GLP-1-based therapy for obesity

Hooking a new hormone

In the mid-1970s, Habener, a newly minted endocrinologist, set up his lab at Massachusetts General Hospital, and diabetes soon caught his attention. Normally, glucose goads the pancreas to release insulin, which ushers the sugar out of the bloodstream and into cells. In diabetes, a dearth of insulin keeps blood glucose levels high while cells starve. Although supplying insulin underpinned one form of therapy, researchers were exploring alternative tactics. Another pancreatic hormone, glucagon, boosts blood sugar concentrations, so thwarting it might benefit people with diabetes, the thinking went.

Habener decided to utilize the new tools of molecular biology and isolate the gene that encodes glucagon. NIH guidelines on recombinant DNA research at the time restricted manipulation of mammalian genes, so he pivoted to the anglerfish, which offered an advantage, as it contains a special organ that manufactures generous amounts of glucagon.

Scientists knew that active peptide hormones are liberated from larger proteins by enzymes that snip in specific places. In 1982, Habener reported that the fish glucagon gene encodes a predicted precursor protein that contains glucagon and, in addition, a second peptide that resembles glucagon. The same pair of amino acids, lysine-arginine, that mark cleavage sites in other hormone precursor proteins appears at several spots. Cutting there would free glucagon and the second peptide.

The following year, Graeme Bell (Chiron Corporation) found that the hamster glucagon-encoding gene also encodes a version of the second fish peptide, which he called glucagon-like peptide-1 (GLP-1). Similar results from human and other mammals followed.

An overlooked processing step?

By this time, chemist Svetlana Mojsov at The Rockefeller University had independently become a glucagon aficionado. She wanted to devise a way to produce large quantities of it for mechanistic studies. As part of this effort, she had pored over the hormone’s structure as a graduate student, postdoctoral fellow, and researcher. Mojsov’s thesis advisor, Bruce Merrifield (1969 Albert Lasker Basic Medical Research Award; 1984 Nobel Prize in Chemistry), had invented so-called solid phase protein synthesis, and by the mid-1970s, it had become the technique of choice for rapidly producing ample supplies of clean material.

Unfortunately, due to glucagon’s chemical peculiarities, certain amino acids undergo side reactions with the strong acid that the process requires. For this reason, conventional wisdom held that glucagon did not lend itself to the solid phase method.

This hurdle did not deter Mojsov. She conceived an alternative strategy that avoided strong acids. Furthermore, her approach produced a degree of purity never before achieved for peptide synthesis. The high yields and absence of contaminants would prove crucial for numerous aspects of her future work.

By 1983, when Mojsov moved to Massachusetts General Hospital as director of its peptide synthesis facility, she had improved the scheme and had begun to apply her expertise to GLP-1. The peptide intrigued her in part because she thought that it might fill a longstanding gap. In the early 1900s, scientists proposed that substances in the gut spur the pancreas to churn out hormones. Solid evidence for such “incretins” emerged in 1964, when researchers demonstrated that ingested glucose elicits more insulin release than injected glucose does. Something in the intestine, they concluded, provokes insulin secretion. Such incretins had thus far eluded identification and GLP-1, a previously unknown peptide that resembles a hormone (glucagon) known to influence blood-sugar levels, shined as a candidate.

Mojsov puzzled over the predicted sequence of GLP-1’s supposed 37-amino-acid chain. The presence of the same amino acids in the same positions of different proteins suggests that they perform an important function, yet GLP-1 begins with a stretch of six amino acids that don’t exist in glucagon or other molecular relatives. Mojsov eyed an arginine at position 6. Arginines are clipped by well-known human enzymes, and if GLP-1 started after that amino acid, the resulting peptide—now 31 rather than 37 amino acids in length—would align perfectly with its glucagon family members.

Getting to the crux of the GLP-1 matter

Mojsov set out to determine whether the shorter version of GLP-1[GLP-1 (7-37)] might be liberated from the longer one [GLP-1 (1-37)] and serve as the missing incretin. Toward that end, she synthesized large amounts of each pure peptide in single batches, thus setting herself up to ensure consistency in subsequent studies. She made antibodies that bind a common region; therefore, they recognize both variants. Crucially, she also figured out how to segregate GLP-1 (1-37) and GLP-1 (7-37) from within a mixture, exploiting the charge on the amino acids that are unique to the longer molecule. These innovations equipped her to find GLP-1 in tissue, distinguish GLP-1 (7-37) from GLP-1 (1-37), and pinpoint the active peptide.

Thus, the pièces de résistance in the early stages of the GLP-1 discoveries were these invaluable reagents and methods, which provided scientists with the means to draw unambiguous conclusions about essential aspects of GLP-1 biology.

