Cancellous bone formation response to simulated resistance training during disuse is blunted by concurrent alendronate treatment


  • Joshua M Swift,

    1. Departments of Health and Kinesiology, Texas A&M University, College Station, TX, USA
    Current affiliation:
    1. Radiation Combined Injury Program, Armed Forces Radiobiology Research Institute, Uniformed Services University of the Health Sciences, Bethesda, MD 20889-5603, USA.
    Search for more papers by this author
  • Sibyl N Swift,

    1. Departments of Intercollegiate Faculty of Nutrition, Texas A&M University, College Station, TX, USA
    Search for more papers by this author
  • Mats I Nilsson,

    1. Departments of Health and Kinesiology, Texas A&M University, College Station, TX, USA
    Search for more papers by this author
  • Harry A Hogan,

    1. Departments of Mechanical Engineering, Texas A&M University, College Station, TX, USA
    Search for more papers by this author
  • Scott D Bouse,

    1. Departments of Mechanical Engineering, Texas A&M University, College Station, TX, USA
    Search for more papers by this author
  • Susan A Bloomfield

    Corresponding author
    1. Departments of Health and Kinesiology, Texas A&M University, College Station, TX, USA
    2. Departments of Intercollegiate Faculty of Nutrition, Texas A&M University, College Station, TX, USA
    • Department of Health and Kinesiology, MS 4243, Texas A&M University, College Station, TX 77843-4243, USA.
    Search for more papers by this author


The purpose of this study was to assess the effectiveness of simulated resistance training (SRT) exercise combined with alendronate (ALEN) in mitigating or preventing disuse-associated losses in cancellous bone microarchitecture and formation. Sixty male Sprague-Dawley rats (6 months old) were randomly assigned to either cage control (CC), hind limb unloading (HU), HU plus either ALEN (HU + ALEN), SRT (HU + SRT), or a combination of ALEN and SRT (HU + SRT/ALEN) for 28 days. HU + SRT and HU + SRT/ALEN rats were anesthetized and subjected to muscle contractions once every 3 days during HU (four sets of five repetitions, 1000 ms isometric + 1000 ms eccentric). Additionally, HU + ALEN and HU + SRT/ALEN rats received 10 µg/kg of body weight of ALEN three times per week. HU reduced cancellous bone-formation rate (BFR) by 80%, with no effect of ALEN treatment (−85% versus CC). SRT during HU significantly increased cancellous BFR by 123% versus CC, whereas HU + SRT/ALEN inhibited the anabolic effect of SRT (−70% versus HU + SRT). SRT increased bone volume and trabecular thickness by 19% and 9%, respectively, compared with CC. Additionally, osteoid surface (OS/BS) was significantly greater in HU + SRT rats versus CC (+32%). Adding ALEN to SRT during HU reduced Oc.S/BS (−75%), Ob.S/BS (−72%), OS/BS (−61%), and serum TRACP5b (−36%) versus CC. SRT and ALEN each independently suppressed a nearly twofold increase in adipocyte number evidenced with HU and inhibited increases in osteocyte apoptosis. These results demonstrate the anabolic effect of a low volume of high-intensity muscle contractions during disuse and suggest that both bone resorption and bone formation are suppressed when SRT is combined with bisphosphonate treatment. © 2011 American Society for Bone and Mineral Research


Significant bone loss remains a persistent problem for humans exposed to microgravity, with little evidence of consistent recovery on return to Earth. Recently, Lang and colleagues1 have demonstrated significant reductions in bone mineral density (BMD) and geometry in astronauts aboard the International Space Station (ISS; 4- to 6-month missions), which losses result in increased estimated fracture risk up to 1 year after returning to Earth.2, 3 Additionally, lack of recovery of BMD in ISS and MIR crew members has been documented 6 months after flight,4 with indications that it may not be fully restored for 3 years.5 The ability of current resistance exercise countermeasures installed on ISS [interim resistance exercise device (iRED) and advanced resistance exercise device (aRED)] to mitigate reductions in lower leg bone mass and strength has yet to be documented.6 Furthermore, if a crew member fracture did occur in microgravity, it would be debilitating and could compromise mission objectives, particularly if crews were to be working on the lunar or martian surface as currently planned in the National Aeronautics and Space Administration's (NASA) “Vision for Space Exploration.”7

Similar deleterious effects to bone are demonstrated during prolonged bed rest, when mechanical loading of weight-bearing bones ceases. Long-duration bed rest reduces femoral neck, lumbar spine, and lower body BMD, resulting in decreased bone volume fraction (BV/TV) and trabecular thickness (Tb.Th).8, 9 Additionally, spinal cord injury (SCI) patients incur even more severe reductions in bone mass, predominantly in cancellous bone compartments.10–12

The rodent hind limb unloading (HU) model is a well-established ground-based model for investigating disuse effects on bone and muscle.13 Hind limb unloading results in significant reductions in disuse-sensitive cancellous bone mass, architecture, and material properties owing to early increases in bone resorption followed by prolonged depressions in BFR.14–17

Previously, we demonstrated the significant positive effects of high-intensity muscle contractions as produced during simulated resistance training (SRT) in a rodent HU model. SRT, completed every other day during a period of disuse, results in absolute increases in disuse-sensitive cancellous bone mass and material properties while maintaining muscle strength.17 Furthermore, significant gains in mid-diaphyseal tibial cortical BMD were associated with a fivefold greater periosteal BFR compared with control animals. However, the cellular mechanisms by which our SRT protocol inhibits unloading-induced reductions in cancellous BMD have not yet to be identified.

