Collagen peptides are small pieces of collagen, the main structural protein of skin, bone, tendons and cartilage, cut from cattle, pig, fish or chicken tissue; the middle half of the US collagen-only labels we counted list 2.5 to 11 g per serving.[1][2][5] There is no US recommended amount or upper limit for collagen.[5]
Questions about collagen peptides
What collagen does in the body
Collagen makes up about one third of all protein in the human body; scientists have described 28 types in vertebrates.[1] Type I is the most common, in skin, bone, tendons and ligaments; type II is found in cartilage, and type III in skin, muscle and blood vessels.[1][2] Each molecule is three protein chains wound into a triple helix, with glycine at every third position and many of the other positions held by proline and hydroxyproline.[1]
After people swallow gelatin hydrolysate, some of it reaches the blood as pairs and triplets of amino acids; in a 2005 study they peaked after 1–2 hours and fell to half by 4 hours, and the main one was Pro-Hyp.[3]
As food protein, collagen contains no tryptophan, so collagen alone has a protein quality score of zero.[4] In a 2019 calculation, collagen peptides could replace up to 36% of protein in a typical US diet and the diet's score stayed at 1.0, the maximum.[4]
Where collagen comes from
Collagen for supplements comes from animal bones, tendons, skin, scales and cartilage; cattle are the main source, and pig skin, fish and chicken skin and feet are others.[2] Heating above 40 °C unwinds the triple helix, and enzymes, or in some processes acids or alkalis, cut the chains into peptides of 3–6 kDa, against 285–300 kDa for a whole collagen molecule.[2] The page on marine and bovine collagen compares two US powders.
How much collagen, and limits
| Amount | |
|---|---|
| US recommended amount | None set[5] |
| US upper limit | None set[5] |
| US collagen-only labels, middle half | 2.5–11 g per serving[5] |
| Most common amounts on those labels | 10 g, 2.5 g and 11 g per serving[5] |
| Trial doses on this page | 5–20 g a day[6][7][8][9][10][11] |
US labels list collagen in g or mg per serving, without a percent Daily Value. Of 344 US collagen-only labels, 46% name a bovine source, 15% fish, 7% chicken and 3% porcine; 31% name none. Of 3,203 labels that list collagen, 80% list other ingredients too.[5]
Collagen in combination products
Most labels that list collagen also list other ingredients, so the grams of collagen in a combination product can differ from a collagen-only powder.[5] Brand names on these labels include Verisol on 97 labels, Fortigel on 13 and Peptan on 11.[5]
When collagen intake runs low
Forms of collagen peptides
| Form | Collagen type and tissues | What is known |
|---|---|---|
| Bovine (cattle) | Type I; bones, tendons and other tissues[2] | The main source of collagen extraction; Peptan B is a bovine collagen hydrolysate[2][14] |
| Porcine (pig) | Type I; skin[2] | Fortigel is porcine type I peptides of about 3 kDa, from GELITA, used in studies 1 and 2[6][7] |
| Marine (fish) | Type I; skin, bones and scales[2] | Peptan F is a fish collagen hydrolysate[14] |
| Chicken | Type II in cartilage; also skin and feet[2] | A source of hydrolyzed collagen[2] |
| Undenatured type II (UC-II) | Type II, not hydrolyzed | A separate ingredient from hydrolyzed collagen peptides. |
Sources: León-López 2019, the trial reports and the lab studies cited. Verisol was tested at 2.5 g a day in a skin trial of 114 women.[13] We do not rank forms. In lab tests, collagen preparations differed in their peptides, even between some batches of one product; the authors write that results for one preparation cannot be extrapolated to another.[14]
Before you choose a product
- Animal source. Check whether the collagen comes from cattle, pig, fish or chicken; the source can matter for religious diets and for fish allergy, and every source is an animal.[2]
- Grams per serving. Compare the source and grams per serving on the labels. Trial findings apply to the specific preparations and participants studied.[5][6][11]
Collagen peptides and medicines
The sources used for this page do not report interactions of collagen peptides with specific medicines.[2] People who take medicines regularly can ask their pharmacist before adding a supplement.
What the studies show
The six trials below enrolled adults without a diagnosed joint condition. Trials on this page gave 5 to 20 g a day to healthy adults for 12 to 15 weeks or longer and measured knee pain or tendon size: 2 reported less pain than with placebo, 2 saw no difference and 2 had mixed results, and the maker of the ingredient paid for or supplied the two trials that reported less pain.[6][7][8][9][10][11]
In short:
- Knee pain: two trials of Fortigel 5 g found pain fell more than with placebo; one trial of 10 g found no difference, and one of 10 or 20 g found differences only in some scores (studies 1–4).[6][7][8][9]
- Tendons with strength training: in one trial the kneecap tendon grew more with 5 g; in another, 15 g made no difference (studies 5 and 6).[10][11]
- Skin: a 2025 review of 23 skin trials found differences overall, but the authors found no statistically significant effect in higher-quality trials or in trials without pharmaceutical-company funding; skin reviews are not listed below; the page on skin studies compares them.[12]
1. Knee pain during sport, 12 weeks — Zdzieblik and colleagues, Nutrients, 2021.[6] 218 active adults aged 18–30 with knee pain linked to activity were randomized to 5 g of Fortigel a day or a maltodextrin placebo, and 180 completed the study. Pain during sport on a 100 mm scale fell by 21.9 mm with collagen and 15.6 mm with placebo, statistically significant; pain at rest and after squats did not differ. Funding: GELITA, the maker of Fortigel, provided the peptides and paid for part of the study.
