- Email: [email protected]
Conducted Vascular Responses: Communication across the Capillary Bed
Microvascular Research 56, 43 – 53 (1998) Article No. MR982076
Conducted Vascular Responses: Communication across the Capillary Bed Diane M. Collins, William T. McCullough, and Mary L. Ellsworth Department of Pharmacological and Physiological Science, Saint Louis University Health Sciences Center, St. Louis, Missouri 63104 Received December 22, 1997
Conducted vasomotor responses are important for the effective distribution of blood flow, although the mechanism by which these responses are initiated is not well understood. ATP, a substance which is released from circulating red blood cells in response to low PO2 and low pH, two conditions which are associated with decreased supply relative to demand, has been shown to initiate conducted vasodilation following its intraluminal application in first and second order arterioles. Since such low PO2 and low pH conditions would most likely occur on the venous side of the vasculature, we evaluated the response of the arteriolar and capillary networks to application of ATP into venules in the Saran-covered hamster cheek pouch retractor muscle using in vivo video microscopy. Intraluminal application of 40 and 400 pl of 1006 M ATP resulted in dosedependent increases in arteriolar diameter ú450 mm upstream from the site of application. These changes in arteriolar diameter were accompanied by significant increases in red blood cell flux. In capillaries, red blood cell flux doubled in response to ATP administration. Since NO was previously determined to be involved in the vascular response to intraluminal ATP in arterioles, we evaluated its role in these responses. We found that systemic administration of L-NAME prior to ATP application eliminated any conducted response and this effect of L-NAME was reversed by the systemic administration of L-arginine. These data suggest that ATP, which is released from red blood cells in response to low PO2 and low pH, conditions which would be found in the venular microvasculature, may serve
a role in distributing perfusion in response to alterations in supply. q 1998 Academic Press Key Words: ATP; conducted vasomotor responses; vasodilation; red blood cells; striated muscle; hamsters.
INTRODUCTION Blood flow must be finely regulated for the maintenance of tissue homeostasis. The precise matching of oxygen supply to demand requires that there be a means of communicating tissue needs to its vascular supply. We have previously suggested that the red blood cell (RBC) may serve a role in communicating tissue requirements to the supplying vasculature by releasing ATP in response to low PO2 and low pH (Bergfeld and Forrester, 1992; Ellsworth et al., 1995). RBCs contain millimolar amounts of ATP (Miseta et al., 1993) which is produced in the circulating red blood cell by membrane bound glycolytic pathways. McCullough et al. (1997) showed that micropressure application of micromolar amounts of ATP into first and second order arterioles resulted in a significant conducted vasodilation observed in the same arteriole as far as 1750 mm upstream from the site of application. Since, in the microvasculature, low PO2 and low pH are more likely to exist in the collecting venules, RBCs would be most likely to release ATP in this region. Therefore, it is important to determine if ATP applied into these vessels
0026-2862/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$201
43
06-30-98 13:00:27
mvrg
AP: MVR
44
Collins, McCullough, and Ellsworth
would similarly produce an increase in vascular caliber in the feeding arterioles, resulting in an increase in flow. For such a response to be conducted to the arteriole, the interposed capillary network must be traversed. Using the hamster retractor muscle as a model, we investigated the hypothesis that ATP administered into a second order venule would elicit vasodilation of anatomically connected second order and terminal arterioles and that this vasodilation would be accompanied by increases in perfusion in these arterioles as well as the intervening capillary network. In addition, we also investigated the hypothesis that the effects of ATP on vascular caliber and flow involve NO as suggested previously by us (McCullough et al., 1997) and others (Kaley and Koller, 1995).
MATERIALS AND METHODS
lamp (75 W) with a stabilized power supply to provide constant light intensity. The microvasculature was viewed with a video microscopy system which consisted of a Zeiss Axioskop microscope equipped with long working distance objectives (UD20/0.57 and UD40/0.65). The microscope image was visualized and recorded using a high-resolution, closed circuit video system consisting of a CCD camera (Model CCD-72 with the image intensified using a Genisys II system, Dage MTI), video monitor (Sony, PVM-137), S-VHS video cassette recorder (Panasonic, AG-1970), and a time-date generator (Panasonic, WJ-810). The microscope was also equipped for epifluorescence using a mercury lamp (100 W) with a stabilized power supply and filter combination containing a 535- to 550-nm interference filter and a 580-nm barrier filter.
Red Blood Cell Flux Determinations
Animal Preparation Male Golden hamsters (74.6 { 17.2 g, 32 { 6 days, N Å 36) were initially anesthetized with sodium pentobarbital (6.5 mg/100 g body wt ip) and tracheotomized with PE160 tubing to ensure a patent airway. The animal was allowed to breathe room air spontaneously. For the continuous monitoring of systemic arterial blood pressure, a femoral artery was cannulated with PE10 tubing. A femoral vein was likewise cannulated to enable continuous infusion of supplemental anesthetic (sodium pentobarbital, 0.15 mg/min) and for drug administration. The cheek pouch retractor muscle was prepared as described by Sullivan and Pittman (1982). Briefly, a small incision made in the skin of the back allowed access to the cheek pouch retractor muscle, which was detached at its distal end and positioned ventral side up at its in situ dimensions on a transparent platform. The muscle, which remained stationary throughout the experiment, was covered with Saran (Dow Corning) to impede the exchange of oxygen with the atmosphere and prevent fluid loss. Deep esophageal and muscle temperatures were monitored and maintained at 37 { 17C by separate heat exchangers within the animal platform. The muscle was transilluminated using a xenon
Fluorescently tagged hamster red blood cells were used to determine red blood cell flux. Labeled cells were prepared as described by Sarelius and Duling (1982) with X - rhodamine - 5 - (and - 6) - isothiocyanate (XRITC). In brief, blood was collected from an anesthetized donor hamster by cardiac puncture, using a heparinized 5-cc syringe and 20-gauge needle. The blood was washed using a Tris-buffered Ringer solution of the following composition (in mM): 140.5 NaCl, 4.7 KCl, 2.0 CaCl2 , 1.2 MgSO4 , and 21.0 THAM; 300 mOsm. The packed red blood cells were then added to a solution of 100 mg/ml XRITC (Molecular Probes) in the Trisbuffered Ringer solution, pH 8.0 – 8.1, and incubated at room temperature for 1.5 h. The cells were washed several times in Tris-buffered Ringer solution, pH 7.4, to eliminate excess dye. Aliquots of labeled red blood cells were administered systemically to achieve approximately a 0.1 to 1% labeled cell ratio depending on the portion of the microvasculature being studied. The cells were allowed to circulate for at least 30 min to achieve a uniform distribution. The fraction of labeled cells ( f ) was determined at the end of the experiment by manual counting. The frequency of passage of labeled cells (Nl) was determined by replaying video tapes of epifluorescent microscope scenes. The total red blood cell flux (i.e., the number of RBCs crossing an observation line
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$201
06-30-98 13:00:27
mvrg
AP: MVR
45
ATP Signaling across the Capillary Network
FIG. 1. Schematic of ATP application showing pipette placement and sight of measurement. Distances given throughout the paper are the minimum distances between the feeding arteriole measured and the collecting venule to which the test solution was applied.
across the lumen per second, NRBC) was calculated as NRBC Å Nl/f.
General Protocol A glass pipette with a tip diameter of 1 – 2 mm was back-filled with a solution of magnesium ATP (1006 M; Sigma, St. Louis, MO) dissolved in the Tris-buffered Ringer solution, pH 7.4. Vehicle controls were run using pipettes filled with the same Tris-buffered Ringer solution alone. A hole was made in the Saran by piercing it with a probe with a tip diameter of 3 – 4 mm attached to a micromanipulator. A pipette tip was then passed though the hole and positioned within the venular lumen (Fig. 1). The Saran sealed itself around the pipette minimizing the effect of atmospheric oxygen on the vessel. A site on an arteriole which was anatomically connected to this venule was visualized and videotaped for 2 min prior to application of the test solution. The shortest distance between the measurement site and the application site was at least 450 mm. The solution was ejected from the pipette using a Picospritzer II (General Valve Corporation, Fairfield, NJ) at a pressure of 10 psi for durations of either 50 or 500 ms which corresponds to volumes of 40 and 400 pl, respectively. The volumes were determined by weighing 100 ejections and com-
paring the results to the weight of a known volume of the same fluid. The diameters of secondary and terminal arterioles and red blood cell flux responses in second order arterioles, terminal arterioles, and capillaries were observed on the video monitor and recorded on video tape for subsequent analysis. Vessel diameters were measured directly from the video monitor using a vernier caliper to compare the vessel image size to a videotaped stage micrometer. Baseline diameter and flux measurements were obtained during the 5 s immediately prior to administration of the test solution (i.e., either ATP or control solution). Since changes in diameter and red blood cell flux could not be made simultaneously, we first determined the time of maximal diameter change and then added 1 s to the start and end of the response to obtain the time frame for determination of the maximal flux response. For determination of red blood cell flux in capillaries, we used as a reference time the point of maximal diameter change in the terminal arteriole. The distance was measured from the site of test solution application in the collecting venule to the region that was observed on the second order arteriole, terminal arteriole, or capillary network. In addition, we determined the shortest distance between the application site and measurement site enabling us to evaluate the impact of diffusive transfer of ATP.