The inability of current exercise protocols to prevent microgravity-induced losses in skeletal tissue may be inconsequential if the use of pharmacologic agents in flight proves to be a more effective and less time-intensive countermeasure. Alendronate (ALEN), an antiresorptive bisphosphonate, is currently being tested by NASA with in-flight experiments aboard the ISS.18 ALEN is a nitrogen-containing bisphosphonate that inhibits bone resorption by adsorbing to bone mineral; it interferes with osteoclast activity by inhibiting enzymes of the mevalonate pathway and ultimately contributes to osteoclast apoptosis.19 Additionally, ALEN is currently in use for treatment of various disorders characterized by increased osteoclast-mediated bone resorption and is a proven agent in minimizing bone loss owing to estrogen deficiency and disuse in rats.20–23 Combining antiresorptive therapy with the anabolic effects of resistance exercise may prove to be an effective approach for mitigating bone loss during bed rest (ie, SCI, prolonged rehabilitation after severe injury, etc.) and spaceflight.

The aim of this investigation was to test the hypothesis that combining the anabolic effects of SRT with the antiresorptive effects of ALEN during 28 days of HU in adult rats would more effectively mitigate the usual loss of cancellous bone. Furthermore, we sought to define the effects of our SRT protocol and ALEN on the prevalence of osteocyte apoptosis after unloading because this is one important mechanism for disuse-induced bone loss. We hypothesized that administering ALEN in rats also subjected to SRT during HU will better prevent or ameliorate deleterious changes in cancellous bone than will ALEN or SRT administration alone.

Materials and Methods

Animals and experimental design

Sixty male Sprague-Dawley rats were obtained from Harlan (Houston, TX, USA) at 6 months of age and allowed to acclimate for 14 days prior to initiation of the study. All animals were singly housed after arriving at our animal facility in a temperature-controlled (23 ± 2 °C) room with a 12-hour light/dark cycle in an American Association for Accreditation of Laboratory Animal Care–accredited animal care facility and were provided standard rodent chow (Teklad 8604, Harlan) and water ad libitum. Animal care and all experimental procedures described in this investigation were conducted in accordance with the Texas A&M University Laboratory Animal Care Committee rules.

Five experimental groups were studied: (1) cage control (CC, n = 12), (2) hind limb unloaded (HU, n = 12), (3) HU animals administered 0.01 mg/kg ALEN via subcutaneous injection three times per week (HU + ALEN, n = 12), (4) HU rats subjected to simulated resistance training (HU + SRT, n = 12), and (5) HU rats subjected to both ALEN and SRT (HU + SRT/ALEN, n = 12). HU + SRT and HU + ALEN/SRT animals underwent nine sessions of simulated resistive exercise conducted once every 3 days during the 28-day protocol. The HU group was similarly unloaded for 28 days and exposed to the same duration (25 minutes) of isoflurane anesthesia (Minrad, Inc., Bethlehem, PA, USA) as trained HU rats, whereas the CC animals were allowed normal ambulatory cage activity.

Calcein injections (25 mg/kg of body mass) were given subcutaneously 9 and 2 days prior to killing to label mineralizing bone for histomorphometric analysis. HU animals were anesthetized before removal from tail suspension at the end of the study to prevent any weight bearing by the hind limbs. At the end of the experiment, on day 28, all animals were anesthetized with 50 mg/kg of body weight of ketamine (Fort Dodge Animal Health, Fort Dodge, IA, USA) and 0.5 mg/kg of body weight of medetomidine (Pfizer, New York, NY, USA), and 5 to 6 mL of blood was collected by cardiac puncture and allowed to clot for 20 minutes before centrifugation. Serum was collected and stored at −80 °C until analysis. At necropsy, left soleus, plantaris, and gastrocnemius muscles were excised, and wet weights were recorded. Additionally, proximal left tibias were removed, cleaned of soft tissue, and stored in 70% ethanol at 4 °C for histomorphometric analysis of the proximal tibial metaphysis (PTM). Distal left femurs were fixed in formalin, decalcified, and stored at 4 °C for paraffin embedding.

Hind limb unloading

Hind limb unloading was achieved by tail suspension, as described previously.17 The height of the animal's hindquarters was adjusted to prevent any contact of the hind limbs with the cage floor, resulting in approximately a 30-degree head-down tilt. The forelimbs of the animal maintained contact with the cage bottom, allowing the rat full access to the entire cage. All animals were monitored daily for health, including assessment of tail integrity, and body weights were measured weekly.

