169
Laser-Induced Endothelial Damage Inhibits
Endothelium-Dependent Relaxation in the
Cerebral Microcirculation of the Mouse
William I. Rosenblum, Guy H. Nelson, and John T. Povlishock
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
This study demonstrates endothelium-dependent relaxation in the surface arterioles of the brain. A
helium-neon laser was used to injure endothelium in situ following i.v. injection of Evans blue dye,
which sensitizes the bed to the laser. Areas 18 or 36 /xm in diameter were injured and no longer relaxed
to either 1 ml of acetylcholine chloride or bradykinin triacetate, 80 jug/ml delivered for 60 seconds.
Dilations to sodium nitroprusside (30 /u.g/ml) were unaffected. Normal responses to nitroprusside, plus
electron microscopy, established that vascular smooth muscle was uninjured. Endothelium-dependent relaxation was impaired when only minor ultrastructural damage was present. Dilation was
inhibited downstream and upstream as far as 80 /xm from the center of the laser beam. This suggests a
spread of endothelium injury around the site of laser impact. However, inhibition was somewhat more
marked downstream than upstream, implying that a portion of the downstream response was dependent on a substance released from an upstream site. To date, very few studies have reported endothelium-dependent relaxation in vivo, especially in the microcirculation. The present study accomplishes
this. Moreover, in contrast to in vitro observations of endothelium-dependent relaxation in large
vessels, the in vivo elimination of endothelium-dependent relaxation in the microcirculation required
neither removal of endothelium nor injury to large numbers of endothelium cells. Since endotheliumdependent relaxation in the microcirculation has now been demonstrated using three different techniques to injure endothelium, it is reasonable to conclude that the phenomenon is real. (Circulation
Research 1987;60:169-176)
T
he purpose of this report is to describe loss of
endothelium-dependent relaxing factor(s) following focal endothelial injury produced in
arterioles on the surface of the brain. Endotheliumdependent relaxing factors (EDRF) are produced in
various blood vessels in several species, but may differ
from one vascular bed to another, even within a species.'"3 EDRFs are the agents actually responsible for
vasodilation produced by several agonists, but may
differ from agonist to agonist.4 The classic means of
establishing that an EDRF mediates dilation by an
agonist is to abolish dilation to that agonist through
removal of large amounts of endothelium, while showing that relaxation to other agents remains intact. u
This has almost always been demonstrated in vitro
using large arteries. Acetylcholine is the classic example of an EDRF-mediated relaxant. The chemical identity of EDRFs is undetermined.1'4-*
It was recently demonstrated that microvessels also
produce EDRF. This was shown in mesenteric7 and
brain microvessels of mice 89 and in brain microvessels
of the cat7 by simply damaging, but not removing, the
endothelium in situ. Damage was produced by exposFrom the Department of Pathlogy (Neuropathology) and Anatomy (JTP), Medical College of Virginia and School of Basic Sciences, Virginia Commonwealth University, Richmond, Va.
This work was supported by grants HL-35935 and RR01773.
Address for reprints: Dr. William I. Rosenblum, Medical College of Virginia, Neuropathology, Box 17, MCV Station, Richmond, VA 23298.
Received June 3, 1986; accepted October 13, 1986.
ing the entire microscopic field to light from a mercury
lamp while sodium fluorescein circulated through the
vasculature following intravenous injection. With that
technique, the entire length of the vessel(s) within the
field was exposed to the noxious stimulus.