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
46
Collins, McCullough, and Ellsworth
Role of NO in Vascular Response to ATP In 26 hamsters, the nitric oxide synthase inhibitor, Nv-nitro-L-arginine methyl ester (L-NAME; 0.4 mg/100 g body wt iv: Sigma) was administered systemically 10 min before injecting ATP. In these animals, the response to ATP was determined on the same vessels prior to and following administration of L-NAME. To confirm that the effect of L-NAME was related to its ability to act as a competitive inhibitor of NO synthesis, in 8 animals we administered L-arginine (50 mg/100 g body wt iv: Sigma) 50 min prior to administration of L-NAME. In these animals, the vascular responses to 1006 M ATP were first obtained in the absence of L-arginine and LNAME. L-Arginine was then administered and the data were again collected 30 min later using the same vessels. Finally, after L-NAME administration, an additional response to ATP was obtained using the same vessels.
Data Analysis The maximal change in vessel diameter was computed as [(DR 0 DB)/DB] 1 100 where the subscripts R and B refer to response and baseline diameters, respectively. Changes in total red blood cell flux were determined in a similar manner with flux data reported as cells per second. Capillaries in which baseline flux during the observation period was zero were not included in the calculation of percentage change but have been included in the determination of mean values. Data are presented as means { SD. N denotes number of animals; n denotes number of observations with 1 to 3 measurements obtained on each arteriole or capillary. To assess statistical significance we used an analysis of variance followed by a Wilcoxon sign rank test to evaluate the change in diameter following each application and to compare the effect of ATP concentration and LNAME administration. All data analysis was done using a statistical software package (Instat, Graph-Pad Software, San Diego, CA). Significance was assigned at P õ 0.05.
RESULTS Application of 40 and 400 pl of 1006 M ATP into second order collecting venules resulted in a significant
dose-dependent vasodilation of second order arterioles measured 591 { 63 mm (range 450 – 655, N Å 17) and 587 { 66 mm (N Å 15) upstream from the site of application, respectively. In these arterioles, diameter increased from 33 { 4 mm (range 27 – 40 mm) to 36 { 4 and 37 { 4 mm for the two ATP application volumes, corresponding to increases of 9 { 2% (range 5 – 13%) and 12 { 2% (range 9 – 16%), respectively (Tables 1 and 2; Fig. 2). The maximal change in diameter was observed 12 { 2 s (range 8 – 16 s) and 12 { 3 s (range 8 – 16 s) following application of ATP for the two doses. These changes in diameter were accompanied by significant increases in red blood cell flux of 48 { 9% (range 25 – 60%, N Å 11, n Å 15) and 63 { 9% (range 47 – 74, N Å 10, n Å 14), for applications of 40 and 400 pl, respectively. In response to administration of 40 pl of 1006 M ATP, terminal arteriolar diameters (N Å 11) increased from 8 { 1 mm (range 6 – 9 mm) to 9 { 1 mm (range 7 – 10 mm) representing a change of 9 { 3% (Table 1, Fig. 3). The response to 400 pl of the same concentration of ATP was significantly greater with diameter increasing 13 { 1% (range 11 – 16%, Fig. 3, Table 2). These changes in diameter were determined 487 { 52 mm (range 400 – 550 mm) upstream from the site of application with the maximal change occurring 10 { 1 s following application. Increases in red blood cell flux were 39 { 9 and 50 { 11% for 40 and 400 pl applications, respectively, with flux increasing from 12 { 3 cells/s (range 8 – 17 cells/s) to 17 { 4 cells/s (range 11 – 22 cells/s) and 18 { 4 cells/s (range 14 – 26 cells/s), respectively. In addition to the measurements of arteriolar diameter and flow, we also determined the effect of 1006 M ATP on capillary red blood cell flux (Tables 1 and 2). We found that the response was not different for the two application volumes so the data have been combined for analysis and presentation (Fig. 4). Baseline red blood cell flux was 6 { 3 cells/s (range 0 – 11, N Å 13, n Å 227). Following application of ATP, we observed a significant increase to 11 { 4 cells/s (range 2 – 19 cells/ s), an increase of 91 { 44% (range 0 – 200%). Systemic administration of L-NAME resulted in a significant increase in mean arterial blood pressure from 97 { 9 to 111 { 8 mm Hg with no effect on baseline arteriolar diameters or fluxes. All conducted vasodilatory and red blood cell flux responses to 1006 M ATP were eliminated. However, in the presence of an excess
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
47
ATP Signaling across the Capillary Network
TABLE 1 Diameter and Red Cell Flux Measurements Made Prior and Subsequent to 50 ms (40 pl) of Venular Application of 1006 M ATP Diameter (mm) Vessel type 27 Arterioles Terminal arterioles
RBC Flux (cells/s)
Baseline
Response
% Change
Baseline
Response
% Change
33 { 4 (32) 8{1 (15)
36 { 4* (32) 9 { 1* (15)
9 { 2** (32) 9 { 3** (15)
80 { 15 (15) 12 { 3 (22) 6{ 3 (113)
118 { 23* (15) 17 { 4* (22) 10 { 4* (113)
48 { 9** (15) 39 { 9** (22) 81 { 40 (108)
Capillaries
Note. Numbers in parentheses denote the number of observations with one to three obtained on each arteriole or capillary. Values are means { 1 SD. * Significant difference from baseline value. ** Significant difference from 500 msec application (Table 2).
of systemically administered L-arginine, systemic LNAME had no effect on the conducted diameter and flux responses to intraluminal application of 1006 M ATP. Systemic administration of L-arginine alone had no effect on the vasodilatory response or the increase in red blood cell flux seen following application of 1006 M ATP. The inclusion of experiments in which the effects of L-NAME were prevented by the administration of L-arginine was essential in light of reports that arginine analogs may have actions unrelated to inhibition of the formation of NO (Buxton et al., 1993; Cocks and Angus, 1991; Peterson et al., 1992). However, such effects of L-NAME would not be expected to be prevented by administration of L-arginine (Buxton et al., 1993; Cocks and Angus, 1991).
Intraluminal applications of 40 and 400 pl of the control vehicle (Tris-buffered Ringer) either alone or following the systemic administration of L-NAME did not alter either conducted vessel diameter or red blood cell flux.
DISCUSSION In striated muscle, the delivery of appropriate amounts of metabolic substrate and removal of metabolic waste products requires that the tissue be appropriately perfused. To ensure this, some mechanism must exist to communicate tissue needs to the vasculature. When oxygen demand exceeds supply, a low PO2 and low pH environment results which should in-
TABLE 2 Diameter and Red Cell Flux Measurements Made Prior and Subsequent to 500 ms (400 pl) of Venular Application of 1006 M ATP Diameter (mm) Vessel type 27 Arterioles Terminal arterioles
RBC Flux (cells/s)
Baseline
Response
% Change
Baseline
Response
% Change
33 { 4 (27) 8{1 (14)
37 { 4* (27) 9 { 1* (14)
12 { 2** (27) 12 { 2** (14)
78 { 16 (14) 12 { 3 (20) 6{ 3 (114)
127 { 26 (14) 18 { 4* (20) 12 { 1* (114)
63 { 9 (14) 50 { 11** (20) 92 { 41 (111)
Capillaries
Note. Numbers in parentheses denote the number of observations with one to three obtained on each arteriole or capillary. Values are means { 1 SD. * Significant difference from baseline value. ** Significant difference from 50-ms application (Table 1).