Simulated resistance training (SRT) paradigm

SRT was completed as described previously.17 Briefly, left plantarflexor muscles from animals in the HU + SRT group were trained once every 3 days during 28-day HU using a custom-made rodent isokinetic dynamometer. Animals were anesthetized with isoflurane gas (∼2.5%) mixed with oxygen before removal from tail suspension to prevent any weight bearing by the hind limbs. Each rat then was placed in right lateral recumbency on a platform, the left foot was secured onto the foot pedal, and the left knee was clamped so that the lower leg was perpendicular to the foot and the femur and tibia were at right angles to each other. This was referred to as the resting, 0-degree position. For isometric contractions, the foot pedal was held fixed in this position. For eccentric contractions, the footplate was rotated in synchrony with muscle stimulation by a Cambridge Technology lever system (Model 6900) interfaced with an 80486 66-MHz PC using custom software written in TestPoint (Version 4.0, Capital Equipment Corp., Billerica, MA, USA). Torque generated around the footplate pivot (at the rat's ankle joint) was measured by the lever system's servomotor. Plantarflexor muscle stimulation was performed with fine-wire electrodes consisting of insulated chromium-nickel wire (Stablohm 800B, H-ML Size 003, California Fine Wire Co., Grover Beach, CA, USA) inserted intramuscularly straddling the sciatic nerve in the proximal thigh region. The stimulation wires then were attached to the output poles of a Grass Instruments stimulus isolation unit (Model SIU5, Astro-Med, Inc., West Warwick, RI, USA) interfaced with a stimulator (S88, Astro-Med) that delivered current to the sciatic nerve and induced muscle contraction.

Voltage optimization to achieve peak isometric torque and stimulation frequency optimization of the eccentric torque were performed at the beginning of each session, as described previously.17 The eccentric phase of the muscle contraction was titrated to equal approximately 75% of each animal's daily peak isometric torque. The HU + SRT and HU + SRT/ALEN animals completed a combined isometric plus eccentric SRT exercise paradigm consisting of four sets of five repetitions once every 3 days during HU (n = 9 total exercise sessions). Each stimulation in the training paradigm consisted of a 1000-ms isometric contraction, immediately followed by a 1000-ms eccentric contraction.

Bisphosphonate treatment

Animals in the HU + ALEN, and HU + SRT/ALEN groups were administered 10 µg/kg of ALEN (Merck and Co., Rathway, NJ, USA) via subcutaneous injection three times per week for the duration of the 28-day study. The ALEN dose of 30 µg/kg per week was the lowest dose found to effectively mitigate reductions in cancellous volumetric bone mineral density (vBMD) at the proximal tibia during 28 days of HU (unpublished data) and is similar to the 30 µg/kg of ALEN (15 µg/kg twice weekly) shown to maintain femur and lumbar spine bone mass and strength after ovariectomy (OVX) in rats.24 Furthermore, the ALEN dose of 30 µg/kg per week that we chose to use in this study is lower than the dose (100 µg/kg per day) used in previously published clinical studies in OVX rats demonstrating pronounced increases in bone mass and strength.25 Rats in the CC, HU, and HU + SRT groups were administered an equal volume of vehicle (phosphate-buffered saline).

Peripheral quantitative computed tomography (pQCT)

On days −1 and 28 of the study, pQCT scans were performed in vivo at the proximal tibial metaphysis with a Stratec XCT Research-M device (Norland Corp., Fort Atkinson, WI, USA) using a voxel size of 100 µm and a scanning beam thickness of 500 µm. Daily calibration of this machine was performed with a hydroxyapatite standard cone phantom. Transverse images of the left tibia were taken at 5.0, 5.5, and 6.0 mm from the proximal tibia plateau. A standardized analysis for metaphyseal bone (ie, contour mode 3, peel mode 2, outer threshold of 0.214 g/cm3, and inner threshold of 0.605 g/cm3) was applied to each section.

Values of total vBMD (includes cortical shell and cancellous bone), total bone mineral content (BMC), total bone area, and cancellous vBMD were averaged across the three slices at the proximal tibia to yield a mean value. Machine precision (based on the manufacturer's data) is ±3 mg/cm3 for cancellous vBMD. Coefficients of variation were ±0.6%, ±1.6%, ±1.9%, and ±2.13% for in vivo proximal tibia total vBMD, total BMC, total area, and cancellous vBMD, respectively, as determined from three repeat scans on each of six adult male rats.