In this study, the vascular bed was "sensitized" by
intravenous injection of Evans blue dye, and a laser
beam either 18 or 36 /Am diameter was used to injure a
single point along the length of a pial arteriole. This
technique permitted us to test the postinjury response
not only at the injured site but also at various distances
upstream and downstream from that site. It was hoped
that the data would reveal an asymmetric gradient of
lost response to the EDRF-dependent dilators. Such a
gradient, with greater loss downstream than upstream
from the site of injury, might imply that the relaxation
of a very short vascular segment was partially dependent on the relaxing factor washed down from the
immediately adjacent upstream site. In, addition, and
just as important, the laser/Evans blue technique could
confirm the two important conclusions of an earlier
report: first, endothelial injury in the pial microcirculation impairs dilation to acetylcholine and bradykinin,
thereby implying mediation of their relaxing action by
EDRF(s); and second, removal of endothelial cells is
not required to impair endothelium-mediated relaxation of pial arterioles. Indeed, in the earlier study
using mercury light/sodium fluorescein to injure vessels, only minor ultrastructural changes were associated with the elimination of the response, a situation that
apparently differs from that in large arteries.l>2-9 In vivo
170
impairment of endothelium-dependent relaxation in
the microcirculation has only been reported by one
other group of investigators.10
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
Materials and Methods
In Vivo Methods
Male mice, ICR strain (Dominion Labs, Va.) were
anesthetized with urethane and the pial arterioles exposed by craniotomy as previously described.""13 The
body temperature was maintained at 37° C. The cerebral surface was continuously suffused at 1 ml/min
with artificial cerebrospinal fluid14 at pH 7.35. 12 An
arteriole 30-50 fim in diameter was arbitrarily selected
for observation through a Leitz microscope. The microscopic field was illuminated from the side by a
halogen lamp and fiber optic guide. The microscope
was fitted with the objective lens turret of a Leitz
metallurgic illuminator. The metallurgic illuminator
itself was disconnected from the turret and not used.
Removal of the illuminator exposed a side port on the
turret at right angles to the objective. The beam of a
6mW helium-neon laser (Spectra Physica, Mountainview, Calif.) was directed through the side port and
downward through the objective lens by die optics
within the turret. Thus, the laser beam epi-illuminated
its target, and appeared as a red dot. Infinity-corrected
objectives were used. The beam was 18 /xm diameter
with the 20 X objective and 36 fim diameter with the
10 x objective. The 6 mW intensity of the laser beam
was confirmed with a silicon diode detector (United
Detector Corp., Santa Monica, Calif.). The calculated
energy delivered at a spot 18 /um in diameter was 2 x
109 /iW/cm 2 . The vessel was sensitized to the laser by
intravenous Evans blue 2% solution in normal saline, 1
ml/100 g body wt given 30 minutes before the start of
the study. The dye rapidly binds to albumin and continues to circulate.
Two extensive series of experiments were performed. In the first, the response of the arteriole was
monitored directly through the microscope with the aid
of an ocular micrometer in a 10 X wide-angle eyepiece. A 20 x objective was used in this set of studies.
In the second, TV microscopy was employed to monitor the vascular bed, and diameters were measured
using a Baez image-shearing device to shear the image
of the vessel on the TV monitor at a total magnification
of 750 X .15 Output from the image splitter was recorded on a strip chart for a permanent record of the responses.
TV microscopy was not initially employed in these
studies because it was discovered that light of sufficient intensity to provide a satisfactory image with an
ordinary TV camera was itself injurious to the vessels
once Evans blue was injected into the mice. For example, in the absence of laser injury the light coming
through the fiber optic guide from the 150 watt halogen
lamp paralyzed the vessel and/or led to subsequent
formation of a "waist" at the site of laser injury as soon
as the laser beam struck the vessel. Therefore, the light
used for observations was greatly reduced but was
sufficient for measurements made directly through the
Circulation Research
Vol 60, No 2, February 1987
eyepiece with an ocular micrometer. However, the
light was not bright enough to produce a satisfactory
image on a CTC-6000 camera (GBC Corp., New
York, N.Y.) with an SG-102B-6 camera tube. An image-intensified silicon TV tube and camera (SIT camera) were required to obtain adequate images with the
Baez image splitter and TV monitor. Therefore, all
studies were originally performed without TV microscopy and then repeated when an SIT camera was
obtained (model 66, with a 737058-01 silicon tube,
Dage-MTI, Michigan City, Ind.).
Two maneuvers were required to eliminate the toxic
effects of light from the halogen lamp in the Evansblue-sensitized bed. First, the light was drastically
reduced with an iris diaphragm. Second, the reduced
light intensity was used only when monitoring the vessel and was turned off during "rest" periods between
observations.
