The authors state that they have no conflicts of interest.
IGF-I plays an important anabolic role in stimulating bone formation and maintaining bone mass. We show that the pro-proliferative, anti-apoptotic, and functional responses to IGF-I in bone and BMSCs decrease with aging. These changes are associated with impaired receptor activation and signal transduction through the MAPK and PI3K pathways.
Introduction: IGF-I is a potent anabolic agent having effects across diverse tissues and cell types. With aging, bone becomes resistant to the anabolic actions of IGF-I. To examine the effects of aging on bone responsiveness to IGF-I, we measured the pro-proliferative, anti-apoptotic, and functional responses of bone and bone marrow stromal cells (BMSCs) to IGF-I and evaluated IGF-I signal transduction in young, adult, and old mice.
Materials and Methods: Male C57BL/6 mice 6 wk (young), 6 mo (adult), and 24 mo (old) were treated with IGF-I for 2 wk using osmotic minipumps, and osteoblast proliferation (BrdU labeling) in vivo, and osteoprogenitor number (BMSC culture and calcium nodule formation) were measured. Proliferation, apoptosis, and expression of key osteoblast factors (alkaline phosphatase, collagen, osteocalcin, RANKL, osteoprotegerin (OPG), macrophage-colony stimulating factor [M-CSF]) and IGF-I signaling elements and their activation in IGF-I—treated cells were studied using QRT-PCR and Western blot analysis. Data were analyzed using ANOVA.
Results: Aging decreased the basal and IGF-I—stimulated number of BrdU-labeled osteoblasts and reduced the ability of IGF-I to stimulate osteoprogenitor formation (calcium nodule number) by 50%. The pro-proliferative and anti-apoptotic actions of IGF-I were blunted in cells from old animals. These changes were accompanied by age-related alterations in the ability of IGF-I to regulate alkaline phosphatase, collagen, osteocalcin, RANKL, OPG, and M-CSF expression. IGF-I binding was normal, but IGF-I receptor mRNA and protein expression was increased in aged animals by 2- and 10-fold, respectively. The age-related changes in proliferation, apoptosis, and function were accompanied by loss of IGF-I—induced signaling at the receptor level and at key regulatory sites along the MAPK (ERK1/2) and PI3K (AKT) pathways.
Conclusions: Our data show that aging is accompanied by loss of bone and BMSC/osteoblast responsiveness to IGF-I and that these changes are associated with resistance to IGF-I signaling that involve receptor activation and downstream signaling events.
IGF-I is a potent anabolic agent having effects across diverse tissues and cell types.(1,2) In bone, IGF-I increases osteoprogenitor number, stimulates proliferation, reduces bone marrow stromal cell (BMSC) apoptosis, and encourages recruitment and migration of osteoblasts to the bone surface.(3,4) IGF-I also augments recruitment of osteoclasts to the bone surface through increased osteoblast expression of RANKL.(5–7) Overall, IGF-I administration increases bone formation. Ligand binding to the IGF-I receptor (IGF-IR) stimulates receptor tyrosine phosphorylation and initiation of the IGF-I signaling cascade.(1,2) Receptor activation triggers phosphorylation of the extracellular signal-related kinase 1/2 (ERK1/2) and cyclin-dependent kinase (AKT), respectively. These combine to promote cell proliferation and survival.
Global loss of IGF-I results in short bones and low BMD.(8) In mice with osteoblast loss-of-function mutations in the IGF-IR, weight is normal, but cancellous bone volume, trabecular thickness, and connectivity are dramatically reduced.(9) Bone-targeted overexpression of IGF-I increases bone formation and BMD.(10) Collectively, these observations show the importance of preserving normal IGF-I function for the maintenance of bone mass and structure.
With aging, tissue responsiveness to IGF-1 in general is altered.(11–15) Aging is associated with decreases in IGF-IR and IGF-IR phosphorylation in muscle,(12) senescence of heart myocytes is accompanied by a reduction in the phosphorylated form of AKT,(13) and in the liver, insulin receptor substrate (IRS) protein levels decrease, and some, but not all investigators, found reduced levels of IGF-1R.(14) Fibroblast DNA synthesis and proliferation in response to IGF-1 decrease with age.(15)
Bone responsiveness to IGF-I also decreases with age.(6,16–20) The skeletal anabolic response to IGF-1 administration is weakened in the elderly, and osteoblasts grown from biopsy specimens from patients of different ages respond to IGF-I by increasing proliferation, but the dose needed to elicit a response is an order of magnitude greater for cells from aged patients.(17) Tanaka et al.(19) reported similar changes in dose-responsiveness in the rat but found no alterations in IGF-IR expression or binding affinity.
