© 1995 American Heart Association, Inc.
Articles |
Effects of Second Messengers on the Permeability and Morphology of Eel Rete Capillaries
From the Department of Medicine, Hôpital Notre-Dame, and the Department of Anatomy, Université de Montréal (E.A.R., M.B.), and the McGill University Medical Clinic and the Department of Medicine, Montreal General Hospital (C.A.G.), Montreal, Quebec, Canada.
Correspondence to Dr Carl A. Goresky, McGill University Medical Clinic, Rm C10.148, Montreal General Hospital, 1650, Cedar Ave, Montreal, Quebec, Canada, H3G 1A4.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Abstract The effects of second-messenger concentration changes on capillary diffusion capacity (permeability–surface area product [PS]) to cellular and paracellular tracers and on capillary ultrastructure were studied during countercurrent perfusion of the rete of the eel swim bladder. Cyclic nucleotide effects were investigated with isoproterenol, forskolin, and dibutyryl cAMP. Isoproterenol (5x10-6 mol/L) did not modify water and solute permeability or capillary structure. Forskolin (10-4 mol/L) immediately raised the concentrations of cAMP in the rete and produced interstitial edema but did not change permeability. The addition of dibutyryl cAMP (10-6 mol/L) to the perfusate had rapid effects: it reduced the PS of [3H]water and oxygen and increased the PS of [125I]albumin, [14C]sucrose, and 22Na. No structural changes were observed. Phosphoinositide effects were studied with 1,2-dioctanoyl-sn-glycerol (DG) and phorbol 12-myristate 13-acetate (PMA). DG (10-5 mol/L) had no effect on the permeability of the rete to water and solutes, while inducing cell membrane vacuolization. PMA (10-5 mol/L) progressively reduced the PS of [3H]water. In contrast, PS values of [125I]albumin, [14C]sucrose, and 22Na rose gradually. Membrane vacuoles bulging into the lumen and in the cytoplasm were a common feature. The Ca2+ effect was investigated with the Ca2+ ionophore A23187. At 5x10-6 mol/L, unsteady permeability changes and extensive cytolysis were observed. At 5x10-7 mol/L, the PS of [125I]albumin, [14C]sucrose, and 22Na rapidly increased. The PS values for water were not modified. No structural changes were identified. It is concluded that increments of second-messenger concentrations in the rete induce characteristic selective effects on the paracellular and transcellular pathways of transport and create significant but nonselective alterations of capillary structure.
Key Words: rete mirabile • capillary permeability • capillary ultrastructure • second messengers
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The intracellular second messengers, cyclic nucleotides,1 polyphosphoinositides,2 and Ca2+3 are ubiquitous substances used by cells to regulate their form and function in response to various stimuli. The microvascular endothelia constitute a barrier between the vascular and interstitial compartments. The question that arises is whether changes in the endothelial permeability of intact vessels, which occur in response to a variety of stimuli, are mediated by the action of second messengers on the shape and surface properties of endothelial cells.
Investigations using cultured endothelial cells have frequently linked the action of second messengers on monolayer permeability to changes in cytoskeletal structures and in integral membrane proteins affecting junctional integrity and the surface area available for transfer across the monolayers.4 5 6 7 8 9 It may not be warranted, however, to extrapolate from conclusions derived from in vitro studies to the in vivo endothelial barrier. The cytoskeletal arrangement and function in endothelial cells cultured in vitro is likely to be determined more by conditions associated with growth, locomotion, confluence, and anchoring to the extracellular matrix than by situations governing cell physiology in vivo, such as shear stress.10 11 At the same time, permeability values for endothelial cell monolayers are generally higher than those observed in vivo, by two orders of magnitude.12 The reason for the difference has not been completely elucidated, but discontinuities in the monolayer may be involved.
In vivo organ studies of the possible role of second messengers, whether released by agonists interacting with receptors or provided directly, in the regulation of capillary permeability are difficult to interpret because of the multiple interactions between endothelial cells, blood elements, and surrounding tissues, which confound the delineation of the mechanisms involved. Investigations with single microvessels tend to avoid these disparities. Using electrophysiological approaches, Olesen13 has shown that in venous capillaries of the frog brain, a selective increase in the cytosolic concentration of cAMP does not, per se, change ion permeability, whereas a rise in cytosolic Ca2+ level increases it. Similarly, Curry14 has demonstrated that the magnitude of Ca2+ influx into intact microvessels determines the increase in their permeability.
