Disuse has been shown to cause a rapid and dramatic loss of skeletal mass and strength in the load-bearing bones of young and mature animals and humans. However, little is known about the skeletal effects of disuse in aged mammals. The present study was designed to determine whether the skeletal effects of disuse are maintained with extreme age. Fischer 344/Brown Norway male rats (6 and 32 mo old) were hindlimb suspended (HS) or housed individually for 2 wk. Trabecular volume and microarchitecture in the proximal tibia were significantly decreased by HS only in young rats. HS significantly reduced cortical bone mineral density and increased cortical porosity only in old rats by inducing new pore formation. Cortical pore diameter was also increased in old rats, regardless of loading condition. Ex vivo osteogenic and adipogenic cultures established from each group demonstrated that age and HS decreased osteoblastogenesis. Age, but not HS, decreased sensitivity to endogenous bone morphogenetic protein stimulation, as measured by treatment with exogenous Noggin. Adipocyte development increased with age, whereas HS suppressed sensitivity to peroxisome proliferator-activated receptor-γ-induced differentiation. Serum insulin-like growth factor I levels were reduced with HS in young rats and with age in control and HS rats. These results suggest that the site of bone loss due to disuse is altered with age and that the loss of osteogenic potential with disuse in the old rats may be due to the combined effects of decreased insulin-like growth factor I levels and sensitivity, as well as diminished bone morphogenetic protein production.
- hindlimb suspension
- bone morphogenetic protein
- insulin-like growth factor I
aging is associated with functional changes in many tissues, such as a decline in muscle mass and strength and reduced bone density (8, 16, 30, 38, 42, 59). With aging, bone and muscle mass regress in a parallel fashion (51). The osteopenia has far-reaching consequences on the ability of the elderly to perform tasks of daily living and to be functionally independent. Indeed, one of the major concerns of aging is the loss of bone mass and strength, which is associated with an increased risk of fractures (1, 9, 18, 51). Moreover, the elderly are often subjected to periods of inactivity, such as bed rest, due to diseases or surgeries. The association of inactivity with decreases in muscle mass and bone density in young people and animals is well documented, and the mechanisms responsible for these losses have been the subject of intense investigation (2, 20, 23, 39, 40, 49, 51).
Recent evidence suggests that the increased bone fragility that occurs with aging is multifactorial: decreased bone mineral density (BMD), as well as changes in bone microarchitecture and strength, contribute to the increase in fracture risk that is associated with osteopenia (9, 12, 18). The cellular basis for age-related bone loss is similarly multifactorial: reductions in bone formation and increases in bone resorption (59). Studies in aging humans (11) and mice (62) have shown that the decreases in bone formation are related to a decrease in osteoblast progenitor number, proliferation, differentiation, and half-life. An increase in cellular apoptosis in aging bone is more prevalent in osteocytes, which are believed to act as the mechanosensors in bone (51). A similar increase in apoptosis of osteoblasts on the bone surface has also been described (11). Interestingly, the aging mesenchymal progenitors in mice (31, 61) and humans (46) also appear to cause a shift in lineage determination, such that cells preferentially differentiate toward the adipogenic, rather than the osteoblastic, phenotype in the bone marrow (42). However, progenitors from aging rodents appear to maintain some capacity to respond to osteogenic stimuli, such as bone morphogenetic protein (BMP) type 2 (BMP2) (62) and lasofoxifene (32), which protect against these deleterious effects of aging.
Despite the fact that disuse in the elderly poses the risk of exacerbating age-related bone loss, there are few reports on effects of disuse in aged humans or animals (21, 29). Because controlled experimental studies of disuse in elderly patients are ethically questionable, studies on this subject have been limited to animals. In addition, reports of age-related changes in the skeletal response to disuse in rodents are limited to middle-aged (∼1-yr-old) or younger animals (17). Consequently, mechanisms underlying the regulation of bone mass during disuse in aged rodents are not well understood.
