astragalus polysaccharide and the therapeutic effect of metformin. These results suggest that astragalus polysaccharide treatment attenuates HG+PA-induced apoptosis and abnormal insulin secretion.
Well-recognized mediators that maintain pancreatic β-cell survival and function, miRNAs, can become novel targets for the treatment of T2DM as well as potential diagnostic and prognostic markers [25,26]. In this study, astragalus polysaccharide treatment promoted miR-136-5p and miR-149-5p expression, reversed the effect of HG+PA treatment on miR-136-5p and miR-149-5p expression in MIN6 cells. miR-136-5p expression was inhibited in HG, PA, or inflammation-treated human β cell, which suggests that miR-136-5p can respond to high glucose, dyslipidemia, and inflammatory diabetic microenvironment [27]. Additionally, miR-136-5p improves blood glucose, urine protein concentrations, and renal fibrosis in diabetic rats [28]. A previous study found that miR-149-5p expression were remarkably decreased in interleukin-1β-and interferon-γ-treated EndoC-βH1 cells, which could enhance the expression of death protein 5, p53 up-regulated modulator of apoptosis, and Bcl2 associated X protein to increase apoptosis [29]. In human umbilical vein endothelial cells (HUVECs), miR-149-5p expression reduced after high glucose treatment [30]. Expression of miR-149-5p reduced in diabetic nephropathy, which promotes mesangial cell proliferation and fibrosis in mice [31]. miR-149-5p expression also reduced in high glucosetreated MIN6 cells, and the miR-149-5p mimic protected cell survival and insulin secretion while reduced cell apoptosis by inhibiting the expression of Bcl2-like 11 [6]. These studies suggest that miR-136-5p and miR-149-5p play the protective role in patients with diabetes and diabetic syndromes. Silencing miR-136-5p and miR-149-5p expression partially reversed the improve effect of astragalus polysaccharide. Similar with previous studies, the result suggested that miR-136-5p and miR-149-5p play the protective role in HG +PA-treated MIN6 cells. The result also suggested that astragalus polysaccharide treatment attenuates HG +PA-induced apoptosis and abnormal insulin secretion partially by promoting miR-136-5p and miR-149-5p expression.
EFHD2, also known as swiprosin-1, was first found in CD8+ lymphocytes [32]. Previous studies found that EFHD2 promote cell apoptosis in murine B cells and non-small cell lung cancer cells [33,34]. EFHD2 overexpression enhanced mitochondria-dependent apoptosis of glomerular podocytes while silenced EFHD2 reversed this effect in the early stage of mouse diabetic nephropathy [35,36]. In this study, astragalus polysaccharide treatment inhibited EFHD2 protein level, reversed the effect of HG+PA treatment on EFHD2 protein level in MIN6 cells. In addition, EFHD2 was the target gene of miR-136-5p and miR-149-5p in MIN6 cells. Finally, EFHD2 overexpression partially reversed the effect of astragalus polysaccharide in HG+PA-induced MIN6 cells. There results showed that astragalus polysaccharide treatment improved proliferation and insulin secretion in HG+PA-treated MIN6 cells partially by promoting miR-136-5p and miR-149-5p expression to inhibit EFHD2 expression.
However, there are certain limitations to the present study. The effect of astragalus polysaccharide treatment on pancreatic β-cells has not been verified in animals. Apart from EFHD2, mmu-miR-136-5p and mmu-miR-149-5p may have other regulatory targets in MIN6 cells. Furthermore, the downstream signaling pathways regulated by EFHD2 were need further verification. In addition, the functions of astragalus polysaccharide in patients with T2DM needs to be further studied.

Conclusion
HG+PA treatment reduced MIN6 cell viability, insulin secretion, and PDX1, miR-136-5p, and miR-149-5p expression and promoted MIN6 cell apoptosis. Astragalus polysaccharide treatment reversed the effect of HG+PA treatment on MIN6 cells partially by promoting miR-136-5p and miR-149-5p expression to inhibit EFHD2 expression. This study provides a new theoretical basis for the clinical application of astragalus polysaccharide ( Figure 7).

Disclosure statement
No potential conflict of interest was reported by the author(s).

Data availability statement
The original datasets are available from the corresponding author on reasonable request.

Author contributions
Shifang Deng has made contributions to the conception and design of the work, carried out experiment, analyzed the data, and drafted the manuscript. Lei Yang and Ke Ma analyzed and interpreted the data. Wei Bian has made contributions to the conception and design of the work and substantively revised manuscript. All of the authors have read and approved the final manuscript.

Funding
The author(s) reported there is no funding associated with the work featured in this article.