L@- 6 -D
Atherosclerosis, 46 0983) 137-146 1371
Elsevier Scientific Publishers Ireland. Ltd.
Effect of Vasoactive Agents and Applied Stress on
A the Albumin Space of Excised Rabbit Carotid
Arteries
C.G. Caro and mi. Lever
PkYstalorcal Flow Studies Unit. ImperialCollege, London SW72AZtGrear Britain)
(Received 26 July. 1982)
(Revised. received 7 Scptcmher. 1982)
(Accepted 9 September. 1982)
Sumnmry
Experiments were undertaken on excised rabbit common carotid arteries to assess
whether the distribution volume for radioactively labelled albumin is influenced by
smooth muscle tone or externally applied stress. In arteries at relaxed length and
zero transmural pressure. the distribution volume in the media was reduced by
noradrenaline at concentrations exceeding 10' M and increased by both sodium
nitrite and isosorbide dinitrate at concentrations exceeding 10' M. The distribution
volume was lower in both the media and adventitia of segments at in vivo length
pressurized with air to prevent convection through the wall. than in segments at
relaxed length and zero transmural pressure. Noradrenaline decreased and sodium
nitrite increased the medial distribution volume of the tracer in these air-pressurized
segments. The vasoactive agents did not change the ["CrJEDTA space in vessels at
relaxed length and zero transmural pressure, although there were small alterations of
medial water content. Transient conditions raise interpretative difficulties because of
uncertainty about cndothelial permeability and time-dependent changes of wall
properties. The changes of the distribution volume for labelled albumin are thought
to reflect changes in the properties of the interstitium.
Key words: Albumin space - Carotid artery - Haemodynamics - Rabbit - Smooth
muscle - Vasoactive agents
Tbis work was supported by the Medical Research Council, The National Heart Research Fund and
Sanol Schwarz Pharmaceuticals Ltd., U.K.
Correspondence to: Prof. C.G. Caro. P.F.S.U.. Imperial College. Prince Consort Road. London SW7
2AZ, Great Britain.
0021-9150/93/ODOO-0000/SO3.00 0 1983 Elsevier Scientific Publishers Ireland. Ltd.
BATCo document for Province of British Columbia 15 April 1999

138
Introduction
Several workers have reported findings suggestive of movement of water and
various solutes through the arterial wall from the blood to the adventitial lymphatics
[1.21. It has been proposed, furthermore. that this transport is implicated in the
development of atheroma. A quantity which can be expected to reflect the ease of
transport of a material in a tissue is the distribution volume (fractional volume of the
tissue occupied by the material at equilibrium). In the course of investigations on
excised arterial segments incubated at relaxed length and zero transmural pressure.
we observed that the medial concentration of radioactively labelled albumin at first
rises rapidly and then slowly (2) and that the arterial segments dilate (3). The
findings suggested the possibiiit, that the rate of uptake of and distribution volume
for the tracer in the media depend on applied stresses, including those arising from
smooth muscle tone. We have undertaken a study to test for this dependence.
Methods
The experimental procedures are described in detail elsewhere [2). Male or female
New Zealand white rabbits fed a normal diet, were anaesthetized with sodium
pentobarbitone, 30 mg/kg i.v. and given 1000 IU heparin i.v. The common carotid
arteries were then exposed, excised and incubated for varying periods of time at
38 I'C with solutions containing tracer. The tracer was either radioiodinated
human or bovine serum albumin purified by ultrafiltration, or 5, ethylene
dian-Line tetraacetate (15'Cr]EDTA. Ainersham International). The incubating solu-
tion in early experiments was homologous plasma but. since this might contain
vasoactive material, Tyrodes solution containing 4% bovine serum albumin was used
in later experiments. Bubbling 551c, CO, in air through the incubating solution had no
effect on tracer uptake. After incubation the arterial segments were opened axially.
briefly rinsed. rapidly frozen and placed in a cryostat at - 300C. They were then
trimmed and, after determination of surface area from length and breadth. were
sectioned parallel to the lurninal surface at 20-jAm intervals. Label uptake of each
section was expressed as a tissue/plasma radioactivity concentration ratio. CTICPI
cpm - cm -3 wet tissue/cpm - cm-' incubating solution. These data were used to
construct graphs of the concentration ratio profile across the wall. The thickness of
the wall of a given arterial segment depends on factors including transmural
pressure. axial extension and smooth muscle tone. As a means of comparing the
steady concentration ratio profiles of segments from different animals. or of seg-
ments subjected to different experimental conditions. medial thickness was made
equal to 100 units; the average thickness of the media of segments at relaxed length
and zero transmural pressure ('relaxed' segments) was 124jum when fully constricted
by noradrenaline and 99 mm when fully dilated by sodium nitrite. Because the
distinction between the media and adventitia was based on the appearance of the
20-lim sections, there was uncertainty concerning the precise location of the medial:
adventitial boundary. Average medial uptake was therefore calculated from the
C,/C, values of the sections lying within 80 units of the luminal surface.
LIKI
L.r
Xh-
Un
BATCo document for Province of British Columbia 15 April 1999

