The canonical target of the glucagon-like peptide-1 receptor (GLP-1R), Protein Kinase A (PKA), has been shown to stimulate mechanistic Target of Rapamycin Complex 1 (mTORC1) by phosphorylating the mTOR-regulating protein Raptor at Ser791 following β-adrenergic stimulation. The objective of these studies is to test whether GLP-1R agonists similarly stimulate mTORC1 via PKA phosphorylation of Raptor at Ser791 and whether this contributes to the weight loss effect of the therapeutic GLP-1R agonist liraglutide. We measured phosphorylation of the mTORC1 signaling target ribosomal protein S6 in Chinese Hamster Ovary cells expressing GLP-1R (CHO-Glp1r) treated with liraglutide in combination with PKA inhibitors. We also assessed liraglutide-mediated phosphorylation of the PKA substrate RRXS*/T* motif in CHO-Glp1r cells expressing Myc-tagged wild-type (WT) Raptor or a PKA-resistant (Ser791Ala) Raptor mutant. Finally, we measured the body weight response to liraglutide in WT mice and mice with a targeted knock-in of PKA-resistant Ser791Ala Raptor. Liraglutide increased phosphorylation of S6 and the PKA motif in WT Raptor in a PKA-dependent manner but failed to stimulate phosphorylation of the PKA motif in Ser791Ala Raptor in CHO-Glp1r cells. Lean Ser791Ala Raptor knock-in mice were resistant to liraglutide-induced weight loss but not setmelanotide-induced (melanocortin-4 receptor-dependent) weight loss. Diet-induced obese Ser791Ala Raptor knock-in mice were not resistant to liraglutide-induced weight loss; however, there was weight-dependent variation such that there was a tendency for obese Ser791Ala Raptor knock-in mice of lower relative body weight to be resistant to liraglutide-induced weight loss compared to weight-matched controls. Together, these findings suggest that PKA-mediated phosphorylation of Raptor at Ser791 contributes to liraglutide-induced weight loss.
glucagon-like peptide-1mechanistic target of rapamycinraptorprotein kinase Abody weightResearch organismMousehttp://dx.doi.org/10.13039/100000002National Institutes of HealthR01DK097361AyalaJulio Ehttp://dx.doi.org/10.13039/100000002National Institutes of HealthR01DK116625CollinsSheilahttp://dx.doi.org/10.13039/100000002National Institutes of HealthS10OD028455AyalaJulio Ehttp://dx.doi.org/10.13039/100000002National Institutes of HealthR01DK132852AyalaJulio EThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.Author impact statementThe ability of the therapeutic glucagon-like peptide-1 receptor agonist liraglutide to promote weight loss is partly dependent on signaling via PKA-dependent phosphorylation of the mTORC1 regulator protein Raptor.Introduction
Glucagon-like peptide-1 (GLP-1) receptor (GLP-1R) agonists reduce food intake and body weight and are approved for weight loss (Drucker, 2022). Elucidating the mechanisms through which GLP-1R agonists regulate body weight could enhance the therapeutic potential of these compounds or identify novel targets for anti-obesity drug development. We have shown that the GLP-1R agonist Exendin-4 (Ex4) stimulates mechanistic Target of Rapamycin (mTOR) Complex-1 (mTORC1) signaling and that inhibition of mTORC1 signaling in the ventromedial hypothalamus (VMH) with rapamycin blocks the anorectic effect induced by Ex4 in this nucleus (Brown et al., 2018; Burmeister et al., 2017). The present studies aim to define the mechanism(s) by which GLP-1R agonism stimulates mTORC1 signaling and reduces body weight.
mTOR is a Ser/Thr protein kinase that forms two complexes: mTORC1 and mTOR complex 2 (mTORC2; Saxton and Sabatini, 2017). The rapamycin-sensitive mTORC1 has unique accessory proteins including Raptor (regulatory associated protein of mTOR). mTORC1 activity is stimulated by growth factors, nutrients, and hormones by promoting PI3K/Akt-mediated phosphorylation of tuberous sclerosis complex 1/2 (TSC1/2) (Kim et al., 2002). Activation of mTORC1 by GLP-1R agonists seemed unlikely since the GLP-1R is a Gαs-coupled receptor that signals via protein kinase A (PKA; Thorens, 1992). However, activation of the Gαs-coupled β-adrenergic receptor stimulates mTORC1 via Raptor phosphorylation by PKA (Liu et al., 2016), suggesting that GLP-1R activation can similarly stimulate mTORC1 via PKA phosphorylation of Raptor.
Here we demonstrate that GLP-1R activation induces mTORC1 signaling and Raptor phosphorylation in a PKA-dependent manner and show that this contributes to weight loss by the therapeutic GLP-1R agonist liraglutide. This reveals a novel weight-lowering GLP-1R-PKA-mTORC1 axis and broadly suggests that other Gαs-coupled receptors may use this PKA-mTORC1 pathway for certain biological responses.
ResultsGLP-1R agonists activate mTORC1 in a PKA-dependent manner
Similar to Ex4 (Brown et al., 2018; Burmeister et al., 2017), liraglutide increases phosphorylation of the mTORC1 signaling target ribosomal protein S6 in CHO-GLP-1R cells, and this is blocked by the mTORC1 inhibitor rapamycin (Figure 1A; Figure 1—figure supplement 1). The PKA inhibitor H89 also blocks S6 phosphorylation by liraglutide or the adenylyl cyclase/PKA activator forskolin (Figure 1B; Figure 1—figure supplement 1). However, H89 has some off-target effects including inhibiting the S6 kinase p70S6K (Davies et al., 2000), likely explaining why H89 inhibits S6 phosphorylation by insulin (Figure 1B). KT5720, a PKA inhibitor with minimal off-target effects on p70S6K (Davies et al., 2000), also reduces liraglutide- and forskolin-induced phosphorylation of S6 and the PKA target CREB in a dose-dependent manner (Figure 1C; Figure 1—figure supplement 1). Neither dose of KT5720 affected S6 phosphorylation by insulin (Figure 1C). Two doses of the pan-Akt inhibitor MK-2206 that block insulin-stimulated Akt phosphorylation did not affect stimulation of mTORC1 activity by liraglutide (Figure 1D; Figure 1—figure supplement 1). These data suggest that liraglutide stimulates mTORC1 signaling in a PKA-dependent and Akt-independent manner.
The GLP-1R agonist liraglutide increases mTOR activity in a PKA- dependent manner (Resubmission Rev 2 <xref ref-type="supplementary-material" rid="fig1sdata1">Figure 1—source data 1</xref>).
(A) Immunoblot of pS6, S6, pCREB, and CREB and quantification of pS6 expression (pS6 /S6) in CHO-K1 cells transfected with the a vector expressing the hGLP1R and treated with DMSO or the mTORC1 inhibitor rapamycin (100 nM) for 30 min followed by treatment with 1 X PBS (Veh) or liraglutide (Lira, 10 nM) for 1 hr (Interaction: F (1, 8) = 6.926, p=0.0301; Inhibitor: F (1, 8) = 6.926, p=0.0301; Treatment: F (1, 8) = 22.50, p=0.0015). (B–C) Immunoblot of pS6, S6, pCREB, and CREB and quantification of pS6 expression (pS6/S6) in CHO-K1 cells stably expressing the hGLP1R treated with DMSO (Veh) or either of the PKA inhibitors, (B) H89 (10 µM) or (C) KT5720 (10 µM or 50 μM), for 30 min followed by treatment with vehicle (PBS), Lira (10 nM), forskolin (Fsk, 10 µM), or insulin (Ins, 10 mg/mL) for 1 hr (H89: Interaction: F (1, 12) = 4.782, p=0.0493; Inhibitor: F (1, 12) = 4.782, p=0.0493; Treatment: F (1, 12) = 12.50, p=0.0041). KT5720: (Interaction: F (2, 18) = 15.32, p=0.0001; Inhibitor: F (2, 18) = 15.32, p=0.0001; Treatment: F (1, 18) = 36.14, p<0.0001). Quantification of pCREB expression (pCREB/CREB) is shown in Figure 1—figure supplement 1. (D) Immunoblot of pS6, S6, pAkt, and Akt and quantification of pS6 expression (pS6/S6) in CHO-K1 cells stably expressing the hGLP1R treated with DMSO or the pan-Akt inhibitor MK-2206 (5 μM or 20 µM) for 30 min followed by treatment with vehicle (PBS), Lira (10 nM), Fsk (10 µM), or Ins (10 mg/mL) for 1 hr (Interaction: F (1, 12) = 0.02486, p=0.8773; Inhibitor: F (1, 12) = 0.02486, p=0.8773; Treatment: F (1, 12) = 110.2, p<0.0001). Quantification of pAkt expression (pAkt/Akt) is shown in Figure 1—figure supplement 1. Analysis was done using two-way ANOVA followed by Tukey’s multiple comparisons. Data are normalized to the respective control (Veh or Veh and inhibitor). Absolute data are shown in Figure 1—figure supplement 1. All data are shown as mean ± SEM, ns not significant, * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001, n=3–4 biological replicates.
Graphs of absolute quantification of data from <xref ref-type="fig" rid="fig1">Figure 1</xref> and raw data in Excel spreadsheet.
