Glucagon-like peptide-1 receptor signalling selectively regulates murine lymphocyte proliferation and maintenance of peripheral regulatory T cells
- 2k Downloads
Glucagon-like peptide-1 receptor (GLP-1R) agonists improve glucose control in animals and humans with type 1 diabetes. However, there is little information on the role of the GLP-1R in the immune system. We studied the role of the GLP-1R in immune function in wild-type (WT) and non-obese diabetic (NOD) and Glp1r −/− mice.
Glp1r mRNA expression was examined in sorted immune subpopulations by RT–PCR. The effects of GLP-1R activation were assessed on cAMP production and proliferation, migration and survival of primary immune cells from WT and NOD mice. The ability of primary cells from Glp1r −/− mice to proliferate, migrate or survive apoptosis was determined. Immunophenotyping studies were performed to assess the frequency of immune subpopulations in Glp1r −/− mice.
Ex vivo activation of the GLP-1R resulted in a modest but significant elevation of cAMP in primary thymocytes and splenocytes from both WT and NOD mice. GLP-1R activation did not increase proliferation of primary thymocytes, splenocytes or peripheral lymph node cells. In contrast, Glp1r −/− thymocytes exhibited a hypoproliferative response, whilst peripheral Glp1r −/− lymphocytes were hyperproliferative in response to mitogenic stimulation. Activation or loss of GLP-1R signalling did not modify apoptosis or chemotaxis in primary lymphocytes. Male Glp1r −/− mice exhibited a significantly lower percentage of peripheral regulatory T cells, although no differences were observed in the numbers of CD4+ and CD8+ T cells and B cells in the spleen and lymph nodes of Glp1r −/− mice.
These studies establish that GLP-1R signalling may regulate lymphocyte proliferation and maintenance of peripheral regulatory T cells.
KeywordsGLP-1 GLP-1 receptor Immunology Incretin T cell
CD45R/protein tyrosine phosphatase, receptor type, C
Forkhead box protein 3
Glucagon-like peptide-1 receptor
Macrophage inflammatory protein 3b
Phorbol myristate acetate
Stromal cell-derived factor 1a
T cell receptor
Regulatory T cell
Glucagon-like peptide-1 receptor (GLP-1R) activation, either by preventing degradation of its endogenous ligand, or via administration of long-acting receptor agonists, is used for the treatment of type 2 diabetes . GLP-1R activation induces multiple glucoregulatory actions, including inhibition of glucagon secretion and gastric emptying, both of which reduce postprandial glycaemic excursions in patients with type 1 diabetes . The regenerative and cytoprotective effects of glucagon-like peptide-1 (GLP-1) on beta cells have engendered considerable interest in whether GLP-1 may be useful for the treatment of type 1 diabetes, either in combination with immunomodulation [3, 4, 5, 6] or as adjunctive therapy in islet transplantation [7, 8, 9, 10].
Intriguingly, GLP-1R agonists improve type 1 diabetes in animal models without the need for immunotherapy, thus implicating GLP-1R-dependent signalling pathways in immunoregulatory processes [11, 12, 13, 14]. Nevertheless, little is known about the effects of GLP-1 on the immune system. Glp1r mRNA transcripts have been detected in murine lymphoid tissue , and GLP-1R activation stimulates regulatory T cells (Tregs), by both increasing the frequency and improving the function of these cells in recently diagnosed diabetic NOD mice .
The use of exenatide therapy in patients with type 1 diabetes [16, 17], despite limited knowledge of the putative immunoregulatory role of the GLP-1R, prompted us to ascertain whether the GLP-1R is active in a range of immune cells. We have now assessed immune function following GLP-1R activation in wild-type (WT) and non-obese diabetic (NOD) mice, a model of type 1 diabetes. In complementary studies, we characterised selective functional variables using primary immune cells from Glp1r −/− mice. Our data reveal a role for GLP-1R signalling in the proliferation and regulation of primary murine lymphocytes.
Synthetic exendin-4 (Ex-4) was purchased from California Peptide Research (Napa, CA, USA), dissolved in phosphate buffered saline (pH 7.4) and stored at −80°C until use. GLP-1 (1–37), GLP-1 (7–36) amide and GLP-1 (9–36) were purchased from Bachem (Torrance, CA, USA). Prostaglandin E2 (PGE-2) was purchased from Sigma-Aldrich (St Louis, MO, USA).
