New Results
David J.Edmonds, Jean-Phillipe Fortin, Amit S.Kalgutkar, J. Brent Kuzmiski, Paula M.Loria, Aditi R.Saxena, Scott W.Bagley, Clare Buckeridge, John M.Curto, David R.Derksen, João M.Dias, Matthew C.Griffor, Seungil Han, V. Margaret Jackson, Margaret S.Landis, Daniel J.Lettiere, Chris Limberakis, Yuhang Liu, Alan M.Mathiowetz, David W.Piotrowski, David A.Price, Roger B.Ruggeri, David A.Tess
Peptide agonists of the glucagon-like peptide-1 receptor (GLP-1R) have revolutionized diabetes therapy, but their use has been limited by the requirement for injection. Here we describe the first effective, orally bioavailable small molecule GLP-1R agonists. A sensitized high-throughput screen identified a series of small molecule GLP-1R agonists that were optimized to promote endogenous GLP-1R signaling with nM potency. These small molecule agonists increased insulin levels in primates but not rodents, which is explained by a cryo-EM structure that revealed a binding pocket requiring primate-specific tryptophan 33. Importantly, oral administration of agonist PF-06882961 to healthy humans produced dose-dependent declines in serum glucose. This opens the door to a new era of oral small molecule therapies that target the well-validated GLP-1R pathway for metabolic health.
One Sentence Summary PF-06882961 is an orally administered small molecule that activates the GLP-1 receptor to lower blood glucose in humans.
Glucagon-like peptide-1 (GLP-1), a neuroendocrine hormone, is derived from a proglucagon precursor and secreted by intestinal enteroendocrine L cells in response to nutrient intake, predominantly in the form of GLP-1(7-36) amide (henceforth GLP-1). Activation of the GLP-1 receptor (GLP-1R) by GLP-1 stimulates insulin release and inhibits glucagon secretion in a glucose-dependent manner. Also, GLP-1 delays gastric emptying, increases satiety, suppresses food intake, and reduces weight in humans. Multiple injectable peptidic GLP-1R agonists are approved for the treatment of Type 2 diabetes mellitus (T2DM), including liraglutide which is also approved for the treatment of obesity. Excitement has grown in this drug class, with several GLP-1R agonists demonstrating benefit in cardiovascular outcomes studies. However, a drawback of these medicines has been the necessity for administration by subcutaneous injection, which limits patient utilization and may reduce opportunities for fixed-dose combination treatments with other small-molecule drugs. Importantly, patients prefer, and are more likely to adhere to, an oral drug regimen versus an injectable alternative. An orally administered formulation of the peptidic GLP-1R agonist semaglutide was recently approved for the treatment of T2DM. This peptidic drug is co-formulated with sodium N-[8-(2-hydroxybenzoyl] amino) caprylate (SNAC), a purported gastric absorption enhancer, to promote oral bioavailability. The dosage must be taken once daily in the fasted state with minimal liquid and at a substantially higher dose than the approved once-weekly injectable formulation. Thus, we sought to identify a small-molecule GLP-1R agonist that is orally bioavailable using standard formulations, and has the potential to be combined with other oral small molecule therapeutics.
The GLP-1R is a seven-transmembrane-spanning, class B, G protein-coupled receptor (GPCR). Class B GPCRs, including GLP-1R, are activated by endogenous peptide hormones, and the development of small-molecule agonists of these receptors has proven particularly challenging. Significant prior efforts across the pharmaceutical industry have failed to identify potent and efficacious small-molecule agonists of the GLP-1R. Given the significant therapeutic value of this mechanism, we pursued a novel high-throughput screening strategy that identified a series of small-molecule GLP-1R agonist leads. Optimization of the lead series resulted in potent agonists that activate the GLP-1R in an unprecedented manner. The series includes the clinical development candidate PF-06882961, which we show has robust preclinical efficacy, oral bioavailability, and evidence of glucose-lowering in healthy human participants.
