GLP-1 at a Glance
Glucagon-like peptide-1 is a 30/31 amino acid incretin hormone that plays a central role in glucose homeostasis, appetite regulation, and cardiovascular protection.
Proglucagon Gene
Encoded by the GCG gene on chromosome 2q36.3. Tissue-specific post-translational processing by PC1/3 in L-cells yields GLP-1, GLP-2, oxyntomodulin, glicentin, and IP-2.
Half-life: 1.5–2 min
Native GLP-1 is rapidly degraded by dipeptidyl peptidase-4 (DPP-4) at the N-terminal alanine (position 2), yielding the inactive metabolite GLP-1(9-36) amide.
Class B1 GPCR
GLP-1 receptor is a class B1 G-protein-coupled receptor with a large N-terminal extracellular domain (ECD) that employs the two-domain binding model.
Incretin Effect
Accounts for 50–70% of postprandial insulin secretion. GLP-1 and GIP together mediate the incretin effect — the augmented insulin response to oral vs. IV glucose.
Pleiotropic Actions
Beyond glycaemia: cardioprotection, neuroprotection, appetite suppression, gastric emptying delay, hepatic steatosis reduction, anti-inflammatory effects, and renal protection.
Active Forms
Two bioactive isoforms: GLP-1(7-37) and GLP-1(7-36)NH₂ (predominant circulating form, ~80%). The amidated form has slightly higher receptor affinity.
Cellular Mechanism of GLP-1 Receptor Agonists
Visual demonstration of GLP-1 receptor mechanism of action
Gene Expression & Post-Translational Processing
The proglucagon gene undergoes remarkable tissue-specific differential processing to yield distinct peptide products in different cell types.
The proglucagon gene (GCG) encodes a 180-amino acid precursor protein — preproglucagon — that is expressed in intestinal enteroendocrine L-cells, pancreatic α-cells, and specific neurons of the nucleus tractus solitarius (NTS) and area postrema in the brainstem. The signal peptide is cleaved to generate the 160-amino acid proglucagon, which contains within its sequence glucagon, GLP-1, GLP-2, glicentin, oxyntomodulin, glicentin-related pancreatic polypeptide (GRPP), intervening peptide-1 (IP-1), and intervening peptide-2 (IP-2).
The critical determinant of which peptides are liberated is the proprotein convertase (PC) expressed in each tissue. In pancreatic α-cells, PC2 predominates, yielding glucagon, GRPP, IP-1, and the major proglucagon fragment (MPGF). In intestinal L-cells and brainstem neurons, PC1/3 predominates, liberating GLP-1, GLP-2, oxyntomodulin, glicentin, and IP-2 — while glucagon remains trapped within the glicentin and oxyntomodulin sequences.
GLP-1 binds its receptor via a "two-domain" mechanism: the C-terminal α-helix (residues ~16–36) first docks into the extracellular domain (ECD) cleft — this is the affinity trap. The N-terminal residues (His7-Gly15) then insert into the transmembrane domain (TMD) core junction region to trigger receptor activation and G-protein coupling. His7 is absolutely critical — its removal abolishes signalling. The Ala8 position is the DPP-4 cleavage target, explaining why N-terminal modifications (e.g., Aib substitution in semaglutide) confer DPP-4 resistance.
Stimuli for L-Cell GLP-1 Secretion
Glucose: Transported via SGLT1 (sodium-glucose co-transporter 1) on the apical membrane. The electrogenic Na⁺ entry depolarises the L-cell, opening voltage-gated Ca²⁺ channels (VGCCs) and triggering exocytosis. GLUT2 on the basolateral membrane also contributes. ATP generated from glucose metabolism closes KATP channels (SUR1/Kir6.2), further depolarising the cell — analogous to the β-cell glucose-sensing apparatus.
Lipids: Long-chain fatty acids activate GPR120 (FFAR4) and GPR40 (FFAR1) on L-cells. Monoacylglycerols activate GPR119, coupling to Gαs → cAMP → PKA/Epac2 → granule exocytosis. Oleoylethanolamide (OEA) and 2-oleoylglycerol (2-OG) are potent endogenous GPR119 agonists. Bile acids activate TGR5 (GPBAR1), a key mediator of postprandial GLP-1 release — particularly relevant after bariatric surgery where altered bile acid kinetics dramatically upregulate GLP-1 secretion.
Proteins & Amino Acids: Peptones and individual amino acids (glutamine, phenylalanine, tryptophan) stimulate GLP-1 release via CaSR (calcium-sensing receptor) and GPRC6A. Protein hydrolysates act through PepT1 transporter and intracellular signalling.
Short-Chain Fatty Acids (SCFAs): Produced by colonic microbiota fermentation of dietary fibre. Propionate and butyrate activate FFAR2 (GPR43) and FFAR3 (GPR41) on distal L-cells, linking the gut microbiome to incretin secretion.
