| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biophysics, University of Rochester, Rochester, New York 14642
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
ACKNOWLEDGEMENTS
FOOTNOTES
REFERENCES
ABSTRACT 
Frame, Mary D. S., and Ingrid H. Sarelius. Endothelial
cell dilatory pathways link flow and wall shear stress in an intact
arteriolar network. J. Appl. Physiol.
81(5): 2105-2114, 1996.
Our purpose was to determine whether the
endothelial cell-dependent dilatory pathways contribute to the
regulation of flow distribution in an intact arteriolar network. Cell
flow, wall shear stress (T
),
diameter, and bifurcation angle were determined for four sequential
branches of a transverse arteriole in the superfused cremaster muscle
of pentobaribtal sodium (Nembutal, 70 mg/kg)-anesthetized hamsters
(n = 51). Control cell flow was
significantly greater into upstream than into downstream branches
[1,561 ± 315 vs. 971 ± 200 (SE) cells/s,
n = 12]. Tissue exposure to 50 µM
N-nitro-L-arginine + 50 µM indomethacin (L-NNA + Indo) produced arteriolar constriction of 14 ± 4% and decreased
flow into the transverse arteriole. More of the available cell flow was
diverted to downstream branches, yet flow distribution remained
unequal. Control T was higher
upstream than downstream (31.3 ± 6.8 vs. 9.8 ± 1.5 dyn/cm2).
L-NNA + Indo decreased
T upstream and increased
T downstream to become equal in
all branches, in contrast to flow. To determine whether constriction in
general induced the same changes, 5%
O2 (8 ± 4% constriction) or
10
9 M norepinephrine (NE;
4 ± 3% constriction) was added to the tissue (n = 7). With
O2, flow was redistributed to
become equal into each branch. With NE, flow decreased progressively
more into the first three branches. The changes in flow distribution
were thus predictable and dependent on the agonist. With
O2 or NE, the spatial changes in
flow were mirrored by spatial changes in
T. Changes in diameter and in
cell flux were not related for
L-NNA + Indo (r = 0.45),
O2
(r = 0.07), or NE
(r = 0.36). For all agonists, when the
bifurcation angle increased, cell flow to the branch decreased
significantly, whereas if the angle decreased, flow was relatively
preserved; thus active changes in bifurcation angle may influence red
cell distribution at arteriolar bifurcations. Thus, when the
endothelial cell dilatory pathways were blocked, the changes in flow
and in T were uncoupled; yet when they were intact, flow
and T changed together.
flow-dependent responses; capillary recruitment; bifurcation angle; nitric oxide
INTRODUCTION CHANGES IN TISSUE BLOOD flow are matched to changes in
tissue metabolism during, for example, exercise (15). In striated muscle, increased flow occurs via the recruitment of complete capillary
networks (6, 15, 23, 32). In microvascular studies, the arterioles that
control flow to the capillary networks have been identified (2, 40).
One would predict that, for recruitment to occur, flow into these
controlling arterioles must be different during baseline conditions
and/or these controlling arterioles must have the capacity to
respond differently from each other. In fact, the baseline cell flux
(and wall shear stress) into these arterioles is systematically
different spatially, being higher in the branches controlling capillary
perfusion that arise upstream than in those downstream along the same
transverse arteriole (31). Furthermore, there are predictable spatial
and temporal differences in the diameter changes for these vessels in
response to metabolic or adrenergic stimuli, which suggest that changes in flow distribution are agonist dependent (6, 8). These studies
together imply that the control of flow into capillary networks is
organized spatially and temporally.
Endothelial cell-dependent dilatory pathways modulate flow in many
organs (3, 12-14, 17, 36, 38). These dilatory pathways involve
L-arginine/nitric oxide and
prostaglandin metabolism to various degrees in different tissues and
species (19-22). Interestingly, these pathways are involved in
exercise-induced dilation (14, 38), and it has been postulated that
they control arteriolar diameter to maintain an optimum flow
distribution within the microcirculation (12) and to regulate wall
shear stress (19). These studies thus suggest that the endothelial
cell-dependent dilatory pathways might be involved in the spatial
control of flow distribution (and of wall shear stress) at the
microcirculatory level.
The present study tests whether unequal cell flow (and wall shear
stress) distribution into the arterioles controlling capillary perfusion (31) is altered by blocking the endothelial cell-dependent dilatory pathways, thus investigating whether these pathways
participate in the regulation of flow distribution. This was
accomplished by measuring the red blood cell flux distribution
into sequential branch arterioles along a transverse
arteriole, before and after the endothelial cell-dependent dilatory
pathways are blocked. Inhibition of dilatory pathways results in
constriction. To examine the specificity of the changes in flow
distribution (i.e., to determine that they were not merely reflective
of a generalized constriction), we also tested whether vasoconstriction
due to elevated tissue O2 or
norepinephrine (NE) changed the flow distribution.
METHODS Preparation
Observation Site
Observations were made in branch arterioles arising in sequence along a single transverse arteriole, which was located in each preparation, as described previously (7, 8, 31, 40), and is schematized in Fig. 1. These branches were chosen for study because they control flow into the capillary networks in this muscle (40).
Experimental Protocol
During a 60-min stabilization period, the tetramethyl rhodamine isothiocyanate-labeled cells were administered (33, 35), and the presence of vasoactive tone (brisk dilation to locally applied 104 M adenosine) and
O2 sensitivity (constriction after
transient addition of 10% O2 to
the superfusate) was confirmed in three arterioles (2 randomly chosen
arterioles and 1 at the test site).