Absolute pS6, pCREB, and pAkt expression in CHO-K1 cells stably expressing the <italic>hGLP1R</italic> and treated with liraglutide.
(Resubmission Rev 2 Figure 1—figure supplement 1—source data 1). (A) Quantification of pS6 expression (pS6 /S6) in CHO-K1 cells transfected with the a vector expressing the hGLP1R and treated with DMSO or the mTORC1 inhibitor rapamycin (100 nM) for 30 min followed by treatment with 1 X PBS (Veh) or liraglutide (Lira, 10 nM) for 1 hr (Interaction: F (1, 8) = 31.10, p=0.0005; Inhibitor: F (1, 8) = 50.10, p=0.0001; Treatment: F (1, 8) = 39.17, p=0.0002). (B–C) Quantification of pS6 (pS6/S6) and pCREB (pCREB/CREB) expression in CHO-K1 cells stably expressing the hGLP1R treated with DMSO (Veh) or either of the PKA inhibitors, (B) H89 (10 µM) or (C) KT5720 (10 µM or 50 μM), for 30 min followed by treatment with vehicle (PBS) or Lira (10 nM) for 1 hr (H89 pS6/S6: Interaction: F (1, 12) = 13.40, p=0.0033; Inhibitor: F (1, 12) = 21.00, p=0.0006; Treatment: F (1, 12) = 18.40, p=0.0011. H89 pCREB/CREB: Interaction: F (1, 8) = 356.7, p<0.0001; Inhibitor: F (1, 8) = 578.5, p<0.0001; Treatment: F (1, 8) = 361.3, p<0.0001. KT5720 pS6/S6: Interaction: F (2, 18) = 15.32, p=0.0001; Inhibitor: F (2, 18) = 15.32, p=0.0001; Treatment: F (1, 18) = 36.14, p<0.0001. KT5720 pCREB/CREB: Interaction: F (2, 12) = 30.76, p<0.0001; Inhibitor: F (2, 12) = 43.93, p<0.0001; Treatment: F (1, 12) = 63.66, p<0.0001). Quantification of pS6 and pAkt (pAkt/Akt) in CHO-K1 cells stably expressing the hGLP1R treated with DMSO or the pan-Akt inhibitor MK-2206 (5 μM or 20 µM) for 30 min followed by treatment with vehicle (PBS), Lira (10 nM) or Ins (10 mg/mL) for 1 hr (pS6/S6: Interaction: F (1, 12) = 17.67, p=0.0012; Inhibitor: F (1, 12) = 31.37, p=0.0001; Treatment: F (1, 12) = 181.6, p<0.0001. pAkt/Akt: Interaction: F (2, 12) = 60.57, p<0.0001; Inhibitor: F (1, 12) = 62.79, p<0.0001; Treatment: F (2, 12) = 61.73, p<0.0001). Analysis was done using two-way ANOVA followed by Tukey’s multiple comparisons. Data are presented as absolute densitometry values and are shown as mean ± SEM, ns not significant, * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001, n=3–4 biological replicates.
Quantification of absolute values from data in <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref> and raw data.
Dose response and time course of absolute pS6 and pCREB expression in CHO-K1 cells stably expressing the <italic>hGLP1R</italic> and treated with liraglutide.
(Resubmission Rev 2 Figure 1—figure supplement 2—source data 1). (A) Quantification of pS6 expression (pS6/S6) and pCREB (pCREB/CREB) in CHO-K1 cells stably expressing the hGLP1R and treated with DMSO (Veh) or increasing doses of Lira for 1 h (pS6/S6: F (5, 12) = 70.69, p<0.0001; pCREB/CREB: F (5, 12) = 52.89, p<0.0001). (B) Quantification of pS6 expression (pS6/S6) and pCREB (pCREB/CREB) in CHO-K1 cells stably expressing the hGLP1R and treated with DMSO (Veh) or Lira (10 nM) for different time points (pS6/S6: F (4, 15) = 73.02, p<0.0001; pCREB/CREB: F (4, 15) = 178.2, p<0.0001). Analysis was done using one-way ANOVA followed by Brown-Forsythe test. Data are presented as absolute densitometry values and are shown as mean ± SEM, ** p<0.01, **** p<0.0001, n=3–4 biological replicates.
<xref ref-type="fig" rid="fig1s2">Figure 1—figure supplement 2</xref> and raw blots and raw data for <xref ref-type="fig" rid="fig1s2">Figure 1—figure supplement 2</xref>.
GLP-1R activation induces PKA phosphorylation of Raptor at Ser<sup>791</sup>
The catalytic subunit of PKA directly phosphorylates Ser791 of Raptor (Liu et al., 2016). We tested whether liraglutide promotes PKA phosphorylation of Raptor at Ser791 by expressing Myc-tagged wild-type (WT) Raptor or Myc-tagged mutant Raptor with alanine replacing serine at the 791st position (Ser791Ala Raptor) in CHO-GLP-1R cells. Using an antibody against the PKA motif RRXS*/T*, immunoblotting of immunoprecipitated Myc-proteins shows that liraglutide and forskolin increase PKA motif phosphorylation on Myc-WT Raptor in a PKA-dependent manner (blocked by H89), but liraglutide fails to increase PKA motif phosphorylation on Myc-Ser791Ala Raptor (Figure 2A and B). This was also observed in β-TC3 cells, an immortalized β-cell line endogenously expressing the GLP-1R (Figure 2C). A high concentration of the Akt inhibitor MK-2206 (20 μM) did not affect liraglutide-induced PKA motif phosphorylation of Raptor in CHO-GLP-1R (Figure 2A and B) or β-TC3 (Figure 2C) cells. Thus, liraglutide stimulates PKA phosphorylation of Raptor at Ser791.
PKA phosphorylates Raptor at S791 upon GLP-1R activation (Resubmission Rev 2 <xref ref-type="supplementary-material" rid="fig2sdata1">Figure 2—source data 1</xref>).
(A) Immunoblot and quantification of phospho-PKA substrate (pRRXS*/T*) and Myc in CHO-K1 cells stably expressing the hGLP1R transfected with Myc-wild-type (WT) Raptor or Myc-Ser791Ala Raptor and treated with vehicle (DMSO) or H89 (10 µM) for 30 min followed by treatment with Lira (10 nM), forskolin (Fsk, 10 μM), or insulin (Ins, 10 mg/mL) for 1 hr (Interaction: F (4, 21) = 3.695, p=0.0071; Genotype: F (2, 21) = 10.51, p=0.0007; Treatment: F (2, 21) = 6.328, p=0.0071). (B) Immunoblot and quantification of phospho-PKA substrate and Myc in CHO-K1 cells stably expressing the hGLP1R transfected with Myc-wild-type (WT) Raptor and treated with vehicle (DMSO) or MK-2206 (20 μM) for 30 min followed by treatment with Lira (10 nM), Fsk (10 μM), or Ins (10 mg/mL) for 1 hr (Interaction: F (3, 16) = 0.4599, p=0.7141; Inhibitor: F (1, 16) = 0.04511, p=0.8345; Treatment: F (3, 16) = 44.26, p<0.0001). (C) Immunoblot and quantification of phospho-PKA substrate and Myc in β-TC3 cells transfected with Myc-wild-type (WT) Raptor or Myc-Ser791Ala Raptor and treated with vehicle (DMSO) or MK-2206 (10 µM) for 30 min followed by treatment with Lira (10 nM), forskolin (Fsk, 10 μM), or insulin (Ins, 10 mg/mL) for 1 hr (Interaction: F (3, 16) = 7.345, p=0.0026; Genotype: F (1, 16) = 22.18, p=0.0002; Treatment: F (3, 16) = 10.93, p=0.0004). Analysis was done using two-way ANOVA followed by Tukey’s multiple comparisons. Data are shown as mean ± SEM, no symbol = not significant, * p<0.05, ** p<0.01, *** p<0.001, n=3–4 biological replicates.
<xref ref-type="fig" rid="fig2">Figure 2</xref>, raw blots from Figure 2, and raw data for <xref ref-type="fig" rid="fig2">Figure 2</xref>.
PKA phosphorylation of Raptor at Ser<sup>791</sup> mediates GLP-1R agonist-induced weight loss
We tested the physiological role of PKA phosphorylation of Raptor at Ser791 by measuring body weight and glycemic responses to liraglutide in wild-type (WT) and knock-in mice in which endogenous Raptor is replaced by a PKA-resistant Raptor expressing alanine instead of serine at position 791 in all tissues (CMV-Ser791Ala Raptor). WT and CMV-Ser791Ala Raptor knock-in mice lost weight in response to liraglutide treatment, but liraglutide-induced weight loss was attenuated in the CMV-Ser791Ala Raptor knock-in mice (Figure 3A–D). This was primarily due to a lesser reduction in fat mass in liraglutide-treated CMV-Ser791Ala Raptor knock-in compared to WT mice (Figure 3E–F). There was no significant difference between genotypes in the ability of liraglutide to reduce lean (Figure 3G–H) or free fluid (not shown) mass. Metabolic cage experiments demonstrate that food intake was equally suppressed in WT and CMV-Ser791Ala Raptor knock-in mice during the first day of liraglutide treatment. However, food intake returned to pre-liraglutide treatment levels in CMV-Ser791Ala Raptor knock-in mice by the second day of treatment whereas they did not return to pre-liraglutide treatment levels in WT mice until the fifth day of dosing (Figure 3I–J). Energy expenditure (EE) decreased in response to liraglutide in both genotypes and remained significantly lower in CMV-Ser791Ala Raptor knock-in mice compared to their vehicle counterparts for up to 8 days of dosing (Figure 3K–L). In WT mice, EE only remained lower in liraglutide-treated mice for 3 days (Figure 3K–L). These data suggest that phosphorylation of Raptor at Ser791 is required for the full body weight-lowering effect of liraglutide via regulation of both food intake and EE.