Vagal Afferents: A proximal-to-distal "neuroincretin" axis exists. Nutrients in the proximal gut trigger vagal afferents → vagal efferents → acetylcholine release → M1/M2 muscarinic receptor activation on distal L-cells. This explains the "early phase" GLP-1 secretion (within 10–15 minutes of eating) before nutrients physically reach distal L-cells.
GIP Cross-Talk: GIP released from proximal K-cells may stimulate distal L-cell GLP-1 secretion via a paracrine/endocrine loop, amplifying the incretin signal.
Somatostatin: Tonically inhibits GLP-1 release via SSTR5 receptors on L-cells. Somatostatin from neighbouring D-cells acts as a paracrine brake on incretin secretion.
Sympathetic Nervous System: α-adrenergic stimulation inhibits, while β-adrenergic stimulation enhances GLP-1 secretion — relevant during stress and exercise states.
The gut microbiome profoundly influences GLP-1 secretion through multiple mechanisms. Short-chain fatty acids (SCFAs) — particularly acetate, propionate, and butyrate — produced by bacterial fermentation of dietary fibre activate FFAR2 and FFAR3 on colonic L-cells. Secondary bile acids generated by bacterial bile salt hydrolases (BSH) and 7α-dehydroxylases are potent TGR5 agonists. Indole produced from tryptophan metabolism by commensal bacteria modulates L-cell secretion via voltage-gated K⁺ channel inhibition. Lipopolysaccharide (LPS) from gram-negative bacteria has complex effects — chronic low-grade endotoxaemia impairs GLP-1 secretion, linking dysbiosis to impaired incretin function in obesity and T2DM.
Akkermansia muciniphila has been shown to enhance GLP-1 secretion and improve metabolic parameters, representing a therapeutic target in the microbiome-incretin axis.
GLP-1 Receptor — Structure & Activation
A Class B1 GPCR with unique structural features enabling peptide hormone recognition and diverse downstream signalling.
Interactive Receptor Domains
Click each domain to explore its structure and function in detail.
Extracellular Domain (ECD)
The ECD (~120 residues) forms a distinctive α-β-β-α fold stabilised by three conserved disulfide bonds (C46-C71, C62-C104, C85-C126) and contains a hydrophobic ligand-binding groove. This domain serves as the affinity trap for the C-terminal α-helix of GLP-1.
Key contact residues include Trp39, Tyr88, Leu89, Trp91, Pro90, and Glu128, which form hydrophobic and polar contacts with GLP-1 residues Ala24-Gly35. The ECD exhibits remarkable flexibility, undergoing a ~50° rotation upon ligand binding to present the peptide's N-terminus to the TMD core.
Cryo-EM structures reveal that the ECD acts as a "molecular matchmaker," orienting the peptide ligand for optimal insertion into the TMD orthosteric pocket. Mutations in the ECD, such as W39A, dramatically reduce ligand binding affinity (>100-fold), confirming its essential role.
Intracellular Signalling Cascade
Click each step to explore the molecular detail of the GLP-1R signalling pathway in pancreatic β-cells — from receptor activation to insulin exocytosis.
GLP-1R Activation & Gαs Coupling
GLP-1 binding induces the outward swing of TM6 (~14Å), breaking the TM3–TM6 ionic lock and exposing the intracellular G-protein coupling surface. The Gαs subunit inserts its α5 helix into the exposed cavity formed by ICL2, ICL3, and the cytoplasmic ends of TM3, TM5, TM6, and TM7.
GDP–GTP exchange on Gαs triggers dissociation of Gβγ. The GTP-bound Gαs adopts its active conformation and diffuses along the inner membrane leaflet to activate adenylyl cyclase (AC) isoforms (primarily AC5 and AC6 in β-cells). Simultaneously, free Gβγ dimers can activate PI3Kγ, PLC-β, and ion channels.
The receptor also couples to Gαq/11 in certain contexts, activating phospholipase C-β (PLCβ) → IP₃ + DAG → ER Ca²⁺ release + PKC activation. This dual coupling (Gαs + Gαq) is cell-type dependent and may be regulated by the lipid microenvironment.
cAMP Generation & Compartmentalisation
Gαs-GTP activates membrane-bound adenylyl cyclase (predominantly AC5/AC6 in β-cells), catalysing ATP → cAMP + PPi. GLP-1R stimulation produces a rapid 3–5 fold increase in intracellular cAMP levels within 30–60 seconds.
Crucially, cAMP signalling is spatially compartmentalised by phosphodiesterases (PDEs) — particularly PDE3B and PDE4 — and by A-kinase anchoring proteins (AKAPs) that tether PKA to specific subcellular locations. AKAP79/150 anchors PKA near L-type Ca²⁺ channels; AKAP-Lbc positions PKA near Rho-GTPases; Yotiao-AKAP links PKA to KCNQ1 channels.
