Huberman Lab — Dr. Abu Bakri: A Masterclass on Peptides

Provenance: physician masterclass — mechanism-rich, but mostly animal data + Russian/Croatian literature

Dr. Bakri is an internal-medicine physician with encyclopedic peptide knowledge. Much of what’s below is [anecdote] (clinical/mechanistic framing) or [literature-animal] / [literature-russian] (single-group or Soviet-era data with explicit reliability caveats). Treat dosing as cited research-use information, not medical advice. Research/educational use only; defer to peer-reviewed kb-paper nodes where they conflict. Source: YouTube — _DfqnpSbMfE

Key frameworks from this episode

• Receptor vs. no-receptor peptides (Bakri’s primary lens): GLP-1s have well-defined receptors → strong, predictable clinical effects. BPC-157, TB-500/TB4 have no identified receptor → effects via protein modification, gene-transcription assistance, or still-elusive targets. The Russian (Khavinson) bioregulators are epigenetic modifiers that bind the DNA groove to open/close chromatin — like steroid hormones.
• BPC lineage: Pavlov’s gastric-juice work → Selye’s stress-adaptation (stress shrinks thymus/lymphatics, ulcerates gut, enlarges adrenals) → Croatian group (1991) isolates the 40-kDa BPC protein, then the 15-amino-acid BPC-157 fragment. We do not make BPC-157; we make the big BPC protein.
• Bioregulator model: these tri/tetrapeptides are abundant in youth and decline with age; replenishing them (often only ~10–20 days/year) is claimed to accrue benefits during the off-period by re-tuning gene expression.
• Sourcing tiers: pharma (most stringent) → compounding pharmacies (variable quality, provider price markups) → gray market (“research use only,” all API from China, batch-to-batch unknown) → black market (don’t).
• The “trinity stack” (celebrity/CEO protocol): GLP-1 (insulin sensitivity) + GH secretagogue (e.g. tesamorelin/ipamorelin) + androgen modulation (TRT ± anavar). The “Wolverine stack” = BPC-157 + TB-500; add GHK-Cu → the “glow stack.”

