Peptides & bioregulators

Chemistry first, because the vocabulary gets used loosely. Amino acids join when the acid group of one condenses with the amine group of the next; a molecule of water leaves and an amide bond forms. Inside a chain of amino acids that bond has its own name, the peptide bond, and a chain of them is a peptide. There is no sharp line where a peptide becomes a protein. The working convention is that up to roughly 50 residues counts as a peptide and beyond that it is a small protein, which leaves insulin at 51 residues sitting exactly on the fence and getting called both ^[3].

Sequence is identity. A peptide is written from the free amine end, the N terminus, to the free acid end, the C terminus, so a name like Ala-Glu-Asp-Gly specifies the molecule completely; that particular four-letter string is epithalon. This is why so many peptide entries on this site carry a string of three-letter codes as an alias. For a peptide, unlike a small molecule, the name and the structure are the same object.

Size has consequences that run through everything else. A typical peptide drug weighs between 500 and 5,000 g/mol, against 200 to 500 for a typical oral small molecule, and it carries many charged and hydrogen-bonding groups. That rules out passive diffusion across membranes, which is how most small molecules get into cells and into the brain. So peptides work almost entirely at cell-surface receptors, and a large share of them at G-protein-coupled receptors, which is exactly where the body's own hormone signals land ^[3].

The upside of working at the same receptor as a natural hormone is that the biology is usually already understood before the drug exists. The downside is that a receptor sitting on the outside of a cell is a much narrower target list than the enzymes, transporters and nuclear receptors a small molecule can reach. That is a real structural limit on the category, not a temporary one, and it is worth holding on to when a vendor page implies a peptide can do anything a drug can do.

Three separate systems dismantle a peptide, and an honest account has to name all three. The gut is the obvious one: the stomach's acid and pepsin, then pancreatic proteases, then brush-border enzymes on the intestinal wall, exist specifically to cut protein into fragments small enough to absorb. A peptide swallowed as a drug is treated as lunch. Even if a fragment survives, it still has to cross the intestinal wall, which a large polar molecule does very poorly, and then survive a first pass through the liver ^[1].

The blood is the second. Circulating peptidases clip peptides continuously, and some of them are startlingly quick. Native GLP-1 is cut two residues from its N terminus by the enzyme DPP-4 and is functionally gone in about two minutes. The kidney is the third and the one people forget: the glomerulus filters essentially anything below roughly 30 to 50 kDa, so a small unmodified peptide is not just degraded, it is also thrown away intact within minutes of arriving ^[3].

The whole engineering programme of modern peptide medicine is a set of answers to those three problems, and once you can name the tricks, a lot of otherwise mysterious structures become readable ^[4]. The table below is the short version.

Two of those entries deserve a note. Oral semaglutide works, but not the way people assume: the tablet carries an absorption enhancer called SNAC that raises the local pH against the stomach wall and helps a small fraction of the drug cross there rather than in the intestine. Bioavailability is on the order of one percent, which is why the oral tablet's strength has nothing in common with the injectable's ^[5]. And the last row is the honest one: several times, the winning move has been to abandon the peptide altogether and find a small molecule that fits the same receptor, which is exactly what MK-677 is at the ghrelin receptor.

Practically, most research peptides ship as a lyophilised powder in a sealed vial, because a peptide in water degrades and aggregates. That means the buyer reconstitutes it, and reconstitution is a sterile procedure being carried out outside a pharmacy, which is its own risk and is covered further down this page.

Grouping peptides by what they do is more useful than grouping them by chemistry, because the chemistry is nearly always the same and the evidence base is not. The table sets out the families this site covers, with the honest state of the record for each.

Read the last column first. This is a category where two neighbouring vials on the same vendor page can be separated by a factor of ten thousand in how much human evidence stands behind them, and nothing about the packaging tells you which is which. Semaglutide has been through hard-outcome trials with tens of thousands of participants. BPC-157 has, at the time of writing, no completed randomised controlled trial in humans at all; its entire case is rodent surgery models, largely from one research group ^[10][11].

A second thing the table makes visible: approval status does not track family. Tesamorelin is an approved prescription medicine and sermorelin once was, while their close relatives are sold as research chemicals. The line is drawn by who paid for the trials, not by how exotic the molecule is.

"Bioregulators" are a specific and much more speculative corner of the category: very short peptides, often only two to four amino acids, proposed to act as tissue-specific instructions that help one particular organ regulate itself. The work traces to a single long programme in St Petersburg, begun in the 1970s and associated above all with Vladimir Khavinson ^[2].