Mojsov then conducted the first set of definitive experiments. She radioactively tagged her peptides and deployed them with GLP-1 antibodies to check whether GLP-1 shows up in animals. It does. Mojsov then separated the peptides and established that the truncated GLP-1 (7-37) composes a significant fraction of the total. This smaller peptide thus exists in nature and, notably, in the intestine, Mojsov, Habener, and their collaborators reported in 1986.

Mojsov and Habener teamed up with Gordon Weir (Joslin Diabetes Center) and, in 1987, demonstrated that tiny concentrations of pure GLP-1 (7-37), such as those in the bloodstream, stimulate insulin secretion from isolated rat pancreases that continue to function even when removed from the body. The longer form remains inert even at 10,000-fold higher concentrations. These observations revealed that GLP-1 (7-37), hereafter referred to as GLP-1, is the physiologically relevant peptide.

Habener and Mojsov then advanced to human studies. With David Nathan (Massachusetts General Hospital), they determined that GLP-1 prompts insulin release and lowers circulating blood glucose levels. This 1992 publication built the case that the hormone might provide the foundation for a safe diabetes drug, and several companies, including Novo Nordisk, AstraZeneca, Eli Lilly, and GSK latched onto this idea. Soon GLP-1’s potential would expand.

Fatty acids, plump with potential

A couple of years later, Knudsen took the helm of GLP-1 therapeutic development at Novo Nordisk, and in 1996, a paper caught her attention. Stephen Bloom (Hammersmith Hospital, London) had injected GLP-1 into rats’ brains, and the animals’ food intake plummeted. The peptide, Bloom proposed, sends a satiety signal.

The possibility that the hormone might fight obesity as well as diabetes had already piqued Knudsen’s interest due to hints from earlier animal studies led by Ole Madsen (Hagedorn Research Institute, Copenhagen). The notion that a molecule pegged for one disease could also combat another broke convention, and as Knudsen championed the idea, she faced resistance. To harness it for either use, though, she had to overcome a show-stopping limitation.

In the human body, GLP-1 vanishes minutes after it enters the bloodstream. An enzyme called dipeptidyl peptidase 4 (DPP-4) chews it up and the kidneys purge the rest. To transform GLP-1 into a drug, scientists would need to render it able to survive these assaults. Knudsen aimed to make an agent that would remain active for 24 hours after a single injection under the skin.

After toying with a slow-release formulation and ones that resisted DPP-4-mediated destruction, she settled on a strategy of attaching fatty acids to GLP-1. Fatty acids naturally adhere to an abundant protein in the circulation called albumin, which transports substances around the body. According to Knudsen’s vision, albumin would ferry its GLP-1 cargo through the bloodstream while protecting it from enzymatic destruction and renal filtration. The fatty acid would gradually unleash GLP-1 so that it can grasp its receptor on target cells and trigger its effects.

Knudsen’s team made GLP-1 analogs that varied in several ways, including fatty-acid length and type, attachment site within GLP-1, and chemical linker. Then the investigators assessed the compounds’ behavior. They aimed to stabilize the peptide and lengthen its longevity in animals while maximizing potency and keeping the amino acid sequence as close as possible to that of human GLP-1 to circumvent immunoreactivity. Two papers, in 2000 and 2007, detailed the results.

The researchers zeroed in on a candidate that they named liraglutide (see Figure). They had extended its half-life after subcutaneous injection from 1.2 to 13 hours. It performed well in a 2010 clinical trial of 1300 people with type 2 diabetes, and adverse events were mostly mild or moderate. The European Medicines Agency (EMA) approved liraglutide (Victoza®) to control blood sugar levels in type 2 diabetes in 2009, and the U.S. Food and Drug Administration (FDA) followed the next year. Liraglutide thus became the first once-daily GLP-1-based drug.

A feast of success

In the meantime, accumulating data supported the notion that GLP-1 reduces appetite and body weight, and Knudsen’s team pursued liraglutide for this purpose. In a key study, nondiabetic, obese or overweight subjects lost an average of more than 12 pounds over a year. More than one-third of the individuals in the liraglutide group lost at least 5% of their body weight and almost one-fourth lost more than 10%. Liraglutide makes people feel more satiated and less hungry, so they voluntarily eat less. The FDA and EMA gave it the green light in 2014 and 2015, respectively, and it was the first GLP-1-based drug approved for the treatment of obesity (Saxenda®).

The Novo Nordisk scientists wanted to go a step further; make the medicine last not for a day, but for a week. It would have to stick optimally to albumin: tightly enough to persist longer in the body, but loosely enough that sufficient quantities would let go to bind the receptor.