To further examine the effects of aging on bone cell responsiveness to IGF-I, we measured the proliferative, apoptotic, and metabolic responses of bone and BMSCs to IGF-I, and evaluated IGF-I—mediated signal transduction in young, adult, and old mice. Our results suggest that aging compromises bone cell responsiveness to IGF-I by impairing IGF-IR activation.
MATERIALS AND METHODS
Chemicals and reagents
Ketamine, xylazine HCl, and acepromazine were purchased from Abbott Laboratories (North Chicago, IL, USA), Boehringer Ingelheim (St Joseph, MO, USA), and Animal Health Co. (Kansas City, MO, USA), respectively. αMEM was from Mediatech (Herndon, VA, USA), FBS was from Atlanta Biologicals (Nocross, GA, USA), and the TRACP staining kit was from Sigma (St Louis, MO, USA). RNA-STAT 60 was purchased from Tel-Test (Friendwood, TX, USA). The oligonucleotide primers and TaqMan probes for PCR amplification were designed by the Primer Express software (Version 1.0) from Applied Biosystems and synthesized by Integrated DNA Technologies (IDT, Coralville, IA, USA) with high-performance liquid chromatography (HPLC) purification.
Male C57BL/6 mice, 6 wk (young), 6 mo (adult), and 24 mo (old) of age, were obtained from the National Institute of Aging (NIA) colony of aging rodents (Harlan Sprague-Dawley, Bethesda, MD, USA) and provided a standard rodent diet (8640 Harlan Teklad 22/5 [W], Madison, WI, USA) containing 1.13% calcium and 0.94% phosphorus. Animals were allowed to acclimate in our animal facility for at least 3 days before experimentation. All animals were housed in air-filtered, humidity- and temperature-controlled rooms with equal 12-h light—12-h dark cycles. The animal protocol for these studies was approved by the Animal Care and Use Committee at the Veterans Affairs Medical Center, San Francisco, CA, USA. Animals were maintained and processed in accordance with the NIH Guide for the Care and Use of Laboratory Animals.
In vivo studies
Young, adult, and old animals (n = 6) were infused with human recombinant IGF-I (2.5 mg/kg/d in 10 mM sodium succinate and 140 mM sodium chloride; generously provided by Eli Lilly and Company, Indianapolis, IN, USA) or vehicle for 2 wk using osmotic mini-pumps (Alzet, Palo Alto, CA, USA) implanted subcutaneously on the back.
Total serum IGF-I was measured by radioimmunoassay.(21) The method has been validated repeatedly for mouse serum, and in particular, for multiple samples from inbred strains and congenic mice of varying ages from 1 to 24 mo.(22) In brief, serum IGF-I binding proteins were separated from IGF-I by acid dissociation. This is followed by addition of a neutralization buffer containing excess recombinant human IGF-II. This binds residual IGFBPs before the immunoassay with human anti-IGF-I, a polyclonal antibody from ALPCO, Windham NH, USA. The sensitivity of the assay is 0.01 ng/ml, and the interassay CV for C57B/6 mouse serum is 6%. There is no cross-reactivity with IGF-II.
Five days before death of the animals, a second osmotic mini-pump containing BrdU (0.7 mg/g/d in 50% DMSO and 50% propylene glycol) was implanted on the back to assess cell proliferation as judged by incorporation of BrdU. At death, distal femurs were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer for 24 h at 4°C overnight. Bones were demineralized in 10% EDTA (pH 7.3) and embedded in paraffin, and sections were mounted on positively charged slides (Fischer Scientific, Pittsburgh, PA, USA). BrdU was detected using a commercially available kit (Zymed Laboratories, South San Francisco, CA, USA), and sections were counterstained with hematoxylin. Sections from animals not infused with BrdU served as controls. Osteoblasts (cuboidal cells lining the bone surface) with dark brown nuclear staining were counted as BrdU positive. The number of BrdU+ cells was quantified using Bioquant software and expressed per unit bone perimeter (BrdU+ cells/mm).