Sequential arteriolar branches were videotaped after stabilization
(control) and after exposure of the entire tissue to the experimental
agents. To avoid time-dependent changes, the length of the videotaping
periods was minimized, and the sequence of observations was alternated
between animals to start at the first or the last branch. In one group
of 12 animals the experimental condition was 20 min of exposure to 50 µM
N-nitro-L-arginine
(L-NNA) + 50 µM indomethacin
(Indo). In a second group of 7 animals the experimental conditions were
10 min of exposure to 5% O2
(equilibrated in the control suffusate as 5% CO2-5%
O2-90%
N2 gas), 10 min of recovery, and
then 10 min of exposure to control suffusate + 109 M NE (plus 100 mg/l
ascorbic acid). The cell flow, cell velocity, and diameter data with
O2, but not with NE, have been
reported earlier (31). In the present study we have normalized the
branch flow as a percentage of the feed inflow, used an improved
calculation of the wall shear stress, and added the bifurcation angle
data; none of these was reported previously (31).
To confirm that each branch arteriole had the same intrinsic ability to respond to Indo or L-NNA, local micropipette application of L-arginine, acetylcholine, Indo, or L-NNA to the external surface of a localized region of the branch arterioles was evaluated, as described previously (8, 9). The first, third, and last branch arterioles were tested sequentially in the same preparation, with the exposure sequence alternated to start upstream or downstream; responses were not related to exposure order, and each vessel was exposed only once. Exposure of arterioles to control superfusate from the pipette produced diameter changes of <5%, as in earlier studies (8, 9). The calculated concentration of the test drugs at the vessel wall is twofold less than the concentration in the pipette, as described previously (8, 9); thus the effective concentration of L-NNA or Indo at the vessel wall is estimated to be 50 µM with local micropipette exposure and superfusate tissue-wide exposure.
Measurements
Vessel diameter (in µm) and the angle at which the branch arose from the transverse arteriole (the bifurcation angle, in degrees) were measured off-line from the recorded image with a software program developed for this application (Dept. of Biophysics, University of Rochester, Rochester, NY) calibrated in the x and y directions with a videotaped stage micrometer. Diameter measurements were reproducible to within 0.6 µm, which is 1-2% of the diameter; angle measurements were reproducible to within <5°. For tissue-wide exposures, the change in diameter and in angle (the response) was calculated as the difference between the test condition and the control baseline condition. For local exposures with a micropipette, the diameter change was expressed as the percent change from baseline, where baseline was defined as the 60-s interval before exposure. The angle of bifurcation was measured within one diameter of the bifurcation junction, as defined previously (7). Fluorescent cell fluxes and velocities were measured at the boundaries of the bifurcation region for each segment by use of a video software program developed for this application (American Innovisions).Calculations
Cell flux (F, in cells/s) is given by the following equations: F = Nt/t and Nt = mt/p, where Nt is the number of cells crossing the specified vessel plane in time t, mt is the number of fluorescent cells crossing the plane, and p is the fluorescent cell fraction. Cell flux was used as an index of cell flow. Cell distribution was calculated as the fraction of the total cell flow into the transverse arteriole that was distributed to each branch and was expressed as a percentage. Individual cell velocities (in µm/s) were measured as the distance traveled in a specified number of video fields; fluorescent cells were taken as representative of the total population (32, 35). Individual velocities were measured for all cells crossing the specified sampling plane in a specified time interval, and the mean axial cell velocity (vc) was calculated from their harmonic mean (5). The shear rate (
, in
s1) was calculated as
follows: = 8 · vc/D,
where
vc
was used to approximate the average velocity in the vessel (11) and
D is vessel diameter.
Hematocrit, the volume fraction of cells in the vessel, was calculated
as follows: H = F · Vc /(vc · 
r2),
where Vc is the mean corpuscular
volume [58 × 1012
cm3 (34)] and
r is vessel radius (32). The apparent
viscosity (
app) was
calculated from the relationship among vessel hematocrit, vessel
diameter, and the relative viscosity (29). Shear stress at the vessel
wall (T, in
dyn/cm2) was calculated as
follows: T = app · .
Statistical Analyses
The calculated values were pooled by branch and test condition to determine the population means ± SE. Group comparisons were made between branches by analysis of variance and between experimental conditions by t-tests; correlations were evaluated by linear regression as appropriate (37). For all statistics, differences were considered significant if P
0.05.
RESULTS
Cell Flow Distribution During Constriction
Constriction by blocking endothelial cell dilatory pathways. During control, total cell flow was greater into the upstream branch arterioles than to those arising downstream (Fig. 1). The first branch received 1,561 ± 315 cells/s, which was significantly more than the second branch (363 ± 75 cells/s), third branch (572 ± 150 cells/s), or last branch (971 ± 200 cells/s, n = 12 animals). The total flow into the transverse arteriole was 4,464 ± 637 cells/s during control and decreased to 1,790 ± 395 cells/s during L-NNA + Indo (50 µM L-NNA + 50 µM Indo). Cell flow was significantly redistributed between branches (Fig. 2A), such that more of the available flow went to the last branch during L-NNA + Indo (31%) than during control (22%). In contrast, the first branch received ~30% of the available cell flow in control and L-NNA + Indo. Thus constriction by blocking the endothelial cell dilatory pathways preferentially increased flow into the downstream branches, yet the flow distribution into the branches remained unequal.
-nitro-L-arginine + 50 µM indomethacin (A,
L-NNA + Indo,
n = 12), 5% added
O2
(B), or
109 M norepinephrine
(C, NE,
n = 7). Control data are the same for O2 and for NE. Values are means ± SE, with significance set at P < 0.05. 
