INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PHYSIOLOGICAL ADAPTATIONS that occur with weightlessness include cardiovascular deconditioning and musculoskeletal atrophy. The osteopenic effect of spaceflight first manifested itself when increased urinary calcium excretion was observed in cosmonauts after the Vostok 2 and 3 missions (17). Significant loss of calcaneus mass was detected in astronauts after Gemini, Apollo, and Skylab flights (17). Evaluation of cosmonauts occupying the Mir space station has provided the opportunity to study the effects of long-term exposure to microgravity. Prolonged spaceflight resulted in a marked loss of cancellous and cortical bone of the tibia (7) and a reduced bone mineral density of the femoral neck and trochanter, despite the employment of exercise countermeasures (39).

Extensive analysis of rats sent into orbit on board the Soviet Cosmos Biosatellites and American Spacelab 3 has contributed greatly to the understanding of mechanisms underlying human skeletal adaptations to weightlessness. These rats were characterized by a reduced trabecular mass in the tibial metaphysis (49) and a slowing of periosteal bone formation of the tibial and humeral shafts (48). In spite of these spaceflight experiments, the paucity of animals exposed to weightlessness during spaceflight has led to the development of ground-based models. Hindlimb unloading (HU) provides a model of mechanical unloading of the hindlimb while producing a headward fluid shift similar to that induced by microgravity. Skeletal adaptations in HU rats are similar to those observed in rats exposed to spaceflight, including a loss of cancellous bone mass in the proximal metaphysis of the tibia (47) and reduced periosteal bone formation in the tibia and femur (25).

The cephalic fluid shift that accompanies simulated and actual weightlessness may affect bone blood flow and potentially alter interstitial fluid pressures and flows. For example, previous work has demonstrated that HU-mediated alterations in perfusion pressure and blood flow induce a remodeling of the arterial resistance vasculature in skeletal muscle (9) and cerebral tissue (46). If these alterations were also to occur in the hindlimb bones, then the mechanical environment of osteoprogenitor cells, osteoblasts, and osteoclasts would be compromised. Such a perturbation would presumably result in the altered regulation of these cells and subsequently in changes in the balance between bone formation and bone resorption (4, 33). Therefore, the purpose of this study was to determine whether the unloading of hindlimb bones and the corresponding cephalic fluid shift alter bone perfusion rates. We hypothesized that perfusion would be diminished in the hindlimb bones and increased in skeletal structures of the forelimb and head. Furthermore, we also predicted that reductions in blood flow to the hindlimb bones would occur in regions where unloading-induced bone loss has been reported to occur.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. All procedures performed in this study were approved by the Texas A&M University Institutional Animal Care Committee and conformed to the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals [DHEW Publication No. (NIH) 85-23, Revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD 20892].

Surgical procedures. HU and control rats were anesthetized with pentobarbital sodium (30 mg/kg ip). The HU rats were anesthetized while remaining in the unloaded position to avoid any weight-bearing activity. A catheter (Dow Corning, Silastic; ID 0.6 mm, OD 1.0 mm) filled with heparinized (200 U/ml) saline and connected to a pressure transducer and chart recorder was advanced into the left ventricle of the heart via the right carotid artery as previously described (11). This catheter was subsequently used for the infusion of radiolabeled microspheres to measure tissue blood flow and record arterial pressure. A second polyurethane catheter (Braintree Scientific, Micro-renathane; ID 0.36 mm, OD 0.84 mm), used for the withdrawal of a reference blood sample and recording arterial pressure, was implanted in the caudal artery of the tail and filled with heparinized saline as previously described (11). Both catheters were externalized and secured on the dorsal cervical region.

Experimental protocol. After 24 h of recovery from the surgical procedure, control and HU animals were instrumented for blood flow determination. Blood flow in the control group was first measured while the animals were in a normal standing position. The hindlimbs of control rats were then elevated via tail suspension similar to that of the chronically unloaded rats. After a 10-min HU period, a second blood flow measurement was made in this position. In the 7-day and 28-day HU rats, blood flow was measured with the animals in the unloaded position. After the final microsphere infusion, euthanasia solution (0.22 ml/kg; Euthanasia-5 Solution, Henry Schein) was infused through the carotid catheter. Bones and muscles were then removed from the carcass. The femora from both hindlimbs were sectioned into three regions, the proximal and distal metaphyses and diaphysis; the femoral marrow was removed from the diaphysis and was counted as a fourth region (Fig. 1). The corresponding femoral sections from the left and right hindlimbs were combined for each animal to ensure sufficient microspheres in the marrow sample. The fibula was cut from the tibia and counted as a single sample (Fig. 1). The tibia was sectioned into the proximal and distal metaphyses and diaphysis. The marrow in the tibial shaft was not isolated or counted separately. Likewise, the humerus was sectioned into three regions with the marrow remaining in the humeral diaphysis (Fig. 1). Tissue samples were weighed and placed in counting vials for flow determination. The weight of the femoral marrow was determined by weighing the shaft before and after the marrow was removed.

