| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Biomolecular Transport Dynamics Laboratory, Department of Chemical Engineering and the Bioengineering Program, The Pennsylvania State University, University Park, Pennsylvania 16802-4400
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Significant changes in transvascular pressure occur in pulmonary hypertension, microgravity, and many other physiological and pathophysiological circumstances. Using bovine aortic endothelial cells grown on porous, rigid supports, we demonstrate that step changes in transmural pressure of 10, 20, and 30 cmH2O induce significant elevations in endothelial hydraulic conductivity (Lp) that require 5 h to reach new steady-state levels. The increases in Lp can be reversed by addition of a stable cAMP analog (dibutyryl cAMP), and the increases in Lp in response to pressure can be inhibited significantly with nitric oxide synthase inhibitors (NG-monomethyl-L-arginine and nitro-L-arginine methyl ester). The increase in Lp was not due to pressure-induced stretch because the endothelial cell (EC) support was rigid. It is unlikely that the increase in Lp was due to a direct effect of pressure because exposure of the cells to elevated pressure (25 cmH2O) for 4 h had no effect on the volume flux driven by a transmural pressure of 10 cmH2O. We hypothesize that elevated endothelial cleft shear stress induced by elevated transmural flow in response to elevated pressure stimulates the increase in Lp through a nitric oxide-cAMP-dependent mechanism. This is consistent with recent studies of the effects of shear stress on the luminal surface of ECs. We provide simple estimates of endothelial cleft shear stress, which suggest magnitudes comparable to those imposed by blood flow on the luminal surface of ECs.
endothelial cells; transmural pressure; nitric oxide; adenosine 3',5'-cyclic monophosphate
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
SIGNIFICANT CHANGES in transvascular pressure occur in a variety of physiological and pathophysiological circumstances. High pulmonary pressures induce a variety of forms of pulmonary edema that may have pathophysiological consequences (18). When an astronaut is subjected to microgravity, transmural pressure is reduced in the lower extremities [below the hydrostatic indifference level (HIL)] and increased in the head (above the HIL) (11). This is in part responsible for the cephalad fluid shift observed in astronauts that manifests itself as a reduction in leg volume and a characteristic facial swelling (17).
Transvascular volume flux is expected to increase in proportion to increased transvascular pressure through a classic Starling mechanism in which the hydraulic conductivity (Lp; filtration coefficient) is assumed to be a passive (constant) property of the transport barrier. However, the endothelium, a major component of the transport barrier, is known to respond actively to changes in its mechanical environment (7). Recent studies have shown that the Lp of endothelial monolayers increases in response to increases in fluid wall shear stress on the luminal surface through a cAMP-nitric oxide (NO)-dependent mechanism (5, 20). Parker and Ivey (18) demonstrated that moderate increases in pulmonary venous pressure led to significant increases in capillary filtration coefficient through an endothelial effect that could be inhibited by isoproterenol, a cAMP agonist. The precise role of mechanical forces in this endothelial effect, however, was not determined.
An increase in transvascular pressure can affect the endothelium mechanically in three ways: 1) it can compress the endothelial cells against their substrate; 2) it can induce transverse stretch in the substrate and, in turn, in the endothelial cells; and 3) it can drive flow through the interendothelial junctions that impose fluid wall shear stress on the cell junctions. The present study explores the hypothesis that the above third point is an important mechanism by which endothelial Lp is increased in response to increases in transmural pressure. We use bovine aortic endothelial cell (BAEC) monolayers grown on porous filters with rigid supports to demonstrate increases in Lp in response to increases in pressure that can be reversed by a stable cAMP analog [dibutyryl cAMP (DBcAMP)] and inhibited by nitric oxide synthase (NOS) inhibitors.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Chemicals
The following chemicals were obtained from Sigma Chemical (St. Louis, MO): BSA (30% solution, fraction V), powdered MEM (without L-glutamine and sodium bicarbonate), trypsin, glutamic acid, sodium bicarbonate, DBcAMP, fetal bovine serum (FBS), gelatin (cell culture grade from porcine skin, lyophilized and gamma irradiated), fibronectin, and acetic acid (glacial). NOS inhibitors, NG-monomethyl-L-arginine (L-NMMA) and nitro-L-arginine methyl ester (L-NAME), were obtained from Calbiochem-Novabiochem (La Jolla, CA). Polycarbonate filters (Transwell chambers, 0.4-µm-pore size, 24.5-mm diameter) were obtained from Costar (Cambridge, MA). Collagenase-dispase was obtained from Boehringer-Mannheim Biochemicals (Indianapolis, IN). 1,1'-Dioctadecyl-3,3,3',3'-tetramethylindocyanide perchlorate-acetylated low-density lipoprotein was obtained from Biomedical Technologies (Stoughton, MA).Cell Culture
