| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Department of Mechanical Engineering and 2 Cancer Institute, University of Nevada, Las Vegas, Las Vegas, Nevada 89154
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
To investigate the ultrastructural
mechanisms of acute microvessel hyperpermeability by vascular
endothelial growth factor (VEGF), we combined a mathematical model
(J Biomech Eng 116: 502-513, 1994) with
experimental data of the effect of VEGF on microvessel hydraulic
conductivity (Lp) and permeability of
various-sized solutes. We examined the effect of VEGF on microvessel
permeability to a small solute (sodium fluorescein, Stokes radius 0.45 nm), an intermediate solute (
-lactalbumin, Stokes radius 2.01 nm), and a large solute [albumin (BSA), Stokes radius 3.5 nm]. Exposure to
1 nM VEGF transiently increased apparent permeability to 2.3, 3.3, and
6.2 times their baseline values for sodium fluorescein, -lactalbumin, and BSA, respectively, within 30 s, and all
returned to control within 2 min. On the basis of
Lp (DO Bates and FE Curry. Am J
Physiol Heart Circ Physiol 271: H2520-H2528, 1996) and
permeability data, the prediction from the model suggested that the
most likely structural changes in the interendothelial cleft induced by
VEGF would be a ~2.5-fold increase in its opening width and partial degradation of the surface glycocalyx.
solute permeability; frog; model for interendothelial cleft
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
ALTHOUGH VASCULAR ENDOTHELIAL growth factor (VEGF) was originally identified by its ability to increase extravasation of plasma proteins (solute flux), how it accomplishes this remains unclear. It has been assumed that VEGF increases solute flux by stimulating microvascular endothelial cells to increase microvessel permeability (13), although other parameters, such as microvessel surface area and driving forces, can increase solute flux. To address this problem, numerous studies have used a variety of approaches: measurements of in vivo transcapillary solute flux (31), molecular localization of receptors for VEGF on endothelial cells in vivo (28) and in vitro (6), in vitro measurements of intracellular Ca2+ fluxes and growth assays in endothelial cells derived from various tissues (14), and in vivo growth assays in a variety of vessels (12). None of these studies has conclusively demonstrated that VEGF controls solute flux by acting directly on the capillary wall or on the cells forming the barrier for exchange. The difficulty in explaining the mechanism of solute flux increase, specifically, the transient increase induced by VEGF, arises from a lack of a well-controlled in vivo system that allows measurement of solute flux under known surface area and driving forces.
Three types of permeability coefficients can be directly measured
across the microvascular wall: the solute permeability coefficient, the
solute reflection coefficient, and the hydraulic conductivity (Lp) (15). The first two
coefficients describe the barrier properties of the microvessel wall to
solutes, and the third parameter describes the barrier properties of
the microvessel wall to water. Wu et al. (33) found in ex
vivo coronary venules that VEGF increased apparent albumin
permeability, which reached a maximum 5 min after stimulation and
returned to control levels 5 min later in a nitric oxide-dependent
manner. Exposure to 1 nM VEGF rapidly and transiently increased
Lp within 30 s (to 7.8 times baseline
values) and returned to control within 2 min (8). After
longer times (i.e., 24 h after 10 min of perfusion), VEGF
increased Lp to 6.8 times baseline without
affecting the reflection coefficient to albumin (7). Although there are several reports of increased
Lp in response to VEGF in intact microvessels
(7-11, 20, 26, 27), there are no data showing the
increase and the history of the increase in permeability to
various-sized solutes in response to VEGF. The first aim of the present
study is to show that VEGF acutely increases permeability by measuring
the solute flux for various-sized solutes [small (sodium fluorescein,
mol wt 376, Stokes radius 0.45 nm), intermediate (-lactalbumin, mol
wt 14,176, Stokes radius 2.01 nm), and large (BSA, mol wt 67,000, Stokes radius 3.5 nm)] under conditions of known surface area and
driving forces. Sodium fluorescein can easily penetrate all the
structural barriers in the interendothelial cleft. -Lactalbumin has
difficulty traveling through the surface glycocalyx. BSA is rarely
transferred across the microvessel wall under normal conditions; its
diameter is the cutoff size for the spacing between adjacent fibers in
the surface glycocalyx (19, 32). In conjunction with a
model for the interendothelial cleft (19), we use the
distinct transport characteristics of these probe molecules and
additional information from previous electron microscopy studies and
the measurement of Lp (8) to
predict changes in structural components of the interendothelial cleft
under the influence of VEGF.
The cleft between adjacent endothelial cells (interendothelial cleft)
is widely believed to be the principal pathway for water and
hydrophilic solute transport through the microvessel wall under normal
physiological conditions (23). The interendothelial cleft
is also suggested to be the pathway for transport of
high-molecular-weight plasma proteins, leukocytes, and tumor cells
across microvessel walls in disease. Direct and indirect evidence
(19, 32) indicates that there are tight junction strands
with discontinuous leakages and fiber matrix components (glycocalyx
layer) at the endothelial surface and within the cleft (Fig.
1). These structural components of the
microvessel wall form the barrier between the bloodstream and body
tissues, which maintains the normal microvessel permeability to water
and solutes.
|
The structural mechanisms that have been studied to increase microvessel permeability by VEGF, inflammatory mediators, and physical stimuli involve the formation of gaps between adjacent endothelial cells in venular microvessels (22), vesiculovacuolar organelle pathways (16), transcellular pores (16, 25), and fenestra (16, 29). However, there may be changes in the thickness and organization of the endothelial cell glycocalyx and more subtle changes in junction ultrastructure that do not lead to formation of the above-mentioned pathways, especially in the case of an acute increase.