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
48
Collins, McCullough, and Ellsworth
FIG. 2. Change in second order arteriolar diameter and red blood cell flux following venular application of 1006 M ATP alone, following systemic administration of L-NAME in the presence and absence of systemically administered L-arginine and a Tris-buffered Ringer control solution. The application volume was either 40 (open bar) or 400 (hatched bar) pl representing application durations of 50 and 500 ms, respectively. Data are presented as means { 1 SD. * Significant difference from baseline value. # Significant difference from 40 pl response.
duce an increase in blood flow. We have demonstrated that venular application of a molecule released from red blood cells under conditions of low PO2 and low pH results in both an increase in arteriolar diameter and red blood cell flux in both second order arterioles and terminal arterioles. In addition, red blood cell flux in capillaries also increases significantly. These results support the idea that the red blood cell may serve as an important regulator of striated muscle perfusion. In an earlier study, Berg et al. (1997) demonstrated that stimulation of individual striated muscle fibers induced a significant increase in blood flow only in those vessels directly associated with the stimulated fibers. Our results presented here would provide a mechanism enabling this level of the fine control of blood flow distribution providing additional flow only to the fibers which are metabolically active. On the surface, changes in vessel diameter of 9 – 12% may seem small. However, according to Poiseuille’s law (Milnor, 1982), blood flow is influenced by the
fourth power of the radius. Therefore, a 9% increase in diameter should result in a calculated 19% increase in flow. However, since the responses are conducted, the change in red blood cell flux would be even greater than that predicted by Poiseuille’s law as proposed by Segal and Duling (1987) and demonstrated previously by Friebel et al. (1995) and Kurijiaka and Segal (1995). In the present study, we demonstrate that ATP applied into the venule results in increases in red blood cell flux in secondary and terminal arterioles of 63 and 50% for 400 pl applications, respectively, similar to that suggested by Segal and Duling (1987). In the capillaries, red blood cell flux almost doubles. Although this increase in capillary red blood cell flux is statistically larger than that seen in the second order and terminal arterioles, we suspect that the difference is not physiologically relevant but rather is a consequence of the small number of labeled red blood cells traversing the capillaries in the short time period measured. The large standard deviation associated with this percentage
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
49
ATP Signaling across the Capillary Network
FIG. 3. Change in terminal arteriolar diameter and red blood cell flux following venular application of 1006 M ATP alone, following systemic administration of L-NAME in the presence and absence of systemically administered L-arginine and a Tris-buffered Ringer control solution. The application volume was either 40 (open bar) or 400 (hatched bar) pl representing application durations of 50 and 500 ms, respectively. Data are presented as means { 1 SD. * Significant difference from baseline value. # Significant difference from 40 pl response.
change value would support such a presumption. Clearly, red blood cell flux increases significantly in both the arterioles and the capillaries in response to ATP application into the secondary venule. Conducted vasomotor responses, defined as vascular changes that extend well beyond the region of initiation, provide a mechanism for coordination of microvascular changes. The precise cellular mechanism responsible for the transmission of the conducted response is unclear, although electrotonic spread via gap junctions is likely involved (Berg et al., 1997; Christ et al., 1996; Segal and Duling, 1987, 1989). Gap junctions, sites of low resistance on cell membranes, allow for signal transfer between endothelial cells and smooth muscle cells and at the myoendothelial junction (Segal and Duling, 1987). Our data demonstrating that application of ATP into the lumen of collecting venules resulted in dilation of the feed arterioles are consistent with this hypothesis that the endothelial cells provide the signal pathway, supporting earlier suggestions of
Haas and Duling (1997). The ability of capillaries to conduct signals has been reported previously (Dietrich and Tyml, 1990; Song and Tyml, 1993). The results presented here are unique in that they demonstrate that a venular signal can be conducted across the capillary bed, as suggested previously by Tigno et al. (1989). The precise mechanisms by which ATP induces a conducted response have yet to be determined. However, initiation of the conducted response likely involves the binding of ATP to one of two G-protein-coupled receptors for ATP located on the endothelium, P2Y and P2U (Ralevic and Burnstock, 1996), which then initiate a series of events resulting in local and conducted vasodilation. The mechanisms for the initiation of local and conducted vasodilation are likely quite different (Dietrich et al., 1996). It is of interest that the vasodilator response we observed in second order arterioles following each application of ATP was not monophasic, as we have seen previously where application and measurement were
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
50
Collins, McCullough, and Ellsworth
FIG. 4. Red blood cell flux in capillaries prior and subsequent to venular application of 1006 M ATP alone, following systemic administration of L-NAME in the presence and absence of systemically administered L-arginine, and a Tris-buffered Ringer control solution. Baseline measurements (open bar) were made 5 s prior to ATP or control solution application. Response measurements (solid bar) were determined based on the time of corresponding terminal arteriole diameter change (see Materials and Methods). Data are presented as means { 1 SD. * Significant change from baseline value.
made in the same arteriole (McCullough et al., 1997) but rather occurred as a series of several dilations, providing an oscillatory effect (Fig. 5). The oscillations occurred rapidly and, just as quickly, ended with diameter returning to its original prestimulus level. We found that the number of oscillations seen in a given experimental run coincided well with the number of third order arterioles fed by the second order arteriole being studied (Fig. 6). This result suggests that the vascular architecture plays a role in the conducted response with signals transmitted along pathways of different lengths (Fig. 6). The velocity of signal conduction in response to intraluminal ATP application is on the order of 47
mm/s, significantly slower than that observed for other substances (Segal and Duling, 1989) although similar to values reported by Doyle and Duling (1997). Calculation of the distance that the signal must travel to provide the first oscillation (the distance measured) and the last oscillation reveals values of 450 and 2077 mm, respectively. These estimated distances are well within the range of arteriovenous pathlengths determined in the retractor muscle previously (Ellsworth et al., 1987, and unpublished data). It is important to note that this calculation assumes that conduction through all segments of the vasculature is the same, which may or may not be the case.
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$203
06-30-98 13:00:27
mvrg
AP: MVR
51
ATP Signaling across the Capillary Network
FIG. 5. Response of a single second order arteriole to the intraluminal application of 1006 M ATP into the collecting venule. Distance between application site and the location of diameter measurement was 600 mm. The number of third order arterioles being fed by the secondary vessel was 8.
Since oscillations are generally associated with vasomotion, we were concerned that the oscillations we had observed were simply vasomotion or spasm induced by placement of the pipette. However, the response visualized was not as rhythmic and regular as demonstrated by others (Colantuoni et al., 1985; Funk et al., 1983). On hamster dorsal skin folds, Funk et al. (1983) observed between 1 and 5 cycles per minute in vessels 40 – 100 mm in diameter, with a mean amplitude of change at 20%. In the same vessel size range, we observed between 2 and 8 cycles at a rate of approximately 20 cycles per minute. In addition, applications of a control solution did not induce the oscillatory response, nor had we observed such a response in our previous study on arterioles (McCullough et al., 1997). Thus, the response does not appear to be an artifact of the preparation further suggesting that path length may play a role in this oscillatory response (Fig. 6). Of additional concern was that the vasodilation we observed in the arterioles could reflect diffusive transfer of ATP from the venule to the arteriole as suggested by Hester (1993) rather than a conducted response along the vasculature. Therefore, we computed the time required for ATP to diffuse the minimal distance between the application site in the venule and the response site in the secondary arteriole (Ç450 mm). Using Einstein’s equation (t Å L2t /2D) and the diffusion coef-
ficient (D) for ATP (Ç1006 cm2/s) reported by Hubley et al. (1996), we determined that it would require ú16 min for the signal to spread by diffusion, significantly longer than the 8 – 16 s observed in these studies. Even if one uses the diffusion coefficient for small molecules of 1005 cm2/s, 101 s would be required. In addition, ATP arriving at the arteriole by diffusion would induce vasoconstriction (McCullough, 1997), due to the preponderance of P2x receptors on the smooth muscle side, rather than the observed vasodilation. As a result of these observations, we have ruled out diffusion as the means of communication in these experiments. Systemic administration of L-NAME, an inhibitor of nitric oxide synthase, resulted in the elimination of both the vasodilatory and the increased perfusion response to intraluminal ATP administration suggesting that NO may play a role in this response. Although NO clearly can induce a local vasodilation (Falcone and Meininger, 1997), there is no evidence to suggest that NO itself induces a conducted response to a short stimulus (Doyle and Duling, 1997) thus requiring some addi-
FIG. 6. Relationship between the number of oscillations observed in a second order arteriole and the number of associated third order arterioles. The dotted line is the line of identity. The solid line is the regression line (r 2 Å 0.98). Data are observations of single second order arterioles. Number in parentheses is the number of animals in which the response was observed. Unlabeled points are indicative of a single observation.
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$203
06-30-98 13:00:27
mvrg
AP: MVR
52
Collins, McCullough, and Ellsworth
tional pathway. There are several possible mechanisms which may contribute to the establishment of the conducted response including an effect of NO to increase intracellular cGMP levels resulting in a decrease in intracellular Ca2/ concentration and thus vascular relaxation or enhancing the activity of Na/K/-ATPase (Gupta et al., 1994, 1995) resulting in the endothelial cell and/or smooth muscle becoming hyperpolarized inducing vasodilation. This hyperpolarization could be conducted along the endothelium and/or smooth muscle via gap junctions although our results would suggest the former route. In addition, NO may work in concert with endothelium-derived hyperpolarizing factor (EDHF) to produce the conducted response as suggested previously by Kajita et al. (1996). Recently, Dora et al. (1997) proposed that an elevation of smooth muscle Ca2/ generates a diffusion gradient that drives Ca2/ into the neighboring endothelial cells, initiating NO synthesis which would suggest that NO is continuously generated and affecting the vasculature locally. Our results are consistent with NO playing a role in the conducted vasodilator response to intraluminal ATP, although the mechanism has yet to be determined. In summary, our results indicate that application of ATP into a collecting venule produces a vasodilator response, which is conducted along the vasculature, through the capillary network to the supplying arteriole inducing significant increases in vessel diameter and red blood cell flux. Since the amount of ATP used in this study has been shown to be released from hamster red blood cells in response to low PO2 and low pH (Ellsworth et al., 1995), two conditions more likely present in portions of the venular microvasculature, our data provide strong support for the hypothesis that the red blood cell itself may be a regulator of the distribution of microvascular perfusion in response to tissue need.
