INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PERIPHERAL ARTERIAL OCCLUSIVE disease is a common cardiovascular disease, especially in populations of the elderly in which prevalence can be as high as ~12% (8). Peripheral arterial insufficiency may greatly reduce physical activity, deteriorate quality of life, and increase mortality rate. The mortality rate caused by cardiovascular diseases and coronary heart disease is about 15 times higher in patients with pathological changes in a large peripheral arterial vessel than those age- and sex-matched control individuals who do not have such vessel lesions (9). Interestingly, increased physical activity is one important means of treating patients with peripheral arterial insufficiency (13). Physical training improves collateral blood supply (1, 11, 17) to the ischemic limb as well as blood flow redistribution (38) and muscular aerobic capacity within the affected limbs (26, 38). Patients realize an improved exercise tolerance after physical training (1, 13, 17, 26, 38). However, for patients with exercise contraindication, the treatment options aimed to improve blood supply to the affected limb are limited. Thus identification of treatment strategies that can effectively induce collateral blood flow development will greatly benefit these patients.

Recent research findings indicate that heparin-binding growth factors are potent angiogenic agents that enhance collateral circulation in experimental arterial insufficiency. We find that collateral-dependent blood flow to hindlimb muscles increases ~200% in animals with exogenous basic fibroblast growth factor (bFGF) infused in young adult rats with bilateral femoral artery ligation (32). An even greater collateral-dependent blood flow can be obtained if bFGF infusion is combined with daily exercise at a moderate intensity in young adult rats (32). The effectiveness of bFGF in promoting collateral circulation function is also observed in the ischemic swine heart (4), the rabbit hindlimb (3), and the rodent hindlimb (5, 6) of young adult animals.

Whereas the improvement of collateral circulation with bFGF administration in young adult animals is encouraging, it is unclear whether a similar response can be found in the aged adult because they are most affected by peripheral arterial insufficiency (8). Therefore, the present study was designed to examine the efficacy of exogenous bFGF to expand collateral blood flow in an aged animal model of experimental peripheral arterial insufficiency. We previously showed that exercise training induced similar increases in collateral blood flow in both young adult and Fischer 344 rats (33, 35). Therefore, we hypothesized that the aged rats remain responsive to bFGF administration, and an increased collateral blood flow to the ischemic hindlimb would be found in the aged rats with peripheral arterial insufficiency following intra-arterial bFGF delivery.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal care. This study was approved by the Committee for the Humane Use of Animals of the State University of New York, Health Sciences Center at Syracuse, NY. All experimental procedures were carried out in accordance with National Institutes of Health guidelines.

Seventy-five male Fischer 344 rats weighing 400-425 g (from National Institute of Aging colony, Harlan Sprague Dawley, Indianapolis, IN) at two age groups: 10 mo-old (n = 32) and 22 mo-old (n = 43) were shipped in. Animals were housed three per cage in a temperature- (21°C) and light (12 h:12 h dark/light cycle)-controlled room. Rats were fed Purina rat chow and tap water ad libitum. On arrival, all rats were accustomed to the treadmill protocol by walking at 15-20 m/min and 15% grade for ~5 min, twice a day for ~2 wk. The treadmill protocol included a brief period of turning the treadmill on and off to condition the rats to run at the front of the treadmill when the belt started moving. Our previous experiments indicate that the aforementioned treadmill orientation protocol does not cause any detectable peripheral adaptations in young adult rats (31, 37).

Experiment design. Different doses of bFGF were tested in the two age groups. Within each age group, rats were randomly subdivided into six experimental groups: 1) normal control group with the hindlimb circulation intact, blood flow measurements during treadmill running indicate normal exercising hyperemia; 2) acute ligation group, rats received bilateral ligation of the femoral arteries 3 days before blood flow determination to serve as baseline for collateral flow; and 3) four chronic ligation groups, each with one of the following bFGF doses: 0 (carrier control), 0.5, 5, and 50 µg · kg¹ · day¹. bFGF dose response was determined using collateral-dependent blood flow to the gastrocnemius-plantaris-soleus (GPS) muscles as an indicator. In 10-mo-old animals, there was no bFGF (0.5 µg · kg¹ · day¹ group) because a previous study has shown no effect at this dose in adult animals (30). Carrier solution or bFGF was constantly delivered by osmotic pump (14 days capacity), and blood flow to the ligated hindlimb was determined on the sixteenth day after pump implantation.

