Foods High in Collagen: Gelatin, Bone Broth, and What the Trials Found

What collagen is Foods high in collagen Bone broth vs powder Gelatin vs hydrolyzed The amino acid balance What the trials measured Can men take collagen? How the studies used it FAQ Sources

The foods highest in collagen are the parts of the animal most people throw away: skin, tendons, cartilage, bone, and the gelatin that cooks out of them. Chicken skin, pork rinds, fish skin, oxtail, chicken feet, knuckle bones, and a long-simmered broth that sets firm in the fridge all deliver it. Plant foods contain no collagen at all. This article covers which foods carry the most, how bone broth compares with powders, the difference between gelatin and hydrolyzed collagen, what Ray Peat argued about balancing muscle meat with gelatin, and what the human trials on skin, tendons, and muscle measured.

Key takeaways

• Collagen is concentrated in skin, connective tissue, cartilage, and bone. Muscle meat carries little of it, and plants carry none.
• Home-made bone broth varies widely in collagen amino acids from batch to batch, and lab-tested batches came in below the 20 g reference dose used in supplement trials.³
• Gelatin and hydrolyzed collagen have the same amino acids. Hydrolyzed collagen dissolves in cold liquid and does not gel; gelatin sets.
• Glycine makes up about one third of collagen, and researchers have calculated that the body cannot make enough of it to cover collagen turnover.¹
• Peat's case for gelatin rested on offsetting the tryptophan and cysteine in muscle meat; the methionine-versus-glycine argument comes from rodent lifespan research.^2,4
• Human trials used 2.5 g to 15 g a day and measured skin elasticity, joint pain in athletes, Achilles tendon function, and muscle mass in men. Most were funded by a gelatin manufacturer.

This article is for education only. It does not diagnose, treat, or replace care from your physician. I am a clinical herbalist, not a medical doctor. Do not start or stop any medication or supplement based on what you read here. Collagen and gelatin products are protein foods, and people with kidney disease, gout, protein-restricted diets, or allergies to fish, beef, pork, or eggs should review any addition with a pharmacist or clinician first. Individual responses vary. If you are in crisis, contact your local emergency number or a crisis line.

What collagen is and why glycine sets the limit

Evidence: strong for the biochemistry; the glycine-shortfall calculation is a modeling paper, not a human trial.

Collagen is the structural protein of the animal body. It forms the fibers in skin, tendons, ligaments, cartilage, bone, blood vessel walls, and the gut lining, and it accounts for more than a quarter of all protein in a human.¹ Every third position in the collagen chain is glycine, and the chain is also unusually rich in proline and hydroxyproline. Peat put the figures at 35 percent glycine and 21 percent proline plus hydroxyproline, with alanine at 11 percent.²

Glycine is classed as a non-essential amino acid because the body can make it from serine. In 2009 a team of biochemists worked through the numbers and found a gap: the enzyme that converts serine to glycine can only produce as much glycine as the body has one-carbon units to spare, and that ceiling sits well below what daily collagen turnover requires. They estimated a shortfall of about 10 g of glycine a day in an adult on a typical diet, and argued that glycine should be treated as conditionally essential.¹ That gap is the reason glycine-rich foods are the subject of this article rather than protein in general.

Cooking converts collagen into gelatin. The triple helix unwinds in hot water, the protein dissolves, and on cooling it sets into a gel. A stock that turns solid in the refrigerator is a stock that pulled a meaningful amount of collagen out of the bones and skin.

What foods contain high collagen?

Evidence: strong. Collagen distribution in animal tissue is settled anatomy; the broth measurements come from one laboratory analysis.

Collagen sits in connective tissue, so the collagen content of a food tracks how much skin, sinew, cartilage, and bone it includes. Lean muscle carries a small amount. A boneless, skinless chicken breast or a trimmed steak is mostly muscle protein with a different amino acid profile, heavier in tryptophan, cysteine, and methionine and light in glycine.

