Investigators report that local heat causes an increase in skin blood flow consisting of two phases. The first is solely sensory neural, and the second is nitric oxide mediated. We hypothesize that mechanisms behind these two phases are causally linked by shear stress. Because microvascular blood flow, endothelial shear stress, and vessel diameters cannot be measured in humans, bat wing arterioles (26.6 ± 0.3, 42.0 ± 0.4, and 58.7 ± 2.2 μm) were visualized noninvasively on a transparent heat plate via intravital microscopy. Increasing plate temperature from 25 to 37°C increased flow in all three arterial sizes (137.1 ± 0.3, 251.9 ± 0.5, and 184.3 ± 0.6%) in a biphasic manner. With heat, diameter increased in large arterioles (n = 6) by 8.7 ± 0.03% within 6 min, medium arterioles (n = 8) by 19.7 ± 0.5% within 4 min, and small arterioles (n = 8) by 31.6 ± 2.2% in the first minute. Lidocaine (0.2 ml, 2% wt/vol) and NG-nitro-l-arginine methyl ester (0.2 ml, 1% wt/vol) were applied topically to arterioles (∼40 μm) to block sensory nerves, modulate shear stress, and block nitric oxide generation. Local heat caused only a 10.4 ± 5.5% increase in diameter with neural blockade (n = 8) and only a 7.5 ± 4.1% increase in diameter when flow was reduced (n = 8), both significantly lower than control (P < 0.001). Diameter and flow increases were significantly reduced with NG-nitro-l-arginine methyl ester application (P < 0.05). Our novel thermoregulatory animal model illustrates 1) regulation of shear stress, 2) a nonneural component of the first phase, and 3) a shear-mediated second phase. The time course of dilation suggests that early dilation of small arterioles increases flow and enhances second-phase dilation of the large arterioles.
- nitric oxide
- in vivo
- shear stress
human skin responds to local temperature increases with a biphasic increase in skin blood flow (SkBF). Within the first few minutes after the local application of heat, SkBF reaches a peak and then begins to decline. Three to five minutes after heating, SkBF reaches a nadir and then increases once again (8, 16, 26). The second slower and more sustained response typically rises above the initial peak value. Typically, SkBF is measured noninvasively in human subjects using laser Doppler flowmetry (LDF), which ensures that the complex interaction of thermoregulatory mechanisms controlling vascular responses remains intact. The resulting behavior is ubiquitous and consistent (8, 39). Investigators universally separate the response to local heat into two distinct phases (16, 26).
Investigators believe that these two phases are controlled by two mechanisms (26). When a local anesthetic is used to block sensory input, the first phase of the biphasic flow response is abolished, suggesting sensory neural-mediated microvascular dilation. The second phase is diminished with the addition of nitric oxide (NO) synthase (NOS) inhibitors (8, 16, 26), suggesting NO-dependent microvascular dilation. A direct link between the mechanisms of the neural- and NO-mediated increases in flow has yet to be established; however, shear stress has been postulated to play a role in the biphasic vasodilatory response (16, 26). Although human studies are carried out in an intact, unanesthetized, thermoregulatory vascular bed, this common experimental model does not allow direct measurement of diameter nor does it allow for calculation of arteriolar blood flow or endothelial shear stress.
In animal models allowing direct measurement of microvascular diameter, investigators have repeatedly shown that endothelial shear stress (i.e., endothelial frictional force) results in the production of NO (18, 32). For instance, flow autoregulation depends primarily on two interacting mechanisms: metabolic demand and endothelial shear stress. An increase in metabolites dilates the resistance vessels, increases flow, and meets the increased requirement for oxygen and nutrients. The resulting increase in blood flow through the larger upstream conductance vessels increases endothelial shear stress. The attending increase in endothelial shear stress produces NO, which provides a secondary dilation and increase in flow (20). With NO scavengers and synthase antagonists, the secondary dilation of conductance vessels is inhibited, and skin perfusion is lessened (15, 19, 21). Thus NO is not directly produced by the primary stimulus. Rather, NO production is a secondary effect, resulting from altered hemodynamics. Studying shear-mediated dilation in human SkBF requires an intact neural stimulus (lacking in ex vivo studies) and the ability to record individual vessel diameters (lacking in LDF studies).
