| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Human Physiology, School of Medicine, University of California, Davis, California 95616
 |
ABSTRACT |
|---|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
To investigate the mechanisms whereby guanosine 3',5'-cyclic monophosphate (cGMP) modulates microvessel permeability in vivo, we measured changes in microvessel hydraulic conductivity (Lp) and endothelial cytoplasmic Ca2+ concentration ([Ca2+]i) in response to the cGMP analogs 8-bromo-cGMP (8-BrcGMP) and 8-(p-chlorophenylthio)cGMP (8-pCPT-cGMP) in the presence and absence of inflammatory stimuli in intact individually perfused microvessels in frog and rat mesenteries. The cGMP analog caused a transient increase in Lp and potentiated ATP or bradykinin-induced increases in Lp in frog and rat mesenteric microvessels, respectively. The mean peak value of the test Lp/control Lp after exposure to 8-BrcGMP was 5.3 ± 0.5 in frog microvessels and 2.8 ± 0.4 in rat microvessels. The ATP-induced increase in Lp in frog microvessels was further raised by 8-BrcGMP from 7.0 ± 0.9 to 12.4 ± 1.9 times the control. In rat mesenteric microvessels, the bradykinin-induced increase in Lp was potentiated by 8-BrcGMP from 4.8 ± 0.4 to 8.3 ± 1.3 times the control and was suppressed by the guanylate cyclase inhibitor LY-83583 to 2.6 ± 0.5 times the control. A similar but larger effect was found when using 8-pCPT-cGMP. In contrast to the actions of increased cGMP on microvessel permeability, cGMP analogs had no effect on basal endothelial [Ca2+]i and did not alter the magnitude and time course of ATP or bradykinin-induced increases in endothelial [Ca2+]i. These results suggested that an elevation of cGMP levels in endothelial cells is a necessary step to increase microvessel permeability in intact microvessels, and this regulatory process occurs downstream from Ca2+ influx, which differs from that reported in large-vessel endothelium in culture and in vascular smooth muscle cells. Experiments carried on microvessels in both frog and rat mesenteries provided a direct comparison of the endothelial cell regulatory mechanisms between species.
hydraulic conductivity; guanosine 3',5'-cyclic monophosphate analog; guanylate cyclase inhibitor LY-83583; bradykinin; adenosine 5'-triphosphate
| |
INTRODUCTION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
THE AIM OF THESE experiments was to investigate the mechanisms whereby guanosine 3',5'-cyclic monophosphate (cGMP) modulates the permeability properties of intact venular microvessels. cGMP is a major intracellular mediator of the effects of nitric oxide (NO) and atrial natriuretic peptides (ANP). The activation of either soluble guanylate cyclase (GC) by NO or membrane GC by natriuretic peptides leads to an increase in intracellular cGMP concentration. To date, several possible mechanisms of actions of cGMP in the regulation of permeability have been demonstrated. In the majority of in vitro studies using cultured endothelial cells from large vessels, elevation of cGMP by NO donors, ANP, or cGMP analogs decreases agonist-induced increases in endothelial permeability (1, 7, 23, 26). Experiments in aortic endothelial cells showed that increased cGMP reduced the magnitude of the initial increase in endothelial cell Ca2+ concentration ([Ca2+]i), and this mechanism appeared to be responsible for the decrease in permeability. In human umbilical vein endothelial cells in culture, [Ca2+]i was not modified, but other mechanisms, possibly involving increased cAMP, decreased permeability (7). In contrast to the in vitro results from large-vessel endothelial cells, in vivo studies using isolated porcine coronary venules or intact frog mesenteric venular microvessels demonstrated that both NO-dependent and ANP-dependent mechanisms increased microvessel permeability (12, 20, 27, 28). Thus, in these vessels, increased cGMP was associated with an increased permeability but was not associated with a decrease in permeability. Furthermore, in these vessels, an increase in permeability is often associated with increased Ca2+ influx into the endothelial cells forming the vessel wall (9, 11, 13, 15). However, no studies have been conducted to investigate the endothelial cell [Ca2+]i when cGMP was directly increased in intact microvessels.
Our current study was designed to investigate the mechanisms whereby elevated cGMP in endothelial cells modulates microvessel permeability under basal and activated conditions using intact individually perfused venular microvessels in both frog and rat mesenteries. Changes in microvessel permeability were studied by measuring hydraulic conductivity (Lp) when single perfused microvessels were exposed to a membrane-permeable analog of cGMP in the presence and absence of inflammatory stimuli. To distinguish the mechanisms whereby cGMP regulates microvessel permeability by modifying Ca2+ influx from those regulating processes downstream from Ca2+ entry, changes in endothelial cell [Ca2+]i were measured under the same experimental conditions as those used for measurements of Lp when cGMP levels were changed. This experimental approach enabled the role of cGMP to be examined in intact microvessels that have normal permeability properties. Furthermore, the single vessel perfusion techniques enabled the direct effect of cGMP on permeability to be studied separately from other hemodynamic effects that can change capillary surface area and hydrostatic pressure. Experiments were carried out in both frog and rat mesenteric microvessels, which provided a direct comparison between the permeability regulation mechanisms in these two species.