Microvessel hyperpermeability is the critical step for the abnormal transport of molecules and cells across the blood vessel and, thus, the crucial step for tumor growth and metastasis. Understanding the mechanisms of microvessel hyperpermeability from various approaches is important in combating these malignant diseases. The second aim of our study is to investigate the ultrastructural mechanisms of acute microvascular hyperpermeability induced by VEGF in intact microvessels by combining a theoretical model for the interendothelial cleft (19) and experimental data for microvessel permeability to water and various-sized solutes in intact microvessels. Here we test the hypothesis that VEGF induces acute (0-5 min) increases in microvessel permeability by destroying the integrity of the structural components in the cleft between endothelial cells forming the microvessel wall.
Solute permeability was measured by quantitative fluorescence microscope photometry (3, 17, 21). All experiments were performed on individually perfused venular microvessels in frog mesentery. Each microvessel was perfused via two micropipettes to enable us to switch rapidly from a clear (washout) perfusate to one containing the fluorescently labeled test solute. This method allowed us to repeat measurements of solute permeability under more than one chemical treatment on each microvessel. In addition, permeability was measured by perfusing segments of the microvessel with fluorescently labeled solutes under conditions in which the transmicrovessel differences in solute concentration and hydrostatic pressure were known (21).
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
General
All in vivo experiments were performed on male leopard frogs (Rana pipiens, 2.5-3 in. long; J. M. Hazen, Alburg, VT). The methods used to prepare the frog mesentery, perfusate solutions, and micropipettes for microperfusion experiments have been described in detail elsewhere (3, 17, 21). A brief outline of the methods is given with emphasis on the special features of the present experiments. The protocol (R714-1198-144) for the experiments was approved by IACUC of University of Nevada, Las Vegas.The frog brain was destroyed by pithing, with the spinal cord left
intact. The abdominal cavity was opened, and the mesentery was gently
arranged on the surface of a polished quartz pillar (1 cm diameter;
Heræus-Amersil, Fairfield, NJ) to maintain circulation to the gut and
mesentery of the animal. The upper mesentery was continuously
superfused with frog Ringer solution at room temperature (~23°C).
Venular capillaries, generally 20-35 µm diameter, were chosen
for study. All vessels had brisk blood flow immediately before
cannulation and had no marginating white blood cells. Each of the two
arms of a Y-branched microvessel was cannulated with a beveled glass
micropipette containing perfusion solution. This arrangement allowed
alternate perfusion of the downstream vessel with a washout solution
(containing no fluorescent solute) on one side and the test solution
(containing the fluorescent solute) on the other side. Each pipette was
connected to a water manometer that allowed perfusion at known
pressures. For these experiments, the pressure was 5-10
cmH2O, as determined by balancing the interface between
fluorescent and nonfluorescent solution within the nonflowing arm of
the Y-branched microvessel. In each vessel, permeability was determined
for straight segments, 300-400 µm long, 
100 µm downstream
from the Y-branch junction point.
Frog Ringer solution was used for all dissections, perfusates, and superfusates. The solution composition was (in mM) 111 NaCl, 2.4 KCl, 1.0 MgSO4, 1.1 CaCl2, 0.195 NaHCO3, 5.5 glucose, and 5.0 HEPES, with pH balanced to 7.4 by adjustment of the ratio of HEPES acid to base. In addition, the clear solution and the fluorescent dye solution contained BSA (catalog no. A4378, Sigma) at 10 mg/ml.
Fluorescent Test Solute Preparation
Sodium fluorescein. Sodium fluorescein (catalog no. F6377, Sigma; mol wt 376, Stokes-Einstein radius ~0.45 nm) was dissolved at 0.1 mg/ml in frog Ringer solution containing 10 mg/ml BSA. The solution was made fresh on the day of use to avoid binding to the serum albumin (4, 17).
FITC-labeled -lactalbumin and BSA.
-Lactalbumin (catalog no. L6010, Sigma; mol wt 14,176, Stokes-Einstein radius ~2 nm) and BSA (catalog no. A4378, Sigma; mol wt ~67,000, Stokes-Einstein radius ~3.5 nm) were labeled with FITC
(catalog no. F7250, Sigma; mol wt 389.4), as described elsewhere (17, 21). Briefly, protein (90 mg) was dissolved in 15 ml of borate buffer (0.05 M, pH ~9.3, 20°C) containing 0.4 M NaCl. The
solution was placed in 18-mm-diameter dialysis tubing with a
3,500-mol-wt cutoff (Spectrum Medical Industries) and dialyzed for
12 h with constant stirring at 15°C against 50 ml of the borate buffer containing FITC (0.5 mM). The labeled protein then was dialyzed
against 2 liters of glucose-free frog Ringer solution twice, each time
for 12 h. Then the dialysis procedure was repeated twice with 2 liters of normal frog Ringer solution until there was no free dye. The
influence of free dye on measured permeability to a labeled protein has
been discussed previously (15). FITC-labeled -lactalbumin and BSA were stored frozen and used within 2 wk of
preparation. On the day of use, unlabeled BSA was added to aliquots of
the labeled protein. The final FITC--lactalbumin and FITC-BSA dye
concentration was 2 mg/ml in frog Ringer solution. For this
preparation, the fluorescence intensity of the free FITC dye was 1% of
the solution and was checked using the photometer at the same
instrument settings used in our experiments.
Microscope and Photometer Preparation
A detailed description of the method used to measure permeability of fluorescently labeled solutes has been published previously (17, 21). In the present experiment, we used a different but similar experimental setup. An inverted fluorescence microscope (Eclipse TE-300, Nikon) was used to observe the mesentery. A ×10 lens (NA 0.3, Nikon) gave a ~2-mm-diameter field of view. The tissue was observed with transmitted white light from a light pipe suspended above the preparation or with fluorescent light from a xenon lamp (300 W; Intracellular Imaging) with appropriate filter sets for fluorescein. The filter set (model XF23, Omega) that was used for sodium fluorescein, FITC-BSA, and FITC--lactalbumin consisted of an
excitation filter (model 485DF22), a dichroic mirror (model 505DRLP),
and an emission filter (model 535DF35). The intensity of the xenon
light source was controlled by an attenuator knob on the front panel of
the lamp housing. The intensity of the light output can be adjusted
from 0 to 100%. To reduce the tissue damage due to exposure to
fluorescent light, the light intensity was kept as low as possible.
