The episode opens on what Dr. Abud Bakri calls the celebrity protocol. CEOs and celebrities stack a GLP-1 drug for insulin sensitivity, growth hormone or a growth-hormone secretagogue for muscle, and androgens on top, then post their transformations. They lose a lot of fat and gain a lot of muscle fast. Bakri’s verdict on whether that’s a good idea: “Is that healthy? We’ll find out.”
That line is the honest subtitle for the whole conversation. Bakri is a board-certified internal-medicine physician who works hospital wards (admissions, the ER, the ICU) and also works with peptides in clinic. Huberman says on air that he’s 33. Across nearly three hours they cover BPC-157, the compounding and gray markets, GLP-1 drugs and retatrutide, the Soviet bioregulators, a long stretch on the thymus, growth-hormone secretagogues and GHK-Cu. The episode is “Peptides: The Science, Uses & Safety,” released June 1, 2026.
How we wrote this: Huberman’s official transcript is still marked as under review, so we worked from the episode’s captions. We describe what was said in our own words and give a timestamp for every point so you can hear it yourself. Doses are reported exactly as they were stated on the show, with who said them. They’re what two people said on a podcast, not protocols. Where a claim deserved checking, we checked it against published papers and regulatory records, and those checks are labeled.
This is a full rewrite of the recap we published in June. Going back through the episode, we found several things the first version got wrong, including two misattributed stories. There’s a list of corrections at the end.
The short version
- The organizing idea is receptors. Peptides with a known receptor, like the GLP-1 drugs, have strong, predictable effects. Peptides with no identified receptor, like BPC-157 and TB-500, are where the evidence gets thin (3:33).
- BPC-157’s human evidence is two small enema trials from the early 2000s, published only as abstracts. Nearly all the animal data come from one lab in Zagreb (20:46).
- The dose people inject online has no basis in data. Bakri says it comes from what fits in a vendor’s vial, and that he uses far more himself (54:13).
- The gray market is huge, the raw material all comes from China, and you can’t tell which batch you’ve got (35:25).
- Bakri’s GLP-1 approach is training wheels: the lowest dose that works, alongside diet and exercise. His worst night as a doctor came from testing an Ozempic pen at four times the starting dose (2:32:32).
- Huberman uses pinealon to double his REM sleep. Bakri points out it was designed as a daytime compound and nobody ever measured its effect on REM (1:07:25).
- About 35 minutes go to the thymus, including a $3 blood ratio Bakri says almost no doctors look at (1:55:32).
- A growth-hormone secretagogue wiped out Huberman’s REM sleep and spiked his PSA. He stopped and it came back down (2:10:56).
Jump to: The receptor question · BPC-157 · BPC-157’s legal status · Compounding and the gray market · GLP-1s and retatrutide · Pinealon · Epitalon and the pineal · The thymus · Thymosin alpha-1 · TB-500 · GHK-Cu · Growth hormone and secretagogues · Where to get peptides · My take · Corrections
The receptor question that organizes everything
Huberman asks how a clinician thinks about the word “peptide” at all (3:33). Bakri describes peptides as one of several systems cells use to signal each other, alongside steroid hormones and others. The split he finds useful is between peptides with a known receptor and peptides without one.
The GLP-1 drugs sit firmly in the first group. They bind a well-understood receptor, and their effects in people are large and reproducible. BPC-157 and TB-500 sit in the second. Nobody has identified a receptor for either (4:52). Bakri offers three possibilities for BPC-157: the receptor simply hasn’t been found, the peptide modifies proteins that already exist, or it links proteins together in ways that favor gene transcription. He describes the Russian bioregulators, like pinealon and epitalon, differently again, as compounds thought to bind DNA directly and change which genes are read.
Huberman says he’s never heard peptides broken down this way. It’s a useful lens to carry through the rest of the episode, because the compounds with known receptors are also the ones with real human trial data.
Near the end (2:36:30), Bakri goes further and argues that “peptide” isn’t a useful scientific category at all. Orforglipron, an oral GLP-1 drug that isn’t a peptide, behaves much more like semaglutide, which is one, than semaglutide behaves like BPC-157. Carnosine and retatrutide share the label and almost nothing else. Huberman jokingly nominates him for a nomenclature committee.
BPC-157: interesting biology, thin evidence, and a dosing problem
BPC-157 gets close to an hour, spread across the first half of the episode, and it’s the richest material in the conversation.
What it is, and where it came from
Bakri opens with a correction to something you’ll read on many websites (7:12). Your body doesn’t make BPC-157. It makes a much larger gastric-juice protein, which he puts at around 40,000 daltons, and BPC-157 is a synthetic 15-amino-acid fragment of it.
The origin story runs through Pavlov, who sold dog gastric juice as a remedy (Huberman jokes about a Pavlov discount code), and Hans Selye’s stress experiments, where stressed animals showed enlarged adrenal glands, damaged stomach linings and shrunken thymus glands (6:26). The Croatian group behind BPC-157 reasoned that the gut must hold some protective, anti-stress compound, and went looking for it. The original goal was a gut drug. The tendon and ligament interest came later, set off largely by an Achilles tendon paper that bodybuilders noticed (11:19).
What the animal studies show
The rodent work covers cut tendons, ACL injuries and burns, with BPC given orally or by injection into the abdomen (11:19). Huberman, who spent years in a lab trying to regenerate optic nerves, says he finds a compound that speeds tendon, ligament and nerve repair in animals genuinely strange (15:10).
The proposed mechanisms are increased VEGF signaling and new blood-vessel growth in damaged tissue, more cell migration (especially combined with TB-500), changes in nitric oxide, and more growth-hormone receptors on tendon cells (15:58). Bakri frames BPC as something that speeds up repair, pulling in immune cells and repair factors, rather than as an anti-inflammatory. One detail he highlights: corticosteroids normally stall wound repair in animals, but when BPC was given alongside them, repair was as good or better.
Fact check: the growth-hormone receptor point has independent support. Chang and colleagues at Chang Gung University in Taiwan (Molecules, 2014) found BPC-157 increased growth-hormone receptor expression in rat tendon cells. It’s one of the few BPC-157 findings from outside the Zagreb group.
The brain side, which Bakri finds more interesting
Bakri says the neurological and psychiatric findings interest him more than the joint claims (17:34). In one study, drunk mice given BPC were less impaired in mazes and skipped alcohol withdrawal, which can kill hospital patients who don’t get the right medication. Huberman jumps in quickly to say nobody is endorsing BPC with alcohol (18:21). Later, Bakri adds that mice given BPC alongside methamphetamine couldn’t get very high and didn’t go through withdrawal (2:40:31).