Binding of GLP-1 to its receptor activates the guanine nucleotide-binding (G) alpha stimulatory subunit (Gαs) of the heterotrimeric G protein complex, stimulating adenylate cyclase activity, and thereby increasing intracellular concentrations of cyclic adenosine monophosphate (cAMP). Activation of the GLP-1R by GLP-1 also results in recruitment of β-arrestin-1 (βArr1) and β-arrestin-2 (βArr2), which mediate receptor internalization and may serve as scaffolds for G protein-independent signaling. The human GLP-1R has an unusually high activation barrier and is essentially silent in the absence of an agonist ligand, relative to other class B GPCRs (e.g. the gastric inhibitory polypeptide and parathyroid hormone 1 receptors). We hypothesized that a positive allosteric modulator (PAM) could be used as a tool to lower this activation barrier, thereby increasing assay sensitivity and facilitating the detection of weak agonists in a cell-based functional assay.
The PAM 4-(3-(benzyloxy)phenyl)-2-ethylsulfinyl-6-(trifluoromethyl)pyrimidine (BETP, Fig. 1A) had been reported to potentiate GLP-1R-mediated cAMP signaling in response to weak peptidic agonists, including the GLP-1 metabolite GLP-1(9–36)amide and the GLP-1R/glucagon-receptor dual agonist oxyntomodulin. Further work demonstrated that BETP positively modulates GLP-1R function through covalent modification of cysteine 347 located on the third intracellular loop. We postulated that a BETP-sensitized assay would prove effective for identifying weak agonists and were pleased to observe potentiation of GLP-1R-mediated cAMP signaling in response to the weak non-peptide GLP-1R agonist BOC5 (Fig. S1A) in Chinese hamster ovary (CHO) cells stably expressing the human GLP-1R (Fig. 1B). In particular, the positive impact on maximal effect (E max) in this assay format significantly improves the chances of identifying weak agonists during high-throughput screening (HTS). Further confidence in the assay design came from testing the effects of BETP on receptor-mediated βArr signaling. Peptide agonist 1 (Fig. S1B), identified during our earlier efforts to design orally available peptidic GLP-1R agonists, is a full (E max) cAMP agonist (Fig. 1C), but a partial βArr agonist (Fig. 1D) at the GLP-1R. BETP improved the potencies of peptide 1 for both pathways and potentiated the E max for βArr recruitment (Fig. 1C and 1D, Table S1). The observed potentiation of both the cAMP and βArr signaling pathways was consistent with our hypothesis that BETP treatment resulted in general receptor sensitization, rather than a pathway-specific signaling amplification.
Fig. 1.Identification of small-molecule GLP-1R agonists in a CHO-GLP-1R cellular assay in the absence or presence of the positive allosteric modulator BETP.
(A) Assay concept: Covalent modification of Cys347 in the GLP-1R by BETP lowers receptor activation barrier, enabling the identification of weak agonists. (B–D) Validation of the BETP-sensitized screening assay. (B) BETP potentiates agonist-induced cAMP production of small molecule (BOC5) (Fig. S1). (C–D) BETP potentiates cAMP production (C) and β-arrestin recruitment (D) by peptide 1 (Fig. S1) at the human GLP-1R. (E), cAMP curves of a representative small molecule HTS hit, compound 2 (Fig. 2). Data represent the mean ± SEM. BETP, 4-(3-(benzyloxy)phenyl)-2-ethylsulfinyl-6-(trifluoromethyl)pyrimidine; cAMP, cyclic adenosine monophosphate; DMSO, dimethyl sulfoxide; GLP-1R, glucagon-like peptide-1 receptor; HTS, high-throughput screening; PAM, positive allosteric modulator; SEM, standard error of the mean.
Our BETP-sensitized cAMP screening assay (SA +BETP) was adapted to a single-point format and employed in an HTS of 2.8 million compounds from the Pfizer compound collection. A hit was defined by a threshold of >30% effect (i.e., >30% of the E max of GLP-1) at 10 µM and the screen resulted in a low confirmed hit-rate of 0.013%. A series of pyrimidine derivatives exemplified by 2 emerged from these hits (Fig. 2A). Compound 2 was inactive as a GLP-1R agonist in the unpotentiated cAMP screening assay (SA), which did not include BETP, but demonstrated a ∼70% effect at 20 µM in the presence of BETP (Fig. 1E). This non-traditional screening approach carried the risk that the GLP-1R agonist lead series might remain dependent on the presence of BETP to activate the receptor, and it was unclear at the outset whether we would observe GLP-1R agonism in the absence of BETP. However, as analogs were identified with improved cAMP potency in the BETP-sensitized assay, we also observed a gradual increase in signaling efficacy in the absence of BETP (Fig. 2B) to the point where a dose-response curve and half-maximal effective concentration
