MicroRNA array expression profile data (GSE98043) was downloaded from GEO database (https://www.ncbi.nlm.nih.gov/geo/) (26). GSE98043 was analyzed using Agilent-070156 Human_miRNA_V21.0_Microarray 046064 (Feature Number version) and consisted of 4 plasma samples from pregnant women (2 from normal controls and 2 from GDM patients). Differentially expressed miRNAs (DEmiRNA) were identified through the R ‘limma’ package (Version 4.2), which is a widely used tool that can be used to analyze data from any GEO series and significance analysis of microarray (SAM), to determine the differential expression of microRNAs among groups (27). miRNAs were considered to be differentially expressed according to the P<0.05 threshold from the limma analysis and median false discovery rate <0.05 from SAM. Data were visualized as heat maps using the online tool Morpheus (https://software.broadinstitute.org/morpheus/).

Whole blood samples were collected from 30 women who underwent pregnancies with GDM and 30 healthy women who underwent healthy pregnancies (control) at the Department of Gynecology, Handan Central Hospital (Handan, China) between November, 2019 and November, 2020. GDM was diagnosed according to the World Health Organization/International Association of the Diabetes and Pregnancy Study Groups criteria (28,29). The GDM patients with pre-gestational diabetes, multiple gestations accompanied with further complications and those taking medications were all excluded and pregnancies with GDM who did not fulfill any of the exclusion criteria were recruited in the current study. The control participants without GDM, pre-eclampsia, low birthweight, or preterm delivery were selected in this study. The main clinicopathological parameters are reported and summarized in Table I, and data are presented as means ± SD for both control and GDM groups. The present study was approved by the Ethics Committee of Handan Central Hospital. Briefly, the whole blood samples were obtained from all individuals through venous blood collection prior to breakfast, before they were put into a heparin sodium-containing anticoagulation tube. All blood samples were obtained with informed consent. The blood glucose levels were measured using a glucose assay kit (cat. no. ab65333; Abcam).

The MIN6 murine pancreatic β-cell line, a well-established cellular model for studying islet β-cell function, was purchased from The Cell Bank of Type Culture Collection of the Chinese Academy of Sciences. They were cultured with DMEM (Gibco; Thermo Fisher Scientific, Inc.) containing 12% FBS (Gibco; Thermo Fisher Scientific, Inc.), 50 µM β-mercaptoethanol, penicillin (100 U/ml) and streptomycin (100 U/ml; Sigma-Aldrich; Merck KGaA) at 37˚C in a humidified atmosphere under 5% CO₂.

In the present study, normal glucose (3.3 mM glucose) was set as the control group. MIN6 cells were incubated in complete medium containing 16.7 mM glucose (final concentration in the medium) for 24 h at 37˚C, which was set as the HG group (30). By contrast, cells in the HG + miR-143-3p mimics or mimic-negative control (NC) group were transfected with miR-143-3p mimics or mimics NC for 24 h at 37˚C, before being cultured in 16.7 mM glucose for another 24 h at 37˚C. In addition, cells in the HG + miR-143-3p mimics + pcDNA-TAK1/pcDNA-empty group were co-transfected with the miR-143-3p mimic and the pcDNA-TAK1 plasmid or pcDNA-empty vector at 37˚C for 6 h, and subsequently the transfection reagent was removed and the medium containing 16.7 mM glucose was added for 48 h at 37˚C.

miR-143-3p mimics, mimics negative control (NC), miR-143-3p inhibitor, inhibitor NC and pcDNA-TAK1 were designed and synthesized by Shanghai GenePharma Co., Ltd. and the pcDNA 3.1 vector (pcDNA empty) was used as a control for the overexpression vector pcDNA-TAK1. The miR-143-3p mimics sequence was 5'-UGAGAUGAAGCACUGUAGCUC-3'; whereas the mimics NC sequence was 5'-ACUGAGCUCGUAGUAUGCAAG-3'. The miR-143-3p inhibitor sequence was 5'-GAGCUACAGUGCUUCAUCUCA-3', whilst the inhibitor NC sequence was 5'-AGUCUCGCGCUUACUGAAUCA-3'.