Primary BAECs were harvested from bovine thoracic aortas and subsequently maintained in MEM-10% FBS (MEM-10) as described previously (19). Cells were plated at a density of 2.5 × 105 cells/cm2 in MEM-10 onto pretreated polycarbonate filters. The filters were first placed in 0.5% acetic acid solution maintained at 50°C for 20 min. They were then sterilized overnight under an ultraviolet light. One hour before plating of cells, each filter was incubated at 37°C with 1 ml of 30 µg/ml fibronectin solution. The fibronectin solution was aspirated off just before plating. Experiments were carried out on the seeded filters cultured for 7-10 days with MEM-10. Cells between passages 6 and 12 were used in experiments.Measurement of Water Flux
A detailed description of the experimental apparatus used to measure water flux is presented in Sill et al. (20). Briefly, the entire apparatus was housed within a Plexiglas box and kept at an ambient air temperature of 37°C. A polycarbonate membrane Transwell filter containing the endothelial monolayer was sealed between two pieces of polycarbonate assembly separating the luminal and abluminal compartments. The compartments were continuously provided with positive pressure outgassing of 5% CO2-95% balance air to maintain the pH of the medium. The abluminal chamber was connected to a reservoir via tygon tubing and borosilicate glass tubing. The reservoir could be lowered to a desired height to create the hydrostatic pressure gradient required to drive water flux across the cell monolayer. To eliminate the oncotic pressure gradient, the same medium (MEM-1% BSA) was added to both the luminal and the abluminal compartments. In most experiments, the filter was supported by a glass-fiber prefilter (APFD 40, Millipore) placed on top of a stiff wire gauze. This eliminated strain due to membrane deformation on application of pressure drop. The monolayer was placed slowly into a Teflon support ring and rubber gasket to prevent any hydrostatic pressure gradient during setup. The compression ring was then tightened slowly to adjoin the upper and lower assemblies. After the luminal and abluminal surfaces were sealed, the system was allowed to equilibrate for 1 h without any applied hydrostatic pressure differential. The rotating-disk system used to impose shear stress on the surface of endothelial monolayers (see Ref. 19) was not incorporated into the present experiment.After the equilibration period, the Tygon tubing leading from the abluminal chamber was clamped with a hemostat, and a small air bubble was inserted into the borosilicate tubing. The clamp was then removed, and, after a brief equilibration period (5 min), the abluminal reservoir was lowered slowly (~1 min to apply a pressure of 10 cmH2O) with respect to the monolayer to produce a desired hydrostatic pressure gradient. The applied pressure drop induced fluid flow across the endothelial monolayer.
To measure fluid flux across the monolayer, the motion of the air bubble in the borosilicate glass tubing was tracked with a spectrophotometer mounted on a screw rod that was driven by a stepper motor. This traveling spectrophotometer was interfaced to a computer, and the bubble position was displayed as a function of time on the computer screen. The bubble displacement was then converted to fluid volume flux by the following formula
|
(1) |
d/t)
is the bubble displacement per unit time,
A is the surface area of the
monolayer, and F represents the volume of fluid contained in a known
length of tubing. Because of the balanced protein concentrations on
either side of the monolayer, it was assumed that there was negligible oncotic pressure differential across the monolayer and that
Lp could be
calculated by the following equation
|
(2) |
P is the hydrostatic pressure differential across the monolayer.
There may have been a slight excess of protein near the luminal surface
because of concentration polarization driven by the volume flux, but
this was expected to have a negligible oncotic effect at a protein
concentration of 1% (13).
The resistance of the polycarbonate filter was negligible compared with
the endothelial monolayer. The water flux through the polycarbonate
filter without endothelial cells was at least three orders of magnitude
higher than with cells (8). The filter-support system used consisted of
a glass-fiber prefilter support placed on top of a stiff wire gauze.
The wire gauze provided a stiff surface that did not deform with
pressure. The glass-fiber prefilter provided a smooth, even surface on
which the endothelial monolayer filter could rest. Because the
prefilter also had high filtration rates (10 cm3/s at 10 cmH2O), it did not affect the
volumetric flow through the endothelial layer
(10
5
cm3/s at 10 cmH2O).
Experimental Protocols
Response of volume flux to a step increase in pressure.
After a 1-h equilibration period in MEM-1% BSA in the absence of a
pressure gradient, a 10-cmH2O
pressure head was imposed on the monolayer over a period of 60 s, and
volume flux was then measured for a 1-h period at that pressure to
establish a baseline. Experience has shown that confluent monolayers
have baseline volume flux below 5 × 106 cm/s (20). Data from
monolayers with baseline fluxes above 5 × 106 cm/s were discarded.
