Type 1 and 3 Collagen: What It Means and When the Label Matters

This content is for informational purposes only and is not medical advice. Collagenium is published by a research-driven editorial team, not licensed medical professionals. See our editorial standards and methodology. Consult a registered dietitian, physician, or pharmacist before starting any supplement, especially if you have existing health conditions, are pregnant or breastfeeding, or take medications.

Type I and type III are the two collagen proteins that make up most of your skin, bones, tendons and blood vessel walls. A tub labelled “type 1 and 3 collagen” is nearly always hydrolyzed bovine or porcine hide, because hide contains both. Once that collagen is broken into peptides, the type number describes the raw material, not what your body rebuilds from it. The one place the type genuinely changes the product is undenatured type II.

The rest of this guide explains what the two types do, which sources contain them, what reaches your bloodstream after a scoop, what the trials measured, and how to read the label without paying for a Roman numeral.

What are type I and type III collagen?

Collagen is not one protein but a family of 28, each built from three chains wound into a triple helix (Ricard-Blum, 2011). The type number is decided by the amino acid sequence of those chains and how the finished molecules assemble (Singh, 2023). Type I and type III are among the most abundant members, and they usually occur together.

Type I is the structural one. It forms thick, stiff fibres that give tissue its tensile strength, and those fibres turn over slowly once laid down. It is the main collagen of skin, tendon, ligament and blood vessel walls, and it makes up most of the organic matrix of bone and teeth (Singh, 2023). Cleveland Clinic puts type I at about 90 percent of the collagen in the body.

Type III is the elastic partner. It forms thinner, less durable fibres with a faster turnover, builds the fine reticular mesh in skin, lungs, blood vessels and hollow organs, and sits alongside type I in all of them. The review we draw on describes type I as a stiff fibrillar protein that gives tensile strength and type III as an elastic network that stores kinetic energy as an elastic rebound; it also notes that the maturation of type I fibres depends on type III (Singh, 2023). That is why the two are discussed as a pair and why bovine hide, which is skin, contains both.

If you are new to the topic, our Collagen 101 guide covers the basics of what collagen is and why production falls with age.

Where are the five common collagen types found in the body?

Five of the 28 types are the ones that appear on supplement labels. The table also shows what each type is typically made from and whether the type survives the processing that turns tissue into powder.

Type	Where it is found	Main job	Typical supplement raw material	Does the type survive processing?
I	Skin, bone, tendon, ligament, cornea, blood vessel walls; about 90% of body collagen	Tensile strength, structural framework	Bovine hide and bone, porcine skin, fish skin and scale	No. Hydrolyzed into peptides; type identity is lost
II	Articular (joint) cartilage, vitreous of the eye	Cartilage framework that resists compression	Chicken sternum cartilage	Only if sold undenatured (intact); hydrolyzed “type II” is peptides again
III	Skin, blood vessels, lungs, intestines and other hollow organs, alongside type I	Elastic reticular mesh; prominent early in tissue repair	Bovine hide, porcine skin	No. Hydrolyzed into peptides
IV	Basement membranes, the thin sheets under epithelial layers of skin and lining organs	Filtration and anchoring	Rarely isolated for supplements	Not applicable
V	Cornea, skin, hair, placenta, in small amounts within type I fibres	Regulates type I fibre formation	Eggshell membrane blends (marketed as types I, V and X)	No. Hydrolyzed

Sources for distribution: Gelse (2003), Singh (2023), Cleveland Clinic. Type III’s role in repair is worth a sentence because it explains a marketing phrase you may meet. During wound healing the body lays down more type III early, then replaces it with type I as the tissue matures and regains strength (Singh, 2023). Some brands call type III “juvenile collagen” on the back of this. It is a description of how the body sequences its own repair, not evidence that swallowing type III peptides does anything specific.

Which supplements contain type 1 and 3 collagen?

Bovine (cow) hide is the main one, and porcine (pig) skin is the other mammalian source. Skin of any mammal contains type I with a meaningful share of type III, so collagen peptides made from hide carry both, and the label says so. Fish skin and scale collagen is almost entirely type I, which is why marine products are labelled type I alone (Ohara, 2007; León-López, 2019). Chicken sternum cartilage is type II (Lugo, 2016). Our bovine collagen guide and marine collagen guide go deeper on each source.

One observation from our own product catalogue shows how loosely the label is applied. Of the 25 bovine collagen products we have reviewed, 17 are listed as “types I and III” and 6 as “type I” only, although all of them are hydrolyzed bovine collagen peptides from the same kind of raw material. The difference is a labelling choice, not a formulation difference.

Does the type on a hydrolyzed collagen label matter?

For hydrolyzed products, no, and here is the chain of evidence that leads to that answer.

