Caffeic Acid Dimethyl Ether Ameliorates Excessive Glucose and Lipid-Induced Insulin Secretion Dysfunction in Pancreatic Beta-Cells through the miR-378b–PI3K–AKT Pathway

. An important factor in the progression of type 2 diabetes mellitus is the malfunctioning insulin production by β -cells in the pancreas. Cafeic acid dimethyl ether (CADE) reduces resistance to insulin in alcoholic fatty liver disease, but both the therapeutic efects of CADE on excessive glucose and lipid-induced insulin secretion disorders and the underlying mechanisms are unknown. Te aim of this research was to (i) explore how CADE impacts insulin production issues caused by a surplus of glucose and lipids β -cells and (ii) elucidate the underlying mechanism. Te results of our research demonstrated that insulin production was reduced in the pancreas of mice given a high-fat -diet and streptozotocin, as well as in human 1.1B4 pancreatic β -cells treated with high -glucose and high -fat, with increased activity of miR-378b and decreased expression levels of p110 α , p-AKT1/2, insulin receptor, p-FoxO1, and PDX-1. However, treatment with CADE ameliorated the insulin secretion impairment by decreasing the miR-378b level and reversing the inhibitory efects on the aforementioned factors. Overexpression of miR-378b exacerbated the insulin secretion disorder and inhibited the PI3K-AKT signaling pathway, whereas miR-378b defciency relieved the insulin secretion disorder, activating the PI3K-AKTpathway. In addition, CADE ameliorated the impairment of insulin secretion and reversed the miR-378b overexpression-induced PI3K-AKTpathway inhibition. In conclusion, our study demonstrates that CADE ameliorated insulin secretion dysfunction induced by excess lipid and glucose in β -cells by downregulating miR-378b expression, thus promoting PI3K-AKT activation.


Introduction
Te prevalence of diabetes is on the rise globally.As per estimates, around 536.6 million people sufered from diabetes in 2021, representing approximately 10.5% of the global population.Te numbers are projected to increase to 783.2 million people in 2045, accounting for 12.2% of the global population [1].Diabetes mellitus and its complications greatly afect patients' quality of life and physical health [2].Terefore, fnding efective ways to prevent diabetes is a crucial public health issue [3].Te main causes of type 2 diabetes mellitus (T2DM) are a decreased number of pancreatic β-cells, insulin secretion problems, and insulin resistance [4].
Previous research has demonstrated that chronic hyperglycemia and hyperlipidemia are key risk factors for impaired insulin secretion by β-cells [5].Some evidence suggests that palmitate and glucose reduce the dephosphorylation of AKT and FoxO1 in β-cells, thereby inhibiting insulin secretion [6,7].Furthermore, the phosphatidylinositol 3-kinase (PI3K)/AKT pathway is a vital pathway for regulating insulin secretion in β-cells [8][9][10].
MicroRNAs (miRNAs) are tiny endogenous RNAs that regulate gene expression at the post-transcriptional level [15].Tey play a crucial part in regulating insulin secretion via targeting various proteins participating in the insulin signaling pathway [16].Several reports have shown that miR-7, miR-29, and miR-802 are important regulators of insulin secretion [17][18][19].Moreover, miR-378b improves insulin resistance in the liver by targeting IR and p110α [20,21].Nevertheless, the exact function of miR-378b in β-cells remains unknown.
Currently, the main clinically used insulin secretagogues include DPP-4 inhibitors, sulfonylureas, and GLP-1R agonists.Tey efectively decrease the level of blood glucose to treat T2DM by promoting insulin secretion.However, sulfonylureas increase the risk of cardiovascular disease and hypoglycemia [22,23]; GLP-1R agonists may cause transient gastrointestinal reactions and pancreatitis [24]; and longterm use of DDP-4 inhibitors may also put patients with heart failure at risk of abnormal cardiac remodeling [25,26].Terefore, the search for safe and efective novel drugs to prevent and treat insulin secretion disorders has been a major research focus worldwide.In recent years, natural drugs such as mulberry alkaloids, cordycepin, and berberine have received attention for their potential in improving insulin secretion [27][28][29].Cafeic acid dimethyl ether (CADE), also known as methyl ferulic acid, is a bioactive monomer extracted from traditional Chinese herbs [30].It can improve insulin resistance in mice with alcoholic fatty liver [21] and alleviate hepatic steatosis [31].However, it remains unclear what efect CADE has on impaired insulin secretion in diabetic patients.
Te aim of the present study was twofold: (i) to explore the efect of CADE on the impairment of insulin secretion in β-cells caused by excessive glucose and lipids and (ii) to determine whether CADE ameliorates insulin secretion impairment by modulating the miR-378b-PI3K-AKT pathway.We discovered for the frst time that CADE may ameliorate excessive glucose and lipid-induced insulin secretion dysfunction through the miR-378b-PI3K-AKT pathway.