PKA phosphorylation of Raptor at S791 contributes to the weight-lowering effects of liraglutide in lean mice (Resubmission Rev 2 <xref ref-type="supplementary-material" rid="fig3sdata1">Figure 3—source data 1</xref>).
Body weight, body composition, and energy balance parameters in male wild-type (WT) and CMV-Ser791Ala Raptor knock-in (S791A) mice given vehicle (0.9% saline) or liraglutide (200 μg/kg) subcutaneously twice a day for 14 days. Absolute body weight (A), total mass (C), fat mass (E), and lean mass (G). Change in body weight (B), total mass (D), fat mass (F), and lean mass (H) from baseline (Day 0). Absolute daily food intake (I) and EE (J) and change in daily food intake (K) and EE (L) relative to baseline (Day 0) during a 10-day treatment period in metabolic cages. ((A), Interaction: F (42, 700)=14.10, p<0.0001. (B), Interaction: F (42, 700)=14.10, p<0.0001. (C), Interaction: F (3, 50)=42.94, p<0.0001. (D), F (3, 50)=42.94, p<0.0001. (E), Interaction: F (3, 36)=30.94, p<0.0001. (F), F (3, 36)=30.94, p<0.0001. (G), Interaction: F (3, 36)=12.19, p<0.0001. (H), F (3, 36)=12.19, p<0.0001. (I), Interaction: F (30, 112)=1.370, p<0.0213. (J), Interaction: F (30, 112)=1.706, p<0.0240. (K), Interaction: F (30, 120)=3.827, p<0.0001. (L), Interaction: F (30, 112)=3.827, p<0.0001) Analysis was done using mixed-effects analysis followed by Holm-Sidak’s multiple comparisons between liraglutide-treated WT and CMV-Ser791Ala Raptor mice. Data are shown as mean ± SEM, * p<0.05, ** p<0.01, **** p<0.0001, n=13–14 (A–D), 9–12 (E–H), or 4 (I–L) mice per group.
<xref ref-type="fig" rid="fig3">Figure 3</xref> and raw data for <xref ref-type="fig" rid="fig3">Figure 3</xref>.
Blood glucose prior to drug treatment and blood glucose and insulin post-treatment in male wild-type (WT) and Ser<sup>791</sup>Ala Raptor knock-in mice (Resubmission Rev 2 <xref ref-type="supplementary-material" rid="fig3s1sdata1">Figure 3—figure supplement 1—source data 1</xref>).
(A–D) Total (A), fat (B), lean (C), and free body fluid (D) measurements in WT and CMV-Ser791Ala Raptor mice prior to vehicle or liraglutide administration. (E– F) 5 hr fasting blood glucose in male WT and CMV-Ser791Ala Raptor mice before drug treatment (E) and on treatment day 14 (F). (G) 5 hr fasting insulin in male WT and CMV-Ser791Ala Raptor mice before on treatment day 14. Analysis was done using one-way ANOVA followed by Holm-Sidak’s multiple comparisons. Data are shown as mean ± SEM, ns not significant, n=4–8 mice per group.
<xref ref-type="fig" rid="fig3s1">Figure 3—figure supplement 1</xref> and raw data for <xref ref-type="fig" rid="fig3s1">Figure 3—figure supplement 1</xref>.
Generation of CMV-Ser<sup>791</sup>Ala Raptor knock-in mice.
(A) Targeting vector used by Ingenious Targeting Laboratory, Inc to introduce a Cre-inducible two base pair point mutation in an inverted exon 20 of the Raptor coding sequence. (B) Verification of correct target sequence expression by genomic DNA digestion. Wild-type (Control) DNA contains a SacI sequence that is absent in the Ser791Ala point mutant (red). SacI digestion of a subcloned 911 base pair (bp) genomic DNA fragment obtained from liver tissue of control mice produces two products at 371 bp and 540 bp. The corresponding 989 bp genomic DNA sequence obtained from CMV-Ser791Ala Raptor knock-in mice cannot be digested by SacI.
Powerpoint slide of <xref ref-type="fig" rid="fig3s2">Figure 3—figure supplement 2</xref>.
There was no difference between genotypes in fasting blood glucose prior to liraglutide treatment (Figure 3—figure supplement 1). Two weeks of liraglutide treatment lowered fasting glucose to similar degrees in both genotypes, and there was no difference in fasting plasma insulin levels (Figure 3—figure supplement 1).
We fed a separate cohort of control and CMV-Ser791Ala Raptor knock-in mice a 60% high-fat diet (HFD) to test whether PKA phosphorylation of Raptor at Ser791 is also required for the weight loss effect of liraglutide in the setting of obesity. Unlike lean mice, obese CMV-Ser791Ala mice were not significantly protected from weight loss induced by liraglutide (Figure 4A–D). One caveat to interpreting these results is that CMV-Ser791Ala mice gained significantly more weight and fat mass when fed the 60% HFD prior to liraglutide treatment compared to wild-type mice (Figure 4C and E). We have anecdotally observed that mice with higher baseline body weight and fat mass tend to lose more weight in response to liraglutide. We now support this by showing a positive correlation between starting body weight and liraglutide-induced weight loss in both wild-type and CMV-Ser791Ala mice (Figure 4—figure supplement 1). Although there is no difference in the relationship between starting body weight and liraglutide-induced weight loss between wild-type and CMV-Ser791Ala mice (Figure 4—figure supplement 1), we determined the average baseline (pre-liraglutide treatment) body weight for both genotypes (43.4 g) and divided the results into mice with baseline body weights above vs. below 43.4 g (Figure 4—figure supplement 1). There was no difference in liraglutide-induced weight loss between wild-type and CMV-Ser791Ala mice in mice with baseline body weights above 43.4 g (Figure 4I–J). In mice with baseline body weights below 43.4 g, there was a tendency for CMV-Ser791Ala mice to lose less body weight in response to liraglutide (Figure 4K–L). These results suggest that PKA phosphorylation of Raptor may contribute to liraglutide-induced weight loss in obese mice below a certain threshold body weight.
PKA phosphorylation of Raptor at S791 does not significantly contribute to the weight-lowering effects of liraglutide in obese mice (Resubmission Rev 2 <xref ref-type="supplementary-material" rid="fig4sdata1">Figure 4—source data 1</xref>).
Body weight in male wild-type (WT) and CMV-Ser791Ala Raptor knock-in (S791A) mice fed a 60% HFD for 8–10 weeks and then given vehicle (0.9% saline) or liraglutide (200 μg/kg) subcutaneously twice a day for 9 days. Absolute body weight (A) and change in body weight (B) from baseline (Day 0). Absolute total mass (C), fat mass (E), and lean mass (G) on Days 0 and 9. Change in total mass (D), fat mass (F), and lean mass (H) from baseline (Day 0). Change in body weight in mice with a baseline body weight >43.4 g (I, J) and mice with a baseline body weight <43.4 g (K, L). ((A), Interaction: F (27, 297)=22.22, p<0.0001. (B), Interaction: F (27, 297)=22.22, p<0.0001. (C), Interaction: F (3, 33) = 39.82, p<0.0001. (D), F (3, 33) = 39.82, p<0.0001. (E), Interaction: F (3, 33) = 30.17, p<0.0001. (F), F (3, 33) = 30.17, p<0.0001. (G), Interaction: F (3, 33) = 25.69, p<0.0001. (H), F (3, 33) = 25.69, p<0.0001. (I), Interaction: F (27, 126)=40.94, p<0.0001. (J), F (3, 14) = 131.2, p<0.0001. (K), Interaction: F (27, 135)=17.98, p<0.0001. (L), F (3, 15) = 29.48, p<0.0001). Analysis was done using mixed-effects analysis followed by Holm-Sidak’s multiple comparisons between liraglutide-treated WT and CMV-Ser791Ala Raptor mice. Data are shown as mean ± SEM, * p<0.05, ** p<0.01, n=7–12 mice per group (A – H), 3–5 (I, J), and 2–7 (K, L).
<xref ref-type="fig" rid="fig4">Figure 4</xref> and raw data for <xref ref-type="fig" rid="fig4">Figure 4</xref>.
Relationship between baseline body weight and liraglutide-induced weight loss in male wild-type (WT) and Ser<sup>791</sup>Ala Raptor knock-in mice (Resubmission Rev 2 <xref ref-type="supplementary-material" rid="fig4s1sdata1">Figure 4—figure supplement 1—source data 1</xref>).
