Tier 4 — mechanistic

Assessment of effectiveness of oral administration of collagen peptide on bone metabolism in growing and mature rats

Jian Wu, Maiko Fujioka, Kiho Sugimoto, Gang Mu, Yoshiko Ishimi
Journal of Bone and Mineral Metabolism 2004 22(6):547–553

Bibliography

PubMed
PMID 15490264
Funding
Explicit funding statement not surfaced in the extracted PDF text. Authors were affiliated with the Division of Applied Food Research, National Institute of Health and Nutrition, Tokyo, Japan — a government research institute.
Competing interests
Explicit funding statement not surfaced. Authors were affiliated with the Division of Applied Food Research, National Institute of Health and Nutrition, Tokyo, Japan — a government research institute.

Study snapshot

DesignTwo parallel rat experiments — dose-ranging in growing males and calcium-deficient adult females — with control and multiple collagen peptide doses
ModelGrowing male rats (Sprague-Dawley) for dose-response; ovariectomy-mimicking calcium-deficient adult female rats for efficacy in bone loss
SampleMultiple groups across two experiments — Experiment 1: growing males in control, Coll-1G (0.166 g/kg BW), Coll-10G (1.66 g/kg BW), Coll-100G (16.6 g/kg BW). Experiment 2: adult females in control, Coll-1M, Coll-10M, and normal-calcium groups. Group sizes 6–10 per arm.
InterventionOral collagen peptide at doses equivalent to (0.166 g/kg BW/day), 10-fold (1.66 g/kg BW/day), or 100-fold (16.6 g/kg BW/day) the recommended human supplemental dose, for 4 weeks (growing) or 8 weeks (adult calcium-deficient)
DurationExperiment 1: 4 weeks. Experiment 2: 9 weeks calcium-depletion followed by 8 weeks intervention (17 weeks total)
EndpointsFemoral BMD by DXA; Lumbar spine BMD by DXA; Kidney histology and organ weight; Body weight; Serum bone turnover markers

What the study showed, in plain terms

Before any dietary supplement can be recommended for widespread daily use, someone needs to test what happens at very high doses. If a supplement is safe at ten times or one hundred times the recommended intake, that safety margin gives regulators and consumers confidence that a normal daily dose is nowhere near a danger threshold. For oral collagen peptides, this 2004 Japanese study filled that gap.

Researchers at Japan's National Institute of Health and Nutrition fed collagen peptides to two groups of rats. In the first experiment, growing male rats received either a dose equivalent to the recommended human daily amount, ten times that amount, or one hundred times that amount, every day for four weeks. In the second experiment, adult female rats were first put on a low-calcium diet for nine weeks to make them lose bone density, then treated with collagen peptides for eight weeks.

The results carry two important messages. First, in the growing rats, the top dose (one hundred times recommended) did increase bone density in the leg bones — but it also caused the kidneys to enlarge, a warning sign of kidney stress. Second, in the adult calcium-deficient females, ten times the recommended dose restored bone density in the thigh and lower spine to levels seen in normally-fed rats, with no obvious kidney or other adverse effects. This is the paper cited by Iwai 2005 and later collagen safety reviews to support the position that normal supplemental doses (up to roughly 15 g/day in humans) fall well within the safety margin, but that pushing much higher may risk kidney effects. Every safety discussion of oral collagen ultimately traces back to this study.

Key findings

  • Growing rats at the recommended-equivalent dose (Coll-1G) and 10-fold dose (Coll-10G) showed no significant change in femoral BMD or kidney histology compared to controls.
  • At 100-fold the recommended dose (Coll-100G), femoral BMD increased significantly compared to other groups, but kidneys exhibited hypertrophy — a warning signal for renal safety at extreme doses.
  • In calcium-deficient adult female rats, the 10-fold dose (Coll-10M) restored femoral BMD to levels similar to those in the normal-calcium reference group, and significantly higher than the control (0.2% calcium alone) and Coll-1M groups.
  • Lumbar spine BMD in Coll-10M group was significantly higher than baseline value and significantly higher than control and Coll-1M groups.
  • The recommended-equivalent dose (Coll-1M) did not significantly improve BMD — suggesting the effect requires a supra-standard dose in this animal model of osteoporosis.
  • No obvious adverse effects observed in the Coll-10M group — 10-fold dose appears safe in calcium-deficient adult rats.
  • Provides the primary preclinical safety-margin data cited by subsequent human collagen supplementation safety discussions.

What this study can and cannot tell us

  • Rat study — species differences in intestinal peptide transport, hepatic first-pass metabolism, and renal handling of amino acids all mean direct dose translation to humans is not straightforward.
  • Rat weight-normalised doses do not translate 1:1 to human weight-normalised doses due to different metabolic scaling.
  • The 100-fold dose that caused kidney hypertrophy in growing rats corresponds to a very high human intake — roughly 1 kg/day of collagen for a 60 kg adult — vastly above any realistic supplement dose. The safety implication is favourable for consumers, but the mechanism of the kidney effect was not fully investigated in this trial.
  • Small group sizes typical of preclinical work — statistical power for detecting subtle effects is limited.
  • Only bone and gross kidney outcomes measured — no assessment of liver function, cardiovascular effects, or other organ systems at high dose.
  • Only one collagen peptide source used — generalisability to hydrolysates from other sources is not established.
  • Explicit funding and competing interests statements not surfaced in the extracted text — verify from the full PDF.
Reviewed by , Medical Advisory Board · Last verified against PubMed on 30 August 2026