Young, adult, and old mice were killed using an overdose of a ketamine cocktail (10% ketamine [vol/vol]; 2% xylazine HCl [vol/vol]; 1% acepromazine [vol/vol]). Cells were harvested from both tibias and femurs and cultured as previously described.(23) To generate cultures containing osteoblast-like cells, marrow cells were plated at 10 × 106 nucleated cells/10-cm tissue culture dish. On day 5, nonadherent cells were removed by aspiration, and the culture medium was changed to secondary medium (αMEM supplemented with 10% FBS, 1% antibiotics, 0.1% Fungizone, 50 μg/100 ml l-ascorbic acid, and 3 mM β-glycerophosphate) to induce mesenchymal cells to form osteoblasts. Subsequent media changes were performed every 2 days for up to 28 days. Sequential expression of alkaline phosphatase, collagen, and osteocalcin, and mineralized nodule formation confirmed the osteoblast nature of our cells (data not shown).(23,24) Mineralized calcium nodule formation was assessed at day 28. Plates (n = 3/animal) were stained with 2% alizarin red (Sigma) for 10 min, and the plates were rinsed five times with distilled water to remove loosely bound stain. The concentration of IGF-I in the conditioned culture media was measured at day 10 as described above.
In vitro cell proliferation and apoptosis assays
BMSCs were cultured as described above for 8 days, harvested, and replated in 96-well plates at 1 × 104 cells/well. Eight days after plating, the cells were switched to medium containing 1% FBS. After 24 h, the cells were washed and cultured in serum-free medium with 0–10 ng/ml IGF-I and containing 10 μM BrdU for 24 h. Cells were fixed and washed, and BrdU incorporation was quantified using a commercially available ELISA assay kit (Roche Diagnostics, Mannheim, Germany). BrdU incorporation into cells was normalized to cell number by measuring the absorbance of crystal violet stain in an additional parallel culture plate that was treated in the same manner. Apoptosis was measured using a commercially available assay kit (Roche Diagnostics) and normalized to cell number as described above.
Measurement of mRNA levels
Cells were harvested at day 10 of culture and processed for mRNA as previously described.(23,25) Briefly, cells were rinsed and extracted using RNA-STAT 60. Concentration and purity of the RNA were determined by measuring absorbance at 260 and 280 nm. The levels of mRNA for alkaline phosphatase, α1 collagen, osteocalcin, IGF-IR, RANKL, osteoprotegerin (OPG), and macrophage-colony stimulating factor (M-CSF) were determined by quantitative real-time RT-PCR (Q-PCR). Primers and probes for alkaline phosphatase, collagen, osteocalcin, RANKL, OPG, and M-CSF have been described.(23) The forward and reverse primers for IGF-IR are as follows: F, 5′-CACTCAGGACACAAGGCTGA-3′; R, 5′-GGCACACACGTTACTGTTGG −3′. The sequence of Taqman probe was 5′-CCTCGACCTGATCCTCGGACACA-3′. To confirm the specificity of our primers and to insure that samples were free of DNA contamination, RT-PCR products were run on an agarose gel (2%). Only one distinct band of the predicted size was observed for each gene (data not shown). Amplicons were sequenced and found homologous with expected products.
After RT, the cDNA (2 μl) was amplified and quantified using a Sequence Detection System (SDS 7700) and a PCR universal protocol as follows: AmpliTaq Gold activation at 95°C for 15 s and annealing/extension at 60°C for 1 min. The fluorescence of the double-stranded products accumulated was monitored in real time. The relative mRNA levels were normalized to levels of GAPDH mRNA in the same sample.
