Endothelial regulation of wall shear stress and blood flow in skeletal muscle microcirculation AKOS KOLLER AND GABOR KALEY Department of Physiology, New York Medical College, Valhalla, New York 10595
KOLLER AKOS,AND GABOR KALEY. Endothelial regulation . . n 1. 1 1 of wall shear stress anu oioou J~OW &n sneietai muscle mtcrocvcuZation. Am. J. Physiol. 260 (Heart Circ. Physiol. 29): H862H868, 1991.-In the presence of intact endothelium, in control conditions, calculated wall shear rate (WSR) (means t SE: 2,658 t 123 s-l; n = 21) was independent of arteriolar diameter (16.2-27.2 pm; correlation coefficient: r = 0.12, P > 0.05) in cremaster muscle of pentobarbital-anesthetized rats. An increase in blood flow velocity (due to parallel arteriolar occlusion) elicited a significant increase in WSR (to 4,981 t 253 s-l) followed by a delayed (6-15 s) increase in diameter (from: 22.5 t 0.6 to 29.5 t 0.8 pm), which consequently resulted in a significant decrease in WSR (to 3,879 t 203 s-l). As a result of the increased flow velocity and dilation, calculated arteriolar blood flow increased by 230%. After impairment of the endothelium of arterioles by a light-dye technique, basal WSR became significantly higher (3,604 t 341 s-l), and despite a greater increase in WSR (10,360 t 1,471 s-l) the dilation was absent. Now an inverse linear correlation was found between arteriolar diameter and WSR both before (r = 0.58, P < 0.05) and during increased flow velocity conditions (r = 0.85, P < 0.05). Also, arteriolar blood flow that was already less after impairment of endothelium increased by only 66% during the period of increased flow velocity due to the absence of dilation. Results suggest that an increase in wall shear stress is the stimulus for the endothelium-dependent mechanism that elicits “flow dependent” arteriolar dilation. This mechanism has a role in the negative-feedback control of wall shear stress and also participates in the regulation of resistance and blood flow in the microcirculation. 111
1
l-1
.
1
On the other hand, if pressure increases, blood flow velocity, which is directly proportional to wall shear stress, could increase as well. In our previous studies in skeletal muscle microcirculation (17-20), we found a significant correlation between increases in blood flow velocity and arteriolar diameter in the presence of intact endothelium, suggesting the existence of a shear stresssensitive mechanism that, independent of other factors, could serve to counterbalance the myogenic response and participate in the regulation of resistance and blood flow. Theoretical (29, 31, 36, 41) and experimental (3-5, 11, 14, 27, 28, 30, 32, 33, 37-39) studies also suggest an important relationship between shear stress and vessel radius, but the nature of this relationship and whether it can participate in the regulation of skeletal muscle microcirculation in vivo have not as yet been clarified. Therefore, in the present study, we sought to investigate whether an increase in wall shear stress could be a stimulus for initiating a regulatory mechanism leading to arteriolar dilation and the consequent decrease in wall shear stress and whether arteriolar endothelium has a role in this process in skeletal muscle microcirculation in vivo. For this purpose we calculated wall shear rate in arterioles during control and increased flow velocity conditions in the presence of intact as well as impaired endothelium. METHODS
hemodynamic forces; wall shear rate; autoregulation; occlusion; arteriolar dilation; red blood cell velocity; law
parallel Murray’s
that resistance to blood flow is the target of numerous control mechanisms in the peripheral circulation. Resistance is determined by factors affecting either vascular (diameter, length, network geometry, etc.) or rheological (viscosity, hematocrit, wall shear stress, etc.) parameters. The interaction among these factors to provide the actual resistance, however, is not completely understood. It is also recognized that an effective blood flow regulation requires several regulatory mechanisms that can interact with each other. Besides central factors, two main local mechanisms are believed to be involved in the regulation of microvascular resistance, namely the “pressure-sensitive” myogenic mechanism, which inversely couples pressure and vessel diameter (1, 13), and metabolic vasoregulation, which via changes in resistance matches the blood supply to tissue demand (9).
IT IS KNOWN
H862
0363-6135/91
$1.50 Copyright
Data utilized in the present study to calculate wall shear rate, as an indicator of wall shear stress, were obtained in experiments reported previously (19) and were taken from original strip-chart recordings or by replaying videotape recordings, whenever it was technically feasible. General preparation. The in vivo tissue preparation utilized in these experiments has been described in detail previously (15, 23) and is summarized as follows. Five-to six-week-old male Wistar rats were anesthetized with pentobarbital sodium (35 mg/kg). Arterial blood pressure was monitored with Statham P23 D6 transducer connected to a cannula inserted in the left common carotid artery. The left cremaster muscle was surgically exteriorized, and the muscle was suffused (2 ml/min) with Ringer gelatin solution, containing (in mM) 154 Nacl, 5.6 KCl, and 2.2 CaCIZ and 5 g/l gelatin. The pH of the solution (7.35-7.42) was adjusted with sodium bicarbonate and was equilibrated with ambient air at a temperature of 33°C. Vessels selected for this study were third-
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
ENDOTHELIAL
REGULATION
OF
SHEAR
STRESS
order arterioles, ranging in size from 16.1 to 27.2 pm in diameter. Intravascular diameter of microvessels was visualized by television microscopy (Olympus BH2) and was measured on- or off-line by an image-shearing monitor (IPM model 907, San Diego, CA) calibrated with a stage micrometer. Red blood cell (RBC) velocity was measured along the microvessel’s center line by using a self-tracking correlator (IPM model 102B) (24). The output of the measurements was recorded on a Sensormedics Dynograph Recorder (model 5 RllA). Light-dye treatment. The mercury light-sodium fluorescein (light-dye) treatment was used to impair the vasoactive functions of the endothelium in a localized segment of arterioles. This technique involves the illumination of a segment of an arteriole (80-100 pm in length) in the presence of i ntravascular sodium fluorescein, as described in detail previously (23, 24). Selective endothelial impairment was considered to have been accomplished if dilation to topically applied (100 ~1) endothelium-dependent dilator agents (lo-” M arachidonic acid or 10V6 M acetylcholine) was completely abolished while the arteriolar dilation to direct vascular smooth muscle relaxants (2 x 10B7 M sodium nitroprusside, or 10q5 M adenosine) was not affected (15, 23, 24). Acetylcholine chloride, arachidonic acid, adenosine, sodium nitroprusside, and sodium fluorescein were obtained from Sigma Chemical (St. Louis, MO). Experimental protocol. After surgery, a 30- to 40-min time period was allowed for the preparation to reach a steady-state condition. Control diameter and RBC velocity were measured in a third-order arteriole. Responses of this arteriole to vasoactive agents and to “parallel arteriolar occlusion” (Fig. 1) of 30- to 180-s duration were then examined. Typically this intervention resulted in an increase in RBC velocity in parallel-connected third-order arterioles located proximal from the occluder (only one of these was studied at any given time) (Fig.
AND
MICROVASCULAR
RESISTANCE
H863
2). In phase 1, just before the start of dilation, and phase 2, at peak dilation, RBC velocity and diameter data were obtained as indicated in Fig. 2 (left). As a next step the endothelium of the arteriolar segment studied was impaired by light-dye treatment and arteriolar responses to vasoactive agents and parallel occlusion were reexamined (Fig. 2, right). Because there was no second phase (i.e., no dilation) in the response of the arteriole with impaired endothelium, phase 1 was designated as a point at peak RBC velocity. Calculation of arteriolar wall shear rate and blood flow. From paired velocity and diameter data wall shear rate (WSR), a measure of wall shear stress, was calculated (2, 27) before and after impairment of the endothelium in control and during parallel occlusion (phases 1 and 2) as illustrated in Fig. 2. WSR = 8 (I&/D), where Vm is the mean or bulk velocity (cells plus plasma) calculated from the measured RBC velocity (I&c), using an empirical correction factor, and D is the luminal diameter of the vessel. Based on this relationship, if diameter does not change, an increase in velocity will result in an increase in wall shear rate and stress providing a test of our hypothesis. The relationship between center-line velocity and actual blood flow velocity can vary for a number of reasons (e.g., optical, rheological). We simplified the relationship as V, = I&&l.6 or 1.3, above or below 15 pm in diameter, respectively, although the correction factor is likely to decrease gradually from 1.6 in tubes of >15 pm in diameter to near 1.0 in tubes of the size of capillaries (26). Arteriolar blood flow was calculated according to & = VmrD2/4, expressed in nanoliters per second. Statistical analysis. Data reported here are means t SE, with n indicating the number of experimental interventions. Arteriolar wall shear rate was calculated before and during parallel occlusions with intact (n = 21) and with impaired endothelium (n = 15) in a total of 10 rats. Regression analysis, power curve fitting with use of leastsquares regression test of parallelism, and Student’s t test were performed. P < 0.05 was considered significant. RESULTS
gZZZ
OCCLUD
FIG. 1. Schematic map of a vascular subnetwork of cremaster muscle; arrows indicate direction of flow. For easier understanding we folded out the network at axis running parallel to midline of capillary network (CN). Al-A5 and Vl-V5 indicate different order of arterioles and venules, respectively. In this network third-order arterioles (A3) are located in a parallel circuit. In control conditions and during parallel occlusion using an occluder (before and after endothelial impairment), one of the several arterioles (A3) proximal to occluder was selected (MF, microscopic field, pointed with vertical arrow) for study of changes in diameter and blood flow velocity.