An increased toxicity of the laser was noted in preliminary studies where the suffusate of artificial cerebrospinal fluid (CSF) crossing the craniotomy site was
maintained at 37° C. Therefore the temperature of the
suffusate was reduced to 26° C. At the higher temperature, many arterioles displayed narrowing at the site of
laser damage. At the lower temperature, mis occurred
in only 10% of the mice, which were discarded without
further study. This observation of greater toxic effects
at higher temperature may support the opinion of those
who originally devised the laser/Evans blue technique
that endothelial damage was heat mediated (laser
"burn" with light absorbed and heat generated by the
dye).16
If exposure to the laser is sufficiently long, platelet
aggregation is initiated at the site of injury.16 The object was to stop the injury before platelet aggregation
was initiated, thereby observing the effects of endothelial damage on vascular responsiveness without me
complicating action of substances released by aggregating platelets. This was accomplished by reducing
the duration of exposure to the laser to 20 seconds with
either the 10 x or the 20 x objective. Under these
conditions, 25% of the injured vessels displayed
thrombi at the site of injury, and these mice were
discarded without further study. The experimental design was as follows: After the craniotomy, the flow of
irrigating, mock CSF was started and the Evans blue
dye injected. After 30 minutes, a 1 ml bolus of either
acetylcholine chloride (ACh) or bradykinin triacetate
(BK, 80 /ig/ml) was suffused in the mock CSF for 60
seconds. The change in diameter was measured and, 5
minutes after washout with restoration of baseline diameter, the arteriole was exposed to the laser for 20
seconds. Ten minutes later (15 minutes after initial
application of ACh or BK), a second bolus of dilator
was applied and the change in diameter monitored.
Separate studies were performed at the laser site or at
sites 25, 55, 85, or as much as 105 fim upstream from
the center of the laser beam and also 25, 55, 85, or as
much as 105 /im downstream from the center of the
beam. As described in the "Results" section, as many
as 8 studies were conducted with a given drug, one
Rosenblum et al
Endothdlum-Dependent Dilation in Mlcrocirculation
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
study per site. In the studies employing the ocular
micrometer for vessel measurement, there were 10
mice in each study. Then studies were repeated using
the TV microscope and image splitter. Since the latter
were confirmatory in nature, only 5 mice were used in
each study. The period of 15 minutes between consecutive applications of dilator was chosen because
preliminary data showed no tachyphylaxis with this
interval. The relatively high dose of dilator (80 /Ag/ml
or approximately 10" 4 M) was used because a bolus
application of 60 seconds does not reach steady state
and requires higher concentrations than prolonged suffusions to achieve maximal relaxation. The diameters
reached 117 ± 3% (mean ± SD) of control when 80
/xg/ml was given, 111 ± 2% when 40 /xg/ml was given, and 107 ± 4% of control with 20/i.g/ml (p for dose
effect<0.01 by analysis of variance). The maximal
change in diameter produced by the bolus application
was used for all calculations. Control studies were also
performed using 30 /i.g/ml sodium nitroprusside. This
dilator is not EDRF dependent, and intact responses to
nitroprusside would rule out nonspecific damage to
smooth muscle by laser/Evans blue.1
Electron Microscopic Methods
While mice with thrombi were eliminated from in
vivo studies of response to acetylcholine or bradykinin, thrombi were deliberately produced in vessels
subjected to ultrastructural examination. This was
done because the laser spot was so small that it was
desirable to mark it with a platelet aggregate. Therefore, the damage observed would be greater than in
these in vivo studies, which tested the effect of laser/
Evans blue injury on responses to ACh or BK in the
absence of aggregation. The data will emphasize how
minimal the damage must be.