| |
MATERIALS AND METHODS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Animal Preparation
The experiments were carried out in frog and rat mesenteries. Frogs (male Rana pipiens, 2.5-3.0 in.) were supplied by J. M. Hazen, and rats (female Sprague-Dawley, 250-300 g) were supplied by Hilltop Laboratory Animals (Scottdale, PA). The brain of the frog was destroyed by pithing while leaving the spinal cord intact. Preparation of frog mesenteries for measurement of Lp has been well described (5). For measurement of [Ca2+]i, the frog tray and frog preparation were modified to accommodate the shorter working distance of the lens used for fluorescence intensity (FI) measurements, which have been described in detail (15, 13). The tray for the rat experiment was extended to fit the size of the rat body and mesentery. Rats (age 3-4 mo) were anesthetized with pentobarbital sodium (65 mg/kg body wt) given subcutaneously and were kept warm on a heating pad. Anesthesia was maintained by giving additional pentobarbital sodium (3 mg/dose) as needed. After the rat was anesthetized, a midline surgical incision (1.5-2 cm) was made in the abdominal wall, and the mesentery was gently taken out from the cavity and spread over a pillar for Lp measurements or over a glass coverslip on the tray for [Ca2+]i measurements. The upper surface of the rat mesentery was continuously superfused with mammalian Ringer solution during preparation and experimentation. The temperature of the superfusate was maintained at 18-20°C for frogs and 35-37°C for rats. All of the experiments were carried out in venular microvessels, which were classified as segments in which there is convergent flow, one or two branches distal to true capillaries. Measurements of [Ca2+]i were carried out in the same type of venular microvessels used for Lp measurements. All of the vessels selected for cannulation had brisk blood flow and were free of white cells sticking or rolling along the vessel wall.Measurement of [Ca2+]i
Endothelial [Ca2+]i was measured using the Ca2+-sensitive fluorescent indicator fura 2. The methods used to measure [Ca2+]i in endothelial cells forming the walls of individually perfused microvessels and to conduct the in vitro calibration have been described in detail (13, 15). In brief, venular microvessels in frog or rat mesentery were cannulated and perfused with an albumin-Ringer solution containing 10 µM fura 2-AM for 30-45 min in the dark. After loading, the vessel was recannulated and perfused with albumin-Ringer solution for 10 min to remove all of the fura 2-AM from the vessel lumen and tissue. The vessel was aligned on the optical axis of a Nikon 300 diaphot inverted microscope. The excitation wavelengths for fura 2 were selected by two narrow-band interference filters (340 ± 5 and 380 ± 5 nm; Oriel), and the emission was separated with a dichroic mirror (DM400) and a wide-band interference filter (500 ± 35 nm; Oriel). FI was collected by a dry Nikon Fluor lens (×20, numerical aperture 0.75) from a measuring window (150 × 50 µm) that was placed ~100 µm downstream from the cannulation site of the vessel. In previous studies using a three-dimensional image reconstruction, the cells loaded with fura 2 in single perfused frog mesenteric microvessels were identified as endothelial cells by labeling the endothelial cell surface with fluorescent cationic ferritin (22). Thus the contribution of other types of cells forming the microvessel wall to the fluorescent signals under our experimental conditions was shown to be small. We estimated that the light collected in the measuring window sampled the fluorescent signals from 50 to 100 endothelial cells forming the segment of the vessel wall. The excitation wavelength alternated between 340 and 380 nm every 5 s, and values of FI at 340 nm (FI340) and at 380 nm (FI380) were measured with a 0.25-s exposure at each wavelength. At the end of each experiment, the microvessel was superfused with a modified Ringer solution (5 mM Mn2+ without Ca2+) while being perfused with the same solution containing ionomycin (10 µM). This procedure bleached the Ca2+-sensitive form of fura 2. The background FI due to unconverted fura 2-AM and other Ca2+-insensitive forms of fura 2 were recorded at both wavelengths (B380 and B340) and subtracted from FI340 and FI380. Ca2+ concentrations were estimated by calculating the ratio of FI values [Rexp = (FI340
B340)/(FI380 B380)] and then
comparing the normalized ratio (R = Rexp/Rmin)
with an in vitro calibration curve. Calibration conducted
at 20°C was used for frog experiments, and calibration at 37°C
was used for rat experiments. Rmax and Rmin are
the in vitro ratios of the values of
FI340 and
FI380 at saturating and zero
Ca2+ concentrations, respectively,
after the background fluorescence is corrected. The normalization by
Rmin, measured at the end of each
experiment, is to compensate for possible variations in the lamp
spectrum on different days. To interpret in vivo values of R in terms
of
[Ca2+]i,
we used the relation
[Ca2+]i = K[(R 0.85Rmin)/(0.85Rmax R)], where the factor 0.85 corrects for the
observation that Rmin and
Rmax are smaller within cells than
in nonviscous calibration solutions, and where K = KdIf0/Ifs, where
Kd is an effective dissociation constant and
If0 and Ifs are the values of FI380
for zero Ca2+ and for saturating Ca2+,
respectively.
Measurement of Lp of the Microvessel Wall
All measurements were based on the modified Landis technique, which measures the volume flux of water across the microvessel wall after occlusion of the vessel. The assumptions and limitations of the measurement have been evaluated in detail (5). The initial transcapillary water flow per unit area of the capillary wall (Jv/S)0 was measured at a given pressure (20-30 cmH2O for frog microvessels and 30-60 cmH2O for rat microvessels) within each vessel. Microvessel Lp was calculated as the slope of the relation between (Jv/S)0 and pressure.Solutions
Ringer solutions and albumin-Ringer perfusates. Frog or mammalian Ringer solutions were used for dissecting mesenteries, superfusing tissue, and preparing the perfusion solutions. The frog Ringer solution was prepared as described for all previous experiments (5). The composition of the mammalian Ringer solution was (in mM) 132 NaCl, 4.6 KCl, 2 CaCl2, 1.2 MgSO4, 5.5 glucose, 5.0 NaHCO3, and 20 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid HEPES and Na-HEPES. The pH of Ringer solution was maintained at 7.40-7.45 by adjusting the ratio of Na-HEPES to HEPES. An albumin-Ringer perfusate was prepared by adding bovine serum albumin (BSA), 1 g/100 ml, to Ringer solution. All perfusates used for control and test perfusion contained BSA (1 g/100 ml).Fluorescent dye and test solutions. A stock solution (10 mM) of fura 2-AM (Molecular Probes) was prepared with 100% dry dimethyl sulfoxide (DMSO). The final concentration of fura 2-AM in the loading solution was 10 µM and DMSO was 0.1% (vol/vol). The GC inhibitor 6-anilino-quinoline-5,8-quinone (LY-83583; Calbiochem) was first dissolved in 95% ethanol at a concentration of 10 mM, followed by a 1:1,000 dilution with albumin-Ringer solution. ATP, bradykinin, and 8-bromo-cGMP (8-BrcGMP) were purchased from Sigma, and 8-(p-chlorophenylthio)cGMP (8-pCPT-cGMP) was from BioLog Life Science Institute (La Jolla, CA). All test solutions were freshly prepared in albumin-Ringer solution before each cannulation.
