The Compound Report is an educational resource. Nothing on this site constitutes medical advice or encourages personal use of any compound. Always consult a qualified healthcare provider.

Learn · Foundations

What are peptides?

A peptide is a short chain of amino acids — the same building blocks proteins are made of, just fewer of them linked together. Most of the body’s signaling molecules (insulin, oxytocin, growth-hormone releasing hormone, and many others) are peptides; they tell cells what to do by binding to specific receptors on the cell surface. The compounds catalogued on this site are either copies of natural human peptides or engineered variants designed to last longer, bind more selectively, or resist being digested.

If you only remember three things
Pick your depth

The same answer at three reading levels. Open whichever one matches what you actually want to know.

Your body is built out of tiny lego pieces called amino acids. There are about 20 different shapes. When the body snaps a bunch of them together in the right order, it builds proteins — the stuff your hair, muscles, and skin are made from. A long lego chain is a protein.

A peptide is what we call a really shortlego chain. Maybe ten pieces. Maybe thirty. Short enough that it’s not really a building yet — it’s a message.

Your body uses these little messages all day long to tell cells what to do. A peptide might say:

  • “Hey skin cells — start making more collagen, we’ve got a cut to fix.”
  • “Hey brain — turn off hunger, the stomach is full.”
  • “Hey body — it’s nighttime, start the sleep stuff.”

Every message has a specific receiver. Cells have little locks on their surface, and each peptide is a key shaped to fit just one kind of lock. Wrong key, nothing happens. Right key, the cell does whatever the message says.

The compounds on this site are either copies of messages your body already makes, or new messages that scientists designed to do something useful — like “repair this tendon faster” or “eat less without feeling sad about it.”

One catch: if you swallow a peptide, your stomach treats it like food and chops it up before it can deliver the message. That’s why most of them have to be injected — to skip the stomach entirely and get into the bloodstream intact.

The building blocks

The body uses 20 amino acids as its primary building blocks. Each one has the same backbone (a central carbon attached to an amino group —NH₂, a carboxyl group —COOH, and a hydrogen) plus a unique side chain that gives it different properties: some are positively charged, some negatively charged, some hydrophobic, some can form hydrogen bonds.

How they link up

When two amino acids join, the amino group of one reacts with the carboxyl group of the next, releasing a water molecule and forming what’s called a peptide bond (chemically, an amide bond). String enough of these together and you have a peptide.

The terminology shifts with length:

  • Dipeptide — 2 amino acids
  • Oligopeptide — roughly 2 to 20
  • Polypeptide — more than 20
  • Protein — a folded polypeptide, usually 50+ amino acids

The cutoff between “peptide” and “protein” is fuzzy. Insulin is 51 amino acids — most people call it a protein, but its small size makes it behave a lot like a peptide.

Where they come from

Your body makes peptides in two ways. First, it can build them from scratch using ribosomes — the cellular machines that read genes and assemble amino-acid chains. Second, it can take a larger protein and clip it into smaller active peptides; oxytocin and many hormones are made this way.

Scientists make them in the lab using a technique called solid-phase peptide synthesis — invented by Bruce Merrifield (Nobel Prize, 1984). One amino acid at a time gets snapped onto a growing chain anchored to a resin bead. Modern automated synthesizers can build a 30-amino-acid peptide in hours.

How they work in the body

Peptides almost never enter cells directly. Instead, they bind to receptors on the cell surface — protein structures that recognize one specific peptide shape (think key/lock). When the right peptide binds, the receptor changes shape and triggers a cascade of events inside the cell: gene expression changes, calcium gets released, an enzyme gets activated, etc.

The specificity is part of why peptides are interesting therapeutics: a peptide drug usually hits one receptor type with high precision, instead of broadly affecting many systems the way small-molecule drugs often do.

Why they’re usually injected

Your digestive system is designed to break protein-like things down into their component amino acids — that’s how you extract nutrition from food. So when you swallow a peptide drug, the stomach’s acid and the peptidase enzymes in your gut treat it like dinner and rip it apart before it can be absorbed.

To avoid this, most peptide drugs are injected (usually subcutaneously — into fat just under the skin) or delivered through the nose. A few newer drugs use chemical tricks to survive the gut: oral semaglutide pairs the peptide with a permeation enhancer that lets it slip through the stomach lining quickly.

Formal definition

A peptide is a polymer of α-amino acids joined by amide (peptide) bonds — a condensation reaction between the carboxyl group (—COOH) of one residue and the α-amino group (—NH₂) of the next, releasing one molecule of water per bond. The α-carbon carries an amino group, a carboxyl group, a hydrogen, and a side chain (R group). The 20 proteinogenic amino acids encode the side chains that determine charge, polarity, aromaticity, and steric bulk.

The peptide-vs-protein cutoff is conventional rather than chemical — typically drawn around 50 residues, but molecules in the 30–100 range get classified either way depending on context (insulin’s 51 residues are usually called a protein; somatostatin’s 14 residues, a peptide).

Structural hierarchy

  • Primary structure — the linear amino-acid sequence, conventionally written N-terminus to C-terminus.
  • Secondary structure — local conformations stabilized by backbone hydrogen bonds: α-helices (3.6 residues per turn), β-sheets (parallel or antiparallel), β-turns. Even short peptides can adopt stable secondary structure when bound to their receptor.
  • Tertiary structure — overall 3D fold of a single chain, driven by hydrophobic packing, disulfide bridges (oxidized cysteine pairs), and electrostatic interactions.
  • Quaternary structure — assembly of multiple chains into a functional complex (insulin: A and B chains held together by two interchain disulfides; haemoglobin: four subunits).