2. Knee or hip pain in daily life, 12 weeks — Schulze and colleagues, International Journal of Environmental Research and Public Health, 2024.[7] The trial randomized 182 adults with knee or hip pain. The analysis included the 154 participants who completed the trial. The study compared 5 g of Fortigel with 5 g of maltodextrin against 10 g of maltodextrin.[7] Physician-rated pain at rest and when walking fell more with collagen. Physician-rated pain on stairs did not differ significantly. Participants reported greater reductions in pain on stairs and when kneeling, with statistically significant differences. Joint stiffness did not differ significantly. Funding: GELITA provided the peptides.
3. Knee pain in active older adults, 12 weeks — Bongers and colleagues, Applied Physiology, Nutrition, and Metabolism, 2020.[8] 167 physically active adults, median age 63, with self-reported knee pain took 10 g of collagen peptides a day or placebo. Pain and knee questionnaire scores changed by similar amounts in both groups. Funding: see the publication.
4. Joint pain in lifelong exercisers, 6–9 months — Kviatkovsky and colleagues, Journal of the International Society of Sports Nutrition, 2023.[9] 86 active adults aged 40–65 took 10 g or 20 g of SOLUGEL a day or placebo. A daily-activity score changed differently between groups, but no single group changed significantly; knee pain scores did not differ overall. Funding: PB Leiner, maker of SOLUGEL.
5. Tendon size with strength training, 14 weeks — Jerger and colleagues, European Journal of Sport Science, 2023.[10] 50 moderately active men took 5 g of collagen peptides a day or placebo while training the thigh muscles. The kneecap tendon grew more with collagen, statistically significant; strength and muscle size rose by similar amounts in both groups. Funding: see the publication.
6. Tendon size and stiffness, 15 weeks — Balshaw and colleagues, Medicine and Science in Sports and Exercise, 2023.[11] 39 young men doing leg strength training took 15 g of collagen peptides a day or placebo. Tendon size and stiffness did not differ; stiffness rose by 17.3% with collagen and 20.9% with placebo. Funding: not stated on the abstract page.
Information, not a prescription. Do not start, stop or combine supplements or medicines on your own.
Sources
- Shoulders and Raines. Collagen structure and stability. Annual Review of Biochemistry. 2009. https://doi.org/10.1146/annurev.biochem.77.032207.120833 ↩
- León-López and colleagues. Hydrolyzed collagen: sources and applications. Molecules. 2019. https://doi.org/10.3390/molecules24224031 ↩
- Iwai and colleagues. Journal of Agricultural and Food Chemistry. 2005. https://doi.org/10.1021/jf050206p ↩
- Paul and colleagues. Nutrients. 2019. https://doi.org/10.3390/nu11051079 — no external funding; one author provides research services to a maker of collagen supplements. ↩
- National Academies, Dietary Reference Intakes summary tables; NIH Dietary Supplement Label Database, data release of 25 September 2025, our count of October 2026. https://dsld.od.nih.gov/ ↩
- Zdzieblik and colleagues. Nutrients. 2021. https://doi.org/10.3390/nu13020523 — PubMed 33562729. ↩
- Schulze and colleagues. International Journal of Environmental Research and Public Health. 2024. https://doi.org/10.3390/ijerph21060687 — PubMed 38928934. ↩
- Bongers and colleagues. Applied Physiology, Nutrition, and Metabolism. 2020. https://doi.org/10.1139/apnm-2019-0654 — PubMed 31990581. ↩
- Kviatkovsky and colleagues. Journal of the International Society of Sports Nutrition. 2023. https://doi.org/10.1080/15502783.2023.2243252 — PubMed 37551682. ↩
- Jerger and colleagues. European Journal of Sport Science. 2023. https://doi.org/10.1080/17461391.2023.2232758 — PubMed 37424319. ↩
- Balshaw and colleagues. Medicine and Science in Sports and Exercise. 2023. https://doi.org/10.1249/MSS.0000000000003242 — PubMed 37436929. ↩
- Myung and Park. The American Journal of Medicine. 2025. https://doi.org/10.1016/j.amjmed.2025.04.034 — PubMed 40324552. ↩
- Proksch and colleagues. Skin Pharmacology and Physiology. 2014. https://doi.org/10.1159/000355523 ↩
- Schadow and colleagues, PLoS One, 2013, https://doi.org/10.1371/journal.pone.0053955; Schadow and colleagues, International Journal of Molecular Sciences, 2017, https://doi.org/10.3390/ijms18010207; Simons and colleagues, Scientific Reports, 2018, https://doi.org/10.1038/s41598-018-36046-3. ↩
You may quote, republish or translate this article under CC BY 4.0: credit “Verateca”, link to this page and indicate changes, including translations. Quotes from NIH, FDA and other sources, label texts and prices are not covered by this license.