Data analysis and statistics. All values in the text are means ± SE, except where noted otherwise. Changes in Lp were expressed as the ratio of testing Lp versus control Lp. For statistical comparisons, data obtained from the same vessel were analyzed by paired t-test and nonparametric Wilcoxon signed rank test. Fisher's protected least significant difference test was conducted for data comparison among groups. The significance level was P < 0.05.
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Effect of cGMP Analog 8-BrcGMP on Basal Lp and Endothelial [Ca2+]i in Frog Mesenteric Microvessels
The membrane-permeable cGMP analog 8-BrcGMP at a concentration of 2 mM caused a transient increase in Lp that lasted 2-4 min and then fell toward control levels. Figure 1A shows the time course of changes in Lp in a single experiment. Figure 1B summarizes the results from 23 microvessels. The mean control Lp when perfused with albumin-Ringer solution was 4.2 ± 0.5 × 10
7
cm · s1 · cmH2O1
(range from 1.2 to 9.4 × 107
cm · s1 · cmH2O1).
After 8-BrcGMP was added to the perfusate, the mean of the ratio of the
peak Lp value
within 1-2 min relative to control Lp (calculated
from individual vessel) was 5.3 ± 0.5. The peak Lp in each vessel
was significantly elevated from the control (P < 0.05). After 5 min, the
Lp of 8 of the 23 vessels remained elevated at 1.5-2 times the control value,
whereas the Lp of
the remainder of the vessels fell back to the control levels.
|
In contrast to the effect of 8-BrcGMP on basal Lp, 8-BrcGMP had no effect on basal endothelial cell [Ca2+]i. The mean values of endothelial [Ca2+]i measured before and after exposure to 8-BrcGMP (2 mM) in 10 microvessels were 50 ± 4 and 52 ± 4 nM, respectively (P > 0.05). Figure 2A shows a single vessel experiment, and Fig. 2B summarizes the results from 10 microvessels.
|
Effect of 8-BrcGMP on ATP-Induced Increase in Lp and Endothelial Cell [Ca2+]i in Frog Mesenteric Microvessels
The effect of 8-BrcGMP on ATP-induced increase in Lp was tested in 13 microvessels. The time course of the Lp changes in a single experiment is shown in Fig. 3A. The mean value of control Lp of 13 microvessels was 4.9 ± 0.6 × 107
cm · s1 · cmH2O1.
After ATP (10 µM) was added to the perfusate,
Lp increased to a
mean peak value 7.0 ± 0.9 times that of the control. After ATP was
washed from the vessel lumen for 10 min, the mean
Lp with albumin-Ringer perfusion (the second control) was 4.8 ± 0.7 × 107
cm · s1 · cmH2O1,
which was not significantly different from the initial control value
(P > 0.05). When each
vessel was then exposed to ATP in the presence of 8-BrcGMP, the
ATP-induced Lp
increase was further raised significantly in 9 of the 13 vessels. The
mean value of the ratio peak
Lp/second control
Lp
with both ATP and 8-BrcGMP present was 12.4 ± 1.9, which was
significantly higher compared with the
Lp increase with
ATP alone.
|
To test whether the effect of cGMP on the magnitude of the peak response to ATP might be due to the order of exposure, we first measured the changes in Lp in the presence of both 8-BrcGMP and ATP and then with ATP alone in another five microvessels. The mean value of the peak Lp increase in the presence of both ATP and 8-BrcGMP relative to control in each vessel was 10.0 ± 3.2. The mean of the peak increase in Lp in response to ATP alone relative to the second control was 6.0 ± 1.0. The magnitude of the peak increase in Lp in response to ATP augmented by the presence of 8-BrcGMP was not significantly different from the results when ATP was tested first as described above. Figure 3B summarizes the results from two groups of experiments.
The effect of 8-BrcGMP on the ATP-induced increase in [Ca2+]i was tested in 8 of the 10 microvessels in which the effect of 8-BrcGMP on baseline endothelial [Ca2+]i was also tested. Figure 4A shows the changes in [Ca2+]i as a function of time in a single experiment. Figure 4B summarizes the results from eight microvessels. The mean baseline endothelial [Ca2+]i was 53 ± 4 nM. ATP (10 µM) induced a transient increase in endothelial [Ca2+]i. The mean peak value was 217 ± 15 nM. Each vessel was then perfused with albumin-Ringer perfusate for 10 min, followed by perfusion of 8-BrcGMP (2 mM) for another 10 min. The mean [Ca2+]i after albumin-Ringer perfusion was 51 ± 4 nM and showed no significant changes during 8-BrcGMP perfusion. Next, each vessel was exposed to ATP and 8-BrcGMP. The mean peak increase in endothelial [Ca2+]i was 217 ± 12 nM, which was not significantly different from that with exposure to ATP alone (P > 0.05).
|
Effect of GC Inhibitor LY-83583 on Baseline and ATP-Induced Increase in Endothelial [Ca2+]i
Figure 5A shows the time course and magnitude of changes in [Ca2+]i when a single perfused microvessel was exposed to ATP or LY-83583 alone and to LY-83583 together with ATP. The experiments were conducted in four microvessels. The mean baseline [Ca2+]i was 39 ± 4 nM. ATP (10 µM) induced a transient increase in [Ca2+]i. The mean peak value was 265 ± 44 nM. After washing out ATP with albumin-Ringer solution for 10 min, the mean [Ca2+]i was 46 ± 4 nM. Each vessel was then exposed to LY-83583 (10 µM). There was no significant change in [Ca2+]i during the 10-min exposure. The mean value of [Ca2+]i was 47 ± 6 nM. When each vessel was exposed to ATP and LY-83583, [Ca2+]i transiently increased to a mean peak value of 266 ± 48 nM and fell toward control values in 5-10 min. The magnitude and the time course of the changes in [Ca2+]i in response to ATP in the presence of LY-83583 were not significantly different from that without LY-83583. Figure 5B summarizes results for all vessels.
