PYY is a negative regulator of bone mass and strength

Abstract

Objective: Bone loss in anorexia nervosa and following bariatric surgery is associated with an elevated circulating concentration of the gastrointestinal, anorexigenic hormone, peptide YY (PYY). Selective deletion of the PYY receptor Y1R in osteoblasts or Y2R in the hypothalamus results in high bone mass, but deletion of PYY in mice has resulted in conflicting skeletal phenotypes leading to uncertainty regarding its role in the regulation of bone mass. As PYY analogs are under development for treatment of obesity, we aimed to clarify the relationship between PYY and bone mass.

Methods: The skeletal phenotype of Pyy knockout (KO) mice was investigated during growth (postnatal day P14) and adulthood (P70 and P186) using X-ray microradiography, micro-CT, back-scattered electron scanning electron microscopy (BSE-SEM), histomorphometry and biomechanical testing.

Results: Bones from juvenile and Pyy KO mice were longer (P < 0.001), with decreased bone mineral content (P < 0.001). Whereas, bones from adult Pyy KO mice had increased bone mineral content (P < 0.05) with increased mineralisation of both cortical (P < 0.001) and trabecular (P < 0.001) compartments. Long bones from adult Pyy KO mice were stronger (maximum load P < 0.001), with increased stiffness (P < 0.01) and toughness (P < 0.05) compared to wild-type (WT) control mice despite increased cortical vascularity and porosity (P < 0.001). The increased bone mass and strength in Pyy KO mice resulted from increases in trabecular (P < 0.01) and cortical bone formation (P < 0.05).

Conclusions: These findings demonstrate that PYY acts as a negative regulator of osteoblastic bone formation, implicating increased PYY levels in the pathogenesis of bone loss during anorexia or following bariatric surgery.

Keywords: Bone mineral density; Fracture; Osteoblast; Osteoporosis; PYY.

Fig. 1

Linear growth and endochondral ossification. (A) Femur lengths and caudal vertebral heights from male and female P14, and male P70 and P186 WT and Pyy KO mice (mean ± SEM, n = 7–8 per genotype per age, *P < 0.05, **P < 0.01, ***P < 0.001 versus WT; unpaired t-test). (B) Decalcified sections of P14 proximal tibia stained with Alcian blue (cartilage) and van Gieson (bone matrix); scale bars = 100 μm. Graphs represent the absolute and relative widths of the resting zone (RZ), proliferating zone (PZ) and hypertrophic zone (HZ) chondrocytes in the growth plate of WT and Pyy KO mice (mean ± SEM, n = 5 per genotype, *P < 0.05, versus WT; unpaired t-test). (C) Representative quantitative X-ray microradiographic images of femurs from male and female P14 and male P70 and P186 WT and Pyy KO mice; scale bar = 1 mm. Pseudo-coloured images represent grey scale images using a 16-color interval scheme with low mineral content blue and high mineral content pink. Graphs are relative frequency histograms of bone mineral content (BMC) (n = 7–8 per genotype per age, *P < 0.05, ***P < 0.001 versus WT; Kolmogorov-Smirnov test). (D) Quantitative X-ray microradiographic images of vertebrae from male and female P14 and male P70 and P186 mice; scale bar = 1 mm. Relative frequency histograms of BMC (n = 7–8 per genotype per age, ***P < 0.001 versus WT; Kolmogorov-Smirnov test).

Fig. 2

Bone structure and mineralisation. (A) Transverse micro-CT rendered images of midshaft femur from male P112 WT and Pyy KO mice; scale bar 200 μm. Graphs show cortical bone structural parameters; cortical thickness (C.Th), marrow cavity diameter (Ma.Dm) and bone mineral density (BMD) (mean ± SEM, n = 4–5 per genotype, **P < 0.01 versus WT; unpaired t-test). (B) Representative BSE-SEM images of distal femur trabecular bone from male P70 WT and Pyy KO mice (n = 4 per genotype); scale bar = 200 μm. Graphs show trabecular bone structural parameters determined by microCT; bone volume as a percentage of tissue volume (BV/TV), trabecular number (Tb.N) and trabecular thickness (Tb.Th) in male P112 WT and Pyy KO mice (mean ± SEM, n = 5 per genotype, *P < 0.05 versus WT; unpaired t-test). (C) Quantitative BSE-SEM images of cortical bone from the proximal humeri of male P70 and P186 WT and Pyy KO mice; scale bar = 200 μm. Pseudo-coloured images represent grey scale images using an 8-color interval scheme with low mineral content green/yellow and high mineral content pink/grey. Graphs are relative frequency histograms of bone micro-mineralisation densities (images representative of n = 5 per genotype, ***P < 0.001 versus WT; Kolmogorov-Smirnov test). (D) Quantitative BSE-SEM images of trabecular bone from the proximal humeri of male P70 and P186 WT and Pyy KO mice; scale bar = 200 μm. Relative frequency histograms of bone micro-mineralisation densities (images representative of n = 5 per genotype, ***P < 0.001 versus WT; Kolmogorov-Smirnov test).

