INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MOST MICROVASCULAR BEDS possess the ability to regulate local hemodynamics in response to metabolic requirements or the physical stimuli provided by changes in intravascular pressure or blood flow itself (shear stress). The degree to which local hemodynamics are controlled, or autoregulated, varies among tissues (12). The ability to regulate local blood flow on a moment-to-moment basis requires that under resting conditions, arterioles exist in a state of partial contraction or tone. From this state, vasodilatation can be invoked to increase local blood flow or vasoconstriction can occur to decrease flow. Such vasomotor responses may be initiated by the local release of paracrine factors such as nitric oxide or prostacyclin, the action of neuroeffectors, or direct responses of the vascular smooth muscle to physical stimuli such as those provided by a change in transmural pressure.

With respect to the local control of neural blood flow, it has been suggested that intrinsic autoregulatory ability is minimal and control is provided through the activity of perivascular nerves. The presence of nerves associated with the vasa nervorum has led to the view that nerve blood flow is predominantly under neurogenic influences (1, 3, 18, 27, 31, 35). These studies have largely used immunohistochemistry to identify the presence of neurotransmitters and have not directly examined dynamic aspects of local blood flow control. In studies in which blood pressure was manipulated (21, 32), it was concluded that peripheral nerve arterioles lack intrinsic autoregulatory ability; however, systemic approaches, for example, exsanguination and angiotensin II infusion, were used to alter intravascular pressure. Thus it is difficult to interpret such studies, because the methods used not only alter local hemodynamics but would be expected to stimulate a number of systemic responses.

A potential difficulty in studying local regulation of neural microvascular blood flow is the necessarily invasive nature of many of the approaches taken to measure hemodynamics. It is conceivable that surgical exposure of peripheral nerves results in the release of vasodilator compounds that mediate a hyperemic response, and this in turn may inhibit the expression of an intrinsic autoregulatory capacity. This argument has been used to explain apparent differences in nerve blood flow measured by H₂ polarography and distribution of radioactive microspheres (4, 11).

The significance of an understanding of local blood flow regulation in peripheral nerve relates to an increasing number of studies implicating a role for hemodynamic factors in the pathogenesis of neuropathy. There is growing support, for example, for the idea that ischemia contributes to the development of peripheral diabetic neuropathy in both humans and experimental animal models (10, 33). From the data specifically relating to neural dysfunction in diabetes it is difficult, however, to conclude whether changes in nerve blood flow are a primary factor in the pathogenesis of neuropathy or represent an adjustment to the altered metabolism.

The aim of the present study, therefore, was to develop approaches for examining vasoreactivity and, specifically, the autoregulatory capacity of rat sciatic nerve epineurial arterioles. Single isolated arterioles were studied to examine myogenic responsiveness in the absence of circulating and parenchymal factors and to allow dose-response relationships for vasoactive agents to be examined under conditions of constant pressure and zero flow. In parallel with this approach, in vivo studies were performed using a laser-Doppler flow probe positioned on epineurial vessels similar to those used in the in vitro studies. In these latter studies, particular attention was paid to time-dependent attainment of vascular tone and responsiveness to vasodilators.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All studies used male Sprague-Dawley rats (250-300 g, n = 52). Before use, animals were housed in a dedicated animal facility with controlled environmental conditions and were allowed free access to standard rat chow and drinking water. All protocols and procedure were approved by the RMIT University Animal Care and Use Committee.

Rats were anesthetized with thiopental sodium (100 mg/kg ip, Abbott Laboratories). For in vivo experiments, supplementary doses (10-20% of initial dose) of thiopental sodium were administered as required to maintain anesthesia.

