Tier 3 — preclinical

Oral ingestion of collagen hydrolysate leads to the transportation of highly concentrated Gly-Pro-Hyp and its hydrolyzed form of Pro-Hyp into the bloodstream and skin

Misato Yazaki, Yukihiko Ito, Masayoshi Yamada, Spyros Goulas, Sachiyuki Teramoto, Masa-aki Nakaya, Shigeo Ohno, Kohji Yamaguchi

Journal of Agricultural and Food Chemistry 2017 65(11):2315–2322

Bibliography

DOI: 10.1021/acs.jafc.6b05679
PubMed: PMID 28244315
Funding: Funding statement (verbatim): "This work was partially supported by the fund for Creation of Innovation Centers for Advanced Interdisciplinary Research Areas Program in the Project for Developing Innovation System from Ministry of Education, Culture, Science and Technology."
Competing interests: No explicit competing interests statement disclosed in the paper. Seven of eight co-authors (Yazaki M, Ito Y, Yamada M, Teramoto S, Yamaguchi K, plus Yazaki M and Nakaya M) were employees of FANCL Corporation's Research Institute — FANCL commercially markets collagen supplement products including the high-tripeptide-content collagen hydrolysate (HACP-01) tested in this study. Both study hydrolysates (HACP-01 and CPAH-5) were provided by Jellice (Miyagi, Japan), a further commercial collagen supplier. Remaining co-author Ohno S was at Yokohama City University Graduate School of Medical Science.

Study snapshot

Design	Human single-blind, two-arm crossover for plasma kinetics; parallel murine dose-ranging and pure-peptide administration arms for tissue distribution
Model	Healthy adult humans (4 M, 8 F); male BALB/c mice for tissue distribution
Sample	Human arm: n=12, mean age 31.5 ± 6.5 years. Mouse arm 1: 5 mice per dose per time point. Mouse arm 2: 5 mice per pure peptide per time point.
Intervention	Human arm: single oral dose of 300 mg/kg body weight of either high-tripeptide-content porcine collagen hydrolysate (HTC-col, HACP-01, avg MW 1,500–1,800 Da) or low-tripeptide porcine collagen hydrolysate (LTC-col, CPAH-5, avg MW 4,500–5,500 Da), ≥7-day washout between arms. Mouse arm 1: 900 or 1,800 mg/kg HTC-col by oral gavage after 16 h fast. Mouse arm 2: pure Gly-Pro-Hyp or pure Pro-Hyp at 0.2 mmol/kg. All hydrolysates supplied by Jellice (Miyagi, Japan).
Duration	Single-dose kinetics. Human plasma sampling 0–300 min. Mouse plasma and perfused-skin sampling 0–360 min (Arm 1) and 0–120 min (Arm 2).
Endpoints	Plasma concentration of 17 targeted collagen-derived peptides in humans by LC-MS/MS; Cmax; AUC and Tmax per peptide; Peptide concentration in perfused mouse skin tissue; Comparative bioavailability between HTC-col and LTC-col; Metabolic conversion of Gly-Pro-Hyp to Pro-Hyp in vivo

What the study showed, in plain terms

By 2017, researchers had already shown that some collagen peptides survive digestion and reach the bloodstream. What remained unclear was whether those peptides then actually reached the skin, where most people take collagen hoping to see a benefit. This paper set out to answer that question directly, and to test whether a newer type of collagen hydrolysate — one deliberately manufactured to contain more three-amino-acid pieces than the older two-amino-acid-dominated products — behaves differently in the body.

Twelve healthy volunteers took a single dose of one of two collagen hydrolysates, both made from pig skin but processed differently. One had a high proportion of tripeptides (three amino acids stuck together, HTC-col); the other had almost none (LTC-col). Blood samples were then taken over five hours to measure exactly which peptides showed up and at what levels. The researchers also ran a parallel experiment in mice, where they could take skin samples as well as blood samples, and they even tested what happens when you give the pure tripeptide Gly-Pro-Hyp on its own.

