Hindlimb unloading has a greater effect on cortical compared with ...

J Appl Physiol 94: 642–650, 2003. First published October 18, 2002; 10.1152/japplphysiol.00656.2002.

Hindlimb unloading has a greater effect on cortical compared with cancellous bone in mature female rats MATTHEW R. ALLEN AND SUSAN A. BLOOMFIELD Department of Health and Kinesiology, Texas A&M University, College Station, Texas 77843-4243 Submitted 17 July 2002; accepted in final form 11 October 2002

Allen, Matthew R., and Susan A. Bloomfield. Hindlimb unloading has a greater effect on cortical compared with cancellous bone in mature female rats. J Appl Physiol 94: 642–650, 2003. First published October 18, 2002; 10.1152/ japplphysiol.00656.2002.—This study was designed to determine the effects of 28 days of hindlimb unloading (HU) on the mature female rat skeleton. In vivo proximal tibia bone mineral density and geometry of HU and cage control (CC) rats were measured with peripheral quantitative computed tomography (pQCT) on days 0 and 28. Postmortem pQCT, histomorphometry, and mechanical testing were performed on tibiae and femora. After 28 days, HU animals had significantly higher daily food consumption (⫹39%) and lower serum estradiol levels (⫺49%, P ⫽ 0.079) compared with CC. Proximal tibia bone mineral content and cortical bone area significantly declined over 28 days in HU animals (⫺4.0 and 4.8%, respectively), whereas total and cancellous bone mineral densities were unchanged. HU animals had lower cortical bone formation rates and mineralizing surface at tibial midshaft, whereas differences in similar properties were not detected in cancellous bone of the distal femur. These results suggest that cortical bone, rather than cancellous bone, is more prominently affected by unloading in skeletally mature retired breeder female rats. bone mineral density; simulated microgravity; peripheral quantitative computed tomography, histomorphometry; mechanical properties

THE EFFECTS OF DISUSE on the weight-bearing bones of the skeleton have been well documented in both humans and rodents (for reviews, see Refs. 3, 21, 29). Humans exposed to long-duration spaceflight (⬎4 mo) exhibit a loss of bone density estimated at 1–2% per month, concentrated mostly in the cancellous compartments of the lower appendicular skeleton (7, 33). Complementary changes to those in humans have been documented in animals exposed to short-duration spaceflight (up to 16 days), including altered gene expression of bone matrix proteins (6, 9, 15), decrements in bone formation (6, 19, 30, 35, 36), and reduced mechanical strength (23, 25, 27, 31). One caveat, however, is that, whereas the majority of human spaceflight studies investigate changes in skeletally mature subjects, most animal spaceflight data describe adaptations in young, rapidly growing rats (ⱕ3 mo old). Therefore, these animal studies are likely measuring

Address for reprint requests and other correspondence: S. A. Bloomfield, Dept. of Health and Kinesiology, Texas A&M Univ., College Station, TX 77843-4243 (E-mail: [email protected]). 642

changes caused by disrupted modeling and growth, rather than disrupted remodeling, as occurs in a more mature skeleton (11). To develop appropriate countermeasures for long-duration spaceflight, the effects of unloading on the mature skeleton must be elucidated. Hindlimb unloading (HU) is well established as a model to simulate spaceflight (20). The few published studies using skeletally mature rats (⬎5 mo of age) demonstrate distinct differences from the response of growing animals (⬃3 mo of age). Unloading must be maintained two to three times longer in mature rats, compared with 3-mo-olds, to produce significant declines in bone calcium content, density, and formation rates (4, 9, 13, 25), all of which are affected more adversely in skeletally mature animals (32). The decrement in cortical bone formation rate (BFR) is twofold greater in 6-mo-old rats compared with 3-mo-old rats of similar sex and strain (4, 8). Additionally, this reduction in bone formation persists over the course of 5 wk of unloading in mature rats, whereas younger rat BFRs return toward baseline values after 14 days of HU (13). These distinct differences based on the level of skeletal maturity suggest the importance of careful interpretation of the animal data on skeletal adaptations to unloading with regard to potential mechanisms and countermeasures. In addition to the paucity of skeletal studies on mature animals, the female skeleton’s response to unloading has been rarely examined. Studies on cosmonauts/astronauts include so few women subjects that conclusions specific to females are not feasible (33), and, given that women experience a higher rate of age-related bone loss compared with men, there is a clear need to define the effects of spaceflight on the female skeleton (14). Of the few published rodent studies investigating unloading effects on the female skeleton (2, 6, 24), ovariectomized (OVX) rats are often used, making it difficult to define the effects of reduced loading independent of those due to estrogen deficiency because of the sudden and complete removal of circulating estrogen. Because both human and rodent males exposed to unloading show decreases in sex steroid levels (1), it is likely that reductions in estrogen will occur in intact females. These declines, however, The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

8750-7587/03 $5.00 Copyright © 2003 the American Physiological Society

http://www.jap.org

UNLOADING-INDUCED SKELETAL CHANGES IN FEMALE RATS

should be less severe than OVX, causing intermediate effects on bone given that estrogen effects on the skeleton are highly dependent on circulating levels (18). For this reason, determining the effects of unloading on intact female rat estrogen levels and bone adaptations is vital. Peripheral quantitative computed tomography (pQCT) is increasingly utilized in small animal research because of its ability to provide high-resolution in vivo measures of volumetric bone density (mg/cm3). It also has the ability to differentiate between cortical and cancellous bone compartments, an important consideration because cancellous bone appears to be more affected by unloading (33) and may change independently of changes in cortical bone (4). Thus studies that measure whole bone properties may potentially miss unloading-induced changes because of the masking of the cancellous bone loss by changes in the cortical shell. Therefore, the purpose of this study was to examine the response of skeletally mature female rats to HU using in vivo and ex vivo pQCT, cortical/cancellous histomorphometry, measures of serum estradiol, and mechanical testing to assess changes after 28 days of HU or cage activity. We hypothesized that 28 days of HU would 1) cause declines in circulating serum estradiol levels and 2) cause reductions in bone density and formation rates, particularly in cancellous bone. METHODS