Less than 1st
branch; § different from control; (§)
P > 0.09.
Constriction by metabolic or adrenergic vasoconstrictors. Despite the finding that the baseline cell flux was higher in this separate group of animals, the trends in control cell flux distribution were the same for the two groups of animals. As above, during control, total cell flow into the branch arterioles was not equal. The first branch received more than the second, third, or last branch (2,207 ± 1,124, 1,312 ± 422, 962 ± 368, and 1,073 ± 360 cells/s, respectively, n = 7 animals). The total flow into the transverse arteriole was 7,158 ± 2,977 cells/s during control and decreased to 4,733 ± 2,144 with O2 (5%) or to 4,294 ± 1,598 cells/s with 10
9 M NE. Cell
flow was significantly redistributed between branches for each
condition (Fig. 2, B and
C). With
O2, flow decreased significantly
upstream (from 28% of inflow to 16%) and was preserved downstream
(from 14% to 13%). Thus flow into each branch became equal. With NE,
flow was preserved upstream (from 28% to 27%) and decreased
progressively more for the first three branches and was unchanged
downstream (Fig. 2C). The net result
was a predictable difference in cell flux for these branches. Thus the
flow distribution changes were different in response to
O2 and NE, and neither mimicked the changes due to L-NNA + Indo.
Diameter and flux changes.
The initial control diameter of the branch arterioles was 11.2 ± 0.75 µm for the group that received
L-NNA + Indo
(n = 12) and was larger (15.6 ± 0.8 µm, n = 7) for the group that
received O2 and NE. There was an
overall constriction of the branch arterioles with all conditions; the
constriction as a percentage of the initial diameter was 
14.2 ± 4% (2.2 ± 0.72 µm) for
L-NNA + Indo, 8 ± 4%
(1.33 ± 0.6 µm) for
O2, and 4 ± 3%
(1.0 ± 0.6 µm) for NE. The changes in cell flow (in
cells/s) or in cell distribution (in percentage of inflow) were
unrelated to the magnitude of the diameter changes by linear regression
(L-NNA:
r = 0.45 cell flux, r = 0.15 cell distribution;
O2:
r = 0.07 cell flux,
r = 0.17 cell distribution; NE:
r = 0.36 cell flux,
r = 0.0001 cell distribution). It is
important to note, as we reported previously (8), that for each
condition, some of the branch arterioles dilated, even though the
average response was to constrict. With
L-NNA + Indo, 29 of 36 branches
constricted and 7 dilated. With
O2, 20 of 26 branches constricted
and 6 dilated. With NE, 16 of 26 branches constricted and 10 dilated.
Because the changes in cell flow were different by branch position, we
examined the flow changes for the branches that constricted vs. those
that dilated for the branches upstream and for those downstream (Fig.
3, A and
C). The corresponding diameter
changes for each condition upstream and downstream are shown in Fig. 3, B and
D. For all conditions, cell flux (in
cells/s) decreased significantly upstream whether the arterioles
constricted or dilated but decreased significantly downstream only for
the arterioles that constricted. With
O2, for the upstream branches that
dilated, the cell flux decreased significantly more than did the cell
flux for the upstream branches that constricted. These data clearly show that the diameter changes did not alone account for the changes in
flow and could not explain the changes in flow distribution.
9 M NE
(n = 7). Changes for branches that
constricted and for those that dilated
(n is indicated above bars) are
compared at upstream (A and
B) and downstream
(C and
D) locations along transverse arteriole. Values are means ± SE, with significance set at
P < 0.05. Significant change from
control (i.e., different from zero); § dilators decreased
flow significantly more than constrictors.
Wall Shear Stress During Constriction
The control wall shear stress was significantly higher in the upstream than in the downstream branches in the group of animals that was exposed to L-NNA + Indo (Table 1; n = 12). This was due to a higher shear rate upstream than downstream but could not be attributed to position-dependent differences in the viscosity, which was the same in the first branch and the last branch (Table 1). During L-NNA + Indo, the wall shear stress significantly decreased upstream (1st branch) and significantly increased downstream (Fig. 4). The changes in wall shear stress reflected changes in shear rate and in viscosity. Because of the high physiological variability in the wall shear stress upstream, these changes were not seen when the overall mean values of the wall shear stress for control and L-NNA + Indo are compared (Table 1). However, the mean value of the wall shear stress was significantly increased downstream, where the variability was much less. The net result was that the wall shear stress was not different in a spatial sense for the branches along this transverse arteriole when the endothelial cell dilatory pathways were blocked.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
9 M NE
(n = 7) for each sequential branch
arteriole (1st, 2nd, 3rd, last) along transverse arteriole. Values are
means ± SE, with significance set at
P < 0.05. Significant change from
control (i.e., different from zero).
The wall shear stress in the branch arterioles upstream was not significantly different from that in the branch arterioles downstream for the group of animals that was exposed to O2 and NE, although there was a progressive decline in wall shear stress for the first three branches (Table 1; n = 7). The control wall shear stress and shear rates were significantly higher at the last branch downstream in this group than in the group that received L-NNA + Indo, whereas upstream the wall shear stress was lower, although not significantly (see DISCUSSION). With O2, the wall shear stress decreased overall (Fig. 4); the mean values were significantly decreased at the first and third branches (Table 1). The net result was that the wall shear stress was the same upstream and downstream. With NE, the wall shear stress was significantly decreased at only the third branch (Fig. 4, Table 1). The net result was that the wall shear stress was progressively lower for the first three branches upstream and was unchanged downstream. Each of these changes was reflected by corresponding changes in shear rate, but not in viscosity, which remained unchanged (Table 1). Thus the wall shear stress was different by position and by the agonist used to constrict the arterioles. Importantly, constriction when the endothelial cell dilatory pathways were intact resulted in a marked difference in the wall shear stress spatially that was attributable solely to changes in the shear rate.