When the MIN6 cell confluency reached 75%, miR-143-3p mimics (50 nM) or inhibitor (100 nM) and/or pcDNA-TAK1/pcDNA empty vector (2 µg; accession no. NM_172688.3) were transfected or co-transfected into the cells using the Lipofectamine 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.) by following the manufacturer's protocol. After 48 h in total, the cells were collected for further experiments.

Total RNA from whole blood and cultured cells were extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. For the reverse transcription of miRNA and mRNA, cDNA was synthesized using an miScript II RT kit (Qiagen GmbH) at 37˚C for 60 min and PrimeScript™ RT reagent kit (Takara Bio, Inc.) at 42˚C for 30 min, respectively. Subsequent qPCR was performed using the SYBR^® Premix Ex Taq™ (Takara Bio, Inc.) on an ABI Prism 7500 Sequence Detection System (Thermo Fisher Scientific, Inc.). The following thermocycling conditions were used: Pre-denaturation at 95˚C for 10 sec; followed by 40 cycles of denaturation at 95˚C for 15 sec and annealing at 55˚C for 15 sec, and extension at 72˚C for 40 sec; and a final extension of 10 min at 72˚C. The following primer sequences were used: miR-143-3p forward, 5'-CTGGCGTTGAGATGAAGCAC-3' and reverse, 5'-CAGAGCAGGGTCCGAGGTA-3'; U6 forward, 5'-CTCGCTTCGGCAGCACA-3' and reverse, 5'-AACGCTTCACGAATTTGCGT-3'; TGFβ-activated kinase 1 (TAK1) forward, 5'-AGCTTGATGACACGCTGTTG-3' and reverse, 5'-GAGTTGCTCTGCCCTTCATC-3'; GAPDH forward, 5'-CTCATGACCACAGTCCATGCCATCACTG-3' and reverse, 5'-CATGAGGTCCACCACCCTGTTGCTGTA-3'. miR-143-3p expression was normalized to that of U6, whereas TAK1 expression was normalized to that of GAPDH. Data were analyzed by using the 2^-ΔΔCq method (31).

Cells were incubated in DMEM containing 3.3 or 16.7 mM glucose at 37˚C for 24 h, respectively. After cell sonication as below: 3 sec On and 5 sec OFF on ice for 10 min, the supernatant was collected after centrifugation at 12,000 x g for 10 min at 4˚C. Glucose-stimulated insulin secretion into supernatant was measured using an insulin ELISA kit (Cat no. P01325) from RayBiotech, Inc. according to the manufacturer's protocols.

Cells were seeded into a 96-well plate with 1x10⁴ cells per well. When the MIN6 cell confluency reached 75%, the cells were transfected at 37˚C for 24 h with the miRNA mimics/inhibitor or plasmids using the Lipofectamine 2000 transfection reagent, before being treated with 16.7 mM glucose. After 0, 12, 24 and 48 h treatment, 10 µl CCK-8 solution (Dojindo Molecular Technologies, Inc.) was added to each well, before the cells were continuously incubated for another 2 h at 37˚C and the absorbance was read using a microplate reader (BioTek Instruments, Inc.) at 450 nm according to the manufacturer's protocols.

Cells were seeded into six-well plates at 1x10⁶ cells per well. When the MIN6 cell confluency reached 75%, the cells were transfected at 37˚C for 24 h with the miRNA mimics/inhibitor or plasmids using the Lipofectamine 2000 transfection reagent, before they were treated with 16.7 mM glucose. After 24 h treatment, MIN6 β-cells in each group were detached by trypsinization and collected by centrifugation at 12,000 x g for 10 min at 4˚C. To measure apoptosis, the Annexin V-FITC Apoptosis Staining/Detection (Abcam) was used according to the manufacturer's instructions. The cells (1x10⁵ cells) were then resuspended in pre-cooled 500 µl binding buffer and incubated with 5 µl Annexin V-FITC and 1 µl propidium iodide (PI). After incubation at room temperature in the dark for 15 min, cell apoptosis was analyzed using a FACScan flow cytometer (BD Biosciences). The MultiCycle Software version 5.0 (Phoenix Flow Systems, San Diego, CA, USA) for Windows 7 (Microsoft Corporation, Redmond, WA, USA) was used to analyze the experimental data.

