For most of the modern era of medicine, obesity has been a disease without an effective drug. Stimulants worked but were unsafe. Lipase inhibitors were modest at best. Surgery worked but was invasive. Then GLP-1 agonists arrived, and over the course of about a decade transformed the field. Semaglutide (Ozempic, Wegovy) and Tirzepatide (Mounjaro, Zepbound) are now producing weight loss outcomes that approach bariatric surgery — with a weekly injection. The cultural impact has been enormous, but the underlying science is just as remarkable.
This guide unpacks the biology of incretin hormones, the protein engineering that turned a 2-minute peptide into a weekly drug, and the trial data that established how these molecules actually work in patients. You can explore the molecules directly in the Drug Discovery Lab.
Incretin biology in three minutes
When you eat, your gut releases hormones that prepare the rest of your body for the incoming nutrients. These hormones are called incretins. The two main ones are GLP-1 (glucagon-like peptide-1), released by L cells in the distal small intestine and colon, and GIP (glucose-dependent insulinotropic peptide), released by K cells in the upper small intestine.
GLP-1 does several things in parallel. In the pancreas, it binds the GLP-1 receptor on beta cells and amplifies glucose-dependent insulin release. Crucially, it does this only when blood glucose is elevated, which is why GLP-1 agonists rarely cause hypoglycemia by themselves. It also suppresses glucagon release from alpha cells, slows gastric emptying so glucose enters the bloodstream more gradually, and acts on neurons in the hypothalamus and brainstem to reduce appetite and food intake.
Type 2 diabetes is associated with a blunted incretin response. The therapeutic question that drove drug development for two decades was: what if you could amplify or replace the incretin signal pharmacologically?
The half-life problem
The obstacle was that natural GLP-1 has a plasma half-life of only about 1-2 minutes. The reason is an enzyme called dipeptidyl peptidase-4 (DPP-4), which sits on endothelial cell surfaces and rapidly chops off the first two residues of GLP-1. Once cleaved, GLP-1 loses its receptor activity. You cannot turn a 2-minute peptide into a useful drug without either inhibiting DPP-4 or engineering a peptide that DPP-4 cannot cut.
Both routes have produced approved drugs. DPP-4 inhibitors (Sitagliptin, Saxagliptin, and others) became modestly used oral drugs for type 2 diabetes — they work, but they only amplify the patient's own GLP-1 a little bit, and they don't produce significant weight loss. The other route — engineering a longer-acting GLP-1 agonist — turned out to be the bigger story.
Exenatide and Liraglutide: the first generation
The first GLP-1 agonist to reach the market was Exenatide (Byetta), based on Exendin-4, a peptide isolated from the saliva of the Gila monster. Exendin-4 is naturally resistant to DPP-4 cleavage because its second residue is glycine instead of alanine, and it has reasonable affinity for the human GLP-1 receptor. Exenatide was approved in 2005 and worked, but required twice-daily injections.
Novo Nordisk took a different approach with Liraglutide (Victoza, approved 2010, and later Saxenda for obesity). They kept the natural GLP-1 backbone but added a 16-carbon palmitic acid side chain via a glutamic acid spacer. The fatty acid chain binds reversibly to circulating serum albumin, which does two things: it shields the peptide from DPP-4 and glomerular filtration, and it creates a slow-release reservoir of bound peptide that gradually equilibrates with the free active form. The result was a once-daily injection.
Semaglutide: the breakthrough
Semaglutide built on Liraglutide and pushed the engineering further. Three modifications matter:
- Position 8 substitution. The natural alanine at position 8 is replaced with alpha-aminoisobutyric acid (Aib), a non-natural amino acid that DPP-4 cannot cleave. This blocks the primary degradation route at its source.
- C18 fatty diacid linker. A C18 fatty diacid chain is attached via a longer hydrophilic spacer to the lysine at position 26. The longer chain binds serum albumin much more tightly than Liraglutide's C16, extending the plasma half-life to about a week.
- Position 34 substitution. A lysine at position 34 is changed to arginine to prevent the fatty acid chain from accidentally attaching at that site during synthesis, ensuring clean conjugation only at position 26.