Histomorphometric analysis

Undemineralized proximal left tibias were subjected to serial dehydration and embedded in methyl methacrylate (Aldrich M5, 590-9, St. Louis, MO, USA). Serial frontal sections were cut 8 µm thick and left unstained for fluorochrome label measurements. Additionally, 4-µm-thick sections treated with von Kossa stain, and tetrachrome counterstain was used for measurement of cancellous bone volume normalized to tissue volume (%BV/TV) and osteoid (Os/BS), osteoblast (ObS/BS), and osteoclast (OcS/BS) surfaces as a percent of total cancellous surface. Adipocyte density was calculated as number of adipocytes (Ad.N) divided by the marrow area (Ma.Ar). The histomorphometric analyses were performed by using the OsteoMeasure Analysis System, Version 1.3 (OsteoMetrics, Inc., Atlanta, GA, USA). A defined region of interest was established approximately 1 mm from the growth plate and within the endocortical edges encompassing 8 to 9 mm2 at ×40 magnification. Total bone surface (BS), single-labeled surface (SLS), double-labeled surface (DLS), interlabel distances, bone volume, and osteoid/osteoclast/osteoblast surfaces were measured at ×200 magnification. Mineral apposition rate (MAR, µm/d) was calculated by dividing the average interlabel width by the time between labels (7 days), and mineralizing surface (MS) for cancellous bone surfaces (BS) was calculated by using the formula %MS/BS = {[(SLS/2) + DLS]/surface perimeter}c × 100. BFR was calculated as (MAR × MS/BS). All nomenclature for cancellous histomorphometry follows standard usage.26

Osteocyte apoptosis

Distal left femurs were fixed in 4% phosphate-buffered formalin for 48 hours at 4 °C and then decalcified in 10% EDTA and 4% phosphate-buffered formalin for 14 days. Following decalcification, the distal left femurs were embedded in paraffin, and serial frontal sections were cut 10 µm thick and mounted on slides. Apoptosis of osteocytes was detected by in situ terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) using the DNA fragmentation TdT enzyme and fluorescein-dUTP label (Roche Diagnostics Corp., Indianapolis, IN, USA) in distal femoral sections counterstained with hematoxylin QS (Vector Laboratories). A defined region of interest (ROI) was established approximately 1 mm from the growth plate and within the endocortical edges encompassing 8 to 9 mm2 at ×40 magnification. Quantification of osteocytes residing in trabeculae within the ROI was performed using the OsteoMeasure Analysis System, Version 1.3 (OsteoMetrics). The total number of osteocytes (N.Ot) within the region was first counted (under normal light), followed by identification of TUNEL+ osteocytes using ultraviolet light at ×200 magnification. The percentage of apoptotic osteocytes was calculated as (TUNEL+ Ot/total N.Ot) × 100.

Serum tartrate-resistant acid phosphatase (TRACP5b)

Serum TRACP5b was determined using the RatTRAP Assay (IDS, Fountain Hills, AZ, USA) according to the manufacturer's instructions. Assay results were analyzed using a DTX 880 microplate reader (Beckman Coulter, Brea, CA, USA). The interassay coefficient of variation was found to be 2%.

Statistical analysis

All data were expressed as means ± SEM and evaluated using the statistical package SPSS (Version 15, SPSS, Inc., Chicago, IL, USA). In vivo pQCT, body mass delta scores, muscle mass at euthanization, histomorphometry, and apoptosis assays were first analyzed using a two-factor ANOVA (exercise and ALEN) to compare group differences between HU groups (HU, HU + ALEN, HU + SRT, and HU + SRT/ALEN). A Tukey's post hoc test was used for pairwise comparisons. Subsequently, a one-factor ANOVA was used to compare individual group change scores versus that of the comparator cage control (CC) group (Tukey's post hoc test for pairwise comparisons) and paired t tests to determine (using absolute date values pre/post within group) whether that change score represented a significant change from day 0 (pQCT and body mass data only). For all data, statistical significance was accepted at p < .05.


SRT does not prevent disuse-associated reductions in total-body mass or ankle plantarflexor muscle mass

Hind limb unloading significantly reduced total-body mass (−8% versus day 0), which loss was not affected by ALEN (−5.5%), SRT (−7.5%), or the combination of both ALEN and SRT (−11%) during 28-day HU (Table 1). Unloading resulted in lower gastrocnemius, plantaris, and soleus muscle masses compared with ambulatory controls (24% to 56%; Table 1). These reductions in ankle plantarflexor muscle masses were not affected by SRT alone, although ALEN and SRT/ALEN mitigated the decrease in soleus mass. Additionally, when normalized to body mass, SRT did not prevent reduced total plantarflexor muscle mass of the trained left leg.

Table 1. Effects of Hind Limb Unloading (HU) With or Without Alendronate (ALEN) Treatment and/or Simulated Resistance Training (SRT) on Body Mass and Left Ankle Plantarflexor Muscle Mass
  • Note: Values are group mean ± SEM. The HU groups not sharing the same letter for each variable are significantly different from each other (p < .05). Group means with no labels are not significantly different.

  • Significantly different versus CC (p < .05).

  • *

    p < .05 versus pre value.

