Peptides
Ozempic is a peptide: what that shared chemistry actually explains
Semaglutide and tirzepatide — the molecules in Wegovy, Ozempic and Zepbound — are peptides, the same chemical class as the BPC-157 and TB-500 vials sold online as "research chemicals." That is not a rhetorical comparison; it is what the FDA-approved labels say. Semaglutide is a chain of 31 amino acids carrying a C18 fatty diacid, and its label describes a "peptide backbone … produced by yeast fermentation". Four things stop being arbitrary once you know that: the injection, the refrigerator, why the same sequence from two sources is not the same product, and why one federal statute governs both the weight-loss market and the longevity-clinic peptide market.
What a peptide is, precisely
A peptide is a chain of amino acids joined end to end by peptide bonds. That is the whole chemical definition. Longer chains are called proteins, but the boundary is a naming convention, not a chemical discontinuity — a 2022 review in Signal Transduction and Targeted Therapy defines therapeutic peptides as agents "composed of a series of well-ordered amino acids, usually with molecular weights of 500–5000 Da": a range, not a cliff. We cover the word itself, and the three unrelated markets that use it, in what are peptides.
Where chemistry declines to draw a line, US regulation has drawn one anyway. Under 21 CFR 600.3(h)(6), "a protein is any alpha amino acid polymer with a specific, defined sequence that is greater than 40 amino acids in size." A polymer above that size is a protein and so a biological product; below it, a molecule is generally handled on the drug pathway. Semaglutide, at 31 residues, and tirzepatide, at 39, both sit under the line. The rule sorts regulatory categories; it does not by itself settle which application any particular company files.
The receipts: what the labels say these molecules are
Section 11 of the Wegovy prescribing information describes semaglutide as a GLP-1 receptor agonist whose peptide backbone is grown in yeast, then explains the engineering: "The main protraction mechanism of semaglutide is albumin binding, facilitated by modification of position 26 lysine with a hydrophilic spacer and a C18 fatty di-acid," with a further change at position 8 to resist the enzyme DPP-4. Molecular weight: 4,113.58 g/mol. That fatty-acid tail is why weekly dosing is possible at all — semaglutide is "extensively bound to plasma albumin (greater than 99%)", which slows renal clearance and shields it from breakdown, giving an elimination half-life of approximately one week. The 2015 discovery paper says the same: semaglutide "has two amino acid substitutions compared to human GLP-1 (Aib(8), Arg(34)) and is derivatized at lysine 26".
Tirzepatide is built the same way from a different starting sequence. Its label states it "is based on the GIP sequence and contains aminoisobutyric acid (Aib) in positions 2 and 13, a C-terminal amide, and Lys residue at position 20 that is attached to 1,20-eicosanedioic acid via a linker," at 4,813.53 Da. That C20 fatty diacid "enables albumin binding and prolongs the half-life"; tirzepatide is 99% albumin-bound with an elimination half-life of approximately 5 to 6 days. A 2023 review in American Journal of Therapeutics describes it as "a synthetic chemical structure based on the GIP sequence" that "consists of 39 amino acid peptides."
Both are short amino-acid chains with a fat molecule bolted on, inside the 500–5,000 Da band. Whatever separates them from a vial of BPC-157, it is not the chemical class.
Why you inject them — and what changed
Your digestive tract exists to take protein apart into amino acids, and it cannot tell a steak from a therapeutic peptide — so a swallowed peptide is mostly destroyed before it reaches the bloodstream. A 2026 review of oral peptide delivery puts it plainly: peptides "often exhibit unfavorable physicochemical properties for epithelial permeability, paired with high polarity, charge, molecular size, and proteolytic instability," which "collectively result in low and often insufficient systemic bioavailability." Injecting under the skin skips the gut.
One label quantifies the gap. Wegovy's prescribing information reports that "absolute bioavailability of semaglutide is 89% following subcutaneous administration" — and approximately 1% to 2% following oral tablet administration. Same molecule, roughly fiftyfold difference, entirely because of what the gut does to a peptide.
The pill that is still a peptide
Oral semaglutide exists anyway, and it works by carrying a helper. FDA approved a 25 mg oral Wegovy tablet for weight management on December 22, 2025. The label explains the mechanism: "WEGOVY tablets are co-formulated with SNAC which facilitates the absorption of semaglutide after oral WEGOVY tablet administration. The absorption of semaglutide predominantly occurs in the stomach." SNAC — salcaprozate sodium — appears on the label as an inactive ingredient alongside magnesium stearate; its job is to move an intact peptide across the stomach lining in a narrow window.