Anti-CD3 and anti-CD28 antibodies were from BD Pharmingen (San Jose, CA, USA). Concanavalin A (ConA), phorbol myristate acetate (PMA) and ionomycin were from Sigma-Aldrich. IL-2, chemokine (C-C motif) ligand 19 (CCL19) (also known as macrophage inflammatory protein 3b [MIP-3b]), chemokine (C-X-C motif) ligand 12 (CXCL12) (also known as stromal cell-derived factor 1a [SDF1a]) and chemokine (C-C motif) ligand 25 (CCL25) were from Peprotech (Rocky Hill, NJ, USA). Dexamethasone was from Sigma-Aldrich, staurosporine from Calbiochem (Darmstadt, Germany) and anti-Fas antibody (Jo2) and annexin-V/7-amino-actinomycin D (7-AAD) apoptosis detection kit from BD Pharmingen.
For immunophenotyping and cell-sorting studies, the following monoclonal antibodies were prepared as described previously : purified anti-CD16/CD32 (for Fc blocking), biotinylated anti-CD8α, FITC-labelled anti-CD8α, FITC-labelled anti-CD4, phycoerythrin (PE)-labelled anti-CD4, allophycocyanin (APC)-labelled anti-H57 (T cell receptor [TCR] β chain) and FITC-labelled anti-CD19. Streptavidin–Spectral Red (SPRD) was from Southern Biotech (Birmingham, AL, USA). FITC-labelled anti-IgM and PE-labelled anti-protein tyrosine phosphatase, receptor type, C (B220) were from BD Pharmingen. The mouse Treg-staining kit (FITC-labelled anti-CD4, PE-labelled anti-CD25 and APC-labelled anti-mouse/rat forkhead box protein [FOXP3] [FJK-16s]) was from eBioscience (San Diego, CA, USA).
NOD/Ltj mice were from the Jackson Laboratory (Bar Harbor, MI, USA) and C57Bl/6 mice were from Taconic Farms (Germantown, NY, USA). Mice were acclimatised to the animal facility for at least 1 week prior to analysis. For the immunophenotyping and loss-of-function studies, Glp1r −/− mice and age- and sex-matched littermate controls on a C57Bl/6 background were used. Mice were housed in a pathogen-free facility and maintained on a 12 h light–dark cycle, with free access to standard rodent chow and water. All experiments were carried out in accordance with protocols and guidelines approved by the Animal Care Committees of the Toronto General and Mt Sinai Hospitals.
Primary single-cell suspensions from thymus, bone marrow, lymph nodes and spleen of 6- to 9-week-old male and female NOD/Ltj and age-matched C57Bl/6 mice were prepared by passing the tissue through a 70 µm cell strainer and recovering cells in calcium- and magnesium-free Hanks’ balanced salt solution with 2% (vol./vol.) FBS and 10 mmol/l HEPES. For the bone marrow preparation, cells were flushed from the tibia and femur and clumps were dissociated by repeatedly passing the cells through a 10 ml syringe, then through a 70 µm cell strainer. Erythrocytes were lysed using RBC Lysis Buffer (Biolegend, San Diego, CA, USA) and removed from cell suspensions. Cell numbers were quantified using a haemocytometer. Cells were incubated with antibodies for 30 min at 4°C in staining media. After removing excess antibody, cells were filtered to remove clumps and sorted on a FACS Aria cell sorter (BD Biosciences, San Jose, CA, USA).
Glp1r expression in sorted immune subsets
Following cell sorting, cells were washed and lysed, and total RNA extracted using RNeasy Mini or Micro kits (Qiagen, Mississauga, ON, Canada). Target cDNA was analysed for the expression of mouse Glp1r and β-actin gene mRNA transcripts by the PCR method . Primer pairs are indicated in Electronic supplementary material (ESM) Table 1. The products were visualised with SYBR Safe DNA stain (Invitrogen, Carlsbad, CA, USA) and transferred to a Nytran Super Charge Nylon Membrane (Mandel Scientific, Guelph, ON, Canada). The amplified cDNA was hybridised overnight using internal primers (ESM Table 1), as described previously .
Single-cell suspensions were prepared from the spleen, lymph nodes, thymus and bone marrow of Glp1r −/− and littermate control mice, as described above. Cells (106) were incubated at 4°C with primary and secondary (in the case of biotinylated primary) antibodies in cold staining media. Cells were either resuspended in cold staining media containing propidium iodide (for dead cell exclusion) and analysed directly, or fixed/permeabilised and stained with the anti-FOXP3 antibody according to the manufacturer’s protocol. Cells were acquired on a FACSCalibur flow cytometer (BD Biosciences) and analysed using FlowJo software (Tree Star, Ashland, OR, USA).