The need to relate vascular function to specific changes in vascular structure led to the development of a unique animal model (37, 38). The Pallid bat wing is easily transluminated, allowing investigators to noninvasively measure vessel diameter and blood velocity (9, 12). Investigators have also exploited the thinness of the wing to deliver drugs topically (6, 22, 25). This model has been used to study the microvascular responses to various stimuli in an intact system uncompromised by surgical trauma or general anesthesia. For instance, arterioles have been found to dilate with decreased pressure (the myogenic effect) (12, 38), de-innervation (37), and increased metabolic factors (1, 5). Previous work has highlighted the bat wing as a thermoregulatory organ, and investigators have studied the responses of interstitial fluid pressure and vasomotion to altered internal or local ambient temperatures (7, 17, 33). These researchers show that heat increases venomotion and interstitial fluid pressure in the bat wing. However, no studies of arterial dilation resulting from local heat in the bat wing have been reported, and pervasive interest in NO-mediated regulatory mechanisms arose after the bat wing model fell out of use in the late 1980s (11).
The present work aimed to examine SkBF and shear-mediated responses in the Pallid bat wing to local heating with local anesthesia and NOS inhibition. We hypothesize that the two phases of the biphasic response to local heat are interrelated, and endothelial shear stress is the fundamental link between them.
MATERIALS AND METHODS
Animal care and all experimental procedures were performed in compliance with protocols approved by the Texas A&M Institutional Animal Care and Use Committee. Two colonies of adult, wild Pallid bats (Antrozous pallidus) were caught using mist nets set at drinking and foraging sites or collected by hand from roost sites in a building in west Texas in August 2003 and August 2004. Each colony was held in quarantine for 6 mo after capture. Bats were marked with hair dye for proper identification. The 25 adult bats were kept in two separate rooms, according to the colony with which they were captured. Bats were fed crickets, mealworms, and wax worms once daily and were given free access to water. The light-dark cycle of the room was set at 6 h on and 18 h off, with the light cycle beginning at 10:00 AM and ending at 4:00 PM. The dark cycle (4:00 PM through 10:00 AM) coincided with feeding times for the animals. Bats were then trained to remain sleeping, unanesthetized, inside a plastic box with one wing extended outside of the box (10, 38). Since male bats were resistant to training, only female bats were studied. An established experiment schedule ensured that each bat was used for experiments no more than once per week and for no more than 6 h per experiment.
With a few notable exceptions, experimental procedures were similar to those employed by Davis et al. (9, 10, 12). Bats were placed in a plastic box attached to a heat plate connected with temperature controller (Olympus Tokai Hit Thermo plate, 2004). The wing was visualized with an intravital microscope (Olympus BX61WI Fixed Stage Upright Microscope, 2003), and the image was relayed via camera (Panasonic KR222 S-Video Camera, 2004) to both a DVD player (Sony DVD Recorder RDR-GX7, 2005) and computer (Dell, Dimension 4600 Series, 2005). The computer analyzed and recorded the image with a custom-built diameter tracker software (LabView 7.1) (24). Additionally, an optical Doppler velocimeter (Optical Doppler Velocimeter model no. 4, A&M Health Systems, 2003) was attached to the microscope to relay centerline red blood cell velocity to the computer. Arteriolar diameter and velocity were recorded at 30 Hz.
The bat laboratory room was kept at a constant 24°C at all times. For all experiments, the bats were placed under the intravital microscope, and the thermoregulator of the heat plate was set at 25.0°C. The region of the wing touching the heat plate was restricted to the specific area of study (no larger than 3 cm2). Bat wing and abdominal fur temperature was recorded at the end of every minute using a laser thermometer (Raytek Raynger ST Pro, 2003). A baseline period of 10 min was established at a heat plate temperature of 25.0°C. The heat plate temperature was then increased to 37.0°C for 20 min, the highest temperature increase that does not wake sleeping bats (36). Plate and batwing temperature stabilized at 37.0°C over a period of 60 s, comparable to rates used in previous human SkBF studies (16, 26).
In the first series of experiments, we examined the vascular response to local heat in one long, continuous vascular structure containing one major, visible bifurcation on the lateral portion of the wing (Fig. 1). We studied large arterioles (∼60 μm, n = 6) just medial to the fourth metacarpal bone, medium arterioles (∼40 μm, n = 8) just proximal to a large bifurcation, and small arterioles (∼25 μm, n = 8) just distal to the major bifurcation but above any other bifurcations.