Skin

Chicken skin, pork skin (rinds and cracklings), fish skin, and the skin left on a roast are the most accessible collagen foods. Porcine skin was one of the raw materials in the 2005 Kyoto study that first identified collagen peptides in human blood after a meal.⁵

Tendons, cartilage, and joints

Oxtail, beef shank, chicken feet, chicken wings and drumsticks eaten to the bone, pork hocks, trotters, and knuckle bones all carry tendon and cartilage. Chicken feet and cartilage were the other two raw materials in that same Kyoto study.⁵ Traditional dishes built on these cuts, such as fish head soup, head cheese, chicken-foot soup, and Mexican chicharrón, are the examples Peat used when he described eating "the whole animal."²

Bone and broth

Marrow bones, knuckles, necks, and a whole carcass simmered for hours release collagen into the liquid. The gel test is the simplest measure: if the cooled broth sets firm, it is rich in gelatin; if it stays thin, most of the collagen is still in the bones or was never there.

Gelatin itself

Plain gelatin powder or sheets are purified collagen, usually from pork or beef hide. Gelatin desserts, panna cotta, aspic, marshmallows made with gelatin, and gummies count as collagen foods, provided they are made with gelatin rather than pectin or agar.

What does not contain collagen

No plant contains collagen. Fruits, vegetables, legumes, grains, seeds, and mushrooms supply some of the amino acids and the vitamin C the body needs to build collagen, but not the protein itself. Egg whites and dairy carry small amounts of glycine and no collagen. Products sold as "vegan collagen" are either plant proteins or precursors, not collagen.

Bone broth vs collagen powder

Evidence: moderate. One published laboratory analysis of broth; absorption data from small human feeding studies.

Bone broth is the traditional collagen food, and a well-made batch is a good source. The problem is consistency. In 2019 researchers at the Australian Institute of Sport analyzed commercial and home-made bone broths for glycine, proline, hydroxyproline, hydroxylysine, lysine, and leucine. Broth made to a standardized recipe came in significantly lower in every one of those amino acids than a 20 g serving of the collagen supplements used in research. Broths made without a fixed recipe varied widely from batch to batch, with café-made versions testing highest and commercial shelf broths testing lowest.³ The authors concluded that broth is unlikely to deliver a reliable amount of collagen precursors, and that anyone trying to match a research dose would need a supplement.

That finding does not make broth pointless. It means the amount you get depends on the bones, the ratio of bones to water, the simmer time, and whether skin and feet went into the pot. A gelatinous broth made with plenty of joints and skin will land far higher than a thin stock from a few marrow bones. Broth also brings minerals, fat, and the flavor that makes a person eat it every day, which no powder does.

Powders win on measurement. A scoop delivers a known number of grams, which is why the trials discussed below all used powders or a liquid concentrate rather than food. For a reader who wants a fixed daily amount, a powder is the tool. For a reader who wants collagen inside a meal, broth, skin-on cuts, and gelatin cooking are the tools, and the two approaches are not in conflict.

Whether collagen from food reaches the bloodstream in a usable form was answered in 2005. Volunteers in Kyoto drank gelatin hydrolysates made from pork skin, chicken feet, or cartilage after an overnight fast. Peptides containing hydroxyproline, an amino acid found almost only in collagen, rose in their blood and peaked at 1 to 2 hours, then fell by half at 4 hours. The main peptide was prolyl-hydroxyproline.⁵ That study was co-authored by researchers at a meat-processing company, so it carries an industry connection, but the finding has been repeated by independent groups since.

Gelatin vs hydrolyzed collagen: what the label words mean

Evidence: strong. This is food chemistry, not a health outcome.

Gelatin and hydrolyzed collagen (also sold as collagen peptides or collagen hydrolysate) start from the same raw material and contain the same amino acids in the same proportions. The difference is chain length.

Gelatin

Gelatin is collagen that has been cooked and dried. The protein chains are long enough to tangle into a gel when the liquid cools below about 95°F (35°C). It needs hot liquid to dissolve, or a cold "bloom" in water before heating. Gelatin is the form used in cooking, in gummies, and in the 2017 collagen-synthesis study at UC Davis discussed below.⁶

Hydrolyzed collagen

Hydrolyzed collagen has been treated with enzymes to cut the chains into short peptides. It dissolves in cold water, coffee, or juice, does not gel, and has almost no taste. Nearly every commercial "collagen peptide" product is this form, and most of the clinical trials on skin, joints, and muscle used it. Because the peptides are already short, hydrolyzed collagen reaches the blood slightly faster than gelatin, though both are digested to the same small peptides and free amino acids.