The experimental interventions did not affect systemic arterial pressure (mean 95-120 mmHg) of rats. In the presence of intact endothelium mean control wall shear rate was 2,658 t 123 s-l (mean arteriolar diam 22.5 t 0.6 pm). Parallel arteriolar occlusion elicited an increase in RBC velocity that was followed by a biphasic change in the diameter of arterioles, as illustrated in the diagram in Fig. 2 (left). First, diameter decreased slightly to a mean of 21.8 t 0.7 pm. In this phase just before the start of dilation (see Fig. 2, phase 1, illustrated with dotted line) wall shear rate reached a significantly higher level (mean 4,981 k 253 s-l) than in control as illustrated in Fig. 3. This increase in wall shear rate then was followed by an increase in diameter, with a delay of 6-15 s. The smallest increase in wall shear rate at phase 1 that was followed by dilation was 1,020 s-‘. The increase in diameter reached a peak value (mean 29.5 t 0.8 pm, range 26-36 pm), resulting in a significant decrease in wall shear rate (at phase 2) toward the control value (mean
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
H864
ENDOTHELIAL
REGULATION
OF
SHEAR
STRESS
10
AND
MICROVASCULAR
RESISTANCE
8 I 10
mm/s
5w -------I_ I
0+ CONTROL
-
0-
ENDOTHELIUM
PHl
ENDOTHELIUM
PHl
CONTROL
FIG. 2. Schematic representation of the effect of parallel occlusion on red blood cell (RBC) velocity and diameter of a third-order arteriole. Wall shear rate was calculated at points indicated with dotted lines; in control, at phase 1 (PHI) and phase 2 (PH2) of the arteriolar response during parallel occlusion. Left panel before and right panel after endothelial impairment. See details in METHODS. * P( 0.05
A- f3 T ** P( 0.05 -7
vs CONTROL vs PHASE 1
+ ENDOTHELIUM
0+
B-
r-l 1
*
ENDOTHELIUM A
0
? X
-
&-
9 m w A T
A
V
V A
A
r4 is
0
v A
0 a
I 2-v,
.A
0 1
9 V
0
* w
w
W
V
0 0
0
%
-
V
.
l
0
0
V
P v
V
0
V
0 0
2
CONTROL FIG.
thelium; 1 (PHl)
PHASE
3. Mean values of wall shear rate in control conditions, and during and at phase 2 (PH2). See details
1
PHASE
2
3O 15
in presence of intact endoparallel occlusion at phase in METHODS and RESULTS.
1I 18
I 21 I
ARTERIOLAR
1
I
L
24
27
1
DIAMETER
I 30 1
I 33 1
1 36
(pm)
4. Correlation between wall shear rate and arteriolar diameter in presence of intact endothelium; in control conditions, (solid circles) and during parallel occlusion at phase 1 (PHI, solid triangles) and at phase 2 (PH2, open triangles). Solid lines, regression lines. Correlation coefficient of regression lines indicated no significance (P > 0.05). See detailsin METHODS and RESULTS. FIG.
3,879 t 203 s-l; Fig. 3). In a few instances when parallel occlusion did not elicit an increase in RBC velocity there was also no dilation of the arteriole studied. In the presence of intact endothelium wall shear rate was also analyzed as a function of arteriolar diameter (Fig. 4). In control conditions the distribution of data was monotonic and wall shear rate was not correlated with arteriolar diameter (range 16.1-27.2 pm) as indicated by the regression line (closed circles) y = (-0.02 t 0.04)x + 3.2 (r = -0.12, P > 0.05). Similar results were obtained during parallel occlusion at phase 1, at which point the regression line (filled triangle) was y = (-0.07 t 0.09)x + 6.5 (r = -0.18, P > 0.05), and in phase 2, at which point the regression line (open triangles) was y =
A 7 m r3” l 0 7
* PC 0.05
*
- ENDOTHELIUM 9--
X
P
6 --
L?
CK 4
(-0.02 t 0.06)x + 4.4 (r = -0.06, P > 0.05). Impairment of the vasoactive function of the endothelium elicited a slight decrease in arteriolar diameter (mean diam 17.3 t 0.8 ,um; Fig. 5) and a wall shear rate that was significantly higher (mean 3,604 t 341.8 s-‘, P < 0.05) than in control conditions when the endothelium of the arteriole was intact. During parallel occlusion at peak flow velocity (phase 1) wall shear rate increased further, to a point (10,360.7 t 1,472 s-l, P c 0.05), which was also significantly higher than the corresponding wall shear rate in arterioles with intact endothelium (see Fig. 3, phase 1). This high level of wall shear rate, however,
12-
5 3 s
3
0 L CONTROL
PHASE
1
5. Mean values of wall shear rate after endothelial impairment; in control conditions and during parallel occlusion at phase 1 (PHl). See detailsin METHODS and RESULTS. FIG.
was not followed by instances the slight 15.2 t 1.5 pm) that during the delay time
arteriolar dilation; instead, in most decrease in diameter (mean diam was observed in control conditions now persisted. Therefore wall shear
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
ENDOTHELIAL
REGULATION
OF
SHEAR
STRESS
rate remained elevated for the duration of parallel occlusion. Further analysis of the data (Fig. 6) indicated a significant inverse correlation between wall shear rate and diameter after impairment of arteriolar endothelium both in control (range of diam 12.6-26.3 pm) and during increased blood flow velocity conditions (range of diam 7.8-22 pm). The corresponding regression lines were y = (-0.24 3- 0.09)x + 7.7 (r = -0.58, P < 0.05) and y = (-1.2 t 0.21)x + 28.7 (r = -0.85, P C 0.05), respectively. It is of note that in phase 1 a significant correlation was also found if the three data points on the far left in Fig. 6 were omitted. In the presence of intact endothelium calculated arteriolar blood flow increased by 230% during parallel occlusion (at phase 2). After impairment of the endothelium this increase was only 66% due to the absence of arteriolar dilation (Table 1). In the presence of intact endothelium power curve-fitting analysis indicated a close to third-power relationship between calculated c
24
r)
X
T--x 7 -
ENDOTHELIUM
c)
I I Y 20--
l
-l
16--
u-l
E 12-2 %
W I cn
8--
4 --
4 8 1 I
5
I I
1 r
10
15
i
20
25
ARTERIOLAR DIAMETER (pm) 6. Correlation between wall shear rate and arteriolar diameter after impairment of arteriolar endothelium; in control conditions (solid squares) and during parallel occlusion at phase 1 (PHl, solid diamond). Solid lines, regression lines. Correlation coefficient of both regression lines indicated significance (P < 0.05). See details in METHODS and FIG.
RESULTS.
1. Calculated arteriolar blood flow and its relationship to radius in control conditions and during parallel occlusion before and after impairment of endothelium TABLE
& Intact endothelium Control During parallel occlusion Impaired endothelium Control During parallel occlusion
3.07t0.38 10.07~0.9”
1.77-c-0.2*
2.94+0.4-t
Correlation Coefficient
& = kr”
y = 2.94x y = 2.87x
y = 1.92x* y = 1.47x?
- 6 05 - 5’47 .
- 3.65 - 1.97
0.86 0.78
0.76 0.87
Effect of parallel occlusion on blood flow (&, nl/s) and on the relationship between In flow and In radius are indicated by the slope of the regression lines (n: 2.94, 2.87, 1.92, and 1.47, respectively) before and after impairment of the endothelium. Significant (P < 0.05) differences from control with intact (*) and impaired (j-) endothelium.
