Paper describes how orforglipron activates the GLP-1 receptor

In November 2020, researchers published a detailed account of how a small-molecule drug switches on the human GLP-1 receptor, the same target hit by injectable medicines such as Ozempic and Wegovy ^[1]. The compound, LY3502970, is now known as orforglipron. The paper appeared in the Proceedings of the National Academy of Sciences ^[1].

LY3502970 was described as an orally active nonpeptide agonist of the GLP-1 receptor, meaning it is not a peptide, the drug class that includes semaglutide and tirzepatide ^[1]. Using cryo-electron microscopy, the team captured a high-resolution structure of the compound bound to the active receptor. It sits in a pocket in the upper helical bundle, held in place by the receptor's extracellular domain, extracellular loop 2, and transmembrane helices 1, 2, 3 and 7 ^[1]. That is a different binding pocket and a distinct receptor conformation from the one peptide hormones use ^[1].

LY3502970 is a partial agonist, and it is biased toward activating the G protein pathway rather than recruiting β-arrestin, a signaling protein. The paper suggests the distinct receptor conformation may explain both the partial agonism and the biased signaling ^[1].

In mice engineered to carry the human GLP-1 receptor, oral doses from 0.1 to 10 milligrams per kilogram of body weight lowered blood glucose after a glucose challenge. Measured blood levels of the drug were 1,257 ± 387 nmol/L at 10 mg/kg, 205 ± 18 nmol/L at 1 mg/kg and 24 ± 8 nmol/L at 0.1 mg/kg ^[1]. Injected exenatide (Byetta, Bydureon), used as a positive control, lowered glucose to a similar extent ^[1]. In nonhuman primates, oral LY3502970 produced insulin-releasing and appetite-suppressing effects, again with an effect size comparable to injectable exenatide ^[1]. The authors note that a partial agonist can still produce a full biological response in vivo when enough receptors are occupied ^[1].

The compound showed no activity against other class B G protein-coupled receptors, and it did not activate the mouse version of the GLP-1 receptor. The difference traces to a single amino acid, Trp33, in the receptor's extracellular domain that is specific to primates, which is why the mouse studies used animals carrying the human receptor ^[1].

At the time, approved GLP-1 receptor agonists were peptide drugs given by subcutaneous injection, with one exception: oral semaglutide (Rybelsus). Its bioavailability is less than 1 percent, and the tablet must be taken after an overnight fast with a set amount of water and at least 30 minutes before food or other medicines. The paper noted that approved Rybelsus doses do not achieve the higher drug exposures needed to match injectable Ozempic ^[1].

Why it matters for patients

The appeal of a nonpeptide GLP-1 drug is practical. A small molecule can be formulated as a standard tablet, and the paper states that nonpeptide agonists could offer more standard formulations and simpler dosing than oral semaglutide's fasting and timing rules ^[1]. For people who take several daily medicines, that could matter.

Everything in this paper is laboratory and animal data. It does not show that orforglipron lowers blood sugar or body weight in people, and it does not report human side effects, human doses, or how the drug would compare with approved medicines in patients ^[1]. Orforglipron is also a different molecule from semaglutide, and findings about oral semaglutide do not carry over to it.

What happens next

The paper does not disclose when human trials of LY3502970 might begin, what doses would be tested in people, or whether the drug would need to be taken with food restrictions. Those details had not been reported at the time of publication ^[1].