Since the focus of platelet aggregation was small, it
was desirable to avoid any chance of dislodging the
aggregate by perfusion fixation. Therefore, only topical fixative (2% paraformaldehyde and 2.5% glutaraldehyde in 0.1 M sodium phoshate buffer) was applied
to the craniotomy site of the anesthetized mouse before
it was killed. Prior to fixation, a map of the pial vasculature was drawn and the site of platelet aggregation
(laser impact) noted. The brain was carefully removed
following topical fixation and immersed in the same
fixative. Two hours later the brain was transferred to
0.1 M sodium phosphate buffer, and the cortex beneath the craniotomy was undercut to form a thin slab
with intact surface vasculature. The slab was then
placed in 1% buffered osmic acid for two hours. Then
the slab was transferred to a petri dish containing
0.1 M sodium phosphate buffer. Employing a dissecting microscope, additional cortical tissue underlying
the pial vessels was removed with a microscalpel and
discarded. The result was that a relatively thin volume
of pia was obtained with only a minute amount of
underlying cortex. With the aid of the map, die area
containing the damaged arteriole was identified along
with adjacent undamaged control segments. These
segments were dissected free, dehydrated in chilled
171
ethanol and propylene oxide, and flat-embedded in
Medcast resin (Ted Pella, Tustin, Calif.) for sectioning
with an ultrarnicrotome. Thick sections were cut parallel to the long axis of the vessel. These were stained
with toluidine blue and studied with light microscopy
to positively identify the precise arteriolar segment
containing die platelet aggregate. Next, thin sections
were cut, stained with uranyl acetate followed by lead
citrate, and examined by transmission electron microscopy. Approximately 75 sections were examined from
each arteriole. A total of 6 damaged arterioles were
studied from 6 mice in which the laser was allowed to
remain on continuously for 30-240 seconds after onset
of aggregation. The long exposure to the laser guaranteed that damage would exceed that present just prior
to the onset of aggregation. Control arterioles from the
same mice were also studied.
Results
In Vivo Demonstration of Loss of EDRF
In both the set of studies employing an ocular micrometer and the set employing TV microscopy and
Baez image splitter, laser/Evans blue injury resulted in
complete or nearly complete loss of the relaxing response to ACh and BK. Moreover, a decreasing gradient of response loss was observed both upstream and
downstream from die site of most profound injury (the
laser site).
The ocular micrometer provides reliable detection
of changes only as small as 2 /Am (4-6% of resting
diameter). Therefore, smaller changes were recorded
as no changes. Consequently, it is best to express the
results of our ocular micrometer studies in terms of the
number of times arterioles failed to dilate.
This is shown in Figures 1 and 2 for ACh and BK
respectively. The ratio of success (dilation) to failure
(no dilation) is illustrated at the laser site and points up
and downstream from its center. The ratio is compared
with that found (10/0) prior to injury, using Fisher's
exact test." Each point along the vessel represents data
from 10 different mice (total 70 mice for ACh and 60
additional mice for BK), one arteriole per mouse. For
ACh, dilation is visible on only 1 of 10 occasions at the
laser site, with increasing frequency of dilation moving up and downstream. Dilation is still significantly
impaired 55 /tm downstream and 25 /im upstream.
The results for BK in Figure 2 are similar to those for
ACh in mat only 2 of 10 arterioles showed dilation at
the laser site, and dilation was significantly impaired
as far as 55 /am downstream. Impairment was not
statistically significant upstream, but nevertheless, 3
of 10 arterioles failed to have recognizable dilation as
far as 55 fim upstream.
Control studies showed that even after laser injury
dilation to nitroprusside occurred in 10 of 10 arterioles
at the laser site and in 10 of 10 at each of die following
sites: 25 /i.m downstream; 55 /am downstream; and 55
/am upstream.