In vitro isolated arteriole preparation. The sciatic nerve was exposed through an incision in either the right or left flank. A segment of nerve, distal to the sciatic notch, was removed with its supply vessels and placed in a cooled (4°C) buffer (in mM: 3 3-(N-morpholino)propanesulfonic acid, 145 NaCl, 5 KCl, 2.5 CaCl₂, 1 MgSO₄, 1 NaH₂PO₄, 0.02 EDTA, 2 pyruvate, and 5 glucose and 1% albumin)-filled dissection chamber. A segment of epineurial arteriole (Fig. 1) was cleared of adherent tissue and cannulated on glass micropipettes (40 µm) that were positioned in a superfusion chamber. Similar methods have been used for a variety of arterioles including porcine endocardium (19), rat skeletal muscle (36), and hamster cheek pouch (5). The chamber was then placed on the stage of an inverted microscope (Nikon Diaphot), vessels were pressurized in the absence of intraluminal flow by adjusting the height of a reservoir attached to one cannulation pipette, and responses were studied by video microscopy (final magnification ×760). During the studies vessels were superfused (2-4 ml/min) with a physiological buffer solution containing (in mM) 111 NaCl, 25.7 NaHCO₃, 4.9 KCl, 2.5 CaCl₂, 1.2 MgSO₄, 1.2 KH₂PO₄, 11.5 glucose, and 10 HEPES; the buffer was continually gassed with 95% N₂-5% CO₂.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Arterioles utilized in both in vitro and in vivo studies. Dotted circle represents typical site from which arterioles were isolated.

In vivo preparation. Rats underwent tracheostomy and intubation with polyethylene tubing (PE-240, Intramedic, Becton Dickinson) to ensure a patent airway; the left carotid artery was cannulated (PE-50) for measurement of arterial pressure. Body temperature was measured via a rectal probe and was maintained at 36-37°C with a heating pad placed under the body of the rat. The right sciatic nerve was exposed as described in In vitro isolated arteriole preparation. Adherent fat and connective tissues were gently removed from the epineurial supply vessels as shown in Fig. 1. The musculature surrounding the nerve formed a well into which fresh physiological buffer solution was continually superfused (4 ml/min, 34-36°C) and aspirated via a pump. A laser-Doppler probe (Moor Instruments) with a tip diameter of 0.8 mm was positioned above the epineurial arteriole using a micromanipulator capable of positioning in three dimensions. Positioning of the probe was done under a dissecting microscope to ensure consistency between preparations. Blood flow recordings, measured in arbitrary cell flux units, were digitized and stored using a MacLab A-D system.

Statistical methods. Results are expressed as means ± SE. Comparisons between groups were performed by ANOVA, with the level of significance calculated by the Bonferroni correction. Results from the in vitro studies, including calculated EC₅₀, were tested for significance using unpaired t-test. P < 0.05 was taken to represent statistical significance.

Chemicals. Unless otherwise stated, all chemicals were purchased from Sigma. With the exception of indomethacin and norepinephrine, all drugs were dissolved in distilled water and diluted in physiological buffer solution. Indomethacin was dissolved in distilled water containing 0.1 M Na₂CO₃ and norepinephrine in 1 mM ascorbic acid.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In vitro studies. After an equilibration period of ~60 min the epineurial arterioles showed sustained pressure-dependent vasoconstriction and acute myogenic reactivity (Fig. 2A) over the pressure range of 40-90 mmHg. Below 30 mmHg, arteriolar myogenic activity was partially obscured by passive collapse. Removal of extracellular Ca²⁺ resulted in a total loss of spontaneous tone and passive behavior [Fig. 2A; P < 0.001 (ANOVA), n = 5-10 vessels]. For example, at 60 mmHg the average diameter of the arterioles was 55 ± 6 µm in the presence of extracellular Ca²⁺ compared with 84 ± 6 µm in the absence of Ca²⁺ (P < 0.05), equating to a basal level of tone of ~38% at a physiological pressure.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   A: pressure-dependent vasoconstrictor responses over range of 5-90 mmHg () and passive (0 Ca2+) changes in internal diameter (). y-Axis represents arteriolar internal diameter as ratio of active diameter to passive diameter at 60 mmHg (d/d60). B: pressure-dependent vasoconstrictor response () is not altered during superfusion with phentolamine (106 M; ) over luminal pressure range (x-axis) of 20-100 mmHg. Passive distension in absence of Ca2+ is also shown (, n = 3 vessels).