Body Mass (g)
 Day 0454.58 ± 6.56447.73 ± 15.89438.31 ± 6.39442.42 ± 7.78436.36 ± 6.09
 Day 28478.75 ± 8.50*412.00 ± 13.35*414.15 ± 9.58*409.08 ± 8.28*389.36 ± 8.04*
 Body Mass Change (g)24.17 ± 3.08*(−35.73 ± 5.25ab)(−24.15 ± 5.81a)(−33.33 ± 4.65a)(−47.00 ± 5.24b)
Ankle Plantarflexor Masses (g)
 Gastrocnemius2.312 ± 0.1351.732 ± 0.0351.845 ± 0.0571.607 ± 0.0581.663 ± 0.044
 Plantaris0.513 ± 0.0090.392 ± 0.0100.401 ± 0.0110.374 ± 0.0150.359 ± 0.010
 Soleus0.199 ± 0.0060.088 ± 0.004a0.107 ± 0.007b0.096 ± 0.004ab0.101 ± 0.004b
 Total Mass3.025 ± 0.1282.212 ± 0.0392.353 ± 0.0682.072 ± 0.0642.100 ± 0.051
 Relative Total Mass/BW (mg/g)6.347 ± 0.2975.41 ± 0.165a5.682 ± 0.290a4.662 ± 0.421b4.922 ± 0.465ab

High-intensity muscle contractions performed during unloading inhibit reductions in cancellous bone mass

Hind limb unloading significantly reduced proximal tibia metaphysis (PTM) total vBMD (−5%), total BMC (−8%), and total bone area (−3%). ALEN prevented HU-associated losses in total vBMD but not the reductions in total BMC and bone area (Fig. 1A–C). SRT and SRT/ALEN prevented HU-associated reductions in these parameters, resulting in significant increases versus day 0 in PTM total vBMD (+7%) and total BMC (+8% to 10%). Cancellous vBMD, reduced 8% by HU alone, was maintained with ALEN treatment (Fig. 1D). Simulated resistance training (SRT and SRT/ALEN groups) not only inhibited disuse-associated reductions in cancellous vBMD but also resulted in absolute gains (+8%).

Figure 1.

Effects of hind limb unloading (HU) with or without alendronate (ALEN) treatment and/or simulated resistance training (SRT) on changes in structural and geometric properties of the proximal tibia metaphysis as taken by in vivo peripheral quantitative computed tomographic scans. (A) Total volumetric bone mineral density (vBMD). (B) Total bone mineral content (BMC). (C) Total bone area. (D) Cancellous vBMD. Vertical dashed line indicates separation of CC from the experimental groups for preliminary ANOVA. Those HU groups not sharing the same letter for each variable are significantly different from each other (p < .05). Significantly different versus CC (p < .05). *p < .05 versus pre value.

ALEN reduces the cancellous bone-formation response to SRT during disuse

Hind limb unloading produced significantly lower PTM cancellous bone mineralizing surface (−60%) and MAR (−50%), resulting in 81% lower bone formation than in ambulatory controls (Fig. 2A–C). ALEN administration had no effect on HU-induced reductions in cancellous bone formation. SRT, undertaken during unloading, not only inhibited deficits in cancellous bone formation but also led to significantly greater %MS/BS (+90%), MAR (+25%), and BFR (twofold increase) versus CC. When SRT was completed in combination with ALEN, %MS/BS and BFR were significantly lower (−63% to −70%) than in the HU + SRT group.

Figure 2.

Effects of hind limb unloading (HU) with or without alendronate (ALEN) treatment and/or simulated resistance training (SRT) on cancellous bone dynamic histomorphometric analysis measured at the proximal tibia metaphysis. (A) Mineralizing surface (%MS/BS). (B) Mineral apposition rate (MAR). (C) Bone-formation rate (BFR). Vertical dashed line indicates separation of CC from the experimental groups for preliminary ANOVA. The HU groups not sharing the same letter for each variable are significantly different from each other (p < .05). Significantly different versus CC (p < .05).

SRT improves metaphyseal bone microarchitecture and reduces adipocyte density

Cancellous BV/TV and Tb.Th were lower in both HU and HU + ALEN groups (−9% to −12%), but only the latter group values were significantly different from the CC group (Fig. 3A, B). Greater cancellous bone formation in the HU + SRT and HU + SRT/ALEN groups resulted in enhanced proximal tibia microarchitecture. BV/TV (+15%) and Tb.Th (+32%) were significantly greater in both SRT groups than in the HU group. Furthermore, ALEN + SRT during HU produced smaller Tb.Sp (−17%) and greater Tb.N (+14%) than HU alone (Fig. 3C, D).

Figure 3.

Effects of hind limb unloading (HU) with or without ALEN treatment and/or SRT on cancellous bone microarchitecture. (A) Bone volume (%BV/TV). (B) Trabecular thickness (Tb.Th). (C) Trabecular spacing (Tb.Sp). (D) Trabecular number (Tb.N). Vertical dashed line indicates separation of CC from the experimental groups for preliminary ANOVA. The HU groups not sharing the same letter for each variable are significantly different from each other (p < .05). Significantly different versus CC (p < .05).

Hind limb unloading resulted in significantly lower OS/BS (−40%) and Oc.S/BS (−45%), in addition to a twofold greater adipocyte density than in ambulatory controls (Fig. 4A, C, D). ALEN treatment exacerbated lowered osteoid surface (−80%), demonstrated during HU, and produced a nearly complete suppression of osteoblast surface (−97%; Fig. 4B) compared with the CC group. SRT inhibited disuse-induced reductions in OS/BS and increases in adipocyte density; the addition of ALEN treatment during SRT significantly lessened the beneficial effects of high-intensity muscle contractions. HU + SRT/ALEN rats' OS/BS and Ob.S/BS were 61% to 72% lower than those of controls. On the other hand, Oc.S/BS was reduced to a greater extent in HU + SRT/ALEN rats than in all other HU groups.