The second series of experiments involved the use of lidocaine (0.2 ml of 2% gel) applied on the back and front of the wing on a 2-cm2 area surrounding an arteriole ∼40 μm in diameter and proximal to any major bifurcations. Arteriole diameter and blood velocity inside the area of lidocaine application (n = 8; Fig. 1, “X”) and 1 cm upstream of lidocaine application (n = 8; Fig. 1, arrow) were recorded.
Lidocaine efficacy was ensured by touching the wing with the tip of a cotton swab. A lack of response from the bat indicated local sensory nerve blockade. Conversely, the bats did respond when the area without lidocaine was touched.
In a third series of experiments, we added a 1% (wt/vol) NG-nitro-l-arginine methyl ester, hydrochloride (l-NAME, Calbiochem, 2005) solution to the wing before heating. The solution consisted of 1.0 mg l-NAME in 100 μl of a 1:1 diH2O + dermabase (Paddock Laboratories, 2004) emulsion. The mixture was spread on the wing at location X (Fig. 1), and the heating protocol was carried out as described above.
We performed an arterial occlusion of the vasculature to severely reduce flow and thus endothelial shear stress at location X (Fig. 1). At the conclusion of the 20-min heating period, flow was significantly limited with the use of a finger of a flour-filled nitrile glove pressed onto the vessels of the surface of the wing for 1 min. We recorded, and reported, second-to-second diameter values to demonstrate the immediate effect of reduced flow during the second phase.
In each experiment, blood flow and endothelial shear stress were calculated using the final 10 s for each minute of data, with the assumption that endothelial shear stress = 8·flow·viscosity/radius3, consistent with a Poiseuillian flow approximation (32). Viscosity was assumed to be 3.5 cP, consistent with reported values (4). The final 10 s of each minute of flow, endothelial shear stress, and diameter were averaged for each bat. Each 10-s average per experiment was then averaged across the entire experimental subject population for that particular time.
Time to initial peak diameter was calculated for each arteriole segment per bat and then averaged per arteriole group. An initial peak was defined as an initially elevated diameter followed by two consecutive lower values.
Data are presented as mean values ± SE of the mean. Student's t-test and repeated-measures ANOVA tests were then performed on each of the vessel groups based on vessel size and orientation with respect to drug application. The parameters of diameter, flow, and shear stress were all analyzed at the various time intervals using vessel size as the dependent, or tested, variable. For the first set of experiments, we separately compared diameter, flow, and shear stress for the three vessel sizes at the specific time intervals established. For the lidocaine experiments, we also analyzed diameter, flow, and shear stress comparing vessels at the site of lidocaine application and upstream of lidocaine application to control vessels without lidocaine. We grouped the timing of the experiment into three groups: 1) first minute of heating, 2) first phase (first 4 min), and finally 3) the second phase (last 10 min). The presence and site of lidocaine application were used to group the dependent variables. The Student's t-tests and repeated-measures ANOVA tests were then used to compare the vessel groups for all of the experiments. Significance was established at P < 0.05. Fisher's least significant difference post hoc test was then used to find differences among the groups.
Local Heat in Absence of Other Interventions
Figure 2 illustrates increases in blood flow, endothelial shear stress, and diameter with local heat. Single-arteriole blood flow (Fig. 2B) increased in all three arteriole sizes and increased in a biphasic manner in the large (∼60 μm) and medium-sized (∼40 μm) vessels (Fig. 3). Flow increased 144 ± 25.8% in large vessels and 303.6 ± 52.9% in medium-sized vessels in the first minute. No nadir in flow could be detected in the small arterioles (∼25 μm) (and therefore they lacked a biphasic response).
We compared the percent increase in flow values for all three vessel sizes in the first minute, fifth minute, and the fifteenth minute of heat application (Fig. 3). We found a statistically significant difference between increases in flow values at minute 1 and those at minute 5 (P = 0.009). Additionally, there was a difference between flow increases between minute 5 and minute 15 (P = 0.013). There was no difference between flow values at minute 1 and minute 15 (P = 0.886). From this analysis, we can conclude that there is a statistically significant biphasic response in the large microvessels in the bat wing with local heat application.