|
Effects of cGMP Analogs on Basal Lp and Endothelial [Ca2+]i in Rat Mesenteric Microvessels
To test whether the effects of cGMP analogs on basal Lp and endothelial cell [Ca2+]i in mammalian venular microvessels were similar to those in frog microvessels, two cGMP analogs, 8-BrcGMP and 8-pCPT-cGMP, were used to measure changes in basal Lp in rat mesenteric venular microvessels.Both 8-BrcGMP and 8-pCPT-cGMP induced a transient increase in
Lp. In 10 rat
mesenteric microvessels, the baseline
Lp measured with
albumin-Ringer perfusion was 2.9 ± 0.8 × 107
cm · s1 · cmH2O1.
The mean value of the peak
Lp increase
relative to control after exposure to 8-BrcGMP (2 mM) was 2.8 ± 0.4 times the control values. The magnitude of the
Lp increase was
smaller than the 5.3 ± 0.5-fold increase in
Lp in frog
mesenteric microvessels. However, the transient increase in
Lp in rats lasted
3-8 min before falling toward control levels in comparison with
1-2 min in frog microvessels. We also measured changes in
Lp in another 12 rat mesenteric microvessels using a highly membrane-permeant and
metabolically resistant cGMP analog, 8-pCPT-cGMP. The mean baseline
Lp of 12 vessels
was 3.0 ± 0.6 × 107
cm · s1 · cmH2O1.
After each vessel was exposed to 8-pCPT-cGMP (2 mM), the mean of the
peak Lp relative
to control was 3.8 ± 0.5. The time course of the changes in
Lp in a single
experiment is shown in Fig.
6A. The
summarized results of both analogs are shown in Fig.
6B.
|
Measurements of endothelial cell [Ca2+]i in rat mesenteric venular vessels showed no significant changes in basal endothelial cell [Ca2+]i after exposure to either 8-BrcGMP or 8-pCPT-cGMP at 2 mM. The mean endothelial cell [Ca2+]i before and after exposure to 8-BrcGMP in five microvessels was 80 ± 4 and 78 ± 6 nM, respectively. In another four microvessels, the mean [Ca2+]i before and after exposure to 8-pCPT-cGMP was 68 ± 8 and 71 ± 8 nM, respectively. These results are summarized in Fig. 8B along with other [Ca2+]i measurements.
Effect of cGMP on Bradykinin-Induced Increase in Lp in Rat Mesenteric Microvessels
In rat mesenteric microvessels, both cGMP analogs, 8-BrcGMP and 8-pCPT-cGMP, potentiated the increases in Lp induced by exposure to bradykinin. In contrast, inhibition of GC using LY-83583 attenuated the bradykinin-induced increase in Lp in rat mesenteric microvessels. In 11 rat mesenteric venular microvessels, the mean baseline Lp was 1.8 ± 0.2 × 107
cm · s1 · cmH2O1.
Bradykinin (1 nM) caused a significant transient increase in Lp. The mean
value of the peak
Lp after exposure
to bradykinin relative to control was 4.8 ± 0.4 (range from 2.3 to
6.5 times the control value). Repeated responses to bradykinin in the
same vessel were tested in 4 of 11 vessels. The mean peak value of the
first and second exposures to bradykinin was 5.4 ± 0.3 and 4.8 ± 0.2 times the control, respectively. The differences in magnitude
between the two exposures were not statistically significant (P > 0.05).
The effect of 8-BrcGMP on bradykinin-induced increases in
Lp was tested in
eight microvessels. The mean control
Lp was 1.8 ± 0.2 × 107
cm · s1 · cmH2O1.
When each vessel was exposed to bradykinin in the presence of 8-BrcGMP
(2 mM), Lp
transiently increased to a peak value of 8.3 ± 1.3 times the
control within 2 min and then fell toward the control level in
10-15 min. Next, each vessel was perfused with albumin-Ringer
perfusate for 15 min to wash out the testing agents. When each vessel
was reexposed to bradykinin alone, the peak increase in
Lp was 4.8 ± 0.5 times the control, which was significantly smaller than the
Lp value when
8-BrcGMP was present, but was not significantly different from the mean
peak value obtained from 11 vessels exposed to bradykinin alone, as
described above. Figure 7A shows
the changes in Lp
in a single experiment.
|
Another cGMP analog, 8-pCPT-cGMP, demonstrated a similar effect on the
bradykinin-induced increase in
Lp. The mean
control Lp in
another seven microvessels was 2.0 ± 0.3 × 107
cm · s1 · cmH2O1.
When each vessel was exposed to bradykinin in the presence of 2 mM of
8-pCPT-cGMP, the mean peak
Lp increased to
7.5 ± 0.7 times the control value. The transient response lasted
8-15 min. A second exposure to bradykinin alone was conducted in
four of the seven vessels. The peak increase in
Lp was 4.3 ± 0.6 times the control. The magnitude of the mean peak increase in
Lp in response to
bradykinin in the presence of either 8-BrcGMP or 8-pCPT-cGMP was
significantly higher than the
Lp increase with
bradykinin alone, but the results obtained for the two cGMP analogs
were not significantly different.
The effect of the GC inhibitor LY-83583 on the bradykinin-induced
increase in Lp
was tested in another six microvessels. The mean control
Lp was 1.9 ± 0.3 × 107
cm · s1 · cmH2O1.
LY-83583 alone caused no significant changes in
Lp. The mean value of Lp with
LY-83583 present versus control was 1.3 ± 0.2 (P > 0.05). However, when each
vessel was exposed to bradykinin in the presence of LY-83583, the
magnitude of the increase in Lp in response to
bradykinin was significantly reduced. The peak increase was only 2.6 ± 0.5 times the control value compared with a 4.8-fold increase
with bradykinin alone from 11 microvessels (P < 0.05). Figure
7B summarizes all of the results in
three groups of studies described above.