Figure 4.

Effects of hind limb unloading (HU) with or without ALEN treatment and/or SRT on cancellous bone cell activity. (A) Osteoid surface (OS/BS). (B) Osteoblast surface (Ob.S/BS). (C) Osteoclast surface (Oc.S/BS). (D) Adipocyte density (N.Ad/Ma.Ar). (E) Representative micrographs (×400 magnification; von Kossa and tetrachrome stain) of adipocytes (top row) and osteoblasts (bottom row) from histologic sections of the proximal tibia from all treatment groups. Note the increased number of adipocytes (black arrows) in the HU sample. Also note the increased number of osteoblasts (black arrows) and greater osteoid surface (below osteoblasts) in the HU + SRT sample. Vertical dashed line indicates separation of CC from the experimental groups for preliminary ANOVA. The HU groups not sharing the same letter for each variable are significantly different from each other (p < .05). Significantly different versus CC (p < .05).

ALEN treatment, not SRT, reduces serum TRACP5b after 28 days of disuse

Hind limb unloading did not result in a significantly greater serum concentrations of TRACP5b, a systemic marker of osteoclast number, compared with ambulatory controls (Fig. 5). Furthermore, SRT did not affect TRACP5b levels. ALEN treatment alone (HU + ALEN group) and in combination with SRT (HU + ALEN/SRT group) resulted in a significant reduction in TRACP5b compared with HU (−40% to −45%) and ambulatory controls (−30% to −36%).

Figure 5.

Effects of hind limb unloading (HU) with or without alendronate (ALEN) treatment and/or SRT on serum TRACP5b measured on day 28. Vertical dashed line indicates separation of CC from the experimental groups for preliminary ANOVA. The HU groups not sharing the same letter for each variable are significantly different from each other (p < .05). Significantly different versus CC (p < .05).

Osteocyte apoptosis is maintained independently by both muscle contractions and ALEN during disuse

Unloading resulted in a significantly greater percentage of apoptotic cancellous osteocytes (+74%) compared with ambulatory controls (CC group; Fig. 6). SRT and ALEN, independently and in combination, retarded HU-associated increases in the prevalence of apoptotic osteocytes within cancellous bone.

Figure 6.

Effects of hind limb unloading (HU) with or without ALEN treatment and/or SRT on cancellous bone TUNEL+ osteocytes (%) measured at the distal femur. Vertical dashed line indicates separation of CC from the experimental groups for preliminary ANOVA. The HU groups not sharing the same letter for each variable are significantly different from each other (p < .05). Significantly different versus CC (p < .05)


The main purpose of our study was to test the hypothesis that by combining SRT and ALEN, the metaphyseal bone response would be greater than with either treatment alone. Furthermore, we hypothesized that both SRT and ALEN treatment would mitigate disuse-associated increases in osteocyte apoptosis.

Contrary to our hypothesis, adding ALEN treatment to SRT during disuse did not have a greater positive impact on metaphyseal bone than either of these interventions alone. SRT in ALEN-treated HU rats resulted in similar changes in proximal tibia bone mass and microarchitecture as did SRT alone. Both the HU + SRT and the HU + SRT/ALEN groups experienced significantly larger increases in proximal tibia cancellous vBMD (Fig. 1D) associated with greater bone volume, trabecular thickness, and trabecular number than untreated HU rats (Fig. 3). Additionally, both exercise-trained groups exhibited reduced prevalence of osteocyte apoptosis in cancellous bone (Fig. 6). However, when ALEN treatment was added to SRT during disuse, we observed a 70% suppression of cancellous bone formation (Fig. 2C) and an inhibition of the dramatic response of osteoid and osteoblast surfaces to SRT (Fig. 4A, B), with values for these static markers of bone formation similar to the HU-only group. ALEN administration in combination with SRT did significantly suppress Oc.S/BS (Fig. 4C) compared with the HU group. Furthermore, treatment with ALEN alone or combined with SRT suppressed serum TRACP5b (Fig. 5), a marker of osteoclast number. Given the suppressive effect of ALEN administration on bone-formation activity stimulated by our training paradigm, as measured at the end of this 28-day trial, this suggests that the absolute gains in metaphyseal bone mass in the HU + SRT/ALEN group occurred earlier in the unloading period. The antiresorptive effects of ALEN, as reflected in diminished osteoclast surface and serum TRACP5b, did not provide an additional benefit to cancellous bone mass compared with that observed with SRT alone.