Animal Models and
Treatments.Tirty male C57BL/6J mice (5 weeks old, weight 18-22 g) were provided by SJA Laboratory (Hunan, China).After one week of acclimatization, they were randomly divided into two groups: the control group (n � 6) and the diabetic group (n � 24).Te mice in the control group were fed a normal-fat diet (12% fat, Trophic Animal Feed High-tech Co., Ltd., Nantong, China).Te mice in the diabetic group were fed a high-fat diet (HFD; 60% fat, Trophic Animal Feed High-tech Co., Ltd., Nantong, China).After four weeks of high-fat diet, all mice were fasted for 8 h and then intraperitoneally injected with streptozotocin (STZ, 55 mg/kg).STZ was injected for three days.After a week of the fnal STZ injection, the fasting blood glucose level was examined with a glucometer (Yuyue Medical Equipment and Supply Co., Ltd.Jiangsu, China).Mice with fasting blood glucose >11.1 mM were defned as T2DM mice [32,33].Te T2DM mice were randomly divided into four groups: the diabetic group (DC), the diabetic + 10 mg/kg glibenclamide group (DC + Gliben), the diabetic + 5 mg/kg CADE group (DC + CADE-L), and the diabetic + 10 mg/kg CADE group (DC + CADE-H) [31].Mice were gavaged with CADE, glibenclamide or 0.5% CMC-Na once daily for six weeks.Body weight and fasting blood glucose were recorded weekly.At week 5 of treatment, the oral glucose tolerance test (OGTT) was performed on the mice.
After six weeks of treatment, all mice were fasted for 8 h and anesthetized with 1.25% 2,2,2-tribromoethanol (Cat# M2910, Aibei Biotechnology).Serum was taken for insulin analysis.Finally, the mice were euthanized and the pancreas was isolated for subsequent assays.Te study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Ethical Committee of Guilin Medical University.

OGTT.
Five weeks after CADE administration, all mice were fasted for 8 h and their blood glucose levels were measured.Following this, a glucose solution with 2 g/kg body weight was orally administered to each mouse.Te level of blood glucose was then monitored at 15-, 30-, 60-, and 120-minutes post-infusion.

Hematoxylin and Eosin
Staining.Pancreatic tissue was placed in a 4% paraformaldehyde solution for 24 h.Afterward, it was dehydrated in a graded series of alcohol and placed in wax blocks.Tese blocks were sectioned into 3 µm

2
Journal of Food Biochemistry thick slices and stained with hematoxylin and eosin (H&E).Pathological changes were observed by light microscopy and images of the islet tissue were taken.