6

6

6

9

6

7

9

6

6

4

5

6

View larger version (15K): [in this window] [in a new window]   Fig. 3.   A: concentration-response curve for norepinephrine (109-106 M). C, vessel diameter before application of norepinephrine (n = 5). B: concentration-response curve to acetylcholine (109-106 M) in absence (; n = 6 vessels) and presence () of NG-nitro-L-arginine methyl ester (L-NAME, 3 × 105 M; n = 4 vessels).

In vivo studies. When initially set up, epineurial arterioles showed a cell flux level of 860 ± 29 (arbitrary flux units; n = 7 vessels). Under these conditions vasoconstrictor responses to norepinephrine were observed, whereas responsiveness to acetylcholine was minimal, suggesting that the vessels were maximally dilated. In contrast, acetylcholine produced vasodilator responses after preconstriction with norepinephrine. Over a period of ~2 h, the baseline cell flux level declined to 266 ± 23 flux units [significantly less (P < 0.001) than flux at 0 h]. In a separate group of rats, the preparations were continuously superfused with a mixture of acetylcholine (1 µM) and adenosine (100 µM) over the 2-h period to confirm that the decline in blood flow was not caused by a nonspecific drift of the flux signal. In these experiments, there was a small but significant decrease in epineurial flow from an initial level of 825 ± 21 to 658 ± 36 flux units (n = 3 vessels), possibly because of decay of the drugs over the 2-h period. Coincident with the time-dependent decline in blood flow, arterioles exhibited vasodilator responses to acetylcholine without a requirement for preconstriction (Fig. 4). Group data showing the time-dependent decrease in baseline flow and the development of acetylcholine responsiveness are shown in Fig. 5. Mean arterial pressure remained constant (initial: 126 ± 10, 2 h: 138 ± 4 mmHg; not significant), whereas neural blood flow decreased, suggesting that the reduction in blood flow did not result from a change in systemic hemodynamics.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Representative tracing showing laser-Doppler flux measurements in epineurial arterioles of anesthetized rats. Channel 1, blood flow measured in arbitrary flux units (1 V = 100 flux units); channel 2, blood pressure in mmHg. A: baseline flow and responses to acetylcholine measured immediately after surgical exposure of vessel [time (t) = 0 h]. B: baseline flow measured 2 h after surgery (t = 2 h). Acetylcholine was applied as indicated by numbered boxes (3 and 17, 107 M; 4 and 18, 106 M).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Time-dependent increase in responsiveness to acetylcholine. * Significant increase in epineurial blood flow from baseline (P < 0.05, n = 5 vessels) in response to acetylcholine (105 M) at t = 2 h.

7

8

6

10

8

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Effects of phentolamine (107 M; A) and neuropeptide Y (NPY) fragment 18-36 (109 M; B) on baseline epineurial blood flow. x-Axis represents time in hours after surgical exposure of nerve and epineurial vessels. Open bars, epineurial blood flow immediately after surgery; hatched bars, epineurial blood flow 2 h after surgery in absence or presence of antagonist. At t = 2 h, phentolamine caused a partial reversal of decrease in blood flow (n = 6 vessels) whereas NPY fragment 18-36 had no effect (n = 6 vessels).

10

8

7

5

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Concentration-response curves for substance P (n = 5) and adenosine (n = 5) in epineurial vessels in vivo.

