Abstract
The ubiquitously expressed G protein α-subunit Gsα mediates the intracellular cAMP response to glucagon-like peptide 1 (GLP1) and other incretin hormones in pancreatic islet cells. We have shown previously that mice with β-cell-specific Gsα deficiency (βGsKO) develop severe early-onset insulin-deficient diabetes with a severe defect in β-cell proliferation. We have now generated mice with Gsα deficiency throughout the whole pancreas by mating Gsα-floxed mice with Pdx1-cre transgenic mice (PGsKO). PGsKO mice also developed severe insulin-deficient diabetes at a young age, confirming the important role of Gsα signaling in β-cell growth and function. Unlike in βGsKO mice, islets in PGsKO mice had a relatively greater proportion of α-cells, which were spread throughout the interior of the islet. Similar findings were observed in mice with pancreatic islet cell-specific Gsα deficiency using a neurogenin 3 promoter-cre recombinase transgenic mouse line. Studies in the α-cell line αTC1 confirmed that reduced cAMP signaling increased cell proliferation while increasing cAMP produced the opposite effect. Therefore, it appears that Gsα/cAMP signaling has opposite effects on pancreatic α- and β-cell proliferation, and that impaired GLP1 action in α- and β-cells via Gsα signaling may be an important contributor to the reciprocal effects on insulin and glucagon observed in type 2 diabetics. In addition, PGsKO mice show morphological changes in exocrine pancreas and evidence for malnutrition and dehydration, indicating an important role for Gsα in the exocrine pancreas as well.
Introduction
Both type 1 and type 2 diabetes have been associated with pancreatic β-cell dysfunction, with reduced glucose-stimulated insulin secretion and decreased β-cell mass due to impaired β-cell proliferation and survival (Butler et al. 2003). In addition, type 2 diabetics also have increased glucagon (GCG) secretion from pancreatic α-cells and inappropriately increased serum levels of GCG (Goke 2008). Glucagon-like peptide 1 (GLP1) and other incretin hormones promote glucose-stimulated insulin secretion and β-cell growth, and inhibit GCG secretion from α-cells by activating the G protein α-subunit Gsα, which stimulates intracellular cAMP production (Weinstein et al. 2001, Brubaker & Drucker 2004, Preitner et al. 2004, Ahren 2009). Because of these beneficial effects, the GLP1 receptor (Glp1r) agonist exendin 4 has recently been approved as a therapeutic agent for diabetes.
We have previously shown in mice with β-cell-specific Gsα deficiency (βGsKO mice) that Gsα/cAMP signaling is critical for normal β-cell function and growth (Xie et al. 2007). Although GLP1 tends to inhibit GCG secretion in vivo, these effects may not be due to direct effects of GLP1 on GCG secretion from α-cells as several studies have shown that Gsα/cAMP signaling stimulates both proglucagon gene expression (Drucker et al. 1991, Islam et al. 2009) and GCG release (Hermansen 1985, Ding et al. 1997, Gromada et al. 1997, Ma et al. 2005) in isolated α-cells. Little is known about the role of GLP1 or Gsα/cAMP signaling on the regulation of pancreatic α-cell proliferation.
In this present study, we generated mice with Gsα deficiency throughout the endocrine and exocrine pancreas (PGsKO mice) by mating Gsα-floxed mice with Pdx1-cre mice in order to directly examine the in vivo role of Gsα/cAMP signaling in both pancreatic α- and β-cells. These mice had the same β-cell defect with early-onset insulin-deficient diabetes as that seen in βGsKO mice. In addition, PGsKO had relatively increased numbers of α-cells, and a similar finding was observed in a similar mouse model that was generated using neurogenin 3 (Ngn3) promoter-cre mice. Studies in the α-cell line αTC1 showed that Gsα/cAMP signaling inhibits α-cell proliferation, an effect opposite to that known to occur in β-cells. Therefore, reduced GLP1 actions on both α- and β-cells via Gsα signaling may be an important contributor to the reciprocal effects on insulin and GCG observed in type 2 diabetics. In addition, we show evidence that Gsα is also important in pancreatic exocrine function.
Materials and Methods
Mice
Mice with loxP sites surrounding Gsα exon 1 (E1fl/fl; Chen et al. 2005) were maintained on Black Swiss genetic background. Pdx1-cre+ mice (Lammert et al. 2003) and Ngn3 promoter-cre+ mice (Schonhoff et al. 2004) were crossed with Black Swiss WT mice for three or more generations before the studies in order to obtain a purer genetic background. Female E1fl/fl mice were mated with male E1fl/+:Pdx1-cre+/− mice to generate pancreatic Gsα knockout mice (PGsKO; E1fl/fl:cre+), mice with heterozygous pancreatic Gsα deletion (E1fl/+:cre+), and control cre− littermates, or were mated with Elfl/+:Ngn3-cre+/− mice to generate mice with pancreatic islet-specific Gsα deficiency. Mice were maintained on standard pellet diet (NIH-07, 5% fat by weight) and 12 h light:12 h darkness cycle. Except where noted, studies were performed in 3- to 4-month-old mice, and were approved by the NIDDK Animal Care and Use Committee. Mice were genotyped by PCR as previously described (Sakamoto et al. 2005).
Body composition and food intake
Body composition was measured using the Minispec mq10 NMR analyzer (Bruker Optics Inc., Woodlands, TX, USA). Food intake was measured over 21 days in individually caged mice.