AND
MICROVASCULAR
RESISTANCE
H865
blood flow and radius of the arteriole, both in control and during increased flow conditions (Table 1). In contrast, after impairment of the endothelium the exponent of the radius was significantly lower and decreased even further during parallel occlusion [from 1.92 to 1.47, P < 0.05 (Table 1)] when the flow increments were due only to increases in blood flow velocity. DISCUSSION
The salient findings of the present study are that in the presence of intact endothelium, regardless of the diameter and flow velocity conditions, there is a tendency for a constant wall shear rate in third-order arterioles of rat cremaster muscle and that the mechanisms involved in the maintenance of wall shear stress also have an important role in blood flow regulation by optimizing vascular network geometry. In our experiments arteriolar blood viscosity was not measured (nor was it likely to have changed significantly during the experimental interventions); therefore as a first approximation of wall shear stress (2) we calculated wall shear rate. In the presence of endothelium mean control wall shear rate was 2,658 t 123 s-l, a value close to previous findings by others. In rabbit mesenteric arterioles (diam 17-32 pm) Tangelder et al. (40) reported that the range of wall shear rate was 472-4,172 s-l with a median value of 1,700 s-l. Lipowsky et al. (25) found a considerable spatial variability in wall shear stress in cat mesentery. In rat cremaster muscle preparation, however, Mayrovitz and Roy (27) observed a reasonably constant wall shear rate (mean 4,253 s-l) in vessels from 6 to 108 pm in diameter. Similarly, in pial arterioles, 35.4-177.8 pm in diameter, a constant wall shear rate (range 2,695-2,915 s-l) was found (16). Our results show that, in the presence of intact endothelium, although the diameter and flow velocity varied in control and highflow (velocity) conditions, wall shear rate was independent of diameter and essentially constant in third-order arterioles. These findings favor the idea that wall shear stress is a controlled parameter in vascular networks (27, 29,31,41), although the set points of regulation could be different in different-sized vessels. On the other hand, Smiesko et al. (39) suggested that there may not be a rigid set point for the regulation of wall shear rate, at least in the arcade vessels of rat mesentery. These discrepant results may be explained on the basis of the particular structure and function of different microvascular networks (e.g., mesentery vs. skeletal muscle). However, extreme values of shear stress are known to result in impairment of endothelial cell function (8). We found that if the endothelium was intact, an increase in wall shear rate was followed (with a delay of -8 s) by an increase in diameter (see Fig. 2, at phase 1). This dilation led to a significant decrease in wall shear rate, back toward the control level, further strengthening the idea that an increase in wall shear stress is the stimulus for the observed “flow-dependent” dilation and that it is a regulated variable in microcirculation. One may, however, entertain the possibility of the participation of other mechanisms that could elicit the arteriolar dilation observed during parallel occlusion.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
H866
ENDOTHELIAL
REGULATION
OF
SHEAR
There is no reason to suspect that blood pressure would fall in the vessels proximal to the occluder, which could then elicit a “myogenic” dilation of the arteriole under study (1, 13). In contrast, we observed a constriction during the delay time, which was overcome by a secondary dilation but only if the endothelium was present (1720). Because blood flow increased to the tissue area supplied by the arteriole studied, metabolic regulation (9) also cannot be considered as a reason for the observed dilation. Moreover, endothelial impairment abolished arteriolar dilation to an increase in wall shear rate but did not affect the dilation to adenosine, a putative mediator responsible for coupling tissue metabolism to blood flow. In addition, occlusion itself, in the absence of an increase in flow velocity, did not elicit dilation, indicating further that metabolic factors are unlikely to play a role in the dilation observed. Along this line, because occlusion was used to show that conducted vasomotor responses (35) are independent of blood flow, there is no reason to suppose that such responses originating from other areas in the tissue could be responsible for the observed dilation in our experiments. Finally in our recent study we found that prostaglandins mediate arteriolar dilation to increased blood flow velocity in cremaster muscle of the rat (22), providing an additional argument against the participation of the above-mentioned mechanisms in the dilation observed, since none of them is believed to be mediated by prostaglandins, dilator agents that can, however, be released from cultured endothelial cells in response to increases in shear stress (4). It is noteworthy that, contrary to wall shear rate, red blood cell velocity remained high after arteriolar dilation. It is therefore unlikely that flow velocity itself would be the target of regulation. At the present time, although there is no evidence for increased mass transport of an endothelium-dependent vasoactive substance during high-flow conditions (which would occur independent of changes in shear stress), such an occurrence cannot be excluded as an alternative hypothesis to explain our findings. Nevertheless, our results and the above arguments strongly suggest that changes in wall shear stress initiate a mechanism which results in arteriolar dilation observed in the present experimental circumstances. The exact cellular mechanisms by which increases in wall shear stress elicit dilation are not yet clear, but involvement of several endothelial mediators that can relax vascular smooth muscle has been proposed (3-5, 22, 33). We found that after impairment of the endothelium wall shear rate became significantly higher than in control preparations and inversely correlated with arteriolar diameter. Moreover, further increases in wall shear rate during parallel occlusion did not evoke a secondary dilation. Instead, the slight constriction (most likely a myogenic response to the increase in pressure proximal to the occluder), which in control preparations was overcome by an endothelium dependent dilation, now prevailed. Also, a strong inverse correlation was found between diameter and wall shear rate (Fig. 6) when velocity reached a maximum level (Fig. 2, right, phase 1). These findings demonstrate that the endothelium of arterioles plays an important role in the mediation of the shear stress-sensitive arteriolar dilation.
STRESS
AND
MICROVASCULAR
RESISTANCE
Although there is always an “error signal” present in a regulatory system, it is interesting to speculate why the dilation of the arterioles did not result in a closer return of calculated wall shear rate to control levels. The point denoted as phase 2 in Fig. 2 is the point of maximal dilation, but frequently in the later stages of the period of occlusion, a slight decrease in velocity, which further reduced wall shear rate, was also observed (probably due to the flow distribution in the other parallel vessels). Also, the putative stimulus for dilation is the change in wall shear stress that depends not only on wall shear rate but also on blood viscosity, which in this study was assumed to be constant. In vivo, however, apparent viscosity at the wall may decrease concurrent with the increase in shear rate (2), reducing further the actual shear stress acting on the endothelium. The effects of changes in viscosity on large-vessel diameter (14,28) and on transit time in the cerebral microcirculation (32) have also been observed, providing further support for the idea that a change in wall shear stress is a stimulus for initiating vasomotor responses. Other plausible explanations for the still elevated wall shear rate after dilation could be that a sustained signal may be necessary to maintain the dilated state of the arteriole or that shear stress is regulated over a broad range of values (41). The smallest increase in wall shear rate that already was followed by an arteriolar response was ~1,000 s-l. Greater increases in wall shear rate resulted in greater dilation of arterioles, similar in magnitude to that induced by 10m4 M adenosine (23), indicating that this arteriolar response could in certain circumstances approach maximal. All the above-mentioned reasoning as well as previous (17-20, 31) and present findings are consistent with the idea that in vivo the input stimulus for flow-dependent dilation of skeletal muscle arterioles is the increase in wall shear stress, which in turn is regulated in a negative-feedback manner by the change in diameter. It is therefore likely that alterations in wall shear stress are continuously coupled to changes in vascular resistance. This interaction, in concert with other mechanisms, could participate in the regulation of basal vascular tone, and in many complex circulatory responses, when changes in blood flow velocity, due to variations in pressure drop across the vessels (increased upstream pressure and/or decreased downstream resistance) or changes in viscosity, due to variations in hematocrit or plasma constituents, alter wall shear stress. Besides the modulation of the effects of central or local blood flow regulation, as a consequence of the induced change in blood flow velocity and/or viscosity, this mechanism could also mitigate or in some instances even vitiate the myogenic response (phase 1, Fig. 2, left). For example, an increase in upstream intravascular pressure increases blood flow velocity and wall shear stress, resulting in downstream arteriolar dilation. Consequently, the decrease in resistance would lower upstream pressure, counteracting the stimulus for the myogenic constriction, which if unopposed, could substantially elevate resistance and upstream intravascular pressure (Fig. 7). Concurrently, this mechanism would also maintain an adequate capillary pressure. The increase of pressure in the
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
ENDOTHELIAL
REGULATION
OF
SHEAR
STRESS
IF IM I
ARTERIES
I
ARTERIOLES
CAPILLARIES
7. Illustration of modulatory effect on pressure distribution microcirculation of continuous presence of flow (shear stress) sensitive and pressure (myogenic) (M) -sensitive regulation. FIG.
in (F)-
arterioles on the other hand may serve to counterbalance flow-dependent dilation. Thus it seems that the continuous interplay between the “flow-sensitive” (shear stress ) and pressure-sensitive (myogenic) mechanisms could stabilize and optimize hemodynamic parameters in the microcirculation to place less demand on other regulatory systems (Fig. 7). During the increased flow velocity conditions we observed that upstream and downstream from our point of observation other parallel arterioles also dilated. The presence of “flow sensitivity” in large arteries (6, 10, 11, 14, 30, 33, 34, 37, 38) and mesenteric arcade vessels (39) suggests that this phenomenon may operate at every level of the circulation. The possible physiological importance of this mechanism may be best examplified in functional hyperemia. For example, at the onset of exercise systemic blood pressure increases and the resistance to blood flow in distal elements of skeletal muscle microcirculation decreases, resulting in an increase in blood flow velocity and wall shear stress in upstream vessels that have not yet dilated. This increase in wall shear stress would cause a dilation of arterioles and arteries of a higher order (which probably are not, or not yet, affected directly by changes in parenchymal tissue metabolism), recruiting a part of or the entire microvascular network to amplify the hyperemic response and thereby preventing, in concert with other mechanisms, tissue ischemia. The increase in velocity, which could be viewed as an “ascending signal,” is gradually mitigated in upstream vessels and will, dependent on the extent of the area involved and the magnitude of the original stimulus, eventually dissipate. Time-dependent changes in distal and proximal vascular resistances have been observed previously during exercise of cat gastrocnemius muscle (7); however, at that time the mechanisms responsible for the events had not yet been elucidated. The shear stress- sensitive vasodilator mechanism could be one of the means by which the function of microvascular networks in such and other conditions [e.g., reactive hyperemia (Zl), collateral blood flow] could be optimized (41) The data indicate that the shear stress-sensitive mechanism could provide a considerable increase and redistribution of blood flow without a significant change in perfusion pressure. Blood flow increased greatly (by 230%) during parallel occlusion, but the i.ncrease was