The human side of this is less comfortable. Bakri says patients tell him their Adderall stopped working once they started oral BPC, and they’re annoyed about it (18:21). He also mentions Reddit reports of flat mood, low energy and a loss of pleasure in things (19:08). He reads both as signs that BPC affects dopamine signaling, in either direction.
His most speculative idea, which he admits sounds fringe, is that BPC works as a homeostatic signal (2:41:19). It may act locally in the gut, on the gut-brain axis, and push the body toward a rest-and-digest, convalescent state while it repairs. He doubts oral BPC reaches the bloodstream at all.
The human evidence, all of it
Huberman frames the human evidence as lots of animal data and very little else. Bakri gets specific (20:46). There were two small trials of BPC given as an enema for ulcerative colitis, a Phase 1 and a Phase 2, in the early 2000s, run by the same Croatian group. Doses went up to about 80 mg. Phase 1 found no adverse effects. Phase 2, with roughly 40 patients, showed at least a positive signal. But only abstracts were ever published, and the full data were never released, which makes him cautious (21:33).
In those trials, BPC didn’t show up in the blood after rectal dosing, and Bakri says the same goes for oral dosing. Either it breaks down fast or it stays in the gut lining. Injected BPC does reach the bloodstream: a 2024 paper detected BPC fragments for anti-doping purposes, and it’s now on the WADA prohibited list.
On toxicity, animals have been given roughly a thousand times the usual dose without real adverse effects, and nobody knows the lethal dose (19:59). Bakri says that gap is part of what blocks a path to approval.
Fact check: Bakri is right, and Huberman’s framing undercounts. The Phase 1 study in healthy men was published as an abstract (Veljača et al., Gut, 2003), and the Phase 2 multicentre trial in ulcerative colitis appeared only as a conference abstract (Ruenzi et al., Gastroenterology, 2005), never as a full paper. Since then there have been three small US pilot reports from Lee and colleagues: a retrospective look at 16 patients given knee injections (2021), 12 women given bladder instillations (2024), and an intravenous safety test of up to 20 mg in two adults (2025). A 2025 systematic review in orthopaedics (Vasireddi et al., HSS Journal) found 36 eligible studies, of which 35 were preclinical. There is still no published, controlled human efficacy trial.
The one-lab problem
Bakri raises this first (20:46) and Huberman returns to it later (43:02): nearly all of the animal literature comes from Predrag Sikiric’s group at the University of Zagreb, with a couple of groups in Asia repeating parts of it. Huberman asks, carefully, whether a Croatian trial should count for less. Bakri defends the group as running proper double-blind trials, says Americans tend to be US-centric about research, and then concedes that Soviet-era research culture had incentives to fabricate.
Fact check: true, and it’s the caveat for every BPC-157 claim. The 2025 orthopaedic review notes how heavily the field depends on that single group. Independent replication is how science catches flukes, bias and wishful thinking, and BPC-157 has very little of it.
The dosing chasm
This is the most useful stretch of the BPC discussion, and it’s where our first version got the attribution wrong.
The question about typical use comes from the host side (20:46): most people inject micrograms, about 100 to 200 micrograms a day, possibly more. Bakri agrees that’s the norm. Set that next to the only human trials, which used up to about 80 milligrams by enema, and the gap is several hundredfold, by a different route, for a different condition.
Then Bakri tells his own story (54:13). He tore his triceps, a grade 2 tear (he could still straighten his elbow), with his arm purple from top to bottom, after lifting with people stronger than him. He injected BPC near the injury along with other peptides he doesn’t name, at a dose he says was far higher than the usual microgram amounts. Within three weeks his physical therapist was astonished at the pace, and he was back in three to four weeks instead of the usual three months or so. He’s clear that he can’t know whether he would have recovered that fast anyway, and that he’s seen injuries where BPC didn’t do much (55:47).
His bigger claim is that low doses may be why so many people notice nothing. The standard internet dose, he says, comes from what fits conveniently in a vendor’s vial, not from any data. When an East Coast orthopaedic group he consults for told him they planned to start a BPC trial at 250 micrograms, he told them he doubts they’ll see an effect that low and that they might need to go higher (55:47).
Huberman’s own story points the other way (55:47). He pulled a muscle badly in his upper trapezius, injected about 200 micrograms into it, and two days later the problem was gone, so he stopped. His BPC came by prescription from a compounding pharmacy, labeled BPC-157. He says he has taken it but doesn’t now, because he has no injury and doesn’t want the blood-vessel risk.
Two smaller details round it out. Bakri used BPC for one injury and felt the improvement in another, which is why he thinks trials that compare one limb against the other don’t work (53:25). And he carries oral BPC when he travels, and says his friends get traveller’s diarrhoea while he doesn’t, which he calls an anecdote (50:14).
My take: this exchange is the honest center of the BPC debate. A physician who uses it says the common dose is probably too low, the host who used it says the common dose worked in two days, and the only controlled human data used a dose hundreds of times higher by a different route. Nobody knows the effective human dose. Anyone selling you a protocol is guessing, and both men on this podcast say so in their own way.
Fact check: a real US trial is under way. Bakri mentions a Phase 2 hamstring trial (43:02). It’s listed on ClinicalTrials.gov as NCT07437547 (“BPC-HAMSTR”), sponsored by Hudson Biotech: randomized, double-blind and placebo-controlled, in 120 adults with grade II hamstring strains, with subcutaneous BPC once daily for 14 days. It has been recruiting since February 2026. A rotator-cuff trial (NCT07803250) is also listed.
The blood-vessel question
Huberman asks directly whether a compound that grows blood vessels could feed a tumour someone doesn’t know they have (41:32). Bakri’s answer has two halves. The animal literature shows no tumour signal, and he contrasts BPC with Cardarine (GW501516), a compound that did show a tumour signal in animals and was dropped (42:13). But both men point out that nearly all that reassuring literature comes from the same single lab (43:02).
The anecdotes on the other side: Reddit reports of bruises and blood pooling getting worse, which Bakri says fits the VEGF pathway, and a physician friend of Huberman’s who notices the small spider-like blood vessels on his face get worse whenever he runs BPC for a joint (43:50). Pulling the other way, BPC lowered VEGF in a melanoma cell line (44:37), which Huberman says might even point to an anti-tumour effect, though it would need testing.
Bakri’s framing later in the episode is precise (2:06:57). There’s no mechanism by which BPC damages DNA, so it’s nothing like smoking. The open question is whether a growth factor could speed up a tumour that already exists, and there’s no good data either way.
Fact check: the melanoma finding is real, with the usual caveat. Radeljak, Seiwerth and Sikiric (Melanoma Research, 2004) reported that BPC-157 inhibited growth and VEGF signaling in a human melanoma cell line. It’s a cell-line study from the same Zagreb group. The tumour question is a reasonable caution, not a demonstrated danger, and it hasn’t been tested in people. Our BPC-157 vs TB-500 breakdown covers the two compounds’ evidence side by side.