5

5

View larger version (40K): [in this window] [in a new window]   Fig. 8.   Effects of indomethacin (2.8 × 105 M; n = 4 vessels) and L-NAME (5 × 105 M; n = 4 vessels) on baseline epineurial blood flow. Open bars, epineurial flow immediately after surgery (n = 8 vessels); hatched bars, epineurial blood flow before or after infusion with indomethacin and L-NAME; cross-hatched bars, time control experiments. ** Significant difference (P < 0.001) compared with all other groups.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of these studies indicate that single epineurial arterioles respond to mechanical and chemical stimuli in a manner similar to that of arterioles from other vascular beds. To our knowledge, these studies provide the first data from isolated epineurial arterioles demonstrating inherent myogenic reactivity. The development of spontaneous tone and the demonstration of acute myogenic reactivity indicate that these vessels possess autoregulatory capacity and therefore have the potential to participate in the local regulation of neural hemodynamics, independently of their innervation. This is in contrast to earlier studies suggesting that nerve blood flow is controlled through the innervation of epineurial and endoneurial arterioles.

An important question is whether the vessel isolated from the surface of the sciatic nerve functions as a true resistance vessel or simply as a conduit artery that acts to distribute blood to the endoneurium. Care was taken during dissection to select vessels that could be seen to penetrate the interior of the nerve and were consistent with those described as epineurial vessels in previous studies (3, 18). The diameter of the vessels studied (passive diameter ~100 µm) is similar to that of feed arterioles of a number of microvascular beds of the rat, for example, cremaster (36) and spinotrapezius muscles (6). Vessels from these tissues have been shown to possess intrinsic myogenic reactivity under in vitro conditions and hence have been implicated in the local control of vascular resistance. By analogy with these other tissues, together with the functional behavior observed in the present study, it is likely that epineurial arterioles function as true resistance vessels and contribute to the local control of intraneural hemodynamics.

A complicating factor in examining the determinants of local neural blood flow in vivo has been the necessarily invasive nature of the preparations that have been employed. In recent studies Williamson and colleagues (4, 11) suggested that surgical exposure of nerves results in hyperemia that would artificially raise baseline blood flow measurements and obscure local intrinsic vasoregulatory mechanisms. In support of this suggestion, Kinoshita and Monafo (15) reported that the trauma associated with surgical exposure elevates neural blood flow and reduces responsiveness to norepinephrine. Although sciatic nerve blood flow has been measured previously by a number of other groups (see, e.g., Refs. 14, 24, 29, 34), there have been no reports of a time-dependent decline in blood flow similar to that observed in the present study. The finding that epineurial blood flow, as measured using a laser-Doppler probe, decreases with time after surgical exposure of the nerve is, however, consistent with the development of vascular tone. This could conceivably result from the gradual removal or washout, by continuous superfusion of the preparation, of dilator substances released during surgery. In support of this suggestion, it was observed that a combination of cyclooxygenase and nitric oxide synthase inhibition reduced blood flow in the postoperative period.

Alternatively, constrictor substances may be released from perivascular nerves; however, this seems unlikely because the isolated arteriole preparation also showed time-dependent development of tone. The finding that neither the spontaneous tone nor acute in vitro myogenic reactivity was abolished by the -adrenoceptor antagonist phentolamine indicates that the tone is not adrenergic in nature. This is further supported by the observation that the same dose of phentolamine significantly inhibited responses to exogenously administered norepinephrine.

Of interest was the finding that phentolamine had no effect on the level of spontaneous tone of the in vitro preparation but caused an increase in epineurial flow in the in vivo preparation. Because the increase in flow was submaximal, the data suggest that under in vivo conditions epineurial arteriolar tone results from a combination of intrinsic and neural influences. This is further supported by the finding that TTX did not completely prevent the time-dependent decline in epineurial blood flow. Alternative explanations for the differential effect of phentolamine under in vitro and in vivo conditions are that 1) arteriolar tone increases in response to circulating catecholamines or factors that affect the local release of norepinephrine or 2) arterioles downstream of the site of blood flow measurement have a greater dependence on -adrenergic mechanisms for the development of tone. In regard to the former, studies have shown that topically applied epinephrine causes a marked reduction in neural blood flow consistent with vasoconstriction of epineurial arterioles (24). In contrast, the latter explanation seems unlikely because arterioles often show a decreasing dependency on extrinsic innervation, as diameter decreases, for vasomotor control (30). In addition to examining the effect of -adrenoceptor blockade on epineurial tone, we examined the effects of blockade of the cotransmitter NPY (22), using NPY peptide fragment 18-36. Despite evidence for exogenous NPY acting as a potent vasoconstrictor, and its actions being blocked by the inactive peptide fragment, no evidence was found for this neuropeptide in the development of arteriolar tone.