(A) Relationship between baseline body weight and liraglutide-induced weight loss in HFD-fed wild-type (WT) and CMV-Ser791Ala Raptor knock-in (S791A) mice (Test if slopes are non-zero: WT, F (1,10) = 9.666, p<0.01; S791A Raptor: F (1, 17) = 36.714, p<0.001. Test if slopes are different: WT vs. S791A: F (1, 17) = 1.266, p=0.28). Absolute body weight in mice with >43.3 g baseline body weight (B) and <43.3 g baseline body weight (C). ((A), F (3, 14) = 9.005, p=0.0014. (B), F (3, 15) = 0.3796, p=0.7691). Analysis was done using linear regression or one-way ANOVA followed by Holm-Sidak’s multiple comparisons. Data are shown as mean ± SEM, ** p<0.01, n=3–5 (B) and 2–7 (C) mice per group.
<xref ref-type="fig" rid="fig4s1">Figure 4—figure supplement 1</xref> and its raw data for <xref ref-type="fig" rid="fig4s1">Figure 4—figure supplement 1</xref>.
PKA phosphorylation of Raptor at Ser<sup>791</sup> does not mediate melanocortin-4 receptor (MC4R) agonist-induced weight loss
To test whether Raptor phosphorylation at Ser791 is required exclusively for the weight loss effect of liraglutide or is necessary for other weight loss drugs, particularly those targeting Gαs-coupled receptors, we treated CMV-Ser791Ala Raptor knock-in mice with the MC4R agonist setmelanotide. Like the GLP-1R, the MC4R is also a Gαs-coupled receptor, and setmelanotide is approved for weight loss in individuals with monogenic obesity caused by mutations in pro-opiomelanocortin, proprotein convertase subtilisin/kexin type 1, or leptin receptor deficiency and in individuals with Bardet-Biedl syndrome (Markham, 2021). As shown in Figure 5, CMV-Ser791Ala Raptor knock-in mice were not resistant to weight loss in response to setmelanotide. This suggests that Raptor phosphorylation at Ser791 by PKA contributes to the weight loss effect of GLP-1R agonists but not another Gαs-targeting pharmacological agent such as setmelanotide.
PKA phosphorylation of Raptor at S791 is not required for the weight-lowering effect of setmelanotide (Resubmission Rev 2 <xref ref-type="supplementary-material" rid="fig5sdata1">Figure 5—source data 1</xref>).
Body weight and body composition in male wild-type (WT) and CMV-Ser791Ala Raptor knock-in (S791A) mice given vehicle (0.9% saline) or setmelanotide (4 mg/kg) subcutaneously once a day for 9 days. Absolute body weight (A), total mass (C), fat mass (E), and lean mass (G). Change in body weight (B), total mass (D), fat mass (F), and lean mass (H) from baseline (Day 0). (A), Interaction: F (27, 171)=4.955, p<0.0001. (B), Interaction: F (27, 171)=4.600, p<0.0001. (C), Interaction: F (3, 19) = 12.93, p<0.0001. (D, F) (3, 19) = 11.28, p=0.0002. (E), Interaction: F (3, 19) = 6.087, p=0.0044. (F), F (3, 19) = 6.087, p=0.0044. (G), Interaction: F (3, 19) = 1.311, p=0.2999. (H, F) (3, 1) = 1.311, p=0.2999. Analysis was done using mixed-effects analysis followed by Holm-Sidak’s multiple comparisons between liraglutide-treated WT and CMV-Ser791Ala Raptor mice. Data are shown as mean ± SEM, * p<0.05, ** p<0.01, n=5–7 mice per group.
<xref ref-type="fig" rid="fig5">Figure 5</xref> and raw data for <xref ref-type="fig" rid="fig5">Figure 5</xref>.
Discussion
GLP-1R-stimulated mTORC1 signaling in the VMH reduces food intake (Burmeister et al., 2017). This is consistent with studies showing that activation of hypothalamic mTORC1 reduces food intake and mediates weight lowering by leptin (Blouet et al., 2008; Cota et al., 2008; Cota et al., 2006) . We show that PKA phosphorylates the mTORC1-regulatory protein Raptor at Ser791 in response to liraglutide and that this is responsible for ~50% of liraglutide-induced weight loss. Given the complexity of signaling mechanisms downstream of the GLP-1R (Fletcher et al., 2016), assigning specific pathways to a desirable phenotype could be leveraged towards more effective weight loss therapeutics.
To our knowledge, our study is the first to describe a PKA-mTORC1 interaction and its physiological relevance in the context of GLP-1R signaling. PKA is reported to either stimulate (Liu et al., 2016) or inhibit (Jewell et al., 2019) mTORC1 activity. Similar to our study, (Liu et al., 2016) show PKA-mediated stimulation of mTORC1 in cells incubated in serum-free media. This contrasts with inhibition of mTORC1 in cells cultured in serum-supplemented media reported by Jewell et al., 2019. Since mTORC1 is a cellular energy sensor, PKA exerts differential actions on mTORC1 depending on nutrient availability or cellular signaling and energy status, as originally shown by Scott and Lawrence, 1998 demonstrating that forskolin inhibits insulin-stimulated S6 phosphorylation and also subsequently shown by Liu and colleagues (Liu et al., 2016).
In vivo results using novel PKA-resistant Raptor knockin mice suggest that PKA phosphorylation of Raptor contributes to the weight loss effect of liraglutide in lean but not in obese mice. One limitation of these in vivo studies is that PKA-resistant Raptor is expressed in all tissues. Loss of PKA phosphorylation of Raptor in one or multiple tissues may contribute to our observation that CMV-Ser791Ala mice gained significantly more weight on the HFD compared to control mice. This difference in starting body weight could explain why liraglutide-induced weight loss was not attenuated in obese CMV-Ser791Ala mice, as in lean mice, since heavier mice tend to be more responsive to liraglutide (Figure 4—figure supplement 1). This could offset any reduction in liraglutide responsiveness resulting from expression of PKA-resistant Raptor. We support this by showing that CMV-Ser791Ala mice with lower initial body weights display a tendency towards reduced responsiveness to liraglutide. Future studies will circumvent the limitations of the CMV-Ser791Ala model by introducing the Ser791Ala mutation to specific tissues/cell types that specifically mediate liraglutide-induced weight loss. The brain is a key target for the weight-lowering effects of GLP-1R agonists (Adams et al., 2018; Sisley et al., 2014; Varin et al., 2019). Regulation of energy balance by mTORC1 appears to be most critical in hypothalamic regions such as the arcuate (ARC), paraventricular hypothalamus (PVH), VMH, and suprachiasmatic nucleus (Cota et al., 2006; Inhoff et al., 2010; Proulx et al., 2008), where the GLP-1R is highly expressed (Jensen et al., 2018; Cork et al., 2015). The ARC mediates weight loss effects of liraglutide (Secher et al., 2014) and is engaged by systemic liraglutide (Salinas et al., 2018). Hindbrain regions are also engaged by systemic liraglutide and mediate its weight-lowering effects (Salinas et al., 2018; Fortin et al., 2020). Inhibition of PKA attenuates the anorectic effect of hindbrain-targeted Ex4 in rats (Hayes et al., 2011). Future studies expressing Ser791Ala Raptor in a brain region/cell-type specific manner will enable identification of neuronal population(s) where a PKA-mTORC1 interaction mediates the anorectic effect of GLP-1R agonists.
The observation that CMV-Ser791Ala Raptor knock-in mice were not resistant to the weight loss effect of the MC4R agonist setmelanotide suggests that PKA-mediated phosphorylation of Raptor at Ser791 is not a general mechanism for all Gαs-coupled receptor-targeting weight loss drugs. However, as with our results using liraglutide in HFD-fed mice, a caveat to interpreting our findings with setmelanotide is the use of the whole-body CMV-Ser791Ala mouse model. Since the weight loss effect of setmelanotide is primarily mediated via activation of the MC4R in the PVH, future studies will test the hypothesis that the weight loss effect of setmelanotide will be blunted in PVH MC4R-specific Ser791Ala Raptor knock-in mice. This could establish a role for PKA-mediated regulation of mTORC1 as a key mechanism for weight loss along an ARC (target of GLP-1R agonists)-PVH (target of MC4R agonists) axis.
Our results suggest that liraglutide does not regulate mTORC1 via Akt, as was observed for βAR signaling in adipocytes (Liu et al., 2016). Inhibition of Akt activity in the hindbrain (Rupprecht et al., 2013) or hypothalamus (Yang et al., 2017) attenuates the anorectic effect of Ex4, so identifying targets of Akt mediating the effects of GLP-1R agonists requires further study. AMP-activated protein kinase (AMPK) is also implicated in the metabolic actions of GLP-1R agonists. Studies from our lab and others show that GLP-1R agonists reduce food intake by inhibiting AMPK (Burmeister et al., 2017; Hayes et al., 2011; Burmeister et al., 2013; Sandoval et al., 2012). Since AMPK suppresses mTORC1 by phosphorylating TSC2 (Inoki et al., 2003) or Raptor (Gwinn et al., 2008), GLP-1R activation could stimulate mTORC1 by reducing AMPK activity. AMPK can also be inhibited by mTORC1 signaling (Dagon et al., 2012), so GLP-1R-mediated inhibition of AMPK may be downstream of mTORC1.