Patents, trials and why nobody pays for the big study
There are several BPC patents, split by salt form (47:01). One ended up with Teva when it acquired the original holder, Pliva, and Bakri says Teva makes its money elsewhere and has no interest. Another, held by Sikiric, has about ten years left. Peptide patents are easy to get around, he says: change one amino acid and you have a new compound (47:50). With millions of people already using BPC from research websites, there’s no commercial reason to fund a large trial unless people crowdfund one.
His wishlist (49:26) is a Phase 3 trial in ulcerative colitis using a delayed-release oral capsule, oral BPC against pantoprazole for reflux, a study on addiction and cravings (possibly building on Diego Bohorquez’s gut-neuron work at Duke, which Huberman offers to help fund), and a post-surgical tendon study in a poorly supplied tissue like a biceps or triceps tendon. He’s clear BPC won’t reattach a torn ACL. The question is whether it could cut a 12-month return after surgery to six. He’s also building a registry to pool people’s reports.
The Olympic rumour
Huberman brings up a story that circulates in peptide circles: an Eastern European athlete who fully ruptured an Achilles tendon, injected BPC locally for a few months, and still won a medal (27:57). Bakri calls it unverified hearsay. Any serious athlete in that position would have used every legal tool available, from stem cells to other peptides, not BPC alone, and research-chemical sellers turned the story into marketing.
Fact check: BPC-157 was added by name to WADA’s prohibited list in 2022, so it wasn’t banned at the Tokyo Games, which fits Bakri’s timeline. The medal story itself remains unsourced.
Two bad endpoints, and Huberman’s third
Bakri sums up the stakes (57:27). Either BPC is as good as its fans say, and millions of people are being denied access to something that would have spared them joint replacements and shortened careers, or it’s useless or harmful, and millions are injecting it right now. Either way, he says, people will eventually be angry.
Huberman adds a third outcome, which he thinks is the most likely (59:49). Adverse events start turning up, nobody can tell whether BPC, misuse or a contaminated diluent caused them, and BPC drifts back into the shadows without ever being properly tested. He says his bigger fear isn’t the molecules themselves but how they’re made and delivered (1:00:37), and he expects pushback for saying so.
Bakri’s message for teenagers in the gym is short (58:14). An 18-year-old’s own physiology already supplies what he’d be injecting. He adds that many young men on testosterone don’t need it and are risking their fertility.
BPC-157’s legal status: renamed, delisted, and voted on
The regulatory story moved fast, and some of it happened after the episode was recorded.
The FDA keeps lists of bulk drug substances for compounding pharmacies. Category 1 can be compounded while under review. Category 2 is effectively a do-not-compound list. Bakri says BPC and about 20 other peptides moved to Category 2 in late 2024, after practices had prescribed BPC from roughly 2017 onward (24:45). The compounding industry’s response was a relabel (25:33). Pharmacies began selling the same 15-amino-acid sequence as pentadeca arginate, or PDA. Bakri confirms it’s the identical molecule as an arginate salt rather than an acetate, and he believes the acetate form is the one on the list. He then says BPC came off Category 2 in April 2026 without reaching Category 1, so doctors still prescribe PDA rather than BPC itself (26:20).
Fact check: the date is off by a year, and there’s news since. The FDA added BPC-157 to Category 2 in September 2023, not 2024. It removed BPC-157 and 11 other peptides from Category 2 on April 15, 2026, as Bakri says. After the recording, on July 23 and 24, 2026, the FDA’s Pharmacy Compounding Advisory Committee voted 8 to 6 to recommend adding BPC-157 to the list of substances compounders can use. Rulemaking is still pending. We covered that vote in BPC-157 and the FDA compounding vote, and the wider picture in are peptides legal in 2026.
The rest of the legal discussion is about the physician’s position:
- State boards (26:20). One state where Bakri is licensed wrote to tell him he can’t prescribe non-approved peptides to that state’s residents under any circumstances. Rules vary by state, and for telehealth, the law where the patient lives applies.
- The hospital bind (1:01:26). If he told a hospital patient to take BPC instead of pantoprazole, he’d expect to lose his licence. In clinic, once the rules allow, he wants a frank, consented, monitored conversation with clear goals.
- Malpractice (1:06:14). Mainstream insurers won’t cover non-approved peptides. If something goes wrong, patients sue the doctor, the pharmacy and anyone who recommended it.
- Supplements (1:02:14). Oral BPC is already sold as a supplement. Bakri mentions that RFK Jr. has called these supplements rather than medications, which he doubts the agency welcomed (1:42:50).
- What’s next (2:16:31). Bakri expects around seven peptides, BPC and TB-500 among them, in the first round of Category 1 review, with GHK-Cu in a later round.
Fact check: the hosts mention FDA commissioner Marty Makary’s recent departure. He resigned on May 12, 2026.
Compounding, the gray market and the story that should scare you
This section is the consumer-protection core of the episode, and our first version barely touched it.
Compounded GLP-1s and the markup
During the GLP-1 shortages, compounding pharmacies were allowed to make semaglutide and tirzepatide, and Bakri says the FDA actively encouraged it (30:20). He recounts former compounding pharmacist Brigham Buhler telling him the agency asked pharmacies to make it and then got them in trouble for it. He compares prices: about $150 for a pen in Mexico against about $1,500 in the US (29:41). Anecdotally, he says, people respond better to brand-name pens than to compounded versions. Compounders also add vitamin B12 or B6 to their GLP-1 vials so the product qualifies under the special-patient-need rules.
Then the markup (31:30). A pharmacy might charge a clinician about $150 for a vial, and the clinician can turn around and charge $200 or $800, calling the difference an administrative fee. Bakri doesn’t think that’s legal in most states. Huberman’s advice to listeners is to ask their clinician what they pay the pharmacy and what they’re charging (32:42). He tells clinicians the secret is out.
Bakri predicts the next step: telehealth checkouts selling BPC for $199 or $299 with a rubber-stamp prescriber attached, which he calls e-commerce rather than medicine (32:42).
How the gray market actually works
Huberman makes a personal admission (33:30). He once bought from Peptide Sciences, paying by Venmo to a string of random, changing names. The vials arrived, and he was too scared to use them.
Bakri’s market estimates (35:25): a US gray market of five to ten billion dollars in 2025, with essentially every peptide ingredient, whether for pharma, compounders or research websites, made in China and only finished elsewhere (35:04). Vials that cost around five dollars to make. Affiliate marketing that he calls the most lucrative gig of 2025, with some TikTok sellers earning his annual salary in a month (39:02). Huberman says more than half the market is women, and Bakri compares the wave of soccer-mom affiliates to the multilevel-marketing boom of twenty years ago (39:48). Below the gray market there’s a black market of WhatsApp sellers and bathtub synthesis.