ACKNOWLEDGMENTS The authors thank Dr. Hans H. Dietrich for his assistance with the measurements of propagated responses and for useful comments on the manuscript, Dr. Ingrid H. Sarelius for her assistance with labeling red cells, and Dr. Randy S. Sprague for useful comments on the manuscript. This study was supported by a grant from the National Heart,
Lung, and Blood Institute, U.S. Public Health Service (HL-39226). M.L.E. was supported by a Research Career Development Award from the same institute (HL-02602).
REFERENCES Berg, B. R., Cohen, K. D., and Sarelius, I. H. (1997). Direct coupling between blood flow and metabolism at the capillary level in striated muscle. Am. J. Physiol. 272, H2693 – H2700. Bergfeld, G. R., and Forrester, T. (1992). Release of ATP from human erythrocytes in response to a brief period of hypoxia and hypercapnia. Cardiovasc. Res. 26, 40 – 47. Buxton, I. L. O., Cheek, D. L., Eckman, D., Westfall, D. P., Sanders, K. M., and Keef, K. D. (1993). NG-Nitro-L-arginine methyl ester and other alkyl methyl esters of arginine are muscarinic receptor antagonists. Circ. Res. 72, 387 – 395. Christ, G. J., Spray, D. C., El-Sabban, M., Moore, L. K., and Brink, P. R. (1996). Gap junctions in vascular tissues — Evaluating the role of intercellular communication in the modulation of vasomotor tone. Circ. Res. 79, 631 – 646. Cocks, T. M., and Angus, J. A. (1991). Evidence that contractions of isolated arteries by L-NMMA and NOLA are not due to inhibition of basal EDRF release. J. Cardiovasc. Pharmacol. 17(Suppl), S159 – S164. Colantuoni, A., Bertuglia, S., and Intaglietta, M. (1985). Variations of rhythmic diameter changes at the arterial microvascular bifurcations. Pflu¨gers Arch. 403, 289 – 295. Dietrich, H. H., Kajita, Y., and Dacey, R. G., Jr. (1996). Local and conducted vasomotor responses in isolated rat cerebral arterioles. Am. J. Physiol. 271, H1109 – H1116. Dietrich, H. H., and Tyml, K. (1990). Microvascular flow response to localized application of norepinephrine on capillaries in rat and frog skeletal muscle. Microvasc. Res. 38, 125 – 135. Dora, K. A., Doyle, M. P., and Duling, B. R. (1997). Elevation of intracellular calcium in smooth muscle causes endothelial cell generation of NO in arterioles. Proc. Natl. Acad. Sci. USA 94, 6529 – 6534. Doyle, M. P., and Duling, B. R. (1997). Acetylcholine induces conducted vasodilation by nitric oxide-dependent and -independent mechanisms. Am. J. Physiol. 272, H1364 – H1371. Ellsworth, M. L., Forrester, T., Ellis, C. G., and Dietrich, H. H. (1995). The erythrocyte as a regulator of vascular tone. Am. J. Physiol. 269, H2155 – H2161. Ellsworth, M. L., Liu, A., Dawant, B., Popel, A. S., and Pittman, R. N. (1987). Analysis of vascular pattern and dimensions in arteriolar networks of the retractor muscle in young hamsters. Microvasc. Res. 34, 168 – 183. Falcone, J. C., and Meininger, G. A. (1997). Arteriolar dilation produced by venule endothelium-derived nitric oxide. Microcirculation 4, 303 – 310. Friebel, M., Klotz, K. F., Ley, K., Gaehtgens, P., and Pries, A. R. (1995). Flow-dependant regulation of arteriolar diameter in rat skeletal
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$203
06-30-98 13:00:27
mvrg
AP: MVR
53
ATP Signaling across the Capillary Network
muscle in situ: Role of endothelium-derived relaxing factor and prostanoids. J. Physiol. 483, 715 – 726. Funk, W., Endrich, B., Messmer, K., and Intaglietta, M. (1983). Spontaneous arteriolar vasomotion as a determinant of peripheral vascular resistance. Int. J. Microcirc. Clin. Exp. 23, 11 – 25. Gupta, S., McArthur, C., Grady, C., and Ruderman, N. B. (1994). Stimulation of vascular Na/-K/-ATPase activity by nitric oxide: A cGMP-independent effect. Am. J. Physiol. 266, H2146 – H2151. Gupta, S., Moreland, R. B., Munarriz, R., Daley, J., Goldstein, I., and de Tejada, I. S. (1995). Possible role of Na/-K/-ATPase in the regulation of human corpus cavernosum smooth muscle contractility by nitric oxide. Br. J. Pharmacol. 116, 2201 – 2206. Haas, T. L., and Duling, B. R. (1997). Morphology favors an endothelial cell pathway for longitudinal conduction within arterioles. Microvasc. Res. 53, 113 – 120. Hester, R. L. (1993). Uptake of metabolites by postcapillary venules: Mechanism for the control of arteriolar diameter. Microvasc. Res. 46, 254 – 261. Hubley, M. J., Locke, B. R., and Moerland, T. S. (1996). The effects of temperature, pH, and magnesium on the diffusion coefficient of ATP in solutions of physiological ionic strength. Biochim. Biophys. Acta 1291, 115 – 121. Kaley, G., and Koller, A. (1995). Prostaglandin–nitric oxide interactions in the microcirculation. Adv. Prosta. Throm. Leuko. Res. 23, 485–490. Kajita, Y., Dietrich, H. H., and Dacey, R. G., Jr. (1996). Effects of oxyhemoglobin on local and propagated vasodilatory responses induced by adenosine, adenosine diphosphate, and adenosine triphosphate in rat cerebral arterioles. J. Neurosurg. 85, 908 – 916. Kurijiaka, D. T., and Segal, S. S. (1995). Conducted vasodilation elevates flow in arteriole networks of hamster striated muscle. Am. J. Physiol. 269, H1723 – H1728. McCullough, W. T., Collins, D. M., and Ellsworth, M. L. (1997). Arteriolar responses to extracellular ATP in striated muscle. Am. J. Physiol. 272, H1886 – H1891. Milnor, W. R. (1982) ‘‘Hemodynamics,’’ pp. 11 – 12. Williams & Wilkins, Baltimore.
Miseta, A., Bogner, P., Berenyi, E., Kellermayer, M., Galambos, C., Wheatley, D., and Cameron, I. (1993). Relationship between cellular ATP, potassium, sodium and magnesium concentrations in mammalian and avian erythrocytes. Biochim. Biophys. Acta 1175, 133 – 139. Parsaee, H., McEwan, J. R., Joseph, S., and MacDermot, J. (1992). Differential sensitivities if the prostacyclin and nitric oxide biosynthetic pathways to cytosolic calcium in bovine aortic endothelial cells. Br. J. Pharmacol. 107, 1013 – 1019. Peterson, D. A., and Peterson, D. C., Archer, S., and Weir, E. K. (1992). The nonspecificity of specific nitric oxide synthase inhibitors. Biochem. Biophys. Res. Commun. 187, 797 – 801. Ralevic, V., and Burnstock, G. (1996). Relative contribution of P2Uand P2Y-purinoceptors to endothelium-dependent vasodilation in the golden hamster isolated mesenteric arterial bed. Br. J. Pharmacol. 117, 1797 – 1802. Sarelius, I. H., and Duling, B. R. (1982). Direct measurement of microvessel hematocrit, red cell flux, velocity, and transit time. Am. J. Physiol. 243, H1018 – H1026. Segal, S. S., and Duling, B. R. (1987). Propagation of vasodilation in resistance vessels of the hamster: Development and review of a working hypothesis. Circ. Res. 61(Suppl II): II-20 – II-25. Segal, S. S., and Duling, B. R. (1989). Conduction of vasomotor responses in arterioles: A role for cell-to-cell coupling? Am. J. Physiol. 256, H838 – H845. Song, H., and Tyml, K. (1993). Evidence for sensing and integration of biological signals by the capillary network. Am. J. Physiol. 265, H1235 – H1242. Sullivan, S. M., and Pittman, R. N. (1982). Hamster retractor muscle: A new preparation for intravital microscopy. Microvasc. Res. 23, 329 – 332. Tigno, X. T., Ley, K., Pries, A. R., and Gaehtgens, P. (1989). Venuloarteriolar communication and propagated response: A possible mechanism for local control of blood flow. Pflu¨gers Arch. 414, 450 – 456.