Endothelial shear stress initially spiked and then receded to steady-state values in both large and medium arterioles (Fig. 2C). Large-vessel shear stress increased 62.8 ± 26.4% in the first minute, while medium-vessel shear stress increased 174.7 ± 55.7%. Both large and medium vessels had elevated steady-state values in the final 10 min of local heat (+63.9 ± 16.4 and +78.5 ± 29.8%, respectively). In contrast, small vessels do not exhibit an increase in shear stress values within the first minute. Instead, endothelial shear stress peaked from minutes 5–11 of local heat application, increasing 43.8 ± 16.3% from baseline. Steady-state values for small-vessel shear stress were 32.5 ± 17.0% above baseline.
Small arterioles exhibited significantly greater dilation, and large arterioles dilated to a significantly less degree (Fig. 2D). Large vessels increased 5.4 ± 7.8% in diameter with an increase of 1.0 ± 2.7% in the first minute. Medium vessels increased 20.4 ± 7.1% in diameter with an increase of 17.5 ± 7.7% in the first minute. Small vessels increased 35.6 ± 7.3% in diameter, with an initial increase of 29.9 ± 5.7% in the first minute. Diameter increased in all three sized vessels; however, there was a lack of evidence of a clear, biphasic nature to the increase
Time to initial peak diameter was significantly longer in large and medium vessels, while peak diameter was reached sooner in the small vessels. Large vessels reached their initial peak diameter in 3.8 ± 0.5 min, whereas medium vessels initially peaked at 3.1 ± 0.6 min. Small vessels reached initial peak values at a significantly faster rate (1.6 ± 0.2 min) compared with both large and medium vessels.
Local Heat with Sensory Neural Blockade
The local application of lidocaine decreased the blood flow (Fig. 4A) response to local heat. Similar to control conditions, blood flow increased 117.3 ± 18.1% in the first minute and then slowly increased to a late-phase steady state (124.5 ± 24.6% from baseline). This increase in blood flow was not biphasic in nature and was lower than in the control group (P < 0.001).
With the local application of lidocaine, endothelial shear stress values initially increased and then decreased to a steady-state value elevated from baseline (10.1 ± 0.9 dyn/cm2; Fig. 4B). At the site of lidocaine application, endothelial shear stress increased 109.1 ± 21.0% in the first minute of local heat.
At the site of lidocaine application, diameter increased 6.5 ± 4.1% in the first minute. Final 10-min steady-state diameters increased at the site of lidocaine application 10.4 ± 5.5% from initial baseline. The initial and steady-state increases were significantly lower compared with a vessel of similar size without lidocaine (Fig. 4C).
Local Heat with Reduced Endothelial Shear Stress
Initial and steady-state blood flow in vessels upstream of the lidocaine application was lower than in control conditions (P < 0.001). Vessels upstream from lidocaine increased flow 115.6 ± 20.2% from baseline within the first minute. Late-phase steady-state flows in vessels upstream of lidocaine were increased 132.7 ± 22.7% from baseline (Fig. 4A). Reduced endothelial shear stress did not eliminate the biphasic shape of the curve.
Endothelial shear stress values upstream from the site of lidocaine application increased 110.9 ± 22.0% from baseline. The initial increase in endothelial shear stress at the site of lidocaine or upstream was not statistically different from the control. Endothelial shear stress also reached a steady-state elevated from baseline in the final 10 min of the local heat application (+106.7 ± 11.8%), which was not statistically different from control (Fig. 4B).
Vessels upstream of lidocaine initially increased 7.3 ± 5.5% in diameter, while a steady-state value was established 7.5 ± 4.1% above baseline. The increase in initial and steady-state diameter was significantly lower compared with control vessels but not from vessels at the site of lidocaine application (Fig. 4C).
Local Heat with Sensory Neural Blockade and Reduced Endothelial Shear Stress
To remove the neural and endothelial shear effects, we occluded a vessel in the presence of lidocaine at the conclusion of 20 min of local heat. Diameter immediately (within seconds), and significantly, decreased 21.6 ± 5.4% with a similar, instantaneous decrease in blood flow velocity (Fig. 5). Diameter was reduced with occlusion in three out of three experiments.
Local Heat with NOS Inhibition
To remove the effects of NO on the vasculature, we utilized a 1% (wt/vol) lidocaine solution on the bat wing vasculature in the presence of increased local temperatures. We show a significant reduction in dilation and flow with l-NAME application compared with control vessels. Additionally, blood flow increases in the first phase were reduced compared with increases seen in the second phase. Mean red cell velocity and shear stress were not significantly reduced with l-NAME (Fig. 6).