The rete mirabile of the eel swim bladder is a compact microvascular organ of arterial and venous capillaries that are in close contact by their continuous basement membrane and in which blood flows in opposite directions. The tissue is entirely made up of endothelial cells with no cell contaminants other than pericytes.15 The rete has proven valuable for the study of capillary permeability to a variety of solutes in normal15 16 17 18 19 and pathological conditions.20 21 22 As an initial step in the determination of whether second messengers act as regulators of microvascular permeability, we have examined the action of a direct infusion of agents aimed at cyclic nucleotide effects (isoproterenol, forskolin, and dibutyryl cAMP), phosphoinositide effects (phorbol 12-myristate 13-acetate [PMA] and 1,2-dioctanoyl-sn-glycerol [DG]), and intracellular Ca2+ effects (ionophore A23187). The various agents were used at the concentrations generally used for in vitro and in vivo studies of vascular endothelium. We have measured the arterial to venous capillary passage of various substances expected to traverse the endothelium via paracellular pathways (labeled albumin, sucrose, urea, and sodium) or both paracellular pathways and the whole of the endothelial surface (labeled water and oxygen). The effects of second-messenger activators or analogues on microvascular structure are rarely reported in conjunction with in vivo functional studies. With the use of the electron microscope, we searched for morphological correlates to the permeability changes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The procedures followed in the present study were approved by the institutional ethics committee on animal research.
Permeability Studies
Eels captured from the St Lawrence River during the summer and
adapted to ambient fresh water at 20°C to 25°C were anesthetized
with tricaine methane sulfate (1 g per gallon of water). One of the two
symmetric retia at the surface of the swim bladder was
countercurrent-perfused in situ through catheters inserted into the
arterial input at the cephalic pole and the venous input at the caudal
pole. The perfusion medium was a Krebs-Ringer bicarbonate buffer (pH
7.4) containing (mmol/L) Ca2+ 2.5,
Mg2+ 4.2, glucose 5, and bovine albumin (fraction V,
powder; 96% to 99% albumin, remainder globulins; Sigma Chemical Co)
0.58. At the arterial input, varying combinations of the following
radioactive tracers were added to the medium: human
[125I]albumin (Frosst; >95% of the labeled iodine was
precipitated with 10% trichloroacetic acid),
[U-14C]sucrose (1 to 5 mCi/mmol; New England Nuclear
[NEN]), 22Na (as sodium chloride, 99.9% radionucleidic
purity; NEN), [14C]urea (crystalline solid, 2 to 10
mCi/mmol; NEN), and tritium-labeled water (biological quality, 0.25
mCi/g; NEN). No tracers were added to the medium at the venous input.
To measure oxygen permeability, the medium delivered to the arterial
input was continuously equilibrated with a gas mixture of 95%
O2/5% CO2, providing a partial
pressure of oxygen of 
400 mm Hg. At the venous input, the medium
was equilibrated with a gas mixture of 95% N2/5%
CO2 to achieve a partial pressure of oxygen of 100
mm Hg.
All perfusions were carried out at ambient room temperature that ranged from 20°C to 25°C and were begun with constant and equal flow in arterial and venous directions, with a steady arterial pressure head of 45 mL H2O. Flow averaged 0.5 mL/min, and the average wet weight of the rete was 150 mg.
The experiments began with a control period of observation, during
which baseline measurements of permeability to labeled water and
solutes were carried out. The perfusion medium, at both arterial and
venous inputs, was then enriched with one of various substances at
concentrations expected to induce permeability effects through changes
in the intracellular concentration of second
messengers.1 2 3 13 17 The cyclic nucleotide effect was
studied by the use of a ß-receptor activator, isoproterenol
(hydrochloride; Sigma); an adenylate cyclase stimulator, forskolin
(7ß-acetoxy-8,13-epoxy-1
,6ß,9-trihydroxy-labd-14-ene-11-one;
Calbiochem); and a cAMP analogue, dibutyryl cAMP
(N6-2'-O-dibutyryladenosine-3':5'-cyclic
monophosphate-sodium salt). The phosphoinositide effect was
investigated with phosphokinase C activators, DG (Sigma) and PMA
(anhydrous; Sigma). The Ca2+ effect was induced with
a divalent cation ionophore, A23187 (Calbiochem). A total of five to
eight perfusions were performed for each substance to test its effects
on permeability. In addition, six retia were perfused with forskolin,
without tracers added to the medium. In these perfusions, to gain
insight into changes in overall cAMP economy, concentrations were
measured in medium emerging from arterial and venous capillaries and in
the rete tissue itself.
