Articles
60 reference articles, each cited to primary literature.
mechanisms
Selank: Proposed Anxiolytic Mechanism and the State of the Evidence
Selank is a tuftsin analogue registered as an anxiolytic in Russia and nowhere else. Its proposed mechanism is indirect modulation of GABAergic signalling, and the literature supporting it is real, substantial and genuinely difficult to appraise from outside.
MOTS-c: A Peptide Written by the Mitochondrial Genome, and the AMPK Mechanism Proposed For It
MOTS-c is a 16-amino-acid peptide encoded inside the mitochondrial genome rather than the nuclear one. That single fact makes it a different kind of molecule from almost every other peptide discussed in this field, and it shapes the mechanism proposed for it.
Semaglutide: The GLP-1 Receptor, and Three Modifications That Made a Weekly Drug
Native GLP-1 survives about two minutes in circulation. Semaglutide survives about a week. The engineering that produced that difference is the clearest worked example of peptide drug design in the literature.
GHK-Cu: How a Three-Residue Peptide Carries Copper, and What That Explains
Glycyl-histidyl-lysine binds copper(II) with unusually high affinity. Most of the compound's reported activity is attributed to that single property — and most of its literature comes from one research group.
TB-500 and Thymosin β4: The Mechanism, and Why the Two Are Not Interchangeable
Thymosin β4 is a 43-residue actin-sequestering peptide with a well-characterised cellular function. TB-500 is a shorter synthetic construct sold under its name. Conflating them distorts most of what is written about both.
BPC-157: What the Evidence Says About Its Mechanism of Action
A pentadecapeptide derived from a gastric protein, studied largely in rodent models. Here is what is actually established about how it behaves, and where the evidence stops.
receptor pharmacology fundamentals
Orthosteric or Allosteric: Where a Peptide Ligand Binds and Why It Changes the Pharmacology
A ligand occupying the site the natural hormone uses competes with it. A ligand binding somewhere else modulates it. The two produce different ceilings, different dependence on the endogenous signal, and different failure modes.
Biased Agonism: G-Protein Versus β-Arrestin Signalling at Peptide Receptors
Two agonists can occupy the same receptor and produce different mixtures of downstream signal. That observation dismantled the idea of a receptor as an on-off switch, and it is the reason two compounds with identical targets can behave differently over time.
Concentration–Response: EC50, Emax, and Why Potency and Efficacy Are Not the Same Number
Two compounds can be compared on where their curve sits and on how high it reaches. Those are independent properties, they answer different questions, and treating one as a proxy for the other is the most common misreading in this literature.
Desensitisation, Internalisation and Downregulation: Three Things That Get Called "Tolerance"
A falling response to a repeated stimulus has at least four possible causes operating on different timescales. They are routinely reported under one word, and the word chosen usually reveals nothing about which process was measured.
Selectivity: How a Peptide Chooses Its Receptor, and What Happens When It Does Not
Selectivity is a ratio between numbers measured at two receptors, and it survives only within the range of concentrations where that ratio holds. Most claims of selectivity in this field are missing both the comparison and the range.
trial readouts 2026
Bimagrumab and the Lean-Mass Question: BELIEVE, and the Trial Lilly Stopped
Bimagrumab is an antibody, not a peptide. It belongs in this cluster because of the question incretin therapy raised: how much of the mass lost is not fat, and whether that proportion can be changed.
CagriSema and REDEFINE: What Adding Amylin Agonism Bought Over Semaglutide Alone
The phase 3 programme answered the question the combination was built to test, and then a harder one: how it compares with tirzepatide. Both answers are smaller than the phase 2 data implied.
Orforglipron: The First Non-Peptide GLP-1 Agonist, and Why It Is Not a Peptide Story
A small organic molecule reaches the same receptor as the incretin peptides by a different route, needs no absorption enhancer, and carries no fasting conditions. Its approval is a pharmacology story more than a peptide one.
Retatrutide TRIUMPH-1: What a 2,339-Person Phase 3 Actually Established
An 80-week phase 3 trial, read with both estimands, the discontinuation table, and the most quoted figure put back in its place. What TRIUMPH-1 settled about body-weight change, and what it left open.