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$203
06-30-98 13:00:27
mvrg
AP: MVR
Conducted Vascular Responses: Communication across the Capillary Bed Diane M. Collins, William T. McCullough, and Mary L. Ellsworth Department of Pharmacological and Physiological Science, Saint Louis University Health Sciences Center, St. Louis, Missouri 63104 Received December 22, 1997
Conducted vasomotor responses are important for the effective distribution of blood flow, although the mechanism by which these responses are initiated is not well understood. ATP, a substance which is released from circulating red blood cells in response to low PO2 and low pH, two conditions which are associated with decreased supply relative to demand, has been shown to initiate conducted vasodilation following its intraluminal application in first and second order arterioles. Since such low PO2 and low pH conditions would most likely occur on the venous side of the vasculature, we evaluated the response of the arteriolar and capillary networks to application of ATP into venules in the Saran-covered hamster cheek pouch retractor muscle using in vivo video microscopy. Intraluminal application of 40 and 400 pl of 1006 M ATP resulted in dosedependent increases in arteriolar diameter ú450 mm upstream from the site of application. These changes in arteriolar diameter were accompanied by significant increases in red blood cell flux. In capillaries, red blood cell flux doubled in response to ATP administration. Since NO was previously determined to be involved in the vascular response to intraluminal ATP in arterioles, we evaluated its role in these responses. We found that systemic administration of L-NAME prior to ATP application eliminated any conducted response and this effect of L-NAME was reversed by the systemic administration of L-arginine. These data suggest that ATP, which is released from red blood cells in response to low PO2 and low pH, conditions which would be found in the venular microvasculature, may serve
a role in distributing perfusion in response to alterations in supply. q 1998 Academic Press Key Words: ATP; conducted vasomotor responses; vasodilation; red blood cells; striated muscle; hamsters.
INTRODUCTION Blood flow must be finely regulated for the maintenance of tissue homeostasis. The precise matching of oxygen supply to demand requires that there be a means of communicating tissue needs to its vascular supply. We have previously suggested that the red blood cell (RBC) may serve a role in communicating tissue requirements to the supplying vasculature by releasing ATP in response to low PO2 and low pH (Bergfeld and Forrester, 1992; Ellsworth et al., 1995). RBCs contain millimolar amounts of ATP (Miseta et al., 1993) which is produced in the circulating red blood cell by membrane bound glycolytic pathways. McCullough et al. (1997) showed that micropressure application of micromolar amounts of ATP into first and second order arterioles resulted in a significant conducted vasodilation observed in the same arteriole as far as 1750 mm upstream from the site of application. Since, in the microvasculature, low PO2 and low pH are more likely to exist in the collecting venules, RBCs would be most likely to release ATP in this region. Therefore, it is important to determine if ATP applied into these vessels
0026-2862/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$201
43
06-30-98 13:00:27
mvrg
AP: MVR
44
Collins, McCullough, and Ellsworth
would similarly produce an increase in vascular caliber in the feeding arterioles, resulting in an increase in flow. For such a response to be conducted to the arteriole, the interposed capillary network must be traversed. Using the hamster retractor muscle as a model, we investigated the hypothesis that ATP administered into a second order venule would elicit vasodilation of anatomically connected second order and terminal arterioles and that this vasodilation would be accompanied by increases in perfusion in these arterioles as well as the intervening capillary network. In addition, we also investigated the hypothesis that the effects of ATP on vascular caliber and flow involve NO as suggested previously by us (McCullough et al., 1997) and others (Kaley and Koller, 1995).
MATERIALS AND METHODS
lamp (75 W) with a stabilized power supply to provide constant light intensity. The microvasculature was viewed with a video microscopy system which consisted of a Zeiss Axioskop microscope equipped with long working distance objectives (UD20/0.57 and UD40/0.65). The microscope image was visualized and recorded using a high-resolution, closed circuit video system consisting of a CCD camera (Model CCD-72 with the image intensified using a Genisys II system, Dage MTI), video monitor (Sony, PVM-137), S-VHS video cassette recorder (Panasonic, AG-1970), and a time-date generator (Panasonic, WJ-810). The microscope was also equipped for epifluorescence using a mercury lamp (100 W) with a stabilized power supply and filter combination containing a 535- to 550-nm interference filter and a 580-nm barrier filter.
Red Blood Cell Flux Determinations
Animal Preparation Male Golden hamsters (74.6 { 17.2 g, 32 { 6 days, N Å 36) were initially anesthetized with sodium pentobarbital (6.5 mg/100 g body wt ip) and tracheotomized with PE160 tubing to ensure a patent airway. The animal was allowed to breathe room air spontaneously. For the continuous monitoring of systemic arterial blood pressure, a femoral artery was cannulated with PE10 tubing. A femoral vein was likewise cannulated to enable continuous infusion of supplemental anesthetic (sodium pentobarbital, 0.15 mg/min) and for drug administration. The cheek pouch retractor muscle was prepared as described by Sullivan and Pittman (1982). Briefly, a small incision made in the skin of the back allowed access to the cheek pouch retractor muscle, which was detached at its distal end and positioned ventral side up at its in situ dimensions on a transparent platform. The muscle, which remained stationary throughout the experiment, was covered with Saran (Dow Corning) to impede the exchange of oxygen with the atmosphere and prevent fluid loss. Deep esophageal and muscle temperatures were monitored and maintained at 37 { 17C by separate heat exchangers within the animal platform. The muscle was transilluminated using a xenon
Fluorescently tagged hamster red blood cells were used to determine red blood cell flux. Labeled cells were prepared as described by Sarelius and Duling (1982) with X - rhodamine - 5 - (and - 6) - isothiocyanate (XRITC). In brief, blood was collected from an anesthetized donor hamster by cardiac puncture, using a heparinized 5-cc syringe and 20-gauge needle. The blood was washed using a Tris-buffered Ringer solution of the following composition (in mM): 140.5 NaCl, 4.7 KCl, 2.0 CaCl2 , 1.2 MgSO4 , and 21.0 THAM; 300 mOsm. The packed red blood cells were then added to a solution of 100 mg/ml XRITC (Molecular Probes) in the Trisbuffered Ringer solution, pH 8.0 – 8.1, and incubated at room temperature for 1.5 h. The cells were washed several times in Tris-buffered Ringer solution, pH 7.4, to eliminate excess dye. Aliquots of labeled red blood cells were administered systemically to achieve approximately a 0.1 to 1% labeled cell ratio depending on the portion of the microvasculature being studied. The cells were allowed to circulate for at least 30 min to achieve a uniform distribution. The fraction of labeled cells ( f ) was determined at the end of the experiment by manual counting. The frequency of passage of labeled cells (Nl) was determined by replaying video tapes of epifluorescent microscope scenes. The total red blood cell flux (i.e., the number of RBCs crossing an observation line
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$201
06-30-98 13:00:27
mvrg
AP: MVR
45
ATP Signaling across the Capillary Network
FIG. 1. Schematic of ATP application showing pipette placement and sight of measurement. Distances given throughout the paper are the minimum distances between the feeding arteriole measured and the collecting venule to which the test solution was applied.
across the lumen per second, NRBC) was calculated as NRBC Å Nl/f.
General Protocol A glass pipette with a tip diameter of 1 – 2 mm was back-filled with a solution of magnesium ATP (1006 M; Sigma, St. Louis, MO) dissolved in the Tris-buffered Ringer solution, pH 7.4. Vehicle controls were run using pipettes filled with the same Tris-buffered Ringer solution alone. A hole was made in the Saran by piercing it with a probe with a tip diameter of 3 – 4 mm attached to a micromanipulator. A pipette tip was then passed though the hole and positioned within the venular lumen (Fig. 1). The Saran sealed itself around the pipette minimizing the effect of atmospheric oxygen on the vessel. A site on an arteriole which was anatomically connected to this venule was visualized and videotaped for 2 min prior to application of the test solution. The shortest distance between the measurement site and the application site was at least 450 mm. The solution was ejected from the pipette using a Picospritzer II (General Valve Corporation, Fairfield, NJ) at a pressure of 10 psi for durations of either 50 or 500 ms which corresponds to volumes of 40 and 400 pl, respectively. The volumes were determined by weighing 100 ejections and com-
paring the results to the weight of a known volume of the same fluid. The diameters of secondary and terminal arterioles and red blood cell flux responses in second order arterioles, terminal arterioles, and capillaries were observed on the video monitor and recorded on video tape for subsequent analysis. Vessel diameters were measured directly from the video monitor using a vernier caliper to compare the vessel image size to a videotaped stage micrometer. Baseline diameter and flux measurements were obtained during the 5 s immediately prior to administration of the test solution (i.e., either ATP or control solution). Since changes in diameter and red blood cell flux could not be made simultaneously, we first determined the time of maximal diameter change and then added 1 s to the start and end of the response to obtain the time frame for determination of the maximal flux response. For determination of red blood cell flux in capillaries, we used as a reference time the point of maximal diameter change in the terminal arteriole. The distance was measured from the site of test solution application in the collecting venule to the region that was observed on the second order arteriole, terminal arteriole, or capillary network. In addition, we determined the shortest distance between the application site and measurement site enabling us to evaluate the impact of diffusive transfer of ATP.
