The main peptide in honeybee venom — powerful in a dish, indiscriminate in blood, and never tested in a controlled human trial as an isolated compound.
Melittin is a 26-amino-acid peptide (sequence GIGAVLKVLTTGLPALISWIKRKRQQ, C-terminally amidated) and the single most abundant component of honeybee venom — roughly half of the venom's dry weight, with published estimates running from about 40 to 60 percent. It is the molecule most responsible for what a bee sting does: it punches holes in cell membranes.
That one property explains both halves of melittin's story. A peptide that disrupts membranes will kill bacteria, fungi, enveloped viruses and cancer cells in a dish — and it will also rupture red blood cells and trigger pain, because it cannot tell a target membrane from a host membrane. Nearly everything interesting and everything dangerous about melittin follows from that.
Two identity points worth fixing before anything else. First, melittin is not bee venom: venom is a mixture that also contains phospholipase A2 (the major venom allergen, Api m 1), apamin, hyaluronidase and other components; melittin itself is a minor allergen designated Api m 4. Second, almost all human trial data in this area is about whole venom, usually injected at acupuncture points — a different exposure from isolated melittin, and this guide keeps the two separate throughout.
Not approved as a drug anywhere. No controlled human trial of isolated melittin exists. Research chemical and preclinical drug-delivery candidate.
Melittin is amphipathic: one end of the molecule is water-shy, the other water-friendly. In contact with a lipid membrane it folds into a helix, inserts, and aggregates into pores that leak the cell's contents — the mechanism reviewed in detail by Raghuraman and Chatopadhyay (2007, doi:10.1007/s10540-006-9030-z). This is a physical attack, not a receptor interaction, which is why it works on such a wide range of cells and microbes — and why selectivity is the field's central problem.
Melittin is the venom's major pain-producing substance. It activates phospholipase A2, and the resulting lipid cascade sensitises the TRPV1 pain channel — the same channel heat and capsaicin act on (Chen et al. 2016, PMID 26983715). The sting hurts largely because of this peptide.
Well below the concentrations that rupture membranes, melittin interacts with inflammatory signalling: in cell work it binds the p50 subunit of NF-κB (dissociation constant around 1.2×10⁻⁸ M) and suppresses NF-κB-driven inflammatory gene expression (Park et al. 2004, PMID 15529353). This low-dose anti-inflammatory pharmacology is real but concentration-dependent — it lives in a narrow window below the membrane damage.
Melittin's reported effects cluster in the low micromolar range — and so does its damage to human cells. Haemolysis (rupture of red blood cells) reaches its half-maximal effect at roughly 1.2 to 1.7 µM depending on the assay, squarely overlapping the concentrations used in the antimicrobial, antiviral and anticancer experiments. There is no known concentration at which free melittin is broadly active against target cells and inert toward blood. Every serious research programme on this molecule is, underneath, an attempt to engineer around that overlap.
| Claimed benefit | Evidence |
|---|---|
| Anticancer activity | Consistent in cell culture; positive and negative animal results both published (next section). No human data. Selectivity, not potency, is the open question. |
| Antiviral (HIV) | In vitro only. Free melittin inactivated HIV at ID₅₀ 0.9–1.5 µM (PMID 9568968) — the same range that damages host cells; nanoparticle formulations are the attempted fix. |
| Anti-inflammatory / arthritis | Animal models with whole venom and synthetic melittin (PMIDs 15529353, 21061459). Human trials exist only for whole-venom acupuncture and are low quality (see below). |
| Antimicrobial | Broad in-vitro membrane activity, inseparable from host-cell toxicity for the free peptide. |
| Pain relief | Paradoxical claim: melittin is itself the venom's main pain-producing component (PMID 26983715). Whole-venom acupuncture trials report pain outcomes, with the caveats in the next sections. |
Build your protocol, log every dose, monitor your body's response, and get reminders so you never miss a dose.
Start Tracking FreeMelittin's anticancer literature is the part most often quoted and most often quoted selectively. Here is what the strongest papers actually found, including the result that did not work.
The 2020 Western Australian study (Duffy et al., npj Precision Oncology, PMID 32923684). Researchers at the Harry Perkins Institute tested honeybee venom and melittin against breast cancer cell lines. Venom killed triple-negative (SUM159) and HER2-enriched (SKBR3) cancer cells at IC₅₀ values of 5.58 and 5.77 ng/µL respectively, while normal breast-derived fibroblasts (HDFa) required 22.17 ng/µL — roughly a four-fold window in a dish. In a mouse allograft model, melittin at 5 mg/kg combined with docetaxel at 7 mg/kg, dosed every other day for seven doses, outperformed either agent alone. That is a genuinely interesting preclinical result. It is also cell lines and one mouse model: no human has received melittin as a cancer treatment in a controlled trial.
