Inhibition of AhR improves cortical bone and skeletal muscle function via preservation of neuromuscular junctions

Abstract

The aryl hydrocarbon receptor (AhR) is proposed to mediate the frailty-promoting effects of the tryptophan metabolite kynurenine, which increases with age in mice and humans. The goal of the current study was to test whether administration of pharmacological AhR inhibitors, BAY2416964 and CH-223191, could abrogate musculoskeletal decline in aging mice. Female C57BL/6 mice (18 months old) were treated with vehicle (VEH) or 30 mg/kg BAY2416964 (BAY) via daily oral gavage 5 days/week for 8 weeks. A second AhR antagonist, CH-223191, was administered to 16-month-old male and female C57BL/6 mice via intraperitoneal injections (3.3 mg/kg) 3 days/week for 12 weeks. While grip strength declined over time in VEH-treated mice, BAY preserved grip strength in part by improving integrity of neuromuscular junctions (NMJs), an effect replicated during in vitro studies with siRNA against AhR. Cortical bone mass was also greater in BAY- than VEH-treated mice. Similarly, CH-223191 treatment improved cortical bone and showed beneficial effects in skeletal muscle, including reducing oxidative stress as compared with VEH-treated animals. Transcriptomic and proteomic data from BAY-treated mice supported a positive impact of BAY on molecular targets that affect NMJ function. Taken together, these data support AhR as a therapeutic target for improving musculoskeletal health during aging.

Keywords: Aging; Bone biology; Osteoclast/osteoblast biology; Osteoporosis; Skeletal muscle.

Figure 1. BAY2416964 is an effective AhR antagonist in musculoskeletal cells and tissues.

(A) AhR transcriptional activity, as shown by a luciferase-based reporter, is activated by kynurenine (KYN; 10 μM) in mesenchymal-lineage ST2 cells, but this impact is blunted by cotreatment with BAY2416964 (BAY); the BAY used in these studies was from the same batch of drug used later for in vivo studies. Group means ± SEM are shown; 1-way ANOVA with Fisher’s LSD post hoc analysis P values are shown above the graph, and groups with different superscript letters are significantly (P < 0.05) different from one another as shown by post hoc testing (B and C) RNAscope staining of AhR and Cyp1a1 mRNA expression show effective suppression of these genes in (B) TA and (C) tibias of BAY- as compared with VEH-treated mice. A magnified image, as demarcated in the yellow box in the columns of images labeled “Merge,” is provided in the columns of images labeled “Merge (zoom)” with scale bars indicative of image size.

Figure 2. BAY treatment did not impact body mass but led to modest decreases in relative weights of fast-twitch muscles.

Body mass was measured weekly (A), and showed no impact of BAY treatment over the course of the study (B). Fat mass (C) and lean mass (D) were measured prior to the onset of treatment and immediately prior to sacrifice and showed no impact of BAY treatment. Relative weights of muscles collected at sacrifice showed no effect on TA (E) or soleus (F) muscles, but BAY treatment reduced the relative weights of extensor digitorum longus (EDL) (G) and tended to reduce the weight of quadriceps (H) muscles as compared with VEH treatment. P values from Student’s t tests comparing groups are shown above each graph.

Figure 3. BAY treatment prevented a decline in forelimb grip strength over time.

Grip strength (A–C) was measured prior to the onset of treatment and immediately prior to sacrifice. Longitudinal changes in grip strength for each mouse (A), as well as cross-sectional measurements of grip strength (B) and grip strength normalized to body mass (C) at 8 weeks of treatment were all significantly improved by BAY treatment. In contrast, neither longitudinal changes in hang time for each mouse (D), nor cross-sectional measurements of hang time (E) or hang time normalized to body mass (F) at 8 weeks of treatment were affected by BAY treatment.

Figure 4. BAY treatment improved the integrity of the neuromuscular junction in skeletal muscle.