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
46
Collins, McCullough, and Ellsworth
Role of NO in Vascular Response to ATP In 26 hamsters, the nitric oxide synthase inhibitor, Nv-nitro-L-arginine methyl ester (L-NAME; 0.4 mg/100 g body wt iv: Sigma) was administered systemically 10 min before injecting ATP. In these animals, the response to ATP was determined on the same vessels prior to and following administration of L-NAME. To confirm that the effect of L-NAME was related to its ability to act as a competitive inhibitor of NO synthesis, in 8 animals we administered L-arginine (50 mg/100 g body wt iv: Sigma) 50 min prior to administration of L-NAME. In these animals, the vascular responses to 1006 M ATP were first obtained in the absence of L-arginine and LNAME. L-Arginine was then administered and the data were again collected 30 min later using the same vessels. Finally, after L-NAME administration, an additional response to ATP was obtained using the same vessels.
Data Analysis The maximal change in vessel diameter was computed as [(DR 0 DB)/DB] 1 100 where the subscripts R and B refer to response and baseline diameters, respectively. Changes in total red blood cell flux were determined in a similar manner with flux data reported as cells per second. Capillaries in which baseline flux during the observation period was zero were not included in the calculation of percentage change but have been included in the determination of mean values. Data are presented as means { SD. N denotes number of animals; n denotes number of observations with 1 to 3 measurements obtained on each arteriole or capillary. To assess statistical significance we used an analysis of variance followed by a Wilcoxon sign rank test to evaluate the change in diameter following each application and to compare the effect of ATP concentration and LNAME administration. All data analysis was done using a statistical software package (Instat, Graph-Pad Software, San Diego, CA). Significance was assigned at P õ 0.05.
RESULTS Application of 40 and 400 pl of 1006 M ATP into second order collecting venules resulted in a significant
dose-dependent vasodilation of second order arterioles measured 591 { 63 mm (range 450 – 655, N Å 17) and 587 { 66 mm (N Å 15) upstream from the site of application, respectively. In these arterioles, diameter increased from 33 { 4 mm (range 27 – 40 mm) to 36 { 4 and 37 { 4 mm for the two ATP application volumes, corresponding to increases of 9 { 2% (range 5 – 13%) and 12 { 2% (range 9 – 16%), respectively (Tables 1 and 2; Fig. 2). The maximal change in diameter was observed 12 { 2 s (range 8 – 16 s) and 12 { 3 s (range 8 – 16 s) following application of ATP for the two doses. These changes in diameter were accompanied by significant increases in red blood cell flux of 48 { 9% (range 25 – 60%, N Å 11, n Å 15) and 63 { 9% (range 47 – 74, N Å 10, n Å 14), for applications of 40 and 400 pl, respectively. In response to administration of 40 pl of 1006 M ATP, terminal arteriolar diameters (N Å 11) increased from 8 { 1 mm (range 6 – 9 mm) to 9 { 1 mm (range 7 – 10 mm) representing a change of 9 { 3% (Table 1, Fig. 3). The response to 400 pl of the same concentration of ATP was significantly greater with diameter increasing 13 { 1% (range 11 – 16%, Fig. 3, Table 2). These changes in diameter were determined 487 { 52 mm (range 400 – 550 mm) upstream from the site of application with the maximal change occurring 10 { 1 s following application. Increases in red blood cell flux were 39 { 9 and 50 { 11% for 40 and 400 pl applications, respectively, with flux increasing from 12 { 3 cells/s (range 8 – 17 cells/s) to 17 { 4 cells/s (range 11 – 22 cells/s) and 18 { 4 cells/s (range 14 – 26 cells/s), respectively. In addition to the measurements of arteriolar diameter and flow, we also determined the effect of 1006 M ATP on capillary red blood cell flux (Tables 1 and 2). We found that the response was not different for the two application volumes so the data have been combined for analysis and presentation (Fig. 4). Baseline red blood cell flux was 6 { 3 cells/s (range 0 – 11, N Å 13, n Å 227). Following application of ATP, we observed a significant increase to 11 { 4 cells/s (range 2 – 19 cells/ s), an increase of 91 { 44% (range 0 – 200%). Systemic administration of L-NAME resulted in a significant increase in mean arterial blood pressure from 97 { 9 to 111 { 8 mm Hg with no effect on baseline arteriolar diameters or fluxes. All conducted vasodilatory and red blood cell flux responses to 1006 M ATP were eliminated. However, in the presence of an excess
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
47
ATP Signaling across the Capillary Network
TABLE 1 Diameter and Red Cell Flux Measurements Made Prior and Subsequent to 50 ms (40 pl) of Venular Application of 1006 M ATP Diameter (mm) Vessel type 27 Arterioles Terminal arterioles
RBC Flux (cells/s)
Baseline
Response
% Change
Baseline
Response
% Change
33 { 4 (32) 8{1 (15)
36 { 4* (32) 9 { 1* (15)
9 { 2** (32) 9 { 3** (15)
80 { 15 (15) 12 { 3 (22) 6{ 3 (113)
118 { 23* (15) 17 { 4* (22) 10 { 4* (113)
48 { 9** (15) 39 { 9** (22) 81 { 40 (108)
Capillaries
Note. Numbers in parentheses denote the number of observations with one to three obtained on each arteriole or capillary. Values are means { 1 SD. * Significant difference from baseline value. ** Significant difference from 500 msec application (Table 2).
of systemically administered L-arginine, systemic LNAME had no effect on the conducted diameter and flux responses to intraluminal application of 1006 M ATP. Systemic administration of L-arginine alone had no effect on the vasodilatory response or the increase in red blood cell flux seen following application of 1006 M ATP. The inclusion of experiments in which the effects of L-NAME were prevented by the administration of L-arginine was essential in light of reports that arginine analogs may have actions unrelated to inhibition of the formation of NO (Buxton et al., 1993; Cocks and Angus, 1991; Peterson et al., 1992). However, such effects of L-NAME would not be expected to be prevented by administration of L-arginine (Buxton et al., 1993; Cocks and Angus, 1991).
Intraluminal applications of 40 and 400 pl of the control vehicle (Tris-buffered Ringer) either alone or following the systemic administration of L-NAME did not alter either conducted vessel diameter or red blood cell flux.
DISCUSSION In striated muscle, the delivery of appropriate amounts of metabolic substrate and removal of metabolic waste products requires that the tissue be appropriately perfused. To ensure this, some mechanism must exist to communicate tissue needs to the vasculature. When oxygen demand exceeds supply, a low PO2 and low pH environment results which should in-
TABLE 2 Diameter and Red Cell Flux Measurements Made Prior and Subsequent to 500 ms (400 pl) of Venular Application of 1006 M ATP Diameter (mm) Vessel type 27 Arterioles Terminal arterioles
RBC Flux (cells/s)
Baseline
Response
% Change
Baseline
Response
% Change
33 { 4 (27) 8{1 (14)
37 { 4* (27) 9 { 1* (14)
12 { 2** (27) 12 { 2** (14)
78 { 16 (14) 12 { 3 (20) 6{ 3 (114)
127 { 26 (14) 18 { 4* (20) 12 { 1* (114)
63 { 9 (14) 50 { 11** (20) 92 { 41 (111)
Capillaries
Note. Numbers in parentheses denote the number of observations with one to three obtained on each arteriole or capillary. Values are means { 1 SD. * Significant difference from baseline value. ** Significant difference from 50-ms application (Table 1).