Serum chemistry
Blood was obtained by retro-orbital bleed. Serum insulin (Crystal Chem, Downers Grove, IL, USA), GCG (R&D Systems, Minneapolis, MN, USA), insulin-like growth factor 1 (IGF1; Alpco Diagnostics, Salem, NH, USA), and the circulating active forms of GLP1 (Millipore, Billerica, MA, USA) were measured by ELISA or RIA. The remainder of the chemistries was measured in the NIH Clinical Chemistry Laboratory.
Glucose and insulin tolerance tests and insulin secretion
Glucose and insulin tolerance tests were performed in overnight fasted mice after i.p. injection of glucose (2 mg/g) or insulin (Humulin; 0.75 mIU/g) respectively. Plasma glucose was measured by glucometer.
Immunostaining and islet morphometry
Pancreata were fixed in 4% paraformadehyde and paraffin-embedded, and sections were H&E stained. For immunostaining, 5-μm sections of paraffin blocks were deparaffinized and rehydrated with xylene followed by decreasing concentrations of ethanol; microwaved in 0.01 M sodium citrate, pH 6.0, for 20 min; and permeabilized with 1% Triton X-100 in PBS. Sections were then incubated with mouse anti-insulin (Invitrogen), rabbit anti-GCG, rat anti-somatostatin (Lab Vision, Fremont, CA, USA), rabbit anti-prohormone convertase 1/3 (PC1/3), rabbit anti-PC2 (Millipore), or rabbit anti-Gsα (Simonds et al. 1989) antibodies, followed by secondary antibodies that were labeled with rhodamine or Cy2 (Jackson ImmunoResearch, West Grove, PA, USA). Sections were counterstained with DAPI (Invitrogen). For α-cell proliferation studies, pancreatic sections of 7-day-old mice (n=3/group) were stained with anti-GCG and rabbit anti-Ki67 (Vector, Burlingame, CA, USA) antibodies.
Cell proliferation and GCG secretion in culture
A pancreatic α-cell line (αTC1.9; Powers et al. 1990) was purchased from ATCC (Manassas, VA, USA; catalog # CRL-2350) and grown in 37 °C under 10% CO2 in DMEM modified to contain 0.3% glucose, 15 mM HEPES, 0.1 mM nonessential amino acids, 0.02% BSA, and 10% fetal bovine serum. For RNAi Gsα knockdown assay, Gsα-specific siRNA and negative control siRNA were purchased from Ambion (Austin, TX, USA; catalog #16706, Gene ID46627), and lipofectamine 2000 (Invitrogen) was used for transfection. To test the effect of increased intracellular cAMP on αTC1 growth, cells were seeded into 24-well plates at 2×105 cell/well and incubated overnight before medium containing either forskolin (FSK; 20 mM) and isobutylmethylxanthine (IBMX; 100 mM), or vehicle alone was added and then refreshed every 24 h. Treated and control cells were stained with 0.1% thiazolyl blue tetrazolium bromide (Sigma) in PBS at 37 °C for 30 min. Stained cells were incubated with isopropanol to elute the stain, and absorbances were read at OD560. To test the effects on GCG secretion, media were changed at 24 h after transfection or treatment, and replaced with new media containing 16.7 or 5.6 mM glucose. The cells were then incubated for 4 h, and the GCG concentrations in the culture media were measured by ELISA (R&D Systems).
Quantitative RT-PCR
Islets were isolated and RNA samples were prepared as previously described (Xie et al. 2007), and quantitative real-time PCR was performed using SYBR-Green based detection in an Mx3000P (Stratagene, La Jolla, CA, USA). Gene expressions were normalized by geometric averaging of the levels of β-actin, Hprt, and Gapdh mRNAs. Primer sequences are available upon request.
Statistical analysis
Data are expressed as mean±s.e.m. Statistical significance was determined by unpaired Student's t-test (two-tailed) or one-way ANOVA with Tukey's post hoc test with differences considered significant at P<0.05.
Results
PGsKO mice have reduced body mass and adiposity
PGsKO (E1fl/fl:Pdx1-cre+) mice with pancreatic Gsα deficiency were generated by mating of E1fl/fl females with E1fl/+:Pdx1-cre+/− males, and their phenotypes were compared with cre− control and E1fl/+:Pdx1-cre+ littermates. Coimmunostaining of pancreatic sections with a Gsα antibody and either an insulin or GCG antibody showed robust Gsα expression in both α- and β-cells of control islets which was markedly reduced in PGsKO islets (Fig. 1, panel A). This is in contrast to βGsKO mice, in which Gsα deficiency was confined to β-cells (Xie et al. 2007). Compared with islets, Gsα expression in the surrounding exocrine pancreas was much lower even in control mice (Fig. 1, panel A). PGsKO mice showed no change in Gsα expression in the pituitary or hypothalamus (data not shown). Unlike βGsKO mice, which had significant lethality throughout the first few months of postnatal life (Xie et al. 2007), PGsKO mice showed minimal effects on postnatal survival (102 and 89% of expected survival at 7 days and 4 weeks of age respectively; n=82 and 282 total offspring respectively).
