Short-term pharmacologic RAGE inhibition differentially affects bone and skeletal muscle in middle-aged mice

Abstract

Loss of bone and muscle mass are two major clinical complications among the growing list of chronic diseases that primarily affect elderly individuals. Persistent low-grade inflammation, one of the major drivers of aging, is also associated with both bone and muscle dysfunction in aging. Particularly, chronic activation of the receptor for advanced glycation end products (RAGE) and elevated levels of its ligands high mobility group box 1 (HMGB1), AGEs, S100 proteins and Aβ fibrils have been linked to bone and muscle loss in various pathologies. Further, genetic or pharmacologic RAGE inhibition has been shown to preserve both bone and muscle mass. However, whether short-term pharmacologic RAGE inhibition can prevent early bone and muscle loss in aging is unknown. To address this question, we treated young (4-mo) and middle-aged (15-mo) C57BL/6 female mice with vehicle or Azeliragon, a small-molecule RAGE inhibitor initially developed to treat Alzheimer's disease. Azeliragon did not prevent the aging-induced alterations in bone geometry or mechanics, likely due to its differential effects [direct vs. indirect] on bone cell viability/function. On the other hand, Azeliragon attenuated the aging-related body composition changes [fat and lean mass] and reversed the skeletal muscle alterations induced with aging. Interestingly, while Azeliragon induced similar metabolic changes in bone and skeletal muscle, aging differentially altered the expression of genes associated with glucose uptake/metabolism in these two tissues, highlighting a potential explanation for the differential effects of Azeliragon on bone and skeletal muscle in middle-aged mice. Overall, our findings suggest that while short-term pharmacologic RAGE inhibition did not protect against early aging-induced bone alterations, it prevented against the early effects of aging in skeletal muscle.

Keywords: Aging; Inflammation; Metabolism; Osteoporosis; RAGE; Skeletal muscle.

Copyright © 2019 Elsevier Inc. All rights reserved.

Figures

Figure 1.. AZ suppresses osteoclast differentiation in vitro.

A) Illustration of AZ mechanism of action. B) Mature osteoclasts/well generated in vitro from non-adherent BMCs from young and middle-aged WT female mice (n=6). Representative images are shown, scale bar indicates 50μm.

Figure 2.. Short-term AZ treatment has modest effects on osteoclasts in vivo.

A) Illustration of in vivo experimental design. Osteoclasts on B) femoral cortical mid-diaphysis and C) vertebral cancellous TRAP/T.blue stained sections (n=8–9). Representative images are shown (black arrows point at osteoclasts), scale bar indicates 25μm. D) Osteoclast-related gene mRNA expression in whole tibia samples isolated from young and aged vehicle- and Azeliragon-treated mice (n=8–10). Bars represent mean±SD, black line: p<0.05, for the overall age effect and *p<0.05, vs veh-treated mice at the same age by two-way ANOVA, Tukey.

Figure 3.. Short-term AZ treatment attenuates osteoblast differentiation/activity in vivo.

A) Osteoblasts on von Kossa/McNeal stained vertebra sections (n=7–10). Representative images are shown, scale bar indicates 25μm. B) Circulating serum PINP levels in young and aged veh- and AZ-treated mice (n=8–10). C) mRNA expression of osteoblast-related genes in whole tibia samples isolated from young and aged vehicle- and Azeliragon-treated mice(n=8–10). D) Dynamic histomorphometric parameters in unstained vertebra sections (n=8–10). E) Mineral apposition rate (MAR), mineralizing surface (MS)/BS, and bone formation rate (BFR)/BS measured in unstained sections of the femoral mid-diaphysis from young and aged vehicle- and Azeliragon-treated mice (n=8–10). F) Mineralization assay of primary osteoblasts generated in vitro from adherent-BMCs from young and middle-aged WT mice (n=5). Bars represent mean±SD, black line: p<0.05, for the overall age effect and *p<0.05, vs veh-treated mice at the same age by two-way ANOVA, Tukey.

Figure 4.. Systemic RAGE inhibition with AZ treatment increases osteocyte apoptosis and pro-inflammatory cytokine production.

A) Active-caspase3 positive apoptotic osteocytes and number of empty lacunae scored in active caspase3 stained bone sections (n=8–10). Representative images of active caspase3-positive osteocytes (arrow, black), scale bar indicates 10μm. B) Osteocytic gene and C) cytokine mRNA levels in tibia from young and middle-aged veh and AZ-treated mice (n=7–10). D) Illustration of bone organ culture experimental design. mRNA expression of E) apoptosis-associated and F) pro-inflammatory cytokines in marrow-flushed bone cultures (n=3–5). Bars represent mean±SD, black line: p<0.05, for the overall age effect and *p<0.05, vs veh-treated mice at the same age by two-way ANOVA, Tukey.

Figure 5.. Short-term AZ treatment decreases bone mass accrual, but not bone architecture in young or old mice.

A) BMD of young and middle-aged mice following veh- or AZ-treatment (n=8–10). Cancellous microarchitecture evaluated by μCT in the B) L4 vertebra and C) distal femur (n=8–9). Representative images are shown. D) Cortical bone geometry evaluated by μCT in the femoral mid-diaphysis (n=8–9). Representative images are shown. Bars represent mean±SD, black line: p<0.05, for the overall age effect and *p<0.05, vs veh-treated mice at the same age by two-way ANOVA, Tukey.

Figure 6.. AZ treatment reverses the loss of skeletal muscle mass induced with aging.

A) Body weight, B) fat and lean mass indicated as percentage of body weight (n=8–10), C) muscle weights (n=8–9), D) myogenesis and satellite cell marker mRNA expression in tibialis anterior muscle (n=7–9) of young and middle-aged mice. Bars represent mean±SD, black line: p<0.05, for the overall age effect and *p<0.05, vs veh-treated mice at the same age by two-way ANOVA, Tukey.

Figure 7.. AZ treatment partially improves the cellular processes involved in maintaining skeletal muscle homeostasis in aging.

A) low- and high-magnification images of forearm muscle cross-sections, muscle of interest is circled in black, scale bar indicates 50μm, and B) average myofiber cross-sectional area (CSA) (n=6–8). C) Average myotube diameter per field and D) distribution of myotube diameter of serum-treated C2C12 cells (n=10). Representative images are shown, scale bar indicates 50μm. E) Ubiquitin-conjugated and GAPDH protein patterns in skeletal muscle (n=6). F) mRNA levels of atrophy-related genes in tibialis anterior muscle (n=7–9). Bars represent mean±SD, black line: p<0.05, for the overall age effect and *p<0.05, vs veh-treated mice at the same age by two-way ANOVA, Tukey.

Figure 8.. Systemic inhibition of RAGE signaling reverses the aging-induced metabolic alterations in skeletal muscle, but not in bone.

Glucose transporter GLUT1–4, and glycolytic and oxidative enzyme mRNA levels in tibialis anterior muscle (A-C) and in tibia (D-F) (n=7–9). Bars represent mean±SD, black line: p<0.05, for the overall age effect and *p<0.05, vs veh-treated mice at the same age by two-way ANOVA, Tukey.