Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Feb 24;18(2):e3000603.
doi: 10.1371/journal.pbio.3000603. eCollection 2020 Feb.

Pancreatic β Cell microRNA-26a Alleviates Type 2 Diabetes by Improving Peripheral Insulin Sensitivity and Preserving β Cell Function

Affiliations
Free PMC article

Pancreatic β Cell microRNA-26a Alleviates Type 2 Diabetes by Improving Peripheral Insulin Sensitivity and Preserving β Cell Function

Haixia Xu et al. PLoS Biol. .
Free PMC article

Abstract

Type 2 diabetes (T2D) is characterized by insulin resistance along with pancreatic β cell failure. β cell factors are traditionally thought to control glucose homeostasis by modulating insulin levels, not insulin sensitivity. Exosomes are emerging as new regulators of intercellular communication. However, the role of β-cell-derived exosomes in metabolic homeostasis is poorly understood. Here, we report that microRNA-26a (miR-26a) in β cells not only modulates insulin secretion and β cell replication in an autocrine manner but also regulates peripheral insulin sensitivity in a paracrine manner through circulating exosomes. MiR-26a is reduced in serum exosomes of overweight humans and is inversely correlated with clinical features of T2D. Moreover, miR-26a is down-regulated in serum exosomes and islets of obese mice. Using miR-26a knockin and knockout mouse models, we showed that miR-26a in β cells alleviates obesity-induced insulin resistance and hyperinsulinemia. Mechanistically, miR-26a in β cells enhances peripheral insulin sensitivity via exosomes. Meanwhile, miR-26a prevents hyperinsulinemia through targeting several critical regulators of insulin secretion and β cell proliferation. These findings provide a new paradigm for the far-reaching systemic functions of β cells and offer opportunities for the treatment of T2D.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. MiR-26a is reduced in serum exosomes and islets during obesity.
(A) QRT-PCR analysis of human miR-26a expression in serum exosomes of lean (n = 7) and obese (n = 17, BMI > 25) individuals. (B–E) Correlation between serum exosomal miR-26a levels and BMI (B), HOMA-IR (C), blood fasting glucose (D), or blood fasting insulin (E) in a human cohort (n = 24). (F and G) Levels of mouse miR-26a in serum exosomes of ob/ob mice (n = 4) (F) or WT DIO mice (n = 4) (G). (H–J) Expression of miR-26a in islets of WT DIO mice (n = 4) (H), ob/ob mice (n = 4) (I), or db/db mice (n = 4) (J). (K and L) Expression of pre-miR-26a (K) or pri-miR-26a (L) in islets of WT DIO mice (n = 4), ob/ob mice (n = 3–4), or db/db mice (n = 3–4). The data underlying this figure may be found in S1 Data. Data are shown as mean ± SD. **P < 0.01, ***P < 0.005, Student t test. BLKS, C57BLKS/J, also known as Black Kaliss J; BMI, body mass index; CD, chow diet; db/db mice, leptin-receptor–deficient mice; DIO, diet-induced obese; HFD, high-fat diet; HOMA-IR, homeostatic model assessment index of insulin resistance; ob/ob mice, leptin-deficient mice; pre-miR-26a, precursor miR-26a; pri-miR-26a, primary miR-26a; QRT-PCR, quantitative reverse transcriptase PCR; WT, wild type.
Fig 2
Fig 2. β-cell–specific overexpression of miR-26a prevents obesity-induced hyperinsulinemia and insulin resistance.
(A) QRT-PCR analysis of miR-26a expression in islets of RIP TG mice and WT littermate controls (n = 4). (B–L) Mice were fed an HFD beginning at 6–8 weeks of age. The following measurements were performed during the course of the HFD. (B) Total BW (n = 8–10). Representative picture of RIP TG and WT littermate controls (right panel). (C) Food intake (n = 5–6). (D–G) GTT (n = 7–8) (D), ITT (n = 8–10) (E), blood glucose (n = 9–12) (F), and insulin levels during GTT (n = 7) (G), performed after 8 weeks HFD. (H and I) GTT (n = 5) (H) and ITT (n = 7–10) (I) performed after 16 weeks HFD. (J and K) Blood glucose (n = 8–12) (J) and insulin (n = 5–7) (K) levels of mice that were fed with either CD or HFD for 16 weeks. Random or fasting conditions are noted. (L) HOMA-IR (n = 5–9). The data underlying this figure may be found in S1 Data. Data are shown as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.005, 2-tailed ANOVA (B, D, E, H, I) and Student t test (C, F, G, J–L). BW, body weight; CD, chow diet; GTT, glucose tolerance test; HFD, high-fat diet; HOMA-IR, homeostatic model assessment index of insulin resistance; ITT, insulin tolerance test; QRT-PCR, quantitative reverse transcriptase PCR; RIP, rat insulin promoter; TG, transgenic; WT, wild type.