This compartmentalisation means that GLP-1 produces discrete "cAMP microdomains" near the plasma membrane, at the ER surface, and around insulin granules — each activating different effector sets. Soluble adenylyl cyclase (sAC) in the cytoplasm generates a distinct cAMP pool regulated by bicarbonate and Ca²⁺, linking metabolic sensing to cAMP signalling.
PKA Pathway — Phosphorylation Cascades
cAMP binds to the regulatory subunits (RIα/RIIβ) of PKA, releasing the catalytic subunits (Cα/Cβ) which phosphorylate a cascade of targets:
• SUR1 (KATP channel): PKA phosphorylates S1448 on SUR1, increasing channel
sensitivity to ATP and reducing the threshold for glucose-induced depolarisation. This is the primary
mechanism by which GLP-1 potentiates (rather than independently triggers) GSIS.
• L-type Ca²⁺ channels (Cav1.2/1.3): PKA phosphorylation at S1928 increases channel
open probability, enhancing Ca²⁺ influx during action potentials.
• Snapin: PKA phosphorylation of Snapin at S50 promotes its interaction with SNAP-25,
facilitating SNARE complex assembly for granule exocytosis.
• RIM2α: Phosphorylation enhances Rab3A interaction, promoting granule docking at
active zones.
• CREB: PKA phosphorylates CREB at S133, activating transcription of pro-survival genes
(IRS-2, Bcl-2, Bcl-xL) — the β-cell proliferation and anti-apoptosis programme.
Epac2 Pathway — cAMP-Independent of PKA
Epac2 (also called RAPGEF4 or cAMP-GEFII) is the dominant cAMP sensor for insulin granule priming in β-cells. When cAMP binds its cyclic nucleotide-binding domain (CNBD), Epac2 undergoes autoinhibition release and activates the small GTPase Rap1.
Rap1-GTP activates PLCε → DAG generation → Munc13-1 activation. Munc13-1 is the essential priming factor that converts syntaxin-1 from the closed (Munc18-1-bound) state to the open state, enabling SNARE complex (syntaxin-1/SNAP-25/VAMP2) assembly. This is the rate-limiting step for insulin granule priming.
Epac2 also: directly interacts with SUR1 on insulin granule membranes (granular SUR1), facilitating granule–plasma membrane interactions; activates Rab3A-interacting molecule (RIM2α), promoting granule docking; and stimulates ryanodine receptor 2 (RyR2) Ca²⁺-induced Ca²⁺ release (CICR) from the ER, amplifying the Ca²⁺ signal.
The PKA vs. Epac2 balance determines the nature of the secretory response: PKA mainly potentiates first-phase (readily releasable pool) insulin secretion, while Epac2 preferentially amplifies second-phase (reserve pool) secretion by enhancing granule priming and mobilisation.
Calcium Dynamics & Electrophysiology
GLP-1 profoundly modifies the β-cell electrophysiological landscape through multiple parallel mechanisms:
1. KATP channel closure: PKA-mediated SUR1 phosphorylation lowers the ATP
threshold, synergising with glucose metabolism to depolarise the membrane from resting ~-70mV to the
action potential threshold (~-50mV).
2. Cav channel enhancement: PKA increases L-type (Cav1.2/1.3) and P/Q-type (Cav2.1)
Ca²⁺ channel activity, increasing peak Ca²⁺ current density by 30–50%.
3. TRPM4/5 activation: cAMP activates non-selective cation channels (TRPM4/5),
contributing to sustained depolarisation and burst duration extension.
4. KCNQ1 channel inhibition: PKA phosphorylation of KCNQ1 (via Yotiao-AKAP complex)
reduces delayed rectifier K⁺ current, prolonging action potential duration.
5. Na⁺ channel modulation: cAMP enhances TTX-sensitive Nav1.3/1.7 currents in human
β-cells, contributing to action potential upstroke.
The net electrophysiological effect is: increased burst frequency, prolonged burst duration, higher action potential amplitude, and enhanced Ca²⁺ oscillation amplitude — all of which increase the time-averaged [Ca²⁺]i that drives exocytosis. Critically, all these effects require a background glucose concentration ≥5.6mM, maintaining the glucose-dependency and safety of GLP-1 action.
SNARE-Mediated Insulin Granule Exocytosis
Insulin granule exocytosis is the final effector step, requiring the assembly of the SNARE complex: syntaxin-1A (t-SNARE), SNAP-25 (t-SNARE), and VAMP2/synaptobrevin (v-SNARE). GLP-1 signalling amplifies multiple steps:
Docking: Rab3A (activated by RIM2α) and Rab27A (via granuphilin/Slp4) tether granules
to the plasma membrane at active zones enriched in Cav channels.
