The authors state that they have no conflicts of interest.
Aging Reduces Skeletal Blood Flow, Endothelium-Dependent Vasodilation, and NO Bioavailability in Rats†
Article first published online: 23 APR 2007
Copyright © 2007 ASBMR
Journal of Bone and Mineral Research
Volume 22, Issue 8, pages 1280–1288, August 2007
How to Cite
Prisby, R. D., Ramsey, M. W., Behnke, B. J., Dominguez, J. M., Donato, A. J., Allen, M. R. and Delp, M. D. (2007), Aging Reduces Skeletal Blood Flow, Endothelium-Dependent Vasodilation, and NO Bioavailability in Rats. J Bone Miner Res, 22: 1280–1288. doi: 10.1359/jbmr.070415
- Issue published online: 4 DEC 2009
- Article first published online: 23 APR 2007
- Manuscript Accepted: 17 APR 2007
- Manuscript Revised: 16 MAR 2007
- Manuscript Received: 1 DEC 2006
- trabecular BV/TV;
- marrow blood flow
We determined whether aging diminishes bone blood flow and impairs endothelium-dependent vasodilation. Femoral perfusion was lower in old animals, as well as endothelium-dependent vasodilation and NO bioavailability. These effects could contribute to old age—related bone loss and the increased risk of fracture.
Introduction: Aging has been shown to diminish bone blood flow in rats and humans. The purpose of this study was to determine whether blood flow to regions of the femur perfused primarily through the principal nutrient artery (PNA) are diminished with aging and whether this putative reduction in flow is associated with impaired endothelium-dependent vasodilation.
Materials and Methods: Blood flow was measured in conscious young adult (4–6 mo old) and aged (24–26 mo old) male Fischer-344 rats using radiolabeled microspheres. Endothelium-dependent vasodilation of the PNA was assessed in vitro using acetylcholine (ACh), whereas the contribution of the NO synthase (NOS) and cyclooxygenase (COX) signaling pathways to endothelium-dependent vasodilation was determined using the NOS and COX inhibitors L-NAME and indomethacin, respectively.
Results: Femoral blood flow in the aged rats was 21% and 28% lower in the proximal and distal metaphyses, respectively, and 45% lower in the diaphyseal marrow. Endothelium-dependent vasodilation was reduced with old age (young: 83 ± 6% maximal relaxation; aged: 62 ± 5% maximal relaxation), whereas endothelium-independent vasodilation (sodium nitroprusside) was unaffected by age. The reduction in endothelium-dependent vasodilation was mediated through impairment of the NOS signaling pathway, which resulted in lower NO bioavailability (young: 168 ± 56 nM; aged: 50 ± 7 nM).
Conclusions: These data show that reductions in metaphyseal bone and diaphyseal marrow perfusion with old age are associated with diminished endothelium-dependent vasodilation through an impairment of the NOS mechanism. Such age-related changes in bone perfusion and vascular NO signaling could impact clinical bone loss, increase risk of fracture, and impair fracture healing in the elderly.
Decrements in bone blood flow have been reported in healthy humans >60 yr of age,(1) as well as in senescent rats(2) and rabbits.(3) Such reductions in bone blood flow and capillary rarefaction in bones of aged individuals is associated with osteoporosis.(4) The potential for bone blood flow and the osseous vasculature to have a modulating effect on bone remodeling(5–12) has led to the hypothesis that reductions in skeletal blood flow with advancing age may contribute to a diminution in bone mass and increased risk of fracture in older individuals.(2,13)
Old age—associated reductions in bone blood flow could result from several factors, including enhanced vasoconstrictor responsiveness, vascular remodeling (i.e., arterial rarefaction or the narrowing of the bone resistance artery lumen), or diminished endothelium-dependent and -independent vasodilation of the bone resistance arteries. Impairment of endothelium-dependent vasodilation could be particularly deleterious to the skeleton, because this could diminish NO or prostacyclin (PGI2) release from vascular endothelial cells, both of which are known to play a role in osteoblast and osteoclast activity.(9–17)
Long bones, which exhibit old age—related reductions in blood flow,(1–3) are characteristically perfused through three nutrient arteries: the epiphyseal, metaphyseal, and diaphyseal nutrient arteries.(18,19) The diaphyseal nutrient artery, often referred to as the principal nutrient artery (PNA), perfuses the cortex and marrow of long bones in adult humans and rodents.(18,19) Similar to other organ systems, blood flow rates through bones are determined by resistance arteries, which include nutrient arteries such as the PNA and arterioles that arise from branches of the PNA.(18) Through the control of blood flow, these resistance arteries can consequently have a profound impact on fluid filtration into the bone interstitium and correspondingly the interstitial fluid flow and shear forces acting on bone cells. Given the role of the PNA as a resistance artery,(18,19) the purpose of this study was to determine whether blood flow to regions of the femur are diminished with aging and whether this putative reduction in flow is associated with impairment of endothelium-dependent vasodilation. Because the results showed that endothelium-dependent vasodilation of the PNA is diminished with old age, a secondary purpose was to test the hypothesis that the old age—associated impairment of endothelium-dependent vasodilation occurs through the NO synthase (NOS) or cyclooxygenase (COX) signaling mechanisms.