PGsKO mice have lower body weight in spite of being hyperphagic. (A) Immunostaining of a control and PGsKO islet for either insulin (red) or Gsα (green) (top panels) or for either GCG (red) or Gsα (green) (bottom panels). In both cases, the merged images are shown on the right. Body weight of control, E1fl/+:Pdx1-cre+, and PGsKO mice at (B) 7 days (males only, n=21–36/group) (C) 3.5 weeks (males and females, n=3–8/group), and (D) 3 months (males only, n=6–10/group) after birth. (E) Body length (males and females, n=3–10/group), (F) body mass index (males, n=4–10/group), (G) % fat mass (males, n=5–13/group), and (H) food intake (males, n=4–8/group) in 3-month-old mice. Data are mean±s.e.m. (*P<0.05 or **P<0.01 versus controls. #P<0.05 or ##P<0.01 versus E1fl/+:Pdx1-cre+). Bars equal 50 μm.
Citation: Journal of Endocrinology 206, 3; 10.1677/JOE-10-0030
PGsKO mice had a severe postnatal growth defect. Body weights of control, E1fl/+:Pdx1-cre+, and PGsKO mice were similar on postnatal day 7 (Fig. 1, panel B) but by 3.5 weeks both male and female PGsKO had markedly reduced body weight compared with either Elfl/+:Pdx-cre+ mice or controls (Fig. 1, panel C), and this difference became even greater by 3 months of age (Fig. 1, panel D). Although adult PGsKO mice had slightly reduced body length compared with E1fl/+:Pdx1-cre+ mice and controls (Fig. 1, panel E), their overall body mass index (body weight/body length2) was significantly reduced (Fig. 1, panel F). Body composition analysis showed PGsKO mice to have markedly reduced fat mass (Fig. 1, panel G). While impaired growth in βGsKO mice was associated with reduced circulating levels of IGF1 (Xie et al. 2007), IGF1 levels were similar in PGsKO, E1fl/+:Pdx1-cre+, and control mice (Table 1). In spite of their impaired growth, these mice had significantly increased food intake (Fig. 1, panel H). One possible explanation for the poor growth in PGsKO mice may be malabsorption due to pancreatic exocrine dysfunction, as PGsKO mice showed markedly abnormal histology of the exocrine pancreas (see below) presumably as a result of Gsα deficiency in these regions as well as the endocrine pancreas. Consistent with severe malnutrition, serum albumin levels were significantly decreased in PGsKO mice (Table 1).
Serum chemistries in adult micea
| Control | E1fl/+:Pdx-cre+ | PGsKO | |
|---|---|---|---|
| Glucose (mg/dl) | 184±8 | 238±11** | 602±2**,‡ |
| Cholesterol (mg/dl) | 129±6 | 160±14 | 206±15**,† |
| Triglycerides (mg/dl) | 277±60 | 320±49 | 523±73* |
| Sodium (mEq/l) | 153±1 | 152±3 | 147±4 |
| Potassium (mEq/l) | 5.9±0.2 | 5.8±0.3 | 6.6±0.3 |
| Chloride (mEq/l) | 117±1 | 113±2 | 109±4 |
| Creatinine (mg/dl) | 0.40±0.03 | 0.56±0.09 | 0.90±0.13** |
| BUN (mg/dl) | 24±2 | 22±1 | 49±16 |
| Albumin (g/dl) | 1.52±0.04 | 1.48±0.02 | 1.22±0.07**,‡ |
| IGF1 (ng/ml) | 641±39 | 682±25 | 673±19 |
| GCG (pg/ml) | 155±30 | 160±46 | 185±61 |
| Insulin (ng/ml) | 1.23±0.16 | 1.02±0.14 | 0.35±0.08**,† |
| GLP1 (active) pM | 6.63±0.09 | 6.42±0.23 | 6.40±0.14 |
**P<0.01 versus control; †P<0.05 or ‡P<0.01 versus E1fl/+:Pdx1-cre+ by ANOVA.
Measured in 3–4-month-old males, n=3–12/group.
PGsKO mice develop early-onset insulin-deficient diabetes
Even by day 7 post partum, PGsKO mice were very hyperglycemic as compared with E1fl/+:Pdx1-cre+ and control mice (Fig. 2, panel A). Serum insulin levels on day 5 were ∼50% lower in PGsKO as compared with the other two groups (P=0.08 versus controls; Fig. 2, panel B). In adult mice (3–4 months of age), glucose levels were slightly higher in E1fl/+:Pdx1-cre+ mice as compared with controls, while they were greater than threefold higher in PGsKO mice (Table 1). Insulin levels were markedly reduced in adult PGsKO mice as compared with controls and E1fl/+:Pdx1-cre+ mice (Table 1). Adult male and female PGsKO mice had markedly impaired glucose tolerance compared with E1fl/+:Pdx1-cre+ and control mice (Fig. 3, panels A and B), while insulin tolerance tests showed similar levels of insulin sensitivity between the three groups (Fig. 3, panels C and D). In addition, adult PGsKO mice had increased serum triglyceride and cholesterol levels as compared with the other two groups of mice (Table 1). Serum IGF1, GLP1 (active forms), and GCG levels were similar between the three groups (Table 1). While electrolytes were normal, serum creatinine was significantly increased and blood urea nitrogen (BUN) levels were more than double in PGsKO mice, although this was not statistically significant (Table 1), indicating that PGsKO mice were dehydrated, most likely secondary to severe hyperglycemia. Therefore, PGsKO mice mimic βGsKO mice in terms of developing very early-onset insulin-deficient diabetes, although βGsKO mice had low IGF1 levels (Xie et al. 2007) while IGF1 remained unaffected in PGsKO mice.