Animals. The study protocol and all animal procedures were in compliance with the Texas A&M University Laboratory Animal Care Committee rules and regulations and conformed to the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals [DHEW Publication No. (NIH) 85-23, Revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD 20892]. Female Sprague-Dawley retired breeder rats (6-mo-old; 283 ⫾ 3 g) from Harlan (Indianapolis, IN) were housed individually in a temperature-controlled room (21 ⫾ 2°C) with a 12:12-h light-dark cycle. Animals were provided standard rat chow and water ad libitum throughout the experiment. Food intake was recorded daily starting 1 wk before group assignments. Two weeks after arrival, rats were randomly assigned to either cage control (CC, n ⫽ 8) or HU (n ⫽ 8) groups. A priori calculations, based on data from previous studies (4, 8), indicated that eight animals per group would provide adequate statistical power to detect differences in bone density and bone formation. Unloading of the hindlimbs was achieved by tail suspension as previously described (4, 20). Briefly, the tail was cleaned and dried thoroughly. A thin layer of adhesive (Outdoor Goop; Eclectic Products) was applied to the middle 50% of the tail along the medial and lateral sides. A harness made of standard porous tape (Kendall; Mansfield, MA) was pressed firmly to the glue and allowed to dry. A paper clip was used to attach the animal’s tail harness to a swivel apparatus on the wire spanning the top of an 18-in. ⫻ 18-in. ⫻ 18-in. cage. The height of the animal’s hindlimbs was adjusted to prevent any contact with the cage bottom, resulting in approximately a 30° head-down tilt. The forelimbs of the animal maintained contact with the cage bottom, allowing the rat full access to the entire cage. On days 9 and 2 before death, animals were given subcutaneous injections of calcein (25 mg/kg) to label mineralizing J Appl Physiol • VOL

643

bone for histomorphometric analysis. On the day of death, animals were anesthetized with a ketamine-xylazine cocktail. HU animals were anesthetized before removal from tail suspension to prevent any weight bearing by the hindlimbs. Blood was collected via cardiac puncture, and plasma was stored at ⫺20°C until analysis. After decapitation, right and left tibiae and femora were removed and cleaned of soft tissue. Right tibiae and femora were stored at ⫺80°C in PBS-soaked gauze for ex vivo pQCT and mechanical testing, whereas left tibiae and femora were stored in 70% ethanol for histology. Right soleus muscles were dissected free, and wet weights were obtained as a marker of disuse atrophy. pQCT. Bone scans of the tibiae and femora were performed by use of an XCT Research M (Stratec; Norland, Fort Atkinson, WI). This model has a minimum voxel size of 0.07 mm and a scanning beam thickness of 0.50 mm. Machine calibration was performed daily with the use of a standard hydroxyapatite cone phantom to ensure machine precision over time. Tibiae were scanned in vivo on days 0 and 28 at the proximal metaphysis with three slices taken at 5.0, 5.5, and 6.0 mm distal to the proximal plateau, a site altered after 28 days of HU in similar age and strain male rats (4). Ex vivo scan sites included the tibiae and femora middiaphyses (three slices of each bone located at one-half the total bone length ⫾ 2 mm) and the femora distal metaphysis (three slices located 5.0, 5.5, and 6.0 mm proximal to the distal plateau). Additionally, on the basis of contact radiography visualization postdeath, proximal tibiae were rescanned ex vivo with the slice locations adjusted upward (3.5, 4.0, and 4.5 mm distal to the proximal plateau) to ensure that cancellous bone was present in all analyses. All scans were obtained at a computed tomography speed of 2.5 mm/s with a voxel resolution of 0.10 ⫻ 0.10 ⫻ 0.50 mm for in vivo measures and 0.07 ⫻ 0.07 ⫻ 0.50 mm for ex vivo measures. A standardized analysis for either metaphyseal bone (contour mode 3, peel mode 2, outer threshold of 0.214 g/cm3, inner threshold of 0.605 g/cm3) or diaphyseal bone (separation 1, threshold of 0.605 g/cm3) was applied to each section. The same contour and peel modes and thresholds were used by our laboratory to successfully differentiate cortical and cancellous bone in skeletally mature unloaded animals (4) and are explained in detail elsewhere (10). Values of total bone mineral content (BMC); total, cortical, and cancellous bone mineral density (BMD); and total, cortical bone, and marrow area for each of the three slices were averaged to get a mean value at each site. In addition, middiaphyseal crosssectional moment of inertia (CSMI) was obtained with respect to the neutral bending axis during three-point bending. Machine precision (based on manufacturer’s data) is ⫾3 mg/cm3 for cancellous BMD and ⫾9 mg/cm3 for cortical BMD. Reproducibility in our laboratory for both in vivo and ex vivo measures was determined from five repeat scans with repositioning of the animal or bone between scans. Coefficients of variation for these measurements were 1.24, 2.13, and 1.95% for in vivo proximal tibia total BMD, cancellous BMD, and total area, respectively. Ex vivo distal metaphysis coefficients of variation for the same variables were 0.37, 1.43, and 0.28%, respectively. Cortical histomorphometry. Undemineralized left tibiae were subjected to dehydration using graded ethanols followed by xylene and were finally embedded in methylmethacrylate (Aldrich M5, 590-9). Serial cross-sections (150–200 ␮m) were cut starting 2.5 mm proximal to the tibial-fibular junction by use of a diamond wafer Isomet low-speed saw (Buehler, Lake Bluff, IL). Sections were hand ground to reduce thickness (⬍80 ␮m) and mounted on glass slides. Histological analysis of fluorochrome-labeled bone surfaces