Biosynthesis vs synthetic production

Endogenous peptides are produced either ribosomally (translated from mRNA) or non-ribosomally (by specialized enzyme complexes — common in microbes, rare in mammals). Most signaling peptides are translated as longer prepropeptides that get cleaved by signal peptidases and prohormone convertases into the mature active form. Oxytocin, for example, is excised from a 125-residue precursor that also yields neurophysin I.

Synthetic production overwhelmingly uses solid-phase peptide synthesis (SPPS), originally developed by Bruce Merrifield. The growing chain is anchored to a resin via the C-terminus. Each cycle: deprotect the α-amino group, couple the next protected amino acid, wash off excess reagent. Fmoc protection chemistry dominates modern practice over the older Boc strategy. Cleavage from the resin yields the crude peptide, which is then purified by reversed-phase HPLC and identity-confirmed by mass spectrometry. Manufacturing quality assurance (whether a vial actually contains what the label says) is the central sourcing concern for any non-pharmaceutical peptide.

Receptor mechanisms

Most therapeutic peptides act on cell-surface receptors — they’re too hydrophilic to cross lipid membranes passively. Major target classes:

  • G-protein coupled receptors (GPCRs) — the largest single class of peptide targets. GLP-1 receptor, melanocortin receptors, oxytocin receptor, and most ghrelin and growth-hormone secretagogue receptors are GPCRs. Binding triggers G-protein activation, second-messenger cascades (cAMP, IP₃, Ca²⁺), and ultimately changes in gene expression or ion-channel state.
  • Receptor tyrosine kinases (RTKs) — insulin receptor and IGF-1 receptor. Peptide binding induces receptor dimerization and autophosphorylation, recruiting downstream signaling adaptors.
  • Cytokine receptors and others — growth hormone receptor (a type-I cytokine receptor that uses JAK-STAT signaling), guanylate cyclase receptors (atrial natriuretic peptide), etc.

Affinity is typically nanomolar (binding constant 10⁻⁹ M range), and receptor selectivity is much higher than is usually achievable with small-molecule drugs targeting the same biology — a major reason peptides are favored for targets like incretin receptors where off-target activity carries meaningful toxicity.

Pharmacokinetics — the hard part

Native peptides are pharmacokinetic nightmares:

  • Oral bioavailability is near zero for unmodified peptides. Gastric pH denatures most fold structure, and intestinal peptidases (along with hepatic first-pass clearance after absorption) reduce systemic exposure to a fraction of a percent.
  • Plasma half-lives are short — native GLP-1 has a half-life of about 2 minutes due to rapid DPP-4 cleavage and renal clearance. Native growth-hormone releasing hormone, similar.
  • Most are subcutaneously injected, with bioavailability typically 60–90%. Intranasal works for some small peptides (oxytocin, desmopressin). Pulmonary inhalation has been tried for insulin with mixed commercial success.

Engineering tricks

Modern peptide drug design exists largely to defeat the PK problems above. Common modifications:

  • D-amino acid substitution — flipping the stereochemistry at one or more residues makes the bond invisible to most proteases. The trade-off is reduced receptor affinity if the substitution lands at a binding-critical position.
  • N-methylation of backbone amides — blocks hydrogen bonding that proteases need to recognize the substrate. Also improves membrane permeability somewhat (a step toward oral viability).
  • Cyclization — head-to-tail, side-chain-to-tail, or disulfide-bridged cycles eliminate the free termini that aminopeptidases and carboxypeptidases attack first.
  • Fatty-acid acylation — attaching a C16 or C18 fatty-acid chain makes the peptide bind serum albumin reversibly, extending half-life from minutes to days. This is how semaglutide gets its weekly dosing schedule (t½ ≈ 7 days) from a GLP-1 analog whose native form clears in minutes.
  • PEGylation — covalent attachment of polyethylene glycol chains increases molecular size and slows renal filtration.
  • Fc fusion — fusing the peptide to an antibody Fc region extends half-life via FcRn-mediated recycling (dulaglutide uses this approach).
  • Multi-receptor agonism — engineering a single peptide to bind two or three receptors at different sites. Tirzepatide (GLP-1 + GIP) and retatrutide (GLP-1 + GIP + glucagon) are the visible examples; the design approach is the same.

Why peptides are having a moment

Peptides sit in a sweet spot between two more established drug classes. Small molecules have excellent pharmacokinetics and cheap manufacturing but struggle to selectively engage large, complex receptor surfaces (the GLP-1 receptor binding pocket, for example, is fundamentally a peptide-shaped slot — small molecules can fit there only with considerable engineering compromise). Monoclonal antibodies give exquisite target selectivity but are large, expensive, must be refrigerated, and don’t cross most tissue barriers.

Peptides bridge the gap: synthetic enough that production is tractable at scale, structured enough to engage protein-protein interfaces antibodies handle well, and (with the engineering tricks above) durable enough to dose weekly. The commercial success of the GLP-1 receptor agonist class has pulled massive industry investment into the modality; expect the catalog of approved and investigational peptide drugs to keep growing through this decade.

See also