Media were collected simultaneously from arterial and venous outflows for 10-minute periods throughout the perfusion. The radioactivity of [125I]albumin was measured on a 10% trichloroacetic acid precipitate with a gamma spectrometer; the other radioactive tracers were measured in protein-free supernatant with a liquid scintillation spectrometer (Packard Instrument Co). Values were corrected for background and crossover. The partial pressure of oxygen was measured with an in-line oxygen tension measuring system with a dual oxygen electrode amplifier (model 203, Instech Laboratories). The concentrations of cAMP in the medium during perfusion with forskolin and in the rete at the end of the experiment were determined with a conventional radioimmunoassay kit (NEN). The permeability–surface area product (PS) values (in cubic centimeters per second) were calculated by the following equation: PS=F(Vo/Ao), where F is the flow (in cubic centimeters per second), and Vo and Ao are the concurrent concentrations at the venous and arterial outputs, respectively. The surface area available for capillary exchange was 1 cm2/mg wet wt.15 Oxygen permeability–surface area products values were calculated with the same equation, except the value of oxygen pressure at the venous input was subtracted from Vo and Ao.18 From the above equation, the permeability (P) is calculated as follows: P=(F/S)(Vo/Ao), where S is the surface area. Hence, the ratio values will reflect permeability values and can be used to monitor the pattern and rate of change of permeability. We have previously reported that in the absence of external agents added to the medium during countercurrent perfusion at 25°C, the ratio of Vo to Ao for each tracer remained steady for the 3-hour period over which it was tested.21
Morphological Studies
The rete perfused for permeability measurements and the
contralateral symmetric rete used as control were examined in each eel.
Small fragments of both the experimental and control retia were fixed
by immersion with 1% glutaraldehyde, postfixed with osmium tetroxide,
and processed for transmission electron microscopy as described
previously.15 Thin sections were performed and stained
before examination with an electron microscope (model ME410LS, Philips
Electronic Instruments). About 100 capillaries were examined for each
experimental protocol.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Permeability Studies
The Cyclic Nucleotide Effect
Effect of isoproterenol on the permeability of rete capillaries. The addition of isoproterenol to the perfusate, at a concentration of 5x10-6 mol/L, did not change the PS values of [125I]albumin, [14C]sucrose, 22Na, and [3H]water over 120 minutes (Table 1
).
|
Effect of forskolin on the production of cAMP by rete capillaries. The addition of forskolin (10-4 mol/L) to the perfusate induced an immediate release of cAMP by the venous and arterial capillaries, which reached plateau concentration values throughout the 60 minutes of the perfusion, at 45±18 and 85±20 nmol/L, respectively, from baseline undetectable levels (n=6). At the end of the perfusion with forskolin, the cAMP content of the whole rete was 6.9±2.1 versus 0.4±0.1 µmol/g DNA in the contralateral nonperfused rete removed at the beginning of the experiment (n=6).
Effect of forskolin on the permeability of rete capillaries.When forskolin was added to the medium at a concentration of
10-4 mol/L, there were no significant changes in the PS
values for [125I]albumin, [14C]urea, and
[3H]water over 120 minutes (Table 1
).
Effect of dibutyryl cAMP on the permeability of rete capillaries.Fig 1 shows that when dibutyryl cAMP was added to
the medium at a concentration of 10-6 mol/L, PS values for
[14C]sucrose and 22Na began to rise within 10
minutes and continued to do so until the end of the perfusion. The
pattern for [125I]albumin was similar, although changes
from control values became significant only after 60 minutes. In
contrast, PS values for [3H]water and oxygen were
significantly reduced from baseline values from the 30th minute onward.
On average, PS values for [125I]albumin,
[14C]sucrose, and 22Na doubled during
dibutyryl cAMP perfusion, with reference to baseline control values,
whereas PS values for [3H]water and oxygen were reduced
by 30% and 40% (Table 1).
|
The Phosphoinositide Effect
Effect of DG on the permeability of rete capillaries.DG, added to the medium at a concentration of 10-5
mol/L, did not significantly alter the permeability of the rete to
[125I]albumin, [14C]sucrose,
22Na, and [3H]water over 120 minutes (Table 2).
|
Effect of PMA on the Permeability of Rete Capillaries. Fig 2 shows that when PMA was added to the medium at a
concentration of 10-5 mol/L, there was a steady continuous
rise of PS values for [125I]albumin,
[14C]sucrose, and 22Na, which, on average,
reached twice the corresponding baseline values from 60 to 120 minutes.
For [3H]water, PS values declined 30% with respect to
control values (Table 2).
|
The Ca2+ Effect
Effect of A23187 on the permeability of rete capillaries.Fig 3 gives the results of perfusions for which the
divalent cation ionophore A23187 was added to the medium at a
concentration of 5x10-6 mol/L. This induced an immediate
and uneven drop of the arterial and venous outflows, from steady
control values in both directions averaging 0.5 mL/min to values
ranging between 0.2 and 0.5 mL/min. The outflows progressively and
spontaneously increased during the ensuing 60 minutes, to again reach
baseline control values. At this time,
Vo-to-Ao ratios were again assessed:
they showed that plateau values were achieved, with an increase in the
ratio for [125I]albumin, no change for
[14C]urea, and a decrease for [3H]water.