Huberman describes the supply chain as tiers: pharma, compounders and the gray market all drawing on the same Chinese raw material at different quality levels (37:28). He mocks the research-only label, pointing out that nobody buying these vials is running cell culture at home. Batches vary from one to the next, and from the outside you can’t tell which you have.
The melanotan story, told correctly
Here’s the anecdote everyone remembers, and our first version got it wrong. Bakri recalls a man who went viral on Twitter a few weeks before the recording (36:41). He’d been injecting gray-market “retatrutide,” started noticing his skin getting darker, and worked out that the vial actually held melanotan, a tanning peptide. Bakri tells it as a public case. He never says the man was his patient.
The point stands either way. When the label can’t be trusted, you can’t reason about a dose, because you don’t know what’s in the vial, how much of it there is, or what else came along. Huberman credits GLP-1 drugs with making people comfortable injecting themselves, and says BPC rode in on that wave (40:38). Comfort with needles has outrun knowledge of what’s in them.
This is why every compound we carry is third-party tested by Janoshik, with the results posted on our lab testing page. If you want to understand what those reports measure and why identity testing matters as much as purity, we explained it in what is Janoshik testing.
GLP-1s and retatrutide: a practitioner’s view
The GLP-1 discussion comes late (2:20:25), and it’s the most clinical part of the episode, because this is the drug class Bakri has prescribed the longest.
Why Bakri became a doctor
Exenatide, the first GLP-1 drug, came from Gila monster saliva (2:24:32). Native GLP-1 lasts too briefly to work as a drug, and Bakri credits the pharmaceutical industry for engineering longer-acting versions. His grandmother started exenatide (Byetta) about 25 years ago, and of all her medications it was the one that changed her course: less insulin, weight loss, more energy. He says it’s the reason he went into medicine.
He’s prescribed GLP-1s since residency. In his VA clinic, veterans weighing around 300 pounds lost 50 to 100 pounds, before the drugs were even approved for weight (2:22:06). Bodybuilders, he says, pioneered off-label use in the late 2010s. His broad claim is that the US health system couldn’t have absorbed the wave of excess weight and diabetic disease without these drugs. Older weight drugs produced 5 to 10% weight loss, GLP-1s produce 10 to 20%, and retatrutide’s newer data reach up to about 30% (2:23:43).
Fact check: fair, but still topline. Lilly’s topline TRIUMPH-1 announcement on May 21, 2026 put average weight loss at about 28% on the top dose over 80 weeks, and around 30% in an extension in people with a high BMI. Consider those numbers provisional until the full paper is published.
Training wheels, and the night shift
Bakri describes his prescribing model as training wheels (2:28:32): the lowest dose that works, combined with diet, exercise and lifestyle changes, then tapering off or staying on as needed. Done that way, he says, his patients rarely describe the low mood and misery people post online. He blames those bad experiences on mega-dosing, barely eating (one meal a day), low blood pressure, low electrolytes, low micronutrients, low blood sugar, and losing the social side of meals. He separates those knock-on effects from the drug’s direct effect on the brain, which at the right dose he thinks may even reduce brain inflammation.
His own cautionary tale (2:32:32) is the best story in this section. A family member said an overseas brand-name Ozempic pen wasn’t working. To test it, Bakri injected about 1 mg, when the standard starting dose is 0.25 mg, stepping up to 0.5 mg. He was on a night shift, spent it vomiting in the call room between admissions, and calls it the worst night he can remember. The lesson he draws is to titrate slowly. He adds that newer drugs, retatrutide included, reportedly cause fewer stomach side effects.
Fact check: the 0.25 mg start, stepping to 0.5 mg, matches semaglutide’s label.
Dopamine, stacking and stopping
Studies show people drink less on GLP-1s, which suggests the drugs change dopamine signaling (2:29:20). That worries Bakri when people stack GLP-1s with ADHD medication or stimulant-type peptides like Semax and Selank, and he notes research websites keep upselling more compounds.
On stopping (2:25:20), he says the industry hasn’t given good tapering guidance, and some people regain all the way back to where they started, which he explains with set-point theory. His specific worry is 18- and 19-year-olds on 5 mg of retatrutide who have lost 30 to 40 pounds. Will they need it for life?
Huberman is relaxed about lifelong use, since better delivery methods are coming and excess weight shortens life more (2:26:08). What worries him is developing brains exposed to what he calls a thousandfold or greater increase in GLP-1 signaling. He cites neuroscientist Zach Knight: older diabetic drugs raised GLP-1 signaling two- to fourfold without big weight effects, the new drugs go vastly higher, and nobody knows the long-term effects on neuroplasticity and learning. Bakri pushes back that the brain effects might turn out to be positive. He notes that a big Alzheimer’s study last year showed no good signal, but says diabetic disease speeds cognitive decline, so better insulin control may still help the brain (2:27:44).
Fact check: the thousandfold figure is right for total drug, overstated for the free fraction. In our first version we called it almost certainly a rhetorical exaggeration. That was too harsh. Population pharmacokinetic data (Overgaard et al., Clinical Pharmacokinetics, 2018) put steady-state semaglutide at about 15.8 nmol/L at 0.5 mg weekly and 29.8 nmol/L at 1.0 mg, against natural GLP-1 in the low picomolar range, so total drug really is on the order of a thousand times higher. But more than 99% of semaglutide is bound to albumin, and a 2023 estimate (Smits et al.) puts the free level at roughly 180 to 350 pmol/L, which shrinks the gap to tens-fold. Both men’s framing holds for what’s in the blood. Huberman’s follow-up GLP-1 episode with a scientist who studies these receptors in the memory system is recapped in Huberman’s GLP-1 episode with Scott Kanoski.
Fact check: the Alzheimer’s trial was about early disease, not prevention. Novo Nordisk’s EVOKE and EVOKE+ trials, reported in late 2025, enrolled 3,808 adults with amyloid-confirmed mild cognitive impairment or mild Alzheimer’s disease. Oral semaglutide did not slow progression on the main clinical scale, although some biomarkers improved.
Fertility, prices and patents
Losing fat improves leptin sensitivity, and GLP-1 receptors sit on the brain’s POMC neurons, so women who were less fertile because of excess weight sometimes get pregnant unexpectedly, the so-called Ozempic babies (2:30:56).
On price, Bakri says tirzepatide now runs about $300 a month through websites and the new orforglipron pill about $150 (2:36:30). Fact check: orforglipron reached the US market as Foundayo in April 2026, with self-pay pricing starting at $149 for the lowest dose and rising to $349.