That is why the dosing rules are so fussy. The label directs patients to take the tablet "orally once daily on an empty stomach in the morning with water (up to 4 ounces)" and to "wait at least 30 minutes before eating food, drinking beverages or taking other oral medications." Dilution and food work against a chemistry already running at 1% to 2% efficiency. We compare the oral options and their trial evidence in oral GLP-1 pills in 2026.
The pill that is not a peptide at all
Orforglipron is the sharpest contrast here. FDA approved it as Foundayo on April 1, 2026, described by Eli Lilly as "a once-daily small molecule (non-peptide) oral glucagon-like peptide-1 receptor agonist". Its label gives the formula for orforglipron calcium as C48H47F2N10O5·0.5Ca and the molecular weight as 902.0 g/mol. There is no amino-acid chain in it. And because there is no peptide to digest, the administration instruction collapses to one line: "Take FOUNDAYO orally once daily, with or without food." (Foundayo still has handling rules — the label calls it light-sensitive and keeps it in the original bottle and carton — but none about timing food or water.) Peptide-in-a-pill needs an absorption enhancer and a 30-minute fast. Not-a-peptide needs neither.
Five molecules, one table
| What kind of molecule | How it is taken | FDA status | Compounding from bulk | |
|---|---|---|---|---|
| Semaglutide | Peptide — 31 amino acids, C18 fatty diacid at Lys26; 4,113.58 g/mol | Weekly subcutaneous injection, or a daily tablet co-formulated with SNAC | Approved (Wegovy, Ozempic, Rybelsus); oral Wegovy 25 mg approved Dec 22, 2025 | FDA proposed April 30, 2026 to exclude it from the 503B bulks list; comments closed July 30, 2026; no final determination as of Aug 4, 2026 |
| Tirzepatide | Peptide — 39 amino acids, based on the GIP sequence, C20 fatty diacid at Lys20; 4,813.53 Da | Weekly subcutaneous injection | Approved (Zepbound, Mounjaro) | Same 503B proposal, same open status |
| Retatrutide | Peptide — a non-endogenous GIP / GLP-1 / glucagon triple receptor agonist | Weekly subcutaneous injection (in trials) | Investigational. Lilly: "cannot be legally sold or marketed for human use"; BLA planned Q1 2027 | Not applicable — no approved product. FDA has warned sellers marketing it "for research purposes" |
| Orforglipron | Not a peptide — small molecule, 902.0 g/mol | Daily tablet, with or without food | Approved as Foundayo, April 1, 2026 | Not named in the April 2026 503B proposal, which covered semaglutide, tirzepatide and liraglutide |
| BPC-157 | Peptide — a synthetic sequence with no approved therapeutic use | Sold as vials for injection by unregulated vendors | Not approved for any indication | FDA's Pharmacy Compounding Advisory Committee voted 8–6 on July 23, 2026 to recommend it for the 503A bulks list, against FDA staff advice. The vote is non-binding and FDA has not acted |
Every cell above is sourced in the Sources list at the foot of this page.
The cold chain is a chemistry problem, not a policy preference
Peptides in solution degrade. A 2025 stability study in the European Journal of Pharmaceutics and Biopharmaceutics subjected semaglutide to thermal stress at 25 °C, 40 °C, 60 °C and 80 °C and identified thirteen known impurities. Its opening line is the general case: "Peptides are fragile and susceptible to degradation during formulation, storage, and transportation," forming impurities "that may impact safety, efficacy, immunogenicity, and regulatory compliance."
Approved labels convert that chemistry into rules with numbers attached — and the numbers belong to a specific container in a specific state, not to the drug in general. A Wegovy single-dose pen or syringe is stored at 2 °C to 8 °C and, prior to cap removal, "can be kept from 8°C to 30°C (46°F to 86°F) up to 28 days," protected from light, never frozen; the four-dose Wegovy FlexTouch pen runs on a different clock, discarded once it has been out of the refrigerator for 56 days. Zepbound's single-dose pen and vial are tighter still: refrigerated at 2 °C to 8 °C, kept unrefrigerated at up to 30 °C for no more than 21 days, and once at room temperature "it should not be returned to the refrigerator"; its multi-dose vial and KwikPen run on a 30-day clock instead.