The quantitative effects of A23187 on PS values were as follows: those
for [125I]albumin were increased by 60%, whereas for
[3H]water, they were decreased by 40%, with respect to
corresponding average control values (Table 3).
|
|
Fig 4 gives the results of perfusions for which A23187
was added to the medium at a concentration of 5x10-7
mol/L. With this concentration, there were no changes in the outflows.
Vo-to-Ao ratios increased within 10
minutes for [125I]albumin, [14C]sucrose,
and 22Na and remained steady thereafter. Corresponding PS
values increased by 70%, 20%, and 20%. No significant effect was
detected on [3H]water permeability (Table 3).
|
Morphological Studies
The capillaries of the control tissues are shown in Fig 5 with their normal morphological features. The arterial
capillaries are continuous and lined by a high endothelium with a
well-developed vesicular-tubular system. On the other hand, the venous
capillaries are thin and fenestrated (Fig 5a). In both types of
capillaries, the endothelial cells are joined by tight junctions and
rest on well-defined basement membranes (Fig 5b) containing pericytes.
No other cell types such as mast cells or fibrocytes are present in
the tissue. The interstitial space is small and filled with bundles of
collagen fibers.
|
The Cyclic Nucleotide Effect
The addition to the perfusate of isoproterenol or dibutyryl cAMP,
at the concentrations tested, did not change the normal features of the
rete (data not shown). On the contrary, exposure to forskolin induced
interstitial edema with loosely arranged bundles of collagen fibers and
cell vacuolization (Fig 6).
|
The Phosphoinositide Effect
Perfusions with DG or PMA resulted in endothelial cell membrane
vacuolization. Interstitial edema was seen after PMA (Fig 7).
|
The Calcium Effect
Cytolysis was prominent with A23187 at 5x10-6 mol/L
(Fig 8a), whereas no structural damage was evidenced
with A23187 at 5x10-7 mol/L (Fig 8b).
|
In all the experiments with second messengers carried out in the present study, intercellular junctions remained tight.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The present studies were aimed at elucidating whether activation of the second-messenger system in the endothelial cells and pericytes of the rete mirabile would affect the permeability and structure characteristics of the capillaries. Although pericytes are contractile in vitro, their function in vivo is difficult to assess. Therefore, the effects reported in the present study may result, in part, from specific interactions between pericytes and endothelial cells.23 A summary of the observations is presented in Table 4
. A first finding is that significant
morphological changes are seen during activation of the second
messengers by some but not all agents. When the structure is altered,
common features are membrane vacuolization and interstitial edema,
while the interendothelial cell junctions remain tight. These
structural changes do not correlate with permeability changes. A second
finding is that for either the cAMP or the phosphoinositide pathways
(or Ca2+ activation, which may be their final common
effector when there is a permeability change), the effects are similar,
in general. There is an increase in the permeation of tracers expected
to pass along paracellular pathways, with a simultaneous decrease in
the permeation of tracers that use the whole cellular surface, such as
water and, when examined, oxygen.
|
The Cyclic Nucleotide Effect
The first set of experiments was directed at determining the
effect on rete capillary permeability of increments in the
intracellular concentration of the second messenger cAMP. Forskolin (a
direct activator of adenylate cyclase),24 dibutyryl cAMP
(a lipophilic cAMP analogue), and isoproterenol (a ß-adrenergic
stimulator) were used. Predictably, forskolin, when added to the
perfusate at a concentration of 10-4 mol/L, increases both
the rete cAMP content and the levels of cAMP in the perfusate leaving
the rete. The same concentration of forskolin was used to assess its
effects on permeability: it had none, at least as tested with the
radioactive tracers albumin, urea, and water. A similar lack of effect
was reported by Olesen25 : forskolin, at a concentration of
10-5 mol/L, did not change the electrical wall resistance
of frog venules in vivo. Adenylate cyclase activation with
isoproterenol at 5x10-6 mol/L also had no effects on the
rete capillary permeability as tested by additional tracers such as
[14C]sucrose and 22Na. Furthermore, although
forskolin induced interstitial edema, isoproterenol did not alter
capillary structure. The lack of a direct effect of either forskolin or
isoproterenol on capillary permeability does not preclude a possible
modulatory effect of cAMP on other intracellular processes; indeed, in
several cell types, the action of one second messenger has been found
to enhance or attenuate the effects of another second
messenger.1 Equally, cAMP concentrations may not have been
raised enough in our experiments.