Claims & cautions by peptide

• BPC-157 — [anecdote] Identity: 15-aa fragment of the natural 40-kDa gastric protein BPC; PDA (pentadeca arginate) is the same molecule (acetate vs arginate) sold to dodge the category-2 ban. Strong gut-brain-axis effects (blunted Adderall, anhedonia reports; in mice, reduced intoxication & withdrawal); oral BPC seems to act locally in the gut, not systemically. [literature-animal] Croatian phase 1/2 enema trials (~80 mg) for ulcerative colitis (full data unpublished — and nearly all BPC data come from one group); no animal carcinogen signal but LD50 unknown. [protocol] Bakri used gram-range local doses for a grade-2 triceps tear (fast healing) and argues common microgram doses may be sub-therapeutic. [warning] Anecdotal worsening of angiomas/hematomas (VEGF), though it lowered VEGF in a melanoma line — net cancer effect unknown. [00:07, 00:18–00:55]
• TB-500 — [anecdote] Thymosin β4, 43-aa, acts on the actin cytoskeleton (cell migration); part of the thymic ‘fraction 5’ family; common horse-racing doping agent; pairs with BPC-157 (‘Wolverine stack’). [01:38, 01:52]
• Pinealon — [anecdote] Tripeptide EDR; despite the name it derives from a cortex extract (cortexin), not the pineal. Epigenetic modifier; modulates PPARα/γ, SOD, IRIS → better neuronal oxidative state; reported daytime alertness/HIIT performance and more REM at night, with benefits persisting off-dose. [protocol] Oral ~0.5–3 mg (Russian product ~200 mcg); some inject. [warning] Can lower blood sugar/A1C; vivid dreams. [01:08–01:18]
• Epithalon — [literature-russian] Tripeptide AEDG from the pineal, mimics endogenous epithalamin; restores melatonin in aged animals; normalises clock genes/morning cortisol. Khavinson’s 15-year nursing-home study (epithalamin + thymalin) reported lower CVD/infection/cancer mortality. [literature-animal] Mouse retinal-neuroprotection (RP, glaucoma) via DNA-repair support; Bakri hypothesises melanopsin upregulation. [01:22–01:27]
• Thymosin Alpha-1 — [anecdote] ≥21-aa thymic hormone, ‘jet fuel for T-cells’; FDA-approved historically as Zadaxin (thymus-deficient children); abroad an adjuvant for hep B/C and some cancers. [protocol] Bakri uses ~2.5 mg twice weekly as a travel/ward prophylactic. [01:35–01:37]
• GHK-Cu — [anecdote] Copper tripeptide (Gly-His-Lys + Cu²⁺) in type-1 collagen (Pickart); serum ~200→60 ng youth→65; regulates collagen synthesis and breakdown; a real product is blue. [protocol] Best evidence topical (vs retinol), synergistic with red light, adjunct not replacement; hair effect modest (adjunct to minoxidil); don’t inject the face. [literature-animal] Animal wound repair; Chinese group studying lung regeneration (COPD/long-COVID). [01:51–02:18]
• Tesamorelin — [anecdote] Stronger GHRH-mimetic than sermorelin; + ipamorelin can drive IGF-1 to ~380–390 (puberty levels); addresses somatopause. [warning] Reduces insulin sensitivity / raises A1C (worse stacked with ipamorelin). [02:05–02:14]
• Ipamorelin — [warning] Amplifies GH/IGF-1 with tesamorelin but worsens insulin resistance; ghrelin-agonism drives hunger. [02:13–02:14]
• HGH — [anecdote] Somatopause (GH drops in the 30s); GH is thymo-regenerative (Fahy TRIIM: GH+metformin+DHEA regrew thymus). [warning] Lowers insulin sensitivity; promotes growth of existing tumors (growth factor, not a carcinogen); GH-deficient models live longer (antagonistic pleiotropy). [02:06–02:13]
• IGF-1 LR3 — [anecdote] IGF-1 output varies by dog breed: high-IGF-1 (Great Dane) → shorter lifespan vs low (Chihuahua) — the growth↔longevity trade-off. [02:08]
• HCG — [anecdote] Acts as synthetic LH (LH/hCG receptor); male fertility restoration + female IVF; in animals thymulin sensitises the gonad so hCG yields more testosterone; recently removed from compounders. [01:47, 02:35]
• Retatrutide — [anecdote] Triple agonist (GLP-1 + GIP + glucagon), 39 aa — Lily reportedly aims to push it >40 aa to classify as a biologic (longer patent, blocks compounding); newer agents report less GI nausea. [02:33–02:35]
• Tirzepatide — [anecdote] GLP-1 class traces to Gila-monster saliva; weight governed by a brain set-point; GLP-1 is a strong ‘don’t-eat’ signal. Bakri’s 1 mg Ozempic self-test → severe vomiting (titrate slowly). [02:24–02:33]

New compounds discussed without a note yet (candidates)

• PDA (pentadeca arginate) — functionally identical to BPC-157; worth an alias/redirect note.
• Thymulin — zinc-dependent 9-aa thymic peptide; sensitises end-organs to hormones (hCG→more testosterone); falls with age and is the earliest marker of zinc depletion. Strong candidate.
• Sermorelin (weak GHRH-mimetic; recurring across all 3 episodes), MK-677 (oral non-peptide GHRP, non-pulsatile GH “bleed-out”), hexarelin, thymalin (Russian thymic polypeptide mix), cortexin (pinealon source), cerebrolysin, orforglipron (oral non-peptide GLP-1), cardarine/GW (PPARδ; animal cancer signal), thymosin fraction 5.

Stack candidates surfaced

• Wolverine stack = BPC-157 + TB-500 (repair).
• Glow stack = Wolverine + GHK-Cu (aesthetic/collagen).
• Trinity stack = GLP-1 + GH secretagogue + androgen modulation (body recomposition; “celebrity protocol”).

Full transcript

Sponsor reads (Eight Sleep, Lingo, AG1, Function, Element) and the closing book/newsletter/social outro are trimmed ([sponsor segment omitted]). A few long non-peptide tangents are marked [condensed]; all substantive peptide content is preserved with timestamps.

[00:00] Bakri (cold open): People are now stacking their GLP-1 as their insulin-sensitivity tool, their growth hormone / GHRH, and their androgen-modulation therapies as this trinity stack to get very fit, very healthy quickly. A lot of the transformations you see in CEOs and celebrities use that combination — TRT plus tirzepatide/retatrutide, plus growth-hormone modulation (GH itself if affordable, or tesamorelin). People lose a lot of fat and gain muscle in short order. Is that healthy? We’ll find out — but that’s the celebrity protocol.