Single-cell suspensions from spleens, thymuses and lymph nodes of 10- to 12-week-old normoglycaemic female NOD/Ltj and C57Bl/6 mice were prepared as described above. Primary cells were resuspended in Earle’s balanced salt solution (Invitrogen) with 0.5% (vol./vol.) FBS and 10 mmol/l HEPES (treatment media). Triplicate wells were treated with the appropriate agonists, supplemented with 10 µmol/l 3-isobutyl-1-methylxanthine (Sigma-Aldrich), for 15 min in an incubator. The reaction was terminated by the addition of 65% (vol./vol.) cold ethanol. cAMP levels were measured with a cAMP RIA kit (Biomedical Technologies, Stoughton, MA, USA).
Primary thymocytes and lymph node cells were prepared aseptically in complete RPMI media with 10% (vol./vol.) heat-inactivated FBS, 10 mmol/l HEPES, 0.055 mmol/l 2-mercaptoethanol, 2 mmol/l L-glutamine, 100 U/ml penicillin and 100 µg/ml streptomycin. The indicated reagents were prepared in complete RPMI and added in triplicate in wells of tissue-culture (TC)-treated flat-bottom 96 well plates (BD Labware, Franklin Lakes, NJ, USA); 1 × 106 thymocytes and 5 × 105 lymph node cells were added per well and incubated for 48 h at 37°C in 5% CO2. Cells were then pulsed with 0.037 MBq [3H]thymidine per well (GE Healthcare, Piscataway, NJ, USA) and plates were returned to the incubator for an additional 18 h. Following cell harvesting, the incorporated radioactivity was measured using an automated β-scintillation counter. Data are expressed as a stimulation index, calculated as the mean cpm of treated wells divided by the mean cpm of negative control (untreated) wells.
Transwell migration assays
Assays were performed as previously described, with slight modifications . Primary cells were recovered from spleens, thymuses and peripheral and mesenteric lymph nodes under sterile conditions, and resuspended in migration assay media (RPMI1640 with 0.5% [wt/vol.] BSA and 25 mmol/l HEPES). Chemokines and peptides were prepared in migration assay media and added to the bottom wells of TC-treated six well plates in a total volume of 0.6 ml in duplicate. Input cells (1 × 106) were added to the top of a 5 µm transwell membrane (Corning, Lowell, MA, USA). Where indicated, the transwell membranes were coated overnight with 20 µg/ml fibronectin (BD Biosciences) followed by blocking with 2% (wt/vol.) BSA in PBS. After 4 h, cells that had migrated to the bottom well were collected and counted on a FACSCalibur flow cytometer, using FACS calibrate beads (BD Biosciences) as a normalisation standard. Dead cells were excluded from analysis by the addition of propidium iodide (Sigma-Aldrich).
Primary thymocytes (1 × 106) in complete RPMI media were added to wells of TC-treated round-bottom 96-well plates. For apoptosis rescue experiments, thymocytes were pretreated for 1 h at 37°C with the indicated reagents, following by addition of dexamethasone, anti-Fas antibody or staurosporine. In loss-of-function studies, the aforementioned reagents were added directly to the cells. All treatments were performed in triplicate. Cells were collected 6 h after dexamethasone treatment and 18 h following anti-Fas antibody and staurosporine addition, stained for apoptosis markers using the annexin-V/7-AAD apoptosis detection kit (BD Pharmingen) and analysed on a FACSCalibur flow cytometer.
Results are expressed as means±SE, or where indicated means±SD. All statistical analyses were performed with PrismVersion 4.0 software (GraphPAD Software, San Diego, CA, USA). Statistical significance was assessed by one- or two-way ANOVA. Differences were considered statistically significant at p ≤ 0.05.
Glp1r mRNA expression in sorted immune populations
GLP-1R activation leads to cAMP production in primary immune cells
Loss of GLP-1R signalling impairs proliferative responses
GLP-1R signalling is not required for the migration of primary lymphocytes
GLP-1R signalling does not modify cell death in murine thymocytes
Lymphocyte subsets in Glp1r−/− mice
GLP-1R agonists and dipeptidyl peptidase-4 inhibitors are approved for the treatment of type 2 diabetes  and are being examined in clinical trials of islet transplantation in patients with type 1 diabetes [16, 17], an autoimmune disorder resulting from defective immunoregulation [23, 24]. A subset of patients with type 2 diabetes may also exhibit a dysregulated immune system [25, 26, 27]. Previous work from our laboratory , as well as others  suggested that GLP-1R signalling may regulate T cell subsets, including Tregs. We have now assessed the effects of activating or eliminating the GLP-1R in subsets of murine immune cells.