Our results suggest that local heat produces a shear-mediated biphasic response in the thermoregulatory microcirculation of the Pallid bat wing. Our experimental results in the bat wing suggest that neural- and shear-dependent mechanisms initially dilate the smaller (25–40 μm) vessels; the resulting increase in flow and thus endothelial shear stress in the larger (∼60 μm) arterioles enhances their dilation. The elevated endothelial shear stress in bat wing vessels presumably affects the second phase of the biphasic response through the well-documented, shear-mediated production of NO (15, 20, 32). With the blockade of NO synthesis, both first- and second-phase dilations were inhibited. This is the first work to document a causal link between sensory-neural mechanisms and mechanisms known to produce NO in a mammal and therefore challenges investigators to reevaluate the interpretation of data from human studies.
Endothelial Shear Stress Is Regulated in the Bat Wing
A primary and important finding from this study suggests that sheer is a regulated variable during local heating in this model. While shear regulation has been well documented in a variety of preparations (15, 18–20, 32), this is one of the first demonstrations of this phenomenon in a thermoregulatory microvasculature. Dilation and constriction can adjust endothelial shear stress independent of a given blood flow (34), and, indeed, shear is believed to be regulated in commonly studied microvascular beds (20, 32). Three pieces of evidence suggest that endothelial shear stress is regulated in the skin microvasculature. 1) The initial spike in endothelial shear stress precedes dilation, which then partially restores the shear stress to initial levels (Fig. 2C). 2) Endothelial shear stress values reach a consistent, steady-state value that is the same with and without lidocaine (Fig. 4B), even though there are strikingly different levels of blood flow through the vessel (Fig. 4A). 3) Reducing the neural input and blood velocity (Fig. 5) instantaneously decreases diameter. This evidence suggests that endothelial shear stress in the bat wing vasculature is regulated, and thus shear-mediated vasodilation is a fundamental component of the response to local heat.
The First Phase of the Biphasic Response Is NO and Shear Dependent in the Bat Wing
With NOS inhibition, there is a significant reduction in diameter and flow in the first phase of the biphasic response during local heat application. This provides evidence for the importance of shear-generated NO to the first phase of the biphasic response. It can be shown from fundamental hemodynamic principles (32, 34) that blood velocity can remain constant, if both blood flow and vessel diameter increase simultaneously. In fact, flow-induced shear dilation tends to maintain constant blood velocity in other mammals (20, 32, 34). Our results are consistent with these observations (Fig. 2), since late-phase increases in diameter and flow are accompanied by highly regulated, steady-state shear levels. From our results, we can conclude that there is a slight sensory neural influence and a large shear-generated, NO-mediated dilation in the thermoregulatory circulation of the Pallid bat wing.
The Second Phase of the Biphasic Flow Response Is Shear-Mediated in the Bat Wing
The vascular response to local heat in the Pallid bat wing can potentially be ascribed to a variety of mechanisms, including local sensory nerves, endothelial shear stress, the direct effect of heat on the vascular smooth muscle, metabolic factors, the myogenic response, and conducted stimuli along the vessels. Of these factors, sensory neural stimulation, shear stress, and metabolic factors could cause vessels to dilate with local heat (7, 8, 18). A myogenic response would cause a constriction with increased pressure accompanying heating (7, 10), and the direct effect of heat on vascular smooth muscle has been noted to cause a constriction in some vascular beds (23). A conducted stimulus would cause a constriction or dilation based on the predominant reaction of the vessel adjacent to the studied vessel (14). Our unique animal model allows us to differentially stimulate these various mechanisms to identify the primary influence on the second phase of the biphasic flow response.
Local heat and neural blockade.
Lidocaine was used for sensory neural blockade and is a local anesthetic that acts by blocking fast sodium channels. This inhibits action potential propagation in C-afferent, sensory nerve fibers (35). In these experiments, there was an increase in flow and diameter with heat, although both were less than the response to heat in the control case (i.e., without neural blockade) (Fig. 4). These results are not consistent with the effects of the myogenic response, metabolic factors, a conducted stimulus, or the direct effect of heat on vascular smooth muscle, which would all cause more dilation than the control case. However, because a reduction in endothelial shear stress is accompanied by a reduction in dilation (as compared with the control case), this response is consistent with a shear-mediated dilation, independent of neural influence.