With the Baez image splitter, changes in diameter of
less than 0.5 /am can be reliably detected.1318 Consequently these data are presented in terms of die change
Circulation Research
172
RESPONSES TO ACETYLCHOLINE (80/ig/ml) AFTER LASER
INJURY TO RIAL ARTERIOLE
LASER
SITE
MICRONS DOWNSTREAM
MICRONS UPSTREAM
80
55
25
25
55
80
100
BLOOD VESSEL
Ve
DILATIONS,
>SAIO
' DDILATIONS
I
Vol 60, No 2, February 1987
FIGURE 1. Figure portrays arterioles 30-50ixm
in diameter and displays composite results of 7
separate studies each at a different site, 10 mice
each, 1 arteriolelmouse. Responses to acetylcholine were measured with aid of ocular micrometer at each site before and after laser injury. All
vessels dilated before injury. Repeat testing 10
minutes after injury revealed dilation in only I
vessel at laser site and in an increasing proportion
of vessels upstream or downstream. Proportion of
dilations to no dilations was significantly reduced
(Fisher's exact test) as far as 55 yjn downstream
or 25 \un upstream.
•Significontly different from IO/O by Fisher test (ps.O5)
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
in diameter produced in relation to the baseline diameter, taken to be 100%.
In Table 1, the data are presented for 35 mice tested
with ACh, 5 mice at each site along the arteriole.
Dilation was significantly impaired (paired t test) at the
laser site and as far as 55 pirn downstream. No statistically significant impairment was observed upstream
from the laser site. However, this may be a function of
the small n at each site, since one sees a diminution in
the postlaser response at each upstream point, with the
difference between prelaser and postlaser responses
diminishing in magnitude farther upstream. The TVmicroscopy and image-splitting data illustrate the
same basic finding as the ocular micrometer studies:
dilation to ACh was essentially eliminated at the laser
site. Indeed, some constrictions were observed after
the injury, when ACh was applied. Also, as with the
ocular micrometer data, there was a decreasing gradient of impaired responsiveness up and downstream
from the laser site, more pronounced on the downstream side.
In Table 2 the data are presented for BK responses
observed with TV-microscopy and image splitting.
Forty mice were studied, 5 at each site. The results
show that dilation was eliminated at the laser site. A
statistically significant difference between prelaser and
postlaser responses was observed as far as 80 fim
downstream and 55 /im upstream. Thus, the data are
similar to those in the study employing an ocular micrometer, since in both studies there was severe impairment at the laser site. In the TV-microscopy stud-
ies, the impairment both downstream and upstream
was more pronounced than in the study employing the
ocular micrometer. However, as in the study employing the micrometer, the TV-microscopy study revealed
asymmetrical loss of response around the laser site.
The loss was demonstrable 80 /xm downstream and 55
fim upstream from the center of the laser site.
With sodium nitroprusside, dilations were unaffected at the laser site (127 ± 12% of control diameter
prior to injury and 122 ± 14% (mean ± SD) after injury), once more providing evidence that no smooth
muscle damage occurred.
Electron Microscopic Studies
As stated in the "Materials and Methods" section,
our in vivo studies were carried out in the absence of
platelet aggregation. The laser was cut off before aggregation was induced. Those mice (25%) showing
aggregation in spite of the short exposure were discarded. The fact that 25% did show aggregation indicates
that the 20-second exposure was near the threshold
required to produce sufficient injury to induce aggregation. Indeed, 30-second exposures generally produced
aggregation in all mice, and the aggregates provided a
precise marker for the site of damage. Therefore, electron microscopy (EM) studies were withheld until aggregation occurred. The operating premise of these
EM studies was that any observed vascular damage
must be worse than that in the in vivo studies, which
were carried out with an insult too brief to elicit
aggregation.
RESPONSES TO BRADYKININ ( 8 0 , i g / m l ) AFTER LASER
INJURY TO PIAL ARTERIOLE
LASER
SITE
MICRONS UPSTREAM
80
55
MICRONS DOWNSTREAM
25
25
55
BLOOD VESSEL
DILATIONSIS/,
/DDILATIONS
NO
2
/e'
•Significantly different from 10/0 by Fisher test (p s .05)
80
100
FIGURE 2. Identical to Figure 1 except that
bradykinin is dilator. Following laser injury ratio
of dilations to no dilations was significantly reduced at laser site and as far as 55 fim downstream or 25 fjun upstream.