The nanoparticle study that cuts both ways (Soman et al. 2009, PMID 19726870). Melittin loaded onto perfluorocarbon nanoparticles and given intravenously at 8.5 mg/kg for four doses shrank B16F10 melanoma tumours in mice to 0.23 g versus 1.87 g for saline — a dramatic effect, in five animals. In the same paper, a second tumour model (MDA-MB-435) responded far more weakly, with tumour reduction of only about 25 percent (n=6). Both numbers are the finding: delivery-protected melittin can be powerfully active in one model and modest in another, and the published record does not let you quote only the first.
Why the field talks about delivery, not dose. Free melittin injected into blood meets red blood cells first. The entire modern literature — nanoparticles, targeted conjugates, masked pro-peptides — exists because the molecule's potency was never in doubt and its selectivity always was. A claim that melittin "kills cancer cells" without the delivery caveat is a claim about a petri dish.
When people say melittin has "clinical evidence," they almost always mean bee venom acupuncture (BVA) — diluted whole venom injected at acupuncture points, practised mainly in South Korea and China. Whole venom is not melittin, the doses are far below the concentrations in the cell studies, and the trial quality deserves a direct look.
A 2025 meta-analysis pooling 20 randomised trials (PMID 40295986) reported a statistically significant pain benefit for BVA over sham on one pooled outcome (mean difference −16.93, 95% CI −26.35 to −7.51) — while flagging substantial heterogeneity between trials, and with internal inconsistencies in how the abstract labels its own scales. An earlier systematic review focused on rheumatoid arthritis (Lee et al. 2014, PMID 25380812) screened 304 records and found exactly one eligible randomised trial comparing BVA with conventional acupuncture, rated low quality. Animal arthritis work exists for both whole venom (0.8 and 1.6 µg/kg in rats) and synthetic melittin injected at an acupuncture point (PMIDs 15529353, 21061459), but animal arthritis models are where promising anti-inflammatories are common and human successes are rare.
The safety side of the BVA literature is better quantified than the efficacy side. A systematic review of adverse events (Park et al. 2015, PMID 25996493) found bee venom therapy carried a relative risk of adverse events of 3.61 (95% CI 2.10 to 6.20) versus control. A large practice-based analysis put anaphylaxis incidence at 0.045% of treatments (95% CI 0.028 to 0.062; PMID 35448847) — uncommon, but this is a treatment whose worst-case event is life-threatening, and a published case report describes a fatal anaphylaxis following live-bee sting apitherapy (PMID 29461208).
The fair summary: whole-venom injection has a human trial literature — thin, heterogeneous, mostly small trials — and a documented, non-trivial risk profile. Neither half of that sentence transfers to isolated melittin, for which the human literature is empty.
Pre-filled with a typical Melittin (Bee Venom Peptide) setup. Edit any field — the draw updates live.
Insulin syringe — 100 units = 1 mL
Free account. Saves your reconstitution + schedules doses + tracks every vial.
Dosing cheat sheet, reconstitution reference, and cycle planning — delivered to your inbox.
Melittin-loaded nanoparticles inhibited HIV's CXCR4 and CCR5 co-receptor pathways at IC₅₀ values of 2.4 µM and 3.6 µM in cell culture (Hood et al. 2013, PMID 22954649). In the same work, free melittin damaged VK2 vaginal epithelial cells at every concentration tested — the unformulated peptide could not be used at all. Follow-up work (Jallouk et al. 2014, PMID 24748389) found free melittin toxic to epithelial cells above 2 µM, while the nanoparticle formulation was non-toxic at 10 µM and only became toxic at 20 µM (epithelial cells) to 40 µM (sperm cells). Encapsulation bought roughly a five-to-ten-fold safety window. It did not create an approved product, and a decade later none exists.
For melittin to become a medicine, a delivery system must hold it inactive in blood, release or expose it preferentially at the target, and do so at a scale and cost a drug can carry — in humans, not just in mice. That is a legitimate and active research programme. It is also, as of this guide's date, unfinished: no melittin formulation has entered a controlled human trial.
Melittin's hazards are not idiosyncratic side effects — they are its mode of action applied to the wrong cells. Haemolysis: red blood cell rupture at low-micromolar concentrations overlapping its studied active range. Pain and local tissue damage: melittin is the venom's principal pain-producing component, acting through phospholipase A2 and TRPV1 (PMID 26983715). Allergy: melittin is the Api m 4 allergen; the venom's major allergen is phospholipase A2 (Api m 1), so whole-venom exposure carries both. Anaphylaxis: documented at 0.045% of bee venom treatments in practice data (PMID 35448847), with at least one published fatal case in apitherapy (PMID 29461208). Anyone with a bee sting allergy should treat this entire category as contraindicated.