AND
MICROVASCULAR
H867
RESISTANCE
considerably attenuated after impairment of the endothelium (Table 1). The smaller increase in flow in the endothelium-impaired preparation was due only to the increase in velocity, and the flow decrement (compared with preparations with intact endothelium) was most likely absorbed as an equivalent increase in flow in other parallel arterioles. A further interpretation of our findings is that Murray’s law (27, 29, 36, 41) is manifested in the present experimental conditions, although we fully recognize that the data presented herein are, in fact, only a point of departure for a more definitive course of investigation. Our findings that in the presence of intact endothelium a close to third-power relationship exists between calculated blood flow and arteriolar diameter in both control and during high-flow conditions agree with those of Mayrovitz and Roy (27) but not with those of others (5). This relationship is thought to minimize the work (power loss) required to maintain blood flow in a vascular network (25, 27, 29, 36) and also implies the importance of controlling wall shear stress (41). It then seems feasible that the dilation of consecutive segments of the vascular tree that elicits the decreases in wall shear stress and resistance can provide for substantial increases in blood flow without the expense of further power dissipation, as suggested by theoretical studies (41). After impairment of the endothelium, however, either in control or during “high-flow” conditions, wall shear stress was high and the diameter of the impaired segment of the arteriole deviated from “optimal” (Table 1). Interestingly, a similar decrease in the exponent was found to be related to the degree of disease in human coronary arteries (12). If endothelial dysfunction due to various pathological conditions (atherosclerosis, hypertension, etc.) were present in the whole network, then to maintain optimal flow (i.e., to avoid ischemia) an increase in work (pressure) would be required due to the increased power dissipation (36, 41). This could indeed be the case, since the myogenic response would not be opposed and vascular resistance would increase with a corresponding increase in pressure drop (i.e., an increase in systemic and a decrease in capillary blood pressure) (Fig. 7). In summary the present in vivo study indicates that wall shear stress is a regulated variable in skeletal muscle microcirculation and that an increase in wall shear stress is the underlying stimulus for flow-dependent arteriolar dilation and the consequent flow increase. Our data also provide an additional basis for the assumption that the flow-dependent ascending dilation observed in large vessels is also initiated by changes in wall shear stress. Together with theoretical studies emphasizing the relationship between shear stress and vessel radius, the present study offers in vivo evidence for the idea that wall shear stress is regulated to provide for an optimal structure of vascular networks that accords with their function. The authors acknowledge the superior Ecke. This work was supported by National PO-I-HL-43023 and by New York Heart Address reprint requests to A. Koller. Received
7 February
1990; accepted
secretarial Institutes Association
in final
form
skills
of Annette
of Health Grant Grant 89-062G. 29 October
1990.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
H868
ENDOTHELIAL
REGULATION
OF
SHEAR
REFERENCES 1. BORGSTROM, P., P. 0. GRANDE, AND S. MELLANDER. A mathematical description of the myogenic response in the microcirculation. Acta Physiol. Stand. 116: 363-376, 1982. 2. CHIEN, S. Determination of microvascular viscosity. In: Microcirculatory Technology, edited by C. H. Baker and W. L. Nastuk. New York: Academic, 1986, p. 179-196. 3. DAVIES, P. F. How do vascular endothelial cells respond to flow? News Physiol. Sci. 4: 22-25, 1989. 4. FRANGOS, J. A., S. G. ESKIN, L. V. MCINTIRE, AND C. L. IVES. Flow effects on prostacyclin production by cultured human endothelial cells. Science Wash. DC 227: 1477-1479, 1985. 5. GRIFFITH, T. M., D. H. EDWARDS, R. L. I. DAVIES, T. J. HARRISON, AND K. T. EVANS. EDRF coordinates the behavior of vascular resistance vessels. Nature Lond. 329: 442-445, 1987. 6. FLEISCH, A. Les reflexes nutritifs ascendants producteurs de dilatation arterielle. Arch. Int. Physiol. 41: 141-167, 1935. 7. FOLKOW, B., R. R. SONNENSCHEIN, AND D. L. WRIGHT. Loci of neurogenic and metabolic effects on precapillary vessels of skeletal muscle. Acta Physiol. Stand. 81: 459-471, 1971. 8. FRY, D. L. Acute vascular endothelial changes associated with increased blood velocity gradients. Circ. Res. 22: 165-197, 1968. 9. HADDY, F. J., AND J. C. SCOTT. Metabolically linked vasoactive chemicals in local regulation of blood flow. Physiol. Rev. 48: 688707, 1968. 10. HILTON, S. M. A peripheral arterial conducting mechanism underlying dilation of the femoral artery and concerned in functional vasodilatation in skeletal muscle. J. Physiol. Lond. 149: 93-111, 1959. 11. HINTZE, T. H., AND S. F. VATNER. Reactive dilation of larger coronary arteries in conscious dogs. Circ. Res. 54: 50-57, 1984. 12. HUTCHINS, G. M., M. M. MINER, AND J. K. BOITNOTT. Vessel caliber and branch-angle of human coronary artery branch-points. Circ. Res. 38: 572-576, 1976. 13. JOHNSON, P. C. The myogenic response. In: Handbook of Physiology. The Cardiovascular System. Vascular Smooth Muscle. Bethesda, MD: Am. Physiol. Sot., 1980, sect. 2, vol. II, chapt. 15, p. 409-442. 14. KAISER, L., AND H. V. SPARKS. Effect of hemodilution on endothelium-dependent vasodilation in the in vivo canine femoral artery. Circ. Shock. 23: 107-118, 1987. 15. KALEY, G., J. M. RODENBURG, E. J. MESSINA, AND M. S. WOLIN. Endothelium associated vasodilators in rat skeletal muscle microcirculation. Am. J. Physiol. 256 (Heart Circ. Physiol. 25): H720H725, 1989. 16. KOBARI, M., F. GOTOH, Y. FUKUUCHYI, K. TANAKA, N. SUZUKI, AND D. UEMATSU. Blood flow velocity in the pial arteries of cats, with particular references to the vessel diameter. J. Cereb. Blood Flow Metab. 4: 110-114, 1984. 17. KOLLER, A., AND G. KALEY. Blood flow velocity-induced dilation in skeletal muscle microcirculation in vivo: role of endothelium (Abstract). Physiologist 32(4): 178, 1989. 18. KOLLER, A., AND G. KALEY. Flow velocity dependent regulation of microvascular resistance in vivo. Microvasc. Endothel. Lympathics 5: 519-530,1989. 19. KOLLER, A., AND G. KALEY. Endothelium regulates skeletal muscle microcirculation by a blood flow velocity sensing mechanism. Am. J. Physiol. 258 (Heart Circ. Physiol. 27): H916-H920, 1990. 20. KOLLER, A., AND G. KALEY. Negative feedback control of wall shear stress by the endothelium of skeletal muscle arterioles (Abstract). FASEB J. 4: A1257, 1990.
STRESS
AND
MICROVASCULAR
RESISTANCE
21. KOLLER, A., AND G. KALEY. Role of endothelium in reactive dilation of skeletal muscle arterioles. Am. J. Physiol. 259 (Heart Circ. Physiol. 28): H1313-H1316, 1990. 22. KOLLER, A., AND G. KALEY. Prostaglandins mediate arteriolar dilation to increased blood flow velocity in skeletal muscle microcirculation. Circ. Res. 67: 529-534, 1990. 23. KOLLER, A., E. J. MESSINA, M. S. WOLIN, AND G. KALEY. Effects of endothelial impairment on arteriolar dilator responses in vivo. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): H1484-H1489, 1989. 24. KOLLER, A., E. J. MESSINA, M. S. WOLIN, AND G. KALEY. Endothelial impairment inhibits prostaglandin and EDRF-mediated arteriolar dilation in vivo. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): H1966-H1970,1989. 25. LIPOWSKY, H. H., S. KOVALCHECK, AND B. W. ZWEIFACH. The distribution of blood rheological parameters in the microvasculature of cat mesentery. Circ. Res. 43: 738-749, 1978. 26. LIPOWSKY, H. H., AND B. W. ZWEIFACH. Application of the “twoslit” photometric technique to measurement of microvascular volumetric flow rates. Microvasc. Res. 15: 93-101, 1978. 27. MAYROVITZ, H. N., AND J. ROY. Microvascular blood flow: evidence indicating a cubic dependence on arteriolar diameter. Am. J. Physiol. 245 (Heart Circ. Physiol. 14): H1031-H1038, 1983. 28. MELKUMYANTS, A. M., S. A. BALASHOV, AND V. M. KHAYUTIN. Endothelium dependent control of arterial diameter by blood viscosity. Cardiovasc. Res. 23: 741-747, 1989. 29. MURRAY, C. D. The physiological principle of minimum work. The vascular system and the cost of blood volume. Proc. NatZ. Acad. Sci. USA 12: 207-214,1926. 30. POHL, U., J. HOLTZ, R. BUSSE, AND E. BASSANGE. Crucial role of endothelium in the vasodilator response to increased flow in vivo. Hypertension Dallas 8: 37-44, 1986. 31. RODBARD, S. Vascular caliber. Cardiology 60: 4-49, 1975. 32. ROSENBLUM, W. I. Effects of blood pressure and blood viscosity on fluorescein transit time in the cerebral microcirculation in the mouse. Circ. Res. 27: 825-833, 1970. 33. RUBANYI, G. M., J. C. ROMERO, AND P. M. VANHOUTTE. Flow induced release of endothelium derived relaxing factor. Am. J. Physiol. 250 (Heart Circ. Physiol. 19): H1145-H1149, 1986. 34. SCHRETZENMAYR, A. Uber kreislaufregulatorishe Vorgange an den grossen Arterien bei der Muskelarbeit. Pfluegers Arch. 232: 743748,1933. 35. SEGAL, S. S., AND B. R. DULING. Flow control among microvessels coordinated by intercellular conduction. Science Wash. DC 234: 868-870,1986. 36. SHERMAN, T. F., A. S. POPEL, A. KOLLER, AND P. C. JOHNSON. The cost of departure from optimal radii in microvascular networks. J. Theor. BioZ. 136: 245-265, 1989. 37. SINOWAY, L. I., C. HENDRICKSON, W. R. DAVIDSON, S. PROPHET, AND R. ZELIS. Characteristics of flow mediated brachial artery vasodilation in human subjects. Circ. Res. 64: 32-42, 1989. 38. SMIESKO, V., J. KOZIK, AND S. DOLEZEL. The control of the arterial diameter by blood flow velocity is dependent upon intact endothelium (Abstract). Physiol. BohemsZov. 32: 558, 1983. 39. SMIESKO, V., D. J. LANG, AND P. C. JOHNSON. Dilator responses of rat mesenteric arterioles to increased blood flow velocity. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): H1958-H1965, 1989. 40. TANGELDER, G. J., D. W. SLAAF, AND R. S. RENEMAN. Wall shear rate in arterioles in vivo: least estimates from platelet velocity profiles. Am. J. Physiol. 254 (Heart Circ. Physiol. 23): H1059H1064,1988. 41. ZAMIR, M. Shear stress forces and blood vessel radii in the cardiovascular system. J. Gen. Physiol. 69: 449-461, 1977.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
KOLLER AKOS,AND GABOR KALEY. Endothelial regulation . . n 1. 1 1 of wall shear stress anu oioou J~OW &n sneietai muscle mtcrocvcuZation. Am. J. Physiol. 260 (Heart Circ. Physiol. 29): H862H868, 1991.-In the presence of intact endothelium, in control conditions, calculated wall shear rate (WSR) (means t SE: 2,658 t 123 s-l; n = 21) was independent of arteriolar diameter (16.2-27.2 pm; correlation coefficient: r = 0.12, P > 0.05) in cremaster muscle of pentobarbital-anesthetized rats. An increase in blood flow velocity (due to parallel arteriolar occlusion) elicited a significant increase in WSR (to 4,981 t 253 s-l) followed by a delayed (6-15 s) increase in diameter (from: 22.5 t 0.6 to 29.5 t 0.8 pm), which consequently resulted in a significant decrease in WSR (to 3,879 t 203 s-l). As a result of the increased flow velocity and dilation, calculated arteriolar blood flow increased by 230%. After impairment of the endothelium of arterioles by a light-dye technique, basal WSR became significantly higher (3,604 t 341 s-l), and despite a greater increase in WSR (10,360 t 1,471 s-l) the dilation was absent. Now an inverse linear correlation was found between arteriolar diameter and WSR both before (r = 0.58, P < 0.05) and during increased flow velocity conditions (r = 0.85, P < 0.05). Also, arteriolar blood flow that was already less after impairment of endothelium increased by only 66% during the period of increased flow velocity due to the absence of dilation. Results suggest that an increase in wall shear stress is the stimulus for the endothelium-dependent mechanism that elicits “flow dependent” arteriolar dilation. This mechanism has a role in the negative-feedback control of wall shear stress and also participates in the regulation of resistance and blood flow in the microcirculation. 111
1
l-1
.