Further protection was provided by using an experimental protocol in
which the time of tissue exposure to the excitation light was kept as
short as possible for the permeability measurement. Generally, exposure
time for an individual measurement was 5-20 s. The larger the
molecule, the longer was the exposure time. The fluorescence intensity
(If) in the capillary lumen and surrounding
tissue was measured by aligning the vessel segment within an adjustable
measuring window consisting of a rectangular diaphragm in the light
path. The maximum size of the window was 250 µm wide and 650 µm
long. In our experiment, the dimensions of the measuring window were
generally 100-200 µm wide (~5 times the microvessel diameter)
and 300-500 µm long. The measuring window was set 100 µm
from the base of the vessel to avoid solute contamination from the
sidearm. If measured by a photometer (model
HC135-11, Hamamatsu) was recorded into a computer through an
analog-to-digital board by using InCyt Pm1 Photometry software
(Intracellular Imaging). Compared with previous strip chart recordings,
this type of recording greatly improves the spatial and temporal
resolution of the intensity vs. time curve, which is used to calculate
solute permeability (P) as follows: P = (1/
If 0)[(dIf/dt)0](r/2),
where If 0 is the step increase in
fluorescent light intensity as the test solute fills the
microvessel lumen,
(dIf/dt)0 is the
initial rate of increase in fluorescence light intensity after solute
fills the lumen and begins to accumulate in the tissue, and
r is the microvessel radius (17, 21).
Calibration Experiments
We used in vitro and in vivo calibration experiments to test the assumption in the calculation of solute permeability that If is a linear function of the number of solute molecules in the measuring field. The instrument settings used in the calibration experiments were the same as those used in the permeability measurements. For different-sized test solutes, we used different instrument settings by adjusting the intensity attenuator knob in the lamp housing to obtain enough fluorescent intensity for the measurement but to avoid tissue damage due to exposure to the fluorescent light.In vitro calibrations were performed using two chambers of different
depths: a 100-µm-deep cell-counting chamber (hemocytometer; Hausser
Scientific) and a ~170-µm-deep chamber constructed of glass
coverslips (4, 17). Solutions of fluorescein were applied to fill the chamber by capillarity, and the intensity was measured. The
chamber was cleaned before different concentrations of solutions were
applied. These in vitro calibrations showed that the relation between
the concentration and the fluorescence intensity was linear from 0.05 to 0.2 mg/ml for sodium fluorescein and from 0.3 to 6 mg/ml for
FITC--lactalbumin and BSA.
In the in vivo calibration experiment, a microvessel was cannulated and
perfused as if for measurement of solute permeability. We recorded the
step increase in If 0 as the solute was perfused into the microvessel. Perfusion time was <5 s during each run
to minimize accumulation in the surrounding tissue. The procedure was
repeated on the same microvessel at each of four sodium fluorescein
concentrations: 0.05, 0.1, 0.15, and 0.2 mg/ml. The step increase was
plotted as a function of concentration. For FITC--lactalbumin and
FITC-BSA, different microvessels were used in in vivo calibrations.
Four concentrations of FITC--lactalbumin and FITC-BSA were used:
0.33, 0.67, 2, and 6 mg/ml. In vivo calibration results for the same
concentration ranges used in the in vitro experiment are shown in Fig.
2. Figure 2A shows the results
of an experiment in which a microvessel was perfused with four
concentrations of sodium fluorescein. Figure 2, B and
C, shows the results for FITC--lactalbumin and FITC-BSA.
The concentrations used in our experiments were 0.1 mg/ml for sodium
fluorescein and 2 mg/ml for FITC--lactalbumin and FITC-BSA.
|
Using in vivo experiments, we also determined whether the initial step of If 0 was independent of the width of the measuring window when the capillary was aligned along the centerline of the window, provided the window was wider than and up to five times the diameter of the vessel. This experiment showed no contamination of our measuring window from dye leaked into the tissue during cannulation or diffused into the tissue from adjacent vessels. If there was contamination during the permeability measurement, the increase of If in the measuring window when the dye traveled out of the vessel and accumulated in the tissue would not be linear. When this occurred, the measurement was discarded.
Furthermore, we found in the in vitro photobleaching experiment that
sodium fluorescein (0.1 mg/ml) and FITC--lactalbumin and FITC-BSA (2 mg/ml) If values fell ~0.4, 0.3, and 0.7% of
their original values, respectively, in 1 min. One minute was typically two to four times as long as the exposure time required for an individual solute permeability measurement. The much lower degree of
photobleaching than in previous experiments (3, 17) was due to the reduction in excitation light intensity as a result of
adjusting the attenuator knob of the light source. In addition, the
shorter exposure time for the larger solute (BSA) in an individual permeability measurement was due to much higher sensitivity of the
computer recording system (dI/If 0 ~1/10,000)
than of the previously used strip chart recorder (~1/100). In
general, the present system is 100 times as sensitive as the chart recorder.