94 • FEBRUARY 2003 •

www.jap.org

644


was performed by use of BioQuant TrueColor Windows image processing analysis system (R&M Biometrics, BQTCW98, version 3.50.6). Measures of labeled surfaces were obtained at ⫻100 magnification, whereas interlabel widths were measured at ⫻200. Periosteal and endocortical mineral apposition rates (MAR, ␮m/day) were calculated by dividing the interlabel width by the time between labels (7 days), and mineralizing surface (MS/BS) for both periosteal and endocortical bone surfaces was calculated by using the formula MS/BS ⫽ {[(single labeled surface/ 2) ⫹ double label surface]/surface perimeter} ⫻ 100. BFR was calculated as (MAR ⫻ MS/BS). Cancellous histomorphometry. Undemineralized distal left femora were embedded in methylmethacrylate with the use of a protocol similar to that for the tibiae. Serial frontal sections (8 ␮m) were cut by use of a motorized rotary microtome (HM355, Carl Zeiss; Thornwood, NY) and affixed to slides. Half of the slides were left unstained for fluorochrome label measurements, and the other half were stained by the VonKossa staining procedure. At ⫻20, a defined region of interest was established ⬃0.8 mm from the growth plate and encompassing 7–8 mm2. Total bone surface, single-labeled surface, and double-labeled surface were measured at ⫻100 whereas interlabel distances were measured at ⫻200 magnification. MAR, MS/BS, and BFR were calculated in the same manner as for cortical bone. Stained section region of interest (7–8 mm2) was determined at ⫻20 and then analyzed at ⫻200 for measures of bone volume and osteoblast/osteoclast surface. All nomenclature for both cortical and cancellous histomorphometry follows standard usage (22). Mechanical testing. Tibiae and femora middiaphyseal mechanical properties were determined by three-point bending by use of an Instron 1125 testing machine. Bones were thawed at room temperature and placed posterior side down on metal pin supports located ⫾9 mm from the middiaphysis testing site. With the use of a 1,000-lb. load cell calibrated to 100 lb. maximum load, quasi-static loading (2.5 mm/min) was applied to the anterior surface of both the tibiae and femora until fracture. All specimens were sprayed with PBS just before testing to maintain hydration. Displacements of the servo-controlled Instron were monitored by a linear variable differential transformer interfaced with a personal computer (Gardener Systems software). Raw data, collected at 10 Hz as load vs. displacement curves, were analyzed with TableCurve 2.0 (Jandel; San Rafael, CA). Structural variables were obtained directly from the load/ displacement curves. The maximum load obtained was defined as ultimate load (UL), the slope of the elastic portion of the curve defined as stiffness (S), and area under the curve to fracture defined as total energy absorbed. Material properties were calculated by normalizing structural properties to bone geometry at the site of testing using CSMI (from pQCT), bone diameter (D) measured by calipers, and a support span distance of 18 mm. Appropriate formulas are as follows: modulus ⫽ S ⫻ support span3/(48 ⫻ CSMI ⫻ 1,000); ultimate stress ⫽ [UL ⫻ support span ⫻ (D/2)]/(4 ⫻ CSMI). Contact radiography. Excised tibiae were X-rayed by using contact radiography (General Electric Industrial Machine; Lexington, MA) with Kodak X-Omat TL Film (Eastman Kodak, Rochester, NY). Scans were obtained by using a setting of 25 kV, 1 mA, 30-in. focal film distance, and 50-s exposure time. The developed X-rays were scanned into a personal computer for image analysis. No direct measures were performed on the images. 17␤-Estradiol analysis. Plasma levels of 17␤-estradiol were measured by using a double-antibody radioimmunoassay kit (Diagnostic Products; Los Angeles, CA) as directed. J Appl Physiol • VOL

Briefly, 100 ␮l of unextracted plasma were assayed in duplicate and averaged for reporting by using a single assay kit. Intra-assay coefficient of variations were 4.6 ⫾ 0.5% with a minimal limit of detection of 5 pg/ml. Data analysis. Body weight and food consumption were evaluated by use of repeated-measures ANOVA. When significant interactions were found, simple main effects analysis using Duncan’s multiple range post hoc test were used to determine significant differences within time or group. A P value of ⬍0.05 was used to determine statistical significance in all tests. In vivo pQCT data were evaluated by using a paired Student’s t-test. Ex vivo pQCT, histomorphometry, mechanical testing, and 17␤-estradiol results were compared by using two-tailed Student’s t-tests. Correlations were assessed between 17␤-estradiol levels and pQCT (total, cancellous BMD; total, cortical area), all histological variables, and all mechanical property variables. All values in tables, graphs, and text are presented as means ⫾ SE. All differences referred to in results and discussion are significant at a level of ⬍0.05 unless otherwise noted; trends (P ⬍ 0.10) are individually cited. RESULTS

Animal characteristics. Baseline body masses were not different between groups (CC: 276 ⫾ 3.8 g; HU: 284 ⫾ 6.3 g). After 28 days, body mass was maintained in CC animals, whereas it decreased (⫺6%; P ⬍ 0.01) in HU animals. Daily food consumption was not different between groups at baseline; by week 2, HU animals had a greater daily intake. This difference persisted throughout the study, and by week 4, HU animals were consuming more food per day compared with their baseline levels (⫹33%) and compared with CC (⫹39%) (Fig. 1). Plasma 17␤-estradiol levels at death tended to be lower in HU animals compared with CC (⫺49%, P ⫽ 0.079) (Fig. 2). Soleus muscle mass was lower (P ⬍ 0.001) in HU animals (59 ⫾ 2 mg) relative to CC (126 ⫾ 6 mg), confirming the effectiveness of the HU intervention. In vivo pQCT. There were no differences in pQCTmeasured bone parameters between the two groups at