Neither average muscle fiber cross-sectional area (A) nor minimum Feret diameter (B) in the TA muscle was affected by BAY treatment. Representative images of histological analyses of the neuromuscular junction (NMJ) via α-bungarotoxin staining (C) show superior integrity of the NMJ in BAY-treated mice, original magnification, ×40. The average area of acetylcholine receptors (D) and fragmentation index (E) of BAY-treated mice were superior compared with VEH-treated mice and with 18-month-old control animals. P values from Student’s t test (A and B) or 1-way ANOVA with Fisher’s LSD post hoc analysis (D and E) comparing groups are shown above each graph. Groups with different lower-case letters are significantly (P < 0.05) different from one another as shown by post hoc testing.

Figure 5. Cortical bone mass was mildly improved by BAY treatment.

DXA scans were collected at baseline (18 months of age) prior to the onset of treatment and immediately prior to sacrifice (20 months of age). Longitudinal analyses of whole-body bone mineral density (BMD) showed no impact of BAY treatment (A). However, microCT analyses of the femur mid-diaphysis showed selective improvement in metrics of cortical bone mass in BAY-treated mice. (B) Cortical bone volume fraction, (C) cortical bone volume, (D) cortical bone tissue volume, (E) cortical thickness, and (F) polar moment of inertia. P values from Student’s t test comparing groups are shown above each graph.

Figure 6. BMSCs from BAY-treated mice showed enhanced colony formation and produce more matrix in vitro.

Crystal violet staining of BMSC-derived colonies on (A) day 7 and (B) day 14 in culture show increased colony formation by cells isolated from BAY-treated mice as shown by quantification of crystal violet–stained area normalized to total well area. (C) Alizarin red staining (normalized to total well area) was also enhanced in BMSC-derived osteoblast cultures isolated from BAY- as compared with VEH-treated mice. Each well shown represents 1 technical replicate culture. (D) Gene expression analyses from BMSC-derived osteoblast cultures on day 21 in culture. P values from Student’s t tests comparing groups are shown above each graph.

Figure 7. CH-223191 treatment is beneficial for cortical bone and skeletal muscle.

(A–E) Cortical bone mass, as measured by microCT; cortical bone volume fraction (A), cortical bone tissue volume (C), and cortical bone thickness (D) were not affected by CH-223191 treatment, while cortical bone volume (B) was significantly greater in CH-223191– as compared with VEH-treated mice, and similar trends are shown for polar moment of inertia (E). Trabecular bone (F–I) was largely unaffected by CH-223191 except for a demonstration of increased trabecular number in male but not female CH-223191–treated mice (I). While quadriceps mass was not affected by CH-223191 (J), average muscle fiber size in the TA muscle was increased in male CH-223191–treated mice (K), and metrics of oxidative stress (from Amplex Red assays) were reduced by CH-223191 in skeletal muscle from both sexes (L). P values from 2-way ANOVA with Fisher’s LSD post hoc analysis comparing groups are shown on each graph; group with different superscript letters are significantly (P < 0.05) different from one another as shown by post hoc testing.

Figure 8. Transcriptomic and proteomic data supported the histological data and suggested that BAY preserves skeletal muscle function in part by maintaining molecular mediators of the NMJ in aged mice.

Heatmaps (A and E), volcano plots (B and F), Gene Ontology plots of protein coding genes and biological processes (C and G), and GSEA plots (D and H) and are shown. Genes found in the cholinergic synapse pathway (I) were expressed at higher levels in the BAY-treated as compared with VEH-treated mice. (J) Proteomics analyses of skeletal muscle lysates identified differentially expressed proteins between BAY- and VEH-treated mice.

Figure 9. AhR knockdown by siRNA is beneficial to myofibers.

(A and B) C2C12 cells treated with siRNA against AhR (AhR siRNA) demonstrated an increased number of acetylcholine receptor (AChR) clusters as compared with cultures treated with a nontargeting control siRNA. Representative images of α-bungarotoxin staining are shown in A. Scale bars: 50 μm. (C) C2C12 cells treated with siRNA against AhR and subjected to myogenic culture demonstrated an increase in myotube area as compared with control siRNA–treated cells. Group means ± SEM are shown; P values for Student’s t tests comparing groups are shown above each graph.