The SRT protocol used in this study produced similar positive effects on the unloaded tibia as described in our previous results.17 However, lacking detailed histomorphometric data in our earlier study, we were unable to confirm that our SRT protocol had an anabolic effect on cancellous bone. In this investigation, we used a reduced intensity of combined isometric and eccentric muscle contractions (75% versus 100% peak isometric torque) administered fewer training sessions (9 versus 14) and still demonstrated absolute increases in proximal tibia cancellous bone vBMD (Fig. 1D). This increase in cancellous bone mass was associated with significantly greater bone formation (Fig. 2C), osteoid, and osteoblast surface (Fig. 4A, B), along with a suppression of the increase in adipocyte density observed with unloading (Fig. 4D). Furthermore, SRT resulted in greater trabecular thickness than in both unloaded and ambulatory control animals (Fig. 3B). However, the reduced volume of the training protocol used in this study was not able to mitigate the loss of total left ankle plantarflexor muscle mass (Table 1), as was achieved in our previous study, although reductions in soleus mass were attenuated with combined SRT and ALEN treatment. We were unable to detect any effect of ALEN treatment on unweighted skeletal muscle. Taken together, these data emphasize the dramatic anabolic response of cancellous bone to a low volume of high-intensity muscle contractions engaged during periods of disuse. However, reducing the training intensity and volume (versus our previous protocol) resulted in a loss of the mitigating effect on disuse-induced muscle atrophy.

Hind limb unloading increases adipocyte number while simultaneously inhibiting osteoblast differentiation, resulting in increased adipocyte number and volume and reduced bone mass and strength at the proximal tibia.27–29 Interestingly, we demonstrated that SRT, with or without concurrent ALEN treatment, mitigated unloading-associated increases in marrow area adipocyte density (Fig. 4D). ALEN treatment by itself did not affect adipocyte density during unloading. While we did not measure adipocyte differentiation, these data provide preliminary evidence that high-intensity muscle contractions during disuse may inhibit bone marrow stromal cell (BMSC) differentiation into adipocytes, which would favor an increase in osteoblast progenitors available to support the increased drive for more bone formation (Fig. 2). Mechanical loading has been demonstrated previously to suppress marrow adipocyte populations, resulting from increased Runx2 (transcriptional promoter of osteoblastogenesis) and decreased PPARγ (transcriptional promoter of adipogenesis) expression in BMSCs.30, 31 These cellular effects may be mechanically driven by the increased intramedullary pressure and fluid shear stress on bone cells resulting from the dynamic muscle contractions generated during our SRT protocol.32

Although this is the first study, to our knowledge, to investigate the effects of ALEN treatment in conjunction with exercise (engaged without weight bearing) during hind limb unloading, there are a limited number of investigations on this topic testing the efficacy during estrogen deficiency using OVX rodents. When 30 µg/kg per week of ALEN (same weekly dose as in this study) is administered to OVX rats in combination with moderate treadmill running, the effects of the combined therapy are superior in maintaining bone mass and strength than either individual treatment.24 Similar to the aforementioned study, Tamaki and colleagues33 documented significant interactions between treadmill running exercise and bisphosphonate treatment (etidronate), both individually and in combination positively affecting metaphyseal bone. However, neither of these two investigations assessed bone formation or osteoblast activity, and both studied bone loss owing to estrogen deficiency, which generally is less deleterious to cancellous bone than the disuse of unloading. Furthermore, in contrast to our study, both studies employed moderate-intensity treadmill running, which has been shown to be less effective than resistance exercise at increasing bone mass in rodents.34 In humans, high-impact exercise35 and jumping36 elicit a greater anabolic response on bone than repetitive high-frequency activities such as running or walking.37 Furthermore, in postmenopausal women, 12 months of jump exercise training combined with daily ALEN administration (35 mg/week) produced no additive benefit to bone mass or mechanical properties beyond the positive effects of each independent treatment.38

Combined exercise and ALEN treatment has been investigated in few human studies assessing disuse-induced bone loss typical of that observed during long-duration spaceflight. Administration of an early-generation bisphosphonate (ethane-1-hydroxy-1-disphosphonate) combined with treadmill exercise and cycling (modeling exercise regimens used by cosmonauts) during 360 days of bed rest reduced negative calcium balance but did not significantly mitigate losses in femoral neck BMD.39 Neither flywheel resistance exercise nor pamidronate (another nitrogen-containing bisphosphonate) was able to rescue metaphyseal bone loss in the tibia during 90 days of bed rest but was effective in mitigating loss of mid-diaphysis (cortical) bone.40

Numerous investigations using rodent unloading models and bed rest trials in humans have demonstrated the ability of ALEN to inhibit disuse-induced bone loss. ALEN administered prior to HU (100 µg/kg) prevented reductions in bone mass by decreasing relative osteoclast surface and mitigating reductions in bone formation.41 Furthermore, ALEN treatment during 14-day HU abolished losses in tibia and femur BMD but was unable to rescue disuse-induced reductions in bone strength.20 Apseloff and colleagues21 administered 300 µg/kg (two times per day) of ALEN during 28 days of unloading and found significant reductions in osteoid perimeter, cancellous bone formation, and bone resorption. The resulting increase in bone mass with ALEN treatment resulted from an inhibition of resorptive activity. Bed rest investigations on human subjects have elucidated similar mechanisms of this bisphosphonate's action on bone during periods of disuse. Ruml and colleagues42 administered 20 mg/d of ALEN during 3 weeks of bed rest and observed a reduction in urinary calcium excretion compared with untreated controls. Half that dose (10 mg/d) used during a much longer period of bed rest (17 weeks) successfully prevented reductions in lumbar spine and femoral neck BMD and attenuated increases in urinary markers of bone resorption.43

