1157
In Vivo Studies of Pial Vascular Permeability to
Sodium Fluorescein: Absence of Alterations by
Bradykinin, Histamine, Serotonin, or
Arachidonic Acid
Manabu Watanabe, MD, and William I. Rosenblum, MD
Downloaded from http://stroke.ahajournals.org/ by guest on July 31, 2017
We studied pial vessels in vivo in mice, examining under mercury light the permeability of these
vessels to sodium fluorescein. Topical application of hypertonic solutions of NaCl caused leakage of
fluorescein. However, putative chemical mediators of leakage, such as histamine, serotonin, arachidonic acid, and bradykinin, all failed to increase permeability to the dye. Apparent increases in
permeability only accompanied endothelial damage caused by the dye + light combination, as indicated by production of local platelet aggregates. The technique is useful, provided inadvertent endothelial injury is recognized and avoided. The data in mice suggest that pial vessels may not participate in
the permeability changes reportedly produced in parenchyma! brain vessels by several of the mediators we tested. Therefore, studies of pial vascular permeability are not expected to provide reliable
data concerning the actions of agents that might mediate cerebral edema. (Stroke 1987;18:1157-1159)
F
ew publications selectively describe the effects
of putative mediators of cerebrovascular permeability on pial vessels. '"3 Such observations
are important because these mediators may affect pial
and parenchymal vessels differently.43 Arachidonate
and bradykinin reportedly increase pial vascular permeability to sodium fluorescein, but with a tremendous interanimal variation in threshold dose, 23 suggesting that extraneous factors might be at work in
those studies. Moreover, not everyone finds that arachidonate increases permeability of brain vessels.6 We
reevaluated the effects of arachidonate and bradykinin
and tested other agents as well.
Materials and Methods
Male mice, ICR strain (Dominion Labs, Dublin,
Virginia) were anesthetized with urethane, prepared,
and maintained as previously reported for in vivo microscopic study of pial vessels. 78 Observations were
made with a Leitz Ultropak microscope (Rockleigh,
N.J.) using incident light from a 200-W mercury lamp
heavily filtered with a KG-1 heat filter and a BG-12
blue filter. Observations were made with 10 x widefield eyepieces and an 11 x Ultropak objective with
dipping cone. The space between the cone and the
brain surface was filled with artificial cerebrospinal
fluid at pH 7.35, 9 which was periodically replaced with
fluid containing putative mediators of permeability.
Histamine phosphate, serotonin creatinine sulfate,
bradykinin triacetate (Sigma Chemical Co, St. Louis,
From the Department of Pathology (Neuropathology), Medical
College of Virginia, Virginia Commonwealth University, Richmond, Va.
Supported in part by Grant HL-35935 from the National Heart,
Lung, and Blood Institute.
Address for reprints: William 1. Rosenblum, MD, Medical College of Virginia, Neuropathology, Box 17, MCV Station, Richmond, VA 23298.
Received March 23, 1987; accepted June 3, 1987.
Mo.), or sodium arachidonate (Nuchek, Elysian,
Minn.; 99% pure) were made up in fresh solutions of
artificial cerebrospinal fluid each day; the pH was adjusted to 7.35 by altering its CO2 content as necessary.
Hypertonic solutions were made by dissolving NaCl in
deionized water with no effort made to control pH.
Sodium fluorescein, mol wt 326, was made up as a
2% solution in saline. The leakage of dye was evaluated by counting the leaking spots in a preselected microscopic field containing arterioles and venules. Solutions to be tested for their ability to produce leaks
were left in place at the craniotomy site for either 5 or
30 minutes. After this time, we injected the 2% sodium
fluorescein, 0.2 ml/25 g body wt, via tail vein and
counted leaks in the microscopic field 15 seconds later.
A variable number of control mice displayed leaks,
apparently due to operative trauma. In 2 studies, 1 with
bradykinin and 1 with arachidonate, such mice were
weeded out by injecting a very small preliminary dose
of 2% sodium fluorescein (0.05 ml/25 g rather than 0.2
ml/25 g). This injection was given before applying
arachidonate or bradykinin. Only mice that showed no
leaks of the test dose of dye were used in further studies. In these studies, the test drug was then applied 30
minutes after the preliminary injection of fluorescein
and left in place for 30 minutes. During this time,
fluorescence from the low-dose preliminary injection
of sodium fluorescein faded to very low levels. Consequently, the regular indicator dose of sodium fiuorescein (0.2 ml/25 g) was injected 30 minutes after the
onset of exposure to arachidonate or bradykinin, and
the leaks were counted 15 seconds later as usual.