MATERIALS AND METHODS
The procedures used in this study were approved by the Texas A&M University and West Virginia University School of Medicine Institutional Animal Care and Use Committees and conformed to the Guide for the Care and Use of Laboratory Animals published by the NIH (NIH Publication 85–23, revised 1996). Young adult (4–6 mo old) and aged (24–26 mo old) male Fischer-344 rats were housed in a temperature-controlled (23 ± 2°C) room with a 12:12-h light dark cycle. Water and rat chow were provided ad libitum.
A group of young (n = 12) and aged (n = 12) rats were anesthetized with isoflurane (2%):oxygen balance, and a carotid catheter (Silastic; inner diameter [ID], 0.6 mm, outer diameter [OD], 1.0 mm; Dow Corning) filled with heparinized (Elkins-Sinn; 100 U/ml) saline solution was advanced into the ascending aorta through the right carotid artery as previously described.(2,9,20) This catheter was externalized at the base of the tail and secured on the underside of the tail with 3–0 silk sutures (Davis+Geck;). A second polyurethane catheter (ID, 0.36 mm; OD, 0.84 mm; Braintree Scientific) filled with heparinized saline solution was implanted in the caudal artery (tail) and externalized at the base of the tail as previously described.(2,9,20) This catheter was used to obtain a reference blood sample, which serves as an artificial organ for calculating tissue flows.
After a 24-h recovery period, the carotid and tail catheters were connected to an infusion port and a withdrawal pump, respectively. The animals were allowed to rest quietly for 20 min before the microspheres were infused as previously described.(2,9,20) The rats were anesthetized with sodium pentobarbital (30 mg/kg) infused through the caudal tail artery catheter and killed by decapitation. Femora from both hindlimbs were excised from the carcass and sectioned into three regions: the proximal and distal metaphyses and diaphysis; the diaphyseal marrow was removed from the diaphysis and counted as a fourth region. The corresponding femoral sections from the left and right hindlimbs were combined for each animal to ensure sufficient microspheres in the marrow sample. The procedures for sectioning the femur for blood flow determination have been previously described in detail.(9) The tissue sections were weighed and placed into counting vials for blood flow determination. The mass of the femoral marrow was determined by weighing the diaphysis before and after the marrow was removed.
Blood flow and vascular conductance measurements
Radiolabeled (46Sc and 85Sr) microspheres (Perkin Elmer NEN) with a 15.5 ± 0.2 μm diameter were used for blood flow measurements as previously described.(2,9,20) Specifically, microspheres were suspended in physiological saline with <0.5% Tween 80 and mixed before infusion by 10-min sonication (FS20 Sonicator; Fisher Scientific). A reference blood sample was taken from the caudal artery at a rate of 0.618 ml/min with a Harvard withdrawal pump (model 907; Harvard, Cambridge, MA, USA) while simultaneously, ∼2.5 × 105 microspheres suspended in 0.2 ml saline were infused into the carotid catheter over a 10- to 15-s period. Warm (37°C) saline (0.5 ml) was infused over a 30-s period immediately after microsphere infusion to clear the catheter of residual microspheres; withdrawal of the reference blood sample continued for 20 s after the saline flush. After death and tissue dissection, tissue samples were counted in a γ counter (model 5780; Packard Auto Gamma Spectrometer), and flows were computed from counts per minute and tissue wet weights (blood flows reported in ml/min/100 g). Adequate mixing of the microspheres was verified by showing a <15% difference in blood flows to the right and left kidneys. Mean arterial pressure and heart rate were electronically averaged from pulsatile pressure measurements through a pressure transducer (BP100; ADInstruments). Regional vascular conductances (ml/min/100 g/mmHg) were calculated by dividing tissue flows (ml/min/100 g) by the mean arterial pressure (mmHg).