There is no human safety dataset for isolated melittin at any dose, by any route. No established dose, no pharmacokinetics, no drug-interaction data. Material sold as research-grade melittin is manufactured for laboratory use, and its purity and content are whatever its certificate says — see the Basics guide on what a COA does and does not prove. This guide deliberately carries no dosing protocol section: there is no literature-reported human dosing to summarise, and inventing one would be fabrication.
Melittin (Bee Venom Peptide) is a research peptide not approved by the FDA for human use. It is sold only as a research chemical, and StackTrax does not endorse or facilitate personal use.
Quality varies enormously among research-chemical suppliers. At minimum, look for:
Lyvn does not carry Melittin (Bee Venom Peptide). They are the vendor we recommend elsewhere on this site, and they publish the full independent panel per batch — which makes their catalogue a useful yardstick for the criteria above even on a compound they do not stock. We are not recommending a specific supplier for Melittin (Bee Venom Peptide) itself, because we have no basis to.
For research use only. Not for human consumption. 21+.
Build your protocol, log every dose, monitor your body's response, and get reminders so you never miss a dose.
Start Tracking FreeNot as an isolated compound in any controlled trial. The human trial literature is about whole bee venom, usually injected at acupuncture points (bee venom acupuncture) — a mixture in which melittin is one component among many, at far lower exposures than the cell-culture studies. Those trials are small and heterogeneous, and a 2014 systematic review of the rheumatoid arthritis literature found only one eligible randomised trial, rated low quality (PMID 25380812).
No. Melittin is not approved as a drug by the FDA or any other major regulator, for any indication. It is sold as a research chemical for laboratory use. Bee venom therapy is practised as a traditional/complementary procedure in some countries, which is not the same as an approved melittin product.
There is no human evidence that it can. The strongest published results are preclinical: selective killing of breast cancer cell lines at roughly a four-fold window over normal cells and a benefit combined with docetaxel in one mouse model (PMID 32923684), and dramatic tumour shrinkage in one mouse melanoma model alongside a much weaker result in a second model in the same paper (PMID 19726870). Melittin's unsolved problem is selectivity in a body, not potency in a dish.
Because its mechanism is indiscriminate membrane disruption. It ruptures red blood cells at low-micromolar concentrations that overlap its studied active concentrations, it is the venom component most responsible for sting pain (PMID 26983715), and venom exposure can trigger anaphylaxis — measured at 0.045% of bee venom treatments in a large practice analysis (PMID 35448847), with a published fatal apitherapy case (PMID 29461208). Natural origin says nothing about safety; this molecule evolved to damage tissue.
Bee venom is a mixture — melittin (roughly half the dry weight), phospholipase A2, apamin, hyaluronidase and more. Bee venom therapy injects the diluted whole mixture, historically at acupuncture points. Melittin research uses the single purified peptide, mostly in cell culture or nanoparticle formulations. Evidence and risks do not transfer cleanly between the two: the major venom allergen is phospholipase A2 (Api m 1), not melittin (Api m 4), and whole-venom trials cannot establish what isolated melittin does in humans.
Because free melittin cannot distinguish target cells from blood and tissue cells. Encapsulation holds it inactive until delivery: in one study, free melittin damaged vaginal epithelial cells at every concentration tested while the nanoparticle form inhibited HIV pathways at 2.4 to 3.6 µM (PMID 22954649), and in follow-up work the formulated peptide was non-toxic at 10 µM where the free peptide was toxic above 2 µM (PMID 24748389). The approach widens the safety window in the lab; no formulation has reached human trials.
Disclaimer: This guide is for educational and informational purposes only and is not intended as medical advice, diagnosis, or treatment. The compounds discussed are not FDA approved for human use. Always consult a qualified healthcare provider before starting any new supplement or peptide protocol. StackTrax does not sell peptides or supplements directly — purchase links go to third-party vendors. StackTrax is not responsible for the products, quality, or business practices of any third-party vendor.
© 2026 StackTrax, LLC. All rights reserved.
Privacy · Terms · Do Not Sell or Share My Personal Information
StackTrax guides cover peptides and compounds that are not FDA-approved for the uses discussed. The dosing, reconstitution, and safety information is compiled from published research and community protocols for educational purposes only.
Before using any compound mentioned here, consult a qualified healthcare provider. StackTrax does not sell, prescribe, or recommend these substances for personal use.