Mice develop insulin deficiency and hyperglycemia at a young age. (A) Serum glucose (n=21–38/group) and (B) insulin (n=3–7/group) in control, E1fl/+:Pdx1-cre+, and PGsKO mice at day 7 and 5 post partum respectively. Data are mean±s.e.m. (**P<0.01 versus controls; ##P<0.01 versus E1fl/+:Pdx1-cre+).
Citation: Journal of Endocrinology 206, 3; 10.1677/JOE-10-0030
PGsKO mice are glucose intolerant but have normal insulin sensitivity. Glucose tolerance tests performed in (A) male (n=4–5/group) and (B) female 3-month-old control (□), E1fl/+:Pdx1-cre+(▾), and PGsKO mice (▪) (n=4–7/group). Insulin tolerance tests performed in (C) male (n=6–8/group) and (D) female mice (n=4–6/group). Data are mean±s.e.m. (**P<0.01 versus controls; ##P<0.01 versus E1fl/+:Pdx1-cre+).
Citation: Journal of Endocrinology 206, 3; 10.1677/JOE-10-0030
PGsKO mice have reduced islet size and β-cell mass with increased numbers of α-cells
Total pancreas weights of PGsKO mice were greater than controls (Fig. 4, panel A). When pancreas weights were normalized to total body weights, they were ∼75% greater in PGsKO mice as compared with E1fl/+:Pdx1-cre+ or control mice (Fig. 4, panel A). This was primarily due to marked enlargement of the pancreatic exocrine ducts in PGsKO mice, which were also more eosinophilic as compared with those in controls on H&E staining (Fig. 4, panel B). These changes are probably the result of Gsα deficiency in exocrine pancreas, and suggest that PGsKO mice may have pancreatic exocrine insufficiency and malabsorption.
Pancreatic exocrine and islet pathology in PGsKO mice. (A) Pancreatic weights in mg. (top panel; n=7–15/group) and as % body weight (lower panel; n=4–7/group) in adult male control (open bars), E1fl/+:Pdx1-cre+ (striped bars), and PGsKO mice (solid bars). (B) H&E-stained sections (two each) of an islet and surrounding exocrine pancreas from a control and a PGsKO mouse. (C) Immunostaining of pancreatic sections from 5-day-old, 4-week-old, and 14-week-old control and PGsKO mice for insulin (red) and GCG (green). (D) Two sections showing islets from E1fl/fl:Ngn3-cre+ mice that were immunostained for insulin (red) or GCG (green) with the merged images on the right. (E) Sections of a control and a PGsKO islet immunostained for insulin (red) and somatostatin (green). (F) mRNA expression levels of the Gsα, glucagon (Gcg), GLP1 receptor (Glp1r), glucagon receptor (Gcgr), and insulin (Ins2) genes in islets isolated from control and PGsKO mice (n=7/group). Data are mean±s.e.m. (**P<0.01 versus controls; ##P<0.01 versus E1fl/+:Pdx1-cre+). Bars equal 50 μm.
Citation: Journal of Endocrinology 206, 3; 10.1677/JOE-10-0030
H&E staining of adult pancreas showed that PGsKO mice, like βGsKO mice (Xie et al. 2007), had markedly smaller islets than controls with irregular shapes and increased cellularity (Fig. 4, panel B). Coimmunostaining of islets with antibodies for insulin and GCG showed that at 5 days of age, PGsKO islets were only slightly smaller than controls with a similar distribution of α- and β-cells (Fig. 4, panel C). By 4 weeks, PGsKO islets had a greater ratio of α- to β-cells than controls, although the α-cells remained on the periphery. At 14 weeks, PGsKO islets were much smaller than in controls, and had less β-cells with reduced intensity of insulin staining, similar to what was observed in βGsKO mice (Xie et al. 2007). However, unlike what was observed in βGsKO mice (Xie et al. 2007), PGsKO mice had increased numbers of α-cells with a very significant number of α-cells mixed with β-cells throughout the islet (Fig. 4, panel C). A similar pattern was also observed in E1fl/fl mice:Ngn3-cre+ (Schonhoff et al. 2004), another model with Gsα deficiency throughout the pancreatic islet (Fig. 4, panel D). Coimmunostaining of PGsKO islets with insulin and somatostatin antibodies showed reduced numbers of β-cells, and a slight increase in somatostatin-staining δ-cells with some intermingling of δ-cells with β-cells within the interior of the islet (Fig. 4, panel E). Immunostaining of duodenal sections for serotonin, GLP1, cholecystokinin, or ghrelin showed no obvious change in the number or distribution of endocrine cells in the duodena of PGsKO mice (data not shown).
The ductal enlargement observed in PGsKO mice suggests that the relative greater proportion of islet α-cells may be due to increased α-cell neogenesis. During embryonic development, fetal α-cells transiently express the prohormone convertase PC1/3 which converts proglucagon to GLP1 (Wilson et al. 2002), whereas adult α-cells express PC2 which converts proglucagon to GCG. Both PC1/3 and PC2 are expressed in islet β-cells, and are required for proinsulin processing. To determine whether α-cell neogenesis is increased in PGsKO mice, we looked for the presence of α-cells expressing PC1/3 by coimmunostaining islets for GCG and either PC1/3 or PC2 (Fig. 5, panels A and B respectively). We found that control and PGSKO islets had similar distributions of PC1/3 and PC2, with the absence of PC1/3 and the presence of PC2 within GCG-producing α-cells, and the presence of both PC1/3 and PC2 in the nonglucagon staining cells, which are primarily β-cells. Therefore, we found no evidence for increased α-cell neogenesis in PGsKO mice.