Surgical preparation. Under isoflurane gas anesthesia, both left and right femoral arteries were isolated and ligated twice with 3-0 surgical silk sutures ~1 cm distal to the inguinal ligament. In addition, on the left side, a polyethylene (PE)-60 catheter connected to an osmotic pump (Alzet model 2002, Alza) was inserted into the left common iliac artery through the ligated femoral artery to establish the route for bFGF delivery. Topical antibiotic powder (Neo-Predef, Upjohn) was placed on the wound before closure with skin clips.

Osmotic pumps were prepared according to the instruction of Alza. The miniosmotic pumps were designed for constant delivery at a flow rate of 0.50 ± 0.02 µl/h for 14 days. The pumps were filled with either carrier (10% sodium citrate to prevent coagulation, 1.6% glycerol to stabilize protein, and 10 mM phosphate-buffered saline) or carrier plus different concentrations of bFGF to achieve the following doses: 0.5, 5, and 50 µg · kg¹ · day¹ as done previously (32). The dead space of the femoral catheter was filled with the same solution as in its connected pump. The pump was housed in a tunnel under the subcutaneous tissue in the left groin area; this did not affect the hindlimb movement while rats were walking on the treadmill.

Blood flow determination. On the sixteenth day after the pump installation, isoflurane-anesthetized rats from each age group were surgically prepared for blood flow measurement as done previously (30, 31). Briefly, a PE-50 catheter was placed in the left carotid artery and advanced to the arch of the aorta for monitoring blood pressure (BP) and heart rate and infusing microspheres. A second catheter was placed in the caudal artery to monitor caudal BP and to obtain the reference blood sample during microsphere infusion. The arteries were catheterized early in the day, and the rats only needed a few minutes to recover from the anesthesia. The fully conscious animals were run on the treadmill for blood flow determination after more than 4 h of recovery as described previously (30, 31).

Muscle blood flow was determined by using radiolabeled microspheres (⁸⁵Sr and ¹⁴¹Ce, 15 ± 0.1 µm diameter, NEN, Boston, MA) during the second minute of running at both low (15 m/min at 15% grade) and high speeds (20 m/min at 15% grade) to ensure a maximal vessel dilation. Within this range of exercise, blood flow is proportional to the running intensity (19), and a maximal vascular dilation is achieved if blood flow does not increase at a higher intensity. At the end of the first minute of running at each speed, a well-mixed suspension of microspheres was carefully infused through the carotid catheter, followed by a saline flush over ~20 s. At the same time, a reference blood sample was withdrawn from the caudal artery at a rate of 500 µl/min (beginning 10 s before each microsphere infusion). After the microsphere infusion, animals were killed by a pentobarbital sodium overdose. Tissue samples dissected from both hindlimbs, together with the reference blood flow sample, were counted with a gamma counter (Beckman Autogamma). Muscle blood flow (ml · min¹ · 100 g¹) was calculated as
where RBS is the reference blood sample and CPM is counts per minute. Results of both hindlimbs were averaged after it was determined that no difference existed between the left and right hindlimbs. Furthermore, comparison of kidney blood flows within each animal provided evidence of proper mixing of microspheres. Blood flow to individual tissue sections was summed to assess blood flow to the total, proximal, and distal regions, as previously done (30, 31).