Immunofuorescence Assay.
To investigate insulin expression in pancreatic sections, the parafn-embedded tissues were frst deparafnized and subjected to antigen retrieval by heating in a 3% citric acid solution for 30 min.
After blocking with serum for 30 min, these sections were incubated with anti-insulin (1 : 80) at 4 °C overnight.Afterward, the sections were incubated with a fuorescent secondary antibody (1 : 200) for 30 min, followed by DAPI staining at room temperature for 10 min to label nuclei.Finally, the sections were sealed with a water-soluble blocker.Images were captured by fuorescence microscopy.After the cells formed three-dimensional pseudo-islet masses, they were divided into the control, HG + HF, and low-, medium-, and high-dose CADE groups [34,35].Te control group continued to be cultured in complete medium, the HG + HF group was cultured in complete medium with HG and HF, and the CADE groups were cultured in complete medium with HG, HF and CADE (0.5, 1, or 2 μM).
(ii) Similarly, the transfected cells were split into the following groups: the miR-378b mimics group, the miR-378b NC group, the miR-378b NC + CADE group, the miR-378b inhibitor group, the miR-378b mimics + CADE group, and the miR-378b inhibitor + CADE group.Each group was incubated with HG and HF, and 1 μM CADE was additionally added to the CADE groups.Cells were incubated for 48 h.Ten, fve pseudo-islet masses were isolated under a microscope, placed into EP tubes, and incubated with KRBH bufer with 0.1% bovine serum albumin for 3 h.Afterward, the supernatant was collected.Finally, an ELISA kit (EZ assay Ltd., Guangdong, China) was applied to assess the insulin concentration in the cell supernatant.

Western Blot Analysis.
Total protein was obtained from CADE-treated 1.1B4 cells for Western blot analysis.Proteins from each group were separated by SDS-PAGE and transferred onto nitrocellulose membranes, which were incubated with primary antibodies at 4 °C overnight, washed three times, and incubated with secondary antibody (1 : 10,000) for 1 h at room temperature.Te expression of target protein was analyzed using a highly sensitive chemiluminescent reagent (Shanghai Liji, Shanghai, China) on a ChemiDoc XRS + type gel imager (Bio-Rad, CA, USA).

Real-Time
Quantitative PCR.TRIzol reagent (Tiangen, Beijing, China) was applied to extract total RNA from the 1.1B4 cells and the pancreas.To synthesize the initial cDNA chain of miRNA, the cDNA synthesis kit (GeneCopoeia ™ , Guangzhou, China) was utilized.Afterwards, the cDNA produced was utilized in qRT-PCR for the determination of miR-378b levels.Te 2 −ΔΔCt method was applied to evaluate the comparative expression of miR-378b, with U6 as the reference gene.Te primers used in the PCR experiments were acquired from Sangon Biotech (Shanghai, China) and their sequences can be found in Table 1.

Statistical Analysis.
All experiments were performed at least three times.Unless stated otherwise, the fndings are shown as the mean and standard deviation (SD).Data were analyzed by one-way ANOVA using SPSS 25.P < 0.05 indicates a statistical signifcance.

CADE Ameliorated Insulin Secretion Disorder and Downregulated miR-378b in HFD/STZ Mice.
To explore the impact of CADE on insulin secretion disorder, we constructed a T2DM mouse model by combined HFD/STZ induction and treated the mice with 0.5% CMC-Na, glibenclamide, or CADE for six weeks (Figure 1(a)).Compared with the control group, the DC group showed signifcant weight loss, poor glucose tolerance, and elevated fasting glucose.However, after CADE and glibenclamide treatment, the mice had notably higher body weight, lower fasting glucose, and worse glucose tolerance (Figures Notably, no statistically signifcant changes in body weight were observed in the DC + Gliben group.4 Journal of Food Biochemistry H&E staining was carried out on pancreatic tissue sections to analyze the pathological changes in mouse islets.Te DC group exhibited severe damage to the islet structure compared with the control group, with a signifcant reduction in islet tissue area and irregularly shaped islets.However, the size and shape of the islets tended to be normal, and the structure was intact with clear borders after CADE or glibenclamide intervention (Figure 1(e)).Tese fndings suggest that CADE might protect against histopathological damage to the islets in HFD/STZ mice.
Te synthesis and secretion of insulin in mice were analyzed by tissue immunofuorescence and ELISA.Compared with the control group, a signifcant reduction in insulin fuorescence and lower insulin levels were observed in the DC group, along with a considerable decrease in serum insulin levels.By contrast, CADE treatment resulted in more intense insulin fuorescence, and both islet and serum insulin levels were substantially increased (Figures 1(f ) and 1(g)).Remarkably, the best efect on insulin synthesis and secretion was observed in the DC + CADE-H group.In addition, qRT-PCR analysis demonstrated a dramatic increase in miR-378b levels in the DC group with the control group.However, miR-378b was signifcantly downregulated in the CADE and glibenclamide groups.Te DC + CADE-H group showed the strongest downregulation.Te DC + CADE-H group showed the strongest downregulation (Figure 1(h)).Interestingly, the promotion of insulin synthesis and secretion by CADE was closely associated with the expression of miR-378b.