Arterial pressure was measured from the infusion port connected to the carotid catheter immediately before and after the microsphere infusion and averaged, because simultaneous pressure measurements and microsphere infusion were not possible. Tissue vascular conductance (blood flow normalized to arterial pressure) was determined for each bone segment by dividing the tissue blood flow by the arterial pressure. The measure of vascular conductance therefore reflects local vascular regulation of blood flow by eliminating the influence of potential differences in arterial pressure on bone and marrow perfusion.
In vitro PNA studies
To determine whether the lower femoral bone blood flow with old age was associated with impaired vascular reactivity, in vitro experiments with the femoral PNA were performed. Young (n = 38) and aged (n = 42) male Fischer-344 rats were anesthetized with isoflurane (2%):oxygen balance and killed by excising the heart. The femur and surrounding musculature from both legs were carefully dissected free and placed in cold (4°C) physiological saline buffer solution (PSS) contained in a dissecting dish. Using a stereomicroscope, a segment of the PNA with a length of ∼1 mm was isolated from the femur and the surrounding muscle tissue where the PNA perforates the femur through the femoral foramen. One PNA from each animal was used for in vitro dose-response studies, and the other was either used for in vitro NO bioavailability determination or frozen for mRNA expression as described below. The isolated PNA for in vitro studies was transferred from the dissecting dish to a Lucite chamber containing PSS equilibrated to room temperature as described previously.(21–23) Each end of the PNA was cannulated with a micropipette (60- to 80-μm-diameter tip) and secured with 11–0 nylon microfilament sutures (Alcon). The microvessel chamber was transferred to the stage of an inverted microscope (Olympus IX70) equipped with a video camera (Panasonic BP310), video caliper (Microcirculation Research Institute, Texas A&M University), and data-acquisition system (PowerLab) for the measurement and recording of PNA intraluminal diameter.(21–23) PNAs were pressurized to 60 cmH2O with two hydrostatic pressure reservoirs, and leaks were detected by closing the valves of the reservoirs and verifying that intraluminal diameter remained constant. PNAs that were free from leaks were warmed to 37°C and allowed to develop spontaneous baseline tone (∼1 h). Vessels that leaked or did not develop at least 20% spontaneous tone were discarded. These experimental procedures are similar to those previously used to study skeletal muscle arteriolar vasomotor responses.(21–23)
Evaluation of vasodilator responses
Concentration-response relations to the cumulative addition of acetylcholine (ACh; 10−9–10−4 M) and sodium nitroprusside (SNP; 10−10–10−4 M) were determined in PNAs from young and aged rats. These vasodilators were selected because they produce vasodilation through the endothelium (ACh) or directly through smooth muscle cell relaxation (SNP). After the ACh dose-response was determined, the PNAs were rinsed with warm PSS buffer and allowed to again develop baseline tone before the determination of the SNP dose-response. After completion of the SNP response, maximal PNA diameter was determined by twice washing the PNA with a calcium-free PSS buffer and waiting 15 min after each wash to make the final diameter measure. The calcium-free PSS buffer was similar to PSS-albumin solution except that it contained 2 mM EDTA and CaCl2 was replaced with 2.0 mM NaCl.
Evaluation of endothelium signaling pathways
PNAs from young and aged animals were cannulated and allowed to develop spontaneous tone. Vasodilator response to ACh were evaluated after a 20-min incubation with one of the following: (1) PSS buffer alone, (2) PSS buffer containing the NOS inhibitor NG-nitro-l-arginine methyl ester (L-NAME; 10−5 M), (3) PSS buffer containing the COX inhibitor indomethacin (Indo; 10−5 M), or (4) PSS buffer containing L-NAME (10−5 M) and Indo (10−5 M).(21,24) Separate PNAs from different animals were used for each of these tests.