Experimental Protocol
To test the effect of 1 nM VEGF (human recombinant VEGF-165, Peprotech, Rocky Hill, NJ) on permeability of various-sized solutes, for each test solute, we made several control measurements when the washout pipette was filled with Ringer perfusate containing BSA (10 mg/ml) and the dye pipette was filled with the same perfusate to which the test solute was added. We then replaced washout and test pipettes with new pipettes that also contained VEGF (1 nM). The concentration of VEGF was chosen to be consistent with that used in Lp measurements by Bates and Curry (8). During the replacement of the pipettes, the pressures for washout and dye pipettes were dropped to <1 cmH2O, so there was negligible flow at the tip during recannulation. Then pressure at the washout side was increased to 20-30 cmH2O while pressure at the dye side was increased to 5-10 cmH2O to balance pressure at the washout side. These pressure settings were chosen for the perfusion of washout solution. After 10-15 s of perfusion with washout solution containing VEGF, pressure at the dye side was switched to 20-30 cmH2O while pressure at the washout side was switched to 5-10 cmH2O for dye perfusion. Perfusion of dye solution containing 1 nM VEGF lasted 5-20 s, depending on solute size. From the initial step increase in fluorescence intensity of the test dye solution and its accumulation in the measuring window (see Calibration Experiments), we can calculate solute permeability. In this way, solute permeability was measured every 15-40 s, including dye (5-20 s) and washout perfusion (10-20 s). During the washout perfusion, the test solute was washed out of the measuring window through the vessel, and we were able to make repeated permeability measurements of the same test solute at different times. The alternating perfusion of dye and washout solutions lasted ~5 min.Analysis and Statistics
Permeability measurements during the control period in a vessel were averaged to establish a single value for control permeability. This value was then used as a reference for all subsequent measurements on that vessel. To present data at a specific time, individual measurements were grouped within 15- to 30-s intervals. For example, for permeability at 15 s, individual measurements starting before and at 15 s of washout perfusion were averaged. For sodium fluorescein, measurements were grouped at 15 s (0-15 s), 30 s (16-30 s), and 45 s (36-45 s). For other solutes, measurements were grouped at 15 s (0-15 s), 35 s (16-35 s), and 65 s (36-65 s). The nonparametric Wilcoxon signed-rank test was applied to the averaged permeability data to test statistical significance of the treatment over time. Mann-Whitney's U-test was applied to between-group data to test for permeability differences at specific times. Significance was assumed for probability levels <5%.| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Control Measurement
To ensure that any changes in permeability after cannulation and perfusion with VEGF were not due to the effects of the recannulation procedure alone, we measured permeability at baseline and then immediately after cannulation and perfusion with 1% BSA (same solution as for baseline; Fig. 3). There was no change in permeability immediately after recannulation and perfusion with 1% BSA alone. For sodium fluorescein, the ratio (mean ± SD) of the peak permeability (within 30 s) compared with baseline (calculated from each individual vessel) was 1.19 ± 0.19 (n = 5). The peak-to-baseline ratios were 1.11 ± 0.42 (n = 6) for-lactalbumin and 0.93 ± 0.58 (n = 5) for BSA. These ratios indicate that the peak
changes in permeability, on average, are 11-19% greater or 7%
less than the baseline values. In comparison, if the maximum baseline
values of permeability are expressed as a ratio of the baselines'
means, then the change is 4-17% for these solutes. The variation
in permeability measurement after recannulation and perfusion with 1%
BSA was therefore no greater than the variation measured during
baseline. Thus there was no significant transient increase in
permeability by a second perfusion with a solution containing 1% BSA
alone.
|
Microvessel Solute Permeability Response to VEGF
After baseline measurement in each vessel, we measured the acute response to 1 nM VEGF in 14 vessels for sodium fluorescein permeability, 17 vessels for-lactalbumin permeability, and 15 vessels for BSA permeability. There was an immediate significant increase in apparent sodium fluorescein, -lactalbumin, and BSA permeabilities (Fig. 3). Because of the rapid nature of the response, not all the vessels had permeability measurements in all time bins
(e.g., only 5, 9, and 4 vessels were measured within the first 15 s after perfusion with VEGF for sodium fluorescein, -lactalbumin, and BSA permeabilities, respectively). Permeability (mean ± SE) measured at the peak of the response was 9.6 ± 2.0 × 10
5 cm/s for sodium fluorescein (n = 14, range 2.7-28.3 × 105 cm/s) compared with a
baseline permeability of 4.2 ± 0.4 × 105 cm/s
(n = 14, range 1.6-7.8 × 105
cm/s, P < 0.015, Wilcoxon signed-rank test). This
represents a mean peak increase in sodium fluorescein permeability of
2.3 ± 0.4-fold compared with baseline in the same vessel
(n = 14, range 1- to 5.5-fold). For -lactalbumin,
the peak was 1.7 ± 0.2 × 105 cm/s
(n = 17, range 0.68-3.3 × 105
cm/s) compared with a baseline permeability of 0.55 ± 0.06 × 105 cm/s (n = 17, range
0.15-0.9 × 105 cm/s, P < 0.01), representing a mean peak increase in -lactalbumin permeability of 3.3 ± 0.3-fold compared with baseline in the same vessel (n = 17, range 1.8- to 5.6-fold). For BSA, the
peak was 0.37 ± 0.06 × 105 cm/s
(n = 15, range 0.11-0.74 × 105
cm/s) compared with a baseline permeability of 0.068 ± 0.009 × 105 cm/s (n = 15, range
0.03-0.13 × 105 cm/s, P < 0.01), representing a mean peak increase in BSA permeability of
6.2 ± 1.2-fold compared with baseline in the same vessel
(n = 15, range 1.2- to 17.1-fold). These control values
are slightly larger than previously measured data (3, 17, 21,
30). The most likely reason is that the temperature in our
experiment, ~23°C, is slightly higher than that in previous
experiments, 18-22°C. The increase was rapid, reaching a peak at
25.6 ± 3.2 s for sodium fluorescein (n = 14), 23.6 ± 3.2 s for -lactalbumin (n = 17), and 29.9 ± 3.2 s for BSA (n = 15) after
reperfusion of the vessel with the Ringer solution containing VEGF. The
transient increase was substantially greater than baseline permeability
in 64, 88, and 87% of vessels for sodium fluorescein, -lactalbumin,
and BSA, respectively (see DISCUSSION for definition of
substantial increase). The increase became not significantly different
from baseline ~95 s after reperfusion (P > 0.2 for
all solutes). There was no change in vessel diameter during our
measurements before and after VEGF perfusion for 5 min, as detected
through eyepiece or measured If 0.