Note what is not refrigerated: Wegovy tablets and Foundayo tablets both sit at controlled room temperature. The fragile thing is a peptide dissolved in water — what is in an injection pen, and what arrives in a padded envelope from an unregulated seller. FDA's consumer page makes the point: "Injectable GLP-1 drugs require refrigeration" and arriving warm can affect quality. Every one of those windows — 28 days, 56, 21, 30 — exists because a manufacturer generated stability data for that container. A vial with no stability data has no window; it has a guess. How that plays out in summer is in GLP-1s in a heat wave.
Why an identical sequence is not an identical product
Two vials can hold the same amino-acid sequence and be entirely different medicines, because a drug product is the sequence plus what it is suspended in and what was measured before it shipped. A Zepbound single-dose pen contains tirzepatide plus sodium chloride, sodium phosphate dibasic heptahydrate and water for injection, at a pH of 6.5 to 7.5; the four-dose multi-dose vial and KwikPen instead list benzyl alcohol, glycerin and phenol as well, because a container entered four times needs antimicrobial preservation and a single-dose one does not. Every one of those choices was reviewed. For a research-grade vial there is no label, no released potency assay, no sterility or endotoxin result, and no independent party that has verified any of it.
As of FDA's June 15, 2026 update, the agency had received 990 reports of adverse events associated with compounded semaglutide and more than 730 associated with compounded tirzepatide, both as of May 31, 2026. It has also warned sellers: in a March 31, 2026 letter to Gram Peptides over products sold as "GLP-1-R peptide" (retatrutide) and "GLP-2 peptide" (tirzepatide), FDA said they "are unapproved new drugs under section 505(a) of the Federal Food, Drug, and Cosmetic Act" — a research-use disclaimer does not change what a product legally is. We track that enforcement wave in FDA's GLP-1 telehealth warning letters and the pharmacy-versus-vendor line in compounded versus branded GLP-1s.
One statute, two directions
Compounding runs on two adjacent sections of the same law. Under section 503A a pharmacy compounds for an individual patient, with drugs "compounded based on the receipt of valid patient-specific prescriptions"; under section 503B, registered outsourcing facilities compound without them but are "subject to CGMP requirements" and FDA inspection. Each section has its own list of permitted bulk substances; we set the two lists side by side in what are peptides. Both moved in 2026, in opposite directions, twelve weeks apart.
On the 503B side, FDA announced on April 30, 2026 that it was "proposing to exclude semaglutide, tirzepatide, and liraglutide" from the 503B bulks list, having "not identif[ied] a clinical need for outsourcing facilities to compound" them. The notice published May 1, 2026; the comment period was extended and closed July 30, 2026. A Federal Register check on August 4, 2026 shows no further document in docket FDA-2018-N-3240: the proposal stands as a proposal. What that means for people on a compounded prescription is covered in the FDA compounded-GLP-1 endgame.
On the 503A side, FDA's Pharmacy Compounding Advisory Committee met July 23–24, 2026 to consider seven peptides for the 503A bulks list and recommended six. On the tallies reported by STAT, BPC-157 — taken on day one — was recommended 8–6, and only emideltide was rejected, 6–7; STAT also reports that "FDA staff told the panelists they recommended against giving compounders the green light due to a lack of clinical evidence demonstrating safety and efficacy." FDA's meeting page states the constraint: "Advisory committees make non-binding recommendations to the FDA, which generally follows the recommendations but is not legally bound to do so." Nothing about BPC-157's legal status changed on July 23; the evidence, and the fuller vote record, are in our BPC-157 review.
Retatrutide is the next one in line
Retatrutide is a peptide too — a "not endogenously occurring peptide" that activates the glucagon, GIP and GLP-1 receptors at once. On July 23, 2026, Lilly said it "plans to submit a Biologics License Application (BLA) for retatrutide to FDA in Q1 2027". We report that as the company worded it: the release gives no reason for the choice of application, and we take no view on the pathway FDA will ultimately require. The same release states that "retatrutide is an investigational molecule that cannot be legally sold or marketed for human use." It is nevertheless already for sale online, which is how it came to be named in the Gram Peptides letter four months earlier. Our companion piece covers the TRIUMPH trial results.