Fig 3
Fig 3. Serum exosomes from RIP TG mice enhance insulin signaling.
(A) Schematic of the exosome transfer or coculture experiment. (B) AKT phosphorylation in MPHs cultured with serum from RIP TG and WT littermates fed a CD or an HFD and stimulated with insulin (10 nM) for the indicated times. (C and D) Exosomes were isolated from the islets of RIP TG or WT littermates fed a CD or an HFD and then transferred to MPHs. (C) AKT phosphorylation in MPHs given either WT or RIP TG serum exosomes and stimulated with insulin. (D) Levels of miR-26a in serum exosomes (upper panel) and MPHs (lower panel). Results are representative of 3 replicated independent experiments (B-D) and ImageJ quantification of the pAKT/AKT ratio is shown (B and C). The data underlying this figure may be found in S1 Data and S1 Raw Images. Data are shown as mean ± SD. *P < 0.05, ***P < 0.005, Student t test. AKT, AKT serine/threonine kinase; CD, chow diet; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HFD, high-fat diet; INS, insulin; MPHs, murine primary hepatocytes; pAKT, phosphorylated AKT; RIP, rat insulin promoter; TG, transgenic; WT, wild type.
Fig 4
Fig 4. Exosomal miR-26a derived from β cells regulates insulin signaling.
(A and B) Islets were isolated from either RIP TG or WT littermates fed an HFD for 16 weeks and cocultured with MPHs. (A) Levels of miR-26a in islets (left panel) and MPHs (right panel). (B) AKT phosphorylation in MPHs stimulated with insulin. (C and D) Exosomes were isolated from islets of RIP TG or WT littermates fed an HFD for 16 weeks and then transferred to MPHs. (C) Levels of miR-26a in islet exosomes (left panel) and MPHs (right panel). (D) AKT phosphorylation in MPHs stimulated with insulin. (E) Exosomes secreted from Min6 cells were labeled with PKH26 and then added to MPHs. (F–I) Exosomes were collected from media of Min6 cells transfected with miR-26a mimics (miR-26a) or NCs and then transferred to MPHs. (F) Levels of miR-26a in Min6-derived exosomes (left panel) and MPHs (right panel). (G) Levels of miR-26a in MPHs treated with exosomes from miR-26a–overexpressing Min6 cells that were treated with the exosome inhibitor GW4869 or controls. (H) AKT phosphorylation in MPHs stimulated with insulin. (I) Levels of miR-26a targets in MPHs. Results are representative of 3 replicated independent experiments (A–I), and ImageJ quantification of the pAKT/AKT ratio is shown (B, D, and H). The data underlying this figure may be found in S1 Data and S1 Raw Images. Data are shown as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.005, Student t test. Acsl3/4, acyl-CoA synthetase long chain family member 3/4; AKT, AKT serine/threonine kinase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; INS, insulin; Min6 cells, murine β cells; MPHs, murine primary hepatocytes; NC, negative control; pAKT, phosphorylated AKT; Pkcθ, protein kinase C theta; RIP, rat insulin promoter; TG, transgenic; WT, wild type.
Fig 5
Fig 5. β cell miR-26a preserves the functions of peripheral tissues.