139
In the first and second series of experiments the segments were 'relaxed'. ap-
proximately 0.5 cm in length and each provided a single concentration profile. In the
first series, half the segments were incubated with label for 20 min immediately after
excision and the remainder were incubated with label for the same time after first
being incubated for 3 h in unlabelled plasma. In the second series. the segments were
incubated for 20, 90 or 180 min immediately after excision in labelled Tvrodes
solution containing noradrenaline, sodium nitrite or isosorbide dinitrate (ISDN) at
concentrations up to 10- I M. In the third series. the arteries were cannulated in situ
and mounted in a rig without being allowed to shorten. They were then incubated at
in vivo length and pressurized with saturated air to 70 mmHg ("air-pressurized"
arteries). If an artery is pressurized with plasma. convective flow occurs across the
wall, which may influence both the conformation of the wall and the distribution of
tracer within it. Such convective flow was overcome by air pressurization. yet the
action of vasoactive agents on tracer uptake could still be studied at normal
transmural pressure. Flow of air along the vessels was prevented by clamping the
downstream cannula. The vessels were then immersed in labelled Tyrodes solution
which, to simulate interstitial fluid. contained 21% albumin. Each artery in this series
yielded three segments and an average concentration profile across the wall was
constructed. The water content of the media of the arteries was determined in each
of the three series of experiments; the dry weight of the media] sections was
compared with their wet weight (obtained from their dimensions and measured
specific gravity). In some experiments. the internal diameter of excised arteries was
measured; the vessels were fixed for I h in Karnovsky's fixative either immediately
after excision or after incubation for various time intervals. In other experiments the
external diameter of excised arteries was measured with a travelling microscope,
inunediately after excision and at intervals dunng the 3-h incubation. The diameter
was measured half-way along the vessel and then measured again at 90' to the first
measurement. This approach had the advantages of providing information about
diameter changes in individual segments, allowing observations at frequent intervals
and preventing fixation artefacts.
Results
Series I
The mean medial CTICP value in segments incubated for 20 min with labelled
albumin immediately after excision was 0.036 SEM 0.005 (n = 10). The corre-
sponding value in segments first incubated for 3 h in unlabelled plasma was
0.045 SEM 0.003 (n = 10). The values are significantly different ( p < 0.05) when
examined by paired t-test.
Series 2
The mean medial CTICP Values for labelled albumin for control segments (no
vasoactive agent) and segments incubated with various concentrations of noradrena-
line or sodium nitrite, are shown in Figs. 1-3. The results for the controls are
consistent with those obtained in an earlier study. where mean medial labelled
C=)
LP4
L.n
-4@6
ON
C:)
U1
C711
BATCo document for Province of British Columbia 15 April 1999

140
oce
007 -
005 -
'D 0-05 -
C 00--
003 -
0 02@-
0 01-
I L I
0 (T. 1. I I'. I.- 10,
Mola; ,CT, v I la
12
2 3
010.
009 -
Oce -
008 -
0 07 - 0-ce -
007-
001-
3
53 0 05 10
r
0 04
Do, -
0 03 0 0:0
002 0 02 -
0 01 L I 0.01i I I I . L L I
lot. 0 'O ci 'II" -L. -.1 la
Molar Concentratson Molar Concentration
Figs. 1-3. Mean medial CTICp values SEM (6 < n < 9) for "relaxed- segments incubated with
vasoactive agents at different concentrations. 0, control; IIL sodium nitntc,. a. noradrenaline. Fig. 1: 20
min, Fig. 2: 90 min. Fi& 3: 180 min.
albumin concentration rose at first rapidly and then slowly 12). as in the previous
study, the mean 180-min value is slightly greater than the mean 90-min value,
although the difference is not significant. The values obtained at 20 min with either
vasoactive agent are not significantly different from the control value. However, with
a concentration of noradrenaline exceeding 10-9 K the values at both 90 and 180
min are lower than those of the controls while, with a concentration of sodium
BATCo document for Province of British Columbia 15 April 1999

