What Is Marine Collagen? Source, Dose, Allergy and Evidence

Marine collagen is type I collagen taken from fish skin and scales, broken down by food enzymes into short peptides and dried to a powder. Fish collagen has randomised trials of its own: 5 to 10 g a day for skin readings, 2 g for body fat and 5 g for bone markers, using skin from tuna, catfish and skate.^{6, 7, 12, 13} No trial has compared marine with any other source on a result, and absorption was comparable in one supplier-funded study of six; marine vs bovine collagen has every "versus" answer.^{4, 14} It is a fish product, labelled as a fish allergen, and fish collagen is itself a registered allergen, so a fish allergy rules it out.^{3, 15} Part of the Wellgard collagen guide.

Sources checked on 29 October 2026.

What is marine collagen made from, and how?

The raw material is the skin and scales left when fish are processed for food, and their collagen is type I (types of collagen explained). Fish collagen holds less hydroxyproline than mammalian collagen; its triple helix comes apart at about 29 °C rather than 40 to 41 °C, so the raw material is handled cold.¹ Then the standard process: skins defatted (fatty skins such as salmon repeatedly, or the powder keeps a fishy smell), collagen extracted as gelatine, a food-grade enzyme cutting it into short peptides, filtering and spray drying.² What collagen peptides are explains how the enzyme, not the species, sets size and taste.

Is marine collagen absorbed better?

No. A supplier-funded 2024 randomised crossover gave six healthy adults fish, porcine and bovine hydrolysates and found comparable uptake of hydroxyproline and its peptides from all of them.⁴ Size changes speed, not amount: in 15 men a 500 dalton product peaked at 30 minutes against two hours for a 2,000 to 3,000 dalton hydrolysate.^{4, 5}

What did the marine collagen trials measure?

The 1,000 mg results belong to tripeptide-enriched products, not a standard hydrolysate.^{10, 11} The six-month bone trial is shorter than the 12-month trials with other sources.¹³ Pooling every source, a 2023 meta-analysis of 26 trials found elasticity did not differ by source.¹⁴ Collagen has no authorised health claim in Great Britain, so these are trial results, not product claims.²³

How much marine collagen per day did the trials use?

There is no separate marine dose. The trials above used 5 g a day for skin in four cases and 10 g in one, 2 g for body fat and 5 g for bone markers, inside the 2.5 to 15 g range collagen trials have used.^{6, 7, 8, 9, 12, 13} How much collagen should you take a day has the ranges by outcome.

Can you take marine collagen if you are allergic to fish?

No. Fish is one of the 14 allergens that must be declared on a food or supplement label in Great Britain; the exemptions for fish gelatine used as a vitamin carrier or as a fining agent do not cover collagen peptides.³ Fish collagen is an allergen in its own right: 21 of 101 fish-allergic patients were sensitised to it, and salmon and barramundi collagen are registered allergens.¹⁵ In Japan, half of fish-allergic patients had IgE to mackerel collagen, with 87 to 98% cross-reactivity across 22 species.¹⁶ Two case reports describe anaphylaxis after fish-collagen gummies, one in a nine-year-old girl confirmed by an oral fish-collagen challenge.^{17, 18} A fish allergy therefore rules marine collagen out in every format. The NHS says to see a GP if you think you or your child may have a food allergy, and to call 999 if your lips, mouth, throat or tongue suddenly swell or you are struggling to breathe.¹⁹

Does marine collagen contain mercury or other contaminants?

In a 40-fish study, mercury was highest in kidney, intermediate in muscle and liver, and lowest in brain, skin and scales.²⁰ A 2017 survey found farmed Atlantic salmon lower than wild in dioxins, PCBs, pesticides and mercury, all well below EU maximums.²¹ The law sets the same heavy-metal limits for every collagen sold as food whatever the animal; the pregnancy page lists them. Ask the brand for its batch certificate: which laboratory, which metals, what limit of detection.

Does marine collagen raise calcium, and is it from shellfish?

Neither. It comes from finned fish, not from prawns, crabs or shells. The high-calcium case reports concern cancer patients taking calcium, vitamin D or shark cartilage supplements, none collagen peptides.²² In the six-month trial of 5 g of fish collagen, serum calcium stayed stable in the collagen-only arm.¹³ If a marine or multi-collagen blend adds calcium or vitamin D, that is the ingredient to read.

Why does marine collagen cost more, and what should you check before buying?

Thin skins and scales give a low yield, and cold handling and defatting add cost.^{1, 2} What to check: the species and part declared, not just "marine"; grams of collagen per serving; molecular weight as a range held batch to batch; a batch certificate of analysis for heavy metals; the fish allergen declaration; and a single ingredient. How to choose a collagen supplement works through each.

Is marine collagen halal, vegetarian or pescatarian?

Fish needs no slaughter ritual, so marine collagen is generally accepted as halal; is marine collagen halal has the ruling. It is animal tissue, so not vegetarian or vegan; it suits pescatarians. Is there a vegan collagen explains what plant-based products contain.