Response of volume flux to NOS inhibitors. After the luminal and abluminal compartments of the endothelial monolayer were sealed, either L-NMMA or L-NAME was added to the luminal media to make the effective concentration 100 µM. This dose was chosen because it substantially inhibited increases in volume flux induced by fluid shear stress on the surface of BAEC monolayers in another study (5). A 1-h equilibrium period then allowed the cells to be incubated with the inhibitor. After the equilibration period, the 10-cmH2O baseline flux was measured for 1 h, and the response to a step increase in pressure with the inhibitor present was then measured for an additional 5 h. At the end of this period, 1 mM DBcAMP was added to test for endothelial monolayer integrity.
Calculations of membrane area strain. On application of the 10-cmH2O pressure gradient to an unsupported filter, the bubble in the capillary tubing was very rapidly displaced 5-6 cm as a result of the filter deformation. The displacement was even greater at higher pressure gradients. Because the strain induced by this deformation might have affected the endothelial cells plated on the filter, the filter-support system previously described was introduced. To calculate the strain in the membrane induced by the pressure gradient, it was assumed that the deformed membrane formed the surface of a sphere. The peak deflection of the deformed filter (a spherical cap) from the undeformed filter (of radius a) was denoted, h. By equating the volume of the spherical cap to the volume of fluid displaced in the capillary tube, and invoking basic analytic geometry formulas, one can show that
|
(3) |
) in the deformed spherical cap is then
given by a formula derived by Winston et al. (26)
|
(4) |
Data Presentation and Statistical Analysis
All volumetric flux data are presented as percent deviation from the baseline value 55 min after application of a 10-cmH2O pressure head
|
(5) |
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The average area strain in the membrane supporting the endothelial cells was measured for filters with and without a filter support at three pressure levels (10, 20, and 30 cmH2O). Even at the highest experimental pressure (30 cmH2O), the average area strain in the unsupported filter was very small (~0.42%). The filter support did, however, greatly reduce the average area strain to ~0.04% at 30 cmH2O.
To determine whether this small area strain would affect water flux
across endothelial monolayers, preliminary experiments were conducted
at the highest pressure gradient (30 cmH2O) on filters with and without
the filter support. The results displayed in Fig.
1 show that there was no significant
difference in the response of supported and unsupported filters. These
data were subsequently averaged together for further analysis (see Fig. 6). Apparently, the membrane strain is small enough to have a negligible influence on transendothelial water flux. All of the remaining data (Figs.
2-6)
are based on experiments with supported filters.
|
|
|
|
|
|
A typical water flux response for an experiment in which the pressure
gradient was stepped from 0 to 10 cmH2O at time
0 and from 10 to 20 cmH2O at 60 min is shown in Fig.
2. DBcAMP was added at 360 min. There was a characteristic decrease in
water flux over the first 60 min after the increase in pressure from 0 to 10 cmH2O that has been referred
to in the literature as the "sealing effect" (20). There was an
immediate jump in water flux at 60 min as the pressure was stepped up
to 20 cmH2O. This was followed by
another sealing effect between 60 and 90 min and then a significant transient increase in water flux toward a new steady-state level approached at 360 min. As a point of reference, it is important to
realize that, from 60 min onward, the water flux is proportional to
Lp because the
pressure differential (P) is constant. This means that
Lp is undergoing
large transient variations over a period of 5 h. If
Lp had been
constant at its baseline (55-min) value, then the response would have
been a constant 100% deviation from the baseline flux.
It is also important to note that in Fig. 2 the addition of DBcAMP at 360 min led to a strong reversal of the water flux increase within 30 min. This reversibility means that the large increase in volume flux (Lp) in response to the step change in pressure was not due to the loss of cells from the monolayer by mechanical damage but rather to an alteration of the intrinsic Lp of the endothelial monolayer. That cells were not denuded from the monolayer in response to the pressure change was also confirmed by observation of stained monolayers under the light microscope (×50). All of the experiments reported in subsequent figures incorporated DBcAMP at 360 min and were characterized by strong reversibility of water flux within 30 min, but the reversibility response is not shown in subsequent figures.
Figure 3 displays the control data for monolayers experiencing a 10-cmH2O pressure step change at time 0, with the data for monolayers experiencing a subsequent step change in pressure from 10 to 20 cmH2O at 60 min. The sealing effect is apparent in the control data, and the baseline value is selected at 55 min. There is a slight drift in the control flux over the next 5 h. The 10-20 cmH2O data are characterized by an immediate response to the step change in pressure, followed by a 30-min sealing period, and then a significant transient increase to a new steady-state flux 5 h after the step change. The new steady flux is 250% higher than the 10-cmH2O baseline flux, and this corresponds to an increase in Lp of 75%.