Hindlimb suspension (HS) is a model that is frequently used to study the cellular and molecular mechanisms underlying skeletal muscle atrophy and bone loss. HS is induced by suspending the hindquarters of the animal and preventing normal weight bearing of the hindlimbs (43). In bone, HS is associated with a reduction in periosteal bone apposition rate and reduced bone formation (13), and our data also indicate a loss of BMD with HS (19, 45). Less is known about the effects of disuse on the skeleton in aging animals. However, rats up to 220 days of age have demonstrated osteopenia after 30 days of HS (17), suggesting that older animals maintain an osteopenic response to disuse. Thus any age-dependent changes in the cellular basis for the osteopenic response to HS, as well as the reloading response, are completely unknown.
HS induces increases in bone resorption, which have been indicated by increases in urinary calcium and in breakdown products of the major bone matrix protein collagen I (56, 57). However, several studies have suggested that a large component of the bone loss associated with disuse induced by HS is related to decreased bone formation, inhibition of mineralization, and delay in bone maturation (5, 13, 25, 39, 43, 64, 67). These changes in bone formation and maturation appear to be due, in part, to decreased proliferation and differentiation of osteoblast progenitors (26, 33, 56, 57, 71). In adult rats, the reduction in bone formation rate or osteoblast numbers induced by HS precedes a decrease in trabecular bone volume, which becomes significant at ∼14 days of HS (67). Because osteoblast numbers and bone formation rates are suppressed by HS, osteoblast proliferation, differentiation, and function are critical to the skeletal response to HS (43). Bone marrow stem cells exhibit impaired function with aging (50, 58), but their response to disuse in the aged rat and whether this contributes to impaired recovery of bone loss at old age are unknown. Thus it is important to elucidate the responses of the osteoblast population to disuse in extremely aged models.
Despite the rapidly growing aged population of industrialized nations, the skeletal effects of disuse in >1-yr-old rodents has not been previously reported. Therefore, the goal of this study was to determine the effects of HS on the skeletons of aged rats and to assess changes in the osteoblastogenic capacity of bone marrow cells with advancing age following disuse via HS.
Animals and experimental procedures.
All procedures were performed in accordance with institutional guidelines for the care and use of laboratory animals and were approved by the University of Arkansas Medical School Institutional Animal Care and Use Committee. Male Fischer 344/Brown Norway rats (6 and 32 mo of age) were purchased from the National Institute on Aging. This strain of rat has increased longevity and decreased cumulative lesion incidence compared with other strains; therefore, aging aspects can be studied in the relative absence of disease (4, 38, 65). The different ages were chosen to reflect a mature rat (6 mo) and an old rat at ∼50% mortality (32 mo). Rats of both ages were divided into two groups (n = 6–9): non-HS control and 14-day HS. Rats were allowed free access to food and water. Animals were housed in a 12:12-h light-dark cycle. HS was performed as previously described (22, 38). Briefly, a tail device containing a hook was attached with gauze and cyanoacrylate glue while the animals were anesthetized with pentobarbital sodium (50 mg/kg body wt). After the animal regained consciousness, the tail device was connected via a thin cable to a pulley sliding on a vertically adjustable stainless steel bar running longitudinally above a high-sided cage with standard floor dimensions. The system was designed in such a way that the rats could not rest their hindlimbs against any side of the cage. After 14 days of control housing or HS, rats were anesthetized with pentobarbital sodium (50 mg/kg) and blood was collected through heart puncture; then the rats were euthanized with an overdose of pentobarbital sodium (100 mg/kg).
Both hindlimbs were rapidly removed by disarticulation at the hip. The left femur was isolated and rinsed with 70% ethanol, and bone marrow cells were flushed with α-MEM-15% FBS with antibiotics, as previously described (45), and marrow was cultured (see below). Both tibiae were isolated, and the right tibia was stored in formalin for micro-CT evaluation and decalcified histological processing in paraffin. Histology was evaluated on 5-μm sections stained with hematoxylin and eosin. The left tibia was stored in 70% ethanol and later processed through graded ethanols before it was embedded in methylmethacrylate for nondecalcified histology, as described previously (60). Cellular histomorphometry was performed to determine adipocyte, osteoblast, osteoclast, and osteocyte number normalized to bone area and bone perimeter in the proximal tibia, as previously described (52).
BMD measurements by dual-energy X-ray absorptiometry.