Figure 4 summarizes the control and
VEGF-induced peak values for apparent permeability of all three
solutes. The corresponding values are shown in Table
1 and in the DISCUSSION for
the values after the correction for free dye effect. The mean ratio of
peak to control permeability of the large solute BSA (Stokes radius 3.5 nm) was 6.2, the highest among the three solutes. The ratio for the
intermediate-sized solute -lactalbumin (Stokes radius 2.01 nm) was
3.3, and that for the small solute sodium fluorescein (Stokes radius
0.45 nm) was 2.3. These values for the peak ratio, combined with the
predictions from the previous model for the interendothelial cleft
(19), will help us elucidate the transient structural
changes induced by VEGF.
|
|
Figure 5 presents the corresponding true
diffusive permeability after exclusion of the convective component as a
result of filtration from the apparent permeability. The ratios of mean peak to control values in diffusive permeability are 4.9, 2.8, and 2.3 for BSA, -lactalbumin (before correction for free dye influence),
and sodium fluorescein, respectively.
|
Model Predictions
On the basis of experimental observations (1, 2, 5), Fu et al. (19) proposed a three-dimensional two-pathway model for the interendothelial cleft (Fig. 1). Under normal physiological conditions when cleft width (2B) = 20 nm, large break width (2d) = 150 nm, small slit width (2bs) ~ 1.5 nm, spacing between adjacent large breaks (2D) = 2,640 nm, cleft depth (L) = 400 nm, L1 = L3 = 200 nm, fiber matrix thickness (Lf) = 100 nm, fiber radius (a) = 0.6 nm, and gap spacing between fibers () = 7 nm, this model can successfully explain the data for permeability
to water and solutes ranging from potassium to albumin in frog
mesenteric microvessels.
To test the hypothesis that VEGF induces acute (0-5 min) increases in microvessel permeability by destroying the integrity of the structural components in the cleft between endothelial cells forming the microvessel wall, we changed the values for entrance Lf, 2B, 2D, and 2d in our model.
Figure 6A shows the effect of
entrance Lf. The glycocalyx layer on the
endothelial surface is widely taken as the molecular sieve
(23) to large solutes such as proteins. Removing or
partially removing this layer would greatly increase the microvessel
permeability to the large solute BSA and somewhat increase permeability
to the intermediate-sized solute -lactalbumin and the small solute sodium fluorescein and Lp. Figure 6B
shows the effect of increasing 2B. This would greatly
increase Lp and moderately increase permeability of various-sized solutes. Figure 6C shows the effect of
increasing 2d in the junction strand. This would moderately
increase Lp and permeability to -lactalbumin
but have little effect on permeability to sodium fluorescein and no
effect on permeability to BSA. Figure 6D shows the effect of
decreasing 2D (or increasing the number of junction breaks).
This has an effect that is similar to the effect of increasing the size
of the breaks (Fig. 6C).
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Evaluation of Methods: Free Dye Associated With FITC-Labeled
-Lactalbumin and Albumin
-lactalbumin and
albumin to obtain high quantum yield (ratio of the number of
fluorescence photons emitted to the number of photons absorbed) with
low light excitation. Because FITC (mol wt 389.4) diffuses through
microvessel walls much faster than -lactalbumin (mol wt 14,176) and
albumin (BSA, mol wt 67,000), a small amount of the free FITC would
cause a large overestimation of the permeability to -lactalbumin and
albumin molecules. We therefore measured the amount of free dye in our
labeled -lactalbumin and albumin solutions. After being
ultrafiltered by a clinical centrifuge (1,750 rpm, 444 g)
through a Centricon filter (Millipore, 3,000 mol wt cutoff) from the 2 mg/ml FITC--lactalbumin or FITC-BSA solutions used in our
experiments, the filtrate was checked for fluorescence intensity due to
free FITC (Iff). The method for measuring
Iff is the same as that described in the in
vitro calibration, and the instrument settings are the same as those
used for the solute permeability measurements. The ratio of free dye to
original solutions is ~1% for 2 mg/ml FITC--lactalbumin and
FITC-BSA solutions. If we use measured sodium fluorescein permeability
(4.2 × 105 and 9.6 × 105 cm/s
for control and peak VEGF effect, respectively) for FITC permeability
of microvessels, because the molecular weight of FITC (389.4) is very
close to that of sodium fluorescein (376), provided the
permeability is determined by solute size, we can estimate the
influence of free dye on -lactalbumin and BSA permeability. With the
use of the same method used in the appendix of Fu et al.
(17), Pcorrect = [1/(1 
F)]Pmeasure [F(1 F)]Pfree dye (where
Pmeasure is the measured permeability for
-lactalbumin or BSA, Pfree dye is FITC
permeability, F = 1% is the intensity ratio of free dye filtrate
to original FITC--lactalbumin or FITC-BSA solutions, and
Pcorrect is the corrected permeability for
-lactalbumin or BSA), we estimated that free FITC dye would lead to
an overestimation of -lactalbumin permeability by 7% for control
and 5% for VEGF measurements. However, the free dye would lead to an
overestimation for BSA permeability by 62% for control and 24% for
VEGF measurements. After correction for free dye influence,
-lactalbumin permeability would be 0.51 × 105
cm/s and BSA permeability would be 0.026 × 105 cm/s
under control and peak -lactalbumin permeability would be 1.6 × 105 cm/s and peak BSA permeability would be 0.28 × 105 cm/s under VEGF treatment.
Solvent Drag Contribution to -Lactalbumin and BSA
Permeability
|
(1) |
is the reflection coefficient of microvessel to the
solute, and Pe is the Peclet number, which is
|
(2) |
peff is the effective filtration pressure
across the microvessel wall, which can be expressed as
|
(3) |
p and 
are the hydrostatic and osmotic pressure
drops across the microvessel wall, respectively, and FITC-solute can be
FITC--lactalbumin or FITC-BSA.