Bifurcation Angle and Flux Distribution
There is not a lot of information available about the role of bifurcation angles in modulating flow distribution in vessels of this size in vivo. There is evidence to suggest that, for larger angles or for increases in angle, fewer cells travel to the branch (26, 28). Because the changes in diameter could not explain the changes in flow distribution in the present study, we examined whether the changes in bifurcation angle were correlated with the changes in cell flux distribution. Bifurcation angle and angle change. During control, the bifurcation angle was larger upstream [110 ± 8° (n = 12) and 93 ± 8° (n = 7) for L-NNA + Indo and O2 and NE groups, respectively] and progressively smaller for the downstream bifurcations [57 ± 5° (n = 12) and 45 ± 12° (n = 7), respectively], as shown previously (7, 10). With any change in the vasoactive state, the angles at individual bifucations increased or decreased by as much as 50°, again confirming our earlier finding (7, 10). Furthermore, here we report that, compared with control, the mean angle did not change at any position with O2 or NE, as reported for adenosine (7). However, with L-NNA + Indo, the mean angle of bifurcation became significantly larger at the downstream bifurcations (from 66 ± 8 to 87 ± 8° and from 57 ± 5 to 69 ± 6° for 3rd and last branches, respectively, n = 12) and did not change upstream. Overall, the magnitude of the angle change and the magnitude of the flow change were not correlated (L-NNA + Indo: r = 0.20 cell flux, r = 0.06 flow distribution; O2: r = 0.15 cell flux, r = 0.16 flow distribution; NE: r = 0.16 cell flux, r = 0.20 flow distribution). Angle change and flux change. Change in flow was related to the direction of the change in angle. With L-NNA + Indo, 26 of 36 bifurcations increased while 10 decreased in angle; with O2, 14 of 26 increased and 12 decreased in angle; with NE, 14 of 26 increased and 12 decreased in angle. The 14 that increased in angle (or 12 that decreased) were not the same with O2 and NE. Many of the changes we have reported illustrate differences in behavior upstream vs. downstream along the transverse arteriole. Therefore, we examined how changes in angle and in flux were related upstream and downstream. In general, for all agonists and branches, if the bifurcation angle increased, the cell flux into the branch decreased significantly, whereas if the angle decreased, the cell flux into the branch was relatively preserved (Figs. 5 and 6). Upstream, for bifurcations that increased in angle, the cell flux (control to test) decreased significantly from 1,935 ± 499 to 455 ± 160 cells/s (L-NNA + Indo) and from 1,317 ± 465 to 596 ± 342 cells/s (O2). Downstream, with angle increases, the cell flux also decreased significantly from 1,077 ± 227 to 558 ± 103 cells/s (L-NNA + Indo) and from 999 ± 333 to 343 ± 134 cells/s (O2). For bifurcations upstream where the angle decreased, the cell flux was preserved from 1,037 ± 132 to 593 ± 193 cells/s (L-NNA + Indo) and from 2,091 ± 989 to 1,026 ± 618 cells/s (O2). Likewise, for angle decreases, downstream, the cell flux was not significantly changed from 439 ± 24 to 419 ± 69 cells/s (L-NNA + Indo) and from 1,054 ± 386 to 363 ± 117 cells/s (O2). The single exception was upstream with NE (Fig. 6), where the cell flux into the branch was relatively unchanged if the angle increased (from 1,281 ± 479 to 717 ± 249 cells/s) or decreased (from 2,119 ± 983 to 1,050 ± 550 cells/s). This may merely reflect the relatively small changes in cell flux compared with the variability upstream, because downstream with NE, if the angle increased, then the cell flux into the branch decreased significantly from 1,084 ± 317 to 583 ± 172 cells/s, and if the angle decreased, then the cell flux was relatively unchanged (from 885 ± 435 to 441 ± 150 cells/s). Thus the changes in angle and cell flux were related both upstream and downstream, in contrast to the spatially organized wall shear stress and flux distribution behaviors that we observed.
Significant change from
control.
9 M NE
(n = 7) for upstream vs. downstream
branches along transverse arteriole; data are separated by response of
angle to increase or decrease between control and experimental
condition (n is indicated above bars).
Values are means ± SE, with significance set at
P < 0.05. Significant change from
control (i.e., different from zero).
Local Application of Vasoconstrictors
To rule out the possibility that the responses obtained from the tissue-wide exposure had occurred because the responsivity (i.e., ability to respond to an agonist) of the upstream branches was different from the responsivity of the downstream branches, we determined the local responsivity of each branch to L-NNA and to Indo. [In a previous study we determined that the local responsivity of each branch to NE was the same (8).] The baseline arteriolar diameter was 12.6 ± 0.76 µm (n = 50 vessels in 19 animals). Local micropipette application of Indo, or (separately) L-NNA, produced a similar degree of peak constriction in upstream (Indo:23 ± 5% of
baseline, n = 4 animals;
L-NNA: 27 ± 4%,
n = 5 animals) and downstream (Indo:
27 ± 5%;
L-NNA: 28 ± 5%)
branch arterioles. In other experiments, constriction to
L-NNA (to 22 ± 5% of
baseline) was reversed with simultaneous pipette exposure to
L-arginine (to +13 ± 4% of
baseline, n = 8 vessels in 4 animals).