173
Rosenblum et al Endothellum-Dependent Dilation in Microcirculation
Table 1. Responses to Acetylcholine Before and After Laser/Evans Blue Injury to Pial Arterioles
Diameter (% control)*
Distance from laser center (in /xm)
Upstream
80
55
25
Laser site
Downstream
25
35
80
Before laser/Evans blue
After laser/Evans blue
120±6
123±4
121 + 14
I27±5
116±11
I1O±14
106±14
97 ± 5
>0.25
>0.10
>0.10
<0.01
124±14
112 + 4
112 ± 6
104±22
102 + 2
109±3
<0.05
<0.05
>0.30
•Mean±SD; tpaired / test.
Results of 7 separate studies, 5 mice each, I arteriole 30-50 /xm in diameter/mouse. Each study evaluatedresponseto
acetylcholine at different site. Responses were evaluated with image splitting and TV microscopy at site shown, 5
minutes before, and 10 minutes after, laser injury. Prior to injury, all vessels dilated at observed site. After injury, dialtion
was eliminated or significantly impaired (paired / test) at laser site and 25 or 55 /xm downstream.
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
Electron microscopic examination of the arterioles
containing aggregates revealed no foci of endothelial
denudation. Basal lamina was not exposed even after
laser exposure for 4 minutes following the onset of
local platelet aggregation. During this time, platelet
aggregation was well established and continued to in-
crease. Serial sacrifice of mice from 30 seconds to 4
minutes after the onset of aggregation revealed an initial, local dilation of both the smooth- and rough-surfaced endoplasmic reticulum within the endothelial
cells. Additionally, various vacuoles of unknown origin were observed, dispersed throughout the endothe-
•
P
P
T
( *
SM
S:M
FIOURE 3. Electron micrograph displaying laser-induced endothelial damage consisting of intracytoplasmic vacuoles (V) and
dilated profiles of smooth- and rough-surfaced endoplasmic reticulum (*). Underlying smooth muscle (SM) is normal. Over damaged
endothelium are aggregating, spherical platelets (P), some of which display pseudopods H>) . Note that endothelium-dependent
relaxation is eliminated before platelet aggregation begins.
Circulation Research
174
Vol 60, No 2, February 1987
Table 2. Responses to Bradykinin Before and After Laser/Evans Blue Injury to Pial Arterioles
Diameter (% control)*
Distance from laser center (in
Before laser/Evans blue
After laser/Evans blue
Upstream
116±6
80
111+2
>0.15
104+11
120±4
= 0.05
55
116±3
101 ±10
<0.05
25
116±4
95 ±7
<0.01
Laser site
Downstream
110±17
127±12
<0.05
25
117±8
105
+
8
<0.06
35
115±2
105 ±5
<0.01
80
114±5
110±13
>0.30
105
•Mean±SD; tpaired / test.
Data from 8 separate studies like those in Table 1 except that bradykinin was dilator. Before laser injury, all arterioles
were dilated by bradykinin. After injury, significant (paired t test) impairment or elimination of dilation was observed at
laser site and as far as 55 /xm on upstream side or 80 /xm on downstream side.
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
Hal cytoplasm (Figure 3). Occasionally, these were in
perinuclear location (Figure 4). From 2-4 minutes
after the onset of aggregation, more intense endothelial
damage was manifest by increased vacuolation and in
one case by disintegration of the luminal cell membrane. However, even in this case the abluminal cell
membrane remained intact and attached to the underlying basal lamina. In all cases, the smooth muscle within the media showed no abnormality.
The platelet masses themselves consisted not only of
activated or aggregating platelets but also of large
masses of discoid platelets in proximity to the sites of
FIGURE 4. Electron micrograph displaying aggregating platelets trj m proximity w damaged endotheltum. Ike latter displays
vacuoles (V) and perinuclear swelling (S). Underlying smooth muscle is normal (SM). Note that endothelium-dependent relaxation is
eliminated before platelet aggregation begins.
Rosenblum et al
Endothelium-Dependent Dilation in Microclrculatkm
175
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
^
FIGURE 5. Control arteriole, harvested from site adjacent to laser injury, shows normal red blood cells (RBCs), endothelium, and
smooth muscle (SM).
endothelial damage. Activation or aggregation of
platelets was manifest by central migration of platelet
dense bodies and by pseudopod formation as well as by
degranulation. Red blood cells were dispersed occasionally throughout the aggregates and appeared normal. Fibrin was not observed.