The retatrutide patent discussion is a lesson in how drug law shapes what you can buy (2:33:53). Huberman posted on X that retatrutide will be a trillion-dollar product, because Lilly will keep finding new uses that extend its patent life. Bakri says retatrutide is 39 amino acids long. Above 40, a molecule counts as a biologic, which the hosts say gets much longer protection and can’t be compounded. He says Lilly is trying to get it classified as a biologic, and that compounders already have the formula and Chinese raw material ready the moment they’re allowed.
Fact check: Bakri was right about the filing, even though the rule says otherwise. An FDA rule (Federal Register, February 2020, effective March 23, 2020) defines a protein, and therefore a biologic, as a chain of more than 40 amino acids, and retatrutide is a 39-residue peptide. Even so, Lilly’s July 23, 2026 announcement says it plans to submit a Biologics License Application for retatrutide in the first quarter of 2027, so the company is filing it as a biologic. We don’t know the regulatory reasoning behind that. The other numbers are looser: biologics get 12 years of exclusivity against 5 for a new chemical entity, not the roughly 15 versus 5 to 7 the hosts guessed, and hCG crossed into the biologic category in March 2020 and left compounding then, not recently. For how retatrutide compares with semaglutide and tirzepatide, see peptides for weight loss.
Pinealon: Huberman’s REM experiment, and what it was actually built for
This is the most charming stretch of the episode, and our first version gave it one sentence.
Huberman says he’s tried pinealon a handful of times and doesn’t use it regularly (1:07:25). Taken at the start of the night, it gave him less deep sleep and much more REM, which he considers a bad trade. Taken as a very small injection when he wakes in the middle of the night to use the bathroom, it roughly doubles the REM in his final hours, from about an hour and a half to about three hours in the same total sleep. He sometimes wakes groggy. He uses it about three times a month at most, often forgets about it for months, and believes the effect lingers on the nights in between (1:15:50). Half the recommended dose works well for him. The full dose gives him very deep sleep, elaborate dreams and waking. He sources it from a compounding pharmacy he trusts.
Then Bakri reframes the whole thing. First, the name is confusing (1:10:18). Epitalon traces back to an extract of the pineal gland. Pinealon traces back to a brain-cortex extract (Cortexin), not the pineal. Huberman decides to call it EDR on air, after its three amino acids, glutamic acid, aspartic acid and arginine.
Second, its origin (1:08:39). Bakri tells the story of Vladimir Khavinson being tasked by the Soviet military with protecting soldiers’ eyes, amid fears the US would target them with lasers, and with fixing the rapid ageing seen in nuclear-submarine crews and cosmonauts, who came back looking 10 to 20 years older, with broken circadian rhythms and weak immunity. Ground-up pineal and thymus extracts reportedly reversed this, and Khavinson then isolated the peptides responsible.
Third, and best: none of Khavinson’s pinealon papers mention REM (1:12:42). There were no consumer sleep trackers in 1970s Soviet labs, as Huberman jokes about his own sleep devices. Pinealon was designed as a daytime anti-stress and cognitive compound, sold orally in places like Kazakhstan (1:13:28). In Khavinson’s athlete study, after training to exhaustion, the pinealon group kept performing better than placebo. High doses are sedating, and taken in the morning or before high-intensity exercise, Bakri says, the effect is notable. Huberman laughs at himself for using a daytime cognition compound to chase dreams (1:14:16).
Bakri mentions a friend who runs a nine-figure company and has all his employees on pinealon in the mornings. People report less brain fog.
Doses as stated (1:15:03). Bakri says oral users settle somewhere between 0.5 mg and 3 mg, while the Russian Khavinson products contain about 200 micrograms. Some people inject it. Bakri admits, to laughter, that he took some the night before the recording, and that he takes larger doses when he wants good sleep. Some people feel briefly more awake at first, which he suggests may explain Huberman’s lost deep sleep.
Side effects as discussed (1:16:38). Nothing is flagged in the Russian literature, which Bakri caveats heavily as Russian literature. From the reports he’s collected, he names drops in blood sugar, with some people’s A1C falling, so he says people prone to low blood sugar should be careful. And very vivid, colour-rich dreams, which could be unwelcome for anyone who gets nightmares.
His proposed mechanism is speculative. Pinealon has no known receptor and is thought to bind DNA to expose gene promoter regions. He lists pathways including GDF11, the antioxidant enzymes SOD1 and SOD2, irisin and the PPAR receptors, and his one-line summary is that it improves brain metabolism. As a tripeptide, it’s assumed to cross into the brain and to be absorbed orally through peptide transporters.
Fact check: the evidence base is one national research tradition. The bioregulator literature comes almost entirely from Khavinson’s group and its successors, with little independent Western replication. Huberman’s REM effect is a self-experiment with a sleep tracker, which is interesting but is one person’s data.
Epitalon, the pineal gland and the longevity study
Epitalon is the tetrapeptide AEDG (alanine, glutamic acid, aspartic acid, glycine). Epithalamin is the whole-pineal-extract drug it was derived from (1:10:18).
Huberman changes his mind on air (1:19:55). He says he used to dismiss pineal calcification because Harvey Karten, whom he ranks among the great neuroanatomists, brushed it off. Then, at 1:20:35, he says Karten probably got this one wrong: the literature now supports that the pineal does calcify, and the open question is whether it matters. Our first version reported this backwards. Bakri adds that he sees calcification on brain scans of stroke and head-injury patients, and that by Khavinson’s account pineal decline is more about function than structure.
The claims from Khavinson’s work, as Bakri describes them: epithalamin restored melatonin output in old monkeys but did nothing in young ones, a result later reported in humans. It also boosts the enzyme that makes melatonin, shifts clock-gene expression in white blood cells toward a more rhythmic pattern, and raises morning cortisol. Huberman notes that’s the right direction, since you want cortisol high in the morning and low at night. The bioregulator model is a short course whose benefits supposedly keep accruing after you stop, which Bakri links to Huberman’s lingering REM effect.
The eye material is Huberman’s home turf (1:25:23). His old lab worked on the visual system, and he finds the mouse data on epitalon in models of retinitis pigmentosa and glaucoma-type degeneration credible enough to take seriously. He likes that it targets DNA repair. He also says BPC hasn’t been studied much for optic-nerve repair and should be. Bakri offers his own theory: epitalon may raise levels of melanopsin, the light-sensing pigment in the eye, which could make morning sunlight more effective. He suggests people who say sunlight does nothing for them may not be getting enough darkness at night to regenerate it.
The longevity study (1:24:36). As Bakri tells it, nursing-home residents were split into groups, and one received epithalamin plus thymalin by injection in short courses once or twice a year for 15 years. That group had much lower death rates from heart disease, infections and tumours. Bakri calls it potentially the most interesting longevity study he’s seen, if it holds up, and flags that it’s a Russian study.