1
On the other hand, if pressure increases, blood flow velocity, which is directly proportional to wall shear stress, could increase as well. In our previous studies in skeletal muscle microcirculation (17-20), we found a significant correlation between increases in blood flow velocity and arteriolar diameter in the presence of intact endothelium, suggesting the existence of a shear stresssensitive mechanism that, independent of other factors, could serve to counterbalance the myogenic response and participate in the regulation of resistance and blood flow. Theoretical (29, 31, 36, 41) and experimental (3-5, 11, 14, 27, 28, 30, 32, 33, 37-39) studies also suggest an important relationship between shear stress and vessel radius, but the nature of this relationship and whether it can participate in the regulation of skeletal muscle microcirculation in vivo have not as yet been clarified. Therefore, in the present study, we sought to investigate whether an increase in wall shear stress could be a stimulus for initiating a regulatory mechanism leading to arteriolar dilation and the consequent decrease in wall shear stress and whether arteriolar endothelium has a role in this process in skeletal muscle microcirculation in vivo. For this purpose we calculated wall shear rate in arterioles during control and increased flow velocity conditions in the presence of intact as well as impaired endothelium. METHODS
hemodynamic forces; wall shear rate; autoregulation; occlusion; arteriolar dilation; red blood cell velocity; law
parallel Murray’s
that resistance to blood flow is the target of numerous control mechanisms in the peripheral circulation. Resistance is determined by factors affecting either vascular (diameter, length, network geometry, etc.) or rheological (viscosity, hematocrit, wall shear stress, etc.) parameters. The interaction among these factors to provide the actual resistance, however, is not completely understood. It is also recognized that an effective blood flow regulation requires several regulatory mechanisms that can interact with each other. Besides central factors, two main local mechanisms are believed to be involved in the regulation of microvascular resistance, namely the “pressure-sensitive” myogenic mechanism, which inversely couples pressure and vessel diameter (1, 13), and metabolic vasoregulation, which via changes in resistance matches the blood supply to tissue demand (9).
IT IS KNOWN
H862
0363-6135/91
$1.50 Copyright
Data utilized in the present study to calculate wall shear rate, as an indicator of wall shear stress, were obtained in experiments reported previously (19) and were taken from original strip-chart recordings or by replaying videotape recordings, whenever it was technically feasible. General preparation. The in vivo tissue preparation utilized in these experiments has been described in detail previously (15, 23) and is summarized as follows. Five-to six-week-old male Wistar rats were anesthetized with pentobarbital sodium (35 mg/kg). Arterial blood pressure was monitored with Statham P23 D6 transducer connected to a cannula inserted in the left common carotid artery. The left cremaster muscle was surgically exteriorized, and the muscle was suffused (2 ml/min) with Ringer gelatin solution, containing (in mM) 154 Nacl, 5.6 KCl, and 2.2 CaCIZ and 5 g/l gelatin. The pH of the solution (7.35-7.42) was adjusted with sodium bicarbonate and was equilibrated with ambient air at a temperature of 33°C. Vessels selected for this study were third-
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
ENDOTHELIAL
REGULATION
OF
SHEAR
STRESS
order arterioles, ranging in size from 16.1 to 27.2 pm in diameter. Intravascular diameter of microvessels was visualized by television microscopy (Olympus BH2) and was measured on- or off-line by an image-shearing monitor (IPM model 907, San Diego, CA) calibrated with a stage micrometer. Red blood cell (RBC) velocity was measured along the microvessel’s center line by using a self-tracking correlator (IPM model 102B) (24). The output of the measurements was recorded on a Sensormedics Dynograph Recorder (model 5 RllA). Light-dye treatment. The mercury light-sodium fluorescein (light-dye) treatment was used to impair the vasoactive functions of the endothelium in a localized segment of arterioles. This technique involves the illumination of a segment of an arteriole (80-100 pm in length) in the presence of i ntravascular sodium fluorescein, as described in detail previously (23, 24). Selective endothelial impairment was considered to have been accomplished if dilation to topically applied (100 ~1) endothelium-dependent dilator agents (lo-” M arachidonic acid or 10V6 M acetylcholine) was completely abolished while the arteriolar dilation to direct vascular smooth muscle relaxants (2 x 10B7 M sodium nitroprusside, or 10q5 M adenosine) was not affected (15, 23, 24). Acetylcholine chloride, arachidonic acid, adenosine, sodium nitroprusside, and sodium fluorescein were obtained from Sigma Chemical (St. Louis, MO). Experimental protocol. After surgery, a 30- to 40-min time period was allowed for the preparation to reach a steady-state condition. Control diameter and RBC velocity were measured in a third-order arteriole. Responses of this arteriole to vasoactive agents and to “parallel arteriolar occlusion” (Fig. 1) of 30- to 180-s duration were then examined. Typically this intervention resulted in an increase in RBC velocity in parallel-connected third-order arterioles located proximal from the occluder (only one of these was studied at any given time) (Fig.
AND
MICROVASCULAR
RESISTANCE
H863
2). In phase 1, just before the start of dilation, and phase 2, at peak dilation, RBC velocity and diameter data were obtained as indicated in Fig. 2 (left). As a next step the endothelium of the arteriolar segment studied was impaired by light-dye treatment and arteriolar responses to vasoactive agents and parallel occlusion were reexamined (Fig. 2, right). Because there was no second phase (i.e., no dilation) in the response of the arteriole with impaired endothelium, phase 1 was designated as a point at peak RBC velocity. Calculation of arteriolar wall shear rate and blood flow. From paired velocity and diameter data wall shear rate (WSR), a measure of wall shear stress, was calculated (2, 27) before and after impairment of the endothelium in control and during parallel occlusion (phases 1 and 2) as illustrated in Fig. 2. WSR = 8 (I&/D), where Vm is the mean or bulk velocity (cells plus plasma) calculated from the measured RBC velocity (I&c), using an empirical correction factor, and D is the luminal diameter of the vessel. Based on this relationship, if diameter does not change, an increase in velocity will result in an increase in wall shear rate and stress providing a test of our hypothesis. The relationship between center-line velocity and actual blood flow velocity can vary for a number of reasons (e.g., optical, rheological). We simplified the relationship as V, = I&&l.6 or 1.3, above or below 15 pm in diameter, respectively, although the correction factor is likely to decrease gradually from 1.6 in tubes of >15 pm in diameter to near 1.0 in tubes of the size of capillaries (26). Arteriolar blood flow was calculated according to & = VmrD2/4, expressed in nanoliters per second. Statistical analysis. Data reported here are means t SE, with n indicating the number of experimental interventions. Arteriolar wall shear rate was calculated before and during parallel occlusions with intact (n = 21) and with impaired endothelium (n = 15) in a total of 10 rats. Regression analysis, power curve fitting with use of leastsquares regression test of parallelism, and Student’s t test were performed. P < 0.05 was considered significant. RESULTS
gZZZ
OCCLUD
FIG. 1. Schematic map of a vascular subnetwork of cremaster muscle; arrows indicate direction of flow. For easier understanding we folded out the network at axis running parallel to midline of capillary network (CN). Al-A5 and Vl-V5 indicate different order of arterioles and venules, respectively. In this network third-order arterioles (A3) are located in a parallel circuit. In control conditions and during parallel occlusion using an occluder (before and after endothelial impairment), one of the several arterioles (A3) proximal to occluder was selected (MF, microscopic field, pointed with vertical arrow) for study of changes in diameter and blood flow velocity.