[00:29] Huberman: Welcome to the Huberman Lab podcast. My guest is Dr. Abu Bakri, an internal-medicine physician extremely knowledgeable on peptides — both FDA-approved (the GLP-1 agonists: Ozempic, Mounjaro, retatrutide) and others like BPC-157, which has a long history of human use for gut health and tissue repair with much animal data but essentially one formal human study. We discuss BPC-157, growth-hormone secretagogues (tesamorelin, MK-677), GHK-Cu for collagen/skin/hair, and DNA-repair/longevity peptides like epitalon and pinealon (also touted for REM sleep and cognition). Bakri thinks about peptides through the lens of whether they have known receptors, and their real safety profiles.

[03:31] Huberman: Peptides — huge topic. As an MD/clinician, how do you carve it up? Bakri: Scientifically, it’s one of the body’s languages. DNA → RNA → proteins, which break down to polypeptides and peptides. Peptides are one signalling language; steroid hormones another. Peptides subdivide by whether they have receptors or not — GLP-1s have strong receptor-mediated effects, vs obscure peptides like BPC-157, TB-500, TB4 with no clear target.

[05:01] Huberman: If BPC-157 has no receptor, how does it act? Bakri: Either the receptor is still elusive, or it modifies existing proteins / links proteins favourably for gene transcription. The Russian peptides are epigenetic modifiers — they bind the DNA groove and open/close chromatin at specific regions (like a steroid hormone entering the nucleus to switch on genes; puberty is the classic example). E.g. pinealon shuttles heat-shock proteins with androgen receptors. So: two big categories — with known receptors (GLP-1s) and without (but still biologically active, per animal data).

[06:46] Huberman: Let’s start with BPC-157. What is it? Bakri: For thousands of years we sought medicines in plants; the last ~150 years, in cells/animal tissue (aspirin, metformin, statins are plant/fungal; now we look at animal tissues). A Croatian group in the ’90s isolated BPC, a ~40-kDa protein; BPC-157 is 15 amino acids of it. We don’t make BPC-157 naturally — we make the big BPC protein.

[08:07] Bakri: Why look there? Pavlov — beyond the dogs/bell work — studied dogs’ gastric juices, putting a hole in the stomach, collecting juice, and selling it as a medicine for GI distress and wound healing. Around the same era, Selye’s stress-adaptation theory: stressed animals show enlarged adrenals (more cortisol), destroyed gastric lining, and shrunken thymus/lymphatics. So the Croatian group reasoned there must be a cytoprotective/organoprotective compound in gut juice (the stomach is a rich endocrine tissue — ghrelin, etc.). People historically drank the gastric juice as an elixir of many peptides.

[10:39] Bakri: The 1991 paper identified the 40-Dalton BPC protein, then BPC-157 as the active 15-aa peptide. Every organ has a peptide signature (e.g. carnosine/carnitine found in muscle in the 1800s) — gut peptides may have gut effects.

[11:34] Huberman: They injected the fragment into mice with induced injuries. Bakri: They did all sorts of things — severed tendons, ACLs, burn wounds — and saw faster healing. In burn-wound ICU-type models, topical BPC prevented the gastric ulcers that normally follow; they called it an anti-stress compound. The Achilles paper is what exploded bodybuilder interest (→ MSK use), but the original idea was gastric, not musculoskeletal.

[12:26] [sponsor segment omitted — Eight Sleep, Lingo]

[14:46] Huberman: What’s so striking — my lab worked on optic-nerve/neural regeneration; CNS nerves don’t regenerate. Yet there are animal data showing BPC accelerates healing of tendon, ligament, nerve pathways, and generally promotes ‘repair’ — weird for something with no known receptor. Bakri: It modulates growth/healing pathways: more VEGF signalling (→ angiogenesis, hence the safety controversy), cell migration (especially with TB-500/TB4), more access for healing factors via androgenic pathways, plus an anti-stress effect. They co-administered corticosteroids (which normally stall wound healing) with BPC and healing was the same or better.