EVOKE and EVOKE+: Why Semaglutide Did Not Slow Early Alzheimer's, and What the Biomarkers Showed
Two phase 3 trials enrolling 3,808 people found no difference from placebo on the clinical endpoint at two years. Small biomarker shifts in spinal fluid are the part of the result most likely to be misread.
compounds not yet covered
Dihexa: The HGF/c-Met Claim and the Distance Between a Rodent Model and a Person
Dihexa is a modified angiotensin IV fragment claimed to potentiate hepatocyte growth factor signalling in the brain. The paper that made that claim was retracted in 2025, there are no human data, and the pathway it targets is a proto-oncogene.
DSIP: A 1977 Discovery With a Thin and Contradictory Evidence Record
Delta sleep-inducing peptide was isolated from rabbit blood almost fifty years ago. No gene, precursor or receptor has been found for it, and its small human trials disagree with each other.
PEG-MGF: What Mechano Growth Factor Is, and What PEGylating a Fragment Changes
Mechano growth factor names a splice variant of the IGF-1 gene. The peptide sold as PEG-MGF is a synthetic fragment of it with a polymer attached. Whether the fragment exists naturally is disputed, a pharmaceutical group could not reproduce its core effect, and the PEGylated form has no published studies of its own.
Semax: An ACTH(4–7) Fragment, BDNF, and What the Russian Literature Does and Does Not Show
Semax is a hormone fragment with the hormone's endocrine activity engineered out. Its proposed mechanism runs through brain-derived neurotrophic factor, the evidence for that is rodent work from one programme, and the human data are stroke studies that cannot be appraised to current standards from outside.
VIP: A 28-Residue Neuropeptide With Receptors Almost Everywhere
Vasoactive intestinal peptide was isolated from pig intestine in 1970 and turned out to act on blood vessels, glands, immune cells and the brain's master clock. That reach is why it has been so hard to make into a drug.
peptide pharmacokinetics
DPP-4 and Neprilysin: The Enzymes That Switch Peptide Hormones Off
One enzyme removes exactly two residues from the amino terminus and destroys the signal. The other cuts almost anything hydrophobic, almost anywhere. Together they set the lifetime of most peptide hormones, and both have been turned into drug targets.
Immunogenicity: When the Immune System Notices a Peptide
A short synthetic peptide is a poor immunogen on paper. In practice anti-drug antibodies to peptide products are measured routinely, and the usual culprits are not the sequence itself but aggregation and the impurities that come with making it.
Plasma Protein Binding: Why Only the Free Fraction Acts
A peptide bound to albumin cannot engage a receptor or cross a filter. That single fact explains half-life extension by acylation — and also explains why the most common inference drawn from a protein-binding number is the wrong one.
Renal Filtration: The Size Threshold That Clears Most Peptides in Minutes
The kidney is a size filter that offers a typical peptide no resistance at all, and then a second system that destroys almost everything the filter lets through. This is the renal half of clearance, examined on its own.
Subcutaneous Absorption: Why Molecular Size Decides Whether a Peptide Enters Blood or Lymph
A molecule leaving a subcutaneous depot has two exits, and its size decides which one it takes. Small peptides cross the capillary wall into blood within minutes; anything much above sixteen kilodaltons travels by lymph and arrives hours later.
incretin pipeline pharmacology
Amycretin: One Molecule, Two Receptor Families
Every unimolecular co-agonist before it combined receptors from the same structural family. Amycretin carries GLP-1 and amylin activity on a single chain, and those two receptors are not relatives. That is the design problem, and the source of everything the approach gains and gives up.
Amylin Receptor Agonists: Cagrilintide and a Second Satiety Pathway
Amylin is not an incretin. It is co-secreted with insulin, it enters the brain through a different door, and its receptor is a calcitonin receptor wearing an accessory protein. That separation from GLP-1 is the entire argument for the class.
GIP Agonism or GIP Antagonism: The Question the Field Has Not Settled
Two development programmes are pushing the same receptor in opposite directions, and both are producing weight loss in humans. That is not a paradox anyone has resolved. It is the largest open question in incretin pharmacology.