Binding and internalization studies
Binding: Cells (8 days after plating) were cultured overnight in serum-free medium, washed twice with PBS, and incubated with 125I-IGF-I (Amersham Biosciences, sp. Act. = 2000 Ci/mmol) along with increasing amounts of unlabeled IGF-I (0–100 nM; Cell Sciences, Canton, MA, USA) for 2 h at 4°C. Preliminary time-course studies showed that binding was at equilibrium after 2 h at 4°C, and under these conditions, receptor internalization was negligible. Plates were rinsed three times with ice-cold PBS, and the cells were solubilized in 0.5 ml of 0.5 M NaOH for 10 min. Radioactive IGF-I was quantified using a γ counter. Binding, normalized to protein concentration, was corrected for nonspecific binding, and the data were submitted to Scatchard analysis. All determinations were performed in triplicate, and the binding studies were performed twice with similar results.
Internalization: Cells were cultured overnight in serum-free medium, washed twice with binding buffer (PBS containing 0.1% BSA, pH 7.0), and incubated with 1 nM 125I-IGF-I for 0, 5, 10, 30, 60, and 120 min at 37°C. Cells were washed with ice-cold binding buffer and incubated in 1 ml of 0.2 M acetic acid and 0.5 M NaCl, pH 2.5, for 5 minutes at 4°C. The acid strip buffer was collected as surface bound 125I-IGF-I. Cells were solubilized in 1 M NaOH, and the radioactivity of the lysates was counted as internalized 125I-IGF-I. All data were normalized to total protein.
Signal induction by IGF-I and Western blot analysis
Cells (day 8 of culture) were washed twice with PBS, cultured overnight in serum-free medium, and treated with 10 ng/ml IGF-I for 2 h (IGF-I protein induction in response to IGF-I) or 10 min (signal induction). Cells were washed with PBS (4°C) and solubilized in lysis buffer (RIPA) containing 1% Triton X-100, 0.1% SDS, 1% sodium deoxycholate, 1 mM EDTA, 50 mM HEPES, 0.1% NaF, 0.1% sodium vanadate (Na3VO4), 150 mM NaCl, 100 μg/ml PMSF, and a protease inhibitor cocktail (Complete Mini; Roche). The cell extracts were centrifuged at 10,000g for 10 min, and the supernatant was removed. Protein in the lysate supernatant was measured using the BCA protein assay kit (Pierce, Rockford, IL, USA). Lysate equivalent to 50 μg of protein was solubilized in sample buffer (2% SDS, 10 mM DTT, 10% glycerol, 100 mM Tris-HCl, 0.01% bromphenol blue) and analyzed by SDS-PAGE (7.5% or 4–15% gel). Separated peptides were transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA, USA). After blocking nonspecific binding (5% Blotto, 0.05% Tween 20 in 100 mM Tris-HCl, 0.9% NaCl [TBS], room temperature for 1 h), the membranes were incubated with rabbit antibodies to the IGF-IR β chain (C-20; Santa Cruz Biotechnology) and phospho-IGF-IR (Cell Signaling Technology), ERK 1/2 and phospho-ERK 1/2 (Cell Signaling Technology), and AKT and phospho-AKT (Cell Signaling Technology). Equal protein loading and transfer were verified by staining membranes with Ponceau (Sigma, St Louis, MO, USA) solution. Blots were developed using a chemiluminescent substrate (SuperSignal West Dura; Pierce). Quantification of all blots was performed using Kodak ID Image Analysis software (Eastman Kodak, Rochester, NY, USA).
All experiments were performed at least twice with similar results. Data are reported as mean ± SD and analyzed using Student's t-test or ANOVA and the Student-Newman-Keuls or Holm-Sidak methods (SigmaStat; SPSS, Chicago, IL, USA).
The concentrations of IGF-I in the serum and conditioned culture media (day 10) from young, adult, and old animals are shown in Table 1. Serum IGF-I changed little with age, whereas the concentration in the media increased by 43%.
Table Table 1.. Effects of Age on the Concentrations of IGF-I in Serum and Conditioned Media From Cells at Day 10 of Culture
Values are mean ± SD.
Serum: n = 8–9; media: n = 19.
* p < 0.02 compared with young (one-way ANOVA).
With aging, the number of BrdU-labeled osteoblasts on the bone surface decreased by ∼65% (Fig. 1). Administration of IGF-I in vivo for 2 wk increased the number of labeled osteoblasts in adult (3-fold) and old (1.9-fold), but the response to IGF-I was weaker in the aged animals. The number of labeled osteoblasts after administration of IGF-I in the aged animals reached only the basal level of that in the adults. Remarkably, young animals showed no change in proliferative activity in response to IGF-I.