Priming: Munc13-1 (activated by DAG/Epac2→Rap1→PLCε) converts syntaxin-1 to the open
conformation. Munc18-1 then chaperones SNARE complex assembly. CAPS (Ca²⁺-dependent activator protein
for secretion) further enhances priming of dense-core vesicles.
Fusion: Synaptotagmin-7 (Syt7) is the primary Ca²⁺ sensor for insulin granule
exocytosis (Kd ~15μM Ca²⁺). Ca²⁺ binding to C2A/C2B domains triggers membrane penetration, driving
membrane merger. Complexin acts as a fusion clamp released by Ca²⁺-Syt7.
Granule pool mobilisation: GLP-1/cAMP promotes cortical actin remodelling (via Cdc42 →
N-WASP → Arp2/3), enabling reserve pool granules to access the plasma membrane. This "newcomer" pathway
is particularly important for sustained second-phase secretion.
β-Arrestin Signalling & Receptor Trafficking
Following sustained agonist stimulation, GRK2/5/6 phosphorylate the GLP-1R C-terminal tail and ICL3 at multiple Ser/Thr residues, creating a "phospho-barcode" that recruits β-arrestin-1/2.
Desensitisation: β-arrestin sterically occludes the G-protein binding site, terminating
Gαs signalling at the plasma membrane (homologous desensitisation). This occurs within 5–15 minutes of
agonist exposure.
Internalisation: β-arrestin acts as an adaptor for clathrin and AP2, driving receptor
endocytosis into clathrin-coated pits → early endosomes. The GLP-1R undergoes rapid internalisation (t½
~10 min).
Endosomal signalling: Critically, the internalised GLP-1R continues to signal from
endosomes. The receptor–β-arrestin–Gαs complex generates a sustained
"second wave" of cAMP from endosomal membranes. This endosomal cAMP pool activates a distinct subset of
PKA/Epac2 targets in the perinuclear region, particularly CREB-mediated transcription.
Receptor fate: GLP-1R can be sorted to lysosomes (degradation, reducing surface receptor density) or recycled via Rab4/Rab11 endosomes back to the plasma membrane (resensitisation). The balance depends on ubiquitination status (NEDD4 E3 ligase) and interaction with sorting nexins (SNX27). Prolonged agonist exposure leads to net receptor downregulation — clinically relevant for dose escalation strategies with GLP-1RAs.
Transcriptional & Trophic Effects (Long-Term)
Beyond acute secretory amplification, GLP-1R activation initiates profound long-term transcriptional programmes:
CREB/CRTC2 axis: PKA-phosphorylated CREB recruits co-activator CRTC2 (dephosphorylated
by calcineurin in response to Ca²⁺), driving transcription of IRS-2
(insulin receptor substrate 2), Bcl-2, Bcl-xL, and PDX-1. IRS-2 upregulation
activates the PI3K → Akt → FOXO1 survival pathway.
Wnt/β-catenin: GLP-1R signalling activates TCF7L2 (the strongest T2DM GWAS gene) via
β-catenin stabilisation, promoting β-cell proliferation and incretin gene expression.
mTORC1: cAMP/PKA → TSC2 phosphorylation → mTORC1 activation → S6K1/4E-BP1 → protein
synthesis and cell growth. This mediates GLP-1-induced β-cell hypertrophy.
Anti-apoptotic: GLP-1 suppresses ER stress (reduced CHOP, ATF4, spliced XBP1) and
mitochondrial apoptotic pathways (reduced Bax, cytochrome c release, caspase-3). It upregulates the
unfolded protein response chaperone BiP/GRP78.
Neogenesis/Transdifferentiation: In rodent models, GLP-1R agonism promotes
ductal-to-β-cell neogenesis and α-to-β-cell transdifferentiation via PDX-1 and MafA induction. Evidence
in human islets is emerging but less definitive.
Glucose-Dependency & the Amplification Concept
Understanding why GLP-1's insulinotropic effect is inherently glucose-dependent — the key safety feature.
The most clinically important aspect of GLP-1 physiology is its strict glucose-dependency. Unlike sulfonylureas (which directly close KATP channels regardless of glucose), GLP-1 amplifies an already-initiated glucose signal. At glucose concentrations below ~4.5 mmol/L, GLP-1 has minimal insulinotropic effect because the underlying glucose metabolism–KATP closure–depolarisation–Ca²⁺ entry cascade is not sufficiently activated for cAMP to amplify.
This glucose-dependency arises from the fundamental architecture of the β-cell stimulus-secretion coupling pathway, which operates in two conceptual arms:
Triggering Pathway
Glucose → GLUT1/2 → Glucokinase → ATP/ADP ratio ↑ → KATP closure → Depolarisation → VGCC opening → Ca²⁺ influx → [Ca²⁺]i rise
This is the necessary triggering signal. Without sufficient glucose metabolism, the KATP channels remain open, the membrane stays hyperpolarised, and VGCCs cannot activate. GLP-1/cAMP cannot override this — it modulates the threshold and magnitude, but cannot independently trigger the cascade.