Determination of eNOS, COX-1, and COX-2 mRNA expression
PNAs were frozen and stored at —80°C in 0.5-ml microcentrifuge tubes. Vessels were pulverized in lysate buffer, and RNA was extracted with the RNAqueous filter system (Ambion). Quantitative real-time PCR was performed with TaqMan probes designed with the use of Primer Express from the published sequence for rat eNOS (primers at exon 8–9 junction: forward, GTGACCCTCACCGATACAACATAC; reverse, TGTCCGGGTGTCTAGATCCAT), COX-1 (primers at exon 2–3 junction: forward, CGGTACTGCTCACAGATGCT; reverse, GGTTCTGGCATGGATAGTAACAACA) or COX-2 (primers at exon 8–9 junction: forward, TGTTGAGTCATTCACCAGACAGATTG; reverse, TGTACAGCGATTGGAACATTCCTT) and a TaqMan oligonucleotide probe (probe: CAGACAGCCACATCCTCAA) labeled with a fluorescent reporter dye and a quencher dye (Applied Biosystems). Real-Time PCR was performed in run with 20 μl using GeneAmp 384 well Optical Reaction plates. Each well contained the following: 2 μl cDNA, 10 μl Universal PCR Master Mix, 1.0 μl 20× Target Primers and Probe, and 7 μl DEPC-treated water. All samples were run in triplicate along with two no-template control samples and two reverse transcriptase negative samples. PCR was initiated by a 10-min step (95°C), 40 two-step cycles of 15 s (95°C), and a 1-min step at 60°C. Levels of the target sequence were quantified relative to the cycle number (cycle threshold) at which the target and co-amplified 18S ribosomal RNA reach a fixed threshold as previously described.(24) Briefly, the fluorescence signal from 18S ribosomal RNA served as controls for the differences in total cDNA loading in the wells. The levels of target sequence were quantified by calculating the difference between Ct for the target sequence and coamplified 18S RNA. One sample with the highest ΔCt value was chosen as a calibrator and assigned a relative quantification (RQ) value of 1.0. All other samples were quantified relative to the calibrator.
Determination of NO bioavailability
To determine NO bioavailability of femoral PNAs from young (n = 9) and old (n = 8) rats, in vitro experiments were performed. Isolated segments of the PNA were placed in a vessel chamber containing PSS equilibrated to room air as described above in the In vitro PNA studies section. One end of the PNA was cannulated with a glass micropipette (60- to 80-μm-diameter tip) and the other end cannulated with an NO sensor probe (30-μm-diameter tip; World Precision Instruments, Sarasota, FL, USA). The micropipette and the probe were secured to the PNA with 11–0 nylon microfilament sutures (Alcon). The sensor probe was connected to a free radical analyzer (APOLLO 4000; World Precision Instruments) interfaced with an IBM PC computer. PNAs were pressurized to 60 cmH2O and checked for leaks. Intraluminal diameter was monitored while the vessels were warmed to 37°C. Subsequent to the development of spontaneous tone (≥1 h), a bolus dose of ACh (10−6 M) was added to the bathing medium, and NO concentration was recorded as the peak change from baseline.
Evaluation of trabecular bone volume
To determine age-related changes in bone volume, distal femora from young (n = 10) and old (n = 10) rats were scanned using μCT (μCT-20; SCANCO USA). Scanning regions were confined to secondary spongiosa and were −1.25 mm in thickness (72 slices). Using 2D images, a region of interest was manually drawn near the endocortical surface, and trabecular bone volume (BV/TV, %) was assessed using 3D image reconstructions.
Solutions and drugs
Stock solutions of drugs were prepared in distilled water and frozen. Fresh dilutions of these stock solutions were prepared daily. All drugs were purchased from Sigma Chemical (St Louis, MO, USA).
Vasodilator responses were expressed as a percentage of maximal relaxation according to the following formula:
where Dm is the maximal inner diameter recorded at 60 cmH2O in calcium-free PSS, Ds is the steady-state inner diameter recorded after each addition of the vasodilator substance, and Db is the initial baseline inner diameter recorded immediately before the first addition of ACh or SNP. To evaluate possible differences in the PNA sensitivity to vasodilators, IC50 values were designated as the concentration of ACh or SNP producing 50% of the maximal vasodilator response.
Independent t-tests were used to determine the significance of differences in body mass, bone and marrow blood flows, trabecular BV/TV, PNA spontaneous tone, maximal diameter, IC50, NO bioavailability, and eNOS mRNA expression between young and old rats. Two-way repeated-measures ANOVAs with pairwise comparisons were used to determine the significance of differences between (young versus old) and within (drug concentration) factors. All data are presented as mean ± SE. Significance was defined as p ≤ 0.05.