View larger version (31K): [in this window] [in a new window]   Fig. 1.   The sampling pattern for determining regional blood flows in the femur, tibia-fibula, and humerus.

Blood flow measurements. Radiolabeled (⁸⁵Sr, ¹¹³Sn, and ⁴⁶Sc) microspheres (New England Nuclear) with a 15-µm diameter were used for blood flow measurements as previously described (24). Radioactivity of the samples was measured with a gamma counter (Packard AutoGamma 5780), and flows were computed (PC-GERDA V2.9 Software) from counts per minute and tissue wet weights. To ensure microsphere-blood mixing, flows to head and forelimb were taken only when the carotid catheter tip was confirmed to be in the left ventricle. Head and forelimb blood flows from two control and two 7-day HU rats were excluded when the carotid catheter tip was found to be in the ascending aorta. Microsphere mixing was considered adequate for hindlimb tissue flows when right and left kidney blood flows were within 15% of each other. On the basis of this criterion, no hindlimb tissue flows were excluded from the study.

Central hemodynamics and vascular resistance. Electronically averaged mean arterial pressure and heart rate (pulse pressure rate) were recorded from the caudal catheter immediately before and after the microsphere infusion and were averaged. In addition, pressure recordings were made with the transducer at the level of the heart and hindlimb to estimate thoracic aorta and hindlimb (caudal) arterial pressures, respectively. Aortic pressure estimates were confirmed in the two control and two 7-day HU rats having the carotid catheter tip in the ascending aorta. Arterial pressure was recorded in control rats during normal standing and after 10 min of HU and in HU rats while they remained with elevated hindlimbs. Caudal arterial pressures with the pressure transducer at the level of the midfemur were used to calculate vascular resistance in hindlimb bones and muscles. Arterial aortic pressures (with the pressure transducer at the level of the heart) were used to calculate vascular resistance in head and forelimb bones and muscles. Thus vascular resistance for specific tissues was calculated as the quotient of arterial pressure (caudal or aortic) divided by the tissue blood flow.

Data analysis. A one-way ANOVA was used to compare heart rate, mean arterial pressure, tissue mass, blood flow, and vascular resistance across conditions. The Student-Newman-Keuls method was used as a post hoc test to determine the significance of differences among means. All values are presented as means ± SE. A P < 0.05 was required for significance, although differences among means having a P < 0.1 are acknowledged.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Body and soleus muscle mass. The body masses at death of control (458 ± 9 g), 7-day HU (432 ± 10 g), and 28-day HU (447 ± 13 g) rats were not different. HU reduced soleus muscle mass from 7-day HU (206 ± 9 mg) and 28-day HU (148 ± 15 mg) rats relative to that of control animals (286 ± 34 mg). Soleus muscle mass was not different between the two HU groups. These alterations in soleus muscle mass are similar to the relative changes previously reported in HU rats (8, 9, 30, 46). Furthermore, soleus muscle atrophy, which is characteristic of reduced skeletal muscle weight-bearing activity, confirms the effectiveness of the HU intervention.

Hindlimb tissue blood flow and vascular resistance. Ten minutes of HU significantly reduced blood flow to the proximal femur, the femoral shaft, and the femoral shaft marrow (Fig. 2). After 7 days HU, blood flow to all portions of the femur, including the distal metaphysis, was significantly less than blood flow to femurs of control rats; there were no differences in femoral blood flows between 10 min and 7 days HU. After 28 days of unloading, femoral blood flow continued to be lower than that of control rats. However, flow to the femoral shaft and marrow after 28 days of tail suspension was less than that with 7 days HU. The reductions in femoral flow with 10 min and 7 days of HU resulted from an increase in vascular resistance (Fig. 2), because tissue blood flow is primarily the result of arterial perfusion pressure and vascular resistance, and caudal arterial pressure was not different at these times (Table 1). The further decrease in flow to the femoral shaft with 28 days HU was accompanied by a continued increase in resistance. In contrast, the further reduction in shaft marrow perfusion with 28 days HU did not result from a continued increase in vascular resistance but resulted from a decreased hindlimb perfusion pressure (Table 1).