Dilation to acetylcholine (to +82 ± 16% of baseline) was
partially inhibited by simultaneous pipette exposure to
L-NNA (to +42 ± 16% of
baseline, n = 6 vessels). Separately,
dilation to acetylcholine (to +141 ± 40% of baseline) was
partially inhibited by Indo (to +80 ± 33% of baseline,
n = 6 vessels). Thus the
L-arginine and prostaglandin pathways are present in these arterioles independent of their location
upsteam or downstream.
DISCUSSION
The most important finding of this study is evidence that flow heterogeneity, classically described at the organ level, is in fact not random within the peripheral circulation but is instead modulated spatially within an area as small as that fed by a single transverse arteriole. Constriction by different agonists elicits specific changes in cell flow distribution along this transverse arteriole. We interpret the agonist-specific vascular changes as indicative of separate mechanisms of the vasculature to respond to quite specific changes in the state of the tissue. The implication is that this transverse arteriole and its branches may constitute an arteriolar functional unit for the control of the delivery of blood within a defined microvascular region.
Tissue-Wide Stimuli Evoke Integrative Responses
We have shown that the L-arginine and the prostaglandin pathways influence the baseline arteriolar diameter in these vessels, as found for other tissues and species at this level of the microcirculation (3, 19, 21, 36, 39). Furthermore, as we previously showed for NE (8), there is no difference in the response of the arterioles at different locations (upstream vs. downstream) to locally micropipette-applied L-NNA or Indo. Constriction to L-NNA was specific for an L-arginine pathway, because it was reversed by simultaneous addition of substrate (L-arginine). We also confirm that these pathways are present in the endothelial cell in this tissue, because dilation to acetylcholine was inhibited by L-NNA or Indo. To determine whether these pathways were involved in flow distribution, we exposed the entire tissue to both L-NNA and Indo, because the biochemical pathways involving L-arginine and prostaglandins are linked (4, 24), and we wanted to block all endothelial cell-dependent dilatory capability by these pathways. Clearly, L-NNA and Indo applied across the entire tissue did not uniformly constrict these arterioles (maximum range +37 to65%). Instead, a spatially organized vascular
response resulted.
To test whether the responses were specific to the vasoconstriction that resulted from blocking the endothelial cell-dependent dilatory pathways, we used data collected during an earlier study (31) in which the vasculature was constricted by exposure to low concentrations of NE or O2. Comparing the baseline values for the two groups of animals used (the group that was exposed to L-NNA + Indo vs. the group that was exposed to NE and O2), we find differences in the baseline diameters, cell velocities, and cell fluxes. These differences presumably reflect variations, for example, in the strain of animals and, hence, in their developmental states at the time of use (34). Nevertheless, the baseline flow distributions for these two groups of animals were similar in transverse arterioles that were identified using identical criteria. Microvascular and architectural properties are similar in the two groups of animals (2, 8, 31, 40). We report spatially predictable vascular responses that are different for each of these agonists (L-NNA + Indo, O2, and NE). Thus we conclude that blocking the endothelial cell-dependent dilatory pathways has uncovered the contribution of these pathways to the spatial organization of vascular tone, wall shear stress, and flow distribution.
Control of Flow Distribution
It is clear that endothelial cell-dependent dilatory pathways significantly affect flow within the peripheral circulation (12-14, 17, 27, 36). The effect of these pathways on flow distribution has been illustrated by studies which show that intravascular inhibition of the L-arginine and/or prostaglandin pathway significantly decreases flow through specific tissues, such as skeletal muscle (13, 14, 17, 36). In this study, using observations made at a single microvascular generation within the muscle, we are able to describe how flow distribution is altered within the peripheral circulation. We observed a defined transverse arteriole that supplies blood to the inflow vessels of separate and adjacent capillary networks (2, 40). When the dilatory pathways are intact, the cell flow to the capillary networks arising from the upstream branches is preferentially greater than that into networks arising from the downstream branches along the same transverse arteriole. This upstream vs. downstream difference in vascular capacity supports our previous findings, indicating that mechanisms are in place that regulate the upstream region differently from the downstream region of this transverse arteriole (8, 9, 31). We now conclude that this upstream-to-downstream difference is partially under the control of the endothelial cell-dependent dilatory pathways. In comparison, vasoconstriction by O2 or by NE each elicited distinct patterns of spatial reorganization of flow along this transverse arteriole (Fig. 2). The net result with O2 is a pattern of flow distribution with overall equal flow into each branch (13-16%), whereas with NE more flow was diverted into the first branch and progressively less into the second and third branches. Clearly, the mechanism by which vasoconstriction is induced is important in a determination of how flow is redistributed into the branches and thence into capillary networks.Control of Wall Shear Stress
Endothelial cell-dependent dilatory pathways are involved in flow (shear)-dependent dilatory responses (19-21, 25). Others have shown that these pathways participate in the regulation of wall shear stress in isolated vessel systems (19) and that they optimize vascular resistance to optimize flow distribution over several vascular generations (12). We show here that, within one vascular generation, inhibition of these endothelial cell-dependent dilatory pathways has made the wall shear stress more uniform in the branches along the transverse arteriole. Likewise, constriction with O2 made the wall shear stress more uniform. In contrast, constriction by addition of NE resulted in a marked difference in wall shear stress for the first vs. the third branches. With inhibition of the endothelial cell-dependent dilatory pathways, changes in the wall shear stress were unrelated to the changes in flow distribution (cf. Figs. 2 and 4). For example, despite a significant decrease in wall shear stress in the first branch, the fractional flow to the first branch was maintained; at the third branch, wall shear stress increased significantly while flow distribution did not change; downstream, wall shear stress increased, and the last branch received more of the available flow. In contrast, with O2 or with NE, changes in flow distribution matched the changes in wall shear stress upstream and downstream (Figs. 2 and 4). For example, with O2, the wall shear stress decreased significantly at the first branch, as did the relative flow, and with NE, the wall shear stress decreased significantly at the third branch, as did the flow. The implication is that intact dilatory pathways are necessary to coordinate flow and fluid shearing forces upstream and downstream along this transverse arteriole. One way these responses could be spatially coordinated in an intact arteriolar network is via a vascular communication system, linking the upstream and downstream regions. We have demonstrated that L-arginine initiates a conducted vasodilation along this transverse arteriole (9); this suggests that the response of the vasculature is organized spatially by an L-arginine pathway. Furthermore, we have shown that the L-arginine-induced conducted signal also alters the responsivity of the upstream region but not the downstream region (9). How this L-arginine-linked conducted response is related to the prevailing spatial differences in flow or wall shear stress remains a puzzle to be solved.The observed range in wall shear stress was quite large (2-72, 0-81, 1-25, and 1-36 dyn/cm2 during control, L-NNA + Indo, O2, and NE, respectively). This variability far exceeds the measurement error (±5%), which leads us to conclude that the variability in wall shear stress is physiological in these arterioles. The measurement error is the same at each branch, but the variability (range) was greater in the upstream branches (2-72 dyn/cm2) and significantly less downstream (5-23 dyn/cm2) for the L-NNA + Indo group.