Common questions
Is Ozempic a peptide?
Yes. Ozempic, Wegovy and Rybelsus all contain semaglutide — a 31-amino-acid chain of 4,113.58 g/mol whose FDA-approved label describes a peptide backbone produced by yeast fermentation and modified with a C18 fatty diacid.
If the sequence matches, is a research-vendor peptide equivalent to a prescription one?
No — the sequence is the least informative part of the comparison. An approved product carries a released potency assay, sterility and endotoxin testing, a defined excipient list, stability data supporting specific storage limits, and a manufacturer accountable to FDA inspection. A vial labeled "research use only" carries none of these, and FDA has said in warning letters that the disclaimer does not change the product's legal status.
What follows from all this
The parts of the experience that feel like bureaucracy — the refrigerator, the 21-day window, the empty stomach, the pharmacy — are downstream of chemistry that was measured and published. Where the molecule came from is not a detail; it is most of the product. For specifics see Wegovy versus Zepbound, semaglutide results in women, GLP-1 costs without insurance, what peptide therapy costs and our GLP-1 FAQ. For a head-to-head between the two approved injections, the randomized comparison is SURMOUNT-5 — not two separate trials read side by side.
Sources
- DailyMed (National Library of Medicine). WEGOVY (semaglutide) injection and tablets — full prescribing information, revised 6/2026. Source for the Description of semaglutide's peptide backbone and C18 fatty diacid, >99% albumin binding, ~1-week half-life, 89% subcutaneous and 1–2% oral bioavailability, SNAC co-formulation, tablet dosing instructions, and the single-dose (28-day), FlexTouch (56-day) and tablet storage limits. dailymed.nlm.nih.gov/…/setid=ee06186f…
- US Food and Drug Administration. ZEPBOUND (tirzepatide) injection — full prescribing information (NDA 217806, 2026). Source for the GIP-based structure, Lys20 / 1,20-eicosanedioic acid linkage, 4,813.53 Da, 99% albumin binding, the 5–6 day elimination half-life given in section 12.3, the single-dose and multi-dose excipient lists and pH, and the 21-day (single-dose) and 30-day (multi-dose vial / KwikPen) storage windows. accessdata.fda.gov/drugsatfda_docs/label/2026/217806s002lbl.pdf
- DailyMed (National Library of Medicine). FOUNDAYO (orforglipron) tablets — full prescribing information. Source for the molecular formula, 902.0 g/mol molecular weight, and "with or without food" administration. dailymed.nlm.nih.gov/…/setid=8ac446c5…
- US Government Publishing Office. 21 CFR 600.3(h)(6) (2023 annual edition, title 21 vol. 7) — "A protein is any alpha amino acid polymer with a specific, defined sequence that is greater than 40 amino acids in size." govinfo.gov/…/CFR-2023-title21-vol7-sec600-3.xml
- Wang L, et al. Therapeutic peptides: current applications and future directions. Signal Transduction and Targeted Therapy 2022;7:48 — source for the 500–5,000 Da definition of therapeutic peptides. pubmed.ncbi.nlm.nih.gov/35165272
- Malgave A, et al. Effect of pH, buffers, molarity, and temperature on solution state degradation of semaglutide using LC-HRMS. Eur J Pharm Biopharm 2025;214:114780 — source for semaglutide as a 31-amino-acid peptide with a C18 fatty diacid, and for peptide thermal degradation and impurity formation. pubmed.ncbi.nlm.nih.gov/40490042
- Wong E, Cope R, Dima L, Nguyen T. Tirzepatide: A Dual Glucose-dependent Insulinotropic Polypeptide and Glucagon-Like Peptide-1 Agonist for the Management of Type 2 Diabetes Mellitus. Am J Ther 2023;30(1):e26–e35 — source for tirzepatide's 39 amino acids. pubmed.ncbi.nlm.nih.gov/36516422
- Lau J, et al. Discovery of the Once-Weekly Glucagon-Like Peptide-1 (GLP-1) Analogue Semaglutide. J Med Chem 2015;58(18):7370–7380 — source for the Aib8 / Arg34 substitutions and Lys26 derivatization. pubmed.ncbi.nlm.nih.gov/26308095