Body mass of the old animals from the in vivo and in vitro experiments was significantly greater than that of the young rats (blood flow studies: young, 341 ± 7 g; old, 411 ± 7 g; isolated vessel experiments: young, 349 ± 8 g; old, 427 ± 9 g). Distal femur trabecular bone volume of the old rats was 22% lower compared with that in young animals (young: 17.9 ± 1.4%, old: 13.9 ± 1.4%; p = 0.06).
Bone blood flow and vascular conductance
Blood flow to the proximal and distal metaphyses and the diaphyseal marrow of the femur was significantly lower in the old rats (Fig. 1) and, correspondingly, femoral vascular conductance was lower (Fig. 2). Right and left kidney flows did not differ within groups, indicating sufficient microsphere-blood mixing.
Maximal PNA inner diameters ranged from 100–198 μm in the young rats to 128–223 μm in the aged rats; maximal diameter of PNAs from aged animals (182 ± 6 μm) was greater than that from young rats (160 ± 6 μm). The development of spontaneous tone, however, did not differ between groups (young: 34 ± 4%; old: 41 ± 5%).
Vasodilator responses to ACh and SNP
ACh-induced vasodilation of the PNA was lower in aged rats relative to that in young rats (Fig. 3), whereas the IC50 did not differ between groups (young, 1.45 × 10−7 ± 4.4 × 10−8 M; old, 2.58 × 10−7 ± 4.5 × 10−8 M). Age-related differences in ACh-mediated vasodilation were abolished by NOS inhibition (Fig. 3). COX inhibition significantly lowered the vasodilator response induced by ACh in PNAs from young and old rats (Fig. 4A). However, the old age-related reduction in endothelium-dependent vasodilation was not eliminated by Indo, indicating the difference in endothelium-dependent vasodilation is not associated with the COX signaling pathway. The combination of the NOS and COX inhibitors had little or no further effect on the ACh-mediated vasodilation beyond that with NOS inhibition alone in PNAs from young and old rats, and there were no differences between vasodilator responses of PNAs from young and old animals (Fig. 4B). SNP-induced vasodilation was similar between young and aged rats (Fig. 5) (IC50: young, 1.49 × 10−7 ± 5.1 × 10−8 M; old, 4.33 × 10−7 ± 2.1 × 10−8 M); these results indicate that the impairment of ACh-mediated vasodilation can be attributed to a dysfunction of the vascular endothelial cells rather than an alteration in the responsiveness of the smooth muscle cells.
eNOS, COX-1, and COX-2 mRNA expression and NO bioavailability
Expression of eNOS and COX-1 mRNA in the PNA was not different between young and old animals (Figs. 6A and 6B). However, COX-2 mRNA expression was greater in PNAs from aged rats (Fig. 6C). Intraluminal NO concentration during ACh-mediated (10−6 M) vasodilation was lower in the PNA of old rats (Fig. 7).
Reductions in bone mass and increased fracture risk in elderly individuals have been proposed to be associated with reduced skeletal blood flow.(13) Indeed, old age-related declines in skeletal perfusion have been observed in both humans(1) and rats.(2,25) However, the mechanisms that mediate these decrements in skeletal perfusion remain largely unexplored.(12) The purpose of this study was to determine whether regional blood flows to the femur, which is primarily perfused through the PNA,(18,19) is diminished with old age, and to determine whether these putative reductions in blood flow and vascular conductance are associated with impaired endothelium-dependent or endothelium-independent vasodilator responsiveness of the femoral PNA. The results showed that femoral metaphyseal and marrow blood flow (Fig. 1) and vascular conductance (Fig. 2) are diminished with old age and that these reductions are associated with impaired endothelium-dependent vasodilation of the femoral PNA (Fig. 3). This impairment of endothelium-dependent vasodilation occurs through a dysfunction of the NOS signaling mechanism (Fig. 3) and diminished NO bioavailability (Fig. 7) rather than through an impairment of the COX signaling pathway (Figs. 4A, 6B, and 6C). To our knowledge, these are the first data to show that diminished blood flow and vascular conductance are associated with impaired vascular endothelial cell function and lower NO bioavailability in the skeletal system.
Old age—associated reductions in skeletal blood flow may have adverse affects on bone properties. In healthy long bones, for example, experimental obstruction of nutrient arteries lowers skeletal blood flow, induces osteoporosis, and promotes the development of an alternate periosteal arterial supply for perfusion of cortical bone and marrow, similar to that which occurs with old age.(26) Structural vascular remodeling and attenuated bone and marrow perfusion have been linked to osteoporosis with old age.(2–4,18) In this study, blood flow to the proximal and distal metaphyses and diaphyseal marrow of the femur was reduced by 21%, 28%, and 45%, respectively, in aged rats (Fig. 1). These results are similar to those previously reported by our laboratory where blood flow to the entire femur was 30% lower in old Fischer-344 rats. Furthermore, these reductions in femoral perfusion correspond with lower metaphyseal trabecular bone volume (this study) and impaired femoral bone mechanical properties (i.e., reduced ultimate stress).(2)
Perfusion rates of sites consisting primarily of cancellous bone and marrow were greater than that of regions composed of cortical bone, which is consistent with the more dense vascular network and higher metabolic rates of cancellous bone and marrow.(27,28) It was in these sites of high metabolic (e.g., bone turnover rates) and hematopoietic activity that blood flow was most affected by old age (Fig. 1). Cancellous bone regions consist of large portions of hematopoietic marrow, which declines with aging and is replaced with fatty marrow.(29) Because perfusion of hematopoietic marrow is greater than that of fatty marrow,(29,30) one mechanism for the old age—related decline in cancellous bone and marrow blood flow is a reduction in hematopoietic marrow and the lower metabolic rate associated with the diminished hematopoietic activity.
Another mechanism for the reduced metaphyseal and marrow perfusion with old age is an impairment of endothelium-dependent vasodilation. Results from this study showed that endothelium-dependent vasodilation is ∼20–25% lower in PNAs from old rats (Fig. 3). This may seem to be a rather modest level of impairment. However, as described by Poiseuille's law:
where blood flow (Q) is proportional to the arterio-venous pressure difference (ΔP) and the radius of the vessel to the fourth power (r4) and inversely proportional to the viscosity of the blood (η) and the length of the vessel (l). Therefore, a 20–25% reduction in the radius or diameter of resistance vessels with aging could fully account for the 20–45% reduction in metaphyseal and marrow blood flow (Fig. 1). The findings of endothelial dysfunction in this study are also in agreement with other studies that have reported age-related impairment of endothelium-dependent relaxation in other vascular tissues such as the aorta,(31) mesenteric conduit and resistance arteries,(32,33) and skeletal muscle arterioles.(21,24) Furthermore, old age—associated endothelial dysfunction can occur at multiple levels of the arterial circulation within tissues. For example, endothelial dysfunction in skeletal muscle is present in feed arteries,(34) which are the resistance arteries that lead to the skeletal muscle but are not contained within the muscle, and in downstream resistance arterioles located within the skeletal muscle.(21,24) Thus, age-associated reductions in endothelium-mediated vasodilation are pervasive in the arterial circulation and, like that in the femoral PNA, are likely to occur throughout the osseous arterial circulation. Whereas this study assessed the PNA, one of the three nutrient arteries that perfuse the femur, it is likely that similar vascular changes occur in the epiphyseal and metaphyseal nutrient arteries, as well as vessels downstream to the PNA. Therefore, results from this study provide for the first time a clear link between diminished endothelial function in the bone vasculature and declines in bone perfusion and vascular conductance with advancing age.
Perfusion of the skeleton and skeletal remodeling have been proposed to be linked through several mechanisms.(9) These include (1) bone and marrow perfusion and the consequent rate of fluid filtration into the bone interstitium(9,35) and (2) the coupling of vascular endothelial cells to osteoclasts and osteoblasts activity through release of endothelium-derived substances such as NO and PGI2.(9,10,14,36) In regard to this first mechanism, a transmural pressure gradient exist between the vasculature of the endosteal surface of cortical bone and the periosteal lymphatics, which causes bone interstitial fluid to flow radially through cortical bone.(37) This flow may be greater in species with haversian systems, although flow through the lacunar-canaliculi system in the rat has been proposed to function similarly.(18) Reductions in bone and marrow blood flow and, consequently, capillary pressure, would elicit reductions in capillary and sinusoid filtration, interstitial fluid pressure, interstitial transcortical and trabecular flow, and interstitial shear stress.(9,35) Shear stresses generated by interstitial fluid flow among bone cells are similar to those occurring within the vascular system.(38) Fluid shear forces acting on bone cells have been reported to stimulate release of autocrine/paracrine signals responsible for modulating bone remodeling activity.(37,39–41) For instance, elevated fluid shear forces can stimulate production and release of NO and prostaglandin E2, which can act to promote bone formation and inhibit bone resorption.(41) Therefore, reductions in interstitial fluid shear force could serve to diminish the release of signals promoting bone formation and inhibiting bone resorption. Results from this study support the contention that declines in bone blood flow may reduce bone interstitial fluid flow and shear forces acting on bone cells and may ultimately contribute to the declines in bone mineral density(42) and bone material properties(2) reported previously.
A second possible mechanism linking bone perfusion to bone remodeling is through the coupling of the osseous vasculature to bone cell activity.(9,10,14,36) Blood vessels are located within basic multicellular units that contain osteoclasts and osteoblasts that carry out bone remodeling.(36) These vessels, and in particular the vascular endothelial cells, can release substances such as NO and PGI2 in response to blood flow and intravascular shear stress.(21,43,44) Both NO and PGI2 have the potential to influence bone cell remodeling activity in skeletal tissue resulting in greater bone formation and decreased bone resorption.(12,17,38,39,45–50) This study showed a reduction in endothelium-dependent vasodilation in the bone vasculature of aged rats; this decline in endothelium-dependent vasodilation was mediated through impairment of the NOS signaling pathway (Fig. 3) and results in a reduction in NO bioavailability (Fig. 7). Reduced bioavailability of NO from vascular endothelial cells not only serves to impair blood flow through reduced endothelium-mediated vasodilation,(21,43,44) but also could serve to provide diminished osteoclast inhibition and less stimulation for osteoblast proliferation, resulting in decreased bone formation and bone loss.(9,17,38,39,45–50) Further studies are necessary to definitively establish this coupling relation between the vascular endothelium and bone remodeling activity with aging.
One limitation of studying the rat for age-associated changes in skeletal blood flow is the absence of haversian systems. In humans, as well as other large mammals (e.g., sheep, dogs, rabbits), cortical bone undergoes intracortical remodeling, with each haversian system containing a blood vessel. This forms a vast vascular network within the cortex that could play a role in modulating bone remodeling. The absence of a haversian vascular system in the rat could also alter old age—associated changes in skeletal blood flow relative to that occurring in human bone. For example, in this study, there were no changes in blood flow or vascular conductance in the cortical bone of the femoral diaphysis (Figs. 1 and 2). The presence of haversian systems in cortical bone may have resulted in a compromised perfusion with old age. Despite this difference between rats and humans, the PNA in the rat and human is the main conduit for blood flowing into the bone and marrow, suggesting that the effects noted in this study remain applicable to humans. In fact, the higher level of vasculature in human cortical bone suggests any disruption to flow through alterations in the responsiveness of the PNA and other resistance arterioles may have greater effects on old age—related cortical bone loss than that which occurs in the rat.
In conclusion, this study showed that aging diminishes long bone metaphyseal and marrow blood flow and vascular conductance. One mechanism for the decrease in vascular conductance is impairment of endothelium-dependent vasodilation. Results showed that endothelial dysfunction is present in femoral resistance arteries of aged rats and that the impairment of endothelium-dependent vasodilation occurs through the NOS signaling pathway and diminished NO bioavailability, rather than through a COX mechanism. Such changes in bone blood flow and vascular signaling have been proposed to diminish interstitial fluid flow and reduce critical downstream signaling molecules such as NO within bone, which in turn could have implications with respect to age-associated reductions in bone mass, increased fracture risk, and diminished fracture healing.
This study was supported by National Aeronautics and Space Administration Grants NAG2–1340 and NCC2–1166.
- 71930 The resorption of bone. Arch Surg 20: 355–385.
- 151991 The Endothelium: Modulator of Cardiovascular Function. CRC Press, Boca Raton, FL, USA.,
- 181998 Blood Supply of Bone: Scientific Aspects. Springer-Verlag, New York, NY, USA.,
- 191983 Pathways of transport in bone. In: SheppardJT, AbboudFM (eds.) The Handbook of Physiology: The Cardiovascular System III. Williams & Wilkins, Baltimore, MD, USA, pp. 371–396.
- 272001 Integrated bone tissue physiology: Anatomy and physiology. CowinS (ed.) Bone Mechanics Handbook. CRC Press, Boca Raton, FL, USA, 1-1–1-34.
- 281989 The skeletal tissues. WeissL (ed.) Cell and Tissue Biology: A Textbook of Histology. Urban and Schwarzenberg, Baltimore, MD, USA, 211–259.