Longitudinal bone densitometry was performed using a bone densitometer (Piximus; Lunar, Madison, WI) to obtain measurements of BMD and bone mineral content from the proximal right tibia of each animal, as described previously (19, 52). Scans were performed under anesthesia for attachment of the tail device on day 0, on the day HS (or control housing) began, and on day 14 immediately before the animal was killed. Subsequently, femur and tibia scans were divided into thirds: proximal and distal thirds for analysis of changes in cancellous bone density and the diaphyseal third for analysis of changes in cortical bone density, as previously described (19). The precision and accuracy of the bone densitometer have been determined by repeated measurements of five animals, five times each, as described elsewhere (44, 52). In-house precision analyses have been previously determined for adult rat femoral BMD (19) to have a coefficient of variation of 0.1%.
Culture of bone marrow cells for ex vivo osteoblastogenesis and adipogenesis.
Bone marrow cells were flushed from the left femur of each rat with α-MEM containing 15% FBS and antibiotics and cultured as previously described (45). Briefly, marrow cells from rats were rinsed in flushing medium and resuspended to obtain a single-cell suspension. The nucleated cell count was determined from pooled cells from each group, and cells were seeded in at least triplicate wells for each in vivo and in vitro treatment. Cells were cultured for osteoblastogenesis at densities of 1.5 × 106 cells/well [colony-forming units-fibroblast (CFU-F)] or 2.5 × 106 cells/well [osteoblastic colony-forming units (CFU-OB)] in 12-well plates in 15% FBS containing α-MEM and supplemented with 10 nM dexamethasone, 50 μM ascorbic acid, and 10 mM β-glycerophosphate to support mineralization. For determination of fully mature CFU-OB, the mineral deposits were stained with alizarin red on day 28 of culture, as previously described (45). Similar marrow cultures were established for adipocyte colony formation (CFU-AD), as previously described (42). Cells were seeded at a density of 1.25 × 166 cells per well in 12-well plates and maintained for 10 days with basal medium (α-MEM containing 15% FBS). On day 10 of culture, the medium was changed to medium containing 1 μM rosiglitazone maleate (Avandia; GlaxoSmithKline, King of Prussia, PA) and maintained for 6 days or IHI cocktail [60 μM insulin, 0.5 μM hydrocortisone, and 0.5 mM IBMX (Sigma, St. Louis, MO)] and maintained for 3 days; then IHI medium was replaced with basal medium, and cells were maintained for 3 more days. Cultures were refed every 3 days, harvested on day 16, and stained with oil red O to detect colonies that had undergone adipocyte differentiation, as previously described (35, 36, 42).
Trabecular bone volume and skeletal architecture in the right tibial metaphysis of each rat was measured by ex vivo micro-CT (model μCT40, Scanco Medical, Bassersdorf, Switzerland) using the manufacturer's software. Briefly, cross-sectional images were obtained with a voxel size of 16 μm in all dimensions. Semiautomated contouring was used to select a region of interest (ROI), comprising the secondary spongiosa extending 3.2 mm distal to the primary spongiosa but excluding the cortex and subcortical bone, composed of 150 adjacent 16-μm slices. For calculation of the three-dimensional volume and architecture of the secondary spongiosa, the volume of each slice was stacked before application of an optimized Gaussian noise filter and gray-scale threshold (63). Bone volume per tissue volume (BV/TV), architectural parameters [trabecular thickness (TbTh), number (TbN), and separation (TbSp)], and structural indexes, such as connectivity density (ConnD) and structure model index (SMI), were calculated directly from the reconstructed trabecular structures, as described elsewhere (27, 63).
Similar semiautomated techniques were used to measure and quantify parameters of cortical porosity. Total tissue volume includes any natural porosity that is included in the determination of cortical BV/TV. The cortical ROIs were obtained by contouring the same cross-sectional images described above, except the cortical and subcortical bones were included and the trabecular bone was excluded. ROIs were composed of 100 adjacent 16-μm slices. For calculation of parameters, the volume of each slice was stacked before application of an optimized Gaussian noise filter and gray-scale threshold (63). In these analyses, the structure and size of cortical pores in each bone, as well as cortical porosity, were examined. Cortical porosity was defined as the volume of pores divided by the volume of cortical bone and was expressed as a percentage. The distribution of pore thickness was calculated by using the direct method that was used for trabecular thickness. The thickness of pores (expressed in μm) was then estimated as the median of this distribution (41).