In our experiments, p is 5-10 cmH2O, the
reflection coefficient for albumin or FITC-albumin is 0.84 (7), the reflection coefficient for FITC--lactalbumin
is 0.35 (17, 21), the osmotic pressure drop is 3.6 cmH2O for 10 mg/ml BSA, the osmotic pressure drop is 0.72 cmH2O for 2 mg/ml FITC-BSA, and the osmotic pressure drop
is 3.1 cmH2O for 2 mg/ml FITC--lactalbumin. From these
values, peff calculated from Eq. 3 is
1-6 cmH2O for FITC--lactalbumin and FITC-BSA.
Baseline Lp is 2.6 × 107
cm · s1 · cmH2O1
and peak Lp induced by 1 nM VEGF is 24.6 × 107
cm · s1 · cmH2O1
(8). Table 2 shows the
effect of solvent drag on -lactalbumin and BSA permeability. Under
control, the solvent drag contributes 2% to -lactalbumin
permeability and 3% to BSA permeability when peff is 1 cmH2O and 10% to -lactalbumin permeability and 19% to
BSA permeability when peff is 6 cmH2O. Under
VEGF treatment, the solvent drag contributes 5% to -lactalbumin and
BSA permeability when peff is 1 cmH2O and
31% to -lactalbumin permeability and 35% to BSA permeability when
peff is 6 cmH2O.
|
The much larger contribution of the solvent drag to apparent
permeability of the proteins during VEGF treatment is due to the large
increase in Lp, which transiently jumps from a
mean control of 2.6 × 107
cm · s1 · cmH2O1
to a mean peak of 24.6 × 107
cm · s1 · cmH2O1,
a ~10-fold increase. It may be also due to the decrease in the reflection coefficient of albumin (see below).
Table 3 shows the results after the
correction for effects of solvent drag and free dye. The control
-lactalbumin permeability would be 0.5 × 105
cm/s, and control BSA permeability would be 0.025 × 105 cm/s when peff is 1 cmH2O.
These values would be 0.46 × 105 and 0.021 × 105 cm/s, respectively, when peff is 6 cmH2O. Peak -lactalbumin permeability by VEGF would be
1.5 × 105 cm/s, and peak BSA permeability would be
0.27 × 105 cm/s when peff is 1 cmH2O. These values would be 1.1 × 105
and 0.18 × 105 cm/s, respectively, when
peff is 6 cmH2O.
|
The calculations for the solvent drag effect shown above were made with
the assumption that the reflection coefficient for BSA (0.84) is not
changed during acute VEGF treatment. This assumption is based on the
observation by Bates (7) that, after 24 h of exposure
to VEGF, there was no change in albumin reflection coefficient. We now
assume another limiting situation for a significant change in the
reflection coefficient to albumin, which decreases from a control of
0.84 to 0.5. Under this assumption, the contribution of the solvent
drag to apparent permeability is shown in Tables 2 and 3. Because of
the change in the reflection coefficient to albumin,
peff ranges from 3 to 8 cmH2O when the
perfusion pressure (p) is 5-10 cmH2O. For the case
of reduced reflection coefficient to albumin, the contribution from the
solvent drag term Lp(1 )peff in Eq. 1 would be 0.48 × 105 and 0.369 × 105 cm/s for
-lactalbumin and BSA permeability, respectively, when peff is 3 cmH2O, and 1.3 × 105 and 0.98 × 105 cm/s for
-lactalbumin and BSA permeability, respectively, when peff is 8 cmH2O. These high values for the
solvent drag contribution to BSA permeability are close to or much
greater than the measured apparent permeability (0.37 × 105 cm/s) without the correction for the free dye
contribution. This may induce negative values for diffusive
permeability when peff is 8 cmH2O and when
the apparent BSA permeability is used with the free dye correction
(0.28 × 105 cm/s). The possible explanations are as
follows: 1) the peak values for the apparent BSA
permeability are not really the peak, as for Lp,
but an averaged value over a period, because of the limitation in the
technique for the measurement of solute permeability (the real peak
value should be higher); and 2) the reflection coefficient
of albumin should not be reduced significantly under acute VEGF
treatment. These calculations show that a reliable estimate of
diffusive permeability for BSA may not be possible when the solvent
drag accounts for most of the albumin flux.
Heterogeneity of Response to VEGF
The same criteria used by Bates and Curry (8) were chosen to determine whether a vessel was considered to have a significant response to perfusion with solutions containing VEGF. The rationale for the selection of the criteria was as follows. For example, the maximum increase (mean ± SD) in sodium fluorescein permeability in the control vessels during a second consecutive perfusion with 1% BSA was 1.19 ± 0.19 times the mean sodium fluorescein permeability during the original perfusion. An increase in sodium fluorescein permeability of >2 SDs greater than the mean increase [(2 × 0.19) + 1.19 = 1.57-fold] was therefore accepted as a significant increase in sodium fluorescein permeability. Nine of 14 (64%) vessels increased their permeability to sodium fluorescein by >1.57-fold and were classified as responders. By the same criteria, 88 and 87% of vessels substantially responded to VEGF in-lactalbumin and BSA permeability, respectively.
Not all vessels responded to VEGF in solute permeability according to this classification. This is consistent with the observation in Lp experiments by Bates and Curry (8). In their study, only 70% of microvessels responded to perfusion with solutions containing VEGF. This indicated a considerable degree of heterogeneity in the ability of the vessels to respond, although we always chose postcapillary venules for experiments.