Step 1: hydrolysis destroys the structure that defines the type. Hydrolyzed collagen is a mixture of short peptides with molecular weights of roughly 3 to 6 kilodaltons, produced by cutting the collagen chains with enzymes (León-López, 2019). The triple helix is gone and the fibres are gone. Since the type is defined by the full chain sequence and how whole molecules assemble (Singh, 2023), a bag of fragments no longer has a type in any functional sense. It has an amino acid profile and a peptide size distribution. Our hydrolyzed collagen guide explains the process in more detail.

Step 2: what reaches your blood is dipeptides, tripeptides and free amino acids. When volunteers drank 9.4 to 23 grams of gelatin hydrolysate from porcine skin, chicken feet or cartilage, the peptide form of hydroxyproline in their plasma peaked one to two hours later, and the dominant circulating peptide was prolyl-hydroxyproline (Pro-Hyp), with smaller amounts of Ala-Hyp, Pro-Hyp-Gly and a few others (Iwai, 2005). A follow-up crossover in five men compared fish scale, fish skin and porcine skin hydrolysates and found the same small peptides, with the mix depending on the source (Ohara, 2007). What these studies detected were small fragments, not intact collagen of any type.

Step 3: your body builds its own type I and III from scratch. Fibroblasts assemble procollagen I and III inside the cell from amino acids, trim it, export it and cross-link it into fibres (Singh, 2023). They decide which type to make based on the tissue and the signals they receive, not based on which type the amino acids came from. As Cleveland Clinic puts it, ingested collagen peptides “can’t be directed to where you want them to be used”; the body uses them for whatever it needs.

Step 4: the peptides may act as signals, but the signal is not type-specific. Pro-Hyp, the main peptide found in blood after a collagen drink, increased fibroblast growth in mouse skin explants in a dose-dependent way (Shigemura, 2009). This is the leading proposed explanation for how collagen peptides might act beyond supplying amino acids. It is a preclinical finding, and the same peptide appeared in volunteers’ blood after porcine skin, chicken feet and cartilage hydrolysates alike (Iwai, 2005), so it is not specific to any type.

Putting the steps together, here is how we grade the claims you will see:

• “Type I and III peptides rebuild your type I and III collagen.” Marketing claim. No study has traced type-specific rebuilding, and steps 1 to 3 explain why it is implausible.
• “Hydrolyzed collagen peptides, from bovine or fish sources, improve skin hydration and elasticity.” Some evidence. Three meta-analyses of randomized trials pooled positive results (de Miranda, 2021; Pu, 2023; Myung, 2025), but the most recent found no effect in trials without industry funding or in high-quality trials (Myung, 2025).
• “The animal source changes which peptides circulate.” Preliminary. True in a five-person crossover (Ohara, 2007), with no link yet to a clinical outcome.
• “A type I plus III product works better than a type I-only product.” Marketing claim. We could not find a randomized trial that tested this comparison on any outcome.

Do you need both type I and type III?

No trial shows that you do. The simplest test is to look at fish-derived peptides, which are type I only. A 12-week randomized trial of 64 adults taking 1 gram a day of a low-molecular-weight fish collagen peptide found higher skin hydration at 6 and 12 weeks and improved wrinkle measures at 12 weeks versus placebo (Kim, 2018). A triple-blind trial in women aged 45 to 60 taking hydrolyzed fish collagen for 12 weeks reported a 35 percent reduction in wrinkle score from baseline (Evans, 2021; the ingredient company holds the paper’s copyright).

Those results point in the same direction as the trials of hydrolyzed collagen peptides whose animal source is not specified in the published abstracts (Proksch, 2014; Asserin, 2015).

A 2023 meta-analysis of 26 randomized trials looked directly at source and found no significant difference between collagen sources for skin elasticity, though the hydration effect did vary by source (Pu, 2023). If having type III in the raw material mattered for skin, the type I-only fish products should underperform, and they do not.

The same logic applies to multi-collagen blends that list five types. Listing more types describes more raw materials; it does not add evidence. If you want one product that covers bovine, fish and chicken sources for dietary variety, a blend is reasonable, but buy it for that reason rather than for the type count.

When does the collagen type genuinely matter?

Undenatured type II collagen is the exception, and it is the opposite of a hydrolyzed product. It is chicken sternum cartilage processed so that the triple helix stays intact, and it is dosed at 40 milligrams a day rather than 5 to 10 grams. The rationale is that the intact molecule, not its amino acids, is the active ingredient, which is why the type and the “undenatured” word both matter here.

