Absence of P2Y2 Receptor Does Not Prevent Bone Destruction in a Murine Model of Muscle Paralysis-Induced Bone Loss

Abstract

Increased incidence of bone fractures in the elderly is associated with gradual sarcopenia. Similar deterioration of bone quality is seen with prolonged bed rest, spinal cord injuries or in astronauts exposed to microgravity and, preceded by loss of muscle mass. Signaling mechanisms involving uridine-5'-triphosphate (UTP) regulate bone homeostasis via P2Y₂ receptors on osteoblasts and osteoclasts, whilst dictating the bone cells' response to mechanical loading. We hypothesized that muscle paralysis-induced loss of bone quality would be prevented in P2Y₂ receptor knockout (KO) mice. Female mice injected with botulinum toxin (BTX) in the hind limb developed muscle paralysis and femoral DXA analysis showed reduction in bone mineral density (<10%), bone mineral content (<16%) and bone area (<6%) in wildtype (WT) compared to KO littermates (with <13%, <21%, <9% respectively). The femoral metaphyseal strength was reduced equally in both WT and KO (<37%) and <11% in diaphysis region of KO, compared to the saline injected controls. Tibial micro-CT showed reduced cortical thickness (12% in WT vs. 9% in KO), trabecular bone volume (38% in both WT and KO), trabecular thickness (22% in WT vs. 27% in KO) and increased SMI (26% in WT vs. 19% in KO) after BTX. Tibial histomorphometry showed reduced formation in KO (16%) but unchanged resorption in both WT and KO. Furthermore, analyses of DXA and bone strength after regaining the muscle function showed partial bone recovery in the KO but no difference in the bone recovery in WT mice. Primary osteoblasts from KO mice displayed increased viability and alkaline phosphatase activity but, impaired bone nodule formation. Significantly more TRAP-positive osteoclasts were generated from KO mice but displayed reduced resorptive function. Our data showed that hind limb paralysis with a single dose of BTX caused profound bone loss after 3 weeks, and an incomplete reversal of bone loss by week 19. Our findings indicate no role of the P2Y₂ receptor in the bone loss after a period of skeletal unloading in mice or, in the bone recovery after restoration of muscle function.

Keywords: P2Y2 receptor; bone; botulinum toxin; skeletal reloading; skeletal unloading.

Figure 1

(A) The study design to show the unloading (weeks 17 – 19) and remobilization (weeks 20 – 35) phases in BTX injected 16-week old female BALB/cJ mice. Image created by Biorender.com. (B) Digit abduction score (DAS) assay to confirm BTX- induced muscle paralysis in both WT and KO mice from day 1 of BTX injection (dotted line) compared to the saline injected mice (solid line). Disability was monitored for 21 days to show the mean peak DAS values at day 4 in both WT and KO. BTX KO, a lower DAS value at day 7 in BTX KO compared to BTX WT (a, P < 0.05) and recovery of muscle function in both WT and KO at the end of 21 days. Statistical significance at day 7 was tested using Student’s unpaired t- test, data shows mean ± SEM of n = 15- 16 mice (WT) and 18- 19 mice (KO) in each treatment condition. (C) Weight changes during the unloading phase, expressed as a percentage change from baseline body weight (BL, 16-week old) in WT and KO mice injected with either saline (solid line) or BTX (dotted line). Significant body weight reduction after BTX (a, P < 0.05 in WT and b, P < 0.05 in KO) compared to the saline controls and, in BTX KO compared to the BL (c, P < 0.05). Statistical significance was tested using 2- way ANOVA with Tukey’s test for multiple comparisons. Data shows mean ± SEM of n = 16 mice (WT) and 19 mice (KO) in each treatment group. (D) DXA assessments of femoral bone mineral density (BMD, grams/cm²), bone mineral content (BMC, grams) and bone area (cm²) and, (E) femoral maximum load (N), at metaphysis and diaphysis in both WT and KO after BTX- induced muscle paralysis. Statistical significance was tested with either Student’s unpaired t-test/Mann-Whitney test (effect of KO) or, 2-way ANOVA (effect of BTX and KO) with Tukey’s test for multiple comparisons; n.s., no significance. Data shows mean ± SEM of n = 15-18 mice (WT) and 23-24 mice (KO) for DXA and n= 6-15 mice (WT) and 15-19 mice (KO) for bone strength.

Figure 2

Tibial microstructural parameters determined after the unloading phase in both WT and KO at 3- weeks after BTX- induced muscle paralysis. (A) cortical bone mineral density (BMD, mg HA/ccm), bone volume (BV/TV, %), pore size (µm) and thickness (µm), and (B) representative images to show the changes in cortical indices. (C) Trabecular BMD (mg HA/ccm), BV/TV (%), structure model index (SMI, -), trabecular number (1/mm), trabecular thickness (µm), trabecular separation (µm), and trabecular degree of anisotropy (DA, -) and (D) representative images to show the changes in trabecular indices. (E) Histomorphometric indices to show to Md.S/BS (mineralized surfaces as percentage of bone surface), MAR (mineral apposition rate in µm/day) and ES/BS (eroded surface as percentage of bone surface) in vivoin both WT and KO mice at 3- weeks after BTX- induced muscle paralysis. (F) Illustrations of calcein labeled and (G) Goldner’s trichrome stained slices to quantify the histomorphometric indices. Statistical significance was tested with either Student’s unpaired t-test/Mann-Whitney test (effect of KO) or, 2-way ANOVA (effect of BTX and KO) with Tukey’s test for multiple comparisons. n.s., no significance. Data shows mean ± SEM of n = 6-7 mice (WT) and 8-9 mice (KO) for Micro-CT and, n= 8- 10 mice (WT) and 7-9 mice (KO) for histomorphometry.