Serum insulin-like growth factor I.
At the time of the animal's death, blood was collected by cardiac puncture, allowed to clot on ice, and centrifuged to obtain serum, which was stored at −80°C until analysis. Insulin-like growth factor I (IGF-I) levels were assessed by ELISA using a kit from ALPCO (Windham, NH), as previously described (53).
Percent changes in BMD were calculated from longitudinal dual-energy X-ray absorptiometry (DEXA) measurements obtained on days 0 and 14 (see above) and analyzed via mixed-models ANOVA with unequal variances, with the Satterthwaite approximation to estimate denominator degrees of freedom for the overall ANOVA, the tests of change from baseline within a loading condition, and the post hoc comparisons for differences in group changes from baseline. Percent change in BMD tests within each loading condition and post hoc comparisons among groups were conducted at 2% significance levels to adjust for the multiple comparisons without undue inflation of type II error. Micro-CT-derived bone architectural parameters were analyzed via one- and two-way ANOVAs, as appropriate, with post hoc comparisons conducted using the Student-Newman-Keuls procedure at α = 5%. Marrow culture experiments were analyzed using one-way ANOVAs by age group, with Student-Newman-Keuls at 5% significance for the post hoc comparisons among treatment groups.
Age and HS cause decreases in bone volume and increases in marrow fat accumulation.
Examination of hematoxylin-and-eosin-stained central sagittal sections of the proximal tibia at low magnification (Fig. 1) revealed the expected age-related increase in the adipocytic content of marrow in the proximal tibia (control-young vs. control-old; Table 1) (42). Increased adipocytic content was accompanied by age-dependent decreases in osteoblast number without changes in the number of osteoclasts or osteocytes (Table 1). Interestingly, HS produced a similar increase in BMD in young animals (Fig. 1, Table 1). These cellular findings are consistent with an age-dependent decrease in bone area in control-old and HS-old rats (Table 1), although the HS-induced decrease in young rats did not reach significance. Given the limited sampling of histomorphometric methods, these observations led us to more sensitively and cumulatively quantify bone volume in the proximal tibiae by micro-CT (see ⇓Fig. 3) and to evaluate adipogenic and osteogenic potential using ex vivo bone marrow cultures (see below).
Cortical + trabecular BMD in the proximal tibia of old rats is reduced after 2 wk of HS.
Effects on apparent BMD were determined from serial DEXA measurements in the proximal one-third of the tibia of old and young adult rats at baseline and after 2 wk of treatment. This measurement encompasses cortical and trabecular bone. Table 2 shows the baseline BMD, the BMD after 2 wk of HS or control housing (after treatment), and the percent change in BMD from baseline. As has been previously demonstrated in 6-mo-old rats (7), control-young rats significantly increased BMD (Table 2, P = 0.018), whereas increases in BMD in control-old rats did not reach significance. Interestingly, exposure to HS for 2 wk was sufficient for detection of significant longitudinal decreases in BMD in old rats (P = 0.017), and this decreased BMD was significantly different from the change in BMD in control-old rats (P = 0.013). In contrast, exposure to HS for 2 wk was not sufficient to cause significant decreases in young rats, as we (19) and others (7) have shown in young adult rats exposed to 4 wk of HS. These data demonstrated an age-dependent responsiveness to disuse in the mixed cortical + trabecular bone of the proximal tibia and led us to use micro-CT to characterize the selective compartment-specific effects of age and HS.
HS reduces trabecular bone in young rats and increases cortical porosity in the proximal tibia in aged rats.