Two findings stand out. First, the high-tripeptide hydrolysate produced roughly ten times more Gly-Pro-Hyp in the bloodstream than the standard product — a genuinely large difference. Second, in mice, both Gly-Pro-Hyp and its two-amino-acid breakdown product Pro-Hyp did reach the skin, appearing there within about half an hour of ingestion. Giving pure Gly-Pro-Hyp also raised skin Pro-Hyp, showing that the tripeptide is partly broken down into the dipeptide inside the body. Together these findings are the closest thing yet to a direct chain of evidence from swallowed collagen supplement to peptides reaching skin tissue — though the crucial skin-uptake step was measured in mice, not humans.

Key findings

In humans, HTC-col produced a plasma Cmax of Gly-Pro-Hyp of 21.1 nmol/mL versus 2.1 nmol/mL after LTC-col — approximately 10-fold higher. AUC was approximately 8-fold higher (2,599 vs undetectable in the LTC group's AUC for this peptide).
Ala-Hyp was the most abundant plasma peptide overall in humans across both hydrolysates (Cmax ~26 nmol/mL for HTC-col, ~24 nmol/mL for LTC-col — not significantly different).
Of 24 candidate peptides screened, 17 were reliably detected in human plasma after collagen ingestion.
In mice at 900 mg/kg HTC-col, plasma Gly-Pro-Hyp reached Cmax 103 nmol/mL at Tmax 15 min; at 1,800 mg/kg it reached 191 nmol/mL — dose-dependent absorption.
Both Gly-Pro-Hyp and Pro-Hyp accumulated in perfused mouse skin tissue, with Pro-Hyp reaching approximately 15 nmol/g (900 mg/kg dose) — measurable evidence that ingested collagen peptides reach skin.
Tmax in skin (30–45 min) lagged Tmax in plasma (15 min), consistent with peptides moving from blood into skin tissue.
The plasma-to-skin ratio inverted between peptides: Gly-Pro-Hyp was 1.78–2.63-fold more concentrated in plasma than in skin, whereas Pro-Hyp was 3–5-fold more concentrated in skin than in plasma — consistent with Gly-Pro-Hyp being partially hydrolysed to Pro-Hyp during or after skin uptake.
Administration of pure Gly-Pro-Hyp raised both Gly-Pro-Hyp and Pro-Hyp in plasma and skin, directly confirming in vivo conversion of the tripeptide to the dipeptide.

What this study can and cannot tell us

Human sample of n=12 is small — the paper reports paired t-test p-values without multiple-comparison correction across 17 measured peptides, so the strongest single-peptide finding (Gly-Pro-Hyp) is robust but many of the smaller differences should be treated as exploratory.
Single-dose kinetics only. No evidence on how peptide plasma or skin levels behave under repeated daily dosing over weeks or months, which is how consumers actually take these products.
The critical skin-uptake evidence — the paper's most novel contribution — was measured in mice, not humans. Species differences in skin structure and peptide transporter expression mean the extrapolation to human skin is plausible but not proven.
Presence of a peptide in skin tissue does not itself demonstrate biological activity there. The paper does not measure downstream outcomes such as fibroblast proliferation, collagen synthesis, hyaluronic acid production, or clinical endpoints.
Substantial industry conflict: seven of eight co-authors are FANCL employees, HTC-col is a FANCL commercial product, and both hydrolysates were provided by a commercial supplier (Jellice). No explicit competing interests statement provided.
Both hydrolysates are porcine skin type I. The findings do not necessarily extend to marine, bovine, or chicken collagen sources.
The dose used in mice (900–1,800 mg/kg) is far higher on a body-weight basis than the typical human supplemental dose of ~2.5–10 g/day — direct dose translation to human tissue exposure is not straightforward.

Tier 3

Identification of food-derived collagen peptides in human blood after oral ingestion of gelatin hydrolysates

Koji Iwai et al. · Journal of Agricultural and Food Chemistry · 2005