[17:00] Huberman: Is BPC anti-inflammatory or does it preserve the needed pro-inflammatory repair response? Bakri: More gas pedal on the regenerative processes — bringing in immune/healing factors. In one tendon model it increased growth-hormone receptors on the tendon (→ more GH docking → outgrowth). Downstream it modulates nitric-oxide synthesis (vasodilation, cell recruitment). On the neuro side they made mice ‘drunk’ and BPC made them less drunk / eased alcohol withdrawal (normally deadly without benzodiazepines). The gastric and neuro/psychiatric effects (gut-brain axis) are arguably more interesting than the MSK hype. Patients report BPC blunting their Adderall, and Reddit has anhedonia discussions (depressed/low energy) — so it touches dopaminergic signalling on both sides.

[19:47] Huberman: Known adverse events (excluding contamination)? Bakri: In animals, ~1000× doses with no real adverse effects; we don’t even know the LD50 (which blocks FDA approval). Two small phase 1/2 rectal-enema trials for ulcerative colitis (early 2000s, same Croatian group), up to 80 mg — far more than the 100–200 mcg/day people inject. Phase 1: no adverse effects; phase 2 (~40 patients): a positive ulcerative-colitis signal — but full data unpublished (abstracts only), which gives pause. Orally/rectally administered BPC doesn’t appear to go systemic (not found in blood — broken down fast or stays local).

[24:17] [condensed — discussion of whether Croatian/Soviet-era trials are trustworthy: Bakri notes the groups run rigorous RCTs; US-centric bias and possible Soviet-era pro-fabrication incentives both exist; Russians prioritised performance (athletes/soldiers), the US a subscription-drug model.]

[25:25] Huberman: Is BPC-157 legal in the US? Bakri: Legally research-purposes-only. Never FDA-approved → compounding categories: cat-1 (OK to compound), cat-2 (do not compound), cat-3. In late 2024, BPC-157 and ~20 peptides moved to category 2. Compounders relabel it as PDA (pentadeca arginate) — the same molecule (acetate vs arginate). In April the acetate came off cat-2 but isn’t yet cat-1; PDA is being prescribed. State medical boards vary — some forbid prescribing any non-FDA-approved peptide regardless; telehealth law applies where the patient is.

[28:02] Huberman: The Olympic-Achilles-transection rumour (athlete podiumed after local BPC). Bakri: Hearsay; they surely used many things (exosomes, stem cells, other peptides) — BPC wasn’t on banned lists then (it now is on the WADA list; a 2024 paper can detect prior use). The research community ran with the story as a marketing tool. Context: GLP-1 supply shortages (2021–22) pushed people to alternative sources (e.g. Mexico: ~$150vs$1,500 for an Ozempic pen); compounders legally stepped in during the shortage (the FDA even asked them to), then later cracked down.

[31:12] [condensed — compounding economics: provider sets the price and pockets the markup (vial costs ~$150fromthepharmacy,soldtopatientfor$200–$800); Bakri argues patients have every right to ask the clinician’s cost. Standard pharma pens involve no such markup but have indirect incentives.]

[34:17] Huberman: Gray market — ‘research use only,’ Venmo/Zelle, ever-changing names. I obtained some, was too cautious to use much; I have tried BPC via a compounding pharmacy. Bakri: ~$5–10 B/yr gray-market peptide spend in the US (2025). All API comes from China — there are no American-made peptides (finishing happens here). Compounders vary in grading/sterility; they must add something (B12/B6) to meet compounding rules. Anecdotally people respond better to pharma pens than compounded; research-grade is all over the place (sometimes wrong substance — e.g. a man who ‘took retatrutide’ got darker: he was injecting melanotan). Batch-to-batch variability is the core problem. Black market = buying raw vials from China/WhatsApp or bathtub synthesis — do not.

[40:50] [condensed — destigmatization: GLP-1 self-injection normalised needles; >half the peptide market is women (skin/collagen via GHK-Cu); “soccer-mom” affiliate selling on TikTok.]

[41:44] Huberman: The angiogenesis/tumor concern — any evidence tumors actually grow faster? Bakri: No animal carcinogen signal for BPC (cf. cardarine/GW, scrapped for an animal cancer signal) — but all from one group, so caution. A US phase-2 hamstring trial is reportedly underway; an East-Coast orthopedic group is planning a BPC trial. On Reddit, hematomas worsening is a reported signal (consistent with VEGF). A physician friend reports facial angiomas worsening on BPC. But in a melanoma cell line BPC decreased VEGF — so it’s not a uniform angiogenesis upregulator; net cancer effect unknown. ICU use (preventing gastric ulcers) would be compelling if proven.