Prohormone convertase 1/3 (PC1/3) is absent in α-cells from control and PGsKO mice. Immunostaining of a control and PGsKO islet for (A) either GCG (red) or PC1/3 (green) or for (B) either GCG (red) or PC2 (green). In both cases, the merged images are shown on the right. Bars equal 20 μm.
Citation: Journal of Endocrinology 206, 3; 10.1677/JOE-10-0030
Gene expression studies in control and PGsKO islets by quantitative RT-PCR (Fig. 4, panel F) showed a significant reduction in Gsα mRNA in PGsKO islets, as expected. Insulin (Ins2) mRNA was also markedly reduced, consistent with the marked reduction in β-cells. Expression of the proglucagon gene (Gcg) which generates GCG was similar in control and PGsKO islets. Given that PGsKO islets have a greater proportion of α-cells, this suggests that expression levels per α-cell may be decreased, although we were unable to detect a change in Gcg mRNA levels in cultured α (αTC1)-cells transfected with Gsα RNAi (data not shown). Glp1r gene expression was also very low in PGsKO islets. This is probably mostly due to the reduced proportion of β-cells, as these receptors were recently shown to be exclusively expressed within β-cells in mouse islets (Tornehave et al. 2008). In addition, Glp1r expression in islets has been shown to be inhibited by hyperglycemia, but to be unaffected by Gsα/cAMP signaling (Abrahamsen & Nishimura 1995). Finally, GCG receptor (Gcgr) expression was up-regulated in PGsKO islets. This is consistent with prior studies showing that Gsα/cAMP signaling downregulates while high extracellular glucose levels upregulate this gene in pancreatic islets (Abrahamsen & Nishimura 1995) and cultured hepatocytes (Abrahamsen et al. 1995).
Gsα deficiency leads to increased proliferation of pancreatic α-cells
We have previously shown that Gsα deficiency in β-cells leads to markedly reduced β-cell proliferation and reduced β-cell mass (Xie et al. 2007), and our findings in PGsKO mice are consistent with these prior observations. However, in PGsKO mice, which also had Gsα deficiency in α-cells, costaining of islets from 7-day-old mice with antibodies to GCG and the proliferative marker Ki67 showed no evidence for a proliferative defect in α-cells, and in fact, the proportion of proliferating α-cells in PGsKO mice tended to be slightly higher than in controls, although the differences were not statistically significant (Fig. 6, panels A and B).
Gsα deficiency leads to increased pancreatic α-cell proliferation. (A) Staining of a control and PGsKO islet for GCG (red) and Ki-67 (green). (B) Percentage of GCG-staining cells that are also Ki-67+ in islets examined from three control and three PGsKO mice. (C) Proliferation of αTC1 cells after transfection with either control (scrambled) or Gsα RNAi (n=6/group). (D) Proliferation of αTC1 cells after treatment with either forskolin plus IBMX (FSK–IBMX) or with vehicle alone (n=12/group). For panels C and D, cell counts are expressed as % of cell number at time 0. (E) GCG secreted into media of αTC1 cells from hours 24–28 after transfection with either control (scrambled) or Gsα RNAi in the presence of 16.7 or 5.6 mM glucose (n=6/group). (F) GCG secreted into media of αTC1 cells from hours 24–28 after treatment with either forskolin plus IBMX (FSK–IBMX) or with vehicle alone in the presence of 5.6 mM glucose (n=6/group). Data are mean±s.e.m. (**P<0.01 versus control or vehicle). Bars equal 50 μm.
Citation: Journal of Endocrinology 206, 3; 10.1677/JOE-10-0030
We further examined the effects of Gsα signaling on proliferation in the α-cell line αTC1. αTC1 cells were treated with either a control (scrambled) RNAi or a Gsα RNAi which reduced Gsα expression by ∼70% as determined by quantitative RT-PCR (data not shown). Cells treated with the Gsα RNAi had a markedly higher proliferative rate than the control cells with the Gsα RNAi-treated cells doubling at 72 h while the control cell number only increased by ∼30% during the same time period (Fig. 6, panel C). We also performed the opposite experiment, in which αTC1 cells were treated with the adenylyl cyclase stimulator FSK and the cAMP phosphodiesterase inhibitor IBMX, which lead to increased intracellular cAMP levels (FSK–IBMX). In contrast to the cells treated with Gsα RNAi, proliferation of FSK–IBMX-treated cells was significantly lower than controls (Fig. 6, panel D). These in vitro results confirm that Gsα signaling pathways have an antiproliferative effect in α-cells, which is opposite to their known proproliferative effect in β-cells.
We next examined the effect of perturbing Gsα/cAMP signaling on GCG secretion in αTC1 cells. Cells were transfected with either control or Gsα RNAi, and GCG secretion into the media was measured from 24 to 28 h later at both 5.6 and 16.7 mM glucose to avoid the potential confounding effects of differing glucose concentrations on GCG secretion. While GCG secretion was increased after Gsα RNAi transfections at both glucose concentrations (Fig. 6, panel E), the percent increase was similar to the percent increase in αTC1 cell numbers at 24 h after Gsα RNAi treatment (Fig. 6, panel C), and therefore, this does not represent an increase in the amount of GCG secreted per cell. Interestingly, αTC1 cells treated with FSK–IBMX showed a more marked twofold increase in GCG secretion at 24–28 h in the presence of 5.6 mM glucose (Fig. 6, panel F), despite the fact that there was no increase in cell numbers (Fig. 6, panel D), indicating that excess Gsα/cAMP signaling directly promotes GCG secretion in α-cells.