Data analysis. All data are expressed as means ± SE. After initial analysis of variance (ANOVA) showed no difference between the left and right hindlimb, values of left and right hindlimb blood flow were combined into a single value for each speed for an animal. Another ANOVA was then used to determine whether there was a speed effect within each animal. The analysis showed that there was no difference between the low- and high-speed values. Therefore, we combined the values of low and high speeds for each tissue. Thus we obtained a single blood flow value for each tissue for each rat. This value was used for ANOVA with two factorial designs across the treatment groups (not including bFGF group with 0.5 µg · kg¹ · day¹). ANOVA with repeated measures were used across the treatment groups for BP, heart rate, kidney, and psoas muscle blood flow data. A P value < 0.05 is recognized as significantly different. Statistical differences between or within treatment were identified by Tukey's procedure (27).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Body weight of aged Fischer 344 animals (~23 mo old) remains similar to that of younger animals. Although the body weights were not different between the two age groups within any treatment, both young adult and aged animals of the 50 µg · kg¹ · day¹ bFGF group were smaller than the normal adult (11 mo) animals (P < 0.05) (Table 1). The total hindlimb and proximal hindlimb weights were lower in two adult animal (11 mo) groups (bFGF 0 and 50 µg · kg¹ · day¹; Table 1). However, the weight of the collateral flow-dependent hindlimb portion, distal hindlimb, and GPS muscles showed no difference across treatments or age groups (Table 1).

Bilateral femoral artery ligation increased BP before and during high-speed exercise (cf. Table 2) (P < 0.025). This observation was found in both acutely and chronically ligated rats. Furthermore, BP in the ligated adult animals tended to be higher than the BP in the aged groups (cf. Table 2). Exercise further elevated BP in the ligated groups (both adult and aged) (P < 0.05) but not in the normal control group. The ligation procedure did not affect heart rate. The aged animals tended to have a lower heart rate during each exercise condition compared with the younger animals (P < 0.05). Increased heart rate was found in both young adult and aged groups during exercise, especially when the animals were running at high speed (Table 2).

The excellent mixing of microspheres was confirmed by the similar blood flows to the left and right kidneys; the value of the left kidney blood flow versus the right kidney blood flow is 1.07 ± 0.07 ml · min¹ · 100 g¹ in the young adult animal group (60 observations) and 1.03 ± 0.08 ml · min¹ · 100 g¹ in the aged group (85 observations). Similarly, blood flow to the left versus right total hindlimb is 1.02 ± 0.03 ml · min¹ · 100 g¹ in the young adult animal group (60 observations) and 0.99 ± 0.02 ml · min¹ · 100 g¹ in the aged group (85 observations). Therefore, the values of the left and right hindlimb are combined into one value for presentation. Furthermore, blood flows determined at low and high speeds showed no difference in the ligated hindlimbs; this observation suggests that a peak blood flow was achieved in the ligated hindlimb during the low-speed running because muscle blood flows are increased proportionally to the treadmill intensity at this intensity range (19). Thus blood flows of low and high speeds are combined into one value as presented in Tables 3 and 4 and Figs. 1-3.

The aging process (up to 23 mo) did not affect the blood flow distribution within the hindlimb of animals with normal hindlimb blood circulation (cf. normal in Table 3). Acute femoral artery ligation greatly reduced blood flow to the distal hindlimb and GPS muscle group (from ~160 to ~10 ml · min¹ · 100 g¹) in both age categories (P < 0.001). Collateral-dependent blood flow was only moderately increased in the carrier control group (bFGF, 0.0 µg · kg¹ · day¹) by comparing with the acute ligation group (Table 3).

Collateral-dependent blood flow (blood flow to the weight-bearing muscle; GPS muscle group) in both age groups exhibited a sigmoid relationship to varied bFGF dose on a semi-log scale (Fig. 1). Our previous data show that in young adult animals, the bFGF dose of 0.5 µg · kg¹ · day¹ does not increase collateral-dependent blood flow (30). Therefore, this bFGF dose was omitted from the young adult groups in the present study. Similarly, bFGF (0.5 µg · kg¹ · day¹) did not enhance collateral blood flow in the aged animal group compared with the flow in the carrier control group (Fig. 1). bFGF infused at 5 µg · kg¹ · day¹ effectively increased the collateral-dependent blood flow (Fig. 1) in both age groups. At this dose, flow to the GPS muscle group was increased by ~66% in the young adult and by ~100% in the aged group in comparison to the carrier controls (P < 0.001). Higher concentration of bFGF (50 µg · kg¹ · day¹) failed to further increase collateral-dependent blood flow in either age group (Fig. 1). However, at 50 µg · kg¹ · day¹ of bFGF infusion, the percent increase of collateral-dependent blood flow to GPS muscle was ~82% in the adult (11 mo) and ~139% in the aged (23 mo) animals compared with the same age carrier control group.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Collateral-dependent blood flow (ml · min1 · 100 g tissue wt1) in gastrocnemius-plantaris-soleus (GPS) muscle group in response to different basic fibroblast growth factor (bFGF) dose infusions for 14 days. bFGF, 0 µg · kg1 · day1, represents carrier-only group. Number of observations in each group refers to Table 1. * Significantly different from bFGF 0 and 0.5 µg · kg1 · day1 groups (P < 0.05).