Local heat and reduced endothelial shear stress.
To limit the influence of endothelial shear stress resulting from local heating, we recorded the response of the vessel segment upstream of lidocaine application. In this location, the flow was reduced (as compared with the control case), but the sensory neural response remained intact. In these experiments, the increase in flow and diameter with heat were less than the response to heat in the control case. This diminished flow was accompanied by a reduced dilation (Fig. 4C). These results were not consistent with metabolic factors or a sensory neural effect, which both would cause more dilation than the control case. Differential dilation, despite similar temperatures, makes a direct effect of heat unlikely. Because the response to heat takes minutes (Fig. 2), these results are not consistent with the myogenic response, which occurs on the order of seconds (12). However, because a reduction in endothelial shear stress is accompanied by a reduction in dilation (as compared with the control case), this response is consistent with a shear-mediated influence on the second phase of the biphasic response in the bat wing.
Local heat, reduced endothelial shear stress, and sensory neural blockade.
We used lidocaine and a partial vascular occlusion to eliminate the influence of sensory nerves and strongly reduce the shear stress stimulus (Fig. 5). With both of these interventions, we saw an immediate constriction in diameter to near preheating values. These data are not consistent with responses to myogenic, metabolic, or conducted stimuli, since these stimuli would be expected to cause dilation. Because the temperature change was the same before and after intervention, these results are not consistent with the direct effect of heat on vascular smooth muscle. The drop in diameter immediately subsequent to a reduction in blood velocity (Fig. 5) is consistent with the hypothesis that endothelial shear stress plays a crucial role in the second phase of the biphasic response in the bat wing.
Local heat with NOS inhibition.
l-NAME, a NOS inhibitor, was used to eliminate NO-mediated dilation with local heat. In this set of experiments (Fig. 6), NOS inhibition resulted in a markedly reduced dilation with local heat, despite significantly elevated shear stress. From these data, we infer that NO generation via shear stress is essential for the second phase of the biphasic flow response in the bat wing. Because flow increased dramatically in the absence of local vasodilation, it is likely that vessels downstream of lidocaine application affected flow. Furthermore, our experiments that blocked either the neural or NO-mediated phases of the biphasic response did not completely inhibit the vascular flow response to local heat. It remains possible that there is an unknown factor that contributes, or even drives, the biphasic response to local heat.
In this series of five experiments, the results are consistent with a significant contribution of shear-mediated dilation to the second phase of the biphasic response. None of the five alternative mechanisms described above (local sensory nerves acting alone, endothelial shear stress, the direct effect of heat on vascular smooth muscle, metabolic factors, the myogenic response, nor conducted stimuli along the vessels) are consistent with the results of all five experiments (Figs. 4 and 5). There may be yet other alternative possibilities that have not been considered; however, they would have to be consistent with the results of these four disparate experiments to be considered a likely cause of the late-phase response in the bat wing.
Limitations of Bat Wing Studies
Our unique animal model introduces unavoidable methodological complications and possible interpretive limitations. For instance, sensory innervation may not have been entirely abolished with application of topical lidocaine. The lack of response to touch on the bat wing, however, is an accepted test of neural blockade in human studies (2). Given the extreme sensitivity of the bat to touch on its wing, it is likely that this test is more sensitive for bats than humans. Another limitation of our studies is that the partial vascular occlusion did not reduce endothelial shear stress in a controlled manner. By reducing flow below normothermic values, this intervention may have exaggerated the role of endothelial shear stress. Gathering systemic parameters such as body temperature and systemic mean arterial pressure during the experiments poses another limitation in dealing with live animal experiments. Our pilot studies revealed little change in bat fur temperature and/or heart rate with local heat. These findings are not surprising, as we are challenging only a fraction of the bat surface area and blood flow with local heat. Widespread interventions might be expected to cause global hemodynamic changes; this, however, is not the case in our particular experiments.
A particular limitation of the Pallid bat wing model is the inability to measure NO generation in response to endothelial shear stress. The small blood volume of the Pallid bat (1.5 ml) (28) does not allow sufficient, consecutive NO measurements without hemodynamically compromising the model. We instead chose to inhibit NOS. Previous studies by Davis (11) report that increased flow in isolated bat wing venules leads to increased diameter, and this dilation is lessened with NOS inhibitors. Additionally, our data indicate a similar, significant decrease in heat-induced dilation when NOS is inhibited. We interpret these data as evidence that the bat wing vasculature produces NO during periods of elevated shear stress.