Results
Studies with Hypertonic NaCl
As a positive control demonstrating our ability to
detect leaks, we applied hypertonic NaCl to the crani-
1158
otomy site10 in 4 groups of 10 mice each. Each group
received a different concentration of NaCl for 5 minutes. Table 1 shows that the total number of mice with
leaks in each group decreased with declining concentration of NaCl. Normotonic NaCl caused minor leakage in only 2 of 10 mice. The amount of leakage at the
high dose was massive, with 99 separate leaks counted
as opposed to only 3 leaks counted with 0.9% NaCl.
Since 0.9% NaCl is isotonic, these mice may be considered controls, and the small number of leaks in this
group indicates that some leakage may result from
surgical trauma rather than from the nature of the superfused solution.
Downloaded from http://stroke.ahajournals.org/ by guest on July 31, 2017
Studies with Histamine or Serotonin
Preliminary studies using 1.8,18, or 36 /xg/ml histamine for 5-minute applications showed leaks in 1 of 3
mice at each dosage. Five additional mice were then
tested with 36 /ig/ml and compared with 5 untreated
controls. Only 2 of the treated and 1 of the control mice
leaked. The lack of a dose effect and the similar degree
of leakage in treated and control mice led us to conclude that the leakage was probably related to otherwise undetectable trauma rather than to histamine.
For similar reasons, we concluded that serotonin
produced no leaks. Three groups of 3 mice each received 5, 10, and 40 /Ag/ml serotonin, respectively, for
5 minutes. In each group, only 1 mouse leaked, as did
1 mouse of each group of 3 controls paired with the
experimental mice.
Studies with Sodium Arachidonate
Arachidonate was continuously in contact with the
pial surface for 30 minutes3 before testing for leaks.
Since we found a variable but small percent of control
mice with leaks in the preceding studies, we screened
out leaking mice by giving a small preliminary dose of
sodium fluorescein before beginning the experiment
(see "Materials and Methods"). However, in an occasional mouse even this small dose of fluorescein triggered endothelial damage and platelet aggregation in
pial vessels exposed to mercury light for as long as 30
minutes.""13 These few mice were also discarded, and
nontraumatized mice replaced them. The craniotomy
site of 1 group of 10 mice received 100 /ag/ml sodium
arachidonate for 30 minutes. Another group of 5 mice
received 1,400 /ig/ml, an extraordinarily high concentration stated to break the blood-brain barrier in some
animals described by others.3 The usual indicator dose
Table 1. Effect of Hypertonk NaCl on Permeability of Mouse
Hal Vessels to Fluoresceta
NaCl
1.6%
0.9%
4.8%
3.2%
3/7
2/8
Leaking/nonleaking
10/0
9/1
Hypertonic solutions of NaCl caused leakage of dye from mouse
pial vessels, with declining effect as tonicity decreased. Normotonic NaCl caused minor leakage in only 2 of 10 mice, probably
reflecting operative trauma.
Stroke
Vol 18, No 6, November-December 1987
Table 2. Effect of Sodium Arachidonate on Permeability of
Mouse Pial Vessels to Fluorescein
Sodium arachidonate
100 yiglm\
Leaking/nonleaking
0/10
1500 Mg ml
0/5
In these groups of mice, traumatized animals leaking dye, were
identified prior to the start of the experiment and were replaced by
nontraumatized mice. Sodium arachidonate, applied for 30 minutes, even in exceedingly high dose, failed to elicit leaks.
of fluorescein was then injected. No mice in either the
low- or high-dose group leaked (Table 2).
Studies with Bradykinin Triacetate
A 30-minute exposure of bradykinin triacetate was
used because this was reported to be effective in the
literature.2 A preliminary 2-dose study was performed
in which we made no effort to weed out mice that
might leak as a result of surgical trauma. Three mice
received 30 /i.g/ml bradykinin triacetate, 10 received
150 /Ag/ml, and 10 served as controls. Essentially the
same proportion of mice (20-33%) leaked in each
group. Nevertheless, we decided to study many more
mice at the higher dose of bradykinin triacetate.