Trial	Participants	Dose and duration	Result	Notes
Lugo, 2013	55 healthy adults with knee discomfort on exertion	40 mg UC-II daily, 120 days	Knee extension improved vs placebo; longer pain-free exercise	Authors employed by the manufacturer
Lugo, 2016	191 adults with knee osteoarthritis	40 mg UC-II vs glucosamine plus chondroitin vs placebo, 180 days	WOMAC symptom score improved vs both comparators	Two manufacturer authors; safety did not differ
Crowley, 2009	Adults with knee osteoarthritis	40 mg UC-II vs glucosamine plus chondroitin, 90 days	WOMAC down 33% vs 14%	No placebo arm

We grade undenatured type II for joint comfort as some evidence: consistent results in randomized trials, but mostly manufacturer-run and in a few hundred people in total. A 2024 meta-analysis of 35 randomized trials covering all collagen derivatives found small-to-moderate improvements in osteoarthritis pain and function, with moderate to high certainty (Liang, 2024). Our undenatured type II collagen guide covers it in full.

The caution for shoppers: a product that says “hydrolyzed type II collagen” or lists chicken collagen in a multi-type blend is back to being peptides. It has not been shown to do what the 40 mg undenatured studies showed, and the type label on it means no more than it does on a bovine tub.

What do the trials of hydrolyzed collagen peptides actually show?

The benefits attached to “type 1 and 3” in marketing are the benefits of hydrolyzed collagen peptides in general. The table lists what the trials measured, in whom, at what dose, and how confident we are. Doses are what the trials used, not recommendations.

Outcome	Study	Participants	Dose and duration	Source	Result	Funding or caveat	Our grade
Skin elasticity	Proksch, 2014	69 women, 35 to 55	2.5 g or 5 g daily, 8 weeks	Specific collagen peptides (source not stated in abstract)	Elasticity improved vs placebo at both doses; moisture not significant	Funding not stated in abstract	Some evidence
Skin hydration, dermal collagen density	Asserin, 2015	Two placebo-controlled trials	Daily collagen peptides, 8 to 12 weeks	Collagen peptides (source not stated in abstract)	Hydration up at 8 weeks; dermal density up and fragmentation down from 4 weeks	Authors from the manufacturer	Some evidence
Skin hydration and wrinkles	Kim, 2018	64 adults	1 g daily, 12 weeks	Fish (type I only)	Hydration higher at 6 and 12 weeks; wrinkle measures improved at 12	One author from a collagen manufacturer; no conflict declared	Some evidence
Wrinkles	Evans, 2021	Women 45 to 60	Hydrolyzed fish collagen daily, 12 weeks	Fish (type I only)	35% wrinkle score reduction from baseline	Ingredient company (Vinh Horn) holds the paper’s copyright	Some evidence
Skin aging, pooled	de Miranda, 2021; Pu, 2023	19 trials, 1,125 people; 26 trials, 1,721 people	Mostly 2.5 to 10 g daily, about 90 days	Mixed	Hydration, elasticity and wrinkles improved vs placebo	Pu found no source difference for elasticity; biases noted	Some evidence
Skin aging, by funding	Myung, 2025	23 trials, 1,474 people	Mixed	Mixed	Pooled benefit, but no effect in independently funded or high-quality trials	Authors conclude no clinical evidence supports collagen for skin aging	Downgrades skin to some evidence
Bone mineral density	König, 2018	131 postmenopausal women	5 g daily, 12 months	Specific collagen peptides	Spine and femoral neck BMD rose vs control; bone formation marker up	One trial; one author at a collagen research institute; no conflict declared	Some evidence (single trial)
Pain and function in people with osteoarthritis	Liang, 2024	35 trials, 3,165 people	Mixed, incl. hydrolyzed and undenatured	Mixed	Small-to-moderate benefit; moderate to high certainty	University authors; funding not stated in abstract	Well-supported for a modest effect

Two points stand out. First, the fish-only (type I) trials and the unspecified-source peptide trials point the same way, and the one meta-analysis that tested source found no elasticity difference, which is the practical case against paying for the type count. Second, the 2025 meta-analysis is a real problem for the skin story: when the authors separated trials by funding and quality, the benefit vanished (Myung, 2025). We therefore grade skin outcomes as some evidence rather than well-supported, and we would not call any type-specific skin claim supported at all.

Bovine vs marine vs chicken: which source should you choose?

Pick the source by diet, allergen, taste and price. The type column in the table below is accurate about the raw material and tells you nothing about results once the product is hydrolyzed, except for undenatured type II.

