Osteocalcin
Osteocalcin, bone gamma-carboxyglutamate protein (BGLAP), including undercarboxylated osteocalcin
Written by Aaron CuhaReviewed Sep 2026
Also known as: ucOC, Undercarboxylated osteocalcin, Bone Gla protein, BGLAP
The bone hormone that controlled muscle, glucose and testosterone in mice, until two independent laboratories made new osteocalcin knockout strains in 2020 and found none of those phenotypes. In humans the causal arrow, as far as genetics can resolve it, appears to run the other way.
Overview
Osteocalcin is the most abundant non-collagen protein in bone, made by osteoblasts and gamma-carboxylated in a vitamin K dependent step. From the mid-2000s it became something much more interesting: a bone-derived endocrine hormone. Mice lacking osteocalcin were reported to be glucose intolerant and infertile, and a 2016 Cell Metabolism paper reported that osteocalcin signalling in muscle fibres is both necessary and sufficient for optimum adaptation to exercise. Osteocalcin released during exercise, acting on muscle, is the most attractive version of the exerkine story: bone talking to muscle.
Then the knockouts were remade. In May 2020 two laboratories published back to back in PLoS Genetics. A Japanese group deleted both mouse osteocalcin genes and found osteocalcin was required for the alignment of bone apatite crystallites, and therefore for bone strength, but was not involved in glucose metabolism, testosterone synthesis or muscle mass. A US group independently made a CRISPR double knockout with no detectable circulating osteocalcin and found no difference from wild-type littermates in serum glucose or male fertility. Their paper says plainly that they cannot explain the absence of endocrine effects seen in the older strain.
The human evidence never got past association, and where genetics can speak it points the other way. A 2021 genome-wide association and Mendelian randomisation study in 5,169 Chinese participants found circulating osteocalcin correlated with better metabolic profiles in the expected direction, but the causal analysis supported type 2 diabetes causing lower osteocalcin, not osteocalcin protecting against diabetes. The muscle association in humans is one post hoc analysis in older women, in which the ratio of undercarboxylated to total osteocalcin, but not the absolute concentration, correlated with hip and quadriceps strength.
Mechanism of action
Osteocalcin is a small gamma-carboxylated peptide secreted by osteoblasts. Its established function is structural: it aligns apatite crystallites parallel to collagen fibrils, and without it bone strength falls even though bone quantity is normal. The proposed endocrine function, decarboxylation during bone resorption releasing undercarboxylated osteocalcin that acts at GPRC6A on pancreatic beta cells, Leydig cells and myofibres, is the part that failed to replicate in new knockout strains. Whether the discrepancy comes from effects of the original targeting construct on neighbouring genes, genetic background or environment is stated as unexplained by the authors who found the null.
Human evidence
All human data is observational or genetic. Nobody has given osteocalcin to a person, so there is no interventional human evidence at all.
- Post hoc analysis in women over 70: undercarboxylated osteocalcin as a percentage of total correlated with hip and quadriceps strength; the absolute concentration did not.
- Mendelian randomisation in East Asian cohorts: the causal direction supported is type 2 diabetes reducing circulating osteocalcin, rather than osteocalcin improving glucose control.
- Circulating osteocalcin is a widely used bone formation marker, and it does track bone turnover reliably, which is a different claim from the hormone story.
- No trial has administered osteocalcin or an undercarboxylated osteocalcin preparation to a human.
What this does not tell you: Observational osteocalcin concentrations move with bone turnover, which itself moves with exercise, age, vitamin K and vitamin D status, and metabolic disease. That makes cross-sectional correlations with muscle strength or glucose close to uninterpretable as causal claims. Mendelian randomisation constrains the direction of effect but rests on a small proportion of variance explained by the identified loci, and was run in East Asian cohorts. Nothing in the human literature tests whether raising osteocalcin does anything.
Reading the research record
This is a textbook case of a striking mouse result narrowing under replication, and it is worth reading carefully because the same pattern recurs across exerkine biology.
The original endocrine findings came from one knockout allele made by homologous recombination in ES cells. Two independent groups, publishing in the same journal on the same day in May 2020, made new osteocalcin-null mice by different methods and found the bone phenotype but none of the endocrine phenotypes. Both papers are careful, both say the discrepancy is unexplained, and one of them deliberately deposited the strain publicly so anyone can test it. Candidate explanations include effects of the original construct on neighbouring gene transcription, genetic background, and environment.
What survived is not nothing. Osteocalcin turned out to be necessary for the crystallographic alignment that gives bone its strength, which is a real and useful finding about bone quality. What did not survive is the part the supplement and biohacking literature repeats: that osteocalcin is the bone hormone that drives muscle adaptation, testosterone and glucose handling.