When dibutyryl cAMP was added to the perfusion medium at a
concentration of 10-6 mol/L, direct effects on the rete
capillary permeability were observed, but no significant alteration of
endothelial cell structure could be detected. PS values for albumin,
sucrose, and sodium rose steadily, reaching twice baseline values by
the end of the experiment, whereas PS values for oxygen and water
decreased by 40%. The intracellular levels of dibutyryl cAMP
achieved by these perfusions were probably higher than those of cAMP
produced in response to forskolin or isoproterenol. In vitro, increased
albumin permeability following stimulation of cAMP production has
similarly been reported in confluent monolayers of microvascular
endothelial cells.26 In contrast, decreased passage rates
of albumin,26 peroxidase,27 and mannitol and
polyethylene glycol28 have been observed in monolayers of
endothelial cells isolated from macrovessels. In vivo, direct effects
of cAMP on permeability have been reported in the brain29
and retina,30 with a breakdown of the barrier to albumin.
On the other hand, cAMP analogues seem to attenuate microvessel
permeability increase in the presence of ATP31 and to
prevent ischemia/reperfusion injury in pulmonary
capillaries32 but not in skeletal muscle
capillaries.33 Therefore, it appears that the sensitivity
of the vascular endothelial paths of transport to cAMP may vary
considerably from one vascular endothelium to another, as well as with
the agent used to elicit the cyclic nucleotide effect.
The observation of a parallel decrease in the PS values for labeled water and oxygen indicates that oxygen could share a common pathway with labeled water across capillary endothelial cells and particularly across their cell membranes.18 34 Alternatively, oxygen could use a primarily lipid pathway35 of limited access to water,36 in which case the parallel changes in PS for water and oxygen would result from simultaneous effects of dibutyryl cAMP on two different pathways.
The Phosphoinositide Effect
Another important system of intracellular signaling is
represented by degradation products of phosphatidyl
inositol.2 37 To test their effect on the rete capillary
permeability, we used a phorbol ester, PMA, at a concentration of
10-5 mol/L. Although phorbol esters have multiple effects
on cellular metabolism,37 their main function is to
substitute for DG and to activate the kinase C pathway in intact cells.
PMA, in our perfusions, increased the permeability to albumin, sucrose,
and sodium: the effect was rapid, unabated throughout the experiment,
and quantitatively equivalent for the three tracers, as if they shared
the same path of diffusion. This may be represented by a
functionally widened interendothelial junction. However, since albumin
is thought to be primarily transported by plasmalemmal
vesicles,38 some of which may form transendothelial
channels, a separate effect of PMA on this system is also possible. In
contrast, labeled-water permeability was reduced: the effect was small
and significant only in the later times of the infusion. Thus, as with
dibutyryl cAMP, PMA must have acted on paths of water transport not
shared by albumin, sucrose, and sodium. These may be
represented by the entire cell membrane
surface34 including the fenestrations,39
which are especially associated with high water permeability. There are
very few data in the literature on the direct effects of PMA on
capillary permeability in vivo. In the pial venules of the perfused
frog brain, PMA at 10-6 mol/L did not modify ion
permeability.40 Another way of showing the possible
effects of exogenous activators of kinase C is to use DGs that
penetrate the plasma membrane and directly interact with kinase C. The
possible role of protein kinase C activation on capillary permeability
has been studied particularly in the lung, where both
PMA41 and DG42 can induce edema. In this
model, however, PMA is a potent stimulator of enzymes in
polymorphonuclear leukocytes and macrophages and is an aggregator of
platelets, which confound its direct action on pulmonary
capillaries.43 We used 10-5 mol/L DG and were
not able to detect any significant effect on the PS values of labeled
albumin, sucrose, sodium, and water. The discrepancy between the
results obtained with PMA and DG may be due to differences in magnitude
of the effector intracellular concentrations or to the fact that PMA
has effects other than that of activating protein kinase C, such as
lowering both the production of inositol triphosphate and the
subsequent endogenous release of cytosolic
Ca2+.37 It is of interest that both PMA
and DG had a significant effect on capillary cell membrane structure,
with the appearance of vacuoles protruding into the lumen and the
cytoplasm.
The Ca2+ Effect
There are indications that changes in
[Ca2+]i in endothelial cells may be
important signals for the regulation of permeability in
venules13 and single capillaries.14 To study
the possible role of increases in
[Ca2+]i on the rete capillary
permeability, we have used the divalent cation ionophore A23187 at
concentrations of 5x10-6 and 5x10-7 mol/L,
with CaCl2 (2.5 mmol/L) in the medium. Ionophore A23187
transports Ca2+ and Mg2+ ions
passively across lipid membranes into the cell, in exchange for
2H+, without depolarizing the cell, down a very
steep concentration gradient. In the experiments with the
countercurrent-perfused rete, we observed that after the addition of
5x10-6 mol/L A23187 to the medium, there were rapid,
variable, and asymmetric reductions in arterial and venous outflows.
