Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 May 21;11(1):2538.
doi: 10.1038/s41467-020-15935-0.

Raptor determines β-cell identity and plasticity independent of hyperglycemia in mice

Affiliations

Raptor determines β-cell identity and plasticity independent of hyperglycemia in mice

Qinglei Yin et al. Nat Commun. .

Abstract

Compromised β-cell identity is emerging as an important contributor to β-cell failure in diabetes; however, the precise mechanism independent of hyperglycemia is under investigation. We have previously reported that mTORC1/Raptor regulates functional maturation in β-cells. In the present study, we find that diabetic β-cell specific Raptor-deficient mice (βRapKOGFP) show reduced β-cell mass, loss of β-cell identity and acquisition of α-cell features; which are not reversible upon glucose normalization. Deletion of Raptor directly impairs β-cell identity, mitochondrial metabolic coupling and protein synthetic activity, leading to β-cell failure. Moreover, loss of Raptor activates α-cell transcription factor MafB (via modulating C/EBPβ isoform ratio) and several α-cell enriched genes i.e. Etv1 and Tspan12, thus initiates β- to α-cell reprograming. The present findings highlight mTORC1 as a metabolic rheostat for stabilizing β-cell identity and repressing α-cell program at normoglycemic level, which might present therapeutic opportunities for treatment of diabetes.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Increased α/β cell ratio in βRapKOGFP mice.
a Heatmap of genes critical to endocrine cell in wild type (WT) and βRapKO islets (n = 4–5). b The number of Ins+ cells per islet (n = 3, p = 0.0008) and c the number of Gcg+ cells per islet were calculated (n = 3, p = 0.023). At least, 46 islets were used for quantifications. d The proportion of pancreatic α-cell mass among pancreatic α- and β-cell mass was shown (n = 4, p = 0.0017). e The proliferation of Gcg+ cell was determined by quantification of the percentage of Ki67+ cells in Gcg+ cells (n = 3). At least, 50 islets were used for quantifications. f Pancreatic sections from 8-week-old WT, βRapKOGFP mice were immunostained for insulin (white), glucagon (green), and GFP (red). Nuclei were stained with DAPI (blue) (n = 3). The yellow arrows indicated insulin, glucagon, and GFP co-stained cells. Scale bars, 20 μm. g The percentage of GFP+Gcg+ cells among GFP+ cells in 8-week-old WT and βRapKOGFP mice (n = 3, p = 1.83046E-05). At least, 1837 GFP+ cells over three mice each group were used for quantifications. h Representative pancreatic sections showed expression of α-cells marker—Arx (red), insulin (white), and glucagon (green) in WT and βRapKOGFP mice at 8 weeks of age (n = 3). Scale bars, 20 μm. i Immunostaining showed expression of α-cells marker—MafB (red), insulin (white), and glucagon (green) in control and βRapKOGFP mice at 8 weeks of age (n = 3). Yellow boxes showed the specific areas of the islet, which were enlarged and represented by arrows on the right to demonstrate protein expression within specific cells. Scale bars, 20 μm. j Transmission electron microscopy (TEM) of pancreatic islets reveals a combination of mature insulin, immature insulin, and glucagon-like granules in the same cell in βRapKOGFP mice (n = 3 independent samples). Granule identity is indicated by colored arrows: glucagon (red), mature insulin (blue), and immature insulin (yellow). Data represent means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 by two-sided Student’s t-test.
Fig. 2
Fig. 2. Impaired proinsulin synthesis and ATP levels for GSIS.