In summary, we have uncovered a novel signaling mechanism downstream of the GLP-1R characterized by PKA-mediated phosphorylation of Raptor and increased mTORC1 signaling that contributes to the weight-lowering effect of liraglutide. Given the variability in response and adverse effects (e.g. nausea) of current GLP-1-based drugs, the identification of therapeutically relevant signaling profiles downstream of the GLP-1R holds promise for improving efficacy and reducing side effects of these drugs. Furthermore, our findings provide a new framework for future investigations into the significance of the PKA-mTORC1 interaction to other beneficial metabolic effects of GLP-1R agonists. GLP-1R agonists continue to be used as glucose-lowering agents for type 2 diabetes based on their insulinotropic effect via pancreatic β-cells (Drucker, 2022). The PKA-mTORC1 interaction could contribute to this or another β-cell phenotype associated with GLP-1R agonists (e.g. proliferation/cytoprotection). Furthermore, GLP-1R agonists also display cardioprotective effects that could be mediated directly via effects on the heart and vasculature and/or be secondary to the ability of these drugs to improve glycemia, body weight, lipid profiles, and inflammation (Ussher and Drucker, 2023). This provides an opportunity to investigate whether the PKA-mTORC1 interaction contributes to the cardioprotective effects of GLP-1R agonists via direct or indirect mechanisms.
Materials and methodsCell culture
Chinese Hamster Ovary cells expressing human GLP-1R (CHO-GLP-1R; DiscoveRx, Fremont, CA) were maintained at 37 °C (5% CO2, humidified) in Ham’s F12 media containing 10 mM glucose, 1 mM pyruvate, and 1 mM glutamine (Cat#11765054, Gibco) supplemented with 10% fetal bovine serum (FBS; Cat#16140071, Gibco), 10 ml/l penicillin-streptomycin (Cat#91, Hyclone), and 0.8 mg/ml Geneticin G418 Sulfate (Cat#10131027, Gibco). β-TC3 cells (Efrat et al., 1988) were gifted by Dr. Richard O’Brien and cultured in DMEM media containing 25 mM glucose, 1 mM pyruvate, and 1 mM glutamine (Cat#11995065, Gibco) supplemented with 15% FBS, 2% horse serum (Cat#26050088, Gibco) and 10 ml/l penicillin-streptomycin. All cells were authenticated by STR profiling (ATCC) and were verified to be free of mycoplasma. Cells were washed twice with serum-free Ham’s F12 (CHO-GLP-1R) or DMEM (β-TC3) media containing 5.5 mM glucose for 3 hr and were treated with vehicle (1 X PBS) or liraglutide (Victoza, Novo Nordisk) for 1 hr with or without 30 min pre-treatment with vehicle (DMSO), PKA inhibitors H89 (Cat#2910, Tocris Bioscience) or KT5720 (Cat#1288, Tocris Bioscience), or the Akt inhibitor MK-2207 (Cat#S1078, Selleck Chemicals LLC) at concentrations stated in the figure legends. The time and dose of liraglutide treatment was determined by pilot dose response and time course experiments (Figure 1—figure supplement 2).
Immunoblotting
Cells were washed once with ice-cold 1 X PBS and lysed in RIPA buffer (Cat#R0278, Sigma-Aldrich) containing 1 X protease and phosphatase inhibitors. After centrifugation of cell homogenates (17,000xg, 10 min, 4 °C), supernatant protein concentration was determined by BCA protein assay (Pierce BCA Protein Assay Kit, Cat#23225, Thermo Fisher Scientific). Twenty μg/lane was run (200 V, 45 min) in 4–20% Tris-glycine gels (Invitrogen) and transferred (100 V, 1 hr, 4 °C) onto PVDF membranes (Bio-Rad). Blots were blocked in buffer containing 5% BSA (Cat#A9418, Sigma-Aldrich) in 1 X Tris-Buffered Saline and 0.1% Tween–20 for 1 hr and incubated in primary antibodies (phospho-S6 Ser235/236, Cat#4858, RRID:AB_916156; total-S6, Cat#2317, RRID:AB_2238583; p-CREB, Cat#9198, RRID:AB_2561044; total-CREB, Cat#9104, RRIS:AB_490881; phospho-Akt Thr308, Cat#9275, RRID:AB_329828; total-Akt, Cat#9272, RRID:AB_329827; and Myc tag, Cat#2276, RRID:AB_2314825; Cell Signaling Technology) overnight at 4 °C and secondary antibodies (IRDye 800CW Goat anti-Rabbit IgG, Cat#926–32211, RRID:AB_621843 and IRDye 680RD Goat anti-Mouse IgG, Cat#926–68070, RRID:AB_10956588 LI-COR Biosciences) for 1 hr at room temperature. Blots were imaged with an Odyssey Imaging System and full lanes were analyzed using ImageJ.
Immunoprecipitation
Myc-tagged wild-type (WT) and PKA-resistant mutant (Ser791Ala) Raptor vectors (Liu et al., 2016) were transfected into CHO-GLP-1R cells using lipofectamine-3000 (Invitrogen). Forty-eight hours later, cells were washed 2 X and maintained in serum-free media for 3 hr. Cells were treated with DMSO, H89 or MK-2207 for 30 min and treated with 1 X PBS, liraglutide, forskolin, or insulin for 1 hr at concentrations stated in the figure legend. Cells were washed once with ice-cold 1 X PBS and lysed in lysis buffer (25 mM Tris HCl, 131 150 mM NaCl, 1 µM EDTA, 1 µM EGTA, 2.5 mM sodium pyrophosphate, 100 mM sodium fluoride, 1 mM sodium orthovanadate) containing 1 X protease and phosphatase inhibitors. After centrifugation of cell homogenates (17,000xg, 10 min, 4 °C) supernatant protein concentration was determined by BCA protein assay. A total of 200 μg protein was incubated with anti–c-Myc affinity gel (EZview Red Anti-c-Myc Affinity Gel, #E6654, Sigma-Aldrich) overnight. Beads were washed in lysis buffer and immunoblotting was done as described above using a PKA substrate monoclonal antibody (Cat#9624, Cell Signaling Technology).
Animals
Mice were kept on a 12 hr/12 hr light/dark cycle with ad libitum access to water and chow (5L0D, LabDiet) from weaning and were studied at 12–16 weeks old. Procedures were approved by the Institutional Animal Care and Use Committee at Vanderbilt University (Protocol #M1500158). Ser791Ala Raptor knock-in mice were generated by inGenious Targeting Laboratory Inc using the targeting vector shown in Figure 3—figure supplement 2. The indicated loxP sites were engineered into the genomic DNA and a two base pair point mutation was introduced into a second inverted exon 20 of the Raptor coding sequence to change the codon for Ser at position 791 to Ala. Raptorfl/fl mice (013188, The Jackson Laboratory) were crossed to mice expressing the Ingenious construct, and offspring were then crossed to CMV-Cre (006054, The Jackson Laboratory) mice to generate hemizygous wild-type mice or hemizygous mice with global knock-in of Raptor Ser791Ala. This strategy was chosen because the Ingenious construct in homozygosity prior to Cre recombinase inadvertently created a hairpin loop in the transcript between the two exon 20 regions that was spliced out. Genotyping and cDNA sequencing confirmed expression of Ser791Ala Raptor in place of endogenous Raptor (Figure 3, Figure 3—figure supplement 2). Male mice were switched to either a low-fat diet (D12450J, Research Diets, Inc) for 4 weeks or a 60% HFD (D12492, Research Diets, Inc) for 8–10 weeks prior to initiation of dosing. Mice were maintained on the low-fat diet and were administered vehicle (0.9% saline), liraglutide (200 μg/kg body weight, BID), or setmelanotide (4 mg/kg body weight once daily) subcutaneously for 9–14 days. Body weight was measured daily, and body composition was obtained by NMR (Minispec 235 LF90II-TD NMR Analyzer, Bruker) prior to and on the last day of dosing. Four hours fasting blood glucose and insulin levels were measured using a Contour Next EZ glucose meter (Cat#BD15789, Bayer) and an Ultra Sensitive Mouse Insulin ELISA kit (Crystal Chem), respectively. A separate cohort of male mice was acclimated to a Promethion System (Sable Systems International, Inc) for one week prior to undergoing dosing with vehicle or liraglutide as described above for 10 days. Food intake and energy expenditure were measured continuously.
Data analysis
Data were analyzed using Prism-9 (GraphPad Software, Inc). One-way or two-way ANOVA or mixed-effects analysis followed by Tukey’s, Holm-Sidak’s, or Brown-Forsythe multiple comparisons was used when appropriate and indicated in the figure legends. Linear regression analysis was also used as indicated in the figure legends. Sample size was determined based on the variability of the primary measurements of interest (phosphorylation of S6 for cell culture studies and body weight for in vivo studies) and the objective of detecting significance (p<0.05) with 90% certainty. Treatments were randomized and blinded to the experimenter when possible.
Formal analysis, Investigation, Writing - review and editing
Investigation
Methodology
Formal analysis
Resources, Validation, Writing - review and editing
Conceptualization, Data curation, Supervision, Funding acquisition, Validation, Project administration, Writing - review and editing
Ethics
Procedures were approved by the Institutional Animal Care and Use Committee at Vanderbilt University (Protocol #M100158).