While not a limitation of the model, we chose to use resting diameter, flow, and shear stress values as a baseline. Because the tone is different in different sized vessels, the best way to compare vessel responses to local heat is a percent change in diameter from resting tone at 25°C. Resting tone is the most appropriate baseline to characterize the dilation resulting from elevated shear stress in the microvasculature (3), when it is possible to obtain experimentally. This choice of baseline is notably different than that used for isolated vessel experiments or in situ preparations that may be influenced by trauma or anesthesia. Typically, these studies rely on maximal dilation as a point of comparison. Although we did not determine maximum diameter of the vessels, from the variation in diameter minute to minute, we can infer that the maximum capacity of the vessels to dilate had not been reached. The ability to consistently obtain normal parameters in an intact, in vivo, microcirculatory preparation is one noteworthy strength of our Pallid bat model.
Comparing Bat Wing to Human SkBF Studies
We have found a series of behaviors of the bat wing microvasculature that are analogous to those reported in human SkBF studies. First, the response of flow was biphasic in nature. Second, the first phase of the response was inhibited by lidocaine. Third, the second phase was inhibited by l-NAME. These novel findings in the bat wing could be used to offer essential insight into unanswered questions surrounding local heating and SkBF.
Despite these striking similarities, it would be inappropriate to infer that bat wing vasculature is completely analogous to human SkBF studies. The Pallid bat wing is structurally different. Given the surface area-to-mass ratio, the thermoregulatory capabilities are likely to be different than those of human skin. Furthermore, little molecular characterization has been performed on the Pallid bat wing, complicating comparison of vascular and interstitial components to human skin. Because we heated an extremely focused region of interest, fewer vessels are involved than conventional human SkBF studies. Nonetheless, the vascular regulation in the bat wing has been well characterized (7, 10, 37, 38), and this study leverages decades of research to establish the bat wing as a novel experimental model to study thermoregulation.
Possible Reinterpretation of Human SkBF Studies
In human SkBF studies, the first phase of the biphasic response is believed to be due to a sensory neural mechanism, and the second phase is NO dependent. Our results indicate that these two mechanisms may also be related in human SkBF. Investigators have asserted that the sensory nerves are solely responsible for the initial increase in vascular flow and diameter with local heat (8) in humans. Their interpretation is based on direct measurements of blood velocity using LDF. While our results apparently conflict with this interpretation, we directly measured different variables than those conventionally measured in human skin. Previous studies have demonstrated that topically applied lidocaine abolishes the primary increase in perfusion resulting from local heating. However, perfusion is indirectly inferred from LDF measurements of blood velocity, not from direct measurement of blood flow (8, 31). Furthermore, changes in vascular diameters are indirectly inferred from LDF velocity divided by mean arterial pressure, not from direct measurements of peripheral resistance. It is possible that blood flow and vessel diameter may actually increase in human studies after neural blockade, consistent with our results. The Pallid bat model, allowing simultaneous, direct measurement of blood velocity and vessel diameter, challenges the strength of the conclusions drawn from human studies.
Recent work has examined the possibility that local heat could be used as a noninvasive test for the severity of particular pathologies (27); however, inconsistent results have yielded few clinical applications (13, 29). Patients with spinal chord injuries and long-term, unmonitored diabetes, for instance, show altered neural portion of the biphasic responses to local heat (30). In our bat wing experiments, we demonstrate that a shear-mediated component to the initial phase of the biphasic response may be a confounding factor that limits the ability of first phase of the biphasic flow response to yield a meaningful index of intact neurovascular function. If shear dilation is a necessary component of a normal biphasic response in humans, local heating of skin may be a clinically useful screening test with high sensitivity for endothelial dysfunction.
The present work has 1) established the Pallid bat wing as a novel model for studying intact, thermoregulatory microvascular beds; 2) shown endothelial shear stress is regulated with local heat; 3) demonstrated that the first phase of the biphasic response is not entirely sensory neural in origin; and 4) provided evidence that endothelial shear stress is an integral part of the second phase of the biphasic flow response.
Portions of this work were supported by National Heart, Lung, and Blood Institute Grant K25 HL-070608, American Heart Association Grant 0365127Y, and Centers for Disease Control Grant CDC-620069.
We thank Dr. Christine Heaps for insightful comments.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society