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Effect of hindlimb unloading on regional femoral blood flow (A) and vascular resistance (B). Values are means ± SE; n = 11 (for control standing and 10 min and 7 day unloaded) and n = 7 (for 28 day unloaded). * Mean is different from control standing mean (P < 0.05);  mean is different from 10-min unloaded mean (P < 0.05);  mean is different from 7-day unloaded mean (P < 0.05).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Effect of hindlimb unloading on hindlimb muscle blood flow (A) and vascular resistance (B). Values are means ± SE; n = 11 (for control standing and 10 min and 7 day unloaded) and n = 7 (for 28 day unloaded). * Mean is different from control standing mean (P < 0.05);  mean is different from 10-min unloaded mean (P < 0.05);  mean is different from 7-day unloaded mean (P < 0.05).

Forelimb and head tissue blood flow and vascular resistance. Perfusion of the humerus was acutely elevated with 10 min HU but returned to control levels by 7 days HU (Fig. 4). The acute increase in humeral blood flow was not the result of a decrease in vascular resistance (Fig. 4) but resulted from an elevated perfusion pressure (Table 1). The subsequent return of blood flow to control levels with 7 days HU resulted from an elevation in vascular resistance. In contrast, the normalized humeral blood flow with 28 days HU was not the result of a heightened resistance but of a normalization of central arterial perfusion pressure. An almost identical pattern of blood flow response to unloading occurred in the mandible and clavicle (Table 2). The skull similarly demonstrated an increase in blood flow with acute HU. However, unlike in the humerus, mandible, and clavicle, this increased perfusion was associated with a decreased vascular resistance.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Effect of hindlimb unloading on regional humeral blood flow (A) and vascular resistance (B). Values are means ± SE; n = 9 (for control standing and 10 min and 7 day unloaded) and n = 7 (for 28 day unloaded). * Mean is different from control standing mean (P < 0.05);  mean is different from 10-min unloaded mean (P < 0.05).

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Effect of hindlimb unloading on forelimb and facial muscle blood flow (A) and vascular resistance (B). Values are means ± SE; n = 9 (for control standing and 10 min and 7 day unloaded) and n = 7 (for 28 day unloaded). * Mean is different from control standing mean (P < 0.05).

Skeletal mass. Femoral mass was reduced by 28 days HU (Table 3). In addition, there was a tendency for tibial and fibular mass to be reduced with prolonged HU (P < 0.1). In contrast, there was an increased clavicular and mandibular mass with 28 days HU and a tendency for a greater humeral mass (P < 0.1). Mass of the pelvis, radius/ulna, scapula, and skull did not change with HU.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The primary purpose of this study was to test the hypothesis that mechanical unloading of hindlimb bones and the decreased arterial perfusion pressure to the hindlimbs decreases skeletal perfusion. In addition, we sought to determine whether the unloading-induced cephalic fluid shift alters skeletal blood flow in forelimb, shoulder, and head bones. The results indicate that HU rapidly diminishes blood flow to the femoral and tibial metaphysis (cancellous bone), diaphysis (cortical bone), and marrow, and that prolonged unloading further decreases perfusion of the femoral shaft and marrow. Chronic unloading was also necessary to decrease fibular perfusion rate. In contrast, HU acutely elevates blood flow to the humerus, clavicle, skull, and mandible. The decline in blood flow to the hindlimb bones appears to coincide with a diminished mineral apposition rate, density, and mass of both cortical and cancellous bone observed in HU rats (present study, 5, 8, 16, 25, 31, 38, 44). Correspondingly, the acute increase in blood flow to forelimb, shoulder, and head bones appears to coincide with some reports of increased bone mass in HU rats (present study, 38). Therefore, our data support the hypothesis that HU-induced changes in bone perfusion lead to alterations in the balance between bone resorption and bone formation.