- Khalid N-U-A, Rivera-Delgado E, von Erlach T. Navigating the complexity of oral peptide delivery: challenges and strategies to enhance oral bioavailability. Front Drug Deliv 2026;6:1809842 — source for proteolytic instability and low oral bioavailability of peptides. pubmed.ncbi.nlm.nih.gov/41953894
- Neumann J, Ahlrep U, Hofmann B, Gergs U. Inotropic effects of retatrutide in isolated human atrial preparations. Naunyn Schmiedebergs Arch Pharmacol 2026;399(1):317–327 — source for retatrutide being a "not endogenously occurring peptide" acting at GCGR, GIPR and the GLP-1 receptor. pubmed.ncbi.nlm.nih.gov/40613938
- US Food and Drug Administration. FDA's Concerns with Unapproved GLP-1 Drugs Used for Weight Loss (page updated June 15, 2026) — 990 adverse-event reports for compounded semaglutide and more than 730 for compounded tirzepatide, both as of May 31, 2026; refrigeration warning; "for research purposes" warning letters. fda.gov/drugs/drug-alerts-and-statements/fdas-concerns-unapproved-glp-1-drugs-used-weight-loss
- US Food and Drug Administration. Warning Letter to Gram Peptides, March 31, 2026 — retatrutide and tirzepatide sold as "research use only"; unapproved new drugs under section 505(a). fda.gov/…/warning-letters/gram-peptides-721806-03312026
- US Food and Drug Administration. Human Drug Compounding Laws — descriptions of sections 503A and 503B. fda.gov/drugs/human-drug-compounding/human-drug-compounding-laws
- US Food and Drug Administration. FDA Proposes to Exclude Semaglutide, Tirzepatide and Liraglutide from 503B Bulks List, April 30, 2026. fda.gov/news-events/press-announcements/fda-proposes-exclude-semaglutide-tirzepatide-and-liraglutide-503b-bulks-list
- Federal Register. List of Bulk Drug Substances for Which There Is a Clinical Need Under Section 503B; Extension of Comment Period, doc. 2026-12937, published June 26, 2026 — comment period closed July 30, 2026; docket FDA-2018-N-3240. federalregister.gov/documents/2026/06/26/2026-12937/…
- Federal Register. List of Bulk Drug Substances for Which There Is a Clinical Need Under Section 503B, doc. 2026-08552, published May 1, 2026 (91 FR 23431) — the underlying proposal. federalregister.gov/documents/2026/05/01/2026-08552/…
- US Food and Drug Administration. July 23–24, 2026 Meeting of the Pharmacy Compounding Advisory Committee — the seven peptides considered for the 503A bulks list, and the statement that advisory committee recommendations are non-binding. fda.gov/advisory-committees/…/july-23-24-2026-meeting-pharmacy-compounding-advisory-committee-07232026
- STAT. FDA peptide compounding panel backs epitalon, rejects emideltide, July 24, 2026 — source for the individual vote tallies, including BPC-157 at 8–6, and for FDA staff's contrary recommendation. statnews.com/2026/07/24/fda-peptide-compounding-panel-backs-epitalon-rejects-emideltide
- Eli Lilly and Company. FDA approves Lilly's Foundayo (orforglipron), April 1, 2026 — "a once-daily small molecule (non-peptide) oral glucagon-like peptide-1 receptor agonist" that "can be taken any time of the day without restrictions on food and water intake." investor.lilly.com/…/fda-approves-lillys-foundayotm-orforglipron-only-glp-1-pill
- Eli Lilly and Company. Retatrutide successful in two additional Phase 3 obesity trials, July 23, 2026 — BLA planned for Q1 2027; retatrutide "cannot be legally sold or marketed for human use." investor.lilly.com/…/lillys-triple-agonist-retatrutide-successful-two-additional
- Novo Nordisk. Wegovy pill approved in the US as the first oral GLP-1 for weight management, December 22, 2025 — 25 mg once-daily dose, US launch expected early January 2026. novonordisk.com/news-and-media/…/id=916472
- Aronne LJ, et al. Tirzepatide as Compared with Semaglutide for the Treatment of Obesity (SURMOUNT-5). NEJM 2025 Jul 3 — the randomized head-to-head between the two approved injections. pubmed.ncbi.nlm.nih.gov/40353578
This article is general information, not medical advice; discuss any weight-loss medication decision with a licensed clinician who knows your health history.