The experimental interventions did not affect systemic arterial pressure (mean 95-120 mmHg) of rats. In the presence of intact endothelium mean control wall shear rate was 2,658 t 123 s-l (mean arteriolar diam 22.5 t 0.6 pm). Parallel arteriolar occlusion elicited an increase in RBC velocity that was followed by a biphasic change in the diameter of arterioles, as illustrated in the diagram in Fig. 2 (left). First, diameter decreased slightly to a mean of 21.8 t 0.7 pm. In this phase just before the start of dilation (see Fig. 2, phase 1, illustrated with dotted line) wall shear rate reached a significantly higher level (mean 4,981 k 253 s-l) than in control as illustrated in Fig. 3. This increase in wall shear rate then was followed by an increase in diameter, with a delay of 6-15 s. The smallest increase in wall shear rate at phase 1 that was followed by dilation was 1,020 s-‘. The increase in diameter reached a peak value (mean 29.5 t 0.8 pm, range 26-36 pm), resulting in a significant decrease in wall shear rate (at phase 2) toward the control value (mean
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
H864
ENDOTHELIAL
REGULATION
OF
SHEAR
STRESS
10
AND
MICROVASCULAR
RESISTANCE
8 I 10
mm/s
5w -------I_ I
0+ CONTROL
-
0-
ENDOTHELIUM
PHl
ENDOTHELIUM
PHl
CONTROL
FIG. 2. Schematic representation of the effect of parallel occlusion on red blood cell (RBC) velocity and diameter of a third-order arteriole. Wall shear rate was calculated at points indicated with dotted lines; in control, at phase 1 (PHI) and phase 2 (PH2) of the arteriolar response during parallel occlusion. Left panel before and right panel after endothelial impairment. See details in METHODS. * P( 0.05
A- f3 T ** P( 0.05 -7
vs CONTROL vs PHASE 1
+ ENDOTHELIUM
0+
B-
r-l 1
*
ENDOTHELIUM A
0
? X
-
&-
9 m w A T
A
V
V A
A
r4 is
0
v A
0 a
I 2-v,
.A
0 1
9 V
0
* w
w
W
V
0 0
0
%
-
V
.
l
0
0
V
P v
V
0
V
0 0
2
CONTROL FIG.
thelium; 1 (PHl)
PHASE
3. Mean values of wall shear rate in control conditions, and during and at phase 2 (PH2). See details
1
PHASE
2
3O 15
in presence of intact endoparallel occlusion at phase in METHODS and RESULTS.
1I 18
I 21 I
ARTERIOLAR
1
I
L
24
27
1
DIAMETER
I 30 1
I 33 1
1 36
(pm)
4. Correlation between wall shear rate and arteriolar diameter in presence of intact endothelium; in control conditions, (solid circles) and during parallel occlusion at phase 1 (PHI, solid triangles) and at phase 2 (PH2, open triangles). Solid lines, regression lines. Correlation coefficient of regression lines indicated no significance (P > 0.05). See detailsin METHODS and RESULTS. FIG.
3,879 t 203 s-l; Fig. 3). In a few instances when parallel occlusion did not elicit an increase in RBC velocity there was also no dilation of the arteriole studied. In the presence of intact endothelium wall shear rate was also analyzed as a function of arteriolar diameter (Fig. 4). In control conditions the distribution of data was monotonic and wall shear rate was not correlated with arteriolar diameter (range 16.1-27.2 pm) as indicated by the regression line (closed circles) y = (-0.02 t 0.04)x + 3.2 (r = -0.12, P > 0.05). Similar results were obtained during parallel occlusion at phase 1, at which point the regression line (filled triangle) was y = (-0.07 t 0.09)x + 6.5 (r = -0.18, P > 0.05), and in phase 2, at which point the regression line (open triangles) was y =
A 7 m r3” l 0 7
* PC 0.05
*
- ENDOTHELIUM 9--
X
P
6 --
L?
CK 4
(-0.02 t 0.06)x + 4.4 (r = -0.06, P > 0.05). Impairment of the vasoactive function of the endothelium elicited a slight decrease in arteriolar diameter (mean diam 17.3 t 0.8 ,um; Fig. 5) and a wall shear rate that was significantly higher (mean 3,604 t 341.8 s-‘, P < 0.05) than in control conditions when the endothelium of the arteriole was intact. During parallel occlusion at peak flow velocity (phase 1) wall shear rate increased further, to a point (10,360.7 t 1,472 s-l, P c 0.05), which was also significantly higher than the corresponding wall shear rate in arterioles with intact endothelium (see Fig. 3, phase 1). This high level of wall shear rate, however,
12-
5 3 s
3
0 L CONTROL
PHASE
1
5. Mean values of wall shear rate after endothelial impairment; in control conditions and during parallel occlusion at phase 1 (PHl). See detailsin METHODS and RESULTS. FIG.
was not followed by instances the slight 15.2 t 1.5 pm) that during the delay time
arteriolar dilation; instead, in most decrease in diameter (mean diam was observed in control conditions now persisted. Therefore wall shear
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
ENDOTHELIAL
REGULATION
OF
SHEAR
STRESS
rate remained elevated for the duration of parallel occlusion. Further analysis of the data (Fig. 6) indicated a significant inverse correlation between wall shear rate and diameter after impairment of arteriolar endothelium both in control (range of diam 12.6-26.3 pm) and during increased blood flow velocity conditions (range of diam 7.8-22 pm). The corresponding regression lines were y = (-0.24 3- 0.09)x + 7.7 (r = -0.58, P < 0.05) and y = (-1.2 t 0.21)x + 28.7 (r = -0.85, P C 0.05), respectively. It is of note that in phase 1 a significant correlation was also found if the three data points on the far left in Fig. 6 were omitted. In the presence of intact endothelium calculated arteriolar blood flow increased by 230% during parallel occlusion (at phase 2). After impairment of the endothelium this increase was only 66% due to the absence of arteriolar dilation (Table 1). In the presence of intact endothelium power curve-fitting analysis indicated a close to third-power relationship between calculated c
24
r)
X
T--x 7 -
ENDOTHELIUM
c)
I I Y 20--
l
-l
16--
u-l
E 12-2 %
W I cn
8--
4 --
4 8 1 I
5
I I
1 r
10
15
i
20
25
ARTERIOLAR DIAMETER (pm) 6. Correlation between wall shear rate and arteriolar diameter after impairment of arteriolar endothelium; in control conditions (solid squares) and during parallel occlusion at phase 1 (PHl, solid diamond). Solid lines, regression lines. Correlation coefficient of both regression lines indicated significance (P < 0.05). See details in METHODS and FIG.
RESULTS.
1. Calculated arteriolar blood flow and its relationship to radius in control conditions and during parallel occlusion before and after impairment of endothelium TABLE
& Intact endothelium Control During parallel occlusion Impaired endothelium Control During parallel occlusion
3.07t0.38 10.07~0.9”
1.77-c-0.2*
2.94+0.4-t
Correlation Coefficient
& = kr”
y = 2.94x y = 2.87x
y = 1.92x* y = 1.47x?
- 6 05 - 5’47 .
- 3.65 - 1.97
0.86 0.78
0.76 0.87
Effect of parallel occlusion on blood flow (&, nl/s) and on the relationship between In flow and In radius are indicated by the slope of the regression lines (n: 2.94, 2.87, 1.92, and 1.47, respectively) before and after impairment of the endothelium. Significant (P < 0.05) differences from control with intact (*) and impaired (j-) endothelium.