Although GLP-1 action was first localised to beta cells , the GLP-1R is detected in extrapancreatic tissues, including liver, lung, kidney, heart, brain and intestine [29, 30]. We previously demonstrated that Glp1r transcripts are expressed in spleen, thymus and lymph nodes of NOD and C57Bl/6 mice . Our current findings illustrate that the Glp1r is more widely expressed in immune subpopulations from the NOD, relative to the C57BL/6 background. Chiu et al. demonstrated a higher frequency of activated B and T cells in the NOD mouse compared with other strains, including the C57BL/6 mouse . Hence one potential explanation for the differential Glp1r expression could be the activation status of lymphocytes, as we did not discriminate between rested and activated B and T cells during the sorting procedure. Further studies are required to examine the significance of and mechanisms underlying the differential Glp1r expression on the NOD vs C57 background.
Intriguingly, there were sex-specific differences in Glp1r expression. There are no reports examining sex differences in GLP-1R abundance in immune cells, or in other tissues. Perhaps not surprisingly there are reported sex-specific patterns of gene expression in the NOD mouse . Hence the observed differences in GLP-1R abundance may reflect both species-specific and sex-specific differences in the control of GLP-1R production. As diabetes development demonstrates a strong sex bias in the NOD mouse, the implications arising from differences in GLP-1R levels require further examination.
Our analyses reveal that several immature lymphocyte subsets in the thymus and bone marrow express Glp1r transcripts. In particular, CD19+ B cells from the bone marrow of female but not male C57 mice are positive for Glp1r transcripts. Moreover, aged female Glp1r −/− mice have a significantly increased percentage of immature B220+IgM− B cells in their bone marrow, implicating GLP-1R signalling in B cell maturation. Further studies are required to determine the timing and significance of GLP-1R production during haematopoiesis and lymphocyte maturation.
Glp1r transcripts were also detected in sorted CD4+CD25+ cells, a cell subset mostly comprising Tregs . Glp1r transcripts are present in CD4+CD25+ cells isolated from the spleens of male C57 mice. Immunophenotyping analysis revealed a significant deficiency in the percentage of CD4+CD25+FOXP3+ Tregs from the lymph nodes of male Glp1r −/− mice, thus suggesting that GLP-1R signalling could play a role in the maintenance and functioning of Tregs. Conversely, we were not able to detect Glp1r mRNA transcripts in splenic samples of female C57 mice, and we did not find a difference in the percentage of splenic and lymph node Tregs of female Glp1r −/− mice.
On the NOD background, CD4+CD25+ cells from both male and female mice were positive for Glp1r transcripts. Very little is known about a potential role for GLP-1R in the frequency and suppressive capacity of Tregs in the setting of an autoimmune attack. Previous work from our laboratory demonstrated a decrease in the number of thymic Tregs in NOD mice receiving chronic Ex-4 treatment, although we did not detect any differences in peripheral Tregs isolated from lymph nodes . Intriguingly, Suarez-Pinzon et al. have shown that treatment of diabetic NOD mice with a combination of GLP-1 and gastrin results in a diminished autoimmune response during the transplantation of syngeneic islets. The authors observed an induction of suppressive cytokine secretion in islet-infiltrating leucocytes, and concluded that the GLP-1 and gastrin combination therapy may regulate immune pathways in the NOD mice . A recent study investigated the role of GLP-1R activation in the number and function of Tregs in NOD mice. Treatment with 200 ng Ex-4 for 30 days increased the number of splenic Tregs. Furthermore, a trend towards enhanced Treg function was observed in CD4+CD25+ splenocytes isolated from Ex-4-treated mice .
Our data demonstrated that GLP-1R activation leads to a modest but significant increase in cAMP accumulation in mixed leucocyte populations. Elevated cAMP has been shown to be a mechanism by which Tregs induce immunosuppression. In fact, Tregs express high levels of cAMP, and are able to induce cAMP accumulation in activated target cells by multiple mechanisms [34, 35]. Taking into consideration that GLP-1R activation modulates Treg function , it is possible to speculate that this mechanism is cAMP dependent. Further experiments are required to elucidate potential mechanism(s) by which GLP-1R activation augments Treg function.
GLP-1R activation did not enhance proliferation of primary thymocytes and lymph node cells, either alone or as a costimulatory signal when combined with anti-CD3. In contrast, loss of GLP-1R signalling in primary thymocytes resulted in a significantly diminished proliferation response upon ConA and PMA plus ionomycin stimulation, but there was no difference on the proliferative response when thymocytes were stimulated with anti-CD-3 and anti-CD-3 plus anti-CD-28 activation. Stimulation of T cells with ConA is dependent on changes in cellular cAMP [36, 37]. Hence it is possible that GLP-1R activation has an additive effect on cAMP accumulation following activation with ConA, and the absence of this additive effect, as is the case in immune cells lacking functional GLP-1R, results in a defective proliferative response.