Figures 4 and 5 compare the responses of endothelial monolayers to step changes in pressure from 10 to 20 cmH2O with and without the NOS inhibitor L-NMMA (Fig. 4) and L-NAME (Fig. 5). The control data (no drug) are the same in both figures. The NOS inhibitors had no statistically significant influence on the sealing effect but did provide a statistically significant attenuation of the water flux at longer times. For the control experiments, the flux increased ~250% at t = 360 min, whereas, in the presence of L-NMMA (L-NAME), the fluxed increased by only 110% (130%) at t = 360 min. These attenuated increases in flux at 360 min in the presence of NOS inhibitors correspond to increases in Lp of only 5 (L-NMMA) and 15% (L-NAME).
The responses of endothelial monolayers to step changes in pressure from 10 to 30 cmH2O in the presence and absence of the NOS inhibitor L-NMMA are compared in Fig. 6. Again, there is a sealing effect after the step change in pressure that is not affected by the NOS inhibitor. This is followed by a significant transient increase in water flux out to t = 360 min, where the maximum response is 420% higher than the baseline value (10 cmH2O) without the NOS inhibitor and 200% higher with the NOS inhibitor. These increases in water flux at t = 360 min correspond to increases in Lp of 73 (no drug) and 0% (drug).
Volume flux was measured in paired experiments in which one endothelial
cell monolayer was held at an elevated hydrostatic pressure (25 cmH2O), whereas the companion
monolayer was held at a low hydrostatic pressure (0.5 cmH2O). After 4 h, the volume flux
driven by a 10-cmH2O pressure
differential was determined. For the pressurized monolayers, volume
flux was 3.36 ± 0.39 × 106 cm/s
(n = 4), whereas for the unpressurized
monolayers volume flux was 3.51 ± 0.30 × 106 cm/s
(n = 4). The mean values of
volume flux for the pressurized and unpressurized monolayers were not
significantly different (P > 0.75).
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
We have used an in vitro model of the endothelial transport barrier to
examine the effects of step changes in pressure on the
Lp of the
endothelium. The baseline values of
Lp for the in vitro system were in the range 3.1-4.4 × 107
cm · s1 · cmH2O1.
This is the same magnitude reported by Dull et al. (8), Sill et al.
(20), and Chang (5) using the same model (BAECs on polycarbonate
filters with 0.40-µm pores), and to our knowledge these are the
lowest values that have been reported in vitro.
After a step change in pressure from 0 to 10 cmH2O, from 10 to 20 cmH2O, or from 10 to 30 cmH2O, there was a 30- to 45-min period during which Lp decreased to a minimum value (Figs. 1-6). This phenomenon has been termed the sealing effect (21) and has been observed by others using cells in culture (5, 20, 21, 24) and intact arteries (13, 22). Earlier studies of the sealing effect showed that it was not attenuated in monolayers fixed with glutaraldehyde, suggesting that the effect is purely physical in origin, not involving biological activity (20, 24). Our observations that the NOS inhibitors L-NMMA and L-NAME attenuated the long-time increase in Lp but had no significant influence on the sealing phenomenon (Figs. 4-6) are consistent with the hypothesis that sealing is a purely physical process.
To our knowledge, this was the first study using cultured endothelial monolayers to demonstrate an increase in Lp in response to a step increase in pressure over an extended time period (5 h). In our previous study of the effect on volume flux of fluid shear stress on the luminal surface of BAEC monolayers by using a rotating-disk system, there was no detectable pressure differential across the monolayer (20). By comparison, when Lp has been measured in capillaries in vivo by using the modified Landis technique (16), step changes in pressure, which induce changes in transmural flow, have been monitored for time periods that were typically <1 min. Thus the phenomena we have described, which play out over a period of hours, would not have been observed in acute capillary experiments.
When intact arteries cannulated ex vivo were subjected to step changes in pressure, one study showed no effect of pressure (in the range 50-150 mmHg) on Lp measured for 30 min after the step change (2), whereas another study showed a significant decrease in Lp with increasing pressure (in the range 70-180 mmHg) when measurements were carried out for 90 min after the step change (22). These observations in whole arteries do not contradict our data in cultured endothelial cells. Although the total pressure drops were much higher in the artery studies, much of the pressure drop across the wall in an artery is supported by the structures underlying the endothelium. Estimates in the study of Tedgui and Lever (22) suggest that, in the rabbit thoracic aorta at 70 mmHg, the pressure drop across the endothelium is only 18 mmHg, which is in the range of our studies. Our data show that 90 min after a step change in pressure (time used by Tedgui and Lever), the Lp at the elevated pressure is actually lower than at the initial pressure (Figs. 3 and 6). At times >120 min after the 10- to 20-cmH2O step change (Fig. 3) and at times greater than 150 min after the 10- to 30-cmH2O step change (Fig. 6), Lp at the elevated pressure is higher than at the initial pressure. The experiments of Baldwin et al. (2), which did not show a significant effect of pressure on Lp, are more difficult to interpret because they introduced NaNO3 into their arterial bathing solutions to relax arterial smooth muscle cells. The nitrate serves as a NO donor, and in view of NO's tendency to increase Lp that was demonstrated indirectly in the present study and a previous one (5), it is possible that the endothelium contributed little to the overall resistance to water flux in the study by Baldwin et al. (2) and that the pressure independence they observed was characteristic of the underlying structures of the artery wall.