Caspase-3 activity was measured using a Caspase-3 Activity kit (Cat no. C1115, Beyotime Institute of Biotechnology) according to the manufacturer's protocol. Briefly, cells were harvested 24 h after 16.7 mM glucose treatment at 37˚C and lysed in RIPA buffer (cat no. P0013B, Beyotime Institute of Biotechnology). Caspase-3 activity assay was performed in 96-well plates by incubating 10 µl of each cell lysate sample in 80 µl reaction buffer containing 10 µl of the caspase-3 substrate Ac-DEVD-pNA; (2 mM) at 37˚C for 2 h. The samples were measured with a microplate reader (Model 680; Bio-Rad Laboratories, Inc., Hercules, CA, USA) at an absorbance of 405 nm.

In the culture supernatants, TNF-α (cat no. MTA00B), IL-8 (cat no. DY442) and IL-6 (cat no. M6000B) levels were measured using ELISA kits (R&D Systems, Inc.) according to the manufacturer's protocols. Absorbance was detected using an automatic multi-well spectrophotometer (Bio-Rad Laboratories, Inc.) at 450 nm.

TargetScan's bioinformatics prediction (version 7.2; www.targetscan.org/vert_72) was used to predict gene targets for miR-143-3p. Gene ontology enrichment analysis and alternative analyses were carried out using EnrichR (version: January 23rd). An adjusted P-value cut-off of below 0.05 was used to filter results. The fragment of the 3'-untranslated region (UTR) of TAK1 [wild-type (wt) or mutant (mut)] was amplified and cloned into the pGL-control vector (Promega). Site-directed mutagenesis of the TAK1 3'-UTR at the putative miR-143-3p binding site was performed using a QuikChange II Site-Directed Mutagenesis kit (Agilent Technologies, Inc., Santa Clara, CA, USA). MIN6 cells at a density of 1x10⁵ per well were seeded into a 24-well plate until they reached ~60% confluence. In total, 2.5 µg wild-type (wt)-TAK1-UTR-pGL3 or mutant (mut)-TAK1-UTR-pGL3 plasmids was co-transfected with 100 nM miR-143-3p mimics or inhibitors using Lipofectamine^® 2000 (Thermo Fisher Scientific, Inc.). Following 48 h transfection, the luciferase activity was measured using a Dual-luciferase reporter assay system (Promega Corporation). pRL-TK plasmid, containing the Renilla luciferase activity, was transfected as an internal control.

Total cellular proteins were extracted using RIPA buffer (Beyotime Institute of Biotechnology) and protein concentration was determined using the BCA kit (Pierce; Thermo Fisher Scientific, Inc.). Protein isolation and immunoblotting were performed as previously described (32). Briefly, total protein (40 µg/lane) was separated by 8% SDS-PAGE and electrophoretically transferred onto a PVDF membrane (MilliporeSigma). Subsequently, membranes were blocked with 5% skim milk for 1 h at room temperature. The membranes were incubated with specific primary antibodies against TAK1 (cat no. 5206, 1:1,000), cleaved caspase 3 (cat. no. 9661, 1:1,000), inhibitor of NF-κB (IκB) kinase subunit β (IKKβ; cat. no. 8943; 1:1,000), phosphorylated (p-)-IKKβ (cat. no. 2697; 1:1,000), anti-IκBα (cat. no. 4812; 1:1,000), p-IκBα (cat. no. 2859; 1:1,000) or β-actin (cat. no. 4970; 1:1,000) at 4˚C overnight. Following incubation with HRP-conjugated anti-rabbit IgG secondary antibody (cat. no. 7074; 1:2,000) for 1 h at room temperature, the bands were detected using ECL Prime Western Blotting Detection Reagent(Cytiva). All antibodies were purchased from Cell Signaling Technology, Inc. The intensities of the bands of interest were analyzed using the ImageJ software (version 1.46; National Institutes of Health).

Data are represented as the mean ± SD from ≥ three replicates and analyzed using the SPSS 22.0 software package (IBM Corp.). Significance was determined using unpaired Student's t-test between two groups whereas for multiple comparisons among > two groups one-way ANOVA followed by Tukey's post hoc test was used. P<0.05 was considered to indicate a statistically significant difference. Correlation analysis was performed using Pearson's correlation coefficient analysis.