This phenomenon could be explained, to a large extent, by cytolysis, as
also reported with tumor cells,44 and by the appearance of
cellular debris in the capillary lumina. With continuing perfusion of
the rete with A23187 and without modification of the hydrostatic
pressure at the inflows, bulk outflow progressively increased,
achieving the original steady state control values in all experiments
by the 60th minute after the addition of A23187. PS values during the
ensuing hour were stable and were found to have increased significantly
for [125I]albumin to 60% above baseline, to be unchanged
for [14C]urea, and to have decreased for
[3H]water to 40% below baseline. When the concentration
of A23187 was lowered by one order of magnitude to 5x10-7
mol/L, no ultrastructural damage was evidenced. PS values for
[125I]albumin, [14C]sucrose, and
22Na increased by 70%, 20%, and 20%, respectively; they
remained unchanged for [3H]water. Again, what is striking
in this series of experiments is a differential effect of the ionophore
A23187 on the two differing transcapillary paths of transport.
Selective effects on rete permeability have also been reported by
Stray-Pedersen,17 who used the chelating agent EDTA to
reduce [Ca2+] and [Mg2+]: the
capillary permeability to sucrose and potassium increased, whereas that
to [3H]water and lipid-soluble molecules remained
unchanged. The effects of changes in
[Ca2+]i on capillary permeability have
been attributed to contraction or relaxation of microfilaments in the
endothelial cells, which modify their shape and their attachment to one
another and consequently the paths of transport. This theory rests on
the results of observations with cultured endothelial cells, where gap
formation in monolayers has been linked to increased cytoplasmic
Ca2+, rearrangement of the actin filament
network, and phosphorylation of myosin light chain.7 45 46 47 48 49
However, there is a lack of evidence concerning the possibility that
endothelial cell contraction or relaxation takes place in vivo.
Our experiments with second-messenger substitutes dibutyryl cAMP, PMA, and A23187 indicate that for the members of this group of compounds that have an effect, the permeability to paracellular solutes is increased but that to transcellular probes is decreased. These observations are compatible with the hypothesis, originated from studies with epithelia, that the effects of second messengers occur via a common Ca2+-dependent final pathway, leading to activation of the actin-myosin filaments in the endothelial cells. Indeed, cell contraction could induce a reversible enlargement of the junctional structure, resulting in moderate increases of albumin, sucrose, and sodium passage. We have not detected any widening of the junctions, but it is entirely possible that detachment of contacts at junctional complexes is beyond the power of resolution of the electron microscope. If we assume that intercellular clefts provide the only transport route for water, then this mechanism would also increase PS for water when, in fact, the experiments show that it decreased; the dominant change must be occurring in alternate pathways. A concomitant change in cell shape could, for instance, have reduced the membrane surface, thereby curtailing, on the whole, the transcellular passage of water and oxygen. Alternatively, rearrangement of cytoskeletal elements may have reduced the number of vesicle channels or fenestrations available to these substances and, in some instances, may have manifested itself with membrane vacuolization.
|
Acknowledgments |
|---|
This study was supported by grants from the Medical Research Council of Canada, the Quebec Heart Foundation, and the Fast Foundation. Dr Bendayan is a scientist and Dr Goresky is a Career Investigator of the Medical Research Council of Canada. We thank M.-P. Dea for her skillful technical assistance.
|
Footnotes |
|---|
Reprint requests to Dr Eugenio A. Rasio, Hôpital Notre-Dame, Unité Métabolique Pavillon Mailloux, 8e étage 1560, rue Sherbrooke Est, Montréal, Québec, Canada, H2L 4M1.
Received August 5, 1994; accepted December 2, 1994.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. Sutherland EW. Studies on the mechanism of hormone action. Science. 1972;177:401-408.
2. Berridge MJ. Inositol triphosphate and diacylglycerol: two interacting second messengers. Annu Rev Biochem. 1987;56:159-193. [Medline] [Order article via Infotrieve]
3. Exton JH. Mechanism of action of calcium-mobilizing agonists: some variations on a young theme. FASEB J. 1988;2:2670-2676. [Abstract]
4. Crone C. Modulation of solute permeability in microvascular endothelium. Fed Proc. 1986;45:77-83. [Medline] [Order article via Infotrieve]
5.
Majno G, Shea SM, Leventhal M. Endothelial contraction
induced by histaminic-type mediators. J Cell Biol. 1969;42:647-672.