In the larger study, we screened out mice with traumatic leaks (see "Materials and Methods"); 150 /ng/ml
bradykinin triacetate was then applied for 30 minutes
to each of 20 mice, which were compared with 20
controls (Table 3). Seven experimental and 5 control
mice leaked — a difference very far from statistical
significance, showing that bradykinin was not the
cause of leakage. However, platelet aggregates were
seen in a number of mice in each group — resulting
from the long exposure to fluorescein dye plus mercury
light.""13 Such mice had been discarded prior to final
observation in the arachidonate study but were not
discarded in this study because we wished to see
whether the endothelial damage causing these aggregates predisposed the vessels to leaks. When we analyzed the data this way, we found that of the 7 leaking
mice treated with bradykinin, 6 had aggregates; of the
5 leaking control mice, 4 had aggregates. The association of leaks with aggregates was significant at the
0.05 level (Fisher's test of exact probability).
Since there was an association between leakage and
platelet aggregation, we performed another study in
which the confounding factor of damage by light + dye
was eliminated. As before, we screened out mice with
Table 3. Effect of 150 /xg/ml Bradykinin Triacetate on Permeabllity of Mouse Pial Vessels to Fluorescein
With aggregates Without aggregates
With Without
With
Without
leaks
leaks
leaks
leaks ;
4
1
1
14
Control (n = 20)
1
Bradykinin (n = 20)
9
6
4
In 10 of the 12 leaking mice, endothelial injury sufficient to
produce platelet aggregation had been produced by light + dye;
association of aggregation and leakage was significant at p<0.05,
Fisher's test.
Watanabe and Rosenblum Permeability of Pial Vessels
surgical trauma by giving a small initial dose of
fluorescein, but to prevent any endothelial injury induced by the dye combined with the light over the next
hour, we placed neutral density filters in the light path.
Bradykinin triacetate (150 £ig/ml for 30 minutes) then
produced no leaks in 10 mice.
Downloaded from http://stroke.ahajournals.org/ by guest on July 31, 2017
Discussion
At the doses used, neither serotonin, histamine, sodium arachidonate, nor bradykinin caused pial arterioles or venules to leak sodium fluorescein. In contrast,
our data with NaCl confirms classical studies showing
leakage, especially from venules, following local application of hypertonic solutions.10 The negative results with histamine also agree with studies that monitored parenchymal14 or pial vessels,1 though contrary
reports also exist.1316
The negative results with arachidonate and bradykinin conflict with reports in cats, in which fluorescein
was also used to indicate leakage from pial vessels. 23
We have shown in mice that these concentrations of
sodium arachidonate or bradykinin produce vasoconstriction17 or dilation,18 respectively. Consequently,
we know that topically applied drugs reach the pial
arterioles. Since relaxation by bradykinin is mediated
by a factor released only when bradykinin reaches the
endothelium,1819 we also know that our failure to alter
permeability was not due to a failure of the drug to
reach the endothelium. Species differences might explain why the permeability of cat pial vessels was
altered by arachidonate and bradykinin whereas the
permeability of mouse pial vessels was not.
However, we should also consider the possibility
that published results with bradykinin and arachidonate 23 were caused by damaging pial vessels with
light in the presence of intravascular sodium fluorescein.11"13 One must be careful that the excitation of
fluorescein by mercury light does not induce endothelial injury, which is manifest by local aggregation of
platelets. These platelet masses fluoresce and can look
exactly like the clumpy material seen along the venules
in Figure 5a and b of Unterberg et al.2 Other evidence
of endothelial damage would be loss of endotheliumdependent relaxation of arterioles by bradykinin.1819 In
fact, the arterioles in Figure 5 of Unterberg et al2 did
not respond to bradykinin. In view of 2 separate indications of endothelial trauma in the paper of Unterberg
et al, 2 it may be that leakage was produced because
arachidonate3 or bradykinin2 acted on injured rather
than normal vessels. This suggestion is valid even in
cases in which fluorescent platelet aggregates cannot
be recognized in the figures since the aggregates may
have embolized and left the field. In support of the
suggestion relating endothelial trauma to dye leaks was
our own finding of a significant relation between apparent dye leakage and recognizable platelet aggregation in our bradykinin studies.
Once the effects of surgical trauma and damage by
light + dye were taken into account, we failed to find
that histamine, serotonin, arachidonate, or bradykinin
1159
break the barrier of the pial vessels to small molecules
like sodium fluorescein. Nevertheless, one or more of
these agents may mediate edema formation in disease
states, with barrier breakdown occurring in parenchymal vessels. Based on our studies, we suggest that pial
vessels are not suitable for studying the effects on
permeability of mediators potentially important for the
production of cerebral edema.