Firstly, 46 miRNAs, including 20 upregulated and 26 downregulated miRNAs, were found from the GEO dataset with the accession number GSE98043 (Fig. 1A). In this list, 20 miRNAs were upregulated and 26 miRNAs were downregulated in the blood samples between the GDM and Control group. Notably, miR-143-3p, miR-574-5p and miR-3135b were found to be downregulated, whereas miR-429, miR-520 and miR-222 were particularly upregulated, compared with those in the control group (Fig. 1A). Association between miRNAs miR-574-5p, miR-3135b, miR-429, miR-520, miR-222 and GDM has been previously reported (25,26,33,34), suggesting the experimental validity of our microarray findings. Notably, miR-143-3p showed the highest degree of downregulation among other miRNAs screened in this dataset, consistent with previous studies (35,36). Furthermore, miR-143-3p has been previously reported to serve an important role in the inflammatory response and apoptosis in several diseases, such as mycoplasmal pneumonia and autoimmune hepatitis (37,38). However, its role in the development of inflammatory response and apoptosis in GDM remains unclear. Therefore, miR-143-3p was selected for further study.

To validate the expression levels of miR-143-3p, RT-qPCR was performed in blood samples isolated from 30 pregnant patients with GDM and 30 healthy pregnant individuals. As shown in Fig. 1B, miR-143-3p levels were significantly lower in the peripheral blood of patients with GDM compared with those in healthy individuals. Pearson correlation analysis revealed an inverse correlation between the blood glucose concentration and miR-143-3p levels among the patients with GDM (r=-0.8792; Fig. 1C). Subsequently, HG-induced MIN6 cell model was established as previously described (28). RT-qPCR results demonstrated that miR-143-3p expression was significantly decreased in the HG group compared with that in the control group (Fig. 1D). These data suggest that miR-143-3p may be involved in the pathogenesis of GDM.

To investigate the function of miR-143-3p in GDM, MIN6 cells were transfected with the miR-143-3p mimics or mimics NC, before being cultured under 16.7 mM glucose (HG). According to the results of RT-qPCR, miR-143-3p was found to be significantly increased after miR-143-3p transfection compared with that in cells transfected with mimics NC, suggesting that the transfection of miR-143-3p mimics was successful (Fig. 2A). As shown in Fig. 2B and C, HG resulted in a significant increase in insulin concentration and secretion in MIN6 cells compared with those in the control group, which was significantly prevented by miR-143-3p overexpression beforehand. Furthermore, the effects of miR-143-3p on MIN6 cell viability and apoptosis was next assessed. A significant decrease in cell viability in HG-stimulated MIN6 cells was detected, but this inhibitory effect was reversed after the overexpression of miR-143-3p (Fig. 2D). In addition, HG-induced cell apoptosis was significantly reversed by miR-143-3p overexpression in MIN6 cells (Fig. 2E). Similarly, caspase 3 activity assay also revealed that caspase 3 activity was significantly increased by HG, which was likewise significantly reversed by the overexpression of miR-143-3p in MIN6 cells (Fig. 2F and G). The caspase 3 activity assay data were supported by those from western blotting, which revealed significantly increased caspase 3 cleavage in response to HG but was in turn significantly reversed by miR-143-3p overexpression (Fig. 2H). These data indicate that miR-143-3p overexpression can promote insulin secretion whilst inhibiting cell apoptosis in HG-treated MIN6 cells.

Apart from insulin resistance, excessive inflammation is also one of the key factors involved in the pathogenesis of GDM (39,40). Therefore, the effects of miR-143-3p on the inflammatory response in MIN6 cells after HG treatment were next investigated. HG stimulation led to a significant increase in the levels of proinflammatory factors TNF-α, IL-6, and IL-8 (Fig. 3A-C). However, overexpression of miR-143-3p significantly decreased the levels of these proinflammatory factors previously raised by HG in MIN6 cells. These data suggest that the overexpression of miR-143-3p can suppress the HG-induced inflammatory response in MIN6 cells.