Amplifying Pathway
Glucose metabolism → ↑ ATP, ↑ glutamate, ↑ malonyl-CoA, ↑ NADPH → Augmentation of exocytosis at a given [Ca²⁺]i
GLP-1/cAMP acts primarily within this amplifying pathway. It increases Ca²⁺ channel activity (more Ca²⁺ per action potential), enhances granule priming and docking, promotes SNARE complex assembly, and increases the exocytotic efficiency (more insulin released per unit of Ca²⁺). All these require the triggering pathway to already be active.
This glucose-dependent amplification mechanism is why GLP-1 receptor agonists carry a very low risk of hypoglycaemia as monotherapy. As plasma glucose falls toward normoglycaemic levels, the triggering pathway becomes less active, and GLP-1's amplifying effect diminishes proportionally. This creates an elegant physiological "safety valve" — the incretin effect self-limits as glucose normalises. This is fundamentally different from sulfonylureas, which close KATP channels independently of glucose levels.
Multi-Organ Physiological Effects
GLP-1 receptors are expressed in numerous tissues beyond the endocrine pancreas. The pleiotropic actions underpin the cardiovascular and renal benefits observed with GLP-1 receptor agonists.
Pancreatic Islets
GLP-1R is expressed on β-cells, δ-cells, and potentially α-cells (debated — may be indirect via somatostatin paracrine signalling).
- β-cells: Potentiates GSIS via cAMP/PKA/Epac2 cascade (detailed above). Promotes insulin gene transcription (PDX-1, MafA). Enhances proinsulin biosynthesis. Induces β-cell proliferation and inhibits apoptosis.
- α-cells: Suppresses glucagon secretion in a glucose-dependent manner. At hyperglycaemic levels, GLP-1 inhibits glucagon; at hypoglycaemic levels, glucagon counter-regulation is preserved. Mechanism: likely indirect via somatostatin from δ-cells, but direct α-cell GLP-1R expression remains debated.
- δ-cells: Stimulates somatostatin secretion, which paracrinally inhibits α-cell glucagon release (the δ-cell intermediary hypothesis).
Brain & Hypothalamus
GLP-1R is expressed in the hypothalamus (arcuate, paraventricular, dorsomedial nuclei), brainstem (NTS, area postrema), hippocampus, and cortex.
- Appetite regulation: GLP-1 activates POMC/CART anorexigenic neurons and inhibits NPY/AgRP orexigenic neurons in the arcuate nucleus. Central cAMP/PKA signalling reduces food intake and promotes satiety. Area postrema activation mediates nausea (the dose-limiting side effect).
- Reward circuitry: GLP-1R in the mesolimbic dopamine system (VTA, nucleus accumbens) reduces food reward, hedonic eating, and potentially alcohol/substance cravings.
- Neuroprotection: Reduces neuroinflammation (microglial M2 polarisation), enhances BDNF signalling, reduces amyloid-β aggregation, improves mitochondrial function, reduces oxidative stress. Basis for trials in Alzheimer's and Parkinson's disease.
- Vagal afferent pathway: Peripheral GLP-1 activates vagal afferent neurons (expressing GLP-1R) in the nodose ganglion → NTS → higher brain centres. This "gut-brain axis" route may mediate rapid satiety signalling without requiring BBB penetration.
Heart & Vasculature
Cardiac GLP-1R expression is primarily in the atria and sinoatrial node. Vascular effects may be partly GLP-1R-independent.
- Cardioprotection: Reduces myocardial ischaemia-reperfusion injury via cAMP/PKA → increased glucose uptake (GLUT1/4), improved mitochondrial function, reduced apoptosis (Akt/GSK3β pathway), and reduced ROS generation.
- Anti-atherosclerotic: Reduces monocyte adhesion (VCAM-1, ICAM-1 downregulation), inhibits foam cell formation, reduces plaque inflammation (NF-κB suppression), promotes plaque stability. GLP-1R-independent effects via GLP-1(9-36) metabolite may contribute.
- Endothelial function: Enhances NO bioavailability (eNOS activation via AMPK and Akt), reduces endothelial ROS, improves flow-mediated dilation.
- Heart rate: GLP-1R activation in sinoatrial node increases heart rate by 2–4 bpm (cAMP/HCN4 channel modulation) — observed in CVOT trials.
- Blood pressure: Natriuretic action (renal) + vascular relaxation → modest systolic BP reduction (2–6 mmHg).
Stomach & Intestine
GLP-1 profoundly influences GI motility through both central (vagal) and local ENS mechanisms.
- Gastric emptying delay: The "ileal brake" effect — slows gastric emptying by 20–40% via vagal afferent → brainstem → vagal efferent circuits. This flattens postprandial glucose excursions and is a major mechanism for PPG reduction with GLP-1RAs.
