Hydrolysed marine collagen: what reaches the blood
is not collagen, it is a pair of two amino acids

What collagen is, before any hydrolysis
three twisted chains, with glycine in every third place

Collagen is not a protein like albumin or an immunoglobulin. It is a rope. Three chains of amino acids, each twisted to the left, wind around one another to the right and form a triple helix. The total mass comes to around 300,000 daltons, and the functional unit, tropocollagen, measures about 300 nanometres in length and between 1.5 and 2 nanometres in diameter.

The amino acid sequence repeats according to a strict rule: Gly-X-Y. Glycine, the smallest amino acid there is, takes every third position, and X and Y are most often proline and hydroxyproline. This is not a textbook curiosity. Glycine is the only one that fits in the centre of the helix, while proline and hydroxyproline stiffen the chain and keep the structure closed.

The rule shows up in chemical analysis, not just in diagrams. In a composition comparison published in 2025, collagen from cod skin has 266 glycine residues per 1,000 amino acid residues. Roughly one in three, exactly as the rule predicts, plus 63 of proline.

Marine collagen is mostly type I, the same type that predominates in our skin, bones and tendons. Its structure and molecular mass are close to those of bovine or porcine collagen, and the differences lie in the amino acid composition and in the temperature at which the helix comes apart. We have written separately about what this means in practice, in the comparison between marine and bovine collagen.

The consequence that matters here: this rope is hard to unravel. Hydrogen bonds and dense packing make it almost insoluble in water. Nobody drinks native collagen, because it does not dissolve and, even if it did, it would not get through anywhere.

Hydrolysed means cut, in two stages
first gelatine, then peptides

The first stage is denaturation. Under heat, acid or alkali, the three chains come apart from one another, and the result has a name everybody knows: gelatine. A mixture of single chains, each between 50 and 1,000 amino acids long, with masses between 3,000 and 200,000 daltons. The composition stays the same, glycine in every third place, proline and hydroxyproline in the rest.

The second stage is enzymatic hydrolysis proper. Purpose-chosen proteases break the gelatine chains into short peptides. This is where the industrial process stops, and where the powder that goes into the capsule comes from.

The difference in scale between the two ends is what always gets lost in commercial messaging, and it is enormous.

What was actually measured in people's blood
and in what amounts

In 2005, a team led by Iwai published in the Journal of Agricultural and Food Chemistry the first clear identification of collagen peptides in human blood. Healthy volunteers fasted for 12 hours and then drank gelatine hydrolysates, between 9.4 and 23 grams, made from pig skin, chicken feet and cartilage.

Before ingestion, hydroxyproline in peptide form was practically absent from the blood. After ingestion it rose to 20 to 60 nanomoles per millilitre of plasma, peaked at one to two hours and had fallen to half its peak by the fourth hour. The main component was identified as Pro-Hyp, proline bound to hydroxyproline. Alongside it, in small but real amounts: Ala-Hyp, Ala-Hyp-Gly, Pro-Hyp-Gly, Leu-Hyp, Ile-Hyp and Phe-Hyp.

Five years later, Ichikawa and colleagues ran the same peptides through tandem mass spectrometry and counted them one by one. Nine hydroxyproline-containing peptides, all quantified. Pro-Hyp came out at a maximum concentration of 60.65 nanomoles per millilitre, with a standard deviation of 5.74. The other eight lined up much lower, from 23.84 down to 0.67.

In the same paper, the authors note that the level of hydroxyproline-containing peptides in plasma is substantially higher than that observed after taking other food peptides. So something from collagen gets through more easily than from other proteins. The natural question is what exactly, and by which route.

The gate it comes in through
a transporter that accepts pieces of two, not three

The answer came from an experiment on membrane vesicles from pig intestine, published in 2007 by Aito-Inoue and colleagues. They followed what happens to the tripeptide Gly-Pro-Hyp, one of the most common sequences in collagen, right at the edge of the cell.

The verdict is clear-cut: Gly-Pro-Hyp does not cross the apical membrane intact. An enzyme anchored on the edge of the cell, aminopeptidase N, cuts the glycine off the end. What remains, the dipeptide Pro-Hyp, enters the cell through PEPT1, a proton-coupled transporter. It is known to be PEPT1 because transport was blocked by a competing peptide, Gly-Pro, and because it depends on pH.

That resistance is the second half of the explanation. The same authors show that Pro-Hyp is hard for the proteases on the surface of the intestinal lining to break. Most dietary dipeptides are split into free amino acids before they get far. Pro-Hyp is not. That is why it appears in the blood as a dipeptide and not as two separate amino acids, and that is why it can be measured.