CADE Enhanced the Viability of HG/HF-Induced 1.1B4
Cells.Cell viability was analyzed using the MTT assay.As illustrated in Figure 2(a), after 48 h of culture of 1.1B4 cells with 0.1-16 μM CADE, there was no statistically signifcant diference in cell viability compared with the control group.Glucose and sodium palmitate were administered to 1.1B4 cells for 48 h.Te viability of 1.1B4 cells treated with 30 mM glucose and ≥0.2 mM sodium palmitate was signifcantly inhibited compared with that of control cells (Figure 2(b)).To investigate the efects of CADE on HG + HF-induced 1.1B4 cells, 1.1B4 cells were co-cultured with 30 mM glucose, 0.2 mM sodium palmitate, and CADE (0.5, 1, or 2 μM) for 48 h.Cell viability was signifcantly enhanced by CADE compared with the HG + HG group (Figure 2(c)).

CADE Improved HG + HF-Induced Insulin Secretion Dysfunction and Reduced Abnormally Elevated miR-378b
Levels in 1.1B4 Cells.To measure the impact of CADE on insulin secretion dysfunction in HG + HF-induced 1.1B4 cells, we examined the intracellular expression level of insulin by an immunofuorescence assay.In addition, 1.1B4 cells were cultured in low-attachment culture dishes to form three-dimensional pseudo-islet masses (Figure 2(e)), and then insulin secretion was analyzed by ELISA in the cell supernatant.Compared with the control group, both insulin synthesis and secretion were decreased signifcantly in the HG + HF group.CADE enhanced the secretion and synthesis of insulin in a dose-dependent manner (Figures 2(d), 2(f )).Furthermore, the proinsulin mRNA levels in 1.1B4 cells were further analyzed by RT-PCR.Te results indicated that proinsulin mRNA levels were decreased by approximately 50% in HG + HF-induced 1.1B4 cells compared with the control group.CADE increased proinsulin mRNA levels in a dose dependence-dependent manner (Figure 2

(g)).
Tese results suggest that CADE might promote insulin synthesis and secretion by increasing proinsulin mRNA expression, ultimately improving insulin secretion disorders.

CADE Ameliorated HG + HF-Induced Insulin Secretion Dysfunction in 1.1B4 Cells via the PI3K-AKT Signaling
Pathway.Te network bioinformatics analysis revealed that miR-378b has binding sites in the 3′-UTRs of INSR (encoding IR) and PIK3CA (encoding p110α) (Figure 3(a)).Consequently, we reasoned that miR-378b may afect insulin secretion by regulating IR and p110α expression.To validate the association between miR-378b, IR, and p110α, we frst examined the mRNA and protein levels of IR and p110α in HG + HF-induced 1.1B4 cells, as well as the levels of their downstream related factors.Te results revealed that the cellular p85, IR, p110α, and p-IR protein levels were greatly reduced in HG + HF-induced 1.1B4 cells compared with the control group, while the mRNA expression levels of IR and p110α were also signifcantly decreased.Surprisingly, CADE treatment reversed the inhibitory efects of the above factors (Figures 3(b)-3(d)).Further analysis of the downstream factors of IR and p110α revealed that the p-Akt1/Akt1, p-Akt2/Akt2, p-FoxO1/FoxO1, and PDX-1 expression levels were remarkably decreased, whereas the levels of the above proteins were increased after CADE application in a dosedependent manner (Figures 3(e), 3(f )).Tese results suggest that the benefcial efects of CADE on insulin secretion dysfunction are closely associated with the regulation of IR and p110α expression by miR-378b.