Discussion
Previously, we generated mice with β-cell-specific Gsα deficiency (βGsKO mice) using Rip2-cre, and demonstrated that Gsα/cAMP pathways are critical regulators of β-cell function and promote β-cell proliferation (Xie et al. 2007). However, the βGsKO model is limited by the fact that Gsα deficiency occurs in other tissues such as the hypothalamus and pituitary due to the expression of the Rip2 gene in these tissues. In this work, we extended the study of Gsα function in pancreatic cells by deleting Gsα from all major types of pancreatic endocrine and exocrine cells using Pdx1-cre transgenic mice (Gu et al. 2002, Jorgensen et al. 2007). The utilization of Pdx1-cre mice avoids the confounding effects of Gsα deficiency in the central nervous system present in βGsKO mice as Pdx1 is not expressed in the central nervous system. Similar to βGsKO mice, PGsKO mice developed early-onset hypoinsulinemia and hyperglycemia with reduced β-cell mass. However, GCG levels in PGsKO mice remained similar to those in control mice, indicating that Gsα deficiency likely does not lead to a similar effect on α-cell function and hormone production as it does in β-cells.
The production and secretion of GCG can be regulated by many factors including nutrients, hormones, and neurotransmitters. Although the net effect of GLP1 on GCG secretion is inhibitory (Ahren 2009), several studies have shown that Gsα/cAMP-stimulating agents increase both proglucagon gene expression (Drucker et al. 1991, Islam et al. 2009) and GCG release (Hermansen 1985, Ding et al. 1997, Gromada et al. 1997, Ma et al. 2005) in pancreatic α-cells. We also found a stimulatory effect of increased cAMP on GCG release in αTC1 cells, although the effect of lowering cAMP on GCG release was not significant. Overall, these observations would predict that loss of Gsα expression in α-cells of PGsKO mice should lead to a reduction in serum GCG levels. Another factor that might be expected to reduce GCG levels in PGsKO mice is hyperglycemia (Dumonteil et al. 2000), as elevated glucose levels do not appear to affect GCG expression levels but do inhibit GCG secretion from α-cells (Gromada et al. 2007). In fact, this may help to explain the moderate reduction in GCG levels in βGsKO mice (Xie et al. 2007). However, GCG levels and islet GCG mRNA levels were maintained in PGsKO mice, despite the fact that they have more severe hyperglycemia than βGsKO mice. Low insulin levels would be expected to increase GCG levels in PGsKO mice as insulin has an inhibitory effect on α-cell activity (Kawamori et al. 2009). However, insulin levels were similarly reduced in βGsKO mice as in PGsKO mice (Xie et al. 2007).
Maintenance of normal GCG levels and islet GCG mRNA levels in PGsKO mice in spite of the presence of hyperglycemia and loss of Gsα expression in α-cells may be partially explained by an increment of the α-cell population. While it is difficult to accurately measure the extent of total α-cell mass alteration in PGsKO mice as all pancreatic endocrine and exocrine cells were morphologically abnormal and therefore could not be used as reference controls for each other, there are morphological characteristics to suggest that α-cells are overrepresented in PGsKO mice compared with other pancreatic cell types. In normal mouse islets, β-cells are the most prominent cell population, and are located in the interior of the islet while α-cells are less numerous and located at the periphery. In comparison with islets from βGsKO mice which maintained the normal architecture (Xie et al. 2007), PGsKO islets were small and deformed with a significantly greater number of α-cells within the interior of the islet and a greater α- to β-cell ratio. The increment in α- to β-cell ratio happened relatively early in postnatal life, occurring within the first 4 weeks after birth. These findings are consistent with our results in the αTC1 cell line showing that Gsα signaling pathways have an antiproliferative effect in α-cells, which would suggest that α-cells in PGsKO mice may have growth advantage due to loss of Gsα expression. However, using Ki-67/GCG costaining, we were unable to document a clear increase in α-cell proliferation in PGsKO mice, at least at day 7 post partum. We also found no evidence for the presence of fetal α-cells in PGsKO islets, suggesting that increased neogenesis was not an important factor. Hyperglycemia and hyperlipidemia also likely contribute to an increased α/β-cell ratio through β-cell toxicity (Poitout & Robertson 2008), although these factors were also present in βGsKO mice (Xie et al. 2007).
The mechanisms underlying the antiproliferative and proproliferative effects of Gsα/cAMP signaling in α- and β-cells respectively are unknown. One possible mechanism may be the stimulatory effect of Gsα/cAMP signaling on Pdx1 expression (Jhala et al. 2003), as both in vivo and in vitro studies in β-cells have shown that loss of Pdx1 expression leads to conversion of insulin-secreting cells to GCG-secreting cells (Ahlgren et al. 1998, Lottmann et al. 2001, Wang et al. 2001). Conversely, ectopic expression of Pdx1 in cultured α-cells inhibits GCG gene expression (Ritz-Laser et al. 2003). However, Pdx1 is not expressed in mature α-cells, so Pdx1-independent mechanisms may be responsible for the antiproliferative effect of Gsα on α-cells that we observed in vitro. Moreover, β-cell-specific Gsα deficiency did not lead to reduced Pdx1 expression in βGsKO mice (Xie et al. 2007). Further studies will be required to determine whether Pdx1 plays a role in postnatal α-cell neogenesis, and contributes to enlarged α-cell mass. Recent studies have also shown islet-specific microRNA miR-375 to be a powerful regulator of insulin secretion and pancreatic α- and β-cell mass (Poy et al. 2004, 2009). miR-375-null mice develop hyperglycemia with decreased β-cell mass and increased α-cell mass, similar to PGsKO mice, while overexpression of miR-375 in β-cells leads to a β-cell defect with diabetes (Poy et al. 2009).