What it does next is the open question. A 2009 study by Shigemura placed Pro-Hyp, at 200 nanomoles per millilitre, on discs of mouse skin kept in culture, without serum. After 72 hours, the number of fibroblasts that had migrated from the skin onto the plate was significantly higher, and the effect disappeared when mitomycin C was added. On collagen gel, fibroblast growth followed the dose, between 0 and 1,000 nanomoles per millilitre. Mouse skin, in a dish. That is not the same as your skin, and the authors do not claim it is.

Hyaluronic acid and the figure 1,000
what it actually describes

Hyaluronic acid is not a protein and has no structural connection with collagen. It is a polysaccharide: a double unit of D-glucuronic acid and N-acetylglucosamine, linked alternately and repeated thousands of times. One unit weighs about 379 daltons. Unlike the rest of its family, it carries no sulphate groups.

Length is the unusual part. In intact tissues, the molecule usually exceeds 1,000 kilodaltons, which puts it among the largest molecules in the body. The skin holds roughly half of all the hyaluronic acid in the body, and turnover is surprisingly fast for a matrix component: the half-life ranges from half a day to a few days.

Now, the figure. The phrase “binds up to a thousand times its weight in water” circulates in the specialist literature as a description of the polymer, and appears exactly like that, spelled out in words, in a review published in Marine Drugs in 2023. It is a property attributed to the long chain of hyaluronic acid in the matrix between cells, where the molecule weighs millions of daltons and behaves like an open sponge.

So the figure describes the molecule, not a dose. It tells you how much space a chain of hyaluronic acid takes up in water, not what happens to 100 milligrams of powder after you swallow them. These are two different statements, and in almost every text on the internet they are presented as one.

On the second, the one about swallowing, there are measurements. A randomised, double-blind study published in Nutrients in 2021 gave 120 milligrams of hyaluronic acid a day, for 12 weeks, to 40 healthy adults aged 35 to 64, against placebo. Wrinkles, water content of the stratum corneum, water loss through the skin and elasticity were measured, and differences compared with placebo appeared at week 8 and at week 12.

What studies on skin measured
and what their own authors say

Two studies from 2014 are the most often cited, both led by Proksch. In the first, 69 women aged 35 to 55 were given 2.5 or 5 grams of collagen hydrolysate a day, or placebo, for 8 weeks, 23 in each group. Skin elasticity, measured with an instrument, came out significantly better than placebo in both collagen groups. For hydration and water loss through the skin, a positive influence was seen in a subanalysis, but the difference did not reach the threshold of statistical significance. The authors say so explicitly.

In the second, 114 women aged 45 to 65 were given 2.5 grams of a specific collagen peptide, or placebo, again for 8 weeks. The volume of wrinkles around the eyes fell by 20% compared with placebo, and suction blister biopsies measured 65% more type I procollagen and 18% more elastin. Fibrillin rose by 6%, and there the difference was not significant.

The meta-analyses were built on top of them. The 2021 one brought together 19 randomised studies, 1,125 participants aged 20 to 70, 95% of them women, with results favourable compared with placebo for hydration, elasticity and wrinkles. The 2023 one went to 26 studies and 1,721 participants, with significant effects on hydration and elasticity, and with an observation worth reading more carefully than the conclusion: the authors identified several sources of bias in the included studies and call for large randomised studies before the subject can be considered closed.

What is useful to take away from the whole batch: the measurements were made with instruments on the skin, not in the mirror; the doses were in grams and ran for 8 to 12 weeks; and the preparations tested were specific peptides, not collagen in general. About hair, which runs on a different clock and has different studies, we have written in the article on hydrolysed collagen and hair growth.

What our label says
on the collagen line

For us, these figures sit on Full-beauty. A serving of 2 capsules provides 800 mg of hydrolysed marine collagen, peptides from fish skin, types I and III, enzymatically cut down to low molecular weight, plus 100 mg of hyaluronic acid as sodium hyaluronate obtained by fermentation, not of animal origin. Alongside them, 100 mg of bamboo extract standardised to 50% silicon, 100 mg of L-methionine, and the vitamin and mineral part: zinc, vitamin E, vitamin A, B6, biotin and B12. Ten actives, each with its dose stated on the label, no proprietary blends.

Zinc and biotin contribute to the maintenance of normal hair and skin.

Where it fits into this article: the 100 mg of hyaluronic acid is in the same range as the 120 milligrams a day used in the 2021 study, and the collagen peptides are already cut, meaning the hydrolysis we talked about is done before it reaches you.

A practical tip: the capsules are taken with a meal, with water, because the formula contains vitamin A and vitamin E, both fat-soluble, and the fat in food helps them be absorbed. The time of day does not matter, the habit does. If you take anything else as well, we have put the timetable together in which supplements should not be taken together, and we have written about what else people look at when they see hairs in their brush in my hair is falling out, what should I take.

If the whole subject has caught your interest, there is another protein that breaks down along the way and is worth reading about: immunoglobulin G in bovine colostrum. The same question, a different answer.

Frequently asked
questions