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
48
Collins, McCullough, and Ellsworth
FIG. 2. Change in second order arteriolar diameter and red blood cell flux following venular application of 1006 M ATP alone, following systemic administration of L-NAME in the presence and absence of systemically administered L-arginine and a Tris-buffered Ringer control solution. The application volume was either 40 (open bar) or 400 (hatched bar) pl representing application durations of 50 and 500 ms, respectively. Data are presented as means { 1 SD. * Significant difference from baseline value. # Significant difference from 40 pl response.
duce an increase in blood flow. We have demonstrated that venular application of a molecule released from red blood cells under conditions of low PO2 and low pH results in both an increase in arteriolar diameter and red blood cell flux in both second order arterioles and terminal arterioles. In addition, red blood cell flux in capillaries also increases significantly. These results support the idea that the red blood cell may serve as an important regulator of striated muscle perfusion. In an earlier study, Berg et al. (1997) demonstrated that stimulation of individual striated muscle fibers induced a significant increase in blood flow only in those vessels directly associated with the stimulated fibers. Our results presented here would provide a mechanism enabling this level of the fine control of blood flow distribution providing additional flow only to the fibers which are metabolically active. On the surface, changes in vessel diameter of 9 – 12% may seem small. However, according to Poiseuille’s law (Milnor, 1982), blood flow is influenced by the
fourth power of the radius. Therefore, a 9% increase in diameter should result in a calculated 19% increase in flow. However, since the responses are conducted, the change in red blood cell flux would be even greater than that predicted by Poiseuille’s law as proposed by Segal and Duling (1987) and demonstrated previously by Friebel et al. (1995) and Kurijiaka and Segal (1995). In the present study, we demonstrate that ATP applied into the venule results in increases in red blood cell flux in secondary and terminal arterioles of 63 and 50% for 400 pl applications, respectively, similar to that suggested by Segal and Duling (1987). In the capillaries, red blood cell flux almost doubles. Although this increase in capillary red blood cell flux is statistically larger than that seen in the second order and terminal arterioles, we suspect that the difference is not physiologically relevant but rather is a consequence of the small number of labeled red blood cells traversing the capillaries in the short time period measured. The large standard deviation associated with this percentage
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
49
ATP Signaling across the Capillary Network
FIG. 3. Change in terminal arteriolar diameter and red blood cell flux following venular application of 1006 M ATP alone, following systemic administration of L-NAME in the presence and absence of systemically administered L-arginine and a Tris-buffered Ringer control solution. The application volume was either 40 (open bar) or 400 (hatched bar) pl representing application durations of 50 and 500 ms, respectively. Data are presented as means { 1 SD. * Significant difference from baseline value. # Significant difference from 40 pl response.
change value would support such a presumption. Clearly, red blood cell flux increases significantly in both the arterioles and the capillaries in response to ATP application into the secondary venule. Conducted vasomotor responses, defined as vascular changes that extend well beyond the region of initiation, provide a mechanism for coordination of microvascular changes. The precise cellular mechanism responsible for the transmission of the conducted response is unclear, although electrotonic spread via gap junctions is likely involved (Berg et al., 1997; Christ et al., 1996; Segal and Duling, 1987, 1989). Gap junctions, sites of low resistance on cell membranes, allow for signal transfer between endothelial cells and smooth muscle cells and at the myoendothelial junction (Segal and Duling, 1987). Our data demonstrating that application of ATP into the lumen of collecting venules resulted in dilation of the feed arterioles are consistent with this hypothesis that the endothelial cells provide the signal pathway, supporting earlier suggestions of
Haas and Duling (1997). The ability of capillaries to conduct signals has been reported previously (Dietrich and Tyml, 1990; Song and Tyml, 1993). The results presented here are unique in that they demonstrate that a venular signal can be conducted across the capillary bed, as suggested previously by Tigno et al. (1989). The precise mechanisms by which ATP induces a conducted response have yet to be determined. However, initiation of the conducted response likely involves the binding of ATP to one of two G-protein-coupled receptors for ATP located on the endothelium, P2Y and P2U (Ralevic and Burnstock, 1996), which then initiate a series of events resulting in local and conducted vasodilation. The mechanisms for the initiation of local and conducted vasodilation are likely quite different (Dietrich et al., 1996). It is of interest that the vasodilator response we observed in second order arterioles following each application of ATP was not monophasic, as we have seen previously where application and measurement were
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$202
06-30-98 13:00:27
mvrg
AP: MVR
50
Collins, McCullough, and Ellsworth
FIG. 4. Red blood cell flux in capillaries prior and subsequent to venular application of 1006 M ATP alone, following systemic administration of L-NAME in the presence and absence of systemically administered L-arginine, and a Tris-buffered Ringer control solution. Baseline measurements (open bar) were made 5 s prior to ATP or control solution application. Response measurements (solid bar) were determined based on the time of corresponding terminal arteriole diameter change (see Materials and Methods). Data are presented as means { 1 SD. * Significant change from baseline value.
made in the same arteriole (McCullough et al., 1997) but rather occurred as a series of several dilations, providing an oscillatory effect (Fig. 5). The oscillations occurred rapidly and, just as quickly, ended with diameter returning to its original prestimulus level. We found that the number of oscillations seen in a given experimental run coincided well with the number of third order arterioles fed by the second order arteriole being studied (Fig. 6). This result suggests that the vascular architecture plays a role in the conducted response with signals transmitted along pathways of different lengths (Fig. 6). The velocity of signal conduction in response to intraluminal ATP application is on the order of 47
mm/s, significantly slower than that observed for other substances (Segal and Duling, 1989) although similar to values reported by Doyle and Duling (1997). Calculation of the distance that the signal must travel to provide the first oscillation (the distance measured) and the last oscillation reveals values of 450 and 2077 mm, respectively. These estimated distances are well within the range of arteriovenous pathlengths determined in the retractor muscle previously (Ellsworth et al., 1987, and unpublished data). It is important to note that this calculation assumes that conduction through all segments of the vasculature is the same, which may or may not be the case.
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$203
06-30-98 13:00:27
mvrg
AP: MVR
51
ATP Signaling across the Capillary Network
FIG. 5. Response of a single second order arteriole to the intraluminal application of 1006 M ATP into the collecting venule. Distance between application site and the location of diameter measurement was 600 mm. The number of third order arterioles being fed by the secondary vessel was 8.
Since oscillations are generally associated with vasomotion, we were concerned that the oscillations we had observed were simply vasomotion or spasm induced by placement of the pipette. However, the response visualized was not as rhythmic and regular as demonstrated by others (Colantuoni et al., 1985; Funk et al., 1983). On hamster dorsal skin folds, Funk et al. (1983) observed between 1 and 5 cycles per minute in vessels 40 – 100 mm in diameter, with a mean amplitude of change at 20%. In the same vessel size range, we observed between 2 and 8 cycles at a rate of approximately 20 cycles per minute. In addition, applications of a control solution did not induce the oscillatory response, nor had we observed such a response in our previous study on arterioles (McCullough et al., 1997). Thus, the response does not appear to be an artifact of the preparation further suggesting that path length may play a role in this oscillatory response (Fig. 6). Of additional concern was that the vasodilation we observed in the arterioles could reflect diffusive transfer of ATP from the venule to the arteriole as suggested by Hester (1993) rather than a conducted response along the vasculature. Therefore, we computed the time required for ATP to diffuse the minimal distance between the application site in the venule and the response site in the secondary arteriole (Ç450 mm). Using Einstein’s equation (t Å L2t /2D) and the diffusion coef-
ficient (D) for ATP (Ç1006 cm2/s) reported by Hubley et al. (1996), we determined that it would require ú16 min for the signal to spread by diffusion, significantly longer than the 8 – 16 s observed in these studies. Even if one uses the diffusion coefficient for small molecules of 1005 cm2/s, 101 s would be required. In addition, ATP arriving at the arteriole by diffusion would induce vasoconstriction (McCullough, 1997), due to the preponderance of P2x receptors on the smooth muscle side, rather than the observed vasodilation. As a result of these observations, we have ruled out diffusion as the means of communication in these experiments. Systemic administration of L-NAME, an inhibitor of nitric oxide synthase, resulted in the elimination of both the vasodilatory and the increased perfusion response to intraluminal ATP administration suggesting that NO may play a role in this response. Although NO clearly can induce a local vasodilation (Falcone and Meininger, 1997), there is no evidence to suggest that NO itself induces a conducted response to a short stimulus (Doyle and Duling, 1997) thus requiring some addi-
FIG. 6. Relationship between the number of oscillations observed in a second order arteriole and the number of associated third order arterioles. The dotted line is the line of identity. The solid line is the regression line (r 2 Å 0.98). Data are observations of single second order arterioles. Number in parentheses is the number of animals in which the response was observed. Unlabeled points are indicative of a single observation.
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$203
06-30-98 13:00:27
mvrg
AP: MVR
52
Collins, McCullough, and Ellsworth
tional pathway. There are several possible mechanisms which may contribute to the establishment of the conducted response including an effect of NO to increase intracellular cGMP levels resulting in a decrease in intracellular Ca2/ concentration and thus vascular relaxation or enhancing the activity of Na/K/-ATPase (Gupta et al., 1994, 1995) resulting in the endothelial cell and/or smooth muscle becoming hyperpolarized inducing vasodilation. This hyperpolarization could be conducted along the endothelium and/or smooth muscle via gap junctions although our results would suggest the former route. In addition, NO may work in concert with endothelium-derived hyperpolarizing factor (EDHF) to produce the conducted response as suggested previously by Kajita et al. (1996). Recently, Dora et al. (1997) proposed that an elevation of smooth muscle Ca2/ generates a diffusion gradient that drives Ca2/ into the neighboring endothelial cells, initiating NO synthesis which would suggest that NO is continuously generated and affecting the vasculature locally. Our results are consistent with NO playing a role in the conducted vasodilator response to intraluminal ATP, although the mechanism has yet to be determined. In summary, our results indicate that application of ATP into a collecting venule produces a vasodilator response, which is conducted along the vasculature, through the capillary network to the supplying arteriole inducing significant increases in vessel diameter and red blood cell flux. Since the amount of ATP used in this study has been shown to be released from hamster red blood cells in response to low PO2 and low pH (Ellsworth et al., 1995), two conditions more likely present in portions of the venular microvasculature, our data provide strong support for the hypothesis that the red blood cell itself may be a regulator of the distribution of microvascular perfusion in response to tissue need.