The effects of 2 wk of HS and aging on trabecular bone volume and microarchitecture in the proximal tibia were examined using ex vivo micro-CT (Figs. 2 and 3). Representative two-dimensional cross-sectional slices through the proximal tibial metaphysis illustrate the expected age-related decreases in trabecular bone volume (Fig. 2). In addition, quantification of the micro-CT data demonstrates the HS-related decreases in bone volume in young animals (HS-young vs. control-young; Figs. 2 and 3A). In contrast, HS did not further reduce the already low trabecular bone volume observed in old animals (HS-old vs. control-old; Figs. 2 and 3A). However, consistent with the significant loss in total (cortical + trabecular) BMD in HS-old animals (Table 2), HS also appeared to increase cortical porosity in HS-old animals (Fig. 2, arrows), which led us to perform quantitative volumetric measures of cortical porosity by micro-CT. Table 3 demonstrates the increase in porosity with age, which was further increased in old rats subjected to HS. Specifically, in old animals, HS significantly increased percent cortical porosity (P < 0.05) and also increased the number of pores per millimeter (P < 0.05) but did not increase pore diameter compared with control-old animals. However, a significant age-dependent increase in cortical pore diameter was observed when old rats were compared with young rats under the same loading conditions (P < 0.05; Table 3). These data indicate that cortical bone has an age-dependent response to disuse.
HS reduces trabecular bone volume and decreases bone quality in the proximal tibia of young adult, but not aged, rats.
Fractional bone volume (BV/TV) and architectural parameters, including TbN, TbTh, TbSp, ConnD, and SMI were calculated directly from three-dimensional micro-CT reconstructions of the tibial trabecular microstructures (Fig. 3). Analysis revealed that HS significantly decreased trabecular BV/TV, TbN, and ConnD and increased SMI in the proximal tibial metaphysis of young adult animals (all P < 0.05; Fig. 3, A–D). Similar, yet more dramatic, age-related decreases in the proximal tibia were also observed in BV/TV, TbN, and ConnD, as well as in SMI (all P < 0.001, Fig. 3, A, D, and C, respectively). In addition, TbTh and TbSp were increased by aging but were not significantly affected by HS, regardless of age (Fig. 3, E and F). Interestingly, HS had no effect on trabecular bone volume or any architectural parameter in aged animals (Fig. 3).
HS inhibits osteoblast differentiation and aging reduces endogenous BMP stimulation in ex vivo bone marrow stromal cell cultures.
Bone marrow cells from the left femur of rats were cultured in osteoblastic conditions to determine alterations in their capacity to differentiate along the osteoblastic linage. HS significantly decreased the formation of mineralized bone nodules (CFU-OB) in young adult (Fig. 4A) and old rats (Fig. 4B), regardless of in vitro treatment. In addition, the bone marrow from old rats produced fewer CFU-OB than that from young rats. However, all cultures were capable of responding to exogenously added BMP2 (100 ng/ml; R&D Systems, Minneapolis, MN) with an increase in the number of CFU-OB (Fig. 4). The BMP2 responsiveness was observed regardless of age or disuse; however, sensitivity to the osteogenic effects of BMP2 appeared greater in marrow from young than from old rats (Fig. 4).
The addition of the BMP antagonist Noggin (200 ng/ml; R&D Systems) (72) to the cultures significantly decreased the number of CFU-OB in cultures established from control-young and HS-young rats, demonstrating that baseline osteoblastogenesis in control medium was dependent on endogenous BMP production in these young rat cultures (Fig. 4A). In contrast, Noggin did not significantly inhibit osteoblast differentiation in control-old or HS-old cultures (Fig. 4B). Since these cells were capable of responding to exogenous BMP, this indicates that aging impairs the endogenous secretion of BMPs by bone marrow cells.
HS reduces the pool of committed adipocyte precursors, but not the number of potential adipocyte progenitors, in vitro.
Since diminished osteogenic potential of mesenchymal progenitor cells from aging bone marrow has been previously associated with an enhanced adipogenic potential (11, 47), we determined the effects of age and disuse on the formation of adipogenic colonies (CFU-AD) in ex vivo marrow cultures. Cultures from young and old control or HS rats (Fig. 5) were established and stimulated with rosiglitazone, a ligand for peroxisome proliferator-activated receptor (PPAR)-γ2, which stimulates differentiation of committed adipocyte precursors, or an adipogenic cocktail containing insulin, hydrocortisone, and IBMX (IHI), which stimulates recruitment of mesenchymal stem cells into the adipogenic lineage, as described previously (37, 52) and in methods. Oil red O staining was used to determine the extent of adipogenic colony formation. Bone marrow cells from control and HS rats have mesenchymal progenitors that can be recruited toward adipogenesis with the IHI cocktail, although an age-dependent increase in the number of oil red O-positive colonies formed was observed in cells from control-old and HS-old rats compared with control-young and HS-young rats (Fig. 5). Interestingly, HS did not alter the ability of IHI to recruit adipocytes in either age group. However, HS reduced the responsiveness of marrow cells from old and, to a lesser extent, young rats to rosiglitazone-stimulated differentiation of adipogenic colonies in vitro (Fig. 5B).