The demonstration that DBcAMP can rapidly reverse the increase in
Lp induced by a
step increase in pressure (Fig. 2) is consistent with previous studies
using BAEC monolayers that showed rapid reversal of increases in
Lp induced by
fluid shear stresses on monolayer surfaces (5, 20). Similar effects of
DBcAMP in reversing agonist-induced increases in endothelial
permeability have been demonstrated in a variety of endothelial cell
types (reviewed in Ref. 20). Parker and Ivey (18) showed that
isoproterenol, a 
-adrenergic-receptor agonist that stimulates cAMP
production, attenuated the high-vascular-pressure-induced
Lp increases in isolated rat lungs when venous pressure was increased to 30 cmH2O. This suggests that the
results of our study may shed light on the mechanism of edema formation
in pulmonary hypertension.
The substantial attenuation of increases in Lp (associated with step changes in pressure) by NOS inhibitors is another new finding of the present study. It is consistent with the observations of Chang (5), who was able to inhibit increases in BAEC Lp in response to fluid shear stress on the luminal surface by preincubating the monolayers with L-NMMA and L-NAME at the same doses used in the present study. Many other studies have demonstrated attenuation of agonist-induced increases in endothelial permeability by using NOS inhibitors (reviewed in Ref. 5). A connection between NO and cAMP, both of which affect endothelial transport properties, was proposed by Chang, who provided evidence in BAECs that shear-stress-induced increases in NO inhibit glycolysis, which, in turn, reduces cAMP.
The mechanism by which step changes in pressure lead to a long transient response and a substantial increase in Lp is not known. It seems clear, however, that some mechanical stimulus activates a biochemical response within the endothelial cells that, in turn, modulates transport pathways. The mechanical stimulus could be 1) a direct effect of pressure on the endothelial cells, leading to membrane and cytoskeletal deformation; 2) an indirect effect of pressure, inducing transverse stretch in the endothelial cells because of stretching of the cell support; or 3) an indirect effect of pressure that induces increased transmural flow and associated fluid shear stress on the endothelial surfaces of interendothelial junctions.
We have attempted to eliminate the effects of transverse stretch (2 above) by incorporating a filter-support system that reduces the maximum area strain to <0.08%. Studies with supported and unsupported filters at the highest pressure show no significant difference in the response of fluid flux (Fig. 1), indicating that transverse stretch does not influence the results.
We have investigated the direct effect of pressure by comparing the volume flux (induced by a 10-cmH2O pressure differential) across monolayers exposed to hydrostatic pressure of 25 cmH2O for 4 h with control monolayers exposed to only 0.5 cmH2O hydrostatic pressure for 4 h. There was no significant difference in the volume flux between these two groups, indicating that hydrostatic pressure elevation does not contribute directly to the elevation of Lp observed in our experiments.
These observations are consistent with other studies of the direct effect of pressure on endothelial cells. Acevedo et al. (1) plated bovine pulmonary artery endothelial cells in subconfluent monolayers on rigid, impermeable substrates and subjected them to hydrostatic pressures in the range 1.5-15 cmH2O for up to 7 days. Changes in cell morphology, cytoskeletal rearrangement, and cell proliferation were observed after 5 days of exposure to pressure, but no changes could be detected after 3 days of exposure. Hishikawa et al. (12) measured the release of endothelin-1 (ET-1) from cultured human umbilical vein endothelial cells and observed that an increase in hydrostatic pressure to 40 mmHg above atmospheric had no significant effect on ET-1 production after 8 h of exposure. There was, however, a significant increase in ET-1 production after 8 h of exposure to 80-mmHg pressure elevation. These studies support our conclusion that a direct change in pressure is not a factor in explaining the increase in Lp over a 5-h period at a maximum pressure of only 30 cmH2O observed in the present study. However, we cannot be sure that other more subtle biochemical responses to direct pressure changes that would be relevant to the phenomena we have observed do not evolve over shorter time scales.