Fact check: shorter than he said, and unreplicated. The published paper is Khavinson and Morozov (Neuroendocrinology Letters, 2003). It followed 266 elderly people for six to eight years, not 15. It reported mortality 2.0 to 2.1 times lower on thymalin, 1.6 to 1.8 times lower on epithalamin, 2.5 times lower on both, and 4.1 times lower with annual combined courses over six years. Effects that large, from one group, never independently replicated, are a reason for scepticism rather than excitement.
Doses as stated (1:44:25). Bakri says the bioregulator literature uses 10 to 100 micrograms of the purified peptides and about 10 mg of the peptide mixtures. The original sourcing was animal tissue: he says 10 mg of desiccated cow brain might yield a few hundred micrograms of pinealon. That prompts the prion story, which our first version credited to Huberman. It’s Bakri’s: his first patient on the wards in his third year of medical school had Creutzfeldt-Jakob disease. One of the two then makes a plea not to eat brain extracts. Bakri says the prion scares of the early 2000s set back research on animal-extract peptides.
Huberman says he’d like his premium channel to fund independent tests of these compounds designed to disprove them.
The thymus: 35 minutes nobody else covered
Bakri’s thesis is that immune ageing follows the decline of the thymus, and that this matters far more than most people, including most doctors, realize. It runs from 1:29:38 to about 2:03:42.
What it is. The thymus sits above the heart, behind the breastbone. In a baby it can be the size of a baseball, about half the size of the heart. In adults it’s mostly fat with islands of thymic tissue, which surgeons often remove during open-heart surgery (1:30:10). T cells are made in bone marrow and trained in the thymus. A teenager produces a very large number of new naive T cells every day, and those cells live 10 to 15 years. When thymic output bottoms out, Bakri says, tumours, heart disease and autoimmunity rise.
Fact check: the thymectomy data are real and striking. Kooshesh and colleagues (NEJM, 2023) compared 1,146 adults who had their thymus removed during heart surgery with 1,146 matched heart-surgery patients who didn’t. Five-year all-cause mortality was 8.1% against 2.8%, and tumour rates 7.4% against 3.7%. Higher autoimmunity showed up only after excluding people with existing infections, tumours or autoimmune conditions (12.3% against 7.9%). It’s observational, but the signal is strong.
Fact check: the 2026 imaging paper used CT. Bakri cites a 2026 Nature study linking thymic health to lower mortality (1:32:33). There are two companion papers from Hugo Aerts’s group at Mass General Brigham. One applied a deep-learning thymic-health score to the CT scans of 27,612 people from the Framingham Heart Study and the National Lung Screening Trial. High scorers had about 50% lower all-cause mortality, 63% lower cardiovascular mortality and 36% lower lung-tumour risk. Bakri says MRI, but it was CT, and the associations don’t show that rebuilding a thymus would help.
What shrinks it. Sex hormones (androgens, estrogens, progestins) and corticosteroids, Bakri says, and castration partly reverses the shrinkage. Huberman points out these are exactly the hormones people want more of. Pregnancy shrinks the thymus, which protects the fetus from immune attack, and it regrows during breastfeeding under growth hormone and prolactin. Hibernating animals shrink and regrow theirs too.
Infections and convalescence (1:54:53). Every infection shrinks the thymus, Bakri says, and it used to recover during a period of convalescence. Now stress, poor light, low vitamin D, processed food and excess weight stop it recovering, so people get sick again within weeks. Since the pandemic, his colleagues report three or four colds a winter instead of one. He laments that there are no thymologists, as he puts it, since immunology as a specialty doesn’t deal with immunity in the general population.
Regrowing it. Bakri cites Greg Fahy’s TRIIM trial, which gave growth hormone, metformin and DHEA for a year (1:34:09). The thymus grew on imaging and T-cell markers improved. Fact check: Fahy and colleagues (Aging Cell, 2019) did report thymic regeneration on MRI and improved immune markers, with epigenetic age about 1.5 years below baseline after a year. It was nine men aged 51 to 65 with no control group, and growth hormone carries the insulin-resistance cost Bakri describes later.
Extracts and the zinc link. About ten Eastern European groups made their own calf-thymus extract drugs, using six-month-old calves for the largest glands, and published hundreds of papers. The Goldstein lab in the US made thymosin fraction 5, and French researchers discovered thymulin (1:46:02). Thymulin is nine amino acids and depends on zinc, part of a pattern Bakri points out: haemoglobin pairs with iron, GHK with copper, thymulin with zinc. It’s the main marker of thymic function and falls sharply with age. Its half-life is very short, so the goal is to raise your own production, and Bakri says thymulin crashes before standard red-cell or serum zinc tests show any depletion (1:48:27). Don’t confuse it with thymalin, Khavinson’s Russian thymic polypeptide mix. Bakri says peptide websites mix the two up constantly.
The $3 test (1:59:42). This is the most practical, needle-free takeaway in the episode. Bakri says the lymphocyte-to-monocyte ratio, which comes free with any standard blood count with differential (he prices it at about $3), is linked to poor outcomes across many conditions when low. He says risk starts climbing as absolute lymphocytes fall toward 1,000, while young, healthy people sit around 1,500 to 3,000. Asked how many of 100 US physicians know about it, he says none (2:02:07). With various interventions, some of them peptides, he says his patients have moved from a ratio of 4 to 1 up to 8 to 1. He also mentions the CD4 to CD8 ratio, where the youthful pattern is not having more CD8 than CD4, and says he’s lobbying radiologists to report a thymic score on full-body scans.
A warning from history (1:28:33). Bakri tells the story of doctors in the early 20th century irradiating infants’ thymus glands, believing a large thymus pressed on the heart and caused sudden infant death. He estimates at least 10,000 people later died of tumours as a result. His point is that medicine has done real harm with confident interventions before, and that applies to peptides too. Fact check: infant thymus irradiation was practised from 1907 and was common in the US from about 1930 to 1945. Published estimates suggest more than 10,000 later thyroid tumours, so the scale is broadly right, though the figures count tumours rather than deaths.
Thymosin alpha-1: Bakri’s ward-year experiment
Thymosin alpha-1 is part of the thymic hormone family. Bakri describes it as fuel for T cells: it boosts T-cell development inside and outside the thymus and helps them attack pathogens (1:34:57). He says it was approved in the US as Zadaxin for children born with a missing or faulty thymus, though the approval may no longer exist, and that other countries approve it as an add-on for hepatitis B and C and some tumours. He calls the sepsis and infection literature not that promising, though recovery might be quicker when it’s given with antibiotics.
His thought experiment (1:36:31): give everyone in a nursing home thymosin alpha-1 in November and December. Is there less flu in January and February?