Data from these previous investigations parallel findings in this study. We found that a weekly dosage of 30 µg/kg of ALEN given during disuse prevented losses in both total and cancellous vBMD at the proximal tibia despite lower bone-formation and markers of osteoblast activity. Although we were unable to detect significant reductions in osteoclast surface, we did demonstrate a significant suppression of serum TRACP5b, a systemic marker of osteoclast number. In postmenopausal females, serum TRACP5b correlates significantly with changes in BMD and has been accepted as a useful marker for monitoring ALEN treatment.44 Serum TRACP5b was significantly reduced in both HU + ALEN and HU + SRT/ALEN groups compared with unloaded animals (Fig. 5). These data suggest that the number of active osteoclasts across the entire skeleton may have been reduced, even though osteoclast surface measured in a focal region of interest was not affected.

Reductions in metaphyseal bone mass are associated with increased osteocyte and osteoblast apoptosis. Recent data demonstrate a striking increase in apoptotic osteocytes in unloaded bone as early as day 3 of HU and continuing through 14 days.45, 46 Our data confirm a doubling of the prevalence of apoptotic osteocytes in cancellous bone after 28 days of hind limb unloading.47 Previously published data have demonstrated a normalization of osteocyte apoptosis after 2 weeks of weight-bearing recovery following HU.46 To our knowledge, however, no prior investigation has defined the role of mechanical loading engaged during a period of imposed disuse on osteocyte apoptosis within cancellous bone. Surprisingly, our data demonstrate that high-intensity muscle contractions completely prevent disuse-induced increases in osteocyte apoptosis (Fig. 6). ALEN treatment resulted in a similar protective effect. Although previously published studies have confirmed that ALEN inhibits osteocyte apoptosis,48, 49 our data are the first to demonstrate the antiapoptotic effects of ALEN during disuse. Reducing osteocyte apoptosis may be crucial to preserving cancellous bone mass and reducing bone resorption during disuse.

There were a few limitations to this investigation. We were unable to detect significant changes in cancellous bone microarchitecture by 2D histomorphometry. The very selective region of interest provided by 2D sections from the proximal tibia may not accurately reflect activity across the entire volume of the metaphyseal region. Additionally, our study design did not include weight-bearing control animals (CC) receiving either SRT or ALEN; therefore, we are unable to determine the important clinical question as to whether or not our regimen of ALEN treatment combined with SRT has similar negative effects on cancellous bone formation in weight-bearing rodents.

In conclusion, our data suggest that bisphosphonate treatment may impair the osteogenic effect of resistance training engaged in during a period of unloading should the suppression of osteoblast activity we observed be more than a transient phenomenon. The beneficial effect of mechanical loading and ALEN treatment (acting independently and in combination) may be achieved in part by inhibiting disuse-associated increases in cancellous osteocyte apoptosis. The inhibitory effects of bisphosphonate treatment on the cancellous bone-formation response to high-intensity resistance exercise has important implications for the efficacy of exercise countermeasures used during periods of disuse in any population using these pharmaceutical agents.


The views, opinions, and findings contained herein are those of the author and do not necessarily reflect official policy or positions of the Department of the Navy, Department of Defense, nor the United States Government. All the authors state that they have no conflicts of interest.


These studies were funded through the NASA Cooperative Agreement NCC 9-58 with the National Space Biomedical Research Institute (SAB). JMS was supported by a National Space Biomedical Research Institute Graduate Training Fellowship (NSBRI-RFP-05-02). We gratefully acknowledge Ms Janet Stallone for assistance with animal care and Drs Gordon Warren (Georgia State University, Atlanta, GA, USA) and Ken Baldwin (University of California, Irvine, CA, USA) for assistance with the simulated resistive exercise programming and procedures. We also thank Dr Matthew Allen (Indiana University, Indianapolis, IN, USA) for his review of this manuscript and Brandon Macias for assistance with micrographs. Alendronate was generously provided by Merck Pharmaceuticals and Co. (Whitehouse Station, NJ, USA).

Authors' roles: JMS assisted with the design and implementation of experiments; collected and analyzed CT, histology, and immunostaining data; and wrote the manuscript. SNS performed all alendronate dosing, collected and analyzed histology data, and completed serum TRACP5b analyses. MIN assisted with the design and implementation of experiments and collection of CT data. HAH designed experiments, provided critical support for SRT, and interpreted data. SDB assisted with SRT and CT data collection. SAB designed experiments and interpreted data. All authors assisted in revising the manuscript and all approved the final submitted version. JMS and SAB accept responsibility for the integrity of the data analyses.