The half-life of the bone-derived hormone osteocalcin is regulated through O-glycosylation in mice, but not in humans
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
Osteocalcin (OCN) is an osteoblast-derived hormone with pleiotropic physiological functions. Like many peptide hormones, OCN is subjected to post-translational modifications (PTMs) which control its activity. Here, we uncover O -glycosylation as a novel PTM present on mouse OCN and occurring on a single serine (S8) independently of its carboxylation and endoproteolysis, two other PTMs regulating this hormone. We also show that O -glycosylation increases OCN half-life in plasma ex vivo and in the circulation in vivo. Remarkably, in human OCN (hOCN), the residue corresponding to S8 is a tyrosine (Y12), which is not O- glycosylated. Yet, the Y12S mutation is sufficient to O -glycosylate hOCN and to increase its half-life in plasma compared to wildtype hOCN. These findings reveal an important species difference in OCN regulation, which may explain why serum concentrations of OCN are higher in mouse than in human.
Article activity feed
-
-
###This manuscript is in revision at eLife
The decision letter after peer review, sent to the authors on August 10 2020, follows.
Summary
The work establishes that mouse osteocalcin is O-glycosylated at Ser-8, independently of processing and gamma-carboxylation. Human osteocalcin was found not to be O-glycosylated, but mutation to Ser-8 allowed that process to occur. If increased stability of osteocalcin in vitro as a result of O-glycosylation is real, it could be of interest. The paper was found to be well written with figures illustrating the findings. With that said, several key experiments are required that would considerably strengthen the conclusions. Notably, without information on the biological outcomes of O-glycosylation, the paper is seen to be of limited interest.
Essential Revisions
Regarding the ELISA (ref. Ferron et al …
###This manuscript is in revision at eLife
The decision letter after peer review, sent to the authors on August 10 2020, follows.
Summary
The work establishes that mouse osteocalcin is O-glycosylated at Ser-8, independently of processing and gamma-carboxylation. Human osteocalcin was found not to be O-glycosylated, but mutation to Ser-8 allowed that process to occur. If increased stability of osteocalcin in vitro as a result of O-glycosylation is real, it could be of interest. The paper was found to be well written with figures illustrating the findings. With that said, several key experiments are required that would considerably strengthen the conclusions. Notably, without information on the biological outcomes of O-glycosylation, the paper is seen to be of limited interest.
Essential Revisions
Regarding the ELISA (ref. Ferron et al 2010b), capture antibodies can distinguish between carboxylated and non-carboxylated OCN. For the present work, it is essential that the authors show that OCN with or without O-glycosylation at Ser-8 is measured identically in this ELISA. The authors should specify what capture antibodies were used. This query applies also to data with human OCN and the Ser-8 mutant that is O-glycosylated. Furthermore, does the commercial ELISA kit measure glycosylated mutant and normal hOCN as identical?
Figs 3H-J show a statistically significant difference in levels of glycosylated and non-glycosylated mouse OCN. These experiments however do not measure "half life" as claimed in the title and abstract. The in vivo half life of injected O-glycosylated vs wt ucOCN should therefore be compared using timed estimations during the declining phase.
A feature of OCN that interested the authors was the remarkable difference in circulating amounts - more than 10 times higher in the mouse. This work appears to be part of a search for mechanisms to explain this, although they might consider that these are evolutionary changes, including the fact that there is only 65 % conservation of sequence between mouse and human , and the human OCN is not O-glycosylated, whereas the mouse OCN is. The biological significance of this difference in O-glycosylation thus needs to be established. While knocking-in a mutation to abolish O-glycosylation will provide the most definitive answer, the reviewers consider this not to be feasible during the pandemic. Therefore, at the very least, a cell-based assay should be used to compare biological activity. Examples are the dose-dependent increase in insulin mRNA in mouse pancreatic islets (Ferron. et al, PNAS 105: 5266, 2008). Very low dose ucOCN was shown in that work to promote insulin expression in islets in a dose-dependent manner. An alternative approach, arising from the same paper, would be to show at slightly higher doses, a dose-dependent increase in adiponectin expression in mouse adipocytes. The authors might have other possibilities of cell-based assay.
-