Fig. 1. Average daily food intake over 28 days. Data are means ⫾ SE of daily intakes for the preceding week. * Significantly different (P ⬍ 0.05) compared with baseline values. †Significantly different (P ⬍ 0.05) compared with cage-control (CC) values within a specific time point.


www.jap.org


Fig. 2. 17␤-Estradiol plasma concentrations. Data are means ⫾ SE. Values for hindlimb-unloaded (HU) animals tended to be lower than CC (P ⫽ 0.079).

baseline. Total BMC of the proximal tibia metaphysis was increased (⫹6%; P ⬍ 0.001) in CC animals after 28 days (baseline: 7.63 ⫾ 0.14; day 28: 8.09 ⫾ 0.12 mg/ mm) and decreased (⫺4%) in HU animals (baseline: 8.08 ⫾ 0.12; day 28: 7.75 ⫾ 0.21 mg/mm) (Fig. 3). Total BMD increased (⫹4.7%) in CC with no change in HU, whereas cortical bone area increased in CC (⫹5.7%; P ⬍ 0.001) and decreased in HU (⫺4.8%; P ⬍ 0.01) animals (Table 1). Cortical BMD, cancellous BMD, total area, and marrow area of the proximal tibia were all similar to baseline values after 28 days in both CC and HU animals. There was no correlation between pQCT-measured variables at baseline and the amount of change over 28 days nor was there a correlation between 17␤-estradiol levels and the amount of change over 28 days for any pQCT parameters. Contact radiography. Digital images of tibiae from all animals were inspected so as to evaluate the amount of cancellous bone; no direct measures were obtained. Surprisingly, the degree of variability in apparent cancellous bone volume and distribution within the metaphysis was very high, with many CC animals appearing to have lower amounts of bone compared with HU (Fig. 4). Ex vivo pQCT. Results from ex vivo analyses were similar to the in vivo measures, with CC animals having significantly higher total BMC, total BMD, and cortical bone area at the proximal tibia compared with HU rats (data not shown). Cancellous BMD at this more proximal location was not significantly different (P ⫽ 0.40) between CC and HU, with values of 269 ⫾ 16 and 247 ⫾ 17 mg/cm3, respectively. After 28 days, there were no significant differences in pQCT-measured bone parameters between CC and HU animals at the distal femur metaphysis (data not shown) or tibia middiaphysis (Table 2). Femoral middiaphyseal measures of the two groups were similar with respect to BMD and total area, yet the distribution of cortical bone was significantly different, as CSMI was larger (⫹10%) in HU compared with CC animals (Table 2). J Appl Physiol • VOL

645

Mechanical properties. There were no significant differences of tibial or femoral middiaphyseal structural properties (ultimate load, stiffness, energy absorbed) between CC and HU (Table 2). Femoral middiaphysis modulus, a measure of material stiffness, was lower (⫺13%) in HU compared with CC. There was no correlation between any mechanical properties variables and 17␤-estradiol levels. Histomorphometry. There was no difference in cortical bone area between CC and HU as assessed 2.5 mm proximal to the tibial-fibula junction (3.80 ⫾ 0.05 vs. 3.83 ⫾ 0.07 mm2, respectively). Dynamic measures of bone formation, assessed over the final 9 days of the experiment, were dramatically different between HU and CC near the tibia middiaphysis. There was no detectable double label on the periosteal or endocortical surfaces in any HU tibiae, disallowing calculation of MAR and BFR in this group. CC animals had periosteal and endocortical MAR of 1.81 ⫾ 0.51 and 1.43 ⫾ 0.63 ␮m/day, respectively. HU animals also had a significantly lower MS/BS on both the periosteal (⫺68%; P ⬍ 0.001) and endocortical surfaces (⫺53%) compared with CC (Fig. 5). There was no correlation between MS/BS and 17␤-estradiol levels (P value of 0.99 and 0.83 for correlation with periosteal and endocortical MS/BS, respectively). Dynamic and cell surface histomorphometric measures of distal femur cancellous bone were not significantly different between CC and HU for any variable (Table 3). Double label was observed in all cancellous bone sections, confirming that both labels were properly administered to all animals. DISCUSSION

These results provide unique longitudinal data on the effects of long-duration HU on the ovary-intact adult retired breeder female rat skeleton. We demonstrate here that 28 days of unloading produce a significant decline in proximal tibia BMC and cortical bone area with no resultant effect on total or cancellous bone

Fig. 3. In vivo peripheral quantitative computed tomography (pQCT) total bone mineral content from proximal tibia metaphysis measured at baseline and after 28 days (d) of treatment. Data are means ⫾ SE. * Significantly different (P ⬍ 0.05) vs. baseline values within treatment group. ** Significantly different (P ⬍ 0.001) vs. baseline values within treatment group.


www.jap.org

646


Table 1. In vivo pQCT at the proximal tibia metaphysis Cage Control

3

Total BMD, mg/cm Cortical BMD, mg/cm3 Cancellous BMD, mg/cm3 Total area, mm2 Cortical bone area, mm2 Marrow area, mm2

Hindlimb Unloaded

Baseline

28 Days

Baseline

28 Days

636 ⫾ 9 1,134 ⫾ 7 166 ⫾ 8 12.0 ⫾ 0.16 5.82 ⫾ 0.11 6.20 ⫾ 0.15

666 ⫾ 12* 1,150 ⫾ 7 170 ⫾ 9 12.2 ⫾ 0.23 6.15 ⫾ 0.09† 6.05 ⫾ 0.20

640 ⫾ 18 1,142 ⫾ 7 177 ⫾ 9 12.7 ⫾ 0.38 6.04 ⫾ 0.08 6.67 ⫾ 0.39

636 ⫾ 12 1,142 ⫾ 7 184 ⫾ 12 12.2 ⫾ 0.28 5.75 ⫾ 0.14†‡ 6.48 ⫾ 0.21

Values are means ⫾ SE. All data are from proximal tibia metaphysis (5.0, 5.5, and 6.0 mm distal to the proximal tibia plateau) on days 0 and 28. BMD, bone mineral density. * Significant difference (P ⬍ 0.05) from baseline value. † Significant difference (P ⬍ 0.01) from baseline. ‡ Significant difference (P ⬍ 0.05) from 28-day cage-control value.