(A–J) 6- to 8-week–old RIP TG and WT littermate controls were fed an HFD for 16 weeks. (A) Expression of miR-26a in obesity-associated tissues (n = 4–6). (B) Hepatic expression of pri- and pre-miR-26a (n = 4–6). (C) AKT phosphorylation in primary hepatocytes isolated from RIP TG mice and WT littermate controls and stimulated with insulin (10 nM) for the indicated times. Results are representative of 3 separate experiments. (D) Liver body ratio (n = 5–7). (E) Representative HE-stained and Oil-red O-stained liver (scale bar, 100 μm) (n = 4). (F) Hepatic triglyceride (n = 4–5). (G) Heat map of mRNA levels of hepatic genes involved in liver metabolism. Red and blue depict higher and lower gene expression, respectively. Color intensity indicates magnitude of expression differences. All listed genes were differentially expressed in two mouse groups with significance (P < 0.05) (n = 4–6). (H) BAT body ratio (n = 5–7). (I) Representative HE-stained BAT (scale bar, 50 μm) (n = 4). (J) Heat map of mRNA levels of adipogenesis, fatty acid metabolism, and BAT-selective and WAT-selective genes in BAT. Genes with differential expression in two mouse groups were highlighted in boldface (P < 0.05) (n = 4–6). (K) Metabolomic profiling of fatty acid metabolites in BATs with differential levels between two mouse groups (n = 4–6). (L) OCR for primary brown adipocytes isolated from WT and transfected with NCs or miR-26a mimics. Oligomycin, FCCP, and rotenone and antimycin were added at indicated time points. The data underlying this figure may be found in S1 Data and S1 Raw Images. Data are shown as mean ± SD. *P < 0.05, **P < 0.01, Student t test. Acc, acetyl-CoA carboxylase; Acsl, acyl-CoA synthetase long chain; Adipoq, adiponectin precursor; Adipor1, adiponectin receptor 1; Agt, angiotensinogen; AKT, AKT serine/threonine kinase; BAT, brown adipose tissue; Cd36, CD36 molecule; Cebpa, CCAAT/enhancer binding protein (C/EBP) alpha; Cidea, cell death inducing DFFA like effector A; Cox, cytochrome c oxidase; Cpt1a, carnitine palmitoyltransferase 1A; Egr1, early growth response protein 1; Fabp4, fatty acid binding protein 4, adipocyte; Fasn, fatty acid synthase; FCCP, carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone; Gsk3b, glycogen synthase kinase 3 beta; Gys2, glycogen synthase 2; HE, hematoxylin–eosin; HFD, high-fat diet; Ldlr, low-density lipoprotein receptor; NC, negative control; OCR, oxygen consumption rate; Pgc1a, peroxisome proliferator-activated receptor gamma coactivator 1-alpha; Pnpla2, patatin-like phospholipase domain-containing protein 2; Ppara, peroxisome proliferator activated receptor alpha; Pparg, peroxisome proliferator activated receptor gamma; Prdm16, PR domain containing 16; pre-miR-26a, precursor miR-26a; pri-miR-26a, primary miR-26a; Pygl, glycogen phosphorylase L; Retn, resistin; RIP, rat insulin promoter; SAT, subcutaneous adipose tissue; Scd1, stearoyl-CoA desaturase; Tcf7l2, transcription factor 7 like 2; TG, transgenic; Trim14, tripartite motif-containing 14; Ucp1, uncoupling protein 1; VAT, visceral adipose tissue; WAT, white adipose tissue; WT, wild type.
Fig 6
Fig 6. MiR-26a regulates insulin secretion through modulating actin remodeling.
(A) Static insulin secretion performed with islets from mice fed an HFD for 12 weeks at indicated glucose, arginine, and KCl concentrations (n = 3). (B) Time course of GSIS in islets from mice fed an HFD for 14 weeks (n = 4). (C–E) Proteomic analysis on islets of either RIP TG or WT littermate controls fed an HFD for 2 days. (C) Venn diagram comparing islet proteins. The number of proteins that are differentially expressed are shown. RIP TG mice showed an increase of 204 proteins and a decrease of 104 proteins. (D) KEGG pathway analysis on islet proteins deregulated in RIP TG mice. (E) Heat map of protein levels of genes involved in focal adhesion. (F) The protein levels of FLNA and CAV1 in the islets were verified by western blotting. (G and H) Representative IF imaging of F-actin in isolated islets treated with glucose (scale bar, 20 μm) (G). Quantification of F-actin intensity are shown (H) (n = 9 islets from 3 independent pancreatic samples per genotype). (I–L) Min6 cells were transfected with either miR-26a mimics (miR-26a) or NCs, followed by H2O2 treatment (n = 3). (I) Representative IF imaging of F-actin (scale bar, 20 μm). (J) Ratio of G-actin and F-actin fluorescence