Which one to use

For cooking, setting a dessert, or a bedtime cup of warm broth, gelatin. For stirring into a cold drink, hydrolyzed. For the amino acid balance Peat described, either one, since the glycine, proline, and hydroxyproline content is the same. Read the source line on the label: beef hide, pork skin, fish skin, and chicken cartilage each produce a product with a slightly different peptide mix, and the fish and chicken versions are the ones to check for anyone with those allergies.

Ray Peat on gelatin: balancing the amino acids in muscle meat

Evidence: mixed. Peat's mechanism is drawn from physiology and older literature; the lifespan data are from rodents; there is no human trial testing the muscle-meat-plus-gelatin ratio directly.

Peat's essay "Gelatin, stress, longevity" made a specific argument, and it is worth stating it as he wrote it rather than as it gets summarized. Collagen makes up about half the protein in an animal's body, but a much smaller share of the active tissues we eat as meat: muscle, liver, and brain. When a diet consists only of muscle meat, the amino acids entering the blood are heavy in tryptophan, cysteine, and methionine and light in glycine. Peat described that pattern as the same mix produced by severe stress, when cortisol breaks muscle down for fuel, and he attributed several downstream effects to it: tryptophan feeds serotonin production, and he held that both tryptophan and cysteine suppress thyroid function and mitochondrial energy output.²

Gelatin, in his framing, corrects the ratio. It contains no tryptophan, only small amounts of cysteine, methionine, and histidine, and a large share of glycine, proline, and alanine. He cited rodent studies in which restricting tryptophan alone, or cysteine alone, extended lifespan more than most calorie-restriction studies had, and reasoned that using gelatin as a major dietary protein is a practical way to restrict those amino acids without restricting protein.² His preference was for gelatin over isolated glycine, since alanine and proline share some of glycine's cell-protective actions and whole tissues carry other useful compounds along with the collagen. He wrote that he had recommended gelatinous broth at bedtime for years before he tried commercial gelatin himself, and that several grams before sleep let him sleep through the night for the first time in a long period.² That last point is his own account and should be read as one.

Where methionine enters

The version of this argument most people encounter online pits glycine against methionine specifically. That framing comes from the aging research rather than from Peat's essay. Diets low in methionine extend the lifespan of rats and mice, and a small 2011 rat study suggested that adding glycine to a normal diet produced a similar effect. In 2019 the National Institute on Aging's Interventions Testing Program tested that idea in genetically diverse mice at three separate sites. An 8 percent glycine diet extended median lifespan by 4 to 6 percent in both sexes, raised maximum lifespan, and reduced deaths from lung adenocarcinoma, with no increase in any of the 40 pathologies the necropsies tracked.⁴ The effect was small, it was in mice, and the mechanism was not established. The authors framed it as support for the idea that shifting dietary amino acid ratios can affect healthy lifespan.

Put together, Peat's physiology and the rodent lifespan work point in the same direction: a diet built on muscle meat alone leaves glycine short relative to methionine, tryptophan, and cysteine, and adding the collagen-rich parts of the animal restores a ratio closer to what a whole carcass provides. The human trial that would settle it has not been run. Anyone applying the idea is working from mechanism, animal data, and tradition, which is a reasonable basis for a food choice and a thin one for a medical claim.

This is one piece of a larger framework. The Ray Peat diet guide lays out his food list and hormone model in order, the 7-day bioenergetic meal plan shows how gelatin and broth fit into a week of meals, and the sourdough article covers the grain side of the same kitchen.

Start here The Ray Peat Diet Guide Every food he mentioned, his hormone framework in order, and links to his interview audio transcripts, and articles. Open the guide →

What the collagen trials measured

Evidence: moderate. Randomized, placebo-controlled human trials exist for skin, joints, tendons, and muscle, but most are small and most were funded by Gelita, the German gelatin manufacturer.

The marketing around collagen runs far ahead of the research, so it helps to know exactly what was tested, in whom, at what amount, and who paid.