	Bovine (cow hide, bone)	Marine (fish skin, scale)	Chicken (sternum cartilage)
Types in the raw material	I and III	Mostly I	II
Usual form sold	Hydrolyzed peptides, 10 to 20 g scoops	Hydrolyzed peptides, about 10 g scoops; also capsules	Undenatured (40 mg) or hydrolyzed peptides
Doses used in trials	2.5 to 10 g daily for skin; 5 g for bone	1 to 10 g daily for skin	40 mg undenatured for joints
Human trial evidence	Skin elasticity and hydration (some evidence); bone density in postmenopausal women (one trial)	Skin hydration and wrinkles (some evidence)	Joint comfort with undenatured II (some evidence); hydrolyzed II not separately shown
Absorption claim	Baseline	Fish scale hydrolysate gave higher blood peptide levels than porcine skin in 5 men (Ohara, 2007); preliminary	Not applicable to undenatured, which is not meant to be digested into peptides
Amino acid profile	Bovine Achilles tendon collagen had about 30% more hydroxyproline than fish skin collagen (Gauza-Wlodarczyk, 2017)	Less hydroxyproline per gram in the same comparison	Not compared in the sources we checked
Allergen and diet notes	Beef; check halal or kosher certification	Fish, a major allergen; pescatarian-friendly	Poultry
Cost per serving in our catalogue	$1.96 (Vital Proteins, 20 g serving)	$1.00 (Sports Research, 11 g serving)	Varies; undenatured products are capsules at a fraction of a gram

“Marine collagen absorbs better” is one of the most repeated claims in the category. Its basis is the 2007 crossover above: five men, three hydrolysates, and a higher area under the curve for fish scale than for porcine skin peptides (Ohara, 2007). That is a real finding and a thin one. Fish skin collagen also had about 30 percent less hydroxyproline than bovine tendon collagen in one laboratory comparison (Gauza-Wlodarczyk, 2017), and no trial has shown that either difference changes a skin or joint outcome. We grade “marine absorbs better” as preliminary.

What are the side effects of type 1 and 3 collagen?

In the published trials, very few. The 2024 meta-analysis of 35 osteoarthritis trials found collagen derivatives did not increase withdrawals or adverse events compared with controls (Liang, 2024). A 2019 systematic review of 11 dermatology trials at 2.5 to 10 grams a day for 8 to 24 weeks described supplementation as generally safe with no reported adverse events (Choi, 2019), and the 8-week elasticity trial noted none (Proksch, 2014).

The practical risks are not in the trials. Marine collagen is made from fish, a major allergen, and multi-type blends may contain eggshell membrane, so read the allergen line. Bovine and porcine products matter for halal, kosher and vegetarian diets. A scoop is also 10 to 20 grams of protein, going by the serving sizes in our catalogue, which anyone on a protein-restricted diet should raise with their clinician. Finally, the FDA does not require supplements to be tested for safety or effectiveness before sale (Cleveland Clinic), so third-party testing for heavy metals is the quality check that matters.

How should you choose a collagen peptide supplement?

Ignore the Roman numerals and check these in order:

1. Grams of collagen per serving. The skin trials used 1 to 10 grams a day (1 gram only for a concentrated low-molecular-weight fish peptide, 2.5 to 10 grams for standard hydrolysates) and the bone trial used 5 grams. Compare the serving on the label with those doses, not with the front-of-pack claims.
2. Form. “Hydrolyzed collagen” or “collagen peptides” for a powder. If the product claims joint benefits from type II, it should say “undenatured” and list about 40 mg.
3. Source named, allergens listed. Bovine, porcine, marine or chicken, plus fish and egg allergen statements.
4. Blend disclosure. A multi-type product should state how many grams of each source it contains. “Proprietary blend” means it does not.
5. Third-party testing for heavy metals, with the certifier named.
6. Cost per serving, worked out from the label, not the front-of-pack claims.

For the cost check, here are the three products in our catalogue with a listed serving count. Prices are the ones recorded in our reviews and may have changed; the per-10-gram figure assumes the whole serving weight is collagen, which slightly flatters blends with added ingredients.

Product	Source and label	Price	Servings	Serving size	Cost per serving	Cost per 10 g
Vital Proteins Collagen Peptides	Bovine, types I and III	$47.00	24	20 g	$1.96	$0.98
Sports Research Marine Collagen Peptides	Marine, type I	$29.95	30	11 g	$1.00	$0.91
Ancient Nutrition Multi Collagen Protein	Multi-source, types I, II, III, V and X	$44.95	45	9 g	$1.00	$1.11

On a per-gram basis the three land within 20 cents of each other, which is a useful reminder that the type count on the front does not move the price much either. Choose the one that fits your diet and that you will actually take daily, since every positive trial ran for at least 8 weeks.

The bottom line

Type I and type III are real, important and well described: the stiff framework and the elastic mesh of your skin, bones, tendons and vessels. On a hydrolyzed collagen label, though, “type 1 and 3” tells you the powder came from mammalian hide and nothing more, because hydrolysis removes the type and your body builds its own from amino acids and signals. Buy hydrolyzed collagen on grams, source, allergens, testing and price. Reserve your attention to the type for the one product where it matters: undenatured type II, dosed in milligrams, for joints.