a Schematic of insulin pump implantation. b Random blood glucose levels were monitored every other day in WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP mice from 4 weeks to 8 weeks old (n = 6 for WT, n = 7 for diabetic KO, n = 13 for euglycemic KO). c Representative pancreatic sections immunostained for insulin (green) and glucagon (red). The ratio of Gcg+ cells to Ins+ cells was calculated (n = 3). At least 50 islets or 2000 insulin-positive cells were used for quantifications. Scale bars, 20 μm. d Images of islets labeled with insulin (green) and UCN3 (red). MFI of UCN3 in these three groups (n = 3). At least 10 islets from three sections were used for MFI. Yellow boxes showed the expression of UCN3. Scale bars, 20 μm. e In vitro glucose-stimulated C-peptide secretion in islets at 2.8 and 16.7 mM glucose levels for 1 h (reported as percent of C-peptide content) (n = 7 for WT, n = 6 for diabetic KO, n = 6 for euglycemic KO). White, black, and gray circles represented the WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP groups, respectively. Stimulation index of C-peptide secretion in diabetic βRapKOGFP and euglycemic βRapKOGFP mice as compared with controls (n = 7 for WT, n = 6 for diabetic KO, n = 6 for euglycemic KO). f ATP content at 2.8 mM glucose and 16.7 mM glucose (n = 3). g Pancreatic C-peptide content (n = 3 independent samples) and h pancreatic proinsulin content normalized by protein concentration were shown (n = 3 independent samples). i The ratio of pancreatic proinsulin to C-peptide content was determined (n = 3 independent samples). j The ratio of islet proinsulin to C-peptide content per 10 size-matched islets at 2.8 mM and high 16.7 mM glucose levels was determined (n = 6 for WT, n = 5 for diabetic KO, n = 5 for euglycemic KO). k INS-1 cells were transfected with SiRaptor or SiNC for 72 h. Immunoblotting and quantification of insulin and proinsulin in INS-1 cells transfected with SiNC or SiRaptor (n = 4 independent cell experiments). Data represent means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 by two-sided Student’s t-test and one-way ANOVA adjusted by LSD multiple comparison, p values included in source data.
Fig. 3
Fig. 3. Impaired metabolic coupling of insulin secretion.
a The number of differentially expressed genes within the three groups (n = 3–4). b Gene expression profiles regulated by Raptor were subjected to hierarchical clustering (n = 3–4). c GO analysis of differentially expressed genes as identified by RNA-seq of 8-week-old WT (n = 3) and euglycemic βRapKOGFP β-cells (n = 4) was associated with β-cell function. d Visualization of differential expression of genes involved in insulin secretion. e–g Relative expression of selected transcripts associated with β-cell secretion function (n = 5 independent sample for WT, n = 5 independent sample for diabetic βRapKOGFP, n = 4 independent sample for euglycemic βRapKOGFP, p values included in source data) (e), mitochondrial metabolism (n = 5 independent sample for WT, n = 5 independent sample for diabetic βRapKOGFP, n = 4 independent sample for euglycemic βRapKOGFP, p values included in source data) (f), and disallowed genes (n = 5 independent sample for WT, n = 5 independent sample for diabetic βRapKOGFP, n = 4 independent sample for euglycemic βRapKOGFP, p values included in source data) (g) in WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP β-cells by qRT-PCR. h Representative immunofluorescent staining for Glut2 (red) and insulin (green) among 8-week-old WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP mice (n = 3). Insets showed different expression levels of Glut2. Quantification of Glut2 areas in Ins+ areas in WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP mice (n = 3, WT vs diabetic βRapKOGFP: p = 0.002; diabetic βRapKOGFP vs euglycemic βRapKOGFP: p = 0.613; WT vs euglycemic βRapKOGFP: p = 0.001), at least 10 islets from three sections were used for quantifications of Glut2. Scale bars, 20 μm. Data represent means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 by one-way ANOVA.
Fig. 4
Fig. 4. Induction of α-cell-enriched genes in euglycemic βRapKOGFP.
