Nitric oxide modulates epithelial
in the feline small intestine
PAUL
KUBES
Department
of Medical Physiology, University
l
l
blood flow; NG-nitro-L-arginine-methyl
ester; L-argi-
BOWEL DISEASE, ischemic bowel disease, and other intestinal disorders are characterized by
leukocyte infiltration,
increased microvascular
permeability, and mucosal barrier dysfunction. The latter is of
major concern in the clinical setting, because a breakdown of the mucosal barrier can lead to bacterial translocation and ultimately
septic shock. The mechanism
that promotes a leaky mucosal barrier at the onset of
inflammation
is poorly understood; however, reactive
oxygen metabolites, vasocongestion, and vascular leukocytes have all been implicated as potential mediators (4).
Nitric oxide, recently purported to be endothelium-derived relaxing factor (l7), has been suggested to exhibit
beneficial effects when administered
as an exogenous
source in ischemia/reperfusion
of the intestine (1) as
well as after alcohol damage in the stomach (16), and
inhibitors
of nitric oxide synthesis including
NCmonomethyl+arginine
have been reported to exacerbate damage to the intestine during endotoxic shock (8).
These data raise the possibility that nitric oxide may be
an important
endogenous modulator of the sequel associated with inflammation
of the small bowel and its
INFLAMMATORY
G1138
0193~185’7/92
$2.00
Copyright
of Calgary, Calgary, Alberta T2N 4N1, Canada
inactivation
may contribute
to intestinal
dysfunction.
Because superoxide anion effectively inactivates nitric
oxide (7, 19), it is conceivable that inactivation
of
nitric oxide production might be a very important mechanism in the onset of intestinal inflammation.
The role of nitric oxide as a physiological regulator of
intestinal
function is poorly understood. There is evidence to suggest that nitric oxide plays an important
role in regulating blood flow in the splanchnic microcirculation (13). Moreover, recent evidence would also suggest a role for nitric oxide as a nonadrenergic
noncholinergic neurotransmitter
in the gut (2). However, its
role in regulating other intestinal functions has not been
assessed. The primary objective of this study was to test
the hypothesis that endogenous production
of nitric
oxide plays an important
role in the modulation
of permeability across the epithelial barrier. This was accomplished by monitoring
the transmucosal
flux of 51Crlabeled EDTA (51Cr-EDTA) as well as a macromolecule
(dextran with mol wt 17,200) before and after the administration
of 1) inhibitors
of nitric oxide production
and 2) exogenous sources of nitric oxide. The results
revealed that inhibition
of nitric oxide production
greatly increased epithelial permeability,
and so the second objective was to determine whether the increased
epithelial cell permeability
associated with inhibition
of
nitric oxide was mediated by a reduction in blood flow,
alterations in transmucosal fluid flux