To investigate the underlying molecular mechanism involved in the miR-143-3p-mediated suppression of inflammation and apoptosis, the TargetScan software was next used to analyze the putative target genes of miR-143-3p. Bioinformatics analyses predicted that 3'-UTR regions of the TAK1 gene contains a potential binding site for miR-143-3p (Fig. 4A). Therefore, MIN6 cells were next transfected with either the miR-143-3p mimics or miR-143-3p inhibitor. The expression levels of miR-143-3p were significantly increased and decreased after transfection with the miR-143-3p mimics and miR-143-3p inhibitor, respectively, compared with those in their corresponding NC (Fig. 4B). To verify if TAK1 is indeed a direct target of miR-143-3p, a dual-luciferase reporter assay was performed. Compared with that in their corresponding NCs, overexpression of miR-143-3p was found to significantly decrease the luciferase activities of the 3'-UTR segment of TAK1, whilst luciferase activities of the same construct was significantly increased by miR-143-3p knockdown (Fig. 4C). Subsequently, the potential effects of miR-143-3p on the expression of TAK1 was measured on a protein level in MIN6 cells by western blot analysis. As shown in Fig. 4D, compared with that in their corresponding NCs, the expression of TAK1 proteins was significantly reduced following the overexpression of miR-143-3p, but was significantly upregulated following the knockdown of miR-143-3p, in the MIN6 cells. Since the expression levels of miR-143-3p were lower in the whole blood samples of patients with GDM, the expression levels of TAK1 were also detected in the samples from 30 pregnant individuals with GDM and 30 healthy pregnant individuals. As shown in Fig. 4E, TAK1 expression was significantly upregulated in whole blood samples of patients with GDM compared with that of healthy individuals. In addition, Pearson correlation analysis revealed an inverse correlation between TAK1 and miR-143-3p expression among the GDM patients (r=-0.7520; Fig. 4F). These results suggest that TAK1 is a direct functional target of miR-143-3p in MIN6 cells.

To clarify if TAK1 is involved the inhibitory effects of miR-143-3p on HG-induced inflammation and apoptosis in MIN6 cells, the TAK1 expression vector pcDNA-TAK1 and miR-143-3p mimics were co-transfected into MIN6 cells followed by HG treatment. First, the expression of the TAK1 protein was markedly increased in MIN6 cells after transfection with the pcDNA-TAK1 plasmid compared with that in cells transfected with the empty plasmid (Fig. 5A). Insulin concentration and insulin secretion were next measured, both of which were significantly increased by miR-143-3p overexpression in HG-treated MIN6 cells (Fig. 5B and C). However, these miR-143-3p overexpression-mediated increases were in turn significantly reversed by TAK1 overexpression (Fig. 5B and C). In addition, the increased cell viability induced by miR-143-3p mimic transfection was also significantly reversed by TAK1 overexpression in HG-treated MIN6 cells (Fig. 5D). Subsequently, apoptosis and inflammatory response in HG-treated MIN6 cells were measured. As shown in Fig. 5E and F, the inhibitory effects of miR-143-3p overexpression on the apoptosis in HG-treated MIN6 cells were also significantly reversed by TAK1 overexpression. Similarly, the reduction of proinflammatory cytokine secretion caused by miR-143-3p overexpression were all significantly reversed by TAK1 overexpression (Fig. 5G-I). These data therefore strongly supports the notion that miR-143-3p overexpression can protect MIN6 cell function by negatively targeting TAK1.

It has been extensively reported that NF-κB functions as a key regulator in the inflammatory process in a wide range of diseases, particularly inflammation-related diseases including GDM (41,42). The present study showed that miR-143-3p overexpression can negatively regulate the expression of TAK1 in MIN6 cells, which is an important upstream activator of the NF-κB pathway. Therefore, the potential effects of miR-143-3p on the activation of TAK1/NF-κB signaling in vitro were next examined. The protein expression levels of TAK1, p-IKKβ and p-IκBα were detected by western blotting. It was found that HG treatment led to a significant increase in IKKβ and IκBα phosphorylation in addition to increasing TAK1 expression in MIN6 cells compared with those in the control group (Fig. 6A). However, these effects were significantly reversed by miR-143-3p overexpression in MIN6 cells (Fig. 6B). These results suggest that miR-143-3p overexpression blocked the TAK1/NF-κB pathway in MIN6 cells under HG conditions.