AND
MICROVASCULAR
RESISTANCE
H865
blood flow and radius of the arteriole, both in control and during increased flow conditions (Table 1). In contrast, after impairment of the endothelium the exponent of the radius was significantly lower and decreased even further during parallel occlusion [from 1.92 to 1.47, P < 0.05 (Table 1)] when the flow increments were due only to increases in blood flow velocity. DISCUSSION
The salient findings of the present study are that in the presence of intact endothelium, regardless of the diameter and flow velocity conditions, there is a tendency for a constant wall shear rate in third-order arterioles of rat cremaster muscle and that the mechanisms involved in the maintenance of wall shear stress also have an important role in blood flow regulation by optimizing vascular network geometry. In our experiments arteriolar blood viscosity was not measured (nor was it likely to have changed significantly during the experimental interventions); therefore as a first approximation of wall shear stress (2) we calculated wall shear rate. In the presence of endothelium mean control wall shear rate was 2,658 t 123 s-l, a value close to previous findings by others. In rabbit mesenteric arterioles (diam 17-32 pm) Tangelder et al. (40) reported that the range of wall shear rate was 472-4,172 s-l with a median value of 1,700 s-l. Lipowsky et al. (25) found a considerable spatial variability in wall shear stress in cat mesentery. In rat cremaster muscle preparation, however, Mayrovitz and Roy (27) observed a reasonably constant wall shear rate (mean 4,253 s-l) in vessels from 6 to 108 pm in diameter. Similarly, in pial arterioles, 35.4-177.8 pm in diameter, a constant wall shear rate (range 2,695-2,915 s-l) was found (16). Our results show that, in the presence of intact endothelium, although the diameter and flow velocity varied in control and highflow (velocity) conditions, wall shear rate was independent of diameter and essentially constant in third-order arterioles. These findings favor the idea that wall shear stress is a controlled parameter in vascular networks (27, 29,31,41), although the set points of regulation could be different in different-sized vessels. On the other hand, Smiesko et al. (39) suggested that there may not be a rigid set point for the regulation of wall shear rate, at least in the arcade vessels of rat mesentery. These discrepant results may be explained on the basis of the particular structure and function of different microvascular networks (e.g., mesentery vs. skeletal muscle). However, extreme values of shear stress are known to result in impairment of endothelial cell function (8). We found that if the endothelium was intact, an increase in wall shear rate was followed (with a delay of -8 s) by an increase in diameter (see Fig. 2, at phase 1). This dilation led to a significant decrease in wall shear rate, back toward the control level, further strengthening the idea that an increase in wall shear stress is the stimulus for the observed “flow-dependent” dilation and that it is a regulated variable in microcirculation. One may, however, entertain the possibility of the participation of other mechanisms that could elicit the arteriolar dilation observed during parallel occlusion.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
H866
ENDOTHELIAL
REGULATION
OF
SHEAR
There is no reason to suspect that blood pressure would fall in the vessels proximal to the occluder, which could then elicit a “myogenic” dilation of the arteriole under study (1, 13). In contrast, we observed a constriction during the delay time, which was overcome by a secondary dilation but only if the endothelium was present (1720). Because blood flow increased to the tissue area supplied by the arteriole studied, metabolic regulation (9) also cannot be considered as a reason for the observed dilation. Moreover, endothelial impairment abolished arteriolar dilation to an increase in wall shear rate but did not affect the dilation to adenosine, a putative mediator responsible for coupling tissue metabolism to blood flow. In addition, occlusion itself, in the absence of an increase in flow velocity, did not elicit dilation, indicating further that metabolic factors are unlikely to play a role in the dilation observed. Along this line, because occlusion was used to show that conducted vasomotor responses (35) are independent of blood flow, there is no reason to suppose that such responses originating from other areas in the tissue could be responsible for the observed dilation in our experiments. Finally in our recent study we found that prostaglandins mediate arteriolar dilation to increased blood flow velocity in cremaster muscle of the rat (22), providing an additional argument against the participation of the above-mentioned mechanisms in the dilation observed, since none of them is believed to be mediated by prostaglandins, dilator agents that can, however, be released from cultured endothelial cells in response to increases in shear stress (4). It is noteworthy that, contrary to wall shear rate, red blood cell velocity remained high after arteriolar dilation. It is therefore unlikely that flow velocity itself would be the target of regulation. At the present time, although there is no evidence for increased mass transport of an endothelium-dependent vasoactive substance during high-flow conditions (which would occur independent of changes in shear stress), such an occurrence cannot be excluded as an alternative hypothesis to explain our findings. Nevertheless, our results and the above arguments strongly suggest that changes in wall shear stress initiate a mechanism which results in arteriolar dilation observed in the present experimental circumstances. The exact cellular mechanisms by which increases in wall shear stress elicit dilation are not yet clear, but involvement of several endothelial mediators that can relax vascular smooth muscle has been proposed (3-5, 22, 33). We found that after impairment of the endothelium wall shear rate became significantly higher than in control preparations and inversely correlated with arteriolar diameter. Moreover, further increases in wall shear rate during parallel occlusion did not evoke a secondary dilation. Instead, the slight constriction (most likely a myogenic response to the increase in pressure proximal to the occluder), which in control preparations was overcome by an endothelium dependent dilation, now prevailed. Also, a strong inverse correlation was found between diameter and wall shear rate (Fig. 6) when velocity reached a maximum level (Fig. 2, right, phase 1). These findings demonstrate that the endothelium of arterioles plays an important role in the mediation of the shear stress-sensitive arteriolar dilation.
STRESS
AND
MICROVASCULAR
RESISTANCE
Although there is always an “error signal” present in a regulatory system, it is interesting to speculate why the dilation of the arterioles did not result in a closer return of calculated wall shear rate to control levels. The point denoted as phase 2 in Fig. 2 is the point of maximal dilation, but frequently in the later stages of the period of occlusion, a slight decrease in velocity, which further reduced wall shear rate, was also observed (probably due to the flow distribution in the other parallel vessels). Also, the putative stimulus for dilation is the change in wall shear stress that depends not only on wall shear rate but also on blood viscosity, which in this study was assumed to be constant. In vivo, however, apparent viscosity at the wall may decrease concurrent with the increase in shear rate (2), reducing further the actual shear stress acting on the endothelium. The effects of changes in viscosity on large-vessel diameter (14,28) and on transit time in the cerebral microcirculation (32) have also been observed, providing further support for the idea that a change in wall shear stress is a stimulus for initiating vasomotor responses. Other plausible explanations for the still elevated wall shear rate after dilation could be that a sustained signal may be necessary to maintain the dilated state of the arteriole or that shear stress is regulated over a broad range of values (41). The smallest increase in wall shear rate that already was followed by an arteriolar response was ~1,000 s-l. Greater increases in wall shear rate resulted in greater dilation of arterioles, similar in magnitude to that induced by 10m4 M adenosine (23), indicating that this arteriolar response could in certain circumstances approach maximal. All the above-mentioned reasoning as well as previous (17-20, 31) and present findings are consistent with the idea that in vivo the input stimulus for flow-dependent dilation of skeletal muscle arterioles is the increase in wall shear stress, which in turn is regulated in a negative-feedback manner by the change in diameter. It is therefore likely that alterations in wall shear stress are continuously coupled to changes in vascular resistance. This interaction, in concert with other mechanisms, could participate in the regulation of basal vascular tone, and in many complex circulatory responses, when changes in blood flow velocity, due to variations in pressure drop across the vessels (increased upstream pressure and/or decreased downstream resistance) or changes in viscosity, due to variations in hematocrit or plasma constituents, alter wall shear stress. Besides the modulation of the effects of central or local blood flow regulation, as a consequence of the induced change in blood flow velocity and/or viscosity, this mechanism could also mitigate or in some instances even vitiate the myogenic response (phase 1, Fig. 2, left). For example, an increase in upstream intravascular pressure increases blood flow velocity and wall shear stress, resulting in downstream arteriolar dilation. Consequently, the decrease in resistance would lower upstream pressure, counteracting the stimulus for the myogenic constriction, which if unopposed, could substantially elevate resistance and upstream intravascular pressure (Fig. 7). Concurrently, this mechanism would also maintain an adequate capillary pressure. The increase of pressure in the
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
ENDOTHELIAL
REGULATION
OF
SHEAR
STRESS
IF IM I
ARTERIES
I
ARTERIOLES
CAPILLARIES
7. Illustration of modulatory effect on pressure distribution microcirculation of continuous presence of flow (shear stress) sensitive and pressure (myogenic) (M) -sensitive regulation. FIG.
in (F)-