To determine whether aging affects the ability of IGF-I to increase osteoprogenitor number, we measured calcium nodule formation after 2-wk infusion of IGF-I. IGF administration in the adult animals increased the number of osteoprogenitors, as judged by calcium nodule formation, from 15 ± 10 to 169 ± 41 (Fig. 2). The basal level of nodule formation in the old was similar to that in the adult, but the response to IGF-I was reduced to one half. The serum concentrations of IGF-I in vehicle and IGF-I—treated animals in these experiments were 224 ± 46 and 414 ± 86 ng/ml, respectively. To determine the effect of age on the ability of IGF-I to stimulate cell proliferation and prevent cell death in vitro, we treated BMSCs/osteoblasts with increasing levels of IGF-I. Proliferation increased with increasing dose of IGF-I in animals of all ages, but the response in cells from the adult and old donors was blunted (Fig. 3A). Proliferation in cells from young animals more than doubled, but it increased by only 80% and 50% in cells from adult and aged animals, respectively. Although, the basal level of proliferation tended to be higher in the young than the old, the difference did not reach significance.
Apoptosis decreased with increasing dose of IGF-I in animals of all ages and, although the relative change was similar (−30%), the absolute magnitude of change was greater in cells from young donors (Fig. 3B). The response of cells from adults was intermediate between the young and aged responses. Basal levels of apoptosis tended to increase with increasing age.
As a functional assessment of BMSC responsiveness to IGF-I, we measured basal and IGF-I—induced expression of alkaline phosphatase, collagen, osteocalcin, RANKL, OPG, and M-CSF. IGF-I treatment of cells from young animals increased alkaline phosphatase, collagen, and osteocalcin expression in young but not old animals (Table 2). There was a progressive loss of responsiveness with advancing age. A similar pattern was observed with RANKL and M-CSF (Table 3). Treatment with IGF-I induced a modest decrease in OPG expression in cells from the young and adult (−10% to 20%), but a substantial decrease in cells from the old (−35%). To evaluate the combined effects of IGF-I—induced changes in RANKL and OPG on pro-osteoclast activity, we calculated the change in the ratios of RANKL/OPG after treatment with IGF-I. The ratios of RANKL to OPG (normalized to the young) in vehicle-treated cells from young, adult, and old animals were 1, 1.7, and 2.7, respectively. IGF-I treatment increased these ratios to 2.0, 3.4, and 4.4, respectively. M-CSF expression increased ∼2-fold after treatment with IGF-I in cells from young donors but was unaffected in cells from adult and old donors (Table 3).
Table Table 2.. Effects of IGF-I Treatment In Vitro (10 ng/ml, 2 h) on Relative mRNA Expression of Alkaline Phosphatase, αI Collagen, and Osteocalcin in BMSCs From Young, Adult, and Old Animals at Day 10 of Culture
Values are mean ± SD, n = 3.
* p < 0.05 compared with age-matched vehicle (Student's t-test).
Table Table 3.. Effects of IGF-I Treatment In Vitro (10 ng/ml, 2 h) on Relative mRNA Expression of RANKL, OPG, and M-CSF in BMSCs From Young, Adult, and Old Animals at Day 10 of Culture
Values are mean ± SD, n = 5.
* p < 0.05 compared with age-matched vehicle (Student's t-test).
To determine whether the blunted in vitro response to IGF-I in aged animals is associated with changes in IGF-IR binding, receptor internalization, or both, we examined binding and internalization of IGF-I in BMSC/osteoblast cells from young, adult, and old animals. Displacement of 125I-IGF-I by unlabeled IGF-I for each of the ages is shown in Fig. 4A. From Scatchard analysis, the numbers of binding sites in cells from young, adult, and old animals were 115, 130, and 90 nmol/mg protein, respectively, and did not differ significantly across age. Binding affinities were also similar across age (Kd = 0.8–1.1 nM). Receptor internalization, normalized to the young maximum, decreased progressively with age, but the kinetics of internalization did not differ across age (Fig. 4B).
With aging, expression of mRNA and protein for the IGF-IR increased. Message levels nearly doubled between 6 wk and 24 mo of age (Fig. 5), and protein levels increased roughly 10-fold (Fig. 6). IGF-I treatment had no effect on either mRNA or protein.