Oral GLP-1: Absorption Enhancers and the Bioavailability Problem
A 31-residue peptide swallowed into a protein-digestion system should not reach the circulation at all. One does, at roughly one percent, and only because an excipient rewrites the local chemistry of a few square millimetres of stomach lining for a few minutes.
Survodutide: Glucagon Receptor Agonism and the Liver Endpoint
Survodutide engages the glucagon receptor and the GLP-1 receptor, and nothing else. Stripping out the GIP arm makes it the cleanest available test of what glucagon agonism contributes on its own — and the liver is where that contribution should show first.
structure and modification
Cyclisation: Locking a Peptide Into One Shape
A linear peptide in solution is a population of thousands of conformations, only one of which binds. Closing the chain into a ring removes most of the rest before the receptor is ever encountered.
D-Amino Acid Substitution and Protease Resistance
Proteases are stereospecific enzymes. Invert one stereocentre at the bond they cut and the cut stops happening. The cost is that the receptor is stereospecific too, and it does not always forgive the same change.
Disulfide Bonds and Why Some Peptides Have a Shape at All
Most short peptides have no fixed structure. The ones that do usually owe it to a covalent bridge between two cysteines, and that bridge is both the reason they work and the first thing to fail.
Lipidation and Albumin Binding: How a Fatty Acid Buys a Week of Half-Life
A single fatty acid chain, hung off one lysine through a short spacer, turns a peptide that survives two minutes into one that survives a week. It works not by protecting the peptide but by tethering it to albumin.
PEGylation: What a Polymer Chain Buys and What It Costs
Attaching polyethylene glycol was the dominant half-life extension strategy for three decades. It works, the mechanism is well understood, and the field has quietly moved away from it. Both halves of that sentence deserve explaining.
immune and mitochondrial peptides
Humanin and the Mitochondrial-Derived Peptide Family
Humanin was found by accident in a screen of brain tissue from an Alzheimer's patient, and it turned out to be encoded inside the mitochondrial genome. It opened a family of short peptides read from mitochondrial DNA, and almost every quantitative claim about them rests on assays that are difficult to trust.
KPV: A Tripeptide From the C-Terminus of Alpha-MSH
KPV is the last three residues of alpha-melanocyte-stimulating hormone. It keeps most of the parent hormone's anti-inflammatory activity and loses the pigmentary activity entirely, because the receptor-binding core sits elsewhere in the sequence. Nearly all of the evidence is preclinical.
LL-37: The Human Cathelicidin
LL-37 is the only cathelicidin humans make. It kills bacteria by physically disrupting their membranes, signals to immune cells through a formyl peptide receptor, and is transcriptionally controlled by vitamin D. It is also a case study in why antimicrobial peptides keep failing in clinical development.
SS-31 and Elamipretide: Binding Cardiolipin
SS-31 is a four-residue peptide that concentrates in the inner mitochondrial membrane without a transporter and binds cardiolipin, the lipid that gives cristae their shape. Its mechanism was substantially rewritten after its first decade, and its clinical trial record includes several clear failures.
Thymosin Alpha-1: Immunomodulation and Why It Is Not TB-500
Thymosin alpha-1 and thymosin beta-4 are unrelated peptides that share a name because they came off the same chromatography column in the 1970s. One is a licensed immunomodulator in about thirty countries. The other is the parent of TB-500.
gh axis pharmacology
Hexarelin and Receptor Desensitisation
Hexarelin is the most potent of the classical growth hormone releasing peptides, and it is also the one that made the limits of sustained receptor agonism visible. Its second story — a binding site in cardiac tissue — turned out to be the more interesting one.
GHRP-2 and GHRP-6: Two Hexapeptides, Two Different Profiles
The growth hormone releasing peptides came out of work on enkephalin analogues and led researchers to a receptor nobody knew existed. GHRP-6 and GHRP-2 share that receptor and differ in what else they do.
Sermorelin and the Shortest Fragment of GHRH That Still Works
Growth hormone releasing hormone is forty-four residues long. Only the first twenty-nine are needed for full activity. That truncation experiment produced sermorelin, and it tells you something general about how peptide hormones carry their information.
Tesamorelin: A Stabilised GHRH Analogue and the Visceral Fat Endpoint
Tesamorelin is a growth hormone releasing hormone analogue carrying a single chemical modification that protects it from enzymatic cleavage. It is the one compound in this family with a substantial randomised trial record — in a specific patient population, on a specific endpoint.