arterioles on the other hand may serve to counterbalance flow-dependent dilation. Thus it seems that the continuous interplay between the “flow-sensitive” (shear stress ) and pressure-sensitive (myogenic) mechanisms could stabilize and optimize hemodynamic parameters in the microcirculation to place less demand on other regulatory systems (Fig. 7). During the increased flow velocity conditions we observed that upstream and downstream from our point of observation other parallel arterioles also dilated. The presence of “flow sensitivity” in large arteries (6, 10, 11, 14, 30, 33, 34, 37, 38) and mesenteric arcade vessels (39) suggests that this phenomenon may operate at every level of the circulation. The possible physiological importance of this mechanism may be best examplified in functional hyperemia. For example, at the onset of exercise systemic blood pressure increases and the resistance to blood flow in distal elements of skeletal muscle microcirculation decreases, resulting in an increase in blood flow velocity and wall shear stress in upstream vessels that have not yet dilated. This increase in wall shear stress would cause a dilation of arterioles and arteries of a higher order (which probably are not, or not yet, affected directly by changes in parenchymal tissue metabolism), recruiting a part of or the entire microvascular network to amplify the hyperemic response and thereby preventing, in concert with other mechanisms, tissue ischemia. The increase in velocity, which could be viewed as an “ascending signal,” is gradually mitigated in upstream vessels and will, dependent on the extent of the area involved and the magnitude of the original stimulus, eventually dissipate. Time-dependent changes in distal and proximal vascular resistances have been observed previously during exercise of cat gastrocnemius muscle (7); however, at that time the mechanisms responsible for the events had not yet been elucidated. The shear stress- sensitive vasodilator mechanism could be one of the means by which the function of microvascular networks in such and other conditions [e.g., reactive hyperemia (Zl), collateral blood flow] could be optimized (41) The data indicate that the shear stress-sensitive mechanism could provide a considerable increase and redistribution of blood flow without a significant change in perfusion pressure. Blood flow increased greatly (by 230%) during parallel occlusion, but the i.ncrease was
AND
MICROVASCULAR
H867
RESISTANCE
considerably attenuated after impairment of the endothelium (Table 1). The smaller increase in flow in the endothelium-impaired preparation was due only to the increase in velocity, and the flow decrement (compared with preparations with intact endothelium) was most likely absorbed as an equivalent increase in flow in other parallel arterioles. A further interpretation of our findings is that Murray’s law (27, 29, 36, 41) is manifested in the present experimental conditions, although we fully recognize that the data presented herein are, in fact, only a point of departure for a more definitive course of investigation. Our findings that in the presence of intact endothelium a close to third-power relationship exists between calculated blood flow and arteriolar diameter in both control and during high-flow conditions agree with those of Mayrovitz and Roy (27) but not with those of others (5). This relationship is thought to minimize the work (power loss) required to maintain blood flow in a vascular network (25, 27, 29, 36) and also implies the importance of controlling wall shear stress (41). It then seems feasible that the dilation of consecutive segments of the vascular tree that elicits the decreases in wall shear stress and resistance can provide for substantial increases in blood flow without the expense of further power dissipation, as suggested by theoretical studies (41). After impairment of the endothelium, however, either in control or during “high-flow” conditions, wall shear stress was high and the diameter of the impaired segment of the arteriole deviated from “optimal” (Table 1). Interestingly, a similar decrease in the exponent was found to be related to the degree of disease in human coronary arteries (12). If endothelial dysfunction due to various pathological conditions (atherosclerosis, hypertension, etc.) were present in the whole network, then to maintain optimal flow (i.e., to avoid ischemia) an increase in work (pressure) would be required due to the increased power dissipation (36, 41). This could indeed be the case, since the myogenic response would not be opposed and vascular resistance would increase with a corresponding increase in pressure drop (i.e., an increase in systemic and a decrease in capillary blood pressure) (Fig. 7). In summary the present in vivo study indicates that wall shear stress is a regulated variable in skeletal muscle microcirculation and that an increase in wall shear stress is the underlying stimulus for flow-dependent arteriolar dilation and the consequent flow increase. Our data also provide an additional basis for the assumption that the flow-dependent ascending dilation observed in large vessels is also initiated by changes in wall shear stress. Together with theoretical studies emphasizing the relationship between shear stress and vessel radius, the present study offers in vivo evidence for the idea that wall shear stress is regulated to provide for an optimal structure of vascular networks that accords with their function. The authors acknowledge the superior Ecke. This work was supported by National PO-I-HL-43023 and by New York Heart Address reprint requests to A. Koller. Received
7 February
1990; accepted
secretarial Institutes Association
in final
form
skills
of Annette
of Health Grant Grant 89-062G. 29 October
1990.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
H868
ENDOTHELIAL
REGULATION
OF
SHEAR
REFERENCES 1. BORGSTROM, P., P. 0. GRANDE, AND S. MELLANDER. A mathematical description of the myogenic response in the microcirculation. Acta Physiol. Stand. 116: 363-376, 1982. 2. CHIEN, S. Determination of microvascular viscosity. In: Microcirculatory Technology, edited by C. H. Baker and W. L. Nastuk. New York: Academic, 1986, p. 179-196. 3. DAVIES, P. F. How do vascular endothelial cells respond to flow? News Physiol. Sci. 4: 22-25, 1989. 4. FRANGOS, J. A., S. G. ESKIN, L. V. MCINTIRE, AND C. L. IVES. Flow effects on prostacyclin production by cultured human endothelial cells. Science Wash. DC 227: 1477-1479, 1985. 5. GRIFFITH, T. M., D. H. EDWARDS, R. L. I. DAVIES, T. J. HARRISON, AND K. T. EVANS. EDRF coordinates the behavior of vascular resistance vessels. Nature Lond. 329: 442-445, 1987. 6. FLEISCH, A. Les reflexes nutritifs ascendants producteurs de dilatation arterielle. Arch. Int. Physiol. 41: 141-167, 1935. 7. FOLKOW, B., R. R. SONNENSCHEIN, AND D. L. WRIGHT. Loci of neurogenic and metabolic effects on precapillary vessels of skeletal muscle. Acta Physiol. Stand. 81: 459-471, 1971. 8. FRY, D. L. Acute vascular endothelial changes associated with increased blood velocity gradients. Circ. Res. 22: 165-197, 1968. 9. HADDY, F. J., AND J. C. SCOTT. Metabolically linked vasoactive chemicals in local regulation of blood flow. Physiol. Rev. 48: 688707, 1968. 10. HILTON, S. M. A peripheral arterial conducting mechanism underlying dilation of the femoral artery and concerned in functional vasodilatation in skeletal muscle. J. Physiol. Lond. 149: 93-111, 1959. 11. HINTZE, T. H., AND S. F. VATNER. Reactive dilation of larger coronary arteries in conscious dogs. Circ. Res. 54: 50-57, 1984. 12. HUTCHINS, G. M., M. M. MINER, AND J. K. BOITNOTT. Vessel caliber and branch-angle of human coronary artery branch-points. Circ. Res. 38: 572-576, 1976. 13. JOHNSON, P. C. The myogenic response. In: Handbook of Physiology. The Cardiovascular System. Vascular Smooth Muscle. Bethesda, MD: Am. Physiol. Sot., 1980, sect. 2, vol. II, chapt. 15, p. 409-442. 14. KAISER, L., AND H. V. SPARKS. Effect of hemodilution on endothelium-dependent vasodilation in the in vivo canine femoral artery. Circ. Shock. 23: 107-118, 1987. 15. KALEY, G., J. M. RODENBURG, E. J. MESSINA, AND M. S. WOLIN. Endothelium associated vasodilators in rat skeletal muscle microcirculation. Am. J. Physiol. 256 (Heart Circ. Physiol. 25): H720H725, 1989. 16. KOBARI, M., F. GOTOH, Y. FUKUUCHYI, K. TANAKA, N. SUZUKI, AND D. UEMATSU. Blood flow velocity in the pial arteries of cats, with particular references to the vessel diameter. J. Cereb. Blood Flow Metab. 4: 110-114, 1984. 17. KOLLER, A., AND G. KALEY. Blood flow velocity-induced dilation in skeletal muscle microcirculation in vivo: role of endothelium (Abstract). Physiologist 32(4): 178, 1989. 18. KOLLER, A., AND G. KALEY. Flow velocity dependent regulation of microvascular resistance in vivo. Microvasc. Endothel. Lympathics 5: 519-530,1989. 19. KOLLER, A., AND G. KALEY. Endothelium regulates skeletal muscle microcirculation by a blood flow velocity sensing mechanism. Am. J. Physiol. 258 (Heart Circ. Physiol. 27): H916-H920, 1990. 20. KOLLER, A., AND G. KALEY. Negative feedback control of wall shear stress by the endothelium of skeletal muscle arterioles (Abstract). FASEB J. 4: A1257, 1990.
STRESS
AND
MICROVASCULAR
RESISTANCE
21. KOLLER, A., AND G. KALEY. Role of endothelium in reactive dilation of skeletal muscle arterioles. Am. J. Physiol. 259 (Heart Circ. Physiol. 28): H1313-H1316, 1990. 22. KOLLER, A., AND G. KALEY. Prostaglandins mediate arteriolar dilation to increased blood flow velocity in skeletal muscle microcirculation. Circ. Res. 67: 529-534, 1990. 23. KOLLER, A., E. J. MESSINA, M. S. WOLIN, AND G. KALEY. Effects of endothelial impairment on arteriolar dilator responses in vivo. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): H1484-H1489, 1989. 24. KOLLER, A., E. J. MESSINA, M. S. WOLIN, AND G. KALEY. Endothelial impairment inhibits prostaglandin and EDRF-mediated arteriolar dilation in vivo. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): H1966-H1970,1989. 25. LIPOWSKY, H. H., S. KOVALCHECK, AND B. W. ZWEIFACH. The distribution of blood rheological parameters in the microvasculature of cat mesentery. Circ. Res. 43: 738-749, 1978. 26. LIPOWSKY, H. H., AND B. W. ZWEIFACH. Application of the “twoslit” photometric technique to measurement of microvascular volumetric flow rates. Microvasc. Res. 15: 93-101, 1978. 27. MAYROVITZ, H. N., AND J. ROY. Microvascular blood flow: evidence indicating a cubic dependence on arteriolar diameter. Am. J. Physiol. 245 (Heart Circ. Physiol. 14): H1031-H1038, 1983. 28. MELKUMYANTS, A. M., S. A. BALASHOV, AND V. M. KHAYUTIN. Endothelium dependent control of arterial diameter by blood viscosity. Cardiovasc. Res. 23: 741-747, 1989. 29. MURRAY, C. D. The physiological principle of minimum work. The vascular system and the cost of blood volume. Proc. NatZ. Acad. Sci. USA 12: 207-214,1926. 30. POHL, U., J. HOLTZ, R. BUSSE, AND E. BASSANGE. Crucial role of endothelium in the vasodilator response to increased flow in vivo. Hypertension Dallas 8: 37-44, 1986. 31. RODBARD, S. Vascular caliber. Cardiology 60: 4-49, 1975. 32. ROSENBLUM, W. I. Effects of blood pressure and blood viscosity on fluorescein transit time in the cerebral microcirculation in the mouse. Circ. Res. 27: 825-833, 1970. 33. RUBANYI, G. M., J. C. ROMERO, AND P. M. VANHOUTTE. Flow induced release of endothelium derived relaxing factor. Am. J. Physiol. 250 (Heart Circ. Physiol. 19): H1145-H1149, 1986. 34. SCHRETZENMAYR, A. Uber kreislaufregulatorishe Vorgange an den grossen Arterien bei der Muskelarbeit. Pfluegers Arch. 232: 743748,1933. 35. SEGAL, S. S., AND B. R. DULING. Flow control among microvessels coordinated by intercellular conduction. Science Wash. DC 234: 868-870,1986. 36. SHERMAN, T. F., A. S. POPEL, A. KOLLER, AND P. C. JOHNSON. The cost of departure from optimal radii in microvascular networks. J. Theor. BioZ. 136: 245-265, 1989. 37. SINOWAY, L. I., C. HENDRICKSON, W. R. DAVIDSON, S. PROPHET, AND R. ZELIS. Characteristics of flow mediated brachial artery vasodilation in human subjects. Circ. Res. 64: 32-42, 1989. 38. SMIESKO, V., J. KOZIK, AND S. DOLEZEL. The control of the arterial diameter by blood flow velocity is dependent upon intact endothelium (Abstract). Physiol. BohemsZov. 32: 558, 1983. 39. SMIESKO, V., D. J. LANG, AND P. C. JOHNSON. Dilator responses of rat mesenteric arterioles to increased blood flow velocity. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): H1958-H1965, 1989. 40. TANGELDER, G. J., D. W. SLAAF, AND R. S. RENEMAN. Wall shear rate in arterioles in vivo: least estimates from platelet velocity profiles. Am. J. Physiol. 254 (Heart Circ. Physiol. 23): H1059H1064,1988. 41. ZAMIR, M. Shear stress forces and blood vessel radii in the cardiovascular system. J. Gen. Physiol. 69: 449-461, 1977.
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (134.225.001.226) on August 12, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