Bakri’s own use (1:37:20). He used it while travelling to avoid getting sick, and throughout his ward year, when for the first time he stayed free of infections all year. He says he took 2.5 mg twice a week, at any time of day. It’s a self-experiment, not approved for that use, and he describes it as a way to stay well for his patients. Huberman calls the common reasons people take it (feeling run down, travelling, short on sleep, new parents) wild-west indications.
Fact check: it was never approved in the US. Zadaxin (thymalfasin) holds FDA orphan-drug designations, including for DiGeorge anomaly and chronic hepatitis B, but orphan designation is not approval. It is approved in more than 35 countries for hepatitis B and C and as an adjuvant. Bakri says it has at least 21 amino acids, and it has 28. And the sepsis caution is backed by the largest trial: TESTS (BMJ, 2025) enrolled 1,106 adults at 22 Chinese centres, and 28-day mortality was 23.4% on thymosin alpha-1 against 24.1% on placebo, no meaningful difference.
TB-500 and the Wolverine stack
TB-500 gets less time than BPC, and the episode talks about it as if it were the same thing as thymosin beta-4 (1:38:13). Its origin is the Goldstein lab, which isolated thymosin fraction 5 from thymus tissue, a mixture that includes both thymosin alpha-1 and thymosin beta-4. Neither is made only in the thymus. Thymosin beta-4 is 43 amino acids and works on the actin cytoskeleton, the scaffolding cells use to move. Immune cells need to move fast, and it seems to help. In wound models it increases cell migration, especially paired with BPC. Like BPC, it has no identified receptor.
Bakri’s aside (1:38:53): it’s a common doping agent in horse racing, given to the horses.
The Wolverine stack, BPC-157 plus TB-500, is a name Bakri says was invented online (1:53:17). Add GHK-Cu and it becomes the “glow stack.” There’s debate over whether mixing them in one vial degrades the peptides, which he leaves aside. No doses for TB-500 were discussed. Bakri expects it in the first round of FDA Category 1 review alongside BPC.
Fact check: what’s sold as TB-500 is usually a synthetic fragment or analogue of thymosin beta-4, not necessarily the full 43-amino-acid protein, so the two aren’t always identical. We go deeper in BPC-157 vs TB-500.
GHK-Cu: skin, hair and the blue-colour trap
Huberman says most GHK-Cu questions he gets come from women, and he polled the women in his life before the episode (1:50:33). His first request: don’t inject it into your face (1:50:52).
GHK-Cu is a tripeptide (glycine, histidine, lysine) carrying a copper ion. Bakri credits Loren Pickart, who has since died, with discovering it and proposing that it controls both collagen building and breakdown, which proper tissue remodelling needs (1:51:40). He says levels fall from around 200 in youth to the 60s by age 65, and that nearly all the literature comes from Pickart’s lab, a pattern he notes keeps repeating with peptides.
The evidence as discussed: in Pickart’s work, topical GHK-Cu compared well with retinol and vitamin C creams, and people report crow’s feet fading. On hair, a study wasn’t promising, and despite peptide-site claims it isn’t better than minoxidil, though it might be a useful add-on. Injected at a distant site in rats, it sped up skin repair. A Chinese group is studying it for lung fibrosis.
Then the practical part (2:15:36). Formulation matters, and Bakri says a well-made topical from a real skincare brand is the way to go, while research-chemical topicals tend to be poor. GHK-Cu should look blue from the copper, but blue doesn’t prove the peptide is intact, since free copper can supply the colour. Topicals have been around for 30 to 40 years. The injectable has no good human data, and Bakri expects it in a later round of FDA review. He says there’s good human data on cosmetic outcomes, especially combined with red light, and that the consensus is to use it alongside a skincare routine rather than instead of one. Huberman adds that sunlight already supplies red and near-infrared light, as long as you watch UV exposure.
Fact check: right on the decline, wrong on where it was found. Pickart isolated GHK from human plasma in 1973, not from collagen, although the GHK sequence does occur in collagen. The usual figures are about 200 ng/mL at age 20, falling to about 80 by 60. Our first version repeated the collagen error.
Growth hormone, secretagogues and Huberman’s PSA scare
Our first version squeezed this into two sentences and named a compound, CJC-1295, that never came up. The secretagogues actually discussed were MK-677, tesamorelin, ipamorelin and sermorelin.
Huberman sets it up (2:04:01). These compounds raise the body’s own growth-hormone output, differ in how much they drive hunger, and are usually taken before sleep, two to three hours after eating. Most users want fat loss, muscle, faster recovery and a younger look. Are those effects real?
The compounds. MK-677 (ibutamoren) is an oral ghrelin agonist and technically not a peptide. It releases growth hormone continuously instead of in pulses, and Bakri says it has the worst off-target effects of the group: hunger, prolactin and cortisol. Tesamorelin is more selective with fewer hunger effects, ipamorelin is commonly paired with it, and sermorelin is a weaker GHRH analogue (2:13:19). Naturally, a big growth-hormone pulse comes in roughly the first 90 minutes of slow-wave sleep. Miss it and you get small pulses through the day. Whether the big pulse matters is an open question.
The ageing timeline. Growth-hormone output drops sharply in your 30s, which Bakri calls somatopause. His sequence: the pineal ages before puberty, the thymus right after puberty and through the 20s, then somatopause, then adrenopause (less DHEA), then menopause or andropause.
The case for and against. IGF-1 supports skin, sleep, muscle recovery and joints, and growth hormone regrows an aged thymus, as in Fahy’s work. Against: there’s no good data on whether growth hormone speeds up tumours that already exist, though Bakri says there’s no direct mechanism by which it causes them. Animals deficient in growth hormone live longer. It isn’t great for cardiometabolic health. Huberman notes large dog breeds, which make more IGF-1, die younger. Bakri raises antagonistic pleiotropy, the idea that what helps you young may cost you later, and weighs it against possible better immune surveillance from a regrown thymus.
Doses and numbers as stated. Huberman says he has friends who take 1 IU of growth hormone a night, some 2 IU, five nights a week, for years, and report feeling markedly sharper and stronger (2:09:20). He finds that both tempting and alarming, because even good imaging won’t tell you whether you’re feeding a tumour. Bakri says tesamorelin plus ipamorelin together can push IGF-1 into the 380s to 390s, which he calls puberty-level (2:13:19). Bakri also floats what he explicitly calls a theoretical idea, not advice: since nothing in nature runs year-round, a cyclical course of tesamorelin under a clinician, raising IGF-1 to a target, with thymus imaging and CD4/CD8 counts before and after (2:10:56). Later he says nobody should take tesamorelin without checking IGF-1 (2:43:43).