Fig. 3

Osteoclastic bone resorption and osteoblastic bone formation. (A) Decalcified sections of P70 proximal tibia from male P70 WT and Pyy KO mice stained for tartrate-resistant acid phosphatase (TRAP); scale bar = 200 μm. Black arrow indicates examples of red TRAP-stained osteoclasts. Graphs show numbers of osteoclasts per mm bone surface (OcN/BS) and osteoclast surface per mm bone surface (OcS/BS) in male WT and Pyy KO mice (mean ± SEM, n = 3 per genotype). (B) BSE-SEM images of femur endosteal bone surfaces from male P70 WT and Pyy KO mice. Arrows indicate borders between regions of osteoclastic resorption and unresorbed bone surfaces; scale bar = 200 μm. Graphs show endosteal and trabecular resorption surfaces as percentage of total endosteal and trabecular bone surface respectively (mean ± SEM, n = 4 per genotype). (C) Confocal images of trabecular bone from proximal humerus of male P70 WT and Pyy KO mice double-labelled with calcein; scale bar = 200 μm. Graphs show trabecular mineralising surface (MS/BS), mineral apposition rate (MAR) and bone formation rate (BFR) (mean ± SEM, n = 5 per genotype, *P < 0.05, **P < 0.01 versus WT; unpaired t-test). (D) Confocal images of cortical bone from proximal humerus of male mice; scale bar = 200 μm. Graphs show cortical MS/BS, MAR and BFR (mean ± SEM, n = 5 per genotype, *P < 0.05, versus WT; unpaired t-test).

Fig. 4

Bone strength and cortical vascularity. (A) Representative load displacement curves from 3-point bend testing of tibias from male P70 WT and Pyy KO mice. Graphs show yield, maximum and fracture loads, stiffness and energy dissipated prior to fracture (toughness) (mean ± SEM, n = 7–8 per genotype, *P < 0.05, **P < 0.01, ***P < 0.001 versus WT; unpaired t-test). (B) BSE-SEM images of mid-femur endosteal surfaces from male P70 WT and Pyy KO male mice; scale bar = 200 μm. Graphs show endosteal surface vessel density and relative frequency histogram of vessel size (mean ± SEM, n = 4 per genotype, **P < 0.01 versus WT; unpaired t-test). (C) Representative load displacement curves from 3-point bend testing of tibias from male P186 WT and Pyy KO mice. Graphs show yield, maximum and fracture loads, stiffness and energy dissipated prior to fracture (toughness) (mean ± SEM, n = 7–8 per genotype, *P < 0.05, **P < 0.01, versus WT; unpaired t-test). (D) BSE-SEM images of mid-femur endosteal surfaces from female P186 WT and Pyy KO mice; scale bar = 200 μm. Graphs show endosteal surface vessel density and relative frequency histogram of vessel size (mean ± SEM, n = 4 per genotype, ***P < 0.001 versus WT; Kolmogorov-Smirnov test).

Fig. 5

Cortical porosity. (A) Transverse and anterior-posterior micro-CT images showing cortical bone porosity in the mid-femoral diaphysis of male P70 Pyy KO and WT mice; bars = 200 μm. Graphs show mean canal diameter (Ca.Dm) and cortical porosity (Ct.Po) (Pore volume/Cortical bone volume (Po·V/Ct.V)) (mean ± SEM, n = 6 per genotype, **P < 0.01 versus WT; unpaired t-test). (B) Micro-CT images showing cortical bone porosity in distal-femoral diaphysis of male P70 Pyy KO and WT mice; bars = 200 μm. Graphs show mean canal diameter (Ca.Dm) and cortical porosity (Ct.Po) (mean ± SEM, n = 6 per genotype, ***P < 0.001 versus WT; unpaired t-test).