The control vessels were unremarkable except for
extremely rare foci of endothelial vacuolation in 2
mice. Collections of discoid platelets could be seen in
controls, apparently a result of local application of
fixative. Such collections were never seen in our studies of perfusion-fixed vessels."
Since loss of EDRF occurred prior to onset of platelet aggregation these ultrastructural studies of damage
after onset of aggregation establish that elimination of
endothelium-dependent relaxation in pial arterioles occurred prior to disintegration of the endothelial cell and
at a time when only endothelial vacuolation could have
been present. Of added interest is the fact that in these
microvessels aggregation did not depend either on exposure of platelets to basal lamina or on disintegration
of red blood cells.
Discussion
The relaxation to ACh and BK was clearly eliminated by laser/Evans blue injury. The response to nitroprusside was not altered even though relaxation by
nitroprusside, like relaxation by EDRF(s) is ultimately
mediated by elevations of cyclic guanosine 5'-monophosphate in muscle (cGMP). u The normal response
to nitroprusside, like the ultrastructural data, therefore
indicates that vascular smooth muscle was not injured
by laser/Evans blue. The findings are comparable to
those previously reported9 using a different in vivo
insult (blue light/sodium fluorescein). Thus, we confirm that endothelium-dependent relaxation exists in
the microcirculation, at least of the pial vasculature.
In the brain, both ACh and BK20 may be produced
and/or released in large amounts. Therefore, our data
are significant in that they provide reason for suggesting that endothelial damage would reduce dilator influences on resistance vessels of the brain and might
thereby exacerbate harmful constriction or spasm. In
addition, such injury might modify normal dilator responses to locally produced humoral agents or to ACh
released by perivascular nerves to pial arterioles.21 Indeed, constriction is the direct effect of ACh on vascular smooth muscle, in the absence of E D R P and in the
mouse BK can also constrict in the absence of EDRF as
shown here and also in our earlier work.9 Consequently
these agents themselves might contribute to spasm in
the absence of EDRF.
Kontos 10 damaged pial arterioles by generating free
radicals by several methods, including acute hypertension. They also showed loss of relaxation to ACh. In
their studies and in present and earlier investigations9
176
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
loss of EDRF occurred without removal of endothelial
cells and in the presence of only vacuolation as evidence of endothelial injury. Thus our data, as well as
Kontos', indicate a major difference between microcirculation and large arteries. In the latter, loss of
EDRF has thus far only been detectable after complete
removal of a substantial percentage of the endothelial
cells. 12
In the present study, the laser beam reduced endothelium-dependent relaxation over a wider area than
that covered by the beam itself. This may simply reflect a gradient of damage, for example of heat produced by the interaction of laser and Evans blue. Indeed the mechanism of laser/Evans blue injury has
been thought to involve heat production.16 The gradient of diminished responsivity was at least 55 jxm on
either side of the center of the laser beam, so the
overall area of heat damage may have been about 110
pirn in diameter. The loss of endothelium-dependent
dilation was somewhat greater downstream than upstream from the laser site, but this may reflect preferential heating of the downstream site as the flowing
blood carries heat in that direction. Alternatively, there
may be some dependence of the downstream relaxation on upstream EDRF, carried by the flowing
blood. EDRF(s) has not yet been identified chemically
but has a half-life measured in seconds in salt solutions
and may be rapidly inactivated by hemoglobin.'-2-6
Nevertheless, since arteriolar flow velocity is several
millimeters per second22 it is clear that EDRF might be
carried 50-100 ptm downstream without serious degradation. In this case, a downstream segment might suffer not only because it was directly damaged by heat,
but also because it was deprived of EDRF from the
immediately adjacent upstream segment.