Structural Mechanisms of VEGF Effect on Microvessel Permeability
We have compared our measured permeability data with the model predictions when the cleft width is increased (Fig. 7). As shown in Fig. 7, except for measured BSA permeability, all measured peak permeability data almost overlap the corresponding model predictions. This suggests that the acute increase in microvessel permeability induced by VEGF might be caused by a 2.5-fold transient increase in the width of the interendothelial cleft (Fig. 1). Figure 6, C and D, shows that the increase in the junctional break size or number of breaks would not change BSA permeability. However, Fig. 6A shows that removing the surface glycocalyx would induce a large increase in BSA permeability but minor increases in permeability to the other solutes. Therefore, the higher value in measured BSA permeability than the prediction in Fig. 7 would be explained by a partial removal or degradation (~50%) of the surface fiber matrix (glycocalyx). The modeling conclusions apply only in the absence of significant solvent drag due to the possible unreliability in the estimation of diffusive permeability to BSA when the solvent drag accounts for most of the albumin flux.
|
Bates et al. (10) summarized that, under various conditions and in different types of blood vessels, VEGF can stimulate formation of a variety of pathways through the endothelial cell, including gaps between adjacent endothelial cells in venular microvessels (22), vesiculovacuolar organelle formation (16), transcellular pores (16, 25), and fenestra (16, 29). Michel and Neal (25) used electron microscopy to study the mechanism of increased permeability by VEGF. They found that, after ~10 min of perfusion with 1 nM VEGF in microvessels, the pores observed by electron microscopy were almost transcellular, instead of intercellular. Some of these structures may be more consistent with a transport mechanism dominated by the solvent drag. Although these are alternative possibilities of structural mechanisms for VEGF-induced microvessel hyperpermeability, the method of combining the experimental results and the theoretical model predictions provides a useful tool for elucidating the transient structural changes induced by VEGF during the period from a few seconds to a few minutes.
In summary, the present study shows that frog mesenteric microvessels
respond acutely to 1 nM VEGF by a rapid and transient increase in
permeability to sodium fluorescein, -lactalbumin, and BSA.
Combination of these data and previous data for
Lp (8) with a model for the
interendothelial cleft (19) suggests that the possible
structural changes by VEGF would be a ~2.5-fold increase in the
opening width of the interendothelial cleft and a partial degradation
of the surface glycocalyx. Similar to Lp in
previous experiments (8), there is considerable
heterogeneity in the initial response, with significant increases in
permeability to sodium fluorescein, -lactalbumin, and BSA in only
64, 87, and 88% of the vessels, respectively.
| |
ACKNOWLEDGEMENTS |
|---|
This work is supported by National Cancer Institute Grant R15 CA-86847-01, a National Science Foundation Career Award, and the University of Nevada, Las Vegas, Applied Research Initiative.
| |
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: B. M. Fu, Dept. of Mechanical Engineering, University of Nevada, 4505 Maryland Pkwy., Box 454027, Las Vegas, NV 89154 (E-mail: bmfu{at}nscee.edu).
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. Section 1734 solely to indicate this fact.
First published January 30, 2003;10.1152/ajpheart.00894.2002
Received 11 October 2002; accepted in final form 27 January 2003.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Adamson, RH.
Permeability of frog mesenteric capillaries after partial pronase digestion of the endothelial glycocalyx.
J Physiol
428:
1-13,
1990
2.
Adamson, RH,
and
Clough G.
Plasma proteins modify the endothelial cell glycocalyx of frog mesenteric microvessels.
J Physiol
445:
473-486,
1992
3.
Adamson, RH,
Huxley VH,
and
Curry FE.
Single capillary permeability to proteins having similar size but different charge.
Am J Physiol Heart Circ Physiol
254:
H304-H312,
1988
4.
Adamson, RH,
Lenz JF,
and
Curry FE.
Quantitative laser scanning confocal microscopy on single capillaries: permeability measurement.
Microcirculation
1:
251-265,
1994[Medline].
5.
Adamson, RH,
and
Michel CC.
Pathways through the intercellular clefts of frog mesenteric capillaries.
J Physiol
466:
303-327,
1993
6.
Barleon, B,
Hauser S,
Schollmann C,
Weindel K,
Marme D,
Yayon A,
and
Weich HA.
Differential expression of the two VEGF receptors flt and KDR in placenta and vascular endothelial cells.
J Cell Biochem
54:
56-66,
1994[Web of Science][Medline].
7.
Bates, DO.
The chronic effect of vascular endothelial growth factor on individually perfused frog mesenteric microvessels.
J Physiol
513:
225-233,
1997.
8.
Bates, DO,
and
Curry FE.
Vascular endothelial growth factor increases hydraulic conductivity of isolated perfused microvessels.
Am J Physiol Heart Circ Physiol
271:
H2520-H2528,
1996
9.
Bates, DO,
Heald RI,
Curry FE,
and
Williams B.
Vascular endothelial growth factor increases Rana vascular permeability and compliance by different signalling pathways.
J Physiol
533:
263-272,
2001
10.
Bates, DO,
Hillman NJ,
Williams B,
Neal CR,
and
Pocock TM.
Regulation of microvascular permeability by vascular endothelial growth factors.
J Anat
200:
581-597,
2002[Web of Science][Medline].
11.
Bates, DO,
Lodwick D,
and
Williams B.
Vascular endothelial growth factor and microvascular permeability.
Microcirculation
6:
83-89,
1999[Web of Science][Medline].
12.
Callow, AD,
Choi ET,
Trachtenberg JD,
Stevens SL,
Connolly DT,
Rodi C,
and
Ryan US.
Vascular permeability factor accelerates endothelial regrowth following balloon angioplasty.
Growth Factors
10:
223-228,
1994[Web of Science][Medline].
13.
Collins, PD,
Connolly DT,
and
Williams TJ.
Characterization of the increase in vascular permeability induced by vascular permeability factor in vivo.
Br J Pharmacol
109:
195-199,
1993[Web of Science][Medline].
14.
Criscuolo, GR,
Lelkes PI,
Rotrosen D,
and
Oldfield EH.
Cytosolic calcium changes in endothelial cells induced by a protein product of human gliomas containing vascular permeability factor activity.
J Neurosurg
71:
884-891,
1989[Web of Science][Medline].
15.
Curry, FE.
Mechanics and thermodynamics of transcapillary exchange.
In: Handbook of Physiology. The Cardiovascular System. Microcirculation. Bethesda, MD: Am. Physiol. Soc, 1984, sect. 2, vol. IV, pt. 1, chapt. 8, p. 309-373.