Effect of 8-BrcGMP on Bradykinin-Induced Increase in Endothelial Cell [Ca2+]i in Rat Mesenteric Microvessels
To test whether the augmented effect of cGMP analogs on the bradykinin-induced increase in Lp was associated with a corresponding increase in endothelial cell [Ca2+]i, we measured changes in endothelial cell [Ca2+]i in response to both 8-BrcGMP and bradykinin and to bradykinin alone in five rat mesenteric microvessels. Figure 8A shows the magnitude and time course of changes in endothelial [Ca2+]i in response to bradykinin with and without 8-BrcGMP present from a single experiment. In three of the five microvessels, the changes in endothelial [Ca2+]i in response to bradykinin alone were measured first. Next, each vessel was perfused with albumin-Ringer solution for 10 min to wash out bradykinin from the vessel lumen and restore the baseline [Ca2+]i. In addition, in the same vessel, endothelial cell [Ca2+]i was measured after each vessel was exposed to 8-BrcGMP, followed by exposure to 8-BrcGMP and bradykinin. In another two microvessels, the changes in [Ca2+]i in response to 8-BrcGMP and 8-BrcGMP plus bradykinin were measured first and then with bradykinin alone. The magnitude and time course of the changes in [Ca2+]i with and without cGMP present were not significantly different. Alternating the experimental sequence did not change the results. The mean baseline endothelial [Ca2+]i of five vessels was 74 ± 3 nM. After exposure to bradykinin, [Ca2+]i increased to a mean peak value of 339 ± 81 nM. After the first stimulating agents were washed out for 10 min, the baseline [Ca2+]i fell back to 80 ± 4 nM. The mean peak value of [Ca2+]i after exposure to 8-BrcGMP and bradykinin was 369 ± 102 nM, which was not significantly different from the mean values measured with bradykinin alone. Figure 8B summarizes the results.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Our experiments demonstrated a consistent role of elevated cGMP in the modulation of basal and activated microvessel permeability in intact individually perfused venular microvessels in both frog and rat mesenteric microvessels. There are two important new findings. First, we found that elevation of cGMP levels using cGMP analogs transiently increased basal Lp without changing basal endothelial cell [Ca2+]i. This is the second example we have described in which permeability can increase without a corresponding increase in endothelial cell [Ca2+]i in intact microvessels. The first case was that inhibition of NO production under basal conditions caused an increase in Lp without changing endothelial cell [Ca2+]i (14). The second important finding is that the use of cGMP analogs potentiated the agonist-induced increase in Lp. We also showed that experimental conditions that decreased levels of cGMP by inhibition of GC using LY-83583 attenuated the agonist-induced increase in Lp. However, changes in endothelial cell [Ca2+]i measured under the same experimental conditions demonstrated that neither elevated levels of cGMP nor suppression of GC to reduce cGMP levels modified the magnitude and time course of the agonist-induced transient increase in endothelial cell [Ca2+]i. These results suggested that an elevation of cGMP levels in endothelial cells is a necessary step to increase microvessel permeability in intact microvessels, and this regulatory process occurs downstream from Ca2+ influx.
Actions of cGMP on Modulation of Microvessel Permeability Occur Downstream From Ca2+ Influx
We previously demonstrated that the magnitude of the initial increase in [Ca2+]i in endothelial cells forming microvessel walls determined the magnitude of the initial increases in venular microvessel permeability, and the initial increase in [Ca2+]i was due mainly to Ca2+ influx via a passive conductive channel (9, 11, 13, 15). A clear dissociation between an increase in endothelial cell [Ca2+]i and microvessel permeability in intact microvessels was demonstrated when single perfused microvessels were exposed to the cAMP analog or NO synthase (NOS) inhibitors in the presence of either ionomycin or ATP (10, 12). In both cases, increased [Ca2+]i in response to an agonist was unchanged, but the increased permeability was attenuated or abolished. Those results indicated that an increase in endothelial [Ca2+]i was a necessary signaling mechanism to increase microvessel permeability, but other factors may modulate the changes in permeability downstream from Ca2+ entry. Our current results that cGMP levels modulated agonist-induced increases in permeability without modifying agonist-induced increases in endothelial cell [Ca2+]i provided further support for a cascade of mechanisms that modulate microvessel permeability downstream from Ca2+ influx.Furthermore, the fact that elevated cGMP by cGMP analogs increased basal Lp without an increase in basal endothelial cell [Ca2+]i indicated that, in addition to mechanisms that increase permeability after endothelial [Ca2+]i is increased, there are also processes that may bypass the initial Ca2+ mobilization and elicit an increase in permeability without changing endothelial cell [Ca2+]i. Another example in which permeability can increase without an increase in endothelial cell [Ca2+]i is the suppression of basal NO formation using NOS inhibitors (14). The mechanism in this case appears to involve an increase in oxidative stress due to decreased basal NO production.
A unique feature of our experimental approach is that the permeability was studied under the same experimental conditions as endothelial cell [Ca2+]i was measured. Our current experiments provided the first evidence in vivo to evaluate the relation between the actions of cGMP in the modulation of microvessel permeability and that in endothelial cell [Ca2+]i. Specifically, the mechanism whereby increased cGMP modulates microvessel permeability does not involve modification of the initial increase in endothelial cell [Ca2+]i. This result is different from that reported in human aortic endothelial cells in culture (7) and in smooth muscle cells (25).
Comparison Between Species
The experiments in both frog and rat mesenteric microvessels demonstrated that there were no fundamental differences in the cGMP-dependent regulatory processes for permeability between microvessels in rats and frogs, although differences in magnitude and time course of the changes in Lp after exposure to 8-BrcGMP existed. In frog microvessels, after exposure to 8-BrcGMP, the mean peak increase in Lp was 5.2 times the control, and Lp fell toward control values in 5 min, whereas, in rat mesenteric venular microvessels, the mean peak increase in Lp in response to 8-BrcGMP was 2.8 times the control, but Lp fell back to control values in 8-10 min. A similar magnitude difference also existed on the further increase in agonist-induced hyperpermeability by cGMP analogs between these two species. We also note that the averaged basal Lp value in frog mesenteric microvessels is ~5 × 107
cm · s1 · cmH2O1
(4), whereas the averaged basal
Lp value in rat
venular microvessels is ~2 × 107
cm · s1 · cmH2O1.