141
0 51
04
0 3
C'Ac'
01-
C
21 .1 11 i5 160 @i() @O iO
NormaliZed Distance from Lurnen
Fig. 4. Average CTICP profiles of labelled albumin across the walls of'relaxed'segmicrits incubuicd %Lith
a, 10-'.%f sodium nitrite (Pt - 10): S. 10-5 M noradrenaline (n - 9). Bars indicaje SEM.
nitrite exceeding 10-4 M, the values are higher than those of the controls. The 90-
and 180-min values are not significantly different for segments incubated with either
of the vasoactive agents.
Figure 4 shows average CTICP profiles of labelled albumin across the walls of
segments exposed for 90 min to 10-5 M noradrenaline, or 10' M sodium nitrite.
Consistent with findings reported above, the medial values are markedly different
for the two agents. The adventitial values are. however, closely similar. A separate
studv was undertaken in this series to assess in addition the effect of isosorbide
dinitrate (ISDN). The mean medial 90-min CT/Cp values for labelled albumin were:
controls 0.059 SEM 0.004 (n - I 1); 10- 5 M noradrenalinc 0.031 SEM 0.004
(n - II); 10-3 M sodium nitrite 0.066 SEM 0.007 (n = I 1); 10' M ISDN 0.071
SEM 0.007 (n - I 1). The values obtained with sodium nitrite and ISDN are both
significantly greater than that obtained with noradrenalinc (P < 0.00 1). They are
not, however, significantly different from the control value.
Table I shows the values of the internal diameter of segments fixed either
CD
(_P4
C_n
_9@@ft
Un
co
BATCo document for Province of BritiSh Columbia 15 April 1999

142
TABLE I
MEAN INTERNAL DIAMETER OF "RELAXED" SEGMENTS FIXED IMMEDIATELY AH'ER
EXCISION OR AFTER INCUBATION
Results expressed as means (mm)-SD
Immediately No lo-, M lo-IM lo-Im
fter vawactive noradrena- sodium isosorhide
a er agent line nitrite dinitrate
90 mimmcuba"on
0.89 0.05 1.13 0.01 0.90 O.D3 114-0.01 1.28 0.03
/90 min incubanon
0.96 0.02 1.06-0.04 0.99 0.03 1. 17 + D.03 IA8 0.GS
immediately after excision or after incubation for 90 or 180 min. The given value in
each case is the average of 8 measurements: 2 segments were used in each stud\,.
each yielding 4 transverse sections. In both the 90- and 180-min studies the mean
diameter of the control segments exceeded that of the segments immediately after
excision. 10-5 M noradrenaline prevented such dilatation. while both 10-3 NI
sodium nitrite and 10-1 M ISDN increased the dilatation above that seen with the
controls. Figure 5 shows the mean external diameters of segments at different times
after excision. For segments incubated either with sodium nitrite i)r ISL)N the
diameter increases rapidly at first and then slowly. A similar time course is seen 'with
control segments, although the initial increase of diameter was less rapid. The
Ole-
E 017 -
E
Ole -
L
4)
X
W 015
014 -
0 10 60 '20 'lo
Time after excision (minutes)
Fig. 5. Mean external diameters + SEM (j? - 8 in all cases) of segments at different times after excision.
0. control; IL 10-4M NAN(),. 0. 10-5 M noradrenaline-. A, lo-'M isosorbide dinitraL9.
C
L
C)@
4,14 @
LrI C
.V, C
o%
Ln
BATCo document for Province of British Columbia 16 April 1999

CL
TABLE2
MEAN MEDIAL AND ADVENTITIAL 1"CrIEDTA CTICI VALUES FOR 'RELAXED' SEGMENTS INCUBATED FOR 90 MIN
No vasoactive agent 10 5 M noradrenaline 10 - 4 M sodium marile
Media Ady. Media Adv. Media Adv. E
Mean C. IC, SEM 0.345 0.009 0.452 0.012 0.335 0.014 0.466 0.014 0.338 0.012 0.461 0,017 Z
12 12 12
TABLE3
MEAN MEDIAL AND ADVENTITIAL C, ICp VALUES rOR LABELLED ALBUMIN rOR'AIR-IIRESSURIZED'ARTERIES
No vasoactive agent 10 M noradrenaline 10-' M sodium nitme
Media Adv. Media Adv. Media Adv.
40 min
Mean CT 1Cp SEM 0.029 0.004 0.24 0.03 0.022 0.004 0.22 0.03 0.040 O.OD4 0.20 0.02
n 1 7 6
90 min
Mcan C, 1Cp t SEM 0035 +0 001 0.21 0 02 0.027 j 0.003 0 26 0.03 0.051 0 008 0.22 0-04
PI 9 6 6
0
0