Collagen synthesis

The cleanest mechanism study came from Keith Baar's lab at UC Davis in 2017. Eight healthy men drank a placebo, 5 g, or 15 g of vitamin C-enriched gelatin one hour before six minutes of rope-skipping, repeated three times a day for three days. Blood glycine, proline, hydroxyproline, and hydroxylysine peaked an hour after the drink. The men who took 15 g showed double the blood marker of new type I collagen after exercise compared with placebo, and serum drawn after the gelatin dose increased collagen content and mechanical strength in engineered ligaments grown in the lab.⁶ The takeaway from that design is timing: gelatin plus vitamin C an hour before loading the tissue. The study did not measure injuries or performance.

Skin

The most-cited skin trial randomized 69 women aged 35 to 55 to 2.5 g or 5 g of a specific collagen peptide (sold as Verisol) or placebo daily for eight weeks. Both collagen groups showed a statistically significant gain in skin elasticity measured by instrument, and the effect was still present four weeks after they stopped. Skin moisture and water loss trended better but did not reach significance.⁷ The senior author works at the Collagen Research Institute in Kiel, which is tied to Gelita, and the product tested is a Gelita product. A 2023 meta-analysis pooling 26 randomized trials with 1,721 participants found a significant improvement in skin hydration and elasticity across the studies, while flagging bias risks in the trials it included and calling for larger ones.⁸

Joints in athletes

At Penn State, 147 varsity and club athletes with activity-related joint pain but no joint disease were randomized to 10 g of collagen hydrolysate in a liquid or a placebo daily for 24 weeks. Data from 97 could be evaluated. The collagen group reported less joint pain at rest, when walking, standing, carrying, and lifting, and a knee subgroup showed the largest change.⁹ One author was a Gelita employee, and Gelita announced the results in a press release before publication. A third of participants dropped out, which limits the finding.

Tendons

The Australian Institute of Sport ran a crossover pilot in 20 patients with chronic Achilles tendinopathy, 18 of whom finished. Each took 2.5 g of a tendon-targeted collagen peptide or placebo for three months alongside twice-daily calf-strengthening exercise, then switched. Function and pain scores on the standard Achilles questionnaire improved more during the collagen phase, and 12 of the 18 returned to running by the end of the six months. Tendon blood-vessel measurements did not differ between groups.¹⁰ The study was investigator-initiated but Gelita paid for it, and the authors called for a larger trial.

Can men take collagen?

Evidence: moderate. The muscle trials were run in men; the largest one drew a published challenge from independent protein researchers.

Yes. Collagen is a food protein with no sex-specific effect, and the muscle and body-composition trials were run in men rather than women. The marketing has aimed at women's skin, which is why the question keeps getting asked.

The first muscle trial randomized 53 men over 65 with sarcopenia (age-related muscle loss) to 15 g of collagen peptides or placebo daily during 12 weeks of resistance training three times a week. The collagen group gained more fat-free mass, lost more fat, and gained more leg strength than the placebo group.¹¹ Six independent protein researchers, including Stuart Phillips and Luc van Loon, published a letter questioning the size of those changes, pointing out that the difference in lean mass was several times larger than protein supplementation had ever produced in meta-analyses and that the fat loss matched what much longer weight-loss trials achieve.¹² In 2025 the journal issued a correction to the original paper's funding statement: Gelita had supplied the product and paid for participant compensation and sample analysis, and one author was a co-inventor on a patent application for the peptide.¹¹

A 2021 follow-up in 97 untrained middle-aged men compared 15 g of the same collagen peptides, whey protein, or placebo during 12 weeks of training. All three groups got stronger. The collagen group gained more fat-free mass and lost more fat than placebo, with whey landing in between.¹³ A separate German group tested 15 g of collagen peptides in 57 recreationally active young men over 12 weeks of training and also reported a larger gain in fat-free mass than placebo, with no difference in strength.¹⁴

For a man reading this, the practical points are that the trials used 15 g a day, paired it with resistance training, and ran for 12 weeks. Collagen is low in leucine and is not a complete protein, so the researchers themselves noted it does not replace whey or meat for building muscle; it was tested as an addition. A man eating little skin, broth, or gelatin has the same glycine gap described earlier as anyone else, and the Peat argument about balancing muscle meat applies to men who eat a lot of steak and chicken breast at least as much as to anyone.

Handmade in Montana

← →

How the studies used collagen and gelatin

Evidence: these are the protocols from the cited trials, listed so readers can see what was tested. They are not instructions.