Serum IGF-I is reduced in old rats and decreases with HS only in young rats.
Age-related decreases in serum IGF-I have been described in humans (for review see Ref. 73) and in rats (4), and the circulating levels of IGF-I have been shown to directly regulate bone growth and density (68). Thus we investigated whether IGF-I levels were also sensitive to HS in young and old rats. IGF-I levels were determined in serum obtained at the time the animals were killed (Fig. 6). HS significantly decreased IGF-I levels compared with controls only in young animals (Fig. 6). As expected, serum IGF-I concentrations were lower in old than in young rats (Fig. 6). Given the well-described and accepted osteogenic role of systemic and local IGF-I (69), these data are entirely consistent with the age-related bone loss observed in cortical and trabecular bone compartments (Figs. 1–3, Tables 1 and 2) and suggest that the HS-induced changes in trabecular bone volume in young animals may also be mediated, at least in part, by changes in serum IGF-I levels.
The experiments in the present study complement previous muscle studies performed using this established model of aging (the Fischer 344/Brown Norway rat) with animals of the same strain, age, and disuse protocol (22). With age, loss of muscle mass was accompanied by increased myofiber nuclear density, which involved fusion of proliferative satellite cells, resembling ongoing regeneration (22). Disuse (by HS) resulted in similar decreases in the soleus muscle of young and old rats; however, the number of myofiber nuclei were decreased by HS in young animals only, indicating a differential sensitivity of aged muscle cells to disuse (22).
In the present study, the skeletal effects of disuse were characterized in the aging rat. This study demonstrated that HS decreased trabecular bone volume and architecture in young adult rats but did not further reduce trabecular bone volume in aged rats. In contrast, the HS-related decrease in trabecular + cortical BMD in the tibia of old rats suggests that HS decreases cortical bone in old rats via a mechanism that involves the induction of new cortical pores and not simply by enlarging the diameter of existing cortical pores. As expected, aging also increased the diameter of cortical pores. These collective findings suggest that, in addition to effects on trabecular bone in young animals, HS also induces a switch in the compartment specificity of disuse osteopenia to include effects on cortical bone. We interpret the data to suggest that a lower threshold limit of trabecular bone volume has been achieved in the aging animals, primarily due to a loss in trabecular number. The increase in trabecular thickness observed in old rats (HS or control) is likely a compensatory mechanism by remaining trabeculae to increase bone formation on existing trabecular templates. Below this lower threshold limit of bone volume, trabecular bone appears to be insensitive to HS. Consistent with this skeletal threshold concept, long-term (3.5 mo) immobilization of 6-mo-old Sprague-Dawley rats has shown that peak trabecular bone loss was achieved 1.5 mo after immobilization and plateaued for the remainder of the 3.5-mo immobilization period (28). As in humans (55), the aged rats used in this study were found to have dramatically less trabecular bone than young adult rats, as has been found previously in intact 27-mo-old Sprague-Dawley rats (66). Thus disuse-induced cortical bone loss in elderly patients with previously deteriorated trabecular architecture (55) may have a critical effect on bone fragility (14, 15), even though such losses in bone mass may not be easily detectable without serial DEXA measurements.