144
TABLE4
MEAN INTERNAL DIAMETERS (mm) (+SD) OF AIR-PRFSSUR1ZED' ARTE.RILS I'IXI..D
AFTER INCUBATION FOR 90 MIN
No vasmctsve I lo-3M Io - M
agent noradrenaline sodium ni(rite isosorbide din I [,.II
2.10 0.03 1.67-+0.06 2.34 0.09 2-24 - o.07
diameter of segments incubated with noradrenaline decreased within the first 30 min
and thereafter did not change.
The mean medial and adventitial CTICP values of [5 'CrIEDTA for segments
incubated for 90 min are shown in Table 2. Neither the medial nor the adventitial
values are significandy changed by the vasoactivc agents. 10-5 M noradrenaline or
10' M sodium nitrite.
Series 3
Table 3 shows the mean medial and adventitial CI IC, values of labelled albunlin
for air-pressurized arteries incubated for 40 or 90 min. The medial values obtained
at both times with 10-5 M noradrenaline are significantly lower than those or the
controls. and those obtained with 10' M sodium nitrite are significantly higher. As
with 'relaxed' segments, the vasoactive agents did not significantly alter the adventi-
tial values. Table 4 shows the mean internal diameters of 'air-pressurized' arteries
fixed after incubation for 90 min. The value is Si En reduced below that of
the controls by 10-5 M noradrenaline and significantly elevated above it bv both
10' M sodium nitrite and 10-@ M ISDN. Each artery in this experiment was cut
into three segments and several transverse sections were cut from each.
Discussion
The concentration ratio of labelled albumin in the media of the relaxed segments
rose rapidly at first and then slowly as in an earlier study [2]. This time course could
be due to the presence in the tissue of regions with different time constants. The
dilatation of the segments with time and greater uptake with incubation with tracer
3 h after excision, rather than immediately after excision. suggests, however. an
alternative explanation, a time-dependent change of the properties of the wall that
determine tracer uptake.
The tracers employed, radioactively labelled albumin and f"CrJEDTA. are rela-
tively inert and were expected to be confined to the extracelluiar space. It is
appropriate. therefore, to ask whether alteration of wall geometry. by change of
smooth muscle tone or of externally applied stress. influences tracer uptake by the
interstitium. The dilatation with time of 'relaxed' segments is thought to have been
due to a change of the passive mechanical properties of the wall or of smooth muscle
tone. We attempted to assess the dependence of uptake on wall geometry by
exposing vessels to vasoactive agents, or by altering externally applied stress.
C
BATCo document for Province of British Columbia 15 April 1999

145
Consideration is given first to the results of experiments with ' refixed' segments
Tracer could enter the wall through all surfaces and ricar equilibration could he
expected in the longer-term studies. the steady values obtained under these Con
tions were taken to represent the distribution volume for the tracer. The uptake of
labelled albumin in the media appeared to become steady by 90 min in the presence
of the higher concentrations of either noradrenaline or sodium nitrite. It wouldseem.
therefore. that the medial distribution volume for this tracer could be changed by a
factor of about 2.5 by changing smooth muscle tone. The diameter of these segments
also appeared to become steady at 90 min. when it was greater by a factor of about
20% with 10-4 M sodium nitrite than with 10-' M noradrenaline. It is reported that
10' M noradrcnaline, which was associated here with the lowest medial uptake
values for labelled albumin, maximally contracts the excised rabbit aorta. while
10' M sodium nitrite. which was associated here with the highest medial uptake
values for labelled albumin, maximally relaxes it [41,
In arteries pressurized with saturated air. interfacial forces would prevent inflow
of fluid to the wall. Results obtained in another study, indicate that air pressuriza-
tion of excised arteries can cause compaction of the wall 15]. Provided. therefore.
that convection associated with compaction was complete within the time course of
our experiments, tracer concentration in the wall could be expected to approach
equilibrium. The steady values of labelled albumin in both the media and adventitia
in control 'air-pressurized' arteries were significantly lower than those in control
relaxed' segments. Noradrenaline decreased and sodium nitrite increased the steady
medial uptake values of labelled albumin in 'air-pressurized' arteries as in 'relaxed'
segments.
The mechanisms which underlay the observed changes in the distribution volume
for labelled albumin in the artery wall are incompletely understood. but the probable
explanation is an alteration of the properties of the interstititim. The absence of a
significant change in the medial distribution volume for [ 5'Cr]EDTA with alteration
of smooth muscle tone is not surprising. because of the lesser exclusion of this
material than albumin f6j. The possibility must be considered that the labelled
albumin gained access to the cells of these excised arteries. but this would seem
unlikely since the distribution volume for the smaller material. ["CrJEDTA. was
always within the normal range [7]. Consistent, furthermore. with the applied
stresses having altered the properties of the interstitium. they induced changes in the
distribution volume for labelled albumin in the adventitia. which is largely acellular.
The lack of change by vasoactive agents of the adventitial distribution volume for
albumin could reflect the smaller dimensional changes that occurred. compared %, ith
those produced by pressurization of the vessels.
The mean water content of the media of control 'relaxed' segments was 0.79 +
SEM 0.005 (n - I 1). The comparable values with 10' M noradrcnahne and 10 -' M
sodium nitrite were 0.75 SEM 0.01 (n = 8) and 0.81 SEM 0.01 (P; = 6) respec-
tively. The water content of the constricted segments was significantly lower than
that of the dilated segments (P < 0.05), though neither value significantly differs
from that of the controls. These results are consistent with the changes observed in
the distribution volume for labelled albumin. if they reflect alterations of the water
content of the interstitium.
LW
(_n
46
OIN
BATCo document for Province of British Columbia 1 5 April 1999