6. Morgan-Boyd R, Stewart JM, Vavrek RJ, Hassid A. Effects of bradykinin and angiotension II on intracellular Ca2+ dynamics in endothelial cells. Am J Physiol. 1987;253(Cell Physiol 22):C588-C598.
7.
Rotrosen D, Gallin JI. Histaminic type 1 receptor occupancy
increases endothelial cytosolic calcium, reduces F-actin, and promotes
albumin diffusion across cultured endothelial monolayers. J Cell
Biol. 1986;103:2379-2387.
8.
Shasby DM, Shasby SS, Sullivan JM, Peach MJ. Role of
endothelial cell cytoskeleton in control of endothelial permeability.
Circ Res. 1982;51:657-661.
9. Shasby DM, Shasby SS, Welsh MJ. Reversible changes in pulmonary endothelial permeability: role of cell calcium. Fed Proc. 1984;43:1331.
10. Mo M, Eskin SG, Schilling WP. Flow-induced changes in Ca2+ signaling of vascular endothelial cells: effect of shear stress and ATP. Am J Physiol. 1991;260:(Heart Circ Physiol 29):H1698-H1707.
11. Dull RO, Davies PF. Flow modulation of agonist (ATP)-response (Ca2+) coupling in vascular endothelial cells. Am J Physiol. 1991;261:(Heart Circ Physiol 30):H149-H154.
12.
Albelda SM, Sampson PM, Haselton FR, McNiff JM, Mueller SN,
Williams SK, Fishman AP, Levine EM. Permeability characteristics of
cultured endothelial monolayers. J Appl Physiol. 1988;64:308-322.
13. Olesen S-P. An electrophysiological study of microvascular permeability and its modulation by chemical mediators. Acta Physiol Scand. 1989;136(suppl 579):1-28.
14. Curry FE. Modulation of venular microvessel permeability by calcium influx into endothelial cells. FASEB J. 1992;6:2456-2466. [Abstract]
15.
Rasio EA, Bendayan M, Goresky CA. Diffusion permeability of an
isolated rete mirabile. Circ Res. 1977;41:791-798.
16. Stray-Pedersen S, Steen JB. The capillary permeability of the rete mirabile of the eel (Anguilla vulgaris L). Acta Physiol Scand. 1975;94:401-422. [Medline] [Order article via Infotrieve]
17. Stray-Pedersen S. The effect of Ca2+, Mg2+ and H+ on the capillary permeability of the rete mirabile of the eel (Anguilla vulgaris L). Acta Physiol Scand. 1975;94:423-441. [Medline] [Order article via Infotrieve]
18.
Rasio EA, Goresky CA. Capillary limitation of oxygen
distribution in the isolated rete mirabile of the eel (Anguilla
anguilla). Circ Res. 1979;44:498-503.
19. Rasio EA, Goresky CA. Passage of ions and dextran molecules across the rete mirabile of the eel. Circ Res. 1985;56:74-83.
20.
Rasio EA, Bendayan M, Goresky CA. The effect of
hyperosmolality on the permeability and structure of the capillaries of
the isolated rete mirabile of the eel. Circ Res. 1981;49:661-676.
21.
Rasio EA, Bendayan M, Goresky CA. Effect of reduced energy
metabolism and reperfusion on the permeability of the capillaries of an
isolated rete mirabile. Circ Res. 1989;64:243-254.
22. Bendayan M, Rasio EA. Hyperglycemia and microangiopathy in the eel. Diabetes. 1981;30:317-325. [Abstract]
23. Shepro D, Morel NML. Pericyte physiology. FASEB J. 1993;7:1031-1038. [Abstract]
24.
Seamon KB, Padget W, Daly JW. Forskolin: unique diterpene
activator of adenylate cyclase in membranes and in intact cells.
Proc Natl Acad Sci U S A. 1981;78:3363-3367.
25. Olesen S-P. Regulation of ion permeability in frog brain venules: significance of calcium, cyclic nucleotides and protein kinase C. J Physiol (Lond). 1987;387:59-68.
26. Watanabe H, Kuhne W, Schwartz P, Piper HM. A2-Adenosine receptor stimulation increases macromolecule permeability of coronary endothelial cells. Am J Physiol. 1992;262(Heart Circ Physiol 31):H1174-H1181.
27. Langeler EG, van Hinsbergh VW. Norepinephrine and iloprost improve barrier function of human endothelial cell monolayers: role of cAMP. Am J Physiol. 1991;260(Cell Physiol 29):C1052-C1059.
28.
Haselton FR, Mueller SN, Howell RE, Levine EM, Fishman AP.
Chromatographic demonstration of reversible changes in endothelial
permeability. J Appl Physiol. 1989;67:2032-2048.