Figure 3

(A) Weight changes during the unloading and remobilization phase, expressed as a percentage change from baseline body weight (BL, 16-week old) in WT and KO mice injected with either saline (solid line) or BTX (dotted line). Significant body weight reduction after BTX (a, P < 0.05 in WT and b, P < 0.05 in KO) compared to the saline controls and, in compared to the BL (c, P < 0.05 in WT; d, P < 0.05 in KO and e, P < 0.05 in BTX KO). Statistical significance was tested using 2- way ANOVA with Tukey’s test for multiple comparisons. Data shows mean ± SEM of n = 9 - 10 mice (WT) and 10 mice (KO) in each treatment group. (B) DXA assessments of femoral bone mineral density (BMD, grams/cm²), bone mineral content (BMC, grams) and bone area (cm²) and, (C) femoral maximum load (N), at metaphysis and diaphysis in both WT and KO at 19- weeks after BTX- induced muscle paralysis. Statistical significance was tested with either Student’s unpaired t-test/Mann-Whitney test (effect of KO) or, 2-way ANOVA (effect of BTX and KO) with Tukey’s test for multiple comparisons. n.s., no significance. Data shows mean ± SEM of n = 9-10 mice (WT) and 10 mice (KO) for DXA and n= 7-10 mice (WT) and 6-10 mice (KO) for bone strength.

Figure 4

Primary bone cells were derived from 8-week-old WT and KO mice to determine the formation and functional characteristics. (A) Representative images to show no morphological differences between the unstained osteoblasts derived from long bone explants (dark spots represent the chopped bone pieces) at initial out-growth (top panel) and as confluent monolayers before subculture (bottom panel). Scale bar, 200 μm. (B) Osteoblast viability (WST-1 activity, (C) alkaline phosphatase activity (ALP activity, normalized to the amount of DNA per sample) and (D) bone nodule formation (% of AR-S stained area). (E) Representative images to illustrate the ALP staining (blue stain, top panel) and AR-S staining (brown stain showing mineralized bone structures, bottom panel) between WT and KO osteoblasts. Scale bar, 200 μm. (F) Osteoclast number (TRAP positive cells), (G) the resorption area excavated from the dentine substrate (expressed as a percentage of total dentine surface). (H) Representative images to illustrate the multinucleated TRAP positive osteoclasts (black arrows, top panel, scale bar, 50 μm) and resorption pits (white stars, bottom panel, scale bar, 200 μm) excavated by osteoclasts (black arrows, bottom panel) differentiated from the spleen of WT and KO littermate pairs. Statistical significance was tested with either Student’s unpaired t-test/Mann-Whitney test where *P-values < 0.05, **P-values < 0.01 show significance from WT. Data shows mean ± SEM of n = 5 littermate mice (osteoblasts) and n= littermate 4 mice (osteoclasts) repeat experiments with 6-10 replicate wells per experiment.

Figure 5

Expression of the P2Y₂ receptor in (A) whole bone marrow (WBM) and spleen, (B) osteoblast precursors and osteoblasts and, (C) osteoclast precursors and osteoclasts. cDNA was reverse transcribed after tissue lysis and probed using the 2 different primer sets designed between 552 bp – 1149 bp of the P2ry2 gene transcript (corresponding to the region of the targeting vector used to create the gene knock out). Products of 389 bp (corresponding to 676 bp -1065 bp) and 245 bp (corresponding to 612 bp to 857 bp) were amplified in osteoblast precursors and osteoblasts of KO mice. Col1A and Runx2 were used as positive control for osteoblast- lineage and -differentiation, RANK and CathK were used as positive control for osteoclast-differentiation and GAPDH was used as sample loading control. (D) Alignment of the amplified PCR products (389 bp product = PS1 and 245 bp product = PS3) indicates matched regions (green) between the transcript from KO cells, WT cells and the P2ry2 receptor gene sequence. (E) Intracellular release of [Ca²⁺]_i expressed as area under the curve (AUC) with increasing doses (10^-8 - 10^-6 M) of UTP (a potent agonist at the P2Y₂ receptor) and MRS-2768 (a selective P2Y₂ receptor agonist with no affinity for human P2Y₄ or P2Y₆ receptors) shows a functional response in the osteoblast precursors and osteoblasts from WT and KO mice. Statistical significance was tested with one-way ANOVA from no agonist control using (a, P < 0.05 in WT and b, P < 0.05 in KO. Data shows mean ± SEM of n = 3 repeat experiments with 3 replicates per dose of each agonist.