What we don't know

No trial has compared marine collagen with another source on any outcome; the only pooled comparison is a subgroup split of 26 skin trials.¹⁴ No fish-only hydrolysate has a placebo-controlled joint trial, and the one bone trial ran six months with bone markers unchanged.¹³ No skin trial has run a year. No peer-reviewed survey of contaminants in finished marine collagen peptides exists. These are gaps in the evidence, not findings against marine collagen.

Frequently asked questions

What is marine collagen good for?

Skin readings in trials of 31 to 72 people at 5 to 10 g a day over four to twelve weeks, and in two company-run trials at 1 g; body fat in one 90-person trial at 2 g; hair in two trials, shaft diameter at 1 g and shedding at 5 g.^{6, 7, 8, 9, 10, 11, 12, 13}

Is marine collagen better than bovine collagen?

No trial has compared them on any outcome, and skin elasticity did not differ by source in the 2023 meta-analysis.^{4, 14}

Does marine collagen work?

In its own randomised trials, 5 to 10 g a day improved measured skin hydration, elasticity and wrinkle scores over eight to twelve weeks, and 2 g a day lowered body fat in overweight adults in one trial.^{6, 7, 12}

What is the best marine collagen?

The one whose label passes the checks above; no brand name belongs in that answer.

Can you take marine collagen when pregnant?

No trial of collagen peptides of any source has been run in pregnancy or breastfeeding, so nobody can promise it is safe; the pack asks you to consult your GP. Collagen in pregnancy has the NHS wording on fish and mercury.

Sources

1. Hickman D, et al. Journal of Biotechnology, 2000. Laboratory study; funding not stated on PubMed. PubMed 10867185.
2. Nilsuwan K, et al. Foods, 2021. Laboratory processing study; no conflict declared. PubMed 34681306.
3. Regulation (EU) No 1169/2011 (assimilated GB law), Annex II. legislation.gov.uk, checked 29 October 2026.
4. Virgilio N, et al. Frontiers in Nutrition, 2024. Supplier-funded double-blind crossover, six volunteers. PubMed 39149544.
5. Nergiz-Unal R, et al. PeerJ, 2026. Single-blind study, 15 men; manufacturer-employed authors. PubMed 42609988.
6. Morakul B, et al. Journal of Cosmetic Dermatology, 2024. Randomised trial, 72 women, 8 weeks; part supplier-funded. PubMed 39075819.
7. Evans M, et al. Journal of Cosmetic Dermatology, 2021. Randomised trial, 50 women, 12 weeks; supplier-linked. PubMed 32799362.
8. Sangsuwan W, Asawanonda P. Journal of Dermatological Treatment, 2021. Randomised trial, 36 women, 4 weeks; funding not stated on PubMed. PubMed 32009486.
9. Koizumi S, et al. Bioscience, Biotechnology, and Biochemistry, 2023. Randomised trial, 31 adults, 12 weeks; manufacturer-employed author. PubMed 37245058.
10. Kim DU, et al. Nutrients, 2018. Randomised trial, 64 adults, 12 weeks; company-employed authors. PubMed 29949889.
11. Hwang S, et al. Journal of Medicinal Food, 2026. Randomised trial, 114 women, 24 weeks; company-employed author. PubMed 41788055.
12. Tak YJ, et al. Marine Drugs, 2019. Randomised trial, 90 overweight adults, 12 weeks; no external funding declared. PubMed 30866485.
13. Duangjai A, et al. Clinics and Practice, 2025. Randomised trial, 80 menopausal women, 6 months; no conflicts declared. PubMed 41002783.
14. Pu SY, et al. Nutrients, 2023. Meta-analysis, 26 randomised trials, 1,721 people; no conflicts declared. PubMed 37432180.
15. Kalic T, et al. Journal of Allergy and Clinical Immunology: In Practice, 2020. Laboratory study, 101 fish-allergic patients; funding not stated on PubMed. PubMed 32389794.
16. Kobayashi Y, et al. Allergy, 2016. Laboratory study; funding not stated on PubMed. PubMed 26785247.
17. Ueno R, et al. Asia Pacific Allergy, 2020. Case report; no conflicts declared. PubMed 33178560.
18. Abe N, et al. Allergology International, 2020. Case report; funding not stated on PubMed. PubMed 31262631.
19. NHS. Food allergy. nhs.uk, Page last reviewed 20 August 2026.
20. Burger J, et al. Science of the Total Environment, 2013. Tissue survey, 40 fish; funding not stated on PubMed. PubMed 23202378.
21. Lundebye AK, et al. Environmental Research, 2017. Contaminant survey; funding not stated on PubMed. PubMed 28189073.
22. Lagman R, Walsh D. Supportive Care in Cancer, 2003. Case series, eight patients; funding not stated on PubMed. PubMed 12673461.
23. Great Britain nutrition and health claims (NHC) register. gov.uk, last updated 19 May 2026.

Spotted something out of date? Email support@wellgard.co.uk with the page address and the sentence.