The human data never tested that claim at all. It measured associations, and the one genetic analysis able to speak to direction suggests metabolic disease lowers osteocalcin rather than the other way around. No one sells osteocalcin, because it cannot be taken orally and has never been given to a person, but vitamin K supplements are sold on the strength of the undercarboxylated osteocalcin story, which rests on exactly this literature.
The evidence, charted
Fig. 1 · evidence composition
2of 5 citations (40%) are in people
Counted from the citation list on this page. The Human count is the same number shown in the badge at the top. Both understate any literature larger than the sources we cite.
Fig. 2 · evidence over time
Evidence spans 4 distinct years, 2014 to 2021, counted from the citation list on this page.
Fig. 3 · legal status at a glance
US
Not approved
UK
Not approved
AU
Not approved
CA
Not approved
Not approved in any of the four jurisdictions shown. A jurisdiction's classification is a regulatory fact, not a verdict on the science; see Legal status below for the full text and any notes.
Key studies & citations
- Animal2016
Osteocalcin Signaling in Myofibers Is Necessary and Sufficient for Optimum Adaptation to Exercise
The mouse paper that made osteocalcin an exerkine. Osteocalcin signalling in muscle fibres was reported as necessary and sufficient for optimum exercise adaptation in mice. No human intervention component.
Cell Metabolism - Animal2020
Osteocalcin is necessary for the alignment of apatite crystallites, but not glucose metabolism, testosterone synthesis, or muscle mass
A new osteocalcin-deficient mouse made by deleting Bglap and Bglap2. Bone quantity, glucose metabolism, testosterone synthesis and muscle mass were all normal. What was abnormal was the crystallographic orientation of bone apatite, and bone strength fell as a result. The authors conclude osteocalcin does not function as a hormone.
PLoS Genetics - Animal2020
An osteocalcin-deficient mouse strain without endocrine abnormalities
An independent CRISPR double knockout with no immunodetectable circulating osteocalcin. Collagen maturity and carbonate to phosphate ratio changed in cortical bone, but bone mass, three-point bending strength, serum glucose and male fertility did not differ from wild-type littermates. The authors state they cannot explain the absence of the endocrine effects reported in the older strain, and donated the line to Jackson Laboratories so others could check.
PLoS Genetics - Human2014
Undercarboxylated osteocalcin, muscle strength and indices of bone health in older women
Post hoc analysis of a randomised vitamin D trial in women over 70. The ratio of undercarboxylated to total osteocalcin, but not the absolute undercarboxylated concentration, was positively associated with hip flexor, hip abductor and quadriceps strength (all p < 0.05). Association, in an analysis that was not the trial's purpose.
Bone - Human2021
Type 2 Diabetes Is Causally Associated With Reduced Serum Osteocalcin: A Genomewide Association and Mendelian Randomization Study
5,169 Chinese participants with measured osteocalcin, plus bidirectional two-sample Mendelian randomisation against Biobank Japan and East Asian diabetes GWAS. Osteocalcin correlated with lower glucose, insulin resistance, triglycerides, liver fat and BMI, but the causal analysis supported type 2 diabetes lowering osteocalcin (causal effect -0.03, -0.05 to -0.01, p = 0.006 and p = 0.001 in two datasets), not the reverse.
Journal of Bone and Mineral Research
Frequently asked questions
Does exercise raise osteocalcin and build muscle through it?
Exercise does raise osteocalcin, because it raises bone turnover. The claim that osteocalcin then acts on muscle to produce the adaptation comes from mice, and two new osteocalcin knockout strains in 2020 found normal muscle mass. There is no human interventional evidence for the bone to muscle pathway.
Should I take vitamin K2 to increase undercarboxylated osteocalcin?
Vitamin K2 does the opposite: it carboxylates osteocalcin, which lowers the undercarboxylated fraction. The hormone story proposes that the undercarboxylated form is the active one. The two claims are frequently sold together, and they contradict each other.
Is a high osteocalcin blood test a good sign?
It is a bone turnover marker. High values indicate high bone formation activity, which is expected in growth, in healing, and in conditions with high remodelling. It is not a fitness or metabolic health score, and it is interpreted alongside other bone markers by the clinician who ordered it.
Can you buy osteocalcin?
No, not in any form that would work. It is a gamma-carboxylated peptide that has never been given to a human in a trial, and there is no approved product anywhere.
So was the mouse work wrong?
The fairest summary is that it did not replicate in independently generated mouse strains, and the authors who failed to replicate it say openly that they cannot explain why. The structural role of osteocalcin in bone quality held up. The endocrine role did not.