The team that sought this therapeutic, led by chemists Jesper Lau and Thomas Kruse, replaced one amino acid in GLP-1 with a molecule that confers resistance to DPP-4 cleavage and then systematically tested different fatty acids and chemical linker combinations. The investigators combed through about 4000 compounds to home in on one whose half-life grew dramatically—to 165 hours. They named it semaglutide (see Figure).

Semaglutide gained FDA approval for treating diabetes (Ozempic®) in 2017 and obesity (Wegovy®) in 2021. The agent fosters almost twice as much average weight loss as liraglutide does: 28 pounds over 16 months. Semaglutide’s side effects are mostly minor, but serious gastrointestinal problems cause some individuals to discontinue the drug. More than one million people in the U.S. have received prescriptions for Wegovy® since it entered the market.

Liraglutide and semaglutide have opened new avenues to powerful second-generation drugs. Eli Lilly’s tirzepatide, which contains not only GLP-1, but also another incretin called GIP, promotes even more dramatic effects than semaglutide does. The company has added glucagon to further fortify it, and individuals on a candidate called retatrutide lose, on average, more than 20% of their body weight.

Unlike GLP-1’s impact on diabetes, which maps primarily to the pancreas, its appetite-suppression activities lie mainly in the brain, and numerous investigators, including Knudsen, are detailing its behavior there. Researchers are probing its use in a tremendous range of illnesses, including chronic kidney disorders, fatty liver disease, neurodegenerative conditions such as Alzheimer and Parkinson’s diseases, and addiction. GLP-1-based therapies also protect the cardiovascular system, and earlier this year, the FDA approved semaglutide to reduce heart attack and stroke in people who have preexisting cardiovascular disease and are overweight or obese.

In addition to the scientists above, many others made key contributions to the GLP-1 story. These individuals include the late Werner Creutzfeldt (University of Göttingen), Richard DiMarchi (then at Eli Lilly), Daniel Drucker (University of Toronto), John Eng (then at the Bronx Veterans Affairs Medical Center), Jens Holst (University of Copenhagen), Michael Nauck (Ruhr-University Bochum), and Nancy Thornberry (then at Merck).

Through their ambitious and committed endeavors, Habener, Mojsov, and Knudsen have transformed the health prospects for the tremendous number of people whose surplus weight compromises their wellbeing. Their work has launched a new battery of blockbuster drugs that is propelling GLP-1 into a pharmaceutical realm with unprecedented reach.

by Evelyn Strauss

Selected Publications Joel F. Habener, Svetlana Mojsov, and Lotte Bjerre Knudsen

Lund PK, Goodman RH, Dee PC, and Habener JF. (1982). Pancreatic preproglucagon cDNA contains two glucagon-related coding sequences arranged in tandem. Proc. Natl. Acad. Sci. USA. 79, 345-349.

Mojsov S, Heinrich G, Wilson IB, Ravazzola M, Orci L, and Habener JF. (1986). Preproglucagon gene expression in pancreas and intestine diversifies at the level of post-translational processing. J. Biol. Chem. 261, 11880-11889.

Mojsov S, Weir GC, and Habener JF. (1987). Insulinotropin: glucagon-like peptide I (7-37) co-encoded in the glucagon gene is a potent stimulator of insulin release in the perfused rat pancreas. J. Clin. Invest. 79, 616-619.

Mojsov S. (1992). Structural requirements for biological activity of glucagon-like peptide-I. Int. J. Pept. Prot. Res. 40, 333-343.

Nathan DM, Screiber E, Fogel H, Mojsov S, and Habener JF. (1992). Insulinotropic actions of glucagon-like peptide-I (7-37) administered to diabetic and non-diabetic human subjects. Diabetes Care. 15, 270-276.

Knudsen LB, Nielsen PF, Huusfeldt PO, Johansen NL, Madsen K, Pedersen FZ, Thøgersen H, Wilken M, and Agersø M. (2000). Potent derivatives of glucagon-like peptide-1 with pharmacokinetic properties suitable for once daily administration. J. Med. Chem. 43, 1664-1669.

Madsen K, Knudsen LB, Agersoe H, Nielsen PF, Thøgersen H, Wilken M, Johansen NL. (2007). Structure-activity and protraction relationship of long-acting glucagon-like peptide-1 derivatives: importance of fatty acid length, polarity, and bulkiness. J. Med. Chem. 50, 6126-6132.

Knudsen LB, and Lau J. (2019). The discovery and development of liraglutide and semaglutide. Front. Endocrinol. 10, 1-32.