[45:12] [sponsor segment omitted — AG1]

[47:02] Huberman: Formal RCT timing / who holds the patent? Bakri: Multiple BPC-157 patents by salt; one ended up with Teva (generic giant, uninterested); another expires in ~10 years (Dr. Šikirić in Croatia). Patent law for peptides is weak — change one amino acid and it’s a ‘new compound’ (the same problem Lily faces with retatrutide). Millions have already used BPC via research sites, so there’s little financial incentive for a giant trial — unless crowdfunded.

[48:38] [condensed — Bakri’s internal-medicine role: admitting ER→floor→ICU patients, managing complex disease.]

[49:27] Huberman: If we funded an independent trial, what endpoints? Bakri: (1) Complete the ulcerative-colitis phase-3 (oral capsule releasing deep in the gut, not enema). (2) A GERD RCT (oral BPC vs pantoprazole) — Bakri travels with oral BPC and avoids traveler’s diarrhea (anecdote); the gut is the most vulnerable, outside-facing organ, so BPC may be part of its protective armory. (3) Neuropsychiatric/addiction (craving reduction via gut-brain axis — likely local gut signalling shutting down neurons, cf. Bohórquez’s neuropod cells → vagal nodose ganglion → dopamine). (4) Musculoskeletal: a poorly-vascularised tissue that benefits from improved blood flow — e.g. bicep/triceps tendon post-surgical outcome; BPC won’t reattach a torn ACL but might speed recovery (6 vs 12 months) — what athletes use it for.

[53:52] Huberman: One-limb-vs-other control is confounded by systemic transfer. Bakri: Right — and I’ve had results on a different injury than the one I treated. My favourite BPC story: I tore my triceps (grade 2, purple arm), expected 3-month recovery/surgery; injected BPC locally (plus a couple peptides) at a gram-range dose (much higher than microgram protocols) and my PT was stunned at ~3–4-week healing. I think common low doses are sub-therapeutic; the orthopedic group plans 250 mcg and I doubt they’ll see an effect. Huberman: I had a bad trapezius/neck pull, ~200 mcg BPC (compounding-pharmacy, labeled BPC-157 not PDA), gone in 2 days — eerily fast; then stopped. Not everyone responds; placebo can’t be excluded. Two possibilities: BPC is as good as claimed (and millions lack access), or it’s ineffective/harmful (and millions are injecting it) — both are bad endpoints, which is why we need data.

[58:29] [condensed — young men in gyms asking about BPC/testosterone; Bakri: at 18 you have all the peptides you need (cf. parabiosis/young-blood work, Tony Wyss-Coray episode); many young men take exogenous testosterone unnecessarily, risking fertility. ‘Looksmaxing’ trend.]

[01:00:04] [condensed — a third bad outcome: adverse events get blamed on BPC itself when the cause may be the vehicle/contaminant or misuse; HRT/TRT benefit from knowing what to monitor, peptides lack this; the field is ‘Wild West’; Huberman’s worry is the delivery harming people and missing a useful tool.]

[01:01:18] Huberman: As a physician, if a patient’s gut is a mess or they’re post-surgical and clean BPC could help — what position does that put you in? Bakri: Uncomfortable. On the hospital wards, recommending BPC over pantoprazole could cost my license. In clinic it can be an honest physician-patient discussion (promising, not FDA-approved, minimal/no human data, clear endpoints, monitoring). Liability: a patient can sue the physician, the compounding pharmacy, and anyone who recommended it; malpractice insurers won’t cover non-FDA-approved peptides. There are also oral BPC ‘supplements’ now (legal status unclear — RFK Jr. has implied peptides are supplements, not meds; the agency may disagree). If you eat an animal’s stomach you’d get some BPC; Khavinson identified many organ peptides (ligen, ovagen) found in desiccated-organ supplements — di/tripeptides can be orally available if the right shape/size.

[01:07:09] Huberman: Let’s talk pinealon. I’ve tried it before sleep — at lights-out it reduces deep sleep and gives far more REM; better is a small injection mid-night → REM jumps from ~1.5 to ~3 h, and it improves REM on the nights in between (lingering effect). Bakri: That matches patients. Pinealon is a tripeptide EDR, from Khavinson. Soviet research aimed to protect soldiers/astronauts (eyes, accelerated aging — submariners/astronauts return ‘aged,’ poor circadian rhythm, weak immunity). Khavinson ground up pineal and thymus extracts, reversed aging effects, then sequenced the active peptides: from pineal → epitalon; from thymus → vilon, thyrogen, etc.