In conclusion, our studies in βGsKO and PGsKO mice as well as cultured α-cells show that Gsα/cAMP pathways play important and distinct roles in pancreatic α- and β-cell growth and function. While many studies, including those in βGsKO mice, suggest that these pathways stimulate hormone production in both types of cells, our results show that there are opposite effects of Gsα/cAMP on cell proliferation (proproliferative in β-cells and antiproliferative in α-cells). In both βGsKO and PGsKO mice, the primary defect leading to hyperglycemia is hypoinsulinemia due to a primary β-cell defect. However, compared with βGsKO mice, PGsKO mice had higher GCG levels and α-cell mass relative to β-cell mass, leading to a higher serum GCG/insulin ratio which may contribute to the more severe hyperglycemia in PGsKO mice. This is similar to what is typically observed in type 2 diabetes mellitus and opposite to the effects of GLP1 administration (Ahren 2009). As Gsα mediates the signals of GLP1, impaired GLP1 action in α- and β-cells may be an important factor in the observed changes in the islets of type 2 diabetics (Goke 2008). In addition to its role in the endocrine pancreas, morphology in PGsKO mice also indicates an important role in normal maturation of pancreatic exocrine cells.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by the Intramural Research Program of the National Institute of Diabetes, Digestive, and Kidney Diseases, National Institutes of Health, US Department of Health and Human Services.
References
Abrahamsen N & Nishimura E 1995 Regulation of glucagon and glucagon-like peptide-1 receptor messenger ribonucleic acid expression in cultured rat pancreatic islets by glucose, cyclic adenosine 3′,5′-monophosphate, and glucocorticoids. Endocrinology 136 1572–1578.
Abrahamsen N, Lundgren K & Nishimura E 1995 Regulation of glucagon receptor mRNA in cultured primary rat hepatocytes by glucose and cAMP. Journal of Biological Chemistry 270 15853–15857.
Ahlgren U, Jonsson J, Jonsson L, Simu K & Edlund H 1998 β-Cell-specific inactivation of the mouse Ipf1/Pdx1 gene results in loss of the beta-cell phenotype and maturity onset diabetes. Genes and Development 12 1763–1768.
Ahren B 2009 Islet G protein-coupled receptors as potential targets for treatment of type 2 diabetes. Nature Reviews. Drug Discovery 8 369–385.
Brubaker PL & Drucker DJ 2004 Minireview: glucagon-like peptides regulate cell proliferation and apoptosis in the pancreas, gut, and central nervous system. Endocrinology 145 2653–2659.
Butler AE, Janson J, Bonner-Weir S, Ritzel R, Rizza RA & Butler PC 2003 β Cell deficit and increased β cell apoptosis in humans with type 2 diabetes. Diabetes 52 102–110.
Chen M, Gavrilova O, Zhao W-Q, Nguyen A, Lorenzo J, Shen L, Nackers L, Pack S, Jou W & Weinstein LS 2005 Increased glucose tolerance and reduced adiposity in the absence of fasting hypoglycemia in mice with liver-specific Gsα deficiency. Journal of Clinical Investigation 115 3217–3227.
Ding WG, Renstrom E, Rorsman P, Buschard K & Gromada J 1997 Glucagon-like peptide I and glucose-dependent insulinotropic polypeptide stimulate Ca2+-induced secretion in rat α cells by a protein kinase A-mediated mechanism. Diabetes 46 792–800.
Drucker DJ, Campos R, Reynolds R, Stobie K & Brubaker PL 1991 The rat glucagon gene is regulated by a protein kinase A-dependent pathway in pancreatic islet cells. Endocrinology 128 394–400.
Dumonteil E, Magnan C, Ritz-Laser B, Ktorza A, Meda P & Philippe J 2000 Glucose regulates proinsulin and prosomatostatin but not proglucagon messenger ribonucleic acid levels in rat pancreatic islets. Endocrinology 141 174–180.
Goke B Islet cell function: α and β cells – partners towards normoglycaemia International Journal of Clinical Practice 62 Suppl 159 2008 2–7.
Gromada J, Bokvist K, Ding WG, Barg S, Buschard K, Renstrom E & Rorsman P 1997 Adrenaline stimulates glucagon secretion in pancreatic α cells by increasing the Ca2+ current and the number of granules close to the L-type Ca2+ channels. Journal of General Physiology 110 217–228.
Gromada J, Franklin I & Wollheim CB 2007 α Cells of the endocrine pancreas: 35 years of research but the enigma remains. Endocrine Reviews 28 84–116.
Gu G, Dubauskaite J & Melton DA 2002 Direct evidence for the pancreatic lineage: NGN3+ cells are islet progenitors and are distinct from duct progenitors. Development 129 2447–2457.
Hermansen K 1985 Forskolin, an activator of adenylate cyclase, stimulates pancreatic insulin, glucagon, and somatostatin release in the dog: studies in vitro. Endocrinology 116 2251–2258.