intensity. (K) Protein levels of F-actin and G-actin. (L) The activation kinetics of focal adhesion. The data underlying this figure may be found in S1 Data and S1 Raw Images. Data are shown as mean ± SD. *P < 0.05, Student t test. ACTN4, actinin alpha 4; AKT, AKT serine/threonine kinase; CAV1, caveolin 1; CDC42, cell division cycle 42; COL6A, collagen type VI alpha; ECM, extracellular matrix; ERK, extracellular signal-regulated kinase; FAK, focal adhesion kinase; FLNA, filamin alpha; F-actin, filamentous actin; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GSIS, glucose-stimulated insulin secretion; G-actin, globular actin; HFD, high-fat diet; IF, immunofluorescence; KEGG, Kyoto Encyclopedia of Genes and Genomes; LAMB2, laminin subunit beta 2; LAMC1, laminin subunit gamma 1; Min6 cells, murine β cells; MIP, maximum intensity projection; MYLK, myosin light chain kinase; NC, negative control; pAKT, phosphorylated AKT; p-ERK, phosphorylated ERK; p-FAK, phosphorylated FAK; PPAR, peroxisome proliferator activated receptor; RIP, rat insulin promoter; TG, transgenic; SOS, SOS Ras/Rac guanine nucleotide exchange factor; VCL, vinculin; WT, wild type.
Fig 7
Fig 7. MiR-26a inhibits obesity-induced hyperplasia and targets several genes critical for β cell proliferation and insulin secretion.
(A) Representative IF staining for insulin and glucagon in pancreatic islets (scale bar, 200 μm) (n = 4). (B) Representative HE-stained pancreas (scale bar, 200 μm) (n = 4). (C) Representative IHC staining for insulin in pancreatic islets (scale bar, 100 μm) (n = 4). (D) β cell mass (n = 4). (E) Islet area relative to pancreas area (n = 8–10). (F) Percentage of large islets (n = 5–8). (G) mRNA levels of the markers for β cell hyperplasia (n = 3). (H) Quantification of IF staining for insulin and PCNA in pancreas (n = 4). (I) mRNA levels of the markers for β cell replication (n = 3). (J) The top 50 enriched genes in the islet of WT mouse identified by Ago2 RNA immunoprecipitation sequencing are presented in a heat map (n = 3). Red and blue depict higher and lower gene enrichment, respectively. Color intensity indicates magnitude of enrichment differences. (K) Relative luciferase activity in 293T cells transfected with reporter constructs containing the 3′ UTR of target genes and co-transfected with either miR-26a mimics (miR-26a) or NCs. (L) Expression of miR-26a target genes in islets of RIP TG and WT littermates fed an HFD for 16 weeks (n = 3–4). The data underlying this figure may be found in S1 Data. Data are shown as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.005, Student t test. Adamts19, ADAM metallopeptidase with thrombospondin type 1 motif 19; Ago2, argonaute RISC catalytic component 2; AKT, AKT serine/threonine kinase; Amph, amphiphysin; Ap5m1, adaptor related protein complex 5 subunit Mu 1; Atad2b, ATPase family AAA domain containing 2B; Baz2b, bromodomain adjacent to zinc finger domain 2B; Brap, BRCA1 associated protein; Cacna1c, calcium voltage-gated channel subunit alpha1 C; Carmil1, capping protein regulator and myosin 1 linker 1; Ccnd2, cyclin D2; Cep350, centrosomal protein 350; Chordc1, cysteine and histidine rich domain containing 1; Col11a1, collagen type XI alpha 1 chain; Col22a1, collagen type XXII alpha 1 chain; Crebrf, CREB3 regulatory factor; Ctgf, connective tissue growth factor; Dab2, DAB adaptor protein 2; Dapk1, death associated protein kinase 1; DKO, double knockout; E2f7, E2F transcription factor 7; Epha7, EPH receptor A7; Esr1, estrogen receptor 1; Ext1, exostosin glycosyltransferase 1; Foxm1, forkhead box M 1; HE, hematoxylin–eosin; HFD, high-fat diet; Hnrnpul1, heterogeneous nuclear ribonucleoprotein U like 1; IF, immunofluorescence; Igf1r, insulin like growth factor 1 receptor; IHC, immunohistochemistry; Insr, insulin receptor; Irs2, insulin receptor substrate 2; Lman1, lecti mannose binding 1; Lrat, lecithin retinol acyltransferase; Mcc, MCC regulator of WNT signaling pathway; Megf9, multiple