ACKNOWLEDGMENTS The authors thank Dr. Hans H. Dietrich for his assistance with the measurements of propagated responses and for useful comments on the manuscript, Dr. Ingrid H. Sarelius for her assistance with labeling red cells, and Dr. Randy S. Sprague for useful comments on the manuscript. This study was supported by a grant from the National Heart,
Lung, and Blood Institute, U.S. Public Health Service (HL-39226). M.L.E. was supported by a Research Career Development Award from the same institute (HL-02602).
REFERENCES Berg, B. R., Cohen, K. D., and Sarelius, I. H. (1997). Direct coupling between blood flow and metabolism at the capillary level in striated muscle. Am. J. Physiol. 272, H2693 – H2700. Bergfeld, G. R., and Forrester, T. (1992). Release of ATP from human erythrocytes in response to a brief period of hypoxia and hypercapnia. Cardiovasc. Res. 26, 40 – 47. Buxton, I. L. O., Cheek, D. L., Eckman, D., Westfall, D. P., Sanders, K. M., and Keef, K. D. (1993). NG-Nitro-L-arginine methyl ester and other alkyl methyl esters of arginine are muscarinic receptor antagonists. Circ. Res. 72, 387 – 395. Christ, G. J., Spray, D. C., El-Sabban, M., Moore, L. K., and Brink, P. R. (1996). Gap junctions in vascular tissues — Evaluating the role of intercellular communication in the modulation of vasomotor tone. Circ. Res. 79, 631 – 646. Cocks, T. M., and Angus, J. A. (1991). Evidence that contractions of isolated arteries by L-NMMA and NOLA are not due to inhibition of basal EDRF release. J. Cardiovasc. Pharmacol. 17(Suppl), S159 – S164. Colantuoni, A., Bertuglia, S., and Intaglietta, M. (1985). Variations of rhythmic diameter changes at the arterial microvascular bifurcations. Pflu¨gers Arch. 403, 289 – 295. Dietrich, H. H., Kajita, Y., and Dacey, R. G., Jr. (1996). Local and conducted vasomotor responses in isolated rat cerebral arterioles. Am. J. Physiol. 271, H1109 – H1116. Dietrich, H. H., and Tyml, K. (1990). Microvascular flow response to localized application of norepinephrine on capillaries in rat and frog skeletal muscle. Microvasc. Res. 38, 125 – 135. Dora, K. A., Doyle, M. P., and Duling, B. R. (1997). Elevation of intracellular calcium in smooth muscle causes endothelial cell generation of NO in arterioles. Proc. Natl. Acad. Sci. USA 94, 6529 – 6534. Doyle, M. P., and Duling, B. R. (1997). Acetylcholine induces conducted vasodilation by nitric oxide-dependent and -independent mechanisms. Am. J. Physiol. 272, H1364 – H1371. Ellsworth, M. L., Forrester, T., Ellis, C. G., and Dietrich, H. H. (1995). The erythrocyte as a regulator of vascular tone. Am. J. Physiol. 269, H2155 – H2161. Ellsworth, M. L., Liu, A., Dawant, B., Popel, A. S., and Pittman, R. N. (1987). Analysis of vascular pattern and dimensions in arteriolar networks of the retractor muscle in young hamsters. Microvasc. Res. 34, 168 – 183. Falcone, J. C., and Meininger, G. A. (1997). Arteriolar dilation produced by venule endothelium-derived nitric oxide. Microcirculation 4, 303 – 310. Friebel, M., Klotz, K. F., Ley, K., Gaehtgens, P., and Pries, A. R. (1995). Flow-dependant regulation of arteriolar diameter in rat skeletal
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$203
06-30-98 13:00:27
mvrg
AP: MVR
53
ATP Signaling across the Capillary Network
muscle in situ: Role of endothelium-derived relaxing factor and prostanoids. J. Physiol. 483, 715 – 726. Funk, W., Endrich, B., Messmer, K., and Intaglietta, M. (1983). Spontaneous arteriolar vasomotion as a determinant of peripheral vascular resistance. Int. J. Microcirc. Clin. Exp. 23, 11 – 25. Gupta, S., McArthur, C., Grady, C., and Ruderman, N. B. (1994). Stimulation of vascular Na/-K/-ATPase activity by nitric oxide: A cGMP-independent effect. Am. J. Physiol. 266, H2146 – H2151. Gupta, S., Moreland, R. B., Munarriz, R., Daley, J., Goldstein, I., and de Tejada, I. S. (1995). Possible role of Na/-K/-ATPase in the regulation of human corpus cavernosum smooth muscle contractility by nitric oxide. Br. J. Pharmacol. 116, 2201 – 2206. Haas, T. L., and Duling, B. R. (1997). Morphology favors an endothelial cell pathway for longitudinal conduction within arterioles. Microvasc. Res. 53, 113 – 120. Hester, R. L. (1993). Uptake of metabolites by postcapillary venules: Mechanism for the control of arteriolar diameter. Microvasc. Res. 46, 254 – 261. Hubley, M. J., Locke, B. R., and Moerland, T. S. (1996). The effects of temperature, pH, and magnesium on the diffusion coefficient of ATP in solutions of physiological ionic strength. Biochim. Biophys. Acta 1291, 115 – 121. Kaley, G., and Koller, A. (1995). Prostaglandin–nitric oxide interactions in the microcirculation. Adv. Prosta. Throm. Leuko. Res. 23, 485–490. Kajita, Y., Dietrich, H. H., and Dacey, R. G., Jr. (1996). Effects of oxyhemoglobin on local and propagated vasodilatory responses induced by adenosine, adenosine diphosphate, and adenosine triphosphate in rat cerebral arterioles. J. Neurosurg. 85, 908 – 916. Kurijiaka, D. T., and Segal, S. S. (1995). Conducted vasodilation elevates flow in arteriole networks of hamster striated muscle. Am. J. Physiol. 269, H1723 – H1728. McCullough, W. T., Collins, D. M., and Ellsworth, M. L. (1997). Arteriolar responses to extracellular ATP in striated muscle. Am. J. Physiol. 272, H1886 – H1891. Milnor, W. R. (1982) ‘‘Hemodynamics,’’ pp. 11 – 12. Williams & Wilkins, Baltimore.
Miseta, A., Bogner, P., Berenyi, E., Kellermayer, M., Galambos, C., Wheatley, D., and Cameron, I. (1993). Relationship between cellular ATP, potassium, sodium and magnesium concentrations in mammalian and avian erythrocytes. Biochim. Biophys. Acta 1175, 133 – 139. Parsaee, H., McEwan, J. R., Joseph, S., and MacDermot, J. (1992). Differential sensitivities if the prostacyclin and nitric oxide biosynthetic pathways to cytosolic calcium in bovine aortic endothelial cells. Br. J. Pharmacol. 107, 1013 – 1019. Peterson, D. A., and Peterson, D. C., Archer, S., and Weir, E. K. (1992). The nonspecificity of specific nitric oxide synthase inhibitors. Biochem. Biophys. Res. Commun. 187, 797 – 801. Ralevic, V., and Burnstock, G. (1996). Relative contribution of P2Uand P2Y-purinoceptors to endothelium-dependent vasodilation in the golden hamster isolated mesenteric arterial bed. Br. J. Pharmacol. 117, 1797 – 1802. Sarelius, I. H., and Duling, B. R. (1982). Direct measurement of microvessel hematocrit, red cell flux, velocity, and transit time. Am. J. Physiol. 243, H1018 – H1026. Segal, S. S., and Duling, B. R. (1987). Propagation of vasodilation in resistance vessels of the hamster: Development and review of a working hypothesis. Circ. Res. 61(Suppl II): II-20 – II-25. Segal, S. S., and Duling, B. R. (1989). Conduction of vasomotor responses in arterioles: A role for cell-to-cell coupling? Am. J. Physiol. 256, H838 – H845. Song, H., and Tyml, K. (1993). Evidence for sensing and integration of biological signals by the capillary network. Am. J. Physiol. 265, H1235 – H1242. Sullivan, S. M., and Pittman, R. N. (1982). Hamster retractor muscle: A new preparation for intravital microscopy. Microvasc. Res. 23, 329 – 332. Tigno, X. T., Ley, K., Pries, A. R., and Gaehtgens, P. (1989). Venuloarteriolar communication and propagated response: A possible mechanism for local control of blood flow. Pflu¨gers Arch. 414, 450 – 456.
Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MVR 2076
/
6z16$$$203
06-30-98 13:00:27
mvrg
AP: MVR