- Tachyphylaxis: The gastric emptying delay shows significant tachyphylaxis with continuous GLP-1R activation (relevant for long-acting vs. short-acting GLP-1RAs; exenatide BID retains more gastric emptying effect than liraglutide/semaglutide).
- Gastric acid suppression: Reduces pentagastrin-stimulated acid output via central vagal mechanism.
- Intestinal motility: Reduces small intestinal motility and transit, potentially contributing to constipation with some GLP-1RAs.
Kidney
GLP-1R is expressed in the juxtaglomerular apparatus, renal vasculature, and proximal tubular epithelium.
- Natriuresis: GLP-1 directly inhibits NHE3 (Na⁺/H⁺ exchanger 3) in the proximal tubule via PKA-dependent phosphorylation at S552, promoting sodium excretion. This is independent of glucose and insulin.
- Diuresis: Secondary to natriuresis; contributes to the modest blood pressure-lowering effect.
- Renal haemodynamics: Increases renal blood flow and GFR acutely (ANP-like effect). May reduce intraglomerular pressure by afferent arteriolar dilation.
- Anti-inflammatory/anti-fibrotic: Reduces NLRP3 inflammasome activation, NF-κB signalling, TGF-β/Smad3 fibrotic pathway, and oxidative stress in mesangial cells and podocytes. Basis for kidney outcome benefits in FLOW trial with semaglutide.
Liver
Hepatic GLP-1R expression is debated. Many hepatic effects are likely indirect (via insulin/glucagon ratio, weight loss, and improved metabolic milieu).
- Hepatic glucose output: Reduced via glucagon suppression and enhanced insulin signalling → decreased glycogenolysis and gluconeogenesis.
- Lipid metabolism: Reduces de novo lipogenesis (via AMPK activation, SREBP-1c downregulation), enhances fatty acid β-oxidation (CPT1a upregulation), reduces hepatic VLDL-TG secretion.
- MASLD/MASH: GLP-1RAs reduce hepatic steatosis (MRI-PDFF reduction ~30%), lobular inflammation, and potentially fibrosis. Mechanisms include reduced ER stress, improved autophagy (via TFEB nuclear translocation), and reduced NF-κB/JNK inflammatory signalling.
- Direct vs. indirect: Semaglutide's effects on MASH are substantially driven by weight loss, but some evidence supports weight-independent hepatic effects via circulating GLP-1(9-36) metabolite acting on hepatocyte mitochondria.
White & Brown Adipose
GLP-1R expression in adipose tissue is low but present. Many effects are indirect via CNS and metabolic improvements.
- WAT: Reduces inflammation (macrophage M1→M2 polarisation), improves adipokine profile (↑ adiponectin, ↓ leptin, ↓ resistin), may enhance insulin-stimulated glucose uptake.
- BAT/Beige fat: Central GLP-1R activation (hypothalamic AMPK inhibition) → increased sympathetic outflow → BAT thermogenesis (UCP1 upregulation). This may contribute to energy expenditure increases with GLP-1RAs.
- Lipolysis: Complex effects — acute cAMP/PKA may stimulate lipolysis, but overall reduced free fatty acid levels with GLP-1RA treatment (improved insulin sensitivity reducing inappropriate lipolysis).
- Body composition: GLP-1RAs cause preferential visceral adipose reduction, with ~60–75% of weight loss as fat mass and ~25–40% as lean mass.
Bone
GLP-1R is expressed on osteoblasts and bone marrow stromal cells. The incretin axis participates in the gut-bone axis linking feeding to bone remodelling.
- Osteoblasts: GLP-1 stimulates osteoblast differentiation and activity via cAMP/PKA/CREB → increased Runx2, osteocalcin expression. Enhances alkaline phosphatase activity and mineralisation.
- Osteoclasts: Indirect inhibition via calcitonin (GLP-1R on thyroid C-cells stimulates calcitonin release in rodents; relevance in humans is minimal due to low C-cell GLP-1R expression).
- Fracture risk: Meta-analyses suggest neutral to protective effect of GLP-1RAs on fracture risk, in contrast to the increased risk with some other diabetes medications.
GLP-1 Degradation & the DPP-4 Axis
The extraordinarily short half-life of native GLP-1 and its metabolic fate.
L-Cell Secretion
GLP-1(7-36)NH₂ released basolaterally into lamina propria
DPP-4 Cleavage
Membrane-bound DPP-4 on endothelial cells cleaves His7-Ala8 bond within seconds
GLP-1(9-36)NH₂
Inactive metabolite — but has independent cardioprotective/vasodilatory properties
Renal Clearance
Both intact GLP-1 and GLP-1(9-36) cleared by kidney (NEP 24.11 further degrades)
DPP-4 is so efficient that only ~10–15% of secreted GLP-1 reaches the systemic circulation intact. Most degradation occurs even before GLP-1 enters the portal circulation — DPP-4 is abundantly expressed on intestinal capillary endothelial cells. This means that a large proportion of GLP-1's acute effects may be mediated locally via vagal afferent activation in the gut wall (the paracrine/neurocrine hypothesis) rather than by circulating hormone reaching the pancreas via the endocrine route. Hepatic DPP-4 further degrades portal GLP-1, and only a small fraction of the intact peptide reaches the systemic circulation.