To determine whether loss of BMSC/osteoblast responsiveness to IGF-I is associated with alterations in IGF-IR signaling, we measured IGF-IR activation and signal transduction through the MAPK (ERK1/2) and PI3K (AKT) pathways. Treatment of cells for 10 min with IGF-I at a concentration of 10 ng/ml, a concentration shown in preliminary studies to maximally activate the receptor and downstream signaling events, induced phosphorylation of the receptor in cells from all ages, but the level of activation was attenuated in the old (Fig. 6). Basal levels of activation and responsiveness to IGF-I decreased progressively with advancing donor cell age. The levels of total and basally activated ERK1/2 were not different among the different aged animals, but the level of activation induced by IGF-I was ∼40% lower in cells from old animals. A similar pattern was observed for AKT. Total AKT expression was unaffected by age and IGF-I treatment. However, basal phosphorylation drifted down with increasing age. Activation by IGF-I increased phosphorylated AKT by nearly 8-fold in cells from young and adult donors but by only 4-fold in cells from old donors. Comparing only the activated states, the level of phosphorylated AKT in cells from old donors was roughly 20% of that in cells from young and adult donors.
Bone and BMSCs/osteoblasts from mice of all ages responded to IGF-I, but the pro-proliferative, anti-apoptotic, and functional responses were blunted in older animals. These results are consistent with earlier studies in rats and humans that showed bone and bone cells become resistant to the anabolic actions of IGF-I with increasing age. That other tissues share loss of responsiveness to IGF-I with age suggests that common mechanisms may be involved. Reports of muted IGF-IR and AKT activation in muscle after IGF-I treatment are examples.(13,26) Interestingly, aging in the mouse has little effect on the serum concentration of IGF-I, suggesting that the role of IGF-I in age-related bone loss is not mediated through changes in the circulating concentration.
The previously described age-related decrease in osteoblast surface in vivo is consistent with our findings that the number of BrdU-labeled osteoblasts both basally and in response to IGF-I is lower in aged animals. The diminishing number of total osteoblasts associated with aging has been attributed to a loss of osteoprogenitors. Although we did not find fewer osteoprogenitors in the old animals, we did observe a significant blunting of the response to IGF-I (Fig. 2). These data suggest that the in vivo loss of bone responsiveness to IGF-I observed with aging may reflect, in part, a resistance of the osteoprogenitor population to IGF-I.
BMSCs in culture (day 8) also showed signs of resistance to IGF-I. Both the pro-proliferative and anti-apoptotic responses to IGF-I treatment in vitro became impaired with increasing age (Fig. 3). Although the basal level of BrdU uptake was similar across age, the basal number of apoptotic cells was much greater (+50%) in cultures from old animals. All three ages responded to IGF-I to a similar degree (−30%), but the higher background in cells from old donors was sustained. It seems that the propensity of the BMSC to undergo apoptosis increases with age, but the cells remain responsive to IGF-I.
To assess the effect of age on the functional response of BMSCs/osteoblasts to IGF-I, we measured expression of alkaline phosphatase, collagen, osteocalcin, RANKL, OPG, and M-CSF. These were chosen because of their importance as markers of osteoblast formation and activity (alkaline phosphatase, collagen, and osteocalcin) and their role in regulating the balance between bone formation and resorption (RANKL, OPG, and M-CSF). The increase in alkaline phosphatase, collagen, and osteocalcin expression in response to IGF-I in cells from young animals is consistent with the well-established anabolic effects of IGF-I on bone in vivo. Loss of this response in cells from the older animals is consistent with the diminishing effectiveness of IGF-I in eliciting an anabolic response in the elderly.
IGF-I is necessary for maintenance of normal RANKL and M-CSF expression and normal bone turnover.(27) Our results indicate that IGF-I can stimulate expression of these cytokines in young but not old animals. Thus, the normal regulatory role that IGF-I plays in bone turnover is lost with aging. Loss of this responsiveness is consistent with the generalized age-related blunting of other cellular functions to IGF-I.