EGF like domains 9; Memo1, mediator of cell motility 1; Mki67, marker of proliferation Ki-67; Mtpn, myotrophin; Mut, mutant; NC, negative control; Nek10, NIMA related kinase 10; Neto1, neuropilin and tolloid like 1; Nras, neuroblastoma RAS viral oncogene; Nwd2, NACHT and WD repeat domain containing 2; Onecut2, one cut homeobox 2; Pcdh9, protocadherin 9; PCNA, proliferative cell nuclear antigen; Pdx1, pancreatic and duodenal homeobox 1; Pfkfb2, 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 2; Phf21a, PHD finger protein 21A; Pja2, praja ring finger ubiquitin ligase 2; Plcb1, phospholipase C beta 1; Prkag2, protein kinase AMP-activated non-catalytic subunit gamma 2; Prkg1, protein kinase CGMP-dependent 1; Pxylp1, 2-phosphoxylose phosphatase 1; Rb1, RB transcriptional corepressor 1; Rcn1, reticulocalbin 1; Ret, Ret proto-oncogene; Rhoq, Ras homolog family member Q; RIP, rat insulin promoter; Ryk, receptor like tyrosine kinase; Scml4, scm polycomb group protein like 4; Slc26a4, solute carrier family 26 member 4; Sox5, SRY-box transcription factor 5; Spock2, SPARC/Osteonectin, CWCV and Kazal like domains proteoglycan 2; Tab2, TGF-beta activated kinase 1 binding protein 2; Tacc2, transforming acidic coiled-coil containing protein 2; Tet2, tet methylcytosine dioxygenase 2; TG, transgenic; Tnks2, tankyrase 2; Trps1, transcriptional repressor GATA binding 1; Tsc22d2, TSC22 domain family member 2; Ube4b, ubiquitination factor E4B; Usp9x, ubiquitin specific peptidase 9 X-linked; WT, wild type.
Fig 8
Fig 8. Deficiency of miR-26a aggravates obesity-induced glucose intolerance and insulin resistance.
(A–J) Mice were fed an HFD beginning at 6–8 weeks of age. The following measurements were performed during the course of the HFD. (A) Total BW (n = 7). (B) GTT performed after 8 weeks of HFD (n = 78). (C) ITT performed after 9 weeks of HFD (n = 8). (D and E) Blood glucose (n = 8–12) (D) and insulin (n = 4–6) (E) levels of mice that were fed with either CD or HFD for 8 weeks. Random or fasting conditions are noted. (F) HOMA-IR (n = 5–6). (G) Representative IHC for insulin in pancreatic islets (scale bar, 200 μm) (n = 3). (H) Representative IF staining for insulin and glucagon in pancreatic islets (scale bar, 20 μm) (n = 3). (I) Quantification of IF staining for insulin and PCNA in pancreas (n = 4). (J) Expression of miR-26a target genes in islets (n = 3). (K–M) Exosomes were isolated from islets of 26a DKO mice or WT controls and then transferred into MPHs. (K) Levels of miR-26a in islet exosomes (n = 3). (L) AKT phosphorylation in hepatocytes stimulated with insulin. Results are representative of 3 replicated independent experiments and ImageJ quantification of the pAKT/AKT ratio is shown. (M) Levels of miR-26a and its target genes in recipient hepatocytes. (N) Metabolomic profiling of fatty acids in hepatocytes isolated from 26a DKO and WT mice fed an HFD for 3 days (n = 4). Red and blue depict higher and lower metabolites enrichment, respectively. Color intensity indicates magnitude of enrichment differences. The data underlying this figure may be found in S1 Data and S1 Raw Images. Data are shown as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.005, 2-tailed ANOVA (B–D) and Student t test (E–G, J–N). Acsl, acyl-CoA synthetase long chain; AKT, AKT serine/threonine kinase; BW, body weight; CD, chow diet; Ctgf, connective tissue growth factor; Dnmt3a, DNA methyltransferase 3 alpha; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GTT, glucose tolerance test; HFD, high-fat diet; HOMA-IR, homeostatic model assessment index of insulin resistance; IF, immunofluorescence; IHC, immunohistochemistry; Inhba, inhibin subunit beta A; ITT, insulin tolerance test; MPHs, murine primary hepatocytes; n.s., not significant; pAkt, phosphorylated AKT; PCNA, proliferative cell nuclear antigen; Rhoq, Ras homolog family member Q; Tcf7l2, transcription factor 7 like 2; WT, wild type; 26a DKO mice, miR-26a double knockout mice.