Why Was Flow Change Unrelated to Diameter Change?
In the microcirculation, vessel diameter is expected to be a major determinant of red cell partitioning at individual microvascular bifurcations (11). We found that changes in diameter and in cell flow were not correlated for any experimental condition. Furthermore, for all conditions, cell flow decreased significantly upstream whether the arterioles constricted or dilated but decreased significantly downstream only for the arterioles that constricted. From consideration of the hemodynamic or myogenic regulatory mechanisms, we would expect flow to be relatively preserved in a vessel that dilated when the inflow is decreased. Therefore, the real question may be, Why was flow not decreased less for the arterioles that dilated upstream? However, we are observing an intact system, and the changes across the tissue will of course impact on the local flow distribution; this may be why flow and diameter changes are not directly linked at the observation site. This study indicates that, in an intact system, diameter changes alone cannot be used to indicate changes in the distribution of flow. We illustrate this from a previous study, in which we measured only diameter changes during tissue-wide O2 (8). We concluded that because the arterioles were more likely to constrict downstream, flow had been relatively preserved upstream. The present measurements of cell flow show that conclusion to be false; with O2, flow is relatively preserved downstream where the branch constriction is greater. Thus the present study suggests that the spatial position within a defined connected group may be a better indicator than just diameter change of how flow redistributes with a specific stimulus.Do Bifurcation Angles Affect Cell Distribution?
Angle change. Previously we showed that active changes in angle are not dependent on branch location after dilation with adenosine (7). In the present study, we show that angle change is indeed dependent on the branch location after constriction with L-NNA and Indo, but not with O2 or NE. Our data indicate that constriction (over this narrow range of diameters) was not related to angle change with L-NNA + Indo (r = 0.21), O2 (r = 0.19), or NE (r = 0.04), and not all branches that constricted were from bifurcations that increased in angle (P > 0.05, Mann-Whitney U-test). Thus the bifurcation angle and segment diameters, which describe the internal geometric shape of the bifurcation in two dimensions, can change independently of each other. It is possible that the way the angle and diameter change together may be related in three dimensions, but our two-dimensional data indicate that this is not in a direct and simple manner. Therefore, we speculate that the link between vessel tone and bifurcation angle may depend on the vasoactive stimulus and may be related to specific three-dimensional shape changes within the bifurcation that are agonist specific. Angles and flow. Endothelial cell-dependent dilatory pathways optimize flow distribution over many vascular generations within the peripheral circulation by optimizing the diameters and the bifurcation angles (12). In the systemic circulation, it is established that the bifurcation geometry (diameter and angles together) strongly influences the cell and total flow partitioning (16, 30). In contrast, few studies have reported on angle and flow in the microcirculation, where the flow rates are much lower. Such studies are predominantly in vitro, and they suggest in a general sense that bifurcation angle may influence cell (particle) distribution in vessels of this size (1, 26, 28). Interestingly, we show that angle increase was associated with a significant decrease in cell flow (except for NE upstream), whereas angle decrease was always associated with a relatively preserved cell flow. We speculate that angle changes at this level of the vasculature may be one mechanism by which cell flow (and thus O2 supply capability) into capillary networks is actively modulated. In summary, this study shows that constriction significantly changed the spatial organization of cell distribution and wall shear stress within a defined microvascular region. The spatial pattern of flow changes was significantly dependent on the mechanism by which the vasculature was constricted. The changes in cell distribution were mirrored by the changes in wall shear stress only when the endothelial cell-dependent dilatory pathways were intact. Neither the vessel diameter nor diameter changes were directly related to the changes in flow distribution. Instead, flow redistribution was associated with the change in arteriolar bifurcation angle at each position.ACKNOWLEDGEMENTS
We thank P. A. Titus for excellent technical assistance.
FOOTNOTES
Address for reprint requests: M. D. S. Frame, Dept. of Anesthesiology, University of Rochester Medical Center, 601 Elmwood Ave., Rochester, NY 14642.