The sequence of ideas is worth knowing, because it explains why these compounds look the way they do. The programme began with extracts: crude peptide fractions pulled out of animal organs, such as Epithalamin from the pineal gland and Thymalin from the thymus. Analysing the amino-acid composition of those fractions, the group then designed synthetic short peptides intended to reproduce the activity of the extract they came from. That is where epithalon (Ala-Glu-Asp-Gly) and Thymagen (Glu-Trp) come from. The claim is not that these peptides bind a surface receptor; it is that a peptide this small can enter the nucleus and interact with DNA in a sequence-selective way, changing which genes a cell transcribes ^[7][8].

That mechanism is genuinely unusual, and it is the part to be most careful with. Sequence-selective DNA binding by a tetrapeptide is not a mainstream result; it is asserted mainly from within the same programme, using its own assays, and it has not been independently reproduced by unrelated groups with modern methods. The strongest honest statement available is that short peptides have been shown to change gene expression in various cell systems, which is a much weaker and much less specific claim than the one usually made for them ^[7].

The human record has the same shape. There is a real body of clinical work, some of it running for decades, reporting benefits in immune function, retinal disease and general markers of ageing ^[2]. Almost all of it comes from the group that developed the compounds, much of it is published in Russian, most of it is small and single-centre, and very little was registered in advance or replicated abroad. The longevity claims, which are what these peptides are usually sold on, rest on rodent and fruit-fly lifespan experiments from the same programme rather than on any human outcome.

None of that makes the idea false. It does mean that the correct posture is curiosity plus a firm hand on the evidence: the bioregulator literature is not fabricated, it is simply not independent, and independence is the specific thing that turns an interesting body of work into a settled one.

The single most useful skill in this category is telling the ends apart, because the packaging is identical and the evidence is not. Three worked examples make the range concrete.

The strong end. The trial that established semaglutide for weight loss randomised 1,961 adults for 68 weeks against placebo, with pre-registered endpoints, an independent committee and published adverse-event tables ^[13]. Whatever you think of the drug, the question of whether it does what it says has been answered in the way medicine answers questions.

The empty end. BPC-157 is one of the most discussed peptides on the internet. Its published record is a long series of rodent experiments showing accelerated healing of tendon, muscle, gut and nerve injuries, mostly from one group in Zagreb, with a proposed mechanism involving nitric oxide signalling and angiogenic growth factors ^[10][11]. What is missing is not a small piece; it is the entire human half. No completed randomised controlled trial has reported. This is the exact shape of a compound that might work and has not been tested.

The instructive middle. Cerebrolysin has been used clinically for decades and has enough randomised trials to support a Cochrane review. The review found no benefit on death or dependency after ischaemic stroke, and noted more serious adverse events in the treated group ^[12]. That is the case worth remembering, because it breaks the assumption that a peptide with a large human literature must be a peptide that works. Sometimes the trials get done and the answer is no.

One more habit worth building: being a copy of a natural signal proves nothing about benefit. Every hormone in the body is already present at the level the body chose. Raising it is an intervention with a direction and a cost, not a return to some default. The fact that a molecule is "natural" or "endogenous" is a statement about where the idea came from, not about whether more of it helps.

Most research peptides are sold labelled "for research use only", which is a legal position rather than a quality claim. In practice it means that identity, purity and sterility are entirely the buyer's problem, and the buyer is generally not equipped to check any of the three. A vial can contain the right peptide at the wrong amount, a related fragment, a truncated synthesis product, or bacterial endotoxin from the manufacturing water, and none of those is visible in a white powder.

Injection adds its own layer. Reconstitution with sterile water, storage, drawing and injecting are all sterile technique performed outside a clinical setting, and abscess and injection-site infection are the routine complications of getting that wrong. This is a mechanical risk, not a pharmacological one, and it is unrelated to whether the compound itself does anything.

There is also a risk class here that small molecules essentially do not have: immunogenicity. A peptide is protein-like, and the immune system can raise antibodies against it. Those antibodies can neutralise the drug, and in the worst case they can cross-react with the body's own version of the hormone ^[3]. This is a known, managed issue in approved peptide medicine, with assays and monitoring behind it; in the research-chemical market nobody is looking for it.

On regulation, peptides sit in an awkward gap. They are not dietary-supplement ingredients in the United States, and most of the compounds discussed here are not approved medicines anywhere, which is why they are sold under a research label rather than as either. Several are also explicitly prohibited in tested sport, including the growth-hormone secretagogues as a class. See research chemicals for how that legal grey zone works generally.

The reasonable summary is the one the evidence supports: treat the well-studied pharmaceuticals in this category as the serious drugs they are, with real benefits and real adverse-event profiles, and treat the rest as genuinely experimental, including the ones with the most confident marketing. None of this is medical advice.