[01:10:38] Huberman: But pinealon isn’t from the pineal? Bakri: Correct — epitalon is from the pineal; pinealon comes from a ground-up cortex extract (cortexin) — confusingly named. Let’s just call pinealon EDR (its 3-amino-acid sequence). Epitalon restored melatonin in aged monkeys/humans (no effect in young) by upregulating the enzyme converting serotonin → melatonin.

[01:12:18] Huberman: Known effects of EDR / am I imagining the REM increase? Bakri: Khavinson never mentions REM (no sleep trackers in 1970s USSR). EDR binds DNA-groove promoter regions to assist transcription — like a transcription-factor helper — improving brain metabolism via GDF11, SOD1/2, IRIS, PPARα/γ. He made it as an anti-stress / cognitive-performance compound; athletes given EDR sustained performance despite maximal exhaustion. Take it in the morning (or pre-workout); high enough doses sedate. Orally ~0.5–3 mg (Russian product ~200 mcg); some inject (goes systemic via PEPT transporters; likely crosses the blood-brain barrier). [warning] Some get a blood-sugar drop / A1C drop (PPAR activation) and vivid dreams. The mechanism (no clear receptor) shifts the brain’s redox/oxidative state → more daytime alertness, better HIIT, more REM.

[01:18:31] [sponsor segment omitted — Function]

[01:19:53] Huberman: Epitalon and the calcifying pineal. Bakri: Pineal research is under-studied (‘woo’ stigma) but key to aging/longevity. Calcification appears on MRI but the real question is what drives the dramatic decline in melatonin synthesis with age (pineal cysts can cause precocious puberty). Epitalon increases clock-gene expression (measurable in lymphocytes), raises morning cortisol (good — high AM, low PM), restoring a circadian-appropriate hormonal profile. The bioregulator idea: take it briefly and accrue benefits off-dose. Khavinson’s 15-year nursing-home study: one group got epithalamin (whole pineal extract) + a thymus peptide thymalin for ~10–20 days/year → significantly lower mortality (CVD, infection, cancers) — the most interesting longevity study if true (Russian-data caveat).

[01:25:22] Huberman: I’m interested in epitalon for vision — mouse data on slowing neurodegeneration in retinitis pigmentosa and glaucoma; it reaches DNA repair rather than vague downstream receptors. Bakri: It supports the genetic machinery; for eyes it seems to repair photoreceptors, and my theory is it works on melanopsin (upregulating it → morning sunlight more effective, since 5 days of darkness raises retinal melanopsin in animals).

[01:28:23] [condensed — Huberman on the premium-channel/donor-matched research funding model and unbiased replication; Bakri on medicine’s history of ‘grotesque abuses’ (e.g. irradiating children’s thymuses 1910s–40s to prevent SIDS, causing cancers) and ‘first, do no harm.’]

[01:29:36] Huberman: The thymus. Bakri: Grows under youthful hormones (melatonin, GH, DHEA) until puberty, then involutes. Sits above the heart behind the collarbone — baseball-sized in a baby, now mostly fat with thymic residue. Surgeons often remove the residue during open-heart surgery, but NEJM data show a mortality signal (more cancers/autoimmunity) within 5 years of thymus removal. It’s vagally innervated (sympathetic + parasympathetic), secretes hormones, and trains T-cells (naive T-cells → CD4/CD8) so they attack pathogens/cancer, not self. Output (~10⁸ naive T-cells/day at 15) falls with age; those cells live ~10–15 years — roughly timing the mortality window for many diseases. A 2026 Nature paper: higher thymic scores on MRI → lower mortality across CVD/cancer/etc.

[01:33:28] Bakri: Thymic involution is driven mostly by androgens, estrogens, progesterone, corticosteroids (the very hormones people boost) — castration partly reverses it; pregnancy involutes the thymus (then it regrows during breastfeeding under GH/prolactin). Regrowing it: Greg Fahy’s trial — GH + metformin + DHEA for 12 months increased thymic size on imaging, raised CD4/CD8, improved exhaustion markers (PD-1).