Ipamorelin: Selective Ghrelin Receptor Agonism and the Cortisol Question
Ipamorelin is a pentapeptide agonist at the growth hormone secretagogue receptor. Its defining claim is selectivity — releasing growth hormone without the cortisol and prolactin rise seen with earlier secretagogues. Here is where that claim comes from and how far the evidence carries it.
evidence base
HCG: A Real Hormone Attached to a Refuted Diet Claim
Human chorionic gonadotropin has a defined receptor, a well-described mechanism and approved clinical uses in fertility and male endocrinology. It also carries one of the most thoroughly disproved claims in the history of obesity treatment, and the first fact is why the second one lasted.
Growth Hormone in Healthy Older Adults: One Study, and What Came After It
A 1990 report on twelve men over 60 is still the foundation of growth hormone's anti-aging reputation. The randomised trials that followed found something considerably smaller, alongside a list of harms.
Melanotan II: The Documented Safety Signals
A non-selective melanocortin receptor agonist that was never developed as a medicine. What the case-report literature records, and why the absence of controlled safety data is itself the central finding.
Epitalon and Telomerase: An Assessment of the Evidence Base
The claims made for epitalon are specific, mechanistic and testable. The evidence behind them is small, concentrated in a single research programme, and largely unreplicated. That combination is the finding.
AOD9604: A Compound That Worked in Mice and Failed in People
The rodent data were strong, the mechanism was coherent, and the human trials did not replicate it. AOD9604 is the most instructive failure in the peptide literature, and it is rarely described as one.
pharmacology
Peptide Half-Life: Why Clearance Is So Fast, and the Five Ways It Is Slowed
Two independent systems remove peptides from circulation — peptidases cut them and the kidney filters them — and both work quickly. Extending half-life means defeating both, and every method of doing so has a price.
PT-141 (Bremelanotide): Melanocortin Pharmacology and the Development Story
A metabolite of Melanotan II that took the other road: randomised trials, a formulation abandoned over a blood pressure signal, and an approval. The mechanism is central, and the effect size is modest.
Retatrutide: Triple Glucagon, GIP and GLP-1 Receptor Agonism, and the Balance It Has to Strike
Retatrutide adds a glucagon receptor arm to the two incretin receptors tirzepatide already engages. Glucagon raises blood glucose, which makes that the most counterintuitive design decision in the class — and the reason the molecule acts on energy expenditure as well as intake.
CJC-1295, the Drug Affinity Complex, and the Half-Life That Defines It
CJC-1295 with DAC and "CJC-1295 without DAC" are not two versions of one compound. They are different molecules whose half-lives differ by orders of magnitude, and only one of them is what the published literature calls CJC-1295.
Tirzepatide: Dual GIP and GLP-1 Receptor Agonism, and Why the Balance Is Uneven
Tirzepatide engages two incretin receptors instead of one, and it does not engage them equally. The receptor-level asymmetry is the most interesting thing about the molecule, and the part of it that is least settled.
fundamentals
Reading a Peptide Sequence: Direction, Codes, Fragments and Modifications
A peptide sequence is a specification; a peptide name usually is not. The conventions that make a sequence readable — direction, numbering, two symbol sets, fragment ranges and modification notation — are few, fixed, and worth learning once.
L-Carnitine Is Not a Peptide: What It Does and What the Evidence Supports
Carnitine is a single small molecule of about 161 daltons that sits in peptide catalogues for commercial reasons rather than chemical ones. Its role in fatty acid transport is textbook biochemistry; the fat-loss claim built on top of that role is a separate question with a much weaker answer.
Why Most Peptides Are Not Orally Bioavailable
The gastrointestinal tract is a protein-digestion system, and a swallowed peptide is its intended substrate rather than an unlucky bystander. Five barriers stand between the mouth and the circulation, and unmodified peptides clear all five at well under 1%.
Peptide or Protein: Where the Boundary Sits, and Who Decides
Chemically the two are the same class of molecule. The line between them is drawn by convention, by folding behaviour, by how the molecule is manufactured — and, in the United States, at exactly 40 residues by statute.