Received 19 June 1996; accepted in final form 20 June 1996.
REFERENCES
| 1. | Audet, D. M., and W. L. Olbricht. The motion of model cells at capillary bifurcations. Microvasc. Res. 33: 377-396, 1987. |
| 2. | Berg, B. R., and I. H. Sarelius. Functional capillary organization in striated muscle. Am. J. Physiol. 268 (Heart Circ. Physiol. 37): H1215-H1222, 1995. |
| 3. | Boczkowski, J., E. Vicaut, G. Danialou, and M. Aubier. Role of nitric oxide and prostaglandins in the regulation of diaphragmatic arteriolar tone in the rat. J. Appl. Physiol. 77: 590-596, 1994. |
| 4. | Braun, M., and K. Schror. Prostaglandin D2 relaxes bovine coronary arteries by endothelium-dependent nitric oxide-mediated cGMP formation. Circ. Res. 71: 1305-1313, 1992. |
| 5. | Cokelet, G. R. Experimental determination of the average hematocrit of blood flowing in a vessel. Microvasc. Res. 7: 382-384, 1974. |
| 6. | Delashaw, J. B., and B. R. Duling. A study of the functional elements regulating capillary perfusion in striated muscle. Microvasc. Res. 36: 162-171, 1988. |
| 7. | Frame, M. D. S., and I. H. Sarelius. Arteriolar bifurcation angles vary with position and when flow is changed. Microvasc. Res. 46: 190-205, 1993. |
| 8. | Frame, M. D. S., and I. H. Sarelius. Regulation of capillary perfusion by small arterioles is spatially organized. Circ. Res. 73: 155-163, 1993. |
| 9. | Frame, M. D. S., and I. H. Sarelius. L-Arginine induced conducted signals alter upstream arteriolar responsivity to L-arginine. Circ. Res. 77: 695-701, 1995. |
| 10. | Frame, M. D. S., and I. H. Sarelius. Energy optimization and bifurcation angles in the microcirculation. Microvasc. Res. 50: 301-310, 1995. |
| 11. | Goldsmith, H. L., G. R. Cokelet, and P. Gaehtgens. Robin Fahraeus: evolution of his concepts in cardiovascular physiology. Am. J. Physiol. 257 (Heart Circ. Physiol. 26): H1005-H1015, 1989. |
| 12. | Griffith, T. M., and D. H. Edwards. Basal EDRF activity helps to keep the geometrical configuration of arterial bifurcations close to the Murray optimum. J. Theor. Biol. 146: 545-573, 1990. |
| 13. | Habazettl, H., B. Vollmar, M. Christ, H. Baier, P. F. Conzen, and K. Peter. Heterogeneous microvascular coronary vasodilation by adenosine and nitroglycerin in dogs. J. Appl. Physiol. 76: 1951-1960, 1994. |
| 14. | Hirai, T., M. D. Visneski, K. J. Kearns, R. Zelis, and T. I. Musch. Effects of NO synthase inhibition on the muscular blood flow response to treadmill exercise in rats. J. Appl. Physiol. 77: 1288-1293, 1994. |
| 15. | Hudlicka, O. Muscle Blood Flow: Its Relation to Muscle Metabolism and Function. Amsterdam: Swets & Zeitlinger, 1973. |
| 16. | Karino, T., and H. L. Goldsmith. Role of blood cell-wall interactions in thrombogenesis and atherogenesis: a microrheological study. Biorheology 21: 587-601, 1984. |
| 17. | King, C. E., M. J. Melinyshyn, J. D. Mewburn, S. E. Curtis, M. J. Winn, S. M. Cain, and C. K. Chapler. Canine hindlimb blood flow and O2 uptake after inhibition of EDRF/NO synthesis. J. Appl. Physiol. 76: 1166-1171, 1994. |
| 18. | Klitzman, B., and B. R. Duling. Microvascular hematocrit and red cell flow in resting and contracting striated muscle. Am. J. Physiol. 237 (Heart Circ. Physiol. 6): H481-H490, 1979. |
| 19. | Koller, A., S. Sun, and G. Kaley. Role of shear stress and endothelial prostaglandins in flow-and viscosity-induced dilation of arterioles in vitro. Circ. Res. 72: 1276-1284, 1993. |
| 20. | Koller, A., S. Sun, E. J. Messina, and G. Kaley. L-arginine analogues blunt prostaglandin-related dilation of arterioles. Am. J. Physiol. 264 (Heart Circ. Physiol. 33): H1194-H1199, 1993. |
| 21. | Kuo, L., W. M. Chilian, and M. J. Davis. Interaction of pressure- and flow-induced responses in porcine coronary resistance vessels. Am. J. Physiol. 261 (Heart Circ. Physiol. 30): H1706-H1715, 1991. |
| 22. | Lee, L., and R. C. Webb. Endothelium-dependent relaxation and L-arginine metabolism in genetic hypertension. Hypertension Dallas 19: 435-441, 1992. |
| 23. | Lund, N., D. H. Damon, D. N. Damon, and B. R. Duling. Capillary grouping in hamster tibialis anterior muscles: flow patterns and physiological significance. Int. J. Microcirc. Clin. Exp. 5: 359-372, 1987. |
| 24. | Marotta, P., L. Sautebin, and M. DiRosa. Modulation of the induction of nitric oxide synthase by eicosanoids in the murine macrophage cell line J774. Br. J. Pharmacol. 107: 640-641, 1992. |
| 25. | Marshall, J. M. The influence of the sympathetic nervous system on individual vessels of the microcirculation of skeletal muscle of the rat. J. Physiol. Lond. 332: 169-186, 1982. |
| 26. | Mchedlishvili, G., and M. Varazashvili. Flow conditions of red cells and plasma in microvascular bifurcations. Biorheology 19: 613-620, 1982. |