Additional filesData availability
All data generated or analysed during this study are included in the manuscript and supporting files.
Acknowledgements
We thank Wei Zhang for excellent technical assistance and mouse colony management. This work was supported by NIH R01 DK132852 and DK097361 (JEA) and NIH R01 DK116625 (SC). Metabolic cage experiments were conducted by the Vanderbilt Mouse Metabolic Phenotyping Center (VMMPC) supported by NIH U2C059637 and shared instrument grant S10OD028455 (JEA). We would like to thank Merrygay James from the VMMPC for her technical assistance.
ReferencesAdamsJMPeiHSandovalDASeeleyRJChangRBLiberlesSDOlsonDP2018Liraglutide modulates appetite and body weight through glucagon-like peptide 1 receptor-expressing glutamatergic neuronsDiabetes671538154810.2337/db17-138529776968BlouetCOnoHSchwartzGJ2008Mediobasal hypothalamic p70 S6 kinase 1 modulates the control of energy homeostasisCell Metabolism845946710.1016/j.cmet.2008.10.00419041762BrownJDMcAnallyDAyalaJEBurmeisterMAMorfaCSmithLAyalaJE2018Oleoylethanolamide modulates glucagon-like peptide-1 receptor agonist signaling and enhances exendin-4-mediated weight loss in obese miceAmerican Journal of Physiology. Regulatory, Integrative and Comparative Physiology315R595R60810.1152/ajpregu.00459.201729949410BurmeisterMAAyalaJDruckerDJAyalaJE2013Central glucagon-like peptide 1 receptor-induced anorexia requires glucose metabolism-mediated suppression of AMPK and is impaired by central fructoseAmerican Journal of Physiology. Endocrinology and Metabolism304E677E68510.1152/ajpendo.00446.201223341495BurmeisterMABrownJDAyalaJEStoffersDASandovalDASeeleyRJAyalaJE2017The glucagon-like peptide-1 receptor in the ventromedial hypothalamus reduces short-term food intake in male mice by regulating nutrient sensor activityAmerican Journal of Physiology-Endocrinology and Metabolism313E651E66210.1152/ajpendo.00113.2017CorkSCRichardsJEHoltMKGribbleFMReimannFTrappS2015Distribution and characterisation of Glucagon-like peptide-1 receptor expressing cells in the mouse brainMolecular Metabolism471873110.1016/j.molmet.2015.07.00826500843CotaDProulxKSmithKABKozmaSCThomasGWoodsSCSeeleyRJ2006Hypothalamic mTOR signaling regulates food intakeScience31292793010.1126/science.112414716690869CotaDMatterEKWoodsSCSeeleyRJ2008The role of hypothalamic mammalian target of rapamycin complex 1 signaling in diet-induced obesityThe Journal of Neuroscience287202720810.1523/JNEUROSCI.1389-08.200818614690DagonYHurEZhengBWellensteinKCantleyLCKahnBB2012p70S6 Kinase Phosphorylates AMPK on Serine 491 to mediate Leptin’s effect on food intakeCell Metabolism1610411210.1016/j.cmet.2012.05.01022727014DaviesSPReddyHCaivanoMCohenP2000Specificity and mechanism of action of some commonly used protein kinase inhibitorsThe Biochemical Journal3519510510.1042/0264-6021:351009510998351DruckerDJ2022GLP-1 physiology informs the pharmacotherapy of obesityMolecular Metabolism5710135110.1016/j.molmet.2021.10135134626851EfratSLindeSKofodHSpectorDDelannoyMGrantSHanahanDBaekkeskovS1988Beta-cell lines derived from transgenic mice expressing a hybrid insulin gene-oncogenePNAS859037904110.1073/pnas.85.23.90372848253FletcherMMHallsMLChristopoulosASextonPMWoottenD2016The complexity of signalling mediated by the glucagon-like peptide-1 receptorBiochemical Society Transactions4458258810.1042/BST2015024427068973FortinSMLipskyRKLhamoRChenJKimEBornerTSchmidtHDHayesMR2020GABA neurons in the nucleus tractus solitarius express GLP-1 receptors and mediate anorectic effects of liraglutide in ratsScience Translational Medicine12eaay807110.1126/scitranslmed.aay807132132220GwinnDMShackelfordDBEganDFMihaylovaMMMeryAVasquezDSTurkBEShawRJ2008AMPK phosphorylation of raptor mediates a metabolic checkpointMolecular Cell3021422610.1016/j.molcel.2008.03.00318439900HayesMRLeichnerTMZhaoSLeeGSChowanskyAZimmerDDe JongheBCKanoskiSEGrillHJBenceKK2011Intracellular signals mediating the food intake-suppressive effects of hindbrain glucagon-like peptide-1 receptor activationCell Metabolism1332033010.1016/j.cmet.2011.02.00121356521InhoffTStengelAPeterLGoebelMTachéYBannertNWiedenmannBKlappBFMönnikesHKobeltP2010Novel insight in distribution of nesfatin-1 and phospho-mTOR in the arcuate nucleus of the hypothalamus of ratsPeptides3125726210.1016/j.peptides.2009.11.02419961888InokiKZhuTGuanKL2003TSC2 mediates cellular energy response to control cell growth and survivalCell11557759010.1016/s0092-8674(03)00929-214651849JensenCBPykeCRaschMGDahlABKnudsenLBSecherA2018Characterization of the glucagonlike peptide-1 receptor in male mouse brain using a novel antibody and in situ hybridizationEndocrinology15966567510.1210/en.2017-0081229095968JewellJLFuVHongAWYuF-XMengDMelickCHWangHLamW-LMYuanH-XTaylorSSGuanK-L2019GPCR signaling inhibits mTORC1 via PKA phosphorylation of RaptoreLife8e4303810.7554/eLife.4303831112131KimDHSarbassovDDAliSMKingJELatekRRErdjument-BromageHTempstPSabatiniDM2002mTOR interacts with raptor to form a nutrient-sensitive complex that signals to the cell growth machineryCell11016317510.1016/s0092-8674(02)00808-512150925LiuDBordicchiaMZhangCFangHWeiWLiJLGuilhermeAGunturKCzechMPCollinsS2016Activation of mTORC1 is essential for β-adrenergic stimulation of adipose browningThe Journal of Clinical Investigation1261704171610.1172/JCI8353227018708MarkhamA2021Setmelanotide: First ApprovalDrugs8139740310.1007/s40265-021-01470-933638809ProulxKCotaDWoodsSCSeeleyRJ2008Fatty acid synthase inhibitors modulate energy balance via mammalian target of rapamycin complex 1 signaling in the central nervous systemDiabetes573231323810.2337/db07-169018776140RupprechtLEMietlicki-BaaseEGZimmerDJMcGrathLEOlivosDRHayesMR2013Hindbrain GLP-1 receptor-mediated suppression of food intake requires a PI3K-dependent decrease in phosphorylation of membrane-bound AktAmerican Journal of Physiology. Endocrinology and Metabolism305E751E75910.1152/ajpendo.00367.201323900416SalinasCBGLuTT-HGaberySMarstalKAlanentaloTMercerAJCorneaAConradsenKHecksher-SørensenJDahlABKnudsenLBSecherA2018Integrated brain atlas for unbiased mapping of nervous system effects following liraglutide treatmentScientific Reports81031010.1038/s41598-018-28496-629985439SandovalDBarreraJGStefaterMASisleySWoodsSCD’AlessioDDSeeleyRJ2012The anorectic effect of GLP-1 in rats is nutrient dependentPLOS ONE7e5187010.1371/journal.pone.005187023284795SaxtonRASabatiniDM2017mTOR Signaling in Growth, Metabolism, and DiseaseCell16936137110.1016/j.cell.2017.03.03528388417ScottPHLawrenceJC1998Attenuation of mammalian target of rapamycin activity by increased cAMP in 3T3-L1 adipocytesThe Journal of Biological Chemistry273344963450110.1074/jbc.273.51.344969852118SecherAJelsingJBaqueroAFHecksher-SørensenJCowleyMADalbøgeLSHansenGGroveKLPykeCRaunKSchäfferLTang-ChristensenMVermaSWitgenBMVrangNBjerre KnudsenL2014The arcuate nucleus mediates GLP-1 receptor agonist liraglutide-dependent weight lossThe Journal of Clinical Investigation1244473448810.1172/JCI7527625202980SisleySGutierrez-AguilarRScottMD’AlessioDASandovalDASeeleyRJ2014Neuronal GLP1R mediates liraglutide’s anorectic but not glucose-lowering effectThe Journal of Clinical Investigation1242456246310.1172/JCI7243424762441ThorensB1992Expression cloning of the pancreatic beta cell receptor for the gluco-incretin hormone glucagon-like peptide 1PNAS898641864510.1073/pnas.89.18.86411326760UssherJRDruckerDJ2023Glucagon-like peptide 1 receptor agonists: cardiovascular benefits and mechanisms of actionNature Reviews. Cardiology2046347410.1038/s41569-023-00849-336977782VarinEMMulvihillEEBaggioLLKoehlerJACaoXSeeleyRJDruckerDJ2019Distinct Neural sites of GLP-1R expression mediate physiological versus Pharmacological Control of Incretin ActionCell Reports273371338410.1016/j.celrep.2019.05.05531189118YangYChoiPPSmithWWXuWMaDCordnerZALiangNCMoranTH2017Exendin-4 reduces food intake via the PI3K/AKT signaling pathway in the hypothalamusScientific Reports7693610.1038/s41598-017-06951-02876113210.7554/eLife.80944.sa0Editor's evaluationBuettnerChristophReviewing Editorhttps://ror.org/05vt9qd57Rutgers Robert Wood Johnson Medical SchoolUnited States
This manuscript provides support for the importance of PKA-dependent mTORC1 activation for the weight-loss effects of Liraglutide, and presumably many GPCR-dependent anorectic agents, while not affecting the food intake-reducing effects of the MC4Ractivator setmelanotide. The weight loss effect of liraglutide however was not impaired after HFD feeding, a model of obesity where GLP1 agonists are being used.