The costs Bakri actually warns about. Growth hormone and the secretagogues reduce insulin sensitivity, and A1C usually rises, especially on tesamorelin plus ipamorelin (2:12:31). He repeats a bodybuilding joke that you have to already be lean and metabolically healthy to tolerate growth hormone. He claims you can spot growth-hormone users in gyms by unusually glossy, youthful skin and healthier hair (2:10:08). And he flags prostate growth, along with heart and liver growth.
Huberman’s sermorelin story (2:10:56). Years ago he tried sermorelin. It greatly increased his deep sleep and wiped out his REM, the opposite of what pinealon does for him. It also spiked his PSA, which had always been low. He stopped immediately and his PSA came back down. Bakri riffs on the enlarged prostate as the curse waiting for every man in later life, from night-time bathroom trips to hunting for toilets at amusement parks, and mentions that Khavinson’s prostate peptides are being translated.
The trinity stack in full (2:14:06). A GLP-1 (tirzepatide or retatrutide) for insulin sensitivity, plus growth hormone or a secretagogue (tesamorelin and ipamorelin for those who can’t afford growth hormone), plus androgens. It produces fast fat loss and muscle gain in CEOs and celebrities. For women, Huberman guesses a secretagogue plus a GLP-1 is the common version. Bakri adds that some celebrities lost too much weight and needed better clinicians (2:43:43). This is where the opening line comes from. It’s fast, it’s visible, and nobody knows yet whether it’s healthy.
Our peptides for muscle growth guide covers the published research on this family.
Quick hits
Pets (1:38:53). Vets turn out to be more open to peptides than you’d expect, and pet-peptide brands are appearing.
Women’s conditions (2:39:42). Asked about endometriosis and fibroids, Bakri says there’s no animal or human literature on BPC for either. Hormones are a much stronger lever, and he sends people to a good OB-GYN.
Head injuries (2:40:31). There’s no good BPC data for traumatic brain injury, though he finds it theoretically interesting because of the anti-stress angle. Russian studies exist for Cortexin and Cerebrolysin, which he doesn’t expect to reach the US.
NAD (1:39:40). In an aside, Huberman says a new Nature Metabolism paper reports blood NAD doesn’t fall with age, complains that news coverage implied he’d called NAD a longevity drug, and says he uses NMN for morning energy.
Where to get peptides, according to the hosts
The listener question about sources gets a direct answer (2:42:54). Bakri says to find a physician you trust, ideally one who knows peptides, and he expects physician-led, monitored access to arrive within roughly six months to two years. He doesn’t trust research-only websites, citing wrong compounds and batch-to-batch variation, and warns away from black-market vials entirely. And he tells fellow doctors they can’t opt out of learning about peptides anymore.
His closing advice is the least glamorous and probably the most important (2:43:43). Peptides make no sense layered on top of poor sleep, no morning light, no darkness at night and processed food. Get the basics right first. And Huberman’s standing advice from earlier applies to anything compounded: ask what your clinician pays and what they charge.
For our part, we sell research compounds for laboratory use, and we think the episode makes the case for one thing above everything else: knowing what’s in the vial. Every compound we carry is third-party tested by Janoshik for identity and purity, and the reports are posted on our lab testing page. If you’re comparing suppliers, our Canadian peptide supplier guide explains what to check.
The compounds discussed in this episode that Peptigo carries for research use: BPC-157, TB-500, the Wolverine Blend (BPC-157 + TB-500), the Glow Blend, Thymosin Alpha-1, Epitalon, GHK-Cu, Tesamorelin, Ipamorelin, Retatrutide, Semaglutide and Tirzepatide.
My take: was it worth three hours?
Yes, and more than we gave it credit for the first time around.
The value isn’t a list of what to take. It’s a practicing physician and a neuroscientist repeatedly marking where knowledge stops. BPC-157’s human data are two abstracts. Its effective dose is unknown, and the two men disagree about it from personal experience. Pinealon’s most famous effect, at least in Huberman’s house, was never measured by the people who made it. The longevity study ran half as long as it’s remembered. Thymosin alpha-1’s US approval didn’t happen. And the GLP-1 drugs, the class with the strongest evidence by far, are being taken at doses and in combinations that the trials never tested.
The best through-line is Huberman’s: his bigger fear is the supply chain, not the molecules. The melanotan story, the five-dollar vials, the affiliates and the Venmo payments to rotating names all point the same way. The interesting biology is real. What’s in a given vial, and at what dose, is the question most people skip.
If you’re short on time, the BPC dosing exchange (54:13 to 56:37), the gray-market stretch (29:41 to 40:38) and the pinealon reveal (1:07:25 to 1:18:12) are the three parts to hear. If you have longer, the thymus section is the one nobody else has covered properly, and the $3 ratio is the only takeaway in the episode that doesn’t involve a needle.
What we corrected from our first version
We published the first version of this recap on June 1, 2026, working from a faster read of the captions. Going back through the full episode, we found these errors and fixed them:
- The melanotan story was about a man who went viral on Twitter, not one of Bakri’s patients. We also described a supplier funnelling materials into unlabeled bins, which was a misreading of Huberman’s description of supply tiers.
- The BPC dose admission. We reported that Bakri admitted to gram-level doses. The gram figure came from the host’s prompt, possibly mis-captioned, and Bakri never gives a number. The 100 to 200 microgram framing also came from the host’s question, which Bakri agreed with.
- The prion story is Bakri’s, from his time on the wards, not Huberman’s.
- Pineal calcification. We said Huberman brought skepticism via Harvey Karten. He actually said Karten was probably wrong and that calcification is real.
- CJC-1295 never came up. The secretagogues discussed were MK-677, tesamorelin, ipamorelin and sermorelin.
- The growth-hormone caution. Bakri’s specific warnings were insulin resistance, rising A1C, prostate growth and the need to monitor IGF-1, not mainly a tumour link, which he said has no good data either way.
- GHK-Cu was first isolated from plasma, not collagen.
- The thousandfold GLP-1 figure is the right order of magnitude for total drug in the blood. We were too dismissive.
- The regulatory timeline now includes the April 2026 removal from Category 2 and the July 2026 advisory vote.
- The single-lab point was raised first by Bakri, then repeated by Huberman, not the other way around.
This article recaps a public Huberman Lab podcast episode for educational and informational purposes. Statements and doses are attributed to the hosts as said on air, or to the published research named, and many of the findings discussed come from animal studies or individual anecdotes. Nothing here is a protocol or a recommendation. All Peptigo products are sold strictly for laboratory and research use only, not for human or veterinary use, and nothing here is medical advice. Peptigo does not make medical claims. Consult a qualified professional before any decision involving these compounds.
Episode: “Peptides: The Science, Uses & Safety” with Dr. Abud Bakri, Huberman Lab, June 1, 2026. Timestamps refer to the YouTube version.