References
1. Furchgott RF, Cherry PD, Zawadiki JV, Jothianandan D: Endothelial cells as mediators of vasodilation of arteries. J Cardiovasc Pharmacol 1984;6:5336-5343
2. Furchgott RF: Role of endothelium in responses of vascular
smooth muscle. Circ Res 1983;53:557-573
3. Vanhoutte PM, deMay J: Control of vascular smooth muscle
function by endothelial cells. Gen Pharmacol 1983; 14:39-41
4. Kontos HA, Wei EP, Povlishock JT, Christman CW: Oxygen
radicals mediate the cerebral arteriolar dilation from arachidonate and bradykinin in cats. Circ Res 1984;55:295-303
Circulation Research
Vol 60, No 2, February 1987
5. Singer HA, Saye JA, Peach MJ: Effects of cytochrome P-450
inhibitors on endothelium dependent relaxation in rabbit aorta.
Blood Vessels 1984;21:223-230
6. Griffith TM, Edwards DH, Lewis MJ, Newly AC, Henderson
AH: The nature of endothelium-derived relaxing factor. Nature
1984;308:645-647
7. Rosenblum WI, Wei EP, Kontos HA: Microvascular endothelial dependent relaxing factor (EDRF) in vivo: Interference by
minimal endothelial injury (abstract). FedProc 1985;44:1262
8. Rosenblum WI: Minor endothelial damage converts relaxation
of brain arterioles in vivo to constriction (abstract). J Exp
Neurol Neuropath 1985;44:339
9. Rosenblum WI: Endothelial dependent relaxation demonstrated in vivo in cerebral arterioles. Stroke 1986; 17:494-497
10. Kontos HA: Oxygen radicals in cerebral vascular injury. Circ
Res 1985;57:508-516
11. Rosenblum WI, Zweifach BW: Cerebral microcirculation in
the mouse brain. Arch Neurol 1963;9:414-423
12. Rosenblum WI: Constriction of pial arterioles by prostaglandin
F 2a . Stroke 1975;6:293-297
13. Rosenblum WI: Pial arteriolar responses in the mouse brain
revisited. Stroke 1976;7:283-287
14. Elliott KAC, Jasper HH: Physiologic salt solutions for brain
surgery. J Neurosurg 1949;6:140-152
15. BaezS: Recording of microvascular dimensions with an image
splitter television microscope. J Appl Physiol 1966;21:
229-301
16. Kovacs IB, Tigyi-Sebes A, Trombitas K, Gorog P: Evans blue:
An ideal energy-absorbing material to produce intravascular
microinjury by He-Ne gas laser. Microvasc Res 1975;
10:107-124
17. SiegelS: Nonparametric Statisticsfor the Behavioral Sciences,
New York, McGraw-Hill Publishing, 1956
18. Levasseur JE, Wei EP, RaperAJ, Kontos HA, Patterson JR Jr:
Detailed description of a cranial window technique for acute
and chronic experiments. Stroke 1975;6:308-317
19. Povlishock JT, Rosenblum WI, Sholley MM, Wei EP: An
ultrastructural analysis of endothelial change paralleling platelet aggregation in a light/dye model of microvascular insult.
Am J Path 1983;110:148-160
20. Kamitani T, Little MH, Ellis EF: Evidence for a possible role
of the brain kallikrein-kinin system in the modulation of the
cerebral circulation. Circ Res 1985;57:545-552
21. Edvinsson L: Neurogenic mechanisms in the cerebrovascular
bed. Autonomic nerves, amine receptors and their effects on
cerebral blood flow. Ada Physiol Scand (supplement 427)
1975
22. Rosephlum WI: Erythrocyte velocity and a velocity pulse in
minut /essels on the surface of the mouse brain. Circ Res
1969; :887-892
KEY WOR
vessels •
injury
• endothelium-dependent relaxing factor • pial
:erebral microcirculation • laser • endothelial
Laser-induced endothelial damage inhibits endothelium-dependent relaxation in the
cerebral microcirculation of the mouse.
W I Rosenblum, G H Nelson and J T Povlishock
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
Circ Res. 1987;60:169-176
doi: 10.1161/01.RES.60.2.169
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1987 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/60/2/169
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/