16.
Feng, D,
Nagy JA,
Payne K,
Hammel I,
Dvorak HF,
and
Dvorak AM.
Pathways of macromolecular extravasation across microvascular endothelium in response to VPF/VEGF and other vasoactive mediators.
Microcirculation
6:
23-44,
1999[Web of Science][Medline].
17.
Fu, BM,
Adamson RH,
and
Curry FE.
Test of two-pathway model for small solute exchange across the capillary wall.
Am J Physiol Heart Circ Physiol
274:
H2062-H2073,
1998
18.
Fu, B,
Curry FE,
Adamson RH,
and
Weinbaum S.
A model for interpreting the labeling of interendothelial clefts.
Ann Biomed Eng
25:
375-397,
1997[Web of Science][Medline].
19.
Fu, BM,
Tsay R,
Curry FE,
and
Weinbaum S.
A junction-orifice-fiber entrance layer model for capillary permeability: application to frog mesenteric capillaries.
J Biomech Eng
116:
502-513,
1994[Web of Science][Medline].
20.
Hillman, NJ,
Whittles CE,
Pocock TM,
Williams B,
and
Bates DO.
Differential effects of vascular endothelial growth factor-C and placental growth factor-1 on the hydraulic conductivity of frog mesenteric capillaries.
J Vasc Res
38:
176-186,
2001[Web of Science][Medline].
21.
Huxley, VH,
Curry FE,
and
Adamson RH.
Quantitative fluorescence microscopy on single capillaries: -lactalbumin transport.
Am J Physiol Heart Circ Physiol
252:
H188-H197,
1987
22.
McDonald, DM,
Thurston G,
and
Baluk P.
Endothelial gaps as sites for plasma leakage in inflammation.
Microcirculation
6:
7-22,
1999[Web of Science][Medline].
23.
Michel, CC,
and
Curry FE.
Microvascular permeability.
Physiol Rev
79:
703-761,
1999
24.
Michel, CC,
and
Kendall S.
Differing effects of histamine and serotonin on microvascular permeability in anaesthetized rats.
J Physiol
501:
657-662,
1997
25.
Michel, CC,
and
Neal CR.
Openings through endothelial cells associated with increased microvascular permeability.
Microcirculation
6:
45-62,
1999[Web of Science][Medline].
26.
Pocok, TM,
and
Bates DO.
In vivo mechanisms of vascular endothelial growth factor-mediated increased hydraulic conductivity of Rana capillaries.
J Physiol
534:
479-488,
2001
27.
Pocok, TM,
Williams B,
Curry FE,
and
Bates DO.
VEGF and ATP act by different mechanisms to increase microvascular permeability and endothelial [Ca2+]i.
Am J Physiol Heart Circ Physiol
279:
H1625-H1634,
2000
28.
Qu, H,
Nagy JA,
Senger DR,
Dvorak HF,
and
Dvorak AM.
Ultrastructural localization of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) to the albumin plasma membrane and vesiculovacuolar organelles of tumor microvascular endothelium.
J Histochem Cytochem
43:
381-389,
1995[Abstract].
29.
Roberts, WG,
and
Palade GE.
Increased microvascular permeability and endothelial fenestration induced by vascular endothelial growth factor.
J Cell Sci
108:
2369-2379,
1995[Abstract].
30.
Rutledge, JC.
Temperature and hydrostatic pressure-dependent pathways of low-density lipoprotein transport across microvascular barrier.
Am J Physiol Heart Circ Physiol
262:
H234-H245,
1992
31.
Senger, DR,
Perruzzi CA,
Feder J,
and
Dvorak HF.
A highly conserved vascular permeability factor secreted by a variety of human and rodent tumor cell lines.
Cancer Res
46:
5629-5632,
1986
32.
Weinbaum, S,
Tsay R,
and
Curry FE.
A three-dimensional junction-pore-matrix model for capillary permeability.
Microvasc Res
44:
85-111,
1992[Web of Science][Medline].
33.
Wu, HM,
Huang Q,
Yuan Y,
and
Grange HJ.
VEGF induces NO-dependent hyperpermeability in coronary venules.
Am J Physiol Heart Circ Physiol
271:
H2735-H2739,
1996
This article has been cited by other articles:
|
|
 |
|
  C. A. Glass, S. J. Harper, and D. O. Bates The anti-angiogenic VEGF isoform VEGF165b transiently increases hydraulic conductivity, probably through VEGF receptor 1 in vivo J. Physiol., April 1, 2006; 572(1): 243 - 257. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. J. Salmon, C. R. Neal, D. O. Bates, and S. J. Harper Vascular endothelial growth factor increases the ultrafiltration coefficient in isolated intact Wistar rat glomeruli J. Physiol., January 1, 2006; 570(1): 141 - 156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Filion and A. S. Popel Intracoronary administration of FGF-2: a computational model of myocardial deposition and retention Am J Physiol Heart Circ Physiol, January 1, 2005; 288(1): H263 - H279. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Ashrafpour, N. Huang, P. C. Neligan, C. R. Forrest, P. D. Addison, M. A. Moses, R. H. Levine, and C. Y. Pang Vasodilator effect and mechanism of action of vascular endothelial growth factor in skin vasculature Am J Physiol Heart Circ Physiol, March 1, 2004; 286(3): H946 - H954. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
1.5 nm.
B: plane view of model. L, total depth of cleft
(~400 nm); L1 and L3,
depths between junctional strand and luminal and abluminal fronts,
respectively; Lf, thickness of fiber matrix at
cleft entrance; 2D, distance between 2 adjacent breaks in
junctional strand; d, one-half width of large junctional
breaks. Depth of pores in junction strand is ~10 nm. At entrance of
cleft on luminal side, surface glycocalyx structures are represented by
a periodic square array of cylindrical fibers. a, Radius of
these fibers; 