Several mechanisms could serve to link changes in bone perfusion with skeletal remodeling (Fig. 6). The first is the effect that alterations in blood flow may have on interstitial fluid flow. For example, it has been proposed that alterations in bone interstitial fluid flow may influence bone remodeling (41). Interstitial fluid flows radially through cortical bone, driven by a transmural pressure gradient between the vasculature of the endosteal surface and the periosteal lymphatics (36). Mechanical loading of the skeletal system causes fluid flow through cancellous bone and the lacunacanalicular network of cortical bone, exacerbating flow-induced shear stress imposed on surface bone cells (40). The shear stresses generated by bone interstitial fluid flow appear to be of similar magnitude to those occurring at the blood-vascular endothelium interface (37). Bone cells respond to fluid shear forces in a manner similar to vascular endothelial cells, by generating autocrine/paracrine signals that modulate remodeling activity (18, 21, 36, 40). For example, cultured osteoblasts increase production of nitric oxide (NO) and PGE₂ when exposed to elevated flow-induced shear stress, but mechanical strain does not elicit a similar release (40). NO has been shown to be an osteoblast mitogen (18, 37) with an inhibitory effect on osteoclast bone resorption (22, 27, 28, 43). PGE₂ stimulates bone formation and attenuates bone loss with immobilization in vivo (1), presumably via stimulation of osteoblast mitosis, reduction of osteoclast number (20), and inhibition of osteoclast activity (6, 15). Results from the present study of an increased blood flow and perfusion pressure in the forelimb, shoulder, and head are consistent with the hypothesis of an elevated transmural interstitial fluid pressure gradient between the endosteal vasculature and the periosteal lymphatic drainage system. Although blood flow to the skull and fore bones of the HU rat was only acutely elevated, this does not necessarily imply that the stimulus for bone remodeling was brief. For instance, the rise in head and forelimb arterial perfusion pressure would increase vascular transmural pressure in bone, thereby promoting capillary/sinusoid fluid filtration and elevating interstitial fluid pressure. The elevation of interstitial fluid pressure would serve to normalize vascular transmural pressure and return blood flow to control levels. However, the maintenance of a high interstitial fluid pressure, which serves to offset the chronically elevated arterial perfusion pressure (30, present study), may provide a chronic stimulus to increase bone formation through a NO (21, 42) or PGE₂ (34, 36) signaling mechanism.

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Hypothetical sequence of events linking the unloading of hindlimb bones and the reduction in skeletal blood flow to a remodeling imbalance that leads to a net loss of bone. Two possible mechanisms are illustrated, that mediated through a diminished interstitial shear stress and that mediated through a reduced vascular shear stress (see text for details). PGE2, prostaglandin E2; NO, nitric oxide; ED-NO, endothelium-derived nitric oxide; ED-PGI2, endothelium-derived prostacyclin.

The increases in bone mass with HU (present study, 38) are not unique to this model of microgravity. Increases in skull bone density have been reported to occur in humans during bed rest (3, 26) and in cosmonauts after long-duration spaceflight (39). The observations of an increased head bone density in humans are consistent with the hypothesis that headward fluid shifts may alter normal bone remodeling.

Results from the present study also demonstrate a reduced blood flow to hindlimb bones with mechanical unloading. Diminished bone perfusion may have direct effects on interstitial fluid flow and, consequently, bone formation by reducing capillary and sinusoid fluid filtration (Fig. 6). Reductions in interstitial fluid movement have been reported in the rat femur with HU (12). Furthermore, Bergula and collaborators (4) used femoral vein ligation to increase intramedullary fluid pressure and presumably interstitial fluid flow in HU rats. Ligation increased femoral bone mineral content and trabecular density over sham-operated control limbs. The authors suggested that the protective effect of ligation on bone mass was apparently due to increased interstitial fluid flow and the corresponding release of NO and PGE₂. These studies collectively provide support for the notion that hindlimb bone loss in HU rats is associated with reduced interstitial fluid flow.

Altered bone perfusion may also be linked to skeletal remodeling via a vascular mechanism (33). On the basis of the juxtaposition of blood vessels and bone cells, it was first proposed in 1930 that the bone's vascular network might be an active mediator of skeletal adaptation (19). More recently, it has been demonstrated that blood vessels are located in the basic multicellular units containing osteoclasts and osteoblasts that carry out bone remodeling (cf. 33). In addition, these vessels are located no more than 100 µm from the site of each unit's remodeling activity (cf. 33). Vascular endothelial cells release substances in response to changes in fluid flow and shear stress that may act directly on bone cell populations or serve as paracrine modulators of bone cell activity within the basic multicellular unit (2, 14, 22, 28, 33). Specifically, NO and PGI₂ are potent vasodilators that are, in many tissues, released by the vascular endothelium in response to blood flow and, correspondingly, intravascular shear stress (32). The effects of NO on bone formation and resorption are well established (18, 22, 27, 28, 37, 43). PGI₂ has been shown to exert a direct inhibitory effect on osteoclasts (6, 35). This raises the possibility that HU-induced increases and decreases in blood flow and shear stress alter vascular endothelial cell release of these agents, which could subsequently modify the focal balance between osteoblast and osteoclast activity (Fig. 6).