DPP-4 (CD26, adenosine deaminase complexing protein 2) is a type II transmembrane serine protease of the prolyl oligopeptidase family. It exists as a homodimer, with each 766-residue monomer containing an N-terminal cytoplasmic tail, a transmembrane helix, a glycosylated stalk, an 8-bladed β-propeller domain, and a C-terminal α/β-hydrolase catalytic domain containing the Ser630-Asp708-His740 catalytic triad.
Substrate specificity: DPP-4 cleaves dipeptides from the N-terminus of substrates with a penultimate proline or alanine (Xaa-Pro↓ or Xaa-Ala↓). GLP-1 (His-Ala↓-Glu...) is an ideal substrate. Other DPP-4 substrates include: GIP, PYY(3-36), NPY, SDF-1α (CXCL12), BNP, substance P, and multiple chemokines — explaining the pleiotropic effects of DPP-4 inhibitors beyond glycaemic control.
Soluble DPP-4 (sDPP-4): A shed form circulates in plasma (released from adipocytes, hepatocytes, and immune cells) and retains enzymatic activity. sDPP-4 levels are elevated in obesity and T2DM and may serve as an adipokine linking visceral adiposity to incretin degradation.
NEP 24.11 (neprilysin, CD10) is a zinc metallopeptidase that cleaves GLP-1 at multiple internal sites (particularly within the mid-region), generating small inactive fragments. NEP is abundantly expressed in the renal brush border, contributing to renal degradation. NEP also degrades natriuretic peptides (ANP, BNP) — the dual NEP/angiotensin receptor inhibitor sacubitril/valsartan may incidentally increase circulating GLP-1 levels.
GLP-1(9-36)NH₂, long dismissed as inactive, has emerging evidence for independent biological activities:
Cardioprotection: Improves post-ischaemic myocardial function, reduces infarct size (in animal models), likely via GLP-1R-independent mechanisms — possibly through direct mitochondrial effects or an as-yet-unidentified receptor. Activates cGMP/PKG pathway in cardiomyocytes.
Vasodilation: Promotes NO-dependent vasodilation via mechanisms independent of the classical GLP-1R. May involve GLP-1(28-36) or GLP-1(32-36) sub-fragments acting on mitochondrial complex I.
Hepatic effects: Reduces hepatic glucose production in hepatocytes, possibly through direct mitochondrial entry and modulation of oxidative phosphorylation. The pentapeptide GLP-1(32-36) has been shown to enter cells and target mitochondria.
Clinical implication: These findings suggest that DPP-4 cleavage of GLP-1 is not merely degradation but rather a processing step that generates a distinct signalling molecule with complementary biological activities — a paradigm shift in incretin biology.
From Physiology to Therapeutics
How understanding GLP-1 molecular physiology directly informed the development of GLP-1-based therapies.
The ultra-short half-life of native GLP-1 (1.5–2 minutes) necessitated molecular engineering strategies to create therapeutically viable GLP-1 receptor agonists. Four primary approaches have been employed:
| Strategy | Mechanism | Example | Key Modification |
|---|---|---|---|
| DPP-4 resistant peptide | Exendin-4 (from Gila monster venom) has Gly at position 2 instead of Ala, conferring natural DPP-4 resistance | Exenatide | 53% homology to GLP-1; Gly8; C-terminal extension (CE-9) |
| Fatty acid acylation | C16/C18 fatty acid attached via glutamic acid spacer → reversible albumin binding → depot effect + steric shielding from DPP-4 | Liraglutide, Semaglutide | Lira: C16 palmitoyl at Lys26. Sema: C18 di-acid at Lys26 + Aib at position 2 |
| Fc fusion | GLP-1 analogue fused to IgG4 Fc domain → FcRn-mediated recycling extends half-life to ~5 days | Dulaglutide | Two GLP-1 analogue chains fused to modified IgG4 Fc; Gly8, Glu34 |
| Amino acid substitution | α-aminoisobutyric acid (Aib) at position 2 blocks DPP-4 access; Arg34→Lys34 with fatty acid enables albumin binding | Semaglutide | Aib2 (DPP-4 resistance) + C18 fatty diacid (albumin binding) = t½ ~7 days |
Biased agonism (functional selectivity) is the concept that different ligands at the same receptor can preferentially activate distinct downstream signalling pathways. For the GLP-1R, the key dichotomy is G-protein (cAMP) signalling vs. β-arrestin recruitment.