The ex vivo bone marrow cultures indicate that disuse significantly reduces the capacity of marrow stromal cells to be recruited to the osteoblast lineage, a observation that has been reported previously in young growing and young adult rats and mice (24, 34). In addition, adipocyte recruitment with IHI was also increased with age dependence, as expected (31, 42, 61). However, the adipocyte differentiation capacity in response to rosiglitazone was age and HS dependent in a manner that suggested that either aged mesenchymal progenitors are more sensitive to adipocyte recruitment by IHI or endogenous inducers of PPAR-γ-dependent adipogenesis are increased in cultures of aged rats. However, the distinct capacity of IHI and rosiglitazone to stimulate oil red O-positive colony formation in control-old and HS-old rats suggests the possibility of a reduction in the pool of rosiglitazone-sensitive committed adipocyte precursors in bone marrow cells of HS-old rats. Alternatively, or in addition, IHI may also retain the capacity to differentiate earlier noncommitted precursors in cells from HS-old rats that have not yet begun expression of the requisite PPAR-γ2 for rosiglitazone.
Old age also reduced osteoblast number (Table 1), as has been reported for 17- to 27-mo-old Wistar rats (3). Marrow cell populations from the aged Wistar rats showed a reduced capacity for self-renewal in vitro (3). As shown here, such a difference ultimately translates into a reduced number of osteoblasts, which is likely responsible for the decrease in bone formation in aged animals (3). The data presented here suggest that in the Fisher 344/Brown Norway rat model the decrease in osteoblastogenesis is partially due to an age-related decrease in the endogenous production of and/or sensitivity to endogenous BMPs. This study also demonstrated that the osteogenic fate of age-impaired bone marrow-derived stromal cells is further hampered by as little as 2 wk of HS. In addition, the treatment of aged HS marrow cultures with exogenous BMPs enhances CFU-OB formation, but only to about half the level of enhancement in cells from young control or young HS animals.
The significant decrease observed in serum IGF-I is also likely to contribute to the age-related decreases in CFU-OB formation. IGF-I has been shown to synergistically and dose dependently enhance BMP action in stimulating [3H]thymidine incorporation, alkaline phosphatase activity, parathyroid hormone-dependent cAMP level, and bone nodule formation (70). More recently, Osyczka and Leboy (48) demonstrated that IGF-I treatment of bone marrow stromal cells permits BMP-dependent increases in the expression of the early osteoblast-associated genes, alkaline phosphatase, and osteopontin, whereas Celil and Campbell (10) demonstrated that BMP2 and IGF-I utilize different signaling pathways to stimulate osterix expression and osteoblast differentiation. Finally, skeletal unloading leads to resistance to the anabolic actions of IGF-I on bone as a result of failure of IGF-I to activate its own signaling pathways (6, 54). Collectively, these data suggest that decreases in IGF-I levels with age and IGF-I sensitivity with HS contribute to reduced numbers of CFU-OB. We interpret these data to suggest that age-dependent changes in the numbers of osteoprogenitors in ex vivo marrow cultures are likely due to changes in expression of and/or altered responsiveness to BMPs and IGF-I in the bone marrow microenvironment. Such observations may also apply to the responsiveness (or lack thereof) of other osteogenic factors in the bone marrow. Given the synergism of the IGF-I and BMP pathways recently described, it is conceivable the IGF-I-related changes also contribute to the age-associated decreases in endogenous BMP expression and responsiveness to exogenous BMPs, resulting in dramatically reduced CFU-OB formation.
In summary, the results of this study provide critical insight into the mechanism(s) of age- and disuse-associated changes in bone mass and architecture within specific bone compartments that may identify potential new targets for intervention in diseases of age-related bone loss.
Funding was provided by National Institutes of Health Grants AG-20407 and AR-47577.
We thank Charlotte Peterson for helpful comments with preparation of the manuscript and Beata Lecka-Czernik for reviewing the manuscript and providing reagents for ex vivo cultures. We are grateful to Clifford Rosen (Maine Center for Osteoporosis Research) for performing the serum IGF-I assays and to David Findlay (University of Adelaide, Adelaide, Australia) for helpful discussions.
Present address of D. S. Perrien: Biomimetic Therapeutics, 389-A Nichol Mill Ln., Franklin, TN 37067.
Present address of E. Dupont-Versteegden: Div. of Physical Therapy, Dept. of Rehabilitation Sciences, College of Health Sciences, Univ. of Kentucky, Lexington, KY 40536–0200.
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.
- Copyright © 2007 the American Physiological Society