FGF21, not GCN2, influences bone morphology due to dietary protein restrictions

Abstract

Background Dietary protein restriction is emerging as an alternative approach to treat obesity and glucose intolerance because it markedly increases plasma fibroblast growth factor 21 (FGF21) concentrations. Similarly, dietary restriction of methionine is known to mimic metabolic effects of energy and protein restriction with FGF21 as a required mechanism. However, dietary protein has been shown to be required for normal bone growth, though there is conflicting evidence as to the influence of dietary protein restriction on bone remodeling. The purpose of the current study was to evaluate the effect of dietary protein and methionine restriction on bone in lean and obese mice, and clarify whether FGF21 and general control nonderepressible 2 (GCN2) kinase, that are part of a novel endocrine pathway implicated in the detection of protein restriction, influence the effect of dietary protein restriction on bone.

Methods Adult wild-type (WT) or Fgf21 KO mice were fed a normal protein (18 kcal%; CON) or low protein (4 kcal%; LP) diet for 2 or 27 weeks. In addition, adult WT or Gcn2 KO mice were fed a CON or LP diet for 27 weeks. Young New Zealand obese (NZO) mice were placed on high-fat diets that provided protein at control (16 kcal%; CON), low levels (4 kcal%) in a high-carbohydrate (LP/HC) or high-fat (LP/HF) regimen, or on high-fat diets (protein, 16 kcal%) that provided methionine at control (0.86%; CON-MR) or low levels (0.17%; MR) for up to 9 weeks. Long bones from the hind limbs of these mice were collected and evaluated with micro-computed tomography (μCT) for changes in trabecular and cortical architecture and mass.

Results In WT mice the 27-week LP diet significantly reduced cortical bone, and this effect was enhanced by deletion of Fgf21 but not Gcn2. This decrease in bone did not appear after 2 weeks on the LP diet. In addition, Fgf21 KO mice had significantly less bone than their WT counterparts. In obese NZO mice dietary protein and methionine restriction altered bone architecture. The changes were mediated by FGF21 due to methionine restriction in the presence of cystine, which did not increase plasma FGF21 levels and did not affect bone architecture.

Conclusions This study provides direct evidence of a reduction in bone following long-term dietary protein restriction in a mouse model, effects that appear to be mediated by FGF21.