density as assessed by pQCT. Diaphyseal cortical BFR and actively mineralizing surfaces were significantly lower in unloaded animals compared with cage controls. Differences in cancellous bone volume and formation rate at the distal femur, however, were not detectable after 28 days of HU. Additionally, changes in daily food intake and serum estradiol were observed in unloaded animals compared with controls. Both of these parameters potentially have a significant impact on bone adaptation independent of weight-bearing status. To our knowledge this is the first study to address the effect of HU on serum estradiol in skeletally ma-

ture, nonovariectomized female rats. The present study shows that, although the change was not statistically significant (P ⫽ 0.079), serum estradiol was considerably lower (⫺49%) in HU animals compared with age-matched cage controls after 28 days. Previous studies provide support for decreased sex steroid levels with unloading, as male rats exposed to either HU or spaceflight showed significant reductions in serum testosterone compared with controls (17). A proposed mechanism for the decreased testosterone in males is a reduction in testes blood flow, thus compromising the efflux of testosterone into the circulatory system (1). It is plausible that HU could produce similar decrements

Fig. 4. Contact radiographs of selected HU and CC tibiae. Transverse line is approximately at location of in vivo pQCT scan (⬃5.5 mm distal to the proximal plateau). A: HU high cancellous bone mineral density (BMD; 250 mg/cm3). B: HU low cancellous BMD (150 mg/cm3). C: CC high cancellous BMD (191 mg/cm3). D: CC low cancellous BMD (136 mg/cm3). BMD values are from in vivo pQCT analysis on day of death (day 28). J Appl Physiol • VOL


www.jap.org

647


Table 2. Tibiae and femora middiaphysis density, geometry, and strength Tibia

Cortical BMC, mg/mm Cortical BMD, mg/cm3 Total area, mm2 Cortical area, mm2 Marrow area, mm2 CSMI, mm4 Ultimate load, N Stiffness, N/mm Energy to fracture, mJ Ultimate stress, MPa Elastic modulus, GPa

Femur

Cage control

Hindlimb unloaded

Cage control

Hindlimb unloaded

5.83 ⫾ 0.1 1,385 ⫾ 2 5.9 ⫾ 0.1 4.25 ⫾ 0.07 1.67 ⫾ 0.07 2.9 ⫾ 0.12 91.3 ⫾ 3 263 ⫾ 10 49.2 ⫾ 2 204 ⫾ 9 11.2 ⫾ 0.36

5.90 ⫾ 0.1 1,382 ⫾ 3 6.2 ⫾ 0.2 4.38 ⫾ 0.08 1.84 ⫾ 0.14 3.3 ⫾ 0.20 99.0 ⫾ 5 280 ⫾ 21 57.2 ⫾ 6 203 ⫾ 4 10.3 ⫾ 0.64

8.0 ⫾ 0.3 1,405 ⫾ 8 9.61 ⫾ 0.35 5.74 ⫾ 0.20 3.88 ⫾ 0.36 5.48 ⫾ 0.38 118 ⫾ 10 343 ⫾ 17 42.5 ⫾ 7 168 ⫾ 10 7.6 ⫾ 0.4

8.3 ⫾ 0.4 1,407 ⫾ 8 10.1 ⫾ 0.77 5.96 ⫾ 0.32 4.17 ⫾ 0.52 6.18 ⫾ 0.66* 130 ⫾ 6 330 ⫾ 22 58.8 ⫾ 6 159 ⫾ 6 6.6 ⫾ 0.3*

Values are means ⫾ SE. Peripheral quantitative computed tomography (pQCT) scans were obtained at 50% of total bone length. Mechanical properties were assessed at a similar location by using 3-point bending. BMC, bone mineral content; CSMI, cross-sectional moment of inertia. * Significant difference (P ⬍ 0.05) from control value.

in ovarian blood flow, thus accounting for the reduced serum estrogen. Other factors, secondary to HU, could also contribute to the lower serum estradiol values shown in this study. We did not verify whether these rats were at different stages of their estrus cycle on the day of death, a factor that could contribute to differences in serum estrogen. Another potential cause of lower estrogen values is stress from the unloading procedure, although the fact that HU animals increased food intake, maintained body mass, and showed no change in grooming patterns during unloading makes this a less likely explanation. Future studies should be designed to confirm this reduction in serum estrogen in intact

female rats, controlling for factors such as the estrus cycle. Another novel finding of this study was that these unloaded animals significantly increased their daily food intake, while simultaneously maintaining body mass. The acute drop in food consumption during the first week in both unloaded and control animals is likely the result of a reduced intake during the day or two after anesthesia. The changes thereafter, however, could be due to a number of factors, one of which is changes in estrogen levels. Rats that undergo OVX, thereby producing complete estrogen deficiency, eat significantly more food; when exogenous estrogen is provided, food intake returns to baseline levels (34). If declines in serum estrogen began within the first week of HU, this may have provided a stimulus for increased food consumption for the duration of the study. However, unlike the OVX model, in which both food intake and body mass increase, the greater food intake of these HU female rats did not result in an increased body mass. This suggests that HU may negatively affect energy absorption or some other important parameter that may influence body mass. These present data, however, cannot directly answer these issues. The in vivo longitudinal pQCT data presented herein suggest that despite a significant decrease in proximal tibia total mineral content after 28 days of unloading, total metaphyseal bone density was maintained. Unloading does, however, attenuate the normal increase Table 3. Distal femur metaphysis histomorphometry Cage Control