In addition to the increased collateral-dependent blood flow in the GPS muscle group, bFGF administration at 5 and 50 µg · kg¹ · day¹ significantly improved blood flow to each of the total, proximal, and distal hindlimb sections of both adult and aged animals in comparison with the carrier control group (bFGF 0.0 µg · kg¹ · day¹) (P < 0.05) (Table 3). The only exception was found in the proximal hindlimb of both age groups infused with bFGF (5 µg · kg¹ · day¹) (Table 3). Furthermore, the extent of the increase in blood flow to all hindlimb sections was similar in both adult and aged groups after receiving bFGF treatment (at both 5 and 50 µg · kg¹ · day¹, cf. Fig. 2). Collateral-dependent blood flow to the distal hindlimb, which consists of both weight bearing- and non-weight-bearing muscles, increased moderately in carrier control groups compared with their acute ligation counterparts in both age groups (P < 0.01) (Fig. 3). Systematic delivery of bFGF 50 µg · kg¹ · day¹ for 14 days further enhanced the collateral-dependent blood flow to a similar extent in both age groups compared with the carrier group (P < 0.01) (Fig. 3). Although carrier infusion did not increase blood flow to the total and proximal hindlimb region, bFGF infusion improved blood flow to the total and proximal hindlimb in both age groups (Fig. 3).

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Percent increase of blood flow in total, proximal, distal hindlimbs and GPS muscle group in bFGF (5.0 µg · kg1 · day1; A)- and bFGF (50.0 µg · kg1 · day1: B)-infused groups in comparison with the carrier controls (bFGF 0 µg · kg1 · day1). Statistical analysis showed that extent of increased flow is not different between the two age groups.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Blood flow of total (C), proximal (B), and distal (A) hindlimb sections expressed as ml · min1 · 100 g tissue wt1 in acute ligated (3 days before blood flow measurement), carrier-infused (bFGF 0 µg · kg1 · day1) or bFGF (50.0 µg · kg1 · day1)-infused groups. * Significantly different from acute ligation group (P < 0.05).  Significantly different from carrier control group (P < 0.05).

Table 4 lists blood flows to individual muscle sections. Bilateral femoral artery ligation greatly reduced blood flow to individual muscles in the distal hindlimb but showed minimal effect on the proximal hindlimb muscles (with the exception of biceps femoris). bFGF administration (5.0 and 50.0 µg · kg¹ · day¹) partially restored blood flow to the muscles of the distal hindlimb (P < 0.001). Moreover, the blood flow to certain muscles in the proximal hindlimb also increased following bFGF infusion at 5.0 and 50.0 µg · kg¹ · day¹. Blood flow in the white quadriceps muscle of adult animals (received bFGF 5.0 and 50.0 µg · kg¹ · day¹ of infusion) increased about two- to threefold, compared with animals with normal circulation (Table 4).