146
Discussion in this paper is largely confined to the longer-term experiments.
because of the greater complexity of transient conditions. Uptake under transient
conditions would be affected by endothelial permeability, which is susceptible to
vasoactive agents [8), mechanical damage [9) and diffusion distances, which in turn
are influenced by smooth muscle tone and applied stress. Further complexity would
arise if changed smooth muscle tone or applied stress were to alter interstitial
properties and thus cause convection. An additional complication is the change with
time of the sensitivity of the smooth muscle of excised arteries to vasoactive agents
(4].
Our results suggest that smooth muscle tone or applied stress alter the distribu-
tion volume for labelled albumin in the arterial wall. However, it must be em-
phasized that the vessels were excised, at relaxed length and unpressurized. or at in
vivo length and pressurized with air. Although the geometry of the arteries ap-
proximated to normal in the latter studies. convection through the wall had been
obviated, leading. it is believed, to alteration of the properties of the interstitium
[101. The transport properties of the arterial wall in vivo could also depend on
smooth muscle tone and applied stress. Should that be so. the possibility arises that
these factor influence the composition of the interstitium and, hence, the environ-
ment of the smooth muscle cells.
Acknowledgentents
We are indebted for supplies of ISDN to Pharma Schwarz. Monheim. F.R.G.
References
I Bratzler. P.L.. Chisholm, G.M.. Colton, C.K.. Smith. K.A.. Zilversmit. D B. and Lees. R.S.. The
distribution of labelled albumin across the rabbit thoracic aorta in vivo. Circulat. Res., 40 (1977) 182.
2 Caro, C.G.. Lever. M.)., Laver-Rudich. Z.. Meyer. F.. Liron. N.. Ebel. W.. Parker, K.H. and Winlove.
C.P.. Net albumin transport across the wall of the rabbit common carotid artery perfused in situ.
Atherosclerosis. 37 (1980) 497.
3 Caro, C.G. and Lever. M.J.. Effect Of V2soactive agents on the alburr@n space of the arterial wall. J.
Physiol. (Lond.). 305 (1980) 112P.
4 Fufchgott, R.F. and Bhadrakorn. S_ Reactions of strips of rabbit aorta to epinephrine. isopropylar-
terenol. sodium nitrite and other drugs. J. Pharmacol. Fxp. Therap.. 109 (1953) 129.
5 Caro, C.G., Lever. MJ. and Tedgui. A.. 7be distribution volumes of albumin and sucrose in arteries
subjected to normal mechanical stresses, J. Physiol. (Lond ). 320 (1991) 391'.
6 Opton, A.G. and Phelps, C.F.. The partition of solutes between buffer solutions and solutions
containing hyaluronic acid. Biochern. J.. 78 (1961) 829.
7 Harrison. R-G. and Massaro. T.A., Extracellular space of swine aorta measured with "C inulin and
"'C sucrose, Amer. J. Phvsiol.. 231 (1976) 1806.
8 Robertson. A.L. and Khairallah. P.A., Effects of ansiotensin 11 and some analogues on vascular
permeability in the rabbit. Circulat. Res.. 31 (1972) 923.
9 Fry. D.L., Responses of the arterial wall to certain physical factors. la: R. Porter and J. Knight (Eds.).
Atherogenesis: Initiating Factors (Ciba Foundation Symposium No. 12. New Series). Elsevier.
Amsterdam, 1973, P. 93.
10 Lever. MJ. and Tedgui. A.. Oedema and albumin space in the rabbit aorta following intimal damage,
J. Physiol. (Lond.), 319 (1981) 37P.
CY
Cy%
BATCo document for Province of BritiSh Columbia 15 April 1999