An alternate explanation for a time-dependent decrease in neural blood flow could be the gradual formation of an endoneurial edema. Myers et al. (25) suggested that because the perineurial sheath is a relatively indistensible structure, the presence of edema within a nerve would raise intraneural pressure that in turn would compress endoneurial blood vessels and restrict blood flow. Kalichman and Myers (13) further demonstrated that procaine-induced endoneurial edema resulted in constriction of transperineurial arterioles. This would not explain the time-dependent decrease in epineurial blood flow observed in the present study, because a decrease in blood flow caused by a biomechanical effect would not be associated with an increase in acetylcholine responsiveness.

It has also been suggested that anesthesia has a marked effect on peripheral nerve blood flow, depressing vascular responsiveness. In relation to this, it is of interest to note that endoneurial blood flow (measured by laser Doppler) increases in decerebrate rats in response to hypercarbia, suggesting vasodilatation of the epineurial and/or endoneurial arterioles. For reasons stated above, it is difficult to interpret such data, however, because inspiration of CO₂ would be expected to invoke both local and extrinsic regulatory mechanisms. Anesthesia per se cannot be said to completely inhibit local neural blood flow regulatory mechanisms because our in vivo studies were performed under thiopental sodium anesthesia. As with other microvascular preparations, it is conceivable that reactivity is differentially inhibited by various anesthetic agents (20). A possible mechanism by which anesthesia could inhibit local autoregulatory mechanisms is impairment of body temperature regulatory mechanisms. Thus Dines et al. (7) reported that decreased temperatures are associated with increased neural blood flow. In the present studies, specific attention was given to maintaining core temperature at 37°C and temperature of the sciatic nerve superfusion solution at 34-36°C.

The finding that isolated epineurial arterioles constricted to the nitric oxide synthase inhibitor L-NAME suggests that a basal release of nitric oxide occurs independently of shear stress or neurohumoral stimuli. This is consistent with a number of studies of the effects of nitric oxide synthase inhibitors on isolated arteriole preparations (see, e.g., Refs. 9, 26). Alternatively, L-NAME may be exerting effects in addition to nitric oxide synthase inhibition, such as blockade of K⁺ channels (17). The rightward shift in the acetylcholine dose-response relationship indicates that at least a portion of the epineurial arteriole dilator response is caused by stimulation of nitric oxide production. Further studies are required to fully characterize the role of the endothelium in epineurial arteriole vasoreactivity and the contribution of paracrine agents including prostaglandins and endothelium-derived hyperpolarizing factor.

With respect to the existence of local microvascular regulatory mechanisms in other neural tissue, it is well known that the cerebral microcirculation exhibits marked autoregulation of blood flow in response to metabolic stimuli and alterations in perfusion pressure (16). In recent studies, McManis et al. (23) extended these observations to show that blood flow in superior cervical and dorsal root ganglia is unchanged despite alterations in perfusion pressure by acute volume expansion or contraction. Interestingly, the same authors found that blood flow in the sciatic nerve varied with the changes in perfusion pressure. Similarly, Rechthand et al. (28) reported that the dilatory response to hypercarbia was more marked in cerebral vessels compared with that in peripheral nerve. It is unclear why the responses differ between neural tissues and whether this relates to methodological issues or differing metabolic requirements or perhaps reflects the need to confer a survival advantage to particular tissues.

In summary, the present studies indicate that epineurial arterioles exhibit spontaneous myogenic tone and, in addition, suggest that these vessels may contribute to autoregulatory responses and the local control of hemodynamics. The approaches described may provide avenues for investigation of the role of the microvasculature in the development of common neuropathies such as those associated with diabetes and nerve compression.