Different GLP-1RAs show distinct bias profiles: exendin-4 is relatively β-arrestin-biased compared to GLP-1; oxyntomodulin shows G-protein bias; small molecule agonists (e.g., TT-OAD2, danuglipron) show markedly different bias profiles. The clinical significance is emerging: G-protein-biased agonists may produce stronger insulin secretion with less receptor desensitisation/internalisation (potentially maintaining efficacy over time), while β-arrestin signalling mediates endosomal cAMP/CREB activation important for β-cell survival.
The molecular basis of bias lies in the distinct receptor conformations stabilised by different ligands: the orientation of TM6 and TM7, ICL3 phosphorylation pattern, and β-arrestin finger-loop insertion angle all differ between G-protein-biased and β-arrestin-biased ligands. Cryo-EM structures of GLP-1R in complex with different agonists have confirmed these distinct active-state conformations.
The observation that bariatric surgery (RYGB/sleeve) produces metabolic benefits exceeding any single hormone replacement led to the concept of multi-incretin agonism — simultaneously targeting GLP-1R with GIP-R, glucagon-R, or both.
Tirzepatide (GLP-1R/GIPR dual agonist) demonstrates that GIP co-agonism amplifies weight loss and glycaemic control beyond GLP-1R mono-agonism. The mechanism involves: GIP receptor activation on adipocytes enhancing lipid buffering and insulin sensitivity; complementary CNS appetite suppression via distinct hypothalamic circuits; enhanced incretin-mediated insulin secretion through additive cAMP generation; and reduced GLP-1R tachyphylaxis potentially mediated by GIP-driven receptor recycling mechanisms.
Triple agonists (GLP-1R/GIPR/GCGR, e.g., retatrutide) add glucagon receptor agonism to increase energy expenditure (via hepatic FGF21 induction, BAT thermogenesis), enhance amino acid catabolism, and amplify lipolysis — producing the most potent weight loss yet observed in clinical trials (~24% at 48 weeks).
The structural basis for multi-agonism exploits the high sequence homology between GLP-1, GIP, and glucagon (all derived from proglucagon). Strategic amino acid substitutions at positions that determine receptor selectivity (particularly positions 1, 2, 3, 16, 17, 20, 21, 24, 27, 28) allow engineering of molecules with tuneable activity ratios at each receptor.
Oral delivery of peptide GLP-1RAs overcomes the inherent challenge of peptide bioavailability (destruction by gastric acid and proteases, poor intestinal permeability of ~36-amino acid peptides). Two main strategies have been developed:
Oral semaglutide (Rybelsus): Co-formulated with sodium N-[8-(2-hydroxybenzoyl) amino] caprylate (SNAC), an absorption enhancer. SNAC creates a local pH-buffering microenvironment, protects semaglutide from pepsin degradation, and transiently enhances transcellular absorption across gastric epithelium via a lipophilic flux pathway. Bioavailability is ~1% (necessitating high 3/7/14mg oral doses to achieve therapeutic semaglutide levels comparable to 0.25–1mg SC doses). Must be taken fasting with ≤120mL water, 30 min before food/other medications.
Oral small molecules: Non-peptide GLP-1R agonists (danuglipron, orforglipron) bind to an intracellular allosteric site or within the TMD pocket, bypassing the need for peptide stability. These have ~60–80% oral bioavailability, no food-timing restrictions, and conventional oral pharmacokinetics. Their binding site within the TMD (between TM6/TM7 and ECL3) is distinct from the orthosteric peptide site, potentially conferring different signalling bias and efficacy profiles.
GLP-1 Axis Dysfunction in Disease
Type 2 Diabetes
The incretin effect is severely attenuated in T2DM (from ~60% to ~20% of postprandial insulin secretion). Fasting GLP-1 levels are relatively preserved, but postprandial GLP-1 secretion may be modestly reduced (~10–20% lower in some studies; debated). More importantly, the insulinotropic potency of GLP-1 is reduced — β-cells require supra-physiological GLP-1 concentrations to achieve normal insulin responses, reflecting β-cell dysfunction (reduced GLP-1R expression, impaired cAMP/Ca²⁺ coupling, reduced SNARE machinery). Pharmacological GLP-1RA doses overcome this "resistance" by flooding the system.
Obesity
In obesity, elevated circulating DPP-4 (sDPP-4 released from expanded visceral adipose) accelerates GLP-1 degradation. Chronic hyperinsulinaemia may downregulate L-cell secretory capacity. Hypothalamic GLP-1R signalling is impaired by central inflammation (microglial activation, IKKβ/NF-κB) and leptin resistance, reducing the anorexigenic response to endogenous GLP-1. Post-bariatric surgery, dramatically enhanced GLP-1 secretion (10–20× increase) is a key mediator of weight loss and T2DM remission — driven by altered nutrient delivery kinetics and bile acid signalling.