a α-Cell-enriched genes and β-cell-enriched genes were subjected to perform principle component analysis based on the dataset from Qiu et al.. b Venn diagram representation of the subsets of Raptor-regulated genes that were enriched in α-cells. c Heatmap showed the differential expression of 57 overlapped genes in 8-week-old WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP β-cells (n = 3-4). d Relative expression of genes involved in cell identity by qRT-PCR (n = 5 independent sample for WT, n = 5 independent sample for diabetic βRapKOGFP, n = 4 independent sample for euglycemic βRapKOGFP, p values included in source data). e Representative immunofluorescent staining for GFP (red) and glucagon (green) among 8-week-old WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP mice (n = 4 for WT, n = 4 for diabetic βRapKOGFP, n = 3 for euglycemic βRapKOGFP). The yellow arrows indicated glucagon and GFP co-stained cells. Percentage of Gcg+GFP+ cells among GFP+ cells in WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP mice was determined (n = 4 independent sample for WT, n = 4 independent sample for diabetic βRapKOGFP, n = 3 independent sample for euglycemic βRapKOGFP, WT vs diabetic βRapKOGFP: p = 0.007; diabetic βRapKOGFP vs euglycemic βRapKOGFP: p = 0.012; WT vs euglycemic βRapKOGFP: p = 0.009). At least 50 islets were used for quantifications. Scale bars, 20 μm. f Representative immunofluorescent staining for ALDH1A3 (red) and insulin (green) among 8-week-old WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP mice (n = 3). Insets showed different expression levels of ALDH1A3. Percentage of ALDH1A3+Ins+ cells among Ins+ cells in WT, diabetic βRapKOGFP, and euglycemic βRapKOGFP mice was calculated (n = 3, WT vs diabetic βRapKOGFP: p < 0.001; diabetic βRapKOGFP vs euglycemic βRapKOGFP: p = 0.628; WT vs euglycemic βRapKOGFP: p < 0.001). For ALDH1A3 quantification, at least 36 islets were used. Scale bars, 20 μm. Data represent means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 by one-way ANOVA.
Fig. 5
Fig. 5. mTORC1 regulates α-cell transcriptional factor MafB.
a Relative expression of glucagon, Arx, and MafB in 2-week-old WT and βRapKOGFP β-cells by qRT-PCR (n = 4). b Quantification of Gcg+GFP+, Arx+GFP+, and MafB+GFP+ cells in GFP+ cells in 2-week-old WT and βRapKOGFP mice (n = 3, Gcg: p = 0.58; Arx: p = 0.86; MafB: p = 9.00122E-07). At least 1783 GFP+ cells were used for quantifications. c Representative immunofluorescent staining for MafB (green) and GFP (red) in 2-week-old WT and βRapKOGFP mice (n = 3). Yellow box showed the specific area of the islet, which was enlarged and represented by arrows to demonstrate MafB+GFP+ cell. Scale bars, 20 μm. d Relative expression of selected transcripts associated with α-cell-enriched genes in INS-1 cells which were transfected with SiNC or SiRaptor in the presence or absence of SiMafB by qRT-PCR (n = 4 independent cell experiments, p values included in source data). e, f Western blotting showed the expression of C/EBPβ(LAP) in 8-week-old WT and βRapKOGFP mice (n = 3, p = 0.049). g, h INS-1 cells were treated with Rapamycin (25 nM), C/EBPβ(LAP) protein expression was assayed by immunoblot (n = 3 independent cell experiments). i Luciferase reporter gene assays revealed that Rapamycin treatment for 12 h positively regulated the luciferase activity of MafB (n = 5 independent cell experiments, p = 0.0011). j Immunoblotting to evaluate the LAP and LIP protein levels after LAP and LIP overexpression (n = 2 independent cell experiments). k Luciferase reporter assay using a rat MafB promoter reporter. INS-1 cells were transfected with GFP or LAP overexpression vector (n = 3 independent cell experiments, p = 0.001). l qRT-PCR analysis of MafB expression in GFP and LAP overexpressed INS-1 cells (n = 4 independent