10.7554/eLife.80944.sa1Decision letterBuettnerChristophReviewing Editorhttps://ror.org/05vt9qd57Rutgers Robert Wood Johnson Medical SchoolUnited StatesCotaDanielaReviewerhttps://ror.org/02vjkv261InsermFranceRenHongxiaReviewerhttps://ror.org/01kg8sb98Indiana University School of MedicineUnited States
Our editorial process produces two outputs: (i) public reviews designed to be posted alongside the preprint for the benefit of readers; (ii) feedback on the manuscript for the authors, including requests for revisions, shown below. We also include an acceptance summary that explains what the editors found interesting or important about the work.
Decision letter after peer review:
Thank you for submitting your article "Glucagon-Like Peptide-1 Receptor Activation Stimulates PKA-Mediated Phosphorylation of Raptor and this Contributes to the Weight Loss Effect of Liraglutide" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Mone Zaidi as the Senior Editor. The following individuals involved in the review of your submission have agreed to reveal their identity: Daniela Cota (Reviewer #2); Hongxia Ren (Reviewer #3).
The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.
Essential revisions:
The reviewers agreed that the paper has potential, but the authors will need to provide substantial new data to answer the reviewers' concerns.
1) Specifically, provide a comprehensive characterization of the mouse model in addition to the studies provided.
2) Demonstrate specificity of the weight-loss effects to GPCR-PKA signaling.
3) Provide a more comprehensive analysis of the changes in energy expenditure and provide a more comprehensive metabolic phenotyping (insulin levels etc).
Reviewer #1 (Recommendations for the authors):
1. While I understand that the authors intend to publish a full characterization of the model elsewhere, it is impossible to evaluate the results of this manuscript without some information on the genetics of the model, and some data showing the expression of the mutant, however. The authors could add some more detail about the mouse model to this manuscript and send the manuscript describing the model in more detail for the reviewers to examine.
2. The authors should examine the normal regulation of feeding/body weight by non-GPCR-dependent agents (e.g., leptin, GDF15, or many others). They would want to examine the effects of these on PKA/pS6 in cultured cells, as well.
3. The authors should provide direct comparisons of the various conditions, rather than normalizing to individual baseline measurements.
Reviewer #2 (Recommendations for the authors):
1. The authors claim that the difference in efficacy in weight loss in WT vs raptor mutants treated with Liraglutide is due to a very minor difference in food intake during the second day of treatment. By looking at the data, the sustained weight difference cannot be due to a very transitory difference in food intake, which by the end of the treatment has become comparable to the food intake of vehicle-treated mice. Authors should investigate changes in energy expenditure.
2. Related to the above, previously published evidence suggests that central GLP1R agonism stimulates thermogenesis by recruiting the sympathetic nervous system (see Kooijman et al., Diabetologia 2015). It would be relevant to assess whether recruitment of the SNS is impaired in the Raptor mutant mice under liraglutide treatment. Authors could assess molecular changes for SNS markers and markers of increased thermogenesis in the brown adipose tissue and in the white adipose tissue. Authors could also assess possible differences in cold-induced SNS activity in their model.
3. It is unclear why the authors did not test the relevance of their observations by inducing diet-induced obesity in their mouse model. Does the raptor mutant have a specific phenotype under diet-induced obesity?
4. Liraglutide importantly impacts glucose metabolism. Here the authors show that 2-weeks Liraglutide treatment decreases fasting glucose levels independent of genotype. The authors however do not provide any information neither on insulin levels or on insulin sensitivity in their animal model. The mTOR pathway (including raptor) is an important funneling site of the molecular action of insulin. The fact that fasting glucose is decreased similarly in WT and mutant mice, does not preclude possible changes in fasting insulin or during dynamic tests (insulin tolerance tests).
5. Statistical analyses carried out in figure 3 do not consider as independent factors the treatment and the genotype. Authors need to carry out 2-way anovas. Only if there is an interaction, the authors can conclude that for instance the total body mass (Figure 3C) or fat mass (Figure 3E) is different between liraglutide-treated WT and mutant mice.
[Editors’ note: further revisions were suggested prior to acceptance, as described below.]
Thank you for resubmitting your work entitled "Glucagon-Like Peptide-1 Receptor Activation Stimulates PKA-Mediated Phosphorylation of Raptor and this Contributes to the Weight Loss Effect of Liraglutide" for further consideration by eLife. Your revised article has been evaluated by Mone Zaidi (Senior Editor) and a Reviewing Editor.
The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:
All three reviewers and the reviewing editor feel that the revision only partly addressed the concerns raised. Specifically, the work lacks studies in diet-induced obesity which are required to explain the therapeutic mechanisms of GLP1 agonists, the failure to examine the anorectic response to other GPCR agonists to establish the specificity of the phosphorylation site for GLP1's signaling/effects versus that of other signaling pathways as the latter is what the authors imply.
Reviewer #1 (Recommendations for the authors):
The authors have made several useful changes to this revised manuscript, including adding additional information about the genesis of their mouse model. The validation remains a bit less complete than one would like, however- in the methods the authors state that they have sequenced the cDNA from these mice and refer to Supplemental Figure 1B, but the data are not provided here (or elsewhere).
I am also disappointed that the authors have not examined the anorectic response to other agents, as this would provide a useful and important control for the specificity of the phosphorylation site for Lira (or other GPCR agonists).
Reviewer #2 (Recommendations for the authors):
The authors have only partly addressed the concerns of this reviewer.
In particular, although already pointed out by this reviewer, the authors have decided to not carry out studies under diet-induced obesity. This reduces the interest for the data in their present form, which remain preliminary.
Liraglutide is used for weight loss in patients suffering from obesity and not to induce weight loss in lean individuals. The observations provided are made in non-obese mice. Whether what the authors observe still stands true in obesity and helps explain the therapeutic action of liraglutide remains to be demonstrated.
It is possible that mutant mice may have per se a phenotype in DIO, and this information would be relevant. Whether under DIO, mutant mice would respond similarly or differently from WT mice to Liraglutide would also be relevant information, even if this is a whole-body and not a cell-specific mutant, a limitation that could then be discussed by the authors.
Reviewer #3 (Recommendations for the authors):
The authors have made an effort to address the reviewers' comments. As a result, the quality of the manuscript was improved. However, the reviewer has the following questions to be further clarified.
1. Appropriate method of statistical analysis needs to be used for results presented in Figure 1 and the accompanying supplemental figure. The reviewer suggests to use 2-way ANOVA instead.
2. For signaling experiments using chemical probes, it is necessary to show the time course of the signaling dynamics and dose-dependent effect (as opposed to demonstrate the result using single dose and single time point).
10.7554/eLife.80944.sa2Author response
Essential revisions:
The reviewers agreed that the paper has potential, but the authors will need to provide substantial new data to answer the reviewers' concerns.
Reviewer #1 (Recommendations for the authors):
1. While I understand that the authors intend to publish a full characterization of the model elsewhere, it is impossible to evaluate the results of this manuscript without some information on the genetics of the model, and some data showing the expression of the mutant, however. The authors could add some more detail about the mouse model to this manuscript and send the manuscript describing the model in more detail for the reviewers to examine.
We gladly agree to this request and now provide detailed information about the Raptor knock-in mutant model in the Methods section (lines 206-215) and in Figure 3—figure supplement 2.
2. The authors should examine the normal regulation of feeding/body weight by non-GPCR-dependent agents (e.g., leptin, GDF15, or many others). They would want to examine the effects of these on PKA/pS6 in cultured cells, as well.
We agree with the reviewer that this is an important question that will be more appropriately addressed once we have identified the specific tissues/cell types in which Raptor signaling mediates the weight loss effect of GLP-1R agonists and generate the appropriate tissue/cell-specific Raptor mutant mouse line.
3. The authors should provide direct comparisons of the various conditions, rather than normalizing to individual baseline measurements.
We thank the reviewer for this comment. We now provide direct comparisons of the various conditions used in the cell culture experiments in Figure 1 (Figure 1—figure supplement 1) and in vivo experiments (Figure 3).