The potential role of increased flow through interendothelial junctions
(3 above) on endothelial cell function has not been assessed
previously. A plausible hypothesis is that flow through endothelial
cell junctions imposes significant fluid shear stress on the walls of
the junctions that stimulates the cells much like endothelial cells are
stimulated by fluid wall shear stress on the luminal surface imposed by
blood flow. To consider the feasibility of this
hypothesis, we consider two simple models to estimate the endothelial
cleft wall shear stress: 1) an
open-slit model with laminar flow of plasma (viscosity 1 cP); and
2) a slit model that is filled with
the fiber matrix associated with the intercellular cleft junction
proteins. Although the actual flow through the endothelial cleft is
undoubtedly much more complicated than envisioned in these two simple
models (23), they should provide estimates of the order of magnitude of
the wall shear stress (
w).
Open-Slit Model
For laminar flow of a Newtonian fluid through a slit with parallel walls, it is a simple matter to show that (3)
|
(6) |
is
the average velocity of the fluid in the slit, and
B is the half-width of the slit.
Noting that
|
(7) |
|
(8) |
6 cm/s (typical baseline
value at 10 cmH2O), the open-slit
model predicts
|
(9) |
|
(10) |
P is the pressure drop across the slit of length
L (3). This estimate requires
different parameters from those in Eq. 8. Taking P to be the full
10-cmH2O pressure differential applied across the cell-filter system (i.e., negligible pressure drop
across the filter); L to be 1 µm;
and B, again, to be 10 nm,
Eq. 10 predicts
|
(11) |
Fiber Matrix Slit Model
If the slit (of width 2 B) is assumed to be filled with a fiber matrix that is characterized by a (Darcy) hydraulic permeability coefficient, Kp, and it is reasonable to assume
|
(12) |
|
(13) |
|
(14) |
is the spacing between fibers. If we assume that
Kp is determined
primarily by an array of glycosaminoglycan fibers having a fiber radius
of 0.6 nm (14) and that the spacing of the fibers is ~7 nm (25), then
Kp = 7.07 nm2
(
= 2.66 nm). Using
this estimate of
Kp along with
Eqs. 13 and 7, we obtain the estimate
|
(15) |
w that is ~35% higher than
Eq. 13.
All estimates of the endothelial cleft wall shear stress indicate values that are on the same order of magnitude as the shear stress of flowing blood on the luminal surface of the endothelial cells (15). In addition, it is clear in Eqs. 6 and 13 that the cleft wall shear stress is proportional to the volume flux across the endothelium.
In this context, it is also of interest to estimate the order of magnitude of the surface area of the endothelial cleft on which the wall shear stress acts. Bundgaard and Frokjaer-Jensen (4) measured the length of the line of contact between adjacent endothelial cells to be 1,800 cm/cm2 in frog mesenteric capillaries. If we assume a cleft depth on the order of 1 µm, then the cleft surface area is estimated to be ~18% of the luminal surface area, which is significant.
These simple models and estimates support the plausibility of a mechanism in which interendothelial cleft wall shear stress driven by transendothelial volume flux stimulates endothelial cells. Chang (5), using the same in vitro transport model as in the present study, demonstrated that a step change in wall shear stress on the luminal surface of the endothelial monolayer of 20 dyn/cm2 could induce a transient increase in Lp that did not reach a new steady state (i.e., was still increasing) 3 h after the onset of shear. Furthermore, this transient increase could be reversed by the addition of DBcAMP after 3 h of shear and could be substantially inhibited by preincubating the monolayers with the NOS inhibitors L-NMMA and L-NAME. We have observed essentially the same responses in the present study by using changes in transmural pressure (volume flux) to stimulate the endothelial cells. This further supports our hypothesis that an increase in endothelial cleft wall shear stress driven by a change in transmural pressure stimulates the increase in Lp by a NO-cAMP-dependent mechanism.
There are a few subtle mechanisms other than endothelial cleft wall shear stress that could have provided a stimulus in our experiments. Under a pressure gradient, the cells were being pushed down onto the underlying polycarbonate membrane, likely producing localized deformation of the basal cell membrane. Such deformations, however, are expected to have been small, and, on the basis of our experiments with supported and unsupported filters (Fig. 1), seem unlikely to have been significant. The onset of pressurization might also have caused some water expulsion due to the transient pressure difference across the basal cell membrane. However, because we applied the pressure gradient slowly (over a 1-min period) rather than suddenly, such transient pressure differentials should have been minimized.