| 27. | Nakamura, T., and R. L Prewitt. Effect of NG-monomethyl-L-arginine on arcade arterioles of rat spinotrapezius muscles. Am. J. Physiol. 261: H46-H52, 1991. |
| 28. | Pries, A. R., K. H. Albrecht, and P. Gaehtgens. Model studies on phase separation at a capillary orifice. Biorheology 18: 355-367, 1981. |
| 29. | Pries, A. R., T. W. Secomb, P. Gaehtgens, and J. F. Gross. Blood flow in microvascular networks: experiments and simulation. Circ. Res. 67: 826-834, 1990. |
| 30. | Roach, M. R., S. Scott, and G. G. Ferguson. The hemodynamic importance of the geometry of bifurcations in the circle of Willis (Glass model studies). Stroke 3: 255-267, 1972. |
| 31. | Sarelius, I. H. Cell and oxygen flow in arterioles controlling capillary perfusion. Am. J. Physiol. 265 (Heart Circ. Physiol. 34): H1682-H1687, 1993. |
| 32. | Sarelius, I. H. Cell flow path influences transit time through striated muscle capillaries. Am. J. Physiol. 250 (Heart Circ. Physiol. 19): H899-H907, 1986. |
| 33. | Sarelius, I. H. Microcirculation in striated muscle after acute reduction in systemic hematocrit. Respir. Physiol. 78: 7-17, 1989. |
| 34. | Sarelius, I. H., D. N. Damon, and B. R. Duling. Microvascular adaptations during maturation of striated muscle. Am. J. Physiol. 241 (Heart Circ. Physiol. 10): H317-H324, 1981. |
| 35. | Sarelius, I. H., and B. R. Duling. Direct measurement of microvessel hematocrit, red cell flux, velocity, and transit time. Am. J. Physiol. 243 (Heart Circ. Physiol. 12): H1018-H1026, 1982. |
| 36. | Shen, W., X. Xu, M. Ochoa, G. Zhao, M. S. Wolin, and T. H. Hintze. Role of nitric oxide in the regulation of oxygen consumption in conscious dogs. Circ. Res. 75: 1086-1095, 1994. |
| 37. | Snedecor, G. W., and W. G. Cochran. Statistical Methods (6th ed.). Ames: Iowa State University Press, 1967. |
| 38. | Sun, D., A. Huang, A. Koller, and G. Kaley. Short-term daily exercise activity enhances endothelial NO synthesis in skeletal muscle arterioles of rats. J. Appl. Physiol. 76: 2241-2247, 1994. |
| 39. | Sun, D., E. J. Messina, A. Koller, M. S. Wolin, and G. Kaley. Endothelium-dependent dilation to L-arginine in isolated rat skeletal muscle arterioles. Am. J. Physiol. 262: H1211-H1216, 1992. |
| 40. | Sweeney, T. E., and I. H. Sarelius. Arteriolar control of capillary cell flow in striated muscle. Circ. Res. 64: 112-120, 1989. |
| 41. | Weideman, M. P. Blood flow through terminal arterial vessels after denervation of the bat wing. Circ. Res. 22: 83-89, 1968. |
This article has been cited by other articles:
|
|
 |
|
  M. D. Frame, R. J. Rivers, O. Altland, and S. Cameron Mechanisms initiating integrin-stimulated flow recruitment in arteriolar networks J Appl Physiol, June 1, 2007; 102(6): 2279 - 2287. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Anim-Nyame, S. R Sooranna, M. R Johnson, J. Gamble, and P. J Steer A longitudinal study of resting peripheral blood flow in normal pregnancy and pregnancies complicated by chronic hypertension and pre-eclampsia Cardiovasc Res, June 1, 2001; 50(3): 603 - 609. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Bao, C. B. Clark, and J. A. Frangos Temporal gradient in shear-induced signaling pathway: involvement of MAP kinase, c-fos, and connexin43 Am J Physiol Heart Circ Physiol, May 1, 2000; 278(5): H1598 - H1605. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. Frame Increased flow precedes remote arteriolar dilations for some microapplied agonists Am J Physiol Heart Circ Physiol, April 1, 2000; 278(4): H1186 - H1195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. S. Frame Conducted signals within arteriolar networks initiated by bioactive amino acids Am J Physiol Heart Circ Physiol, March 1, 1999; 276(3): H1012 - H1021. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Thorin-Trescases, J. A. Bevan, and W. J. Pearce High Levels of Myogenic Tone Antagonize the Dilator Response to Flow of Small Rabbit Cerebral Arteries • Editorial Comment Stroke, June 1, 1998; 29(6): 1194 - 1201. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. B. Cowan, S. J. Lye, and B. L. Langille Regulation of Vascular Connexin43 Gene Expression by Mechanical Loads Circ. Res., April 20, 1998; 82(7): 786 - 793. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Fox and M. D. Frame Regulation of flow and wall shear stress in arteriolar networks of the hamster cheek pouch J Appl Physiol, May 1, 2002; 92(5): 2080 - 2088. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. Frame, R. J. Fox, D. Kim, A. Mohan, B. C. Berk, and C. Yan Diminished arteriolar responses in nitrate tolerance involve ROS and angiotensin II Am J Physiol Heart Circ Physiol, June 1, 2002; 282(6): H2377 - H2385. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