CADE Afected HG + HF-Induced Insulin Secretion
Dysfunction in 1.1B4 Cells via miR-378b.To determine the direct relationship between CADE and miR-378b in improving insulin secretion dysfunction, 1.1B4 cells were transfected with miR-378b inhibitor or mimics plasmids.Te fuorescence intensity of each group was determined by fuorescence microscopy after 12 h.At least 80% of the cells in each group expressed GFP, so the experimental requirements were met (Figure 4(a)).Te transfected cells were then treated with 0.2 mM sodium palmitate and 30 mM glucose for 48 h; cells in the CADE treatment group were additionally treated with 1 μM CADE.qRT-PCR analysis showed that miR-378b levels were signifcantly downregulated in the miR-378b inhibitor group and upregulated in the miR-378b mimic group (Figure 4(b)).In addition, the miR-378b levels were signifcantly decreased in the miR-378b NC + CADE group compared with the miR-378b NC group.Furthermore, although miR-378b silencing resulted in the downregulation of miR-378b, the diference in miR-378b levels was not signifcant after CADE intervention (Figure 4(c)).Tese results suggest that CADE may improve insulin secretion dysfunction by regulating miR-378b expression.
Immunohistochemistry and ELISA assays were performed to analyze insulin synthesis and secretion in transfected 1.1B4 cells.Te fndings revealed that the insulin secretion and synthesis levels were enhanced in the miR-378b NC + CADE group compared with the miR-378b NC group, and the levels were reduced to diferent degrees when miR-378b was overexpressed, and both were signifcantly increased after the co-application of CADE and miR-378b mimics (Figures 5(a), 5(c)).On the contrary, insulin synthesis and secretion levels were signifcantly upregulated in the miR-378b inhibitor group after silencing, and it is worth mentioning that the levels were not statistically signifcant after the combined application of CADE and miR-378b inhibitor (Figures 5(b), 5(d)).In addition, proinsulin mRNA levels were further analyzed by PCR.Te mRNA levels of proinsulin were consistent with the results of insulin synthesis and secretion described above (Figures 5(e), 5(f )).Tese fndings suggest that miR-378b overexpression impairs the secretory function of β-cells, which can be improved by CADE treatment.

CADE Improved HG + HF-Induced Impairment of Insulin Secretion in 1.1B4 Cells via the miR-378b-PI3K-AKT
Pathway.Our study has demonstrated that CADE efectively inhibits the impairment of insulin secretory function in T2DM models.We found that the regulatory efect of miR-378b on IR and p110α expression is closely related to the therapeutic efect of CADE on insulin secretion dysfunction.To further investigate the function of the miR-378b-PI3K-AKT pathway in this process, we performed validation experiments on 1.1B4 cells.Te results showed that the protein levels of p-IR, p110α, IR, p-Akt1/Akt1, p-Akt2/Akt2, p-FoxO1/FoxO1, and PDX-1 and the mRNA levels of IR and p110α were signifcantly reduced in the miR-378b mimics group compared with the NC group.However, these protein and mRNA levels were partly restored by CADE treatment.Moreover, the miR-378b inhibitor enhanced the protein levels of IR, p110α, p-IR, p-Akt1/Akt1, p-Akt2/Akt2, p-FoxO1/FoxO1, and PDX-1 and the mRNA levels of IR and p110α.After CADE treatment, p-IR, p110α, and PDX-1 were further upregulated, while the diferences in IR and p110α mRNA levels were not signifcant (Figures 6(a)-6(d) and, 7(a), 7(b)).In summary, CADE improved impairment of β-cell insulin secretion by downregulating miR-378b to facilitate IR and p110α expression, which activated the PI3K-AKT signaling pathway.