Similar articles

See all similar articles

References

    1. Zheng Y, Ley SH, Hu FB. Global aetiology and epidemiology of type 2 diabetes mellitus and its complications. Nat Rev Endocrinol. 2018;14(2):88–98. 10.1038/nrendo.2017.151 . - DOI - PubMed
    1. Tian Y, Peng B, Fu X. New ADCY3 Variants Dance in Obesity Etiology. Trends Endocrinol Metab. 2018;29(6):361–3. 10.1016/j.tem.2018.02.004 . - DOI - PubMed
    1. Samuel VT, Shulman GI. Nonalcoholic Fatty Liver Disease as a Nexus of Metabolic and Hepatic Diseases. Cell Metab. 2018;27(1):22–41. 10.1016/j.cmet.2017.08.002 . - DOI - PMC - PubMed
    1. Samuel VT, Shulman GI. The pathogenesis of insulin resistance: integrating signaling pathways and substrate flux. J Clin Invest. 2016;126(1):12–22. 10.1172/JCI77812 . - DOI - PMC - PubMed
    1. Ebner M, Lucic I, Leonard TA, Yudushkin I. PI(3,4,5)P3 Engagement Restricts Akt Activity to Cellular Membranes. Mol Cell. 2017;65(3):416–31 e6. 10.1016/j.molcel.2016.12.028 . - DOI - PubMed

Publication types

MeSH terms

Grant support

This work was supported by the Ministry of Science and Technology of China (2018ZX09201018-005 to XF), the National Natural Science Foundation of China (91540113, 81970561, and 81570527 to XF, and 81502631 to YT), the 1.3.5 Project for Disciplines of Excellence, West China Hospital, Sichuan University (ZYJC18049 to XF), and National Clinical Research Center for Geriatrics, West China Hospital, Sichuan University (Z20191005 to XF).
Feedback