Islam D, Zhang N, Wang P, Li H, Brubaker PL, Gaisano HY, Wang Q & Jin T 2009 Epac is involved in cAMP-stimulated proglucagon expression and hormone production but not hormone secretion in pancreatic alpha- and intestinal L-cell lines. American Journal of Physiology. Endocrinology and Metabolism 296 E174–E181.
Jhala US, Canettieri G, Screaton RA, Kulkarni RN, Krajewski S, Reed J, Walker J, Lin X, White M & Montminy M 2003 cAMP promotes pancreatic β cell survival via CREB-mediated induction of IRS2. Genes and Development 17 1575–1580.
Jorgensen MC, Ahnfelt-Ronne J, Hald J, Madsen OD, Serup P & Hecksher-Sorensen J 2007 An illustrated review of early pancreas development in the mouse. Endocrine Reviews 28 685–705.
Kawamori D, Kurpad AJ, Hu J, Liew CW, Shih JL, Ford EL, Herrera PL, Polonsky KS, McGuinness OP & Kulkarni RN 2009 Insulin signaling in α cells modulates glucagon secretion in vivo. Cell Metabolism 9 350–361.
Lammert E, Gu G, McLaughlin M, Brown D, Brekken R, Murtaugh LC, Gerber HP, Ferrara N & Melton DA 2003 Role of VEGF-A in vascularization of pancreatic islets. Current Biology 13 1070–1074.
Lottmann H, Vanselow J, Hessabi B & Walther R 2001 The Tet-On system in transgenic mice: inhibition of the mouse Pdx1 gene activity by antisense RNA expression in pancreatic β cells. Journal of Molecular Medicine 79 321–328.
Ma X, Zhang Y, Gromada J, Sewing S, Berggren P-O, Buschard K, Salehi A, Vikman J, Rorsman P & Eliasson L 2005 Glucagon stimulates exocytosis in mouse and rat pancreatic α cells by binding to glucagon receptors. Molecular Endocrinology 19 198–212.
Poitout V & Robertson RP 2008 Glucolipotoxicity: fuel excess and β cell dysfunction. Endocrine Reviews 29 351–366.
Powers AC, Efrat S, Mojsov S, Spector D, Habener JF & Hanahan D 1990 Proglucagon processing similar to normal islets in pancreatic α-like cell line derived from transgenic mouse tumor. Diabetes 39 406–414.
Poy MN, Eliasson L, Krutzfeldt J, Kuwajima S, Ma X, Macdonald PE, Pfeffer S, Tuschl T, Rajewsky N & Rorsman P et al. 2004 A pancreatic islet-specific microRNA regulates insulin secretion. Nature 432 226–230.
Poy MN, Hausser J, Trajkovski M, Braun M, Collins S, Rorsman P, Zavolan M & Stoffel M 2009 miR-375 maintains normal pancreatic α and β cell mass. PNAS 106 5813–5818.
Preitner F, Ibberson M, Franklin I, Binnert C, Pende M, Gjinovci A, Hansotia T, Drucker DJ, Wollheim C & Burcelin R et al. 2004 Gluco-incretins control insulin secretion at multiple levels as revealed in mice lacking GLP-1 and GIP receptors. Journal of Clinical Investigation 113 635–645.
Ritz-Laser B, Gauthier BR, Estreicher A, Mamin A, Brun T, Ris F, Salmon P, Halban PA, Trono D & Philippe J 2003 Ectopic expression of the β cell specific transcription factor Pdx1 inhibits glucagon gene transcription. Diabetologia 46 810–821.
Sakamoto A, Chen M, Kobayashi T, Kronenberg HM & Weinstein LS 2005 Chondrocyte-specific knockout of the G protein Gsα leads to epiphyseal and growth plate abnormalities and ectopic chondrocyte formation. Journal of Bone and Mineral Research 20 663–671.
Schonhoff SE, Giel-Moloney M & Leiter AB 2004 Neurogenin 3-expressing progenitor cells in the gastrointestinal tract differentiate into both endocrine and non-endocrine cell types. Developmental Biology 270 443–454.
Simonds WF, Goldsmith PK, Woodard CJ, Unson CG & Spiegel AM 1989 Receptor and effector interactions of Gs. Functional studies with antibodies to the αs carboxyl-terminal decapeptide. FEBS Letters 249 189–194.
Tornehave D, Kristensen P, Rømer J, Knudsen LB & Heller RS 2008 Expression of the GLP-1 receptor in mouse, rat, and human pancreas. Journal of Histochemistry and Cytochemistry 56 841–851.
Wang H, Maechler P, Ritz-Laser B, Hagenfeldt KA, Ishihara H, Philippe J & Wollheim CB 2001 Pdx1 level defines pancreatic gene expression pattern and cell lineage differentiation. Journal of Biological Chemistry 276 25279–25286.
Weinstein LS, Yu S, Warner DR & Liu J 2001 Endocrine manifestations of stimulatory G protein α-subunit mutations and the role of genomic imprinting. Endocrine Reviews 22 675–705.
Wilson ME, Kalamaras JA & German MS 2002 Expression pattern of IAPP and prohormone convertase 1/3 reveals a distinctive set of endocrine cells in the embryonic pancreas. Mechanisms of Development 115 171–176.
Xie T, Chen M, Zhang QH, Ma Z & Weinstein LS 2007 β Cell-specific deficiency of the stimulatory G protein α-subunit Gsα leads to reduced β cell mass and insulin-deficient diabetes. PNAS 104 19601–19606.