The combination of PMA and ionomycin induces proliferation by directly activating protein kinase C (PKC) and increasing intracellular calcium levels, hence bypassing T cell reception activation . The fact that thymocytes from Glp1r −/− mice displayed a hypoproliferative response to PMA plus ionomycin suggests that GLP-1R signalling is required subsequent to activation of PKC. Potential candidates could be transcription factors, as GLP-1R activation in the islet beta cells has been shown to promote nuclear translocation of nuclear factor kappaB and the nuclear factor of activated T cells, and to activate several immediate early-response genes and proto-oncogenes [39, 40, 41], which are common in the signalling pathway of TCR activation .
Paradoxically, Glp1r −/− peripheral lymph node cells exhibited a significantly enhanced proliferative response to mitogens. It has been shown that in mouse islets, genetic disruption of GLP-1R signalling results in compensatory changes in signalling pathways. In the study by Flamez et al., islets lacking a functional GLP-1R also demonstrated elevated cytosolic calcium levels, and this increase in calcium was observed at a significantly lower threshold for glucose concentration . These results might partially explain the hyperresponsiveness of immune cells lacking GLP-1R signalling, as elevation in cytosolic calcium is a triggering signal for lymphocyte proliferation (as reviewed ).
The hypoproliferative responses in thymocytes from Glp1r −/− mice were observed only when thymocytes were stimulated with non-physiological mitogens (PMA and ConA). Hence it seems that GLP-1R signalling is not critical for the endogenous proliferation of thymocytes. However, we did observe a defect in response to anti-CD3 plus anti-CD28 in peripheral lymphocytes, which suggests that GLP-1R signalling may be required for optimal proliferation of peripheral lymphocytes under these experimental conditions. A potential explanation for these differences could be that, in the thymus, absent GLP-1R signalling is compensated by upregulation of other mechanisms not operative in the periphery. Indeed, it was previously reported that glucoregulatory actions of GLP-1 are compensated via other mechanisms in Glp1r −/− mice .
GLP-1R signalling in immune cells was not associated with modulation of anti-apoptotic pathways. In contrast, GLP-1R activation inhibits apoptosis in other cell types including islet beta cells, cardiomyocytes and neurons [46, 47, 48]. However, we cannot exclude the possibility that a modest effect of GLP-1 on cell survival may have been obscured, as the importance of GLP-1R signalling for cytoprotection was studied in mixed immune populations. Hence, further viability studies are required in isolated immune subsets producing the GLP-1R to resolve whether GLP-1R signalling has a role in leucocyte apoptosis.
Similarly, we were not able to detect any effects on lymphocyte migration upon either GLP-1R activation or loss of GP-1R signalling in mixed primary immune populations. This result was also unanticipated, as it is well accepted that G protein-coupled receptor signalling is crucial for chemotaxis [49, 50]. The GLP-1R, although primarily coupled to Gαs, has also been shown to couple to the Gαi protein , which is essential for the migration response. In agreement with our results, Kim et al. also did not find a significant effect on migration of GLP-1-treated splenic CD4+ T cells .
In conclusion, we report that Glp1r mRNA is widely expressed in several immune subpopulations. Activation of the GLP-1R leads to cAMP accumulation, and GLP-1R signalling contributes to the regulation of both thymocyte and peripheral T cell proliferation. Moreover, the GLP-1R is involved in the maintenance of peripheral Tregs. These findings expand our understanding of the role of the GLP-1R in the development, maintenance and function of the immune system.
The authors would like to thank A. Wong for assistance with the immunophenotyping experiments, B. Dong for assistance with the proliferation, apoptosis and migration experiments, D. Holland for excellent technical assistance and J. Zhang for helpful discussions on data analysis and interpretation. In addition, the authors would like to thank the staff at the SickKids-UHN flow cytometry facility (Toronto, ON, Canada), in particular S. Zhao, P.-A. Penttilä, L. Jamieson, M. Tseng and A. Bang from Mt Sinai Flow Cytometry facility for their expertise in cell sorting. These studies were supported in part by the Canada Research Chairs Program, and a grant from the Juvenile Diabetes Association (1-2006-696) and the Canadian Research Chairs Program to D. J. Drucker.
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.
- 7.National Institutes of Health Clinical Center (2009) Effect of AC2993 with or without immunosuppression on beta cell function in patients with type I diabetes. Available from www.clinicaltrials.gov/, accessed 6 September 2009.