29. Joó F. Minireview: regulation by second messengers of permeability in the cerebral microvessels. Neurobiology. 1993;1:3-10. [Medline] [Order article via Infotrieve]
30.
Sen HA, Campochiaro PA. Stimulation of cyclic adenosine
monophosphate accumulation causes breakdown of the blood-retinal
barrier. Invest Ophthalmol Vis Sci. 1991;32:2006-2010.
31. He P, Curry FE. Differential actions of cAMP on endothelial [Ca2+]i and permeability in microvessels exposed to ATP. Am J Physiol. 1993;265(Heart Circ Physiol 34):H1019-H1023.
32.
Adkins WK, Barnard JW, May S, Seibert AF, Haynes J, Taylor AE.
Compounds that increase cAMP prevent ischemia-reperfusion pulmonary
capillary injury. J Appl Physiol. 1992;72:492-497.
33. Korthuis RJ, Grisham MB, Zimmerman BJ, Granger DN, Taylor AE. Vascular injury in dogs during ischemia-reperfusion: improvement with ATP-MgCl2 pretreatment. Am J Physiol. 1988;254(Heart Circ Physiol 23):H702-H708.
34. Crone C, Levitt DG. Capillary permeability to small solutes. In: Renkin EM, Michel CC, eds. Handbook of Physiology, Section 2: The Cardiovascular System, Volume IV, Microcirculation. Bethesda, Md: American Physiological Society; 1984:411-466.
35. Pappenheimer JR, Renkin EM, Borrero M. Filtration, diffusion and molecular sieving through capillary membranes: a contribution to the pore theory of capillary permeability. Am J Physiol. 1951;167:13-46.
36. Diana JK. Transcapillary water flux. Physiologist. 1982;25:365-375. [Medline] [Order article via Infotrieve]
37. Abdel-Latif AA. Calcium mobilizing receptors, polyphosphoinositides, and the generation of second messengers. Pharmacol Rev. 1986;38:227-272. [Medline] [Order article via Infotrieve]
38. Predescu D, Palade GE. Plasmalemmal vesicles represent the large pore system of continuous microvascular endothelium. Am J Physiol. 1993;265(Heart Circ Physiol 34):H725-H733.
39. Granger DN, Taylor AE. Permselectivity of intestinal capillaries. Physiologist. 1980;23:47-52.
40. Olesen S-P, Crone C. Substances that rapidly augment ionic conductance of endothelium in cerebral venules. Acta Physiol Scand. 1986;127:233-241. [Medline] [Order article via Infotrieve]
41. Struhar D, Harbeck RJ. Inhibition of induced acute lung oedema by a novel protein C inhibitor. FASEB J. 1987;1:116-118.[Abstract]
42. Johnson A, Hocking DC, Ferro TJ. Mechanisms of pulmonary edema induced by a diacylglycerol second messenger. Am J Physiol. 1990;258(Heart Circ Physiol 27):H85-H91.
43. O'Flaherty JT, Cousart S, Lineberger AJ, Bond E, Bass DA, Dechatelet LR, Leake ES, McCall CE. Phorbol-myristate-acetate: in vivo effects upon neutrophils, platelets, and lung. Am J Pathol. 1980;101:79-92. [Abstract]
44. Duke RC, Witter RZ, Nash PB, Young JD-E, Ojcius DM. Cytolysis mediated by ionophores and pore-forming agents: role of intracellular calcium in apoptosis. FASEB J. 1994;8:237-246. [Abstract]
45. Alexander JS. Modulation of Endothelial Monolayer Barrier Function and Cytoskeleton by Second Messenger Analogs In Vitro. Boston, Mass: Boston University; 1989. Doctoral dissertation.
46.
Wysolmerski RB, Lagunoff D. Involvement of myosin light-chain
kinase in endothelial cell retraction. Proc Natl Acad Sci
U S A. 1990;87:16-20.
47. Schnittler H-J, Wilke A, Gress T, Suttorp N, Drenckhahn D. Role of actin and myosin in the control of vascular permeability in pig, rat and human vascular endothelium. J Physiol (Lond). 1990;431:379-401.
48. Northover AM, Northover BJ. Changes of vascular endothelial cell shape and of membrane potential in response to the ionophore A23187. Int J Microcirc Clin Exp. 1987;6:137-148. [Medline] [Order article via Infotrieve]
49. Gudgeon JR, Martin W. Modulation of arterial endothelial permeability: studies on an in vitro model. Br J Pharmacol. 1989;98:1267-1274.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  M Bendayan and E. Rasio Transport of insulin and albumin by the microvascular endothelium of the rete mirabile J. Cell Sci., January 7, 1996; 109(7): 1857 - 1864. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||