Fig. 5. Cortical and cancellous histology measures of mineralizing surface (MS/BS). Data are means ⫾ SE. * Significantly different (P ⬍ 0.05) compared with CC group. J Appl Physiol • VOL

BV/TV, % Osteoblast surface, % Osteoclast surface, % Cancellous MAR, ␮m/day Cancellous BFR, ␮m3 䡠 ␮m⫺2 䡠 day⫺1

Hindlimb Unloaded

15.5 ⫾ 1.9 4.1 ⫾ 1.2 3.3 ⫾ 0.59

15.5 ⫾ 0.33 5.5 ⫾ 1.59 3.3 ⫾ 0.88

1.81 ⫾ 0.12

1.61 ⫾ 0.14

133.8 ⫾ 19

108.8 ⫾ 15

Values are means ⫾ SE. BV/TV, bone volume/tissue volume; MAR, mineral apposition rate; BFR, bone formation rate.


www.jap.org

648


in total density that occurs, as control rats exhibit a significant increase of total BMD over 28 days (⫹5%). Cortical bone area decreased over 28 days in HU animals and increased in CC whereas neither group had significant changes in total or marrow area. These changes in both CC and HU cortical bone area are most likely explained by subtle changes in the balance of formation and resorption at periosteal and endocortical surfaces, although the lack of significant change in total and marrow areas render further interpretation difficult. In an attempt to explain changes in BMC and cortical bone area in unloaded animals, correlations between these variables and serum estradiol levels were investigated. Because no significance relationship was found, this suggests that the drop in serum estradiol may not be directly related to the unloading-induced bone adaptations in these retired breeder female rats. Our unloaded rats did not exhibit any decline in cancellous bone density at the proximal tibia, as measured longitudinally over 28 days. This is intriguing because cancellous bone has previously been shown to be adversely affected by a 28-day period of unloading in similar strain and age male rats (4). Interestingly, X-ray images of the excised tibiae revealed significant variation in the amount and distribution of cancellous bone at death, even within treatment groups (Fig. 4). Numerous animals had very little cancellous bone near the in vivo scan location, possibly accounting for the lack of change in cancellous density over 28 days. Ex vivo results, using a more proximal scanning location, confirmed no difference between unloaded and control rats with respect to proximal tibia cancellous bone density. Additionally, there was no difference in distal femur cancellous bone BMD measured ex-vivo at the conclusion of the study (data not shown). The scarcity of cancellous bone in these retired breeder rats is further supported by comparing cancellous bone density values to those our laboratory has measured in CC virgin female rats of the same strain and age. The latter exhibit values for proximal tibia cancellous BMD that are, on average, two times greater (350 ⫾ 15 mg/cm3; unpublished data) than values for the retired breeders of the present study. This lack of change in cancellous bone density after 28 days of unloading in these retired breeder female rats may well be the result of typical animal breeding practices, causing a disruption of bone turnover before the initiation of our study. Before arrival at our facility at the age of 6 mo, these rats had likely gone through multiple breeding cycles, initiated by the vendor at approximately 2 mo of age. Previous studies have shown that both pregnancy and lactation have a negative impact on the rat skeletal system, particularly cancellous bone (5, 16). The effect of lactation on cancellous bone loss is even greater than that caused by ovariectomy, reducing bone volume from a control value of 30% to just 6% after 3 wk; ovariectomy for the same time period reduces cancellous bone volume to 19% (5). Specifically, the size of each litter (ranging from 4 to 17) affects the variability in the degree of J Appl Physiol • VOL

skeletal mass loss (16), with estimates of a 6.7 mg/cm3 reduction in BMD for each pup (28). In addition, the timing of the most recent litter in relation to shipping could have dramatic consequences on the state of the skeleton at time of study initiation. It is plausible that these retired breeder females had so little cancellous bone at baseline that they were less sensitive to unloading and thus did not experience measurable declines in cancellous bone density over 28 days. It should be noted, however, that no correlation existed between pQCT-measured cancellous bone density at baseline and the amount of change in this variable over 28 days. When assessing changes in cancellous BMD (or lack of change) as measured by pQCT, we must also consider changes in marrow area. Because the resolution (70 ␮m) is not high enough to measure the “true” trabecular BMD, “apparent” cancellous BMD is calculated by dividing mineral content of the region by marrow area. Therefore, changes in marrow area could artificially influence this measure of cancellous BMD. However, because we did not detect a significant change over time of proximal tibia marrow area in either group, we feel that the lack of change in cancellous BMD over 28 days is a reliable finding. Assuming that the sparseness of cancellous bone in these rats is not limited to the proximal tibia, it is not surprising that distal femur cancellous bone histological measures were similar between groups after 28 days. Although histologically measured cancellous bone volume has been shown to be unchanged in mature animals after long-term HU (8, 24), declines in MAR and/or BFR are commonly found. Our results suggest similar cancellous bone mineral apposition and formation rates of the distal femur in CC and HU animals, as assessed over the final 9 days of the experiment, although our statistical power to detect a difference was low. Because our differences between groups were less than expected and the variability within groups was greater than expected, conclusions based on these cancellous histomorphometry data must be conservative. It is interesting to note, though, that the differences in MAR (⫺11%), BFR (⫺19%), and mineralizing surface (⫺11%) of the distal femur cancellous bone between HU and CC animals in this study are substantially less than has been previously shown over a similar period of unloading using similar aged animals (8, 24). Because bone formation is highly dependent on the status of the animal at baseline, the effects of pregnancy and lactation on cancellous bone may mask changes caused by HU and cause a wider variation in the response of female rats. There also may be an interaction of declining estrogen levels (which stimulate formation) and unloading (which depress formation) so as to result in no change in cancellous formation rate. Further investigation to separate these two effects is necessary. The lack of density or area differences at the tibial and femoral middiaphyses between unloaded and control rats has been previously shown in similar age and strain male rats (4), suggesting that longer periods of