Reviewer #2 (Recommendations for the authors):
1. The authors claim that the difference in efficacy in weight loss in WT vs raptor mutants treated with Liraglutide is due to a very minor difference in food intake during the second day of treatment. By looking at the data, the sustained weight difference cannot be due to a very transitory difference in food intake, which by the end of the treatment has become comparable to the food intake of vehicle-treated mice. Authors should investigate changes in energy expenditure.
we agree with the Reviewer and have now conducted an experiment in metabolic cages. As now shown in Figure 3K-N, along with a more rapid rebound in food intake in the Ser791Ala Raptor mutant mice during liraglutide treatment, these mutants also display reduced energy expenditure relative to vehicle-treated mice that lasts for a longer duration during the liraglutide treatment period than in wild-type mice. We hypothesize that both of these phenotypes contribute to the relative resistance to liraglutide-induced weight loss in the Raptor mutant mice.
2. Related to the above, previously published evidence suggests that central GLP1R agonism stimulates thermogenesis by recruiting the sympathetic nervous system (see Kooijman et al., Diabetologia 2015). It would be relevant to assess whether recruitment of the SNS is impaired in the Raptor mutant mice under liraglutide treatment. Authors could assess molecular changes for SNS markers and markers of increased thermogenesis in the brown adipose tissue and in the white adipose tissue. Authors could also assess possible differences in cold-induced SNS activity in their model.
Although GLP-1R agonists have been shown to increase sympathetic tone and stimulate BAT thermogenesis in rodents, it does not appear that these phenotypes contribute to parameters that regulate body weight (i.e., food intake and energy expenditure). As shown by our results in Figure 3 and studies conducted by other groups (Baggio LL et al. Gastroenterology 2004; Knauf C et al. Endocrionology 2008) GLP-1R agonist administration actually lowers energy expenditure even in wild-type mice. However, our data do suggest that the duration of liraglutide-mediated reduction of energy expenditure is greater in the Raptor mutant knock-in mice compared to control mice, even though there is no significant difference in absolute energy expenditure between genotypes. One potential interpretation of this is that mechanisms that could contribute to restoring energy expenditure (e.g. thermogenesis) are activated in control mice, but this is impaired in the Raptor mutant knock-in mice. This is a hypothesis that would be better tested in tissue-specific Raptor mutant knock-in mice since this would avoid confounding effects of impaired PKA-mediated regulation of Raptor in multiple tissues (e.g., impaired PKA-Raptor signaling in adipose tissue).
The reviewer does raise an interesting point about whether Raptor signaling downstream of the GLP-1R my play a role in the response to cold exposure. However, testing this is beyond the scope of the present studies that focus on body weight control.
3. It is unclear why the authors did not test the relevance of their observations by inducing diet-induced obesity in their mouse model. Does the raptor mutant have a specific phenotype under diet-induced obesity?
This is a valid point raised by the reviewer, and it is something that we will be testing with tissue/cell-type specific Raptor mutant mice. The concern with using high fat diet-induced obesity with the whole-body Raptor mutant line is that high fat diet may provoke secondary phenotypes in this global mouse line even prior to liraglutide treatment that would complicate interpretation of results from liraglutide treatment.
4. Liraglutide importantly impacts glucose metabolism. Here the authors show that 2-weeks Liraglutide treatment decreases fasting glucose levels independent of genotype. The authors however do not provide any information neither on insulin levels or on insulin sensitivity in their animal model. The mTOR pathway (including raptor) is an important funneling site of the molecular action of insulin. The fact that fasting glucose is decreased similarly in WT and mutant mice, does not preclude possible changes in fasting insulin or during dynamic tests (insulin tolerance tests).
We agree with the reviewer and now include data showing that in addition to no effect of liraglutide on glucose levels in the Raptor mutant mice, there is also no difference between wild-type and Raptor mutant mice on fasting insulin levels (Figure 3—figure supplement 1). We also provide data for the Reviewer in Author response image 1 showing data that glucose tolerance in response to liraglutide treatment is equally improved in wild-type and Raptor mutant mice.
5. Statistical analyses carried out in figure 3 do not consider as independent factors the treatment and the genotype. Authors need to carry out 2-way anovas. Only if there is an interaction, the authors can conclude that for instance the total body mass (Figure 3C) or fat mass (Figure 3E) is different between liraglutide-treated WT and mutant mice.
We thank the Reviewer for this suggestion and have re-analyzed these results using 2-way ANOVA tests. There is a significant genotype and treatment interaction with time for the various comparisons in Figure 3 that show significant differences.
[Editors’ note: what follows is the authors’ response to the second round of review.]
Reviewer #1 (Recommendations for the authors):
The authors have made several useful changes to this revised manuscript, including adding additional information about the genesis of their mouse model. The validation remains a bit less complete than one would like, however- in the methods the authors state that they have sequenced the cDNA from these mice and refer to Supplemental Figure 1B, but the data are not provided here (or elsewhere).
We apologize for the confusion. The results of the sequencing validation are shown in new Figure 3—figure supplement 2. We also provide additional details for the design and creation of this mouse model in the Methods section (lines 280-283).
I am also disappointed that the authors have not examined the anorectic response to other agents, as this would provide a useful and important control for the specificity of the phosphorylation site for Lira (or other GPCR agonists).
Following this suggestion from the reviewer, we have now conducted additional experiments in which we assessed the weight loss effect of the melanocortin-4 receptor (MC4R) agonist setmelanotide in wild-type and CMV-Ser791Ala Raptor knockin mice. Like the Glp1r, the MC4R is a Gαs-coupled receptor, so this provides a valuable comparison to the Glp1r. As we now show in the new Figure 5, unlike what we observed with liraglutide, CMV-Ser791Ala Raptor knockin mice are not resistant to the weight loss effect of setmelanotide. This suggests that the phosphorylation of Raptor by PKA at Ser-791 does not contribute to the weight loss effect of all Gαs -coupled receptor agonists equally. These results are also described in the Results section (lines 141-152) and in the Discussion (lines 193-202).
Reviewer #2 (Recommendations for the authors):
The authors have only partly addressed the concerns of this reviewer.
In particular, although already pointed out by this reviewer, the authors have decided to not carry out studies under diet-induced obesity. This reduces the interest for the data in their present form, which remain preliminary.
Liraglutide is used for weight loss in patients suffering from obesity and not to induce weight loss in lean individuals. The observations provided are made in non-obese mice. Whether what the authors observe still stands true in obesity and helps explain the therapeutic action of liraglutide remains to be demonstrated.
It is possible that mutant mice may have per se a phenotype in DIO, and this information would be relevant. Whether under DIO, mutant mice would respond similarly or differently from WT mice to Liraglutide would also be relevant information, even if this is a whole-body and not a cell-specific mutant, a limitation that could then be discussed by the authors.
Following this suggestion from the reviewer, we have now conducted experiments treating wild-type and CMV-Ser791Ala Raptor knockin mice with liraglutide after having been rendered obese following feeding with a 60% high fat diet (HFD). As shown in the new Figure 4, unlike the results obtained in lean mice, obese CMV-Ser791Ala Raptor knockin mice were not resistant to the weight loss effect of liraglutide. However, it must be noted that the CMVSer791Ala Raptor knockin mice had significantly higher body weight and adiposity following the HFD exposure prior to any treatment (Figure 4C and 4E). It has been our experience that mice with higher starting body weight and adiposity tend to lose more weight in response to liraglutide. We support this with additional data in new Figure 4—figure supplement 1 showing a positive correlation between initial body weight and liraglutide-induced weight loss. This could offset any attenuation of the liraglutide effect in the CMV-Ser791Ala Raptor knockin mice. Therefore, we determined the average baseline body weight of all mice prior to liraglutide treatments (43.4g) and divided results into those from mice with baseline body weights above 43.4g and those with baseline body weights below 43.4g. We show that there is no difference in the weight loss effect of liraglutide between wild-type and CMV-Ser791Ala Raptor knockin mice that have and average baseline body weight above 43.4 g (Figure 4I, 4J); however, there is a tendency for the CMV-Ser791Ala Raptor knockin mice to be resistant to the weight loss effect of liraglutide when looking at mice whose baseline body weights are below 43.4g. These findings suggest that PKA phosphorylation of Raptor at Ser-791 may contribute to the weight loss effect of liraglutide in obese mice below a specific weight threshold. Results are described in the Results section (lines 121-140) and in the Discussion (lines 171-183).
Reviewer #3 (Recommendations for the authors):
The authors have made an effort to address the reviewers' comments. As a result, the quality of the manuscript was improved. However, the reviewer has the following questions to be further clarified.
1. Appropriate method of statistical analysis needs to be used for results presented in Figure 1 and the accompanying supplemental figure. The reviewer suggests to use 2-way ANOVA instead.
Following this recommendation from the reviewer, we have now re-analyzed these data using 2-way ANOVA.
2. For signaling experiments using chemical probes, it is necessary to show the time course of the signaling dynamics and dose-dependent effect (as opposed to demonstrate the result using single dose and single time point).
Following the recommendation from the reviewer, we show both dose response and time course experiments that were conducted in order to select the proper dose and time exposure of liraglutide for the signaling experiments (Figure 1—figure supplement 2, Methods lines 243-245). Different doses of the inhibitors are already shown in Figure 1.