cell experiments, p = 0.0023). m Luciferase reporter assay using a rat MafB promoter reporter. INS-1 cells were transfected with GFP or LIP overexpression vector in the absence and presence of Rapamycin (n = 3 independent cell experiments, GFP + Vehicle vs GFP + Rapamycin, p < 0.001; LIP + Vehicle vs LIP + Rapamycin, p = 0.001). n qRT-PCR analysis of MafB expression in GFP and LIP overexpressed INS-1 cells in the absence or presence of Rapamycin (n = 4 independent cell experiments, GFP + Vehicle vs GFP + Rapamycin, p = 0.002; LIP + Vehicle vs LIP + Rapamycin, p = 0.177). Data represent means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 by two-sided Student’s t-test and one-way ANOVA.
Fig. 6
Fig. 6. Induction of α-cell-enriched genes in 2-week-old βRapKOGFP.
a GO analysis of differentially expressed genes as identified by microarray of 2-week-old WT (n = 3) and βRapKOGFP β-cells (n = 4) was associated with β-cell function and metabolism. b Venn diagram representation of the subsets of Raptor-regulated genes in 2-week-old purified β-cells that were enriched in α-cells and heatmap showed the differential expression of nine overlapped genes in 2-week-old WT and βRapKOGFP β-cells (n = 3–4). c Analysis strategy to identify Raptor-dependent genes in β-cells is shown. d Heatmap of seven Raptor-dependent genes obtained from 8-week-old RNA-seq and 2-week-old microarray. e Volcano plot shows differential genes between 2-week-old WT and βRapKOGFP β-cells. Microarray identification of Etv1 and Tspan12 as significantly upregulated α-cell-enriched genes in 2-week-old βRapKOGFP β-cells. f INS-1 cells were transfected with SiRaptor or SiNC for 48 h. qRT-PCR confirmed Raptor-dependent genes in INS-1 cells (n = 4 independent cell experiments, for Raptor, p = 0.029; for Slc2a2, p = 0.048; for Msln, p = 0.031; for Ppp1r1a, p = 0.28; for Etv1, p = 0.004; for Tspan12, p = 0.008; for Aass, p = 0.02). g, h INS-1 cells were transfected with GFP, Etv1, or Tspan12 overexpression vector. g Luciferase reporter assay using a rat glucagon promoter reporter (n = 3 independent cell experiments, p = 0.039). h qRT-PCR analysis of glucagon expression in vector transfected INS-1 cells (n = 4 independent cell experiments, p = 0.02). Data represent means ± SEM. *p < 0.05, **p < 0.01 by two-sided Student’s t-test.
Fig. 7
Fig. 7. The role of Raptor in β-cell identity maintenance and α-cell gene repression.
Raptor is required for maintaining β-cell identity as well as repressing α-cell signature. At euglycemia, loss of Raptor results in β-cell dedifferentiation (marked as UCN3low, Glut2low, and Aldh1a3high) with compromised β-cell identity, metabolic uncoupling, and regression to multi-hormonal state with α-cell features. We propose Raptor/C/EBPβ/MafB-dependent and MafB-independent ways in silencing α-cell-enriched genes and glucagon expression in healthy β-cells. Raptor-deficient β-cells with compromised identity become dysfunctional and eventually cause diabetes; in turn, hyperglycemia further aggravates dedifferentiation and reprogramming process.

Similar articles

Cited by

References

    1. Weir GC, Bonner-Weir S. Five of stages of evolving β-cell dysfunction during progression to diabetes. Diabetes. 2004;53:S16–S21. - PubMed
    1. Accili D, et al. When β-cells fail: lessons from dedifferentiation. Diabetes Obes. Metab. 2016;18:117–122. - PMC - PubMed
    1. Remedi MS, Emfinger C. Pancreatic β-cell identity in diabetes. Diabetes Obes. Metab. 2016;18:110–116. - PMC - PubMed
    1. Brereton MF, Rohm M, Ashcroft FM. β-Cell dysfunction in diabetes: a crisis of identity? Diabetes Obes. Metab. 2016;18:102–109. - PMC - PubMed
    1. Talchai C, Xuan S, Lin HV, Sussel L, Accili D. Pancreatic β cell dedifferentiation as a mechanism of diabetic β cell failure. Cell. 2012;150:1223–1234. - PMC - PubMed

Publication types

MeSH terms