To determine the mechanisms responsible for the age-related loss of sensitivity to IGF-I, we examined IGF-I binding and signal transduction. We observed no significant changes in binding site number or affinity, suggesting that the resistance to IGF-I in the aged animal is not likely a consequence of changes in cell surface receptor number or affinity. These results are consistent with those of Tanaka et al.,(19) who reported that the number of high affinity binding sites and the affinity constant do not change with age in the rat. That there is no effect of age on binding is perplexing in light of the substantial age-related increase in IGF-IR mRNA and protein expression. The rise in expression in mice contrasts with findings in the rat, where expression has been reported to be unaffected by aging.
Along with increased IGF-IR expression, we found impaired receptor activation. Aging diminishes the ability of the cell to maintain receptor phosphorylation in the presence of ligand. Aberrant IGF-IR activation is accompanied by faulty downstream signaling through both the MAPK and PI3K pathways, as revealed by diminished phosphorylation of ERK1/2 and AKT in response to IGF-I treatment. The changes in ERK1/2 and AKT activation likely reflect the changes in receptor function but may also be independently modulated by other senescence-associated changes. This is supported by the observation that the loss of signal transmission seems to be greater through AKT (80% reduction in P-AKT) than ERK (40% reduction in P-ERK).
The reasons for the deficiency in receptor activation and signaling are unclear. One possibility is that senescence changes in receptor kinase activity, phospho-tyrosine phosphatase activity, or both may shift the kinetic balance between the phosphorylated and unphosphorylated states of the receptor. This could result in a lower level of the phosphorylated state of the receptor after ligand binding. It is also possible that the IGFBPs may play a role. It has been reported that BP-3 expression increases with age in human fibroblasts,(28) and BP-3 is known to inhibit IGF-IR phosphorylation.(29)
The consequences of loss of normal signaling may include the observed rise in receptor expression. Loss of signaling may trigger a feedback response to increase receptor number and restore basal signaling tone. Interestingly, the increase in receptor expression is not manifest in an increase in surface receptor number. This suggests that receptor processing may be impaired with aging.
Our studies were designed to examine the transitional phase between skeletal growth (6 wk) and adulthood (6 mo) and between young adulthood and old age (postmaturational aging). In general, changes in responsiveness to IGF-I occur during both transitions from growth to adulthood and from adulthood to old age. In most, but not all, cases, the responsiveness of the adult falls midway between the young and the old. RANKL expression, for example, is increased by IGF-I in the young and adult but not in the old, suggesting that modulation of RANKL expression by IGF-I is lost during postmaturational aging. M-CSF expression, on the contrary, is increased by IGF-I in the young but not in the adult or old, suggesting that modulation of M-CSF expression by IGF-I is lost during the transition between the young growing animal and the adult.
Interpretation of our findings is confounded by several aspects of our model systems. The age-related changes in bone and BMSCs/osteoblasts may reflect changes in the bone cell environment (systemically and locally), changes in the composition of the BMSC population related to aging, or senescent changes inherent to the cells themselves. In the in vivo studies, the ability of IGF-I to increase osteoprogenitor number is reduced with aging. This could represent changes in the local cell environment that impair IGF-I action (e.g., local concentration of the IGF-I binding proteins), or it could represent senescent changes in the mesenchymal stem cells that impair their responsiveness to IGF-I. It is well established that the BMSC population is heterogeneous. Age-related changes in the composition of this population may contribute to the observed changes in proliferation and function induced by IGF-I in vitro. That cells senescence and the BMSC environment changes with age suggests the likelihood that numerous factors including cell senescence, changes in the composition of the marrow cells, and systemic changes all contribute to the loss of cell responsiveness to IGF-I.
In summary, the effects of aging on bone and bone responsiveness to IGF-I are complex. Our data show that aging is accompanied by alterations in IGF-I signaling that involve receptor activation. Faulty receptor activation is likely to contribute to, but may not completely account for, impaired downstream signaling. The effect of these changes is to reduce responsiveness to IGF-I. The loss of bone that accompanies aging is poorly understood, but the observations that serum IGF-I (in humans) and bone cell responsiveness to IGF-I decrease with advancing age suggests that IGF-I plays an important role in age-related osteopenia.
This work was supported by the Veterans Affairs Merit Review Program and NASA Grant NNAO4CK55G.