Kidney and psoas muscle (trunk muscle) blood flows determined at low and high speeds are listed in Table 5. A reduced kidney blood flow was found during high-speed running (P < 0.05). bFGF (5.0 and 50.0 µg · kg¹ · day¹) infusion prevented reduction of kidney blood flow during high-speed running in the adult rats (Table 5). There are bFGF effects on blood flow to the nonexercised trunk muscle (psoas muscle). No age differences in psoas muscle blood flow were found among the treatment groups (Table 5).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study shows that aged Fischer 344 rats (~23 mo old) are as responsive to exogenous angiogenic factor (bFGF) stimulation of collateral vessel growth to the hindlimbs as that in 11-mo-old adult animals (cf. Fig. 2). Blood flow to the calf muscles during treadmill running increased to a similar extent (~80%) in both adult and aged animals. This increase in blood flow to the calf muscles is most significant, because flow to this muscle group is collateral dependent following proximal occlusion of the femoral artery (34, 35, 37). Furthermore, our experimental protocol was designed to measure maximal collateral-dependent flow during treadmill running. We measured blood flow during two running speeds: the first chosen to be demanding of high flow, and the second demanding a slightly higher speed to stimulate further vasodilatation. The absence of a greater blood flow to the calf muscles at the higher treadmill speed can be interpreted as peak muscle flow, determined by the minimal resistance of the upstream collateral vessels that circumvent the occluded femoral artery in the thigh. Furthermore, the dose response of the aged animals to bFGF was not different from that observed for the 11-mo-old adults (Fig. 1). Thus the aging process, reflected in 23-mo-old Fischer 344 rats, does not compromise the ability to respond to exogenous bFGF stimulation of collateral vessel development following vascular occlusion in the peripheral circulation. This is significant because the aged Fischer 344 rat exhibits an altered vascular function (7) that may be characteristic of dysfunctional vasculature that would be refractory to stimuli.

Recent evidence indicates that vascular remodeling of preexisting vessels in the region surrounding a vascular obstruction is the important process in response to occlusion of the femoral artery (4, 15, 32). This vascular remodeling is enhanced by bFGF as evident by proliferation of endothelial and smooth muscle cells (2, 15), by enlargement of existing conduit vessels (4, 15, 32), and by increases in collateral-dependent blood flow downstream (2, 4, 32, 36). The process of remodeling existing vessels has been named "arteriogenesis," because of active cell proliferation evident within the collateral vessel walls (15) and is distinct from de novo microvascular formation of new capillaries by the process of angiogenesis. This latter process would be effective at expanding the microvasculature within a limited region of influence but not effective at establishing a large decrease in resistance of the upstream collateral vessels within the time course observed experimentally in this study. Thus the vasculature of interest are the existing vessels in the upper thigh that surround the femoral artery occlusion. It is important to note that arterial obstruction appears to be an essential requisite for vascular remodeling (30). Delivery of an efficacious dose of bFGF to the normal vasculature of a limb in the absence of vascular occlusion does not induce arteriogenesis as observed in the contralateral limb, which was subjected to experimental occlusion of the femoral artery (30). The obstruction of a primary conduit artery establishes altered pressure and flow patterns in the surrounding vasculature (16). It has been suggested that stimuli from these changed blood flow dynamics (14) and/or a tissue inflammatory response, but not frank tissue ischemia (2), are important contributors to collateral vessel expansion. Even though the precise mechanisms are presently unclear, our results indicate that the outcome of these stimuli and their response to exogenous bFGF lead to a similar benefit of increased collateral-dependent blood flow to the distal limb muscle in the aged animals.

Our results appear at odds with evidence indicating that aging may blunt the angiogenic process. For example, tumor angiogenesis (18, 20), microvascular adaptations to exercise training in human skeletal muscle (12) and muscle capillarity increases in animals in response to muscle stimulation (29) and exercise (28) have all been tempered in aged as compared with the young counterparts. However, there may be a fundamental difference in the underlying process leading to these vascular adaptations. All of the above studies have focused on microvascular changes related to angiogenesis, whereas, as described above, the process of remodeling preexisting vessels likely involves very different vessels, stimuli, and/or sequence of cellular events. Thus direct comparison of data sets and their outcomes may not be appropriate. Furthermore, it is possible that the magnitude of stimuli prompting vascular adaptation was not equivalent across the studies, for example, even when considering the same outcome of altered capillary density. We reported earlier (33, 35) that aged rats are fully capable of increasing muscle capillary density as adult controls when the prompting stimulus of exercise training was sufficient. Thus the absence of a response in the aged rats may simply reflect the inability to develop an appropriate adaptive stimulus, rather than an inability of the vasculature to respond. Similar findings were recently reported by Rivard et al. (23); angiogenic responses were dulled in aged rabbits and mice, but an equivalent stimulation response of angiogenesis was prompted by exogenous delivery of vascular endothelial growth factor (23).