Alterations in Lp driven by changes in transmural pressure may contribute to fluid volume shifts in microgravity, where removal of the hydrostatic pressure gradient leads to an increased driving force for fluid filtration above the HIL that may contribute to the characteristic facial swelling observed in astronauts (11). Pulmonary edema, driven by hypertension, may also be exacerbated by the mechanism we have described in this paper. Parker and Ivey (18) observed increases in hydraulic permeability in isolated rat lungs in response to increases in venous pressure that could be inhibited by isoproterenol, a cAMP agonist, at venous pressures up to 31 cmH2O. At higher pressures (43 cmH2O) isoproterenol was not an effective inhibitor. This suggests that at higher pressures additional mechanisms contribute to the breakdown of the endothelial transport barrier. It is likely that stretch of the interendothelial junctions and basement matrix is an important additional mechanical consequence of an increase in transmural pressure that contributes to the increase in Lp and may become dominant at higher pressures.
| |
ACKNOWLEDGEMENTS |
|---|
This work was supported by National Aeronautics and Space Administration Grant NAG3-1871 and National Heart, Lung, and Blood Institute Grant HL-57093.
| |
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests and other correspondence: J. M. Tarbell, 155 Fenske Laboratory, The Pennsylvania State Univ., Univ. Park, PA 16802-4400 (E-mail: jmt{at}psu.edu).
Received 3 August 1998; accepted in final form 25 February 1999.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Acevedo, A. D.,
S. S. Bowser,
M. E. Gerritsen,
and
R. Bizios.
Morphological and proliferative responses of endothelial cells to hydrostatic pressure: role of fibroblast growth factor.
J. Cell. Physiol.
157:
603-614,
1993[Medline].
2.
Baldwin, A. L.,
L. M. Wilson,
and
B. R. Simon.
Effect of pressure on aortic hydraulic conductance.
Arterioscler. Thromb.
12:
163-171,
1992
3.
Bird, R. B.,
W. E. Stewart,
and
E. N. Lightfoot.
Transport Phenomena. New York: Wiley, 1960, p. 62.
4.
Bundgaard, M.,
and
J. Frokjaer-Jensen.
Functional aspects of the ultrastructure of terminal blood vessels: a quantitative study of consecutive segments of the frog mesenteric microvascular.
Microvasc. Res.
23:
1-30,
1982[Medline].
5.
Chang, Y. S.
The Mechanism of Shear-Induced Increases in Endothelial Transport Properties (Ph.D. thesis). University Park, PA: The Pennsylvania State University, 1998.
6.
Crone, C.,
and
D. G. Levitt.
Capillary permeability to small solutes.
In: Handbook of Physiology, The Cardiovascular System, Microcirculation. Bethesda, MD: Am. Physiol. Soc., 1984, sect. 2, vol. IV, pt. 1, chapt. 10, p. 411-466.
7.
Davies, P. F.
Flow-mediated endothelial mechanotransduction.
Physiol. Res.
75:
519-559,
1995.
8.
Dull, R. O.,
J. Jo,
H. W. Sill,
T. M. Hollis,
and
J. M. Tarbell.
The effect of varying albumin concentration and hydrostatic pressure on hydraulic conductivity and albumin permeability of cultured endothelial monolayers.
Microvasc. Res.
41:
390-407,
1991[Medline].
9.
Ethier, C. R.,
and
R. D. Kamm.
Flow through partially gel-filled channels.
PhysicoChem. Hydrodynamics
11:
219-224,
1989.
10.
Gilbert, J. A.,
P. S. Weinhold,
A. J. Banes,
G. W. Link,
and
G. L. Jones.
Strain profiles for circular culture plates containing flexible surfaces employed to mechanically deform cells in vitro.
J. Biomech.
27:
1169-1177,
1994[Medline].
11.
Hargens, A. R.,
D. W. Watenpaugh,
and
G. A. Breit.
Control of circulatory functions in altered gravitational fields.
Physiologist
35, Suppl.:
S80-S83,
1992[Medline].
12.
Hishikawa, K.,
T. Nakaki,
T. Marumo,
H. Susuki,
R. Kato,
and
T. Saruta.
Pressure enhances endothelin-1 release from cultured human endothelial cells.
Hypertension
25:
449-452,
1995
13.
Lever, M. J.,
J. M. Tarbell,
and
C. G. Caro.
The effect of luminal flow in rabbit carotid artery on transmural fluid transport.
Exp. Physiol.
77:
553-563,
1992[Abstract].
14.
Levick, J. R.
Flow through interstitium and other fibrous matrices.
Q. J. Exp. Physiol.
72:
409-438,
1987
15.
Lipowsky, H.
Shear stress in the circulation.
In: Flow Dependent Regulation of Vascular Function, edited by J. Bevan,
and G. Kaley. New York: Oxford Univ. Press, 1995, p. 28-45.
16.
Michel, C. C.
Filtration coefficients and osmotic reflexion coefficients of the walls of single frog mesenteric capillaries.
J. Physiol. (Lond.)
309:
341-355,
1980
17.
Moore, T. P.,
and
W. R. Thornton.
Space shuttle inflight and postflight fluid shifts measured by leg volume changes.
Aviat. Space Environ. Med.
58, Suppl.:
A91-A96,
1987[Medline].
18.
Parker, J. C.,
and
C. L. Ivey.
Isoproterenol attenuates high vascular pressure-induced permeability increases in isolated rat lungs.