This mean Lp
value in rats is consistent with another study using the same
techniques in rat mesenteries (17). The
Lp increase in
response to 8-pCPT-cGMP (2 mM) in rat microvessels was higher (3.8-fold) than that for 8-BrcGMP (2 mM), which might be attributed to
its higher lipophilicity and relatively metabolic resistance, as
reported by Butt et al. (3).
Comparison With Other Studies
Our results are consistent with other in vivo studies using isolated porcine venular microvessels and intact individually perfused microvessels in frog mesenteries, demonstrating that elevation of cGMP in endothelial cells by different agents increases microvessel permeability. For instance, in isolated porcine coronary venules, both histamine- and vascular endothelial growth factor (VEGF)-induced hyperpermeability were suppressed by NOS inhibitors and the inhibitor of GC (27, 28). In intact frog mesenteric microvessels, the NOS inhibitor attenuated substance P-induced increases in microvessel Lp (21). Our previous study also demonstrated that ATP-induced increases in Lp were attenuated by NO synthase inhibitors and the GC inhibitor in intact frog mesenteric microvessels (12). Our current results showing that the bradykinin-induced increase in Lp was attenuated by LY-83583 in rat venular microvessels are consistent with the previous studies. It has been demonstrated that both ATP and bradykinin enhance cGMP production in cultured endothelial cells (8). Although no directly measured levels of cGMP or NO are available in in vivo studies, these studies together provide strong evidence indicating that the NO-cGMP-dependent pathway in endothelial cells might be a common pathway to increase microvessel permeability after exposure to inflammatory agents.In addition, elevation of cGMP by activating the membrane-bond particulate form GC using ANP also caused an increase in Lp in frog mesenteric microvessels (20). However, the question of whether the two sources of cGMP generated by different forms of GC have the same downstream pathway remains to be studied. Although it has been consistently demonstrated that cGMP increases microvessel permeability in isolated or individually perfused microvessels, we noted that, in some of the whole organ studies, increased cGMP caused a protective effect on microvessel permeability, such as demonstrated in a feline small intestine preparation (18). The reasons for this difference need further investigation but may involve the interaction between leukocyte and endothelial cells.
In contrast to in vivo studies, the majority of studies in vitro using cultured large-vessel endothelial cells demonstrated opposite results. Evidence showed that an increased cGMP formation by a cGMP analog induced reductions in both the thrombin-enhanced permeability and the rise of [Ca2+]i in human aortic endothelial cells but had no effect on thrombin-induced Ca2+ accumulation in human umbilical vein endothelial cells (7). A similar effect of increased cGMP to decrease agonist-induced permeability increase has been demonstrated in human umbilical and pulmonary artery endothelial cell monolayers (7, 26), in freshly isolated porcine pulmonary artery endothelial cells (23), and in bovine aortic and pulmonary artery endothelial cells (1), although there are exceptions (16).
Our results emphasize that the mechanisms of the action of cGMP to modulate the permeability of cultured endothelial cell barriers should not be extrapolated to all endothelial cells, especially to those intact microvascular endothelial cells. Furthermore, the differences between our results and those from large-vessel endothelium in culture cannot be attributed to differences between species.
Possible Mechanisms of Microvessel Permeability Modulation by Elevated cGMP
Our results indicated that cGMP does not regulate microvessel permeability by regulating endothelial cell [Ca2+]i. Thus the action of cGMP in venular microvessels is quite distinct from the mechanisms whereby NO-cGMP-dependent processes relax smooth muscle cells (25) and also from those in vitro studies using large-vessel endothelial cells (7).The downstream targets for increased levels of cGMP to regulate microvessel permeability may be either a cGMP-dependent protein kinase (cGK) or cGMP-dependent phosphodiesterases, which may modify cAMP levels. The existence of cGK has been demonstrated in bovine aortic endothelial cells, in rat pulmonary microvascular endothelial cells, and in human aortic and pulmonary artery endothelial cells as well as in foreskin microvessel endothelial cells, but low expression was found in human umbilical vein endothelial cells (6, 7, 19). The presence of cGK in endothelial cells suggested a potential mechanism for a cGMP-mediated event in these cells, which might involve the modification of cytoskeleton in endothelial cells (24). Draijer et al. (7) attributed the action of cGMP to decrease Ca2+ influx to cGK because it was present in the cells where Ca2+ influx was reduced but not in those where there was no effect on Ca2+ influx. However, there may be other cGMP-dependent kinase actions to change permeability. Wu et al. (27) demonstrated that an inhibitor of cGK prevented VEGF-induced hyperpermeability in isolated porcine coronary venules. Their results indicated that VEGF modulates microvascular permeability via a signaling cascade involving NO synthesis, GC stimulation, and cGK activation. Identification of functional cGK in venular microvessel endothelial cells is needed to further confirm this signaling pathway. In addition, cGMP may exert its function via modifying cAMP levels through a number of specific phosphodiesterases, which may cause a decrease or an increase of cAMP levels in response to cGMP. For example, a cGMP-stimulated phosphodiesterase that reduces cAMP levels may account for the results in venular microvessels. We note, however, that our observation that 8-pCPT-cGMP, a selective activator of cGMP-dependent protein kinase, gave results similar to those using 8-BrcGMP may indicate that the cGMP-induced decrease in cAMP is not likely to be the dominating mechanism. The validity of this conclusion rests on the specificity of the agent in our preparation, so we would rather not rule out this possibility at this point. Further study using more specific phosphodiesterase inhibitors might provide more definitive information.