The amounts in the research fall into three bands. The skin and tendon trials used 2.5 g a day of a specific peptide product for eight weeks to six months.^7,10 The athlete joint trial used 10 g a day for 24 weeks.⁹ The collagen-synthesis and muscle trials used 15 g, taken an hour before exercise in the UC Davis study and daily alongside a training program in the muscle studies.^6,11,13 The Kyoto absorption study fed 9.4 to 23 g in a single fasted dose.⁵ Peat's own account mentions "several grams" of gelatin at bedtime and friends taking 10 or 15 g a day.²

In food terms, 10 g of gelatin is about one heaped tablespoon of powder, and a cup of firm, home-made bone broth made with feet and skin can land in a similar range, though the Australian analysis shows that number cannot be assumed without testing.³ The 2017 study paired gelatin with vitamin C, which the body needs to cross-link new collagen; orange juice was the vehicle in that trial.⁶

No adverse events were reported in any of the trials cited here. Collagen is still a protein load, and the source animal can be a problem for anyone with a fish, shellfish, beef, or pork allergy. If you take medication or have a kidney, liver, or gout diagnosis, a pharmacist or clinical herbalist can review a gelatin habit against your own list before you build one.

Frequently asked questions

Short answers, with the longer discussion above.

What foods contain the most collagen?

Skin (chicken, pork, fish), tendons and cartilage (oxtail, chicken feet, hocks, knuckle bones), bone broth that sets firm when chilled, and plain gelatin. Muscle meat carries little, and plant foods carry none.

Is bone broth a good source of collagen?

It can be, but the amount varies widely by recipe. A laboratory analysis found standardized home broth below the 20 g supplement dose used in research and commercial broth lower still.³ A broth that gels firmly in the fridge is the higher-collagen kind.

Can men take collagen?

Yes. The muscle and body-composition trials were run in men, using 15 g a day with resistance training for 12 weeks.^11,13,14 Collagen is low in leucine, so the researchers tested it as an addition to a normal protein intake, not a replacement.

What is the difference between gelatin and collagen peptides?

Same amino acids, different chain length. Gelatin gels and needs hot liquid; collagen peptides (hydrolyzed collagen) dissolve cold and do not gel. Either supplies glycine, proline, and hydroxyproline.

Does collagen from food reach the skin?

Collagen peptides do reach the bloodstream after a meal, peaking at 1 to 2 hours.⁵ Whether that translates to skin change was tested with 2.5 g and 5 g of a specific peptide product, which improved instrument-measured elasticity over eight weeks in a manufacturer-funded trial.⁷

Why did Ray Peat recommend gelatin?

To offset the tryptophan and cysteine in muscle meat, which he held suppress thyroid and mitochondrial function, and to supply glycine, which he described as protective across many cell types. Gelatin contains no tryptophan and little cysteine or methionine.²

Do vegetarians have a collagen source?

No plant contains collagen. A vegetarian who eats eggs and dairy gets some glycine from those, and fruit and vegetables supply the vitamin C needed to build collagen, but the protein itself only comes from animal connective tissue or a supplement made from it.

Sources

Every citation was checked against its PubMed or DOI record. Funding and affiliations are noted where they apply.