J. Appl. Physiol.
83:
1962-1967,
1997
19.
Sill, H. W.,
C. Butler,
T. M. Hollis,
and
J. M. Tarbell.
Albumin permeability and electrical resistance as means of assessing endothelial monolayer integrity in vitro.
J. Tiss. Cult. Meth.
14:
253-258,
1992.
20.
Sill, H. W.,
Y. S. Chang,
J. R. Artman,
J. A. Frangos,
T. M. Hollis,
and
J. M. Tarbell.
Shear stress increases hydraulic conductivity of cultured endothelial monolayers.
Am. J. Physiol.
268 (Heart Circ. Physiol. 37):
H535-H543,
1995
21.
Suttorp, N.,
T. Hessz,
W. Seeger,
A. Wilker,
R. Koob,
and
D. Drenckenhan.
Bacterial endotoxins and endothelial permeability for water and albumin in vitro.
Am. J. Physiol.
255 (Cell Physiol. 24):
C368-C376,
1988
22.
Tedgui, A.,
and
M. J. Lever.
Filtration through damaged and undamaged rabbit thoracic aorta.
Am. J. Physiol.
247 (Heart Circ. Physiol. 16):
H784-H791,
1984
23.
Tsay, R.-Y.,
and
S. Weinbaum.
Viscous flow in a channel with periodic cross-bridging fibres: exact solutions and Brinkman approximation.
J. Fluid Mech.
226:
125-135,
1991.
24.
Turner, M. R.
Flows of liquid and electrical current through monolayers of cultured bovine arterial endothelium.
J. Physiol. (Lond.)
449:
1-20,
1992
25.
Weinbaum, S.
1997 Whitaker distinguished lecture: models to solve mysteries in biomechanics at the cellular level; a new view of fiber matrix layers.
Ann. Biomed. Eng.
26:
627-643,
1998[Medline].
26.
Winston, F. K.,
E. J. Macarak,
S. F. Gorfein,
and
L. E. Thibault.
A system to reproduce and quantify the biomechanical environment of the cell.
J. Appl. Physiol.
67:
397-405,
1989
This article has been cited by other articles:
|
|
 |
|
  S. V. Lopez-Quintero, R. Amaya, M. Pahakis, and J. M. Tarbell The endothelial glycocalyx mediates shear-induced changes in hydraulic conductivity Am J Physiol Heart Circ Physiol, May 1, 2009; 296(5): H1451 - H1456. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Burns, Z. Zheng, S. H. Soubra, J. Chen, and R. E. Rumbaut Transendothelial flow inhibits neutrophil transmigration through a nitric oxide-dependent mechanism: potential role for cleft shear stress Am J Physiol Heart Circ Physiol, November 1, 2007; 293(5): H2904 - H2910. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. K. Sumanasekera, G. U. Sumanasekera, K. A. Mattingly, S. M. Dougherty, R. S. Keynton, and C. M. Klinge Estradiol and dihydrotestosterone regulate endothelial cell barrier function after hypergravity-induced alterations in MAPK activity Am J Physiol Cell Physiol, August 1, 2007; 293(2): C566 - C573. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. O. Dull, I. Mecham, and S. McJames Heparan sulfates mediate pressure-induced increase in lung endothelial hydraulic conductivity via nitric oxide/reactive oxygen species Am J Physiol Lung Cell Mol Physiol, June 1, 2007; 292(6): L1452 - L1458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Parker, T. Stevens, J. Randall, D. S. Weber, and J. A. King Hydraulic conductance of pulmonary microvascular and macrovascular endothelial cell monolayers Am J Physiol Lung Cell Mol Physiol, July 1, 2006; 291(1): L30 - L37. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-h. Kim, N. R. Harris, and J. M. Tarbell Regulation of hydraulic conductivity in response to sustained changes in pressure Am J Physiol Heart Circ Physiol, December 1, 2005; 289(6): H2551 - H2558. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-h. Kim, N. R. Harris, and J. M. Tarbell Regulation of capillary hydraulic conductivity in response to an acute change in shear Am J Physiol Heart Circ Physiol, November 1, 2005; 289(5): H2126 - H2135. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. SAJKOV, T. WANG, N. A. SAUNDERS, A. J. BUNE, and R. DOUGLAS MCEVOY Continuous Positive Airway Pressure Treatment Improves Pulmonary Hemodynamics in Patients with Obstructive Sleep Apnea Am. J. Respir. Crit. Care Med., January 15, 2002; 165(2): 152 - 158. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
;
n = 4) and unsupported (
;
n = 4) monolayers after a step change
in pressure from 10 to 30 cmH2O at
60 min. Baselines:
Jv = 3.69 ± 0.24 × 10
60 min.