[01:34:50] Huberman: Thymosin alpha-1? Bakri: Thymic-family hormone (≥21 aa) — increases T-cell development/proliferation and proper pathogen targeting (‘jet fuel for T-cells’). FDA-approved as Zadaxin for thymus-deficient kids (DiGeorge etc.); abroad an adjuvant for hepatitis B/C and cancers. Sepsis/infection literature is mixed. Both thymosin alpha-1 and beta-4 came from the Goldstein lab (thymus ‘fraction 5’). I use ~2.5 mg twice weekly when travelling/on the wards as a prophylactic and didn’t get sick all season (anecdote).

[01:38:13] Huberman: TB-500 vs thymosin alpha-1? Bakri: While Khavinson found thymalin, Goldstein found fraction 5 (containing alpha-1 and beta-4). Thymosin beta-4 (43 aa) helps the actin cytoskeleton (immune cells must rearrange actin to migrate) → upregulates movement; the horse-racing community uses it for doping. Beta-4/TB-500 are made in many tissues, not just the thymus.

[01:39:23] [condensed — morning-sunlight physiology applies to dogs (melanopsin/SCN); a pet-peptide industry is emerging; vets are surprisingly open; Bakri’s family vet gave dogs testosterone with good results.]

[01:43:09] [condensed — NAD/NMN/NR: Huberman augments NAD with NMN (morning energy, thick nails/fast hair) but has never claimed it extends lifespan; the landmark paper found blood NAD doesn’t drop with age; he corrects a NYT misattribution. Thymic atrophy may underlie long-COVID; infection causes transient thymic atrophy, and modern chronic stress prevents recovery.]

[01:46:16] Bakri: Naming chaos: the French isolated the true main thymus hormone thymulin (not thymalin). Thymulin is a 9-aa, zinc-dependent peptide — the marker of thymus function, falling with age. It develops NK/T-cells and sensitises end-organs to their target hormone: e.g. in animals, hCG + thymulin → more testosterone than hCG alone. So immune status (thymulin high in youth, low in age) gates reproduction/steroidogenesis — a negative-feedback loop preventing system overrun.

[01:48:30] Huberman: Boosting thymulin? Bakri: Thymulin has a very short half-life; better to raise endogenous production — and adequate zinc is required (falling thymulin is the earliest sign of zinc depletion, before serum/RBC zinc).

[01:49:00] [sponsor segment omitted — Element]

[01:50:43] Huberman: GHK-Cu — most questions come from women (skin; topical vs oral vs inject?). Bakri: Tripeptide glycine-histidine-lysine + copper, found in type-1 collagen (skin, hair, connective tissue). Discovered by Lauren Pickart (~mid-70s); serum ~200 ng in youth → ~60 by 65. Regulates both collagen synthesis and breakdown (proper remodelling). Animal wound-repair data (like BPC). It’s the basis of the ‘glow stack’ (Wolverine BPC+TB500 + GHK). Topical has the best human data (vs retinol/vitamin-C; crow’s-feet improvement); hair data modest (adjunct to minoxidil, not a minoxidil replacement). A Chinese group studies it for lung regeneration (COPD; possibly long-COVID connective-tissue repair). Real GHK is blue (copper) — but blue ≠ proof of an intact complex. Best with red-light therapy (synergistic) and as an adjunct to skincare; don’t inject the face; injectable isn’t FDA-approved (maybe a ‘second round’ for cat-1 after BPC/TB).

[01:55:20] [condensed — thymic atrophy after every infection; modern life prevents convalescent thymic rejuvenation; we lack ‘thymologists’; post-pandemic interest in immunity; zinc + thymic peptides/extracts as possible population-level tools.]

[01:59:14] Bakri: A cheap immune/thymic proxy: the lymphocyte-to-monocyte ratio on a standard CBC (~$3). Low LMR tracks worse outcomes across CVD/cancer/diabetes; a healthy young person runs ~1,500–3,000 total lymphocytes; hazard rises below ~1,000. High lymphocytes + low monocytes = a more robust immune state. Almost no clinician reports on it (not in guidelines), but peptides have moved patients from ~4:1 to ~8:1 LMR.