Another possible mechanism we cannot exclude at present is that the action of cGMP in the modulation of microvessel permeability may involve other cellular components of the microvessel wall. In frog mesentery, ~40% of microvessel endothelium is ensheathed by pericytes (2). In mammals, cGK has a very high expression in these smooth muscle-type cells. Therefore, a cGMP-induced modification of pericyte contraction might be another possible mechanism that regulates microvessel permeability.
In summary, our experiments demonstrated that elevation of cGMP levels using cGMP analogs transiently increased basal Lp and potentiated the agonist-induced increase in Lp in both frog and rat mesenteric microvessels, whereas inhibition of the GC using LY-83583 attenuated the bradykinin-induced increase in Lp in rat mesenteric microvessels. Changes in endothelial cell [Ca2+]i measured under the same experimental conditions demonstrated that neither elevated levels of cGMP nor suppression of GC to reduce cGMP levels had effects on basal endothelial cell [Ca2+]i and the transient increases in endothelial cell [Ca2+]i induced by ATP or bradykinin. These results suggested that an elevation of cGMP levels in endothelial cells is a necessary step to increase microvessel permeability in intact microvessels and that the regulatory action of cGMP in venular endothelial cells occurs downstream from Ca2+ influx. Our results differ from those reported in large-vessel endothelium in culture and in vascular smooth muscle cells in which cGMP modulates Ca2+ influx into these cells.
| |
ACKNOWLEDGEMENTS |
|---|
This work is supported by American Heart Association National Center Grant-In-Aid 96011510 (P. He) and by National Heart, Lung, and Blood Institute Grants HL-56237 (P. He) and HL-28607 (F. E. Curry).
| |
FOOTNOTES |
|---|
Address for reprint requests: P. He, Dept. of Human Physiology, School of Medicine, Univ. of California Davis, One Shields Ave., Davis, CA 95616.
Received 1 December 1997; accepted in final form 17 February 1998.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
1.
Buchan, K. W.,
and
W. Martin.
Modulation of barrier function of bovine aortic and pulmonary artery endothelial cells: dissociation from cytosolic calcium content.
Br. J. Pharmacol.
107:
932-938,
1992[Medline].
2.
Bundgaard, M.,
and
J. Frokjaer-Jesen.
Functional aspects of the ultrastructure of terminal blood vessels: a quantitative study on consegments of the frog mesenteric microvasculature.
Microvasc. Res.
23:
1-30,
1982[Medline].
3.
Butt, E.,
C. Nolte,
S. Schulz,
J. Beltman,
J. A. Beavo,
B. Jastorff,
and
U. Walter.
Analysis of the functional role of cGMP-dependent protein in intact human platelets using a specific activator 8-para-chlorophenylthio-cGMP.
Biochem. Pharmacol.
43:
2591-2600,
1992[Medline].
4.
Curry, F. E.
Modulation of venular microvessel permeability by calcium influx into endothelial cells.
FASEB J.
6:
2456-2466,
1992[Abstract].
5.
Curry, F. E.,
V. H. Huxley,
and
I. H. Sarelius.
Techniques in microcirculation: measurement of permeability, pressure and flow.
In: Cardiovascular Physiology. Techniques in the Life Sciences, edited by R. T. Linden. New York: Elsevier, 1983, vol. P309/1, p. 1-34.
6.
Diwan, A. H.,
W. J. Thompson,
A. K. Lee,
and
S. J. Strada.
Cyclic GMP-dependent protein kinase activity in rat pulmonary microvascular endothelial cells.
Biochem. Biophys. Res. Commun.
202:
728-735,
1994[Medline].
7.
Draijer, R.,
D. E. Atsma,
A. van der Laarse,
and
V. W. van Hinsbergh.
cGMP and nitric oxide modulate thrombin-induced endothelial permeability. Regulation via different pathways in human aortic and umbilical vein endothelial cells.
Circ. Res.
76:
199-208,
1995
8.
Gosink, E. C.,
and
E. J. Forsberg.
Effects of ATP and bradykinin on endothelial cell Ca2+ homeostasis and formation of cGMP and prostacyclin.
Am. J. Physiol.
265 (Cell Physiol. 34):
C1620-C1629,
1993
9.
He, P.,
and
F. E. Curry.
Depolarization modulates endothelial cell calcium influx and microvessel permeability.
Am. J. Physiol.
261 (Heart Circ. Physiol. 30):
H1246-H1254,
1991
10.
He, P.,
and
F. E. Curry.
Differential actions of cAMP on changes in microvessel permeability and cell calcium after exposure to ATP.
Am. J. Physiol.
265 (Heart Circ. Physiol. 34):
H1019-H1023,
1993
11.
He, P.,
and
F. E. Curry.
Endothelial cell hyperpolarization increases [Ca2+]i and venular microvessel permeability.
J. Appl. Physiol.
76:
2288-2297,
1994
12.
He, P.,
B. Liu,
and
F. E. Curry.
Effect of nitric oxide synthase inhibitors on endothelial [Ca2+]i and microvessel permeability.
Am. J. Physiol.
272 (Heart Circ. Physiol. 41):
H176-H185,
1997
13.
He, P.,
S. N. Pagakis,
and
F. E. Curry.
Measurement of cytoplasmic calcium in single microvessels with increased permeability.
Am. J. Physiol.
258 (Heart Circ. Physiol. 27):
H1366-H1374,
1990
14.
He, P.,
M. Zeng,
and
F. E. Curry.
Effect of nitric oxide synthase inhibitors on basal microvessel permeability and endothelial [Ca2+]i.
Am. J. Physiol.
273 (Heart Circ. Physiol. 42):
H747-H755,
1997
15.
He, P.,
X. Zhang,
and
F. E. Curry.
Calcium entry through conductive pathway modulates receptor-mediated increase in microvessel permeability.
Am. J. Physiol.
271 (Heart Circ. Physiol. 40):
H2377-H2387,
1996
16.
Holschermann, H.,
T. Noll,
A. Hempel,
and
H. M. Piper.
Dual role of cGMP in modulation of macromolecule permeability of aortic endothelial cells.