www.jap.org


unloading may be necessary to cause a perturbation of bone mass or geometry at sites composed solely of cortical bone. This maintenance of density and area occurs despite a complete cessation of measurable bone formation at the tibia middiaphysis. The 68% reduction in periosteal labeled surface was accompanied by a complete lack of any double-labeled surface, suggesting that bone formation had slowed to nondetectable levels by the final 9 days of HU. Despite the cessation of bone formation by the last week of HU, there was no resulting decrement in whole bone strength of the tibial diaphysis. These mechanical strength results are markedly different from those observed in immature rats, which exhibit declines in whole bone strength and stiffness with unloading or spaceflight relative to age-matched controls (23, 25, 27, 31). It would appear that the changes in immature animals caused by unloading represent an impairment of the usual growth-related gains in strength rather than an absolute loss of bone strength. Although whole bone strength was not different between groups for the tibia or femora middiaphyses, elastic modulus was significantly lower in the unloaded femurs (⫺13%) compared with controls. This reduction in an important material property is compensated for by a larger femoral middiaphysis CSMI in the unloaded animals. This is somewhat paradoxical, because formation rates on the tibial diaphysis had ceased by 28 days. Because direct measures of bone formation were not performed at a similar site on the femur, we can only speculate as to why these results were found. The most likely explanation is that either 1) differences existed in CSMI at baseline, as in vivo analysis of this site was not possible; or 2) periosteal bone was deposited early during unloading, so as to increase CSMI, yet this new bone was undermineralized, resulting in lower tissue modulus. It is important to note, however, that two bones (tibiae and femora) that were similarly unloaded appear to respond differently. Further work should focus on these site-specific differences to help explain why they occur. In conclusion, the data reported herein demonstrate that skeletally mature retired breeder female rats experience a loss of proximal tibia total BMC and cortical shell area whereas both total density and cancellous bone density are maintained after 28 days of HU. Additionally, unloaded animals did not exhibit a detectable difference in dynamic or cell surface cancellous bone parameters of the distal femur metaphysis compared with controls. These results are unique in that metaphyseal cancellous bone has previously been shown to be highly sensitive to unloading in both mature male rats and OVX female rats. Unloading did result in a significantly lower tibial midshaft periosteal bone formation, yet this did not translate into differences in density, geometry, or mechanical strength at this site. Because of the uncertain breeding history of retired breeder female rats, future studies should confirm these results by using similar methods in skeletally mature nonovariectomized virgin female rats. Additionally, the role of increased daily food consumption J Appl Physiol • VOL

649

during unloading should be further investigated because it may contribute to cancellous bone maintenance in the mature female unloaded skeleton. We gratefully acknowledge the technical assistance of Jan Stallone with the hindlimb suspension procedure and general lab assistance, Dr. John Stallone with 17␤-estradiol analysis, and Dr. Harry Hogan who assisted with mechanical testing and provided insightful discussion of the results. The National Aeronautics and Space Administration (NASA) supported this work through the NASA Cooperative Agreement NCC 9-58 with the National Space Biomedical Research Institute. REFERENCES 1. Amann RP, Deaver DR, Zirkin RB, Frills GS, Sapp WJ, Veeramachaneni DNR, Clemens JW, Banerjee SD, Folmer J, Cruppi CM, Wolgenmuth DJ, and Williams CS. Effects of microgravity or simulated launch on testicular function in rats. J Appl Physiol 73: 174S–185S, 1992. 2. Bateman TA, Broz JJ, Fleet ML, and Simske SJ. Differing effects of two-week suspension on male and female mouse bone metabolism. Biomed Sci Instrum 34: 374–379, 1998. 3. Bikle DD and Halloran BP. The response of bone to unloading. J Bone Miner Metab 17: 233–244, 1999. 4. Bloomfield SA, Allen MR, Hogan HA, and Delp MD. Siteand compartment-specific changes in bone with hindlimb unloading in mature adult rats. Bone 31: 149–157, 2002 5. Bowman BM and Miller SC. Skeletal adaptations during mammalian reproduction. J Musculoskel Neuron Interact 1: 347–355, 2001. 6. Cavolina JM, Evans GL, Harris SA, Zhang M, Westerlind KC, and Turner RT. The effects of orbital spaceflight on bone histomorphometry and messenger ribonucleic acid levels for bone matrix proteins and skeletal signaling peptides in ovariectomized growing rats. Endocrinology 138: 1567–1576, 1997. 7. Collet P, Uebelhart D, Vico L, Moro L, Hartmann D, Roth M, and Alexandre C. Effects of 1- and 6-month spaceflight on bone mass and biochemistry in two humans. Bone 20: 547–551, 1997. 8. Dehority W, Halloran BP, Bikle DD, Curren T, Kostenuik PJ, Wronski TJ, Shen Y, Rabkin B, Bouraoui A, and Morey-Holton E. Bone and humoral changes induced by skeletal unloading in the mature male rat. Am J Physiol Endocrinol Metab 276: E62–E69, 1999. 9. Evans GL, Morey-Holton E, and Turner RT. Spaceflight has compartment- and gene-specific effects on mRNA levels for bone matrix proteins in rat femur. J Appl Physiol 84: 2132–2137, 1998. 10. Ferretti JL. Perspectives of pQCT technology associated to biomechanical studies in skeletal research employing rat models. Bone 17, Suppl 4: 353S–364S, 1995. 11. Frost HM. The Utah paradigm of skeletal physiology: an overview of its insights for bone, cartilage and collagenous tissue organs. J Bone Miner Metab 18: 305–316, 2000. 12. Garber MA, McDowell DL, and Hutton MC. Bone loss during simulated weightlessness: a biomechanical and mineralization study in the rat model. Aviat Space Environ Med 71: 586–592, 2000. 13. Halloran BP, Bikle DD, Wronski TJ, Globus RK, and Morey-Holton E. The role of 1,25-dihydroxyvitamin D in the inhibition of bone formation induced by skeletal unloading. Endocrinology 118: 948–954, 1986. 14. Harm DM, Jennings RT, Meck JV, Powell MR, Putcha L, Sams CP, Schneider SM, Shackelford LC, Smith SM, and Whitson PA. Invited Review: gender issues related to spaceflight: a NASA perspective. J Appl Physiol 91: 2374–2383, 2001. 15. Harris SA, Zhang M, Kidder LS, Evans GL, Spelsberg TC, and Turner RT. Effects of orbital spaceflight on human osteoblastic cell physiology and gene expression. Bone 26: 325–331, 2000. 16. Honda A, Kurabayashi T, Yahata T, Tomita M, Matsushita H, Takakuwa K, and Tanaka K. Effects of pregnancy and