Discussion
Te occurrence of T2DM increases every year [36], resulting in substantial negative efects on patients' quality of life and a huge fnancial burden on healthcare systems [37].Hence, it is crucial to fnd efective and safe insulin secretagogues while gaining insight into the molecular mechanisms underlying diabetes.Our research provides pioneering evidence that (i) CADE improved weight loss and glucose homeostasis in HFD/STZ mice and ameliorated insulin secretion dysfunction and β-cell injury in T2DM models, (ii) Te miR-378b-PI3K-AKT signaling pathway is involved in insulin secretion dysfunction in β-cells, and (iii) CADE ameliorated insulin secretion dysfunction in a T2DM model by predominantly suppressing miR-378b and consequently activating the PI3K-AKT pathway.
T2DM is usually associated with fasting glucose >11.1 mM, polyphagia, hyperphagia, and weight loss.In the present study, the symptoms induced by HFD/STZ in C57BL/6J mice were similar to those reported in the above studies, indicating the successful establishment of the T2DM mouse model.
Literature search revealed that CADE alleviates insulin resistance in alcoholic fatty liver disease [38].However, it is unclear whether CADE attenuates excessive glucose and     patients with T2DM, but it does not signifcantly alter body weight, plasma insulin levels, or insulin sensitivity [39].Furthermore, long-term proinsulin secretion with little or no synthesis only depletes intracellular insulin stores.Conversely, CADE had marked efects on both insulin synthesis and secretion.Tese results indicate that CADE is a promising drug candidate for the long-term treatment of T2DM by promoting insulin synthesis and secretion.
Previous research has shown the involvement of miR-NAs in insulin secretion from β-cells [51,52].It has been reported that excessive miR-378, a major regulatory insulin signaling factor, can impair the insulin signaling pathway [53].Based on these reports, we found by bioinformatics analysis that there is a miR-378b binding site in the 3′-UTRs of IR and p110α.In addition, we found that CADE ameliorated insulin secretion dysfunction, inhibited the abnormal elevation of miR-378b levels, and promoted the expression of p110α and IR in excessive glucose and lipidinduced β-cells.Tus, we hypothesized that the benefcial efects of CADE on insulin secretion are mediated by the regulation of miR-378b, which in turn afects p110α and IR expression, activating the PI3K-AKT pathway.We confrmed this hypothesis by miR-378b overexpression and silencing assays.Te fndings indicate that the impaired insulin secretion and the inhibitory efects of the PI3K-AKT pathway were exacerbated in miR-378b overexpressing 1.1B4 cells, while the impaired insulin secretion and the inhibitory efects of the PI3K-AKT pathway were ameliorated in miR-378b-silenced 1.1B4 cells.Co-treatment with CADE and miR-378b mimics alleviated the above aggravated insulin secretion impairment and PI3K-AKT pathway inhibition.Taken together, CADE ameliorates insulin secretion by β-cells through modulation of the miR-378b-PI3K-AKT signaling pathway.
To sum up, we present novel evidence that CADE can efectively alleviate insulin secretion dysfunction caused by excessive glucose and lipids, by modulating the miR-378b-PI3K-AKT signaling pathway in β-cells.However, the clinical application of CADE still poses several challenges since existing research is limited to animal and cellular studies.More clinical studies are required to validate the current fndings and to explore new avenues of CADE application for the treatment of T2DM.

Conclusions
Tis study found that CADE has the potential to alleviate insulin secretion dysfunction in β-cells caused by excessive glucose and lipid.CADE, which downregulates miR-378b and activates the PI3K-AKT signaling pathway, is a promising drug candidate for individuals with insulin secretion dysfunction.Tese fndings ofer signifcant support for the development of new therapeutic strategies for T2DM.Journal of Food Biochemistry

Figure 8 :
Figure 8: Diagram of the insulin secretion signaling pathway.

Table 1 :
Primers for the RT-PCR assay.