[02:03:56] Huberman: Batch GH secretagogues (tesamorelin, MK-677) and melanotans — FDA-approved for some indications (short stature, burns, HIV; melanocortin for libido). People take them to lose fat, gain muscle, recover, look youthful. Real effects? Bakri: Parse them. MK-677 — oral, non-peptide GHRP, not FDA-approved; makes you ‘bleed out’ GH non-pulsatially (vs the natural first-90-min slow-wave-sleep pulse) and drives hunger. Somatopause = the GH drop in the 30s (alongside adrenopause, menopause/andropause). The debate: should we replace declining IGF-1? GH helps skin/sleep/muscle/joints and is thymo-regenerative, but there’s an oncogenic-promotion concern (growth factor for existing cancers), GH-deficient models live longer, and dog breeds with higher IGF-1 (Great Dane) live shorter than low-IGF-1 (Chihuahua). Tesamorelin + ipamorelin can push IGF-1 to ~380–390; both reduce insulin sensitivity (A1C↑) — “you have to be lean/healthy enough to take GH.”

[02:11:23] Huberman: I tried sermorelin — dramatically increased deep sleep, nuked REM, and spiked my PSA (reverted off it). Bakri: With age the prostate enlarges (BPH) under DHT/estrogen/GH; there are prostate peptides (Khavinson) being translated. The trinity stack (GLP-1 + GH secretagogue + androgen modulation ± anavar) is the celebrity body-recomposition protocol.

[02:14:35] Huberman: GLP-1s. Bakri: Semaglutide (Ozempic/Wegovy), tirzepatide (Mounjaro/Zepbound). These are transforming medicine — our system was set to collapse under obesity/pre-diabetes/diabetes. Origin: GLP-1 from Gila-monster saliva (too short-acting natively; pharma extended half-life). Weight is a brain set-point computed from gut hormones (GIP, GLP, glucagon, insulin, sex steroids); GLP-1 is a strong ‘don’t-eat’ signal. Stopping at max dose → weight often regains. Microdosing + lifestyle (‘training wheels’) minimises the depression/low-drive and GI complaints people report (often from too-high doses, under-eating, low electrolytes/micronutrients). My own 1 mg Ozempic self-test (skipping titration) → projectile vomiting through a night shift — titrate slowly. GLP-1 also lowers alcohol cravings (dopaminergic), improves fertility via leptin normalisation, and has possible cognitive effects (confounded by social/appetite changes vs direct brain receptors).

[02:33:48] Huberman: Retatrutide. Bakri: Triple agonist (GLP-1 + GIP + glucagon), 39 amino acids; Lily is trying to classify it as a biologic (>40 aa) for a much longer patent and to block compounding (similar to what happened with hCG compounders). Trump-era pricing pressure has dropped some costs (~$300/motirzepatide;oral\ast \ast orforglipron\ast \ast$150/mo — a non-peptide GLP-1). Bakri’s nomenclature point: ‘peptide’ is too broad to be functionally meaningful — what matters is which receptor it touches (orforglipron is closer to semaglutide than BPC is).

[02:37:23] [condensed — Huberman proposes Bakri lead a peptide nomenclature committee (like gene-naming): split by receptor/no-receptor, regenerative vs immunogenic peptides, etc., for public clarity.]

[02:39:17] Huberman (audience Qs): Endometriosis/fibroids — can BPC help/hurt (angiogenesis)? Bakri: No animal/human data; those are hormonal/metabolic — the hormonal lever is far stronger; no reports either way. TBI: Russian cortexin/cerebrolysin studies exist (unlikely in the US); no good BPC-TBI data, but plausible via anti-stress. BPC’s neuro effects are homeostatic — in mice it blunts both highs and withdrawals (can’t get too drunk/high, doesn’t withdraw); a homeostatic gut-brain mechanism may explain the anhedonia reports (a ‘rest-and-digest/convalescent’ mode).

[02:41:54] Huberman: Where should peptide-curious people look / not look? Bakri: Most get them from research-only websites — not reliable (unknown contents/batch). Over the next 6–24 months expect physician-led telehealth options (price competition → clarity on which compounders are good). Get a physician (ideally peptide-educated — doctors must now get educated), be monitored with clear endpoints (e.g. don’t take tesamorelin without checking IGF-1; GLP-1 needs counselling against over-loss). And basics first — morning sunlight, sleep, minimally-processed food — before stacking peptides.

[02:44:35] [closing — thanks; Bakri is 33; new circadian-biology app; book/newsletter/social outro omitted.]