www.jap.org

650

17.

18.

19. 20.

21.

22.

23.

24.

25.

26.


lactation on trabecular bone and marrow adipocytes in rats. Calcif Tissue Int 67: 367–372, 2000. Merrill AH, Want E, Mullins RE, Grindeland RE, and Popova IA. Analysis of plasma for metabolic and hormonal changes in rats flown aboard COSMOS 2044. J Appl Physiol 73: 132S–135S, 1992. Moon LY, Wakley GK, and Turner RT. Dose-dependent effects of tamoxifen on long bones in growing rats: influence of ovarian status. Endocrinology 129: 1568–1574, 1991. Morey ER and Baylink DJ. Inhibition of bone formation during spaceflight. Science 201: 1138–1141, 1978. Morey-Holton ER and Globus RK. Invited Review: hindlimb unloading rodent model: technical aspects. J Appl Physiol 92: 1367–1377, 2002. Morey-Holton ER, Whalen RT, Arnaud SB, and Van Der Meulen MC. The skeleton and its adaptation to gravity. In: Handbook of Physiology. Environmental Physiology. Bethesda, MD: Am. Physiol. Soc., 1996, sect. 4, vol. I, chapt. 31, p. 691–719. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, and Recker RR. Bone histomorphometry: standardization of nomenclature, symbols, and units. J Bone Miner Res 2: 595–610, 1987. Patterson-Buckendahl P, Arnaud SB, Mechanic GL, Martin RB, Grindeland RE, and Cann CE. Fragility and composition of growing rat bone after one week in spaceflight. J Appl Physiol 85: 279–285, 1987. Rubin C, Xu G, and Judex S. The anabolic activity of bone tissue, suppressed by disuse, is normalized by brief exposure to extremely low-magnitude mechanical stimuli. FASEB J 15: 2225–2229, 2001. Shaw SR, Vailas AC, Grinderland RE, and Zernicke RF. Effects of a 1-wk spaceflight on morphological and mechanical properties of growing rats. Am J Physiol Regul Integr Comp Physiol 254: R78–R83, 1988. Smith BJ, King JB, Lucas EA, Akhter MP, Arjmandi BH, and Stoecker BJ. Skeletal unloading and dietary copper deple-

J Appl Physiol • VOL

27. 28.

29. 30. 31.

32.

33.

34.

35. 36.

tion and detrimental to bone quality of mature rats. J Nutr 132: 190–196, 2002. Spengler DM, Morey ER, and Carter DR. Effects of spaceflight on structural and material strength of growing bone. Proc Soc Exp Biol Med 174: 224–228, 1983. Tojo Y, Kurabayashi T, Honda A, Yamamoto Y, Yahata T, Takakuwa K, and Tanaka K. Bone structural and metabolic changes at the end of pregnancy and lactation in rats. Am J Obstet Gynecol 178: 180–185, 1998. Turner RT. Invited Review: what do we know about the effects of spaceflight on bone? J Appl Physiol 89: 840–847, 2000. Turner RT, Evans GL, and Wakley GK. Spaceflight results in depressed cancellous bone formation in rat humeri. Aviat Space Environ Med 66: 770–774, 1995. Vailas AC, Zernicke RF, Grindeland FE, Kaplansky A, Durnova BN, Li KC, and Martinez DA. Effects of spaceflight on rat humerus geometry, biomechanics, and biochemistry. FASEB J 4: 47–54, 1990. Vico L, Bourrin S, Very JM, Radziszowska M, Collet P, and Alexandre C. Bone changes in 6-mo-old rats after head down suspension and a reambulation period. J Appl Physiol 79: 1426–1433, 1995. Vico L, Collet P, Guignandon A, Lafage-Proust MH, Thomas T, Rehailla M, and Alexandre C. Effects of long-term microgravity exposure on cancellous and cortical weight-bearing bones of cosmonauts. Lancet 355: 1607–11, 2000. Wallen WJ, Belanger MP, and Wittnich C. Sex hormone and the selective estrogen receptor modulator tamoxifen modulates weekly body weights and food intakes in adolescent and adult rats. J Nutr 131: 2351–2457, 2001. Wronski TJ and Morey ER. Effect of spaceflight on periosteal bone formation in rats. Am J Physiol Regul Integr Comp Physiol 244: R305–R309, 1983. Wronski TJ, Morey-Holton ER, Doty SB, Maese AC, and Walsh CC. Spacelab 3: histomorphometric analysis of rat skeleton following spaceflight. Am J Physiol Regul Integr Comp Physiol 252: R252–R255, 1987.


www.jap.org