1. Meléndez-Hevia E, de Paz-Lugo P, Cornish-Bowden A, Cárdenas ML. A weak link in metabolism: the metabolic capacity for glycine biosynthesis does not satisfy the need for collagen synthesis. J Biosci. 2009;34(6):853-872. doi:10.1007/s12038-009-0100-9. PMID 20093739. Modeling paper; supported by the authors' own Instituto del Metabolismo Celular, Tenerife.
2. Peat R. Gelatin, stress, longevity. raypeat.com. raypeat.com/articles/articles/gelatin.shtml. Essay by a physiologist; includes his own anecdotal account of gelatin at bedtime.
3. Alcock RD, Shaw GC, Burke LM. Bone broth unlikely to provide reliable concentrations of collagen precursors compared with supplemental sources of collagen used in collagen research. Int J Sport Nutr Exerc Metab. 2019;29(3):265-272. doi:10.1123/ijsnem.2018-0139. PMID 29893587. Australian Institute of Sport and Australian Catholic University.
4. Miller RA, Harrison DE, Astle CM, et al. Glycine supplementation extends lifespan of male and female mice. Aging Cell. 2019;18(3):e12953. doi:10.1111/acel.12953. PMID 30916479. National Institute on Aging Interventions Testing Program; mouse study.
5. Iwai K, Hasegawa T, Taguchi Y, et al. Identification of food-derived collagen peptides in human blood after oral ingestion of gelatin hydrolysates. J Agric Food Chem. 2005;53(16):6531-6536. doi:10.1021/jf050206p. PMID 16076145. Co-authored by researchers at Nippon Meat Packers (industry).
6. Shaw G, Lee-Barthel A, Ross ML, Wang B, Baar K. Vitamin C-enriched gelatin supplementation before intermittent activity augments collagen synthesis. Am J Clin Nutr. 2017;105(1):136-143. doi:10.3945/ajcn.116.138594. PMID 27852613. Eight male subjects; crossover design.
7. Proksch E, Segger D, Degwert J, Schunck M, Zague V, Oesser S. Oral supplementation of specific collagen peptides has beneficial effects on human skin physiology: a double-blind, placebo-controlled study. Skin Pharmacol Physiol. 2014;27(1):47-55. doi:10.1159/000351376. PMID 23949208. Senior author at the Collagen Research Institute, Kiel (Gelita-affiliated); tested product is Gelita's Verisol.
8. Pu SY, Huang YL, Pu CM, et al. Effects of oral collagen for skin anti-aging: a systematic review and meta-analysis. Nutrients. 2023;15(9):2080. doi:10.3390/nu15092080. PMID 37432180. Pooled 26 randomized trials, 1,721 participants; authors identified several biases in the included trials and called for larger studies.
9. Clark KL, Sebastianelli W, Flechsenhar KR, et al. 24-Week study on the use of collagen hydrolysate as a dietary supplement in athletes with activity-related joint pain. Curr Med Res Opin. 2008;24(5):1485-1496. doi:10.1185/030079908X291967. PMID 18416885. Co-author Flechsenhar was a Gelita employee; 97 of 147 participants evaluated.
10. Praet SFE, Purdam CR, Welvaert M, et al. Oral supplementation of specific collagen peptides combined with calf-strengthening exercises enhances function and reduces pain in Achilles tendinopathy patients. Nutrients. 2019;11(1):76. doi:10.3390/nu11010076. PMID 30609761. Investigator-initiated pilot, 18 finishers; financially supported by Gelita AG.
11. Zdzieblik D, Oesser S, Baumstark MW, Gollhofer A, König D. Collagen peptide supplementation in combination with resistance training improves body composition and increases muscle strength in elderly sarcopenic men: a randomised controlled trial. Br J Nutr. 2015;114(8):1237-1245. doi:10.1017/S0007114515002810. PMID 26353786. 2025 corrigendum (doi:10.1017/S000711452510425X) states Gelita AG supplied the product and paid participant compensation and sample analysis, and that co-author Oesser is a co-inventor on a related patent application.
12. Phillips SM, Tipton KD, van Loon LJC, Verdijk LB, Paddon-Jones D, Close GL. Exceptional body composition changes attributed to collagen peptide supplementation and resistance training in older sarcopenic men. Br J Nutr. 2016;116(3):569-570. doi:10.1017/S000711451600221X. Letter to the editor; Phillips disclosed funding from the US National Dairy Council and PepsiCo.
13. Zdzieblik D, Jendricke P, Oesser S, Gollhofer A, König D. The influence of specific bioactive collagen peptides on body composition and muscle strength in middle-aged, untrained men: a randomized controlled trial. Int J Environ Res Public Health. 2021;18(9):4837. doi:10.3390/ijerph18094837. PMID 33946565. Co-author Oesser at the Collagen Research Institute, Kiel (Gelita-affiliated).
14. Kirmse M, Oertzen-Hagemann V, de Marées M, Bloch W, Platen P. Prolonged collagen peptide supplementation and resistance exercise training affects body composition in recreationally active men. Nutrients. 2019;11(5):1154. doi:10.3390/nu11051154. PMID 31126103. Ruhr University Bochum; collagen product supplied by Gelita AG. Authors declare no conflicts; the paper states the funders could suggest phrasing, with final decisions and the right to publish held by the authors.