References
1. Broman T, Lindberg-Broman AM: An experimental study of
disorders in permeability of the cerebral vessels ("the blood
brain barrier") produced by chemical and physico-chemical
agents. Ada Physiol Scand 1945; 10:102-125
2. Unterberg A, Wahl M, Baethmann A: Effects of bradykinin on
permeability and diameter of pial vessels in vivo. J Cereb
Blood Flow Metab 1984;4:574-585
3. Wahl M, Unterberg A, Baethmann A: Effects of arachidonic
acid on blood brain barrier function, in Dietz H, Brock M,
Klinger M (eds): Advances in Neurosurgery, vol 13. Berlin,
Springer, 1985, pp 323-325
4. Hardebo JE, Owman C: Barrier mechanisms for neurotransmitter monoamines and their precursors at the blood-brain
interface. Ann Neurol 1980;8:l-l 1
5. Wei EP, Ellison MD, Kontos HA, Povlishock JT: O2 radicals
in arachidonate induced increased blood brain barrier permeability to proteins. Am J Physiol 1966;251:H693-H699
6. Domer FR, Boertje SB, Sampson JB: Permeability of the blood
brain barrier is not changed by vasoactive doses of arachidonic
acid or prostaglandins in rats. Chem Pathol Pharmacol 1985;
48:27-37
7. Rosenblum WI: Pial arteriolar responses in the mouse brain
revisited. Stroke 1976;7:283-287
8. Rosenblum WI, Zweifach BW: Cerebral microcirculation in
the mouse brain. Arch Neurol 1963;9:414-423
9. Elliott KAC, Jasper HH: Physiologic salt solutions for brain
surgery. J Neurosurg 1949;6:140-152
10. Rapoport SI, Hori M, Klatzo I: Testing of a hypothesis for
osmotic opening of the blood brain barrier. Am J Physiol
1972;223:323-331
11. Rosenblum WI: Fluorescence induced in platelet aggregates as
a guide to luminal contours in the presence of platelet aggregation. Microvasc Res 1978;15:103-106
12. Rosenblum WI, El-Sabban F: Platelet aggregation in cerebral
microcirculation. Effect of aspirin and other agents. Circ Res
1977 ;40:320-328
13. Povlishock JT, Rosenblum WI, Sholley MM, Wei EP: An
ultrastructural analysis of endothelial change paralleling platelet aggregation in a light/dye model of vascular insult. Am J
Pathol 1983;110:148-160
14. Gross PM: Cerebral histamine: Indications for neuronal and
vascular regulation. J Cereb Blood Flow Metab 1982;2:3-23
15. Foldcs I, Kelentei B: Studies on the haemato-encephalic barrier
II. The effects of histamine with special reference to the passage of antibiotics. Ada Physiol Hung 1954;5:149—161
16. Hurst EW, Davies OL: Studies on the blood brain barrier, II.
Attempts to influence the passage of substances into the brain.
BrJ Pharmacol 1950;5:147-163
17. Rosenblum WI: Unsaturated fatty acids and cyclooxygenase
inhibitors: Effects on pial arterioles. Am J Physiol 1982;
242:H629-H632
18. Rosenblum WI: Endothelial dependent relaxation demonstrated in vivo in cerebral arterioles. Stroke 1986;17:494-497
19. Rosenblum WI, Nelson GH, Povlishock JT: Laser-induced
endothelial damage inhibits endothelium-dependent relaxation
in the cerebral microcirculation of the mouse. Circ Res
1987;60:169-176
KEY WORDS • sodium fluorescein • blood-brain barrier •
bradykinin • histamine • serotonin • arachidonic acid •
permeability of brain blood vessels
In vivo studies of pial vascular permeability to sodium fluorescein: absence of alterations by
bradykinin, histamine, serotonin, or arachidonic acid.
M Watanabe and W I Rosenblum
Stroke. 1987;18:1157-1159
doi: 10.1161/01.STR.18.6.1157
Downloaded from http://stroke.ahajournals.org/ by guest on July 31, 2017
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1987 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/18/6/1157
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Stroke 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 Stroke is online at:
http://stroke.ahajournals.org//subscriptions/