Femoral artery occlusion raised the BP in our animals; this was observed particularly during exercise, was most apparent in the adult control animals even before exercise, and tended to be tempered in the aged especially with bFGF treatment. The increased BP induced by femoral artery ligation may be partially explained by a "pressor reflex" initiated by contracting ischemic muscle (24, 25). During exercise, ischemic muscle can produce metabolites that activate chemosensitive afferent nerve fibers in the muscle and trigger a pressor reflex through sympathetic nerve efferent fibers (24, 25). This enhanced sympathetic efferent activity, however, might be expected to reduce renal blood flow to an exaggerated extent, an outcome not apparent in our data set (cf. Table 5). Another factor that is unclear is the elevation in BP in the 11-mo-old with acute ligation and bFGF 5.0 µg · kg¹ · day¹ groups before exercise while the animals were on the treadmill. It is unknown whether any anticipatory response of these animals could have simulated a pressor reflex characteristic of contracting ischemic muscle. An acute hypotensive influence of bFGF, typical of the response with an acute bolus administration (10), probably did not confound our experiment, because the osmotic pump used to infuse bFGF was complete after 14 days, ~2 days before our BP and blood flow determinations. Whereas any hypertension would have altered the pressure at the head of the collateral vessels and thereby enhanced absolute calf muscle blood flow, there was no measurable difference in BP values between the two age groups within the same treatment. Also worthy of mention is the relatively "normal" muscle blood flows for the nonoccluded ~23-mo-old aged rats. Calf muscle blood flow (~175 ml · min¹ · 100 g¹) and muscle fiber-specific flows to the high-oxidative red quadriceps (fast-twitch red; 350 ml · min¹ · 100 g¹) and low-oxidative white quadriceps (fast-twitch white; 30 ml · min¹ · 100 g¹) were not different from the 11-mo-old control values. Furthermore, they are relatively good matches to the absolute values for these muscles in young adult rats measured previously (19, 31) at the relatively low treadmill speeds used. In this regard, it is important to recognize that the running speeds used in this study were "submaximal" for the animals, although challenging because of the peripheral flow limitation. Thus our "normal" muscle blood flows for the normal nonligated aged animals do not relate to the well-known reduction in maximal physical performance and aerobic capacity experienced by the aged animals (21, 22).

In the present study, peripheral arterial insufficiency was created via bilateral ligation of the femoral arteries. It should be recognized that this procedure does not represent the broad range of pathological changes found in patients with peripheral arterial insufficiency. Rather, it may be characteristic of acute-onset occlusive disease associated with a thrombus in a major proximal supply artery. Our animals experienced a greatly reduced flow reserve to the distal limb tissue, a condition characteristic of intermittent claudication. Our results show that aged animals up to 23 mo old remain responsive to an angiogenic growth factor-induced increase in collateral blood flow to the ischemic hindlimb muscles during treadmill exercise. Although we did not evaluate muscle function in the present study, we know that increases in collateral blood flow do improve muscle function (32, 34-36). Therefore, we expect that an increased collateral blood flow will lead to better muscle performance and improve limb function. This would significantly improve the quality of life of affected patients if a similar response were realized clinically. This could have a secondary benefit if the individuals chose to be physically active. Because an improved physical performance should enable an individual to be more physically active, a further enhancement of the vascular remodeling process and increase in collateral-dependent blood flow would be expected as previously shown (36). The vascular improvement with bFGF (3, 5, 6, 30, 32, 36) shows the potential for a new therapeutic avenue in managing selective patients with peripheral arterial insufficiency. The results of the present study suggest that the value of bFGF should not be preempted in the aged, the population most affected by peripheral arterial insufficiency.