Am. J. Physiol.
272 (Heart Circ. Physiol. 41):
H91-H98,
1997
17.
Kendall, S.,
and
C. C. Michel.
The measurement of permeability in single rat venules using the red cell microperfusion technique.
Exp. Physiol.
80:
359-372,
1995[Abstract].
18.
Kubes, P.
Nitric oxide-induced microvascular permeability alterations: a regulatory role for cGMP.
Am. J. Physiol.
265 (Heart Circ. Physiol. 34):
H1909-H1915,
1993
19.
MacMillan-Crow, L. A.,
J. E. Murphy-Ullrich,
and
T. M. Lincoln.
Identification and possible localization of cGMP-dependent protein kinase in bovine aortic endothelial cells.
Biochem. Biophys. Res. Commun.
201:
531-537,
1994[Medline].
20.
Meyer, D. J.,
and
V. H. Huxley.
Capillary hydraulic conductivity is elevated by cGMP-dependent vasodilators.
Circ. Res.
70:
382-391,
1992
21.
Nguyen, L. S.,
A. C. Villablanca,
and
J. C. Rutledge.
Substance P increases microvascular permeability via nitric oxide-mediated convective pathways.
Am. J. Physiol.
268 (Regulatory Integrative Comp. Physiol. 37):
R1060-R1068,
1995
22.
Pagakis, S. N.,
F. E. Curry,
and
J. F. Lenz.
Application of image restoration and the three dimensional visualization techniques to frog microvessels in-situ loaded with fluorescent indicators.
In: Biomedical Image. Processing and Biomedical Visualization, edited by R. S. Acharya,
and D. B. Goldgof. San Jose, CA: SPIE, 1993, vol. 1905, p. 920-931.
23.
Suttorp, N.,
S. Hippenstiel,
M. Fuhrmann,
M. Krull,
and
T. Podzuweit.
Role of nitric oxide and phosphodiesterase isoenzyme II for reduction of endothelial hyperpermeability.
Am. J. Physiol.
270 (Cell Physiol. 39):
C778-C785,
1996.
24.
Vaandrager, A. B.,
and
H. R. de Jonge.
Signalling by cGMP-dependent protein kinases.
Mol. Cell. Biochem.
157:
23-30,
1996[Medline].
25.
Walsh, M. P.,
G. J. Kargacin,
J. Kendrick-Jones,
and
T. M. Lincoln.
Intracellular mechanisms involved in the regulation of vascular smooth muscle tone.
Can. J. Physiol. Pharmacol.
73:
565-573,
1995[Medline].
26.
Westendorp, R. G.,
R. Draijer,
A. E. Meinders,
and
V. W. van Hinsbergh.
Cyclic-GMP-mediated decrease in permeability of human umbilical and pulmonary artery endothelial cell monolayers.
J. Vasc. Res.
31:
42-51,
1994[Medline].
27.
Wu, H. M.,
Q. Huang,
Y. Yuan,
and
H. J. Granger.
VEGF induces NO-dependent hyperpermeability in coronary venules.
Am. J. Physiol.
271 (Heart Circ. Physiol. 40):
H2735-H2739,
1996
28.
Yuan, Y.,
H. J. Granger,
D. C. Zawieja,
D. V. DeFily,
and
W. M. Chilian.
Histamine increases venular permeability via a phospholipase C-NO synthase-guanylate cyclase cascade.
Am. J. Physiol.
264 (Heart Circ. Physiol. 33):
H1734-H1739,
1993
This article has been cited by other articles:
|
|
 |
|
  E. P. Schmidt, M. Damarla, O. Rentsendorj, L. E. Servinsky, B. Zhu, A. Moldobaeva, A. Gonzalez, P. M. Hassoun, and D. B. Pearse Soluble guanylyl cyclase contributes to ventilator-induced lung injury in mice Am J Physiol Lung Cell Mol Physiol, December 1, 2008; 295(6): L1056 - L1065. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Zhu and P. He Platelet-activating factor increases endothelial [Ca2+]i and NO production in individually perfused intact microvessels Am J Physiol Heart Circ Physiol, June 1, 2005; 288(6): H2869 - H2877. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Zhu, D. Schwegler-Berry, V. Castranova, and P. He Internalization of caveolin-1 scaffolding domain facilitated by Antennapedia homeodomain attenuates PAF-induced increase in microvessel permeability Am J Physiol Heart Circ Physiol, January 1, 2004; 286(1): H195 - H201. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. H. Adamson, M. Zeng, G. N. Adamson, J. F. Lenz, and F. E. Curry PAF- and bradykinin-induced hyperpermeability of rat venules is independent of actin-myosin contraction Am J Physiol Heart Circ Physiol, June 5, 2003; 285(1): H406 - H417. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zeng, H. Zhang, C. Lowell, and P. He Tumor necrosis factor-alpha -induced leukocyte adhesion and microvessel permeability Am J Physiol Heart Circ Physiol, December 1, 2002; 283(6): H2420 - H2430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Behrends, A. Mietens, J. Kempfert, M. Koglin, H. Scholz, and R. Middendorff The Expression Pattern of Nitric Oxide-sensitive Guanylyl Cyclase in the Rat Heart Changes During Postnatal Development J. Histochem. Cytochem., October 1, 2002; 50(10): 1325 - 1332. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Pocock, B. Williams, F. E. Curry, and D. O. Bates VEGF and ATP act by different mechanisms to increase microvascular permeability and endothelial [Ca2+]i Am J Physiol Heart Circ Physiol, October 1, 2000; 279(4): H1625 - H1634. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. He, J. Wang, and M. Zeng Leukocyte adhesion and microvessel permeability Am J Physiol Heart Circ Physiol, May 1, 2000; 278(5): H1686 - H1694. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. He, M. Zeng, and F. E. Curry Dominant role of cAMP in regulation of microvessel permeability Am J Physiol Heart Circ Physiol, April 1, 2000; 278(4): H1124 - H1133. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