The combined effect is a peptide that resists DPP-4, hides from renal clearance by binding albumin, and reaches the GLP-1 receptor with high affinity. Semaglutide can be administered as a once-weekly subcutaneous injection — and eventually, as an oral tablet (Rybelsus) using an absorption enhancer to get a small fraction across the gut wall.
Semaglutide was approved for type 2 diabetes (as Ozempic) in 2017 and for chronic weight management (as Wegovy) in 2021. For more on the peptide engineering details, see our companion peptide-focused article on Semaglutide vs Tirzepatide.
The STEP trials and the obesity story
The STEP program (Semaglutide Treatment Effect in People with Obesity) tested Semaglutide at 2.4 mg weekly for chronic weight management. STEP 1, published in 2021, randomized nearly 2,000 adults without diabetes to Semaglutide or placebo, both with lifestyle support, for 68 weeks. Average weight loss in the Semaglutide arm was about 15 percent of body weight, compared to about 2 percent on placebo.
That magnitude of weight loss had never been achieved with a non-surgical drug therapy at that scale and consistency. Bariatric surgery still produces more weight loss on average, but Semaglutide closed the gap dramatically. Subsequent STEP trials confirmed the effect across different patient populations and combinations.
Tirzepatide: dual agonism
Eli Lilly took the GLP-1 story in a different direction with Tirzepatide. Instead of optimizing a pure GLP-1 agonist, they designed a single peptide that activates both the GLP-1 receptor and the GIP receptor — the other major incretin. GIP biology had been considered a dead end for years because GIP receptor agonists alone produced weight gain, not loss. But the combined effect of activating both receptors with one molecule turned out to be greater than either alone.
Tirzepatide is a 39-residue peptide with extensive modifications, including a C20 fatty diacid for albumin binding and several non-natural residues to optimize stability and dual receptor engagement. Approved in 2022 for type 2 diabetes (as Mounjaro) and 2023 for obesity (as Zepbound), it has set new clinical benchmarks for both.
The SURMOUNT-1 obesity trial published in 2022 randomized adults with obesity to Tirzepatide at 5, 10, or 15 mg weekly or placebo, with lifestyle support, for 72 weeks. Average weight loss at the 15 mg dose was about 21 percent of body weight — meaningfully greater than Semaglutide's STEP results, and approaching the lower end of bariatric surgery outcomes for some patients.
How appetite suppression actually works
The most surprising thing about GLP-1 agonists is that the weight loss is dominated by appetite suppression, not by reduced absorption or increased metabolic rate. Patients on Semaglutide and Tirzepatide consistently report feeling less hungry and feeling full sooner. The mechanism involves direct action on neurons in the arcuate nucleus of the hypothalamus and the area postrema in the brainstem, both of which are outside the blood-brain barrier and accessible to circulating GLP-1 analogs.
These neurons are part of the brain's energy homeostasis circuitry. When activated by a GLP-1 signal, they release signals that suppress feeding behavior and increase satiety. It is essentially a pharmacological trick to make patients feel full earlier in the meal.
Side effects and what we don't know
The most common side effects are gastrointestinal — nausea, vomiting, constipation, diarrhea — and they correlate with the dose-escalation phase. Most patients tolerate the drugs reasonably well after slow titration, but a meaningful percentage discontinue because of GI symptoms.
Less common but more serious concerns include pancreatitis, gallbladder disease, and (in rat studies) thyroid C-cell tumors. The thyroid signal has not been clearly replicated in humans, but both Semaglutide and Tirzepatide carry a boxed warning. Effects on muscle mass and bone density are an active area of research given the rapid magnitude of weight loss. Long-term safety data is still accumulating.
Explore the molecules in SciRouter
Both drugs have dedicated workspaces in the Drug Discovery Lab:
- Semaglutide — modified GLP-1 analog with Aib8 and C18 fatty diacid
- Tirzepatide — dual GLP-1/GIP agonist
Bottom line
GLP-1 agonists are the result of three decades of patient work in incretin biology, peptide engineering, and clinical trial design. They work because the underlying physiology is real: GLP-1 already regulates how your body responds to food, and turning that signal up changes both glucose handling and appetite. The protein engineering — Aib8 substitutions, fatty acid albumin tethers, dual receptor agonism — is some of the most elegant work in modern biotech, and the clinical results have rewritten what is possible for obesity and diabetes pharmacotherapy.