There appear to be several mechanisms responsible for the increased and decreased blood flow to the fore and aft skeleton of the HU rat, respectively. The initial decreases in flow to the femora and tibiae were accompanied by increases in vascular resistance. Presumably this increase in resistance is not mediated by a sympathetic neural mechanism, because cephalic fluid shifts engage cardiopulmonary receptors and elicit reflexive decreases in efferent sympathetic nerve activity (45). More likely, the decreased loading of the hindlimb bones diminished the net metabolic rate of the bone cells in these male animals. Metabolic control of vascular resistance is the classic regulatory mechanism whereby tissue metabolism is coupled to oxygen delivery. In addition to increases in vascular resistance, a decrease in hindlimb perfusion pressure also appears to contribute to the decreased blood flow. For example, blood flow to the femoral marrow is lower with 28 days HU than with 10 min or 7 days HU. This decreased flow is not the result of a further increase in vascular resistance, indicating that the lower caudal arterial pressure is responsible for the diminished perfusion. Similarly, fibular perfusion at 28 days HU is lower than at 10 min HU without a corresponding increase in vascular resistance, indicating that the lower flow with prolonged unloading is the result of a diminished perfusion pressure. It is interesting to note that changes in blood flow to hindlimb skeletal muscles consistently correspond to changes in vascular resistance (Fig. 3). These data indicate that autoregulation of vascular resistance, rather than the lower perfusion pressure, appears to be entirely responsible for determining hindlimb muscle perfusion.

Unlike with the hindlimb bones, blood flow to the skull, humerus, clavicle, and mandible increased with acute unloading. The transient increase in skull perfusion resulted from a lower vascular resistance. It is possible that the lower resistance may have resulted from an increased loading of the skull, a consequence of a transient increase in intracranial pressure (29). In contrast, increased perfusion of the humerus, clavicle, and mandible were not accompanied by changes in vascular resistance, indicating that the elevated flow to these bones was a result of the increased thoracic arterial perfusion pressure. With 7 days HU, however, vascular resistance was elevated in these fore bones. Interestingly, blood flow to facial and forelimb muscles did not increase with acute or chronic unloading, owing to the fact that elevations in vascular resistance offset elevations in perfusion pressure. These data suggest that the resistance vasculature in both the fore and aft skeleton is not as adept at compensating for fluid pressure shifts as the skeletal muscle resistance vasculature. Although this apparent inability to precisely regulate blood flow appears to make bone more "susceptible" to fluid shifts, this could serve a functionally important role as a stimulus or signal for skeletal remodeling.

Rat HU is a popular ground-based model to study the effects of microgravity on the musculoskeletal and cardiovascular systems (9, 30). To date, there are no bone blood flow data obtained in a weightless environment to support or refute the findings of an increased and decreased blood flow to the skeleton of the HU rat. However, there are data from space-flown rats that provide indirect evidence of a diminished perfusion of hindlimb bones in microgravity. Doty and colleagues (13) reported that blood vessels on the periosteal surface of the tibial diaphysis in rats flown on Cosmos 1887 had smaller cross-sectional areas. This type of vascular remodeling is indicative of a chronically reduced blood flow through the vessel (9, 23). In addition, these authors reported signs of vascular endothelial degeneration and lipid inclusion and suggested that these vascular alterations may be associated with ischemia. Further work will be required to determine whether skeletal vascular alterations induced by unloading are similar to those reported to occur in microgravity.

In conclusion, the present study demonstrates that mechanical unloading of bone rapidly diminishes blood flow to the femoral and tibial metaphysis and diaphysis and the femoral marrow. Prolonged unloading further decreases perfusion of the fibula and femoral shaft and marrow. These decreases in flow appear to be the result of both increases in vascular resistance and decreases in perfusion pressure. In contrast, HU acutely elevates blood flow to the humerus, clavicle, mandible, and skull. The increased flow to the skull was the result of a decrease in vascular resistance, whereas the increased flow to the other forelimb and facial bones was the result of an increased perfusion pressure. The decline in blood flow to the hindlimb bones appears to coincide with a diminished mineral apposition rate, density, and mass of both cortical and cancellous bone observed in HU rats (present study, 5, 8, 16, 25, 31, 38, 44). Correspondingly, the acute increase in blood flow to forelimb, shoulder, and head bones appears to coincide with some reports of increased mass (present study, 38). Therefore, we speculate that alterations in bone perfusion and associated changes in interstitial or intravascular shear stress may provide a stimulus for altering the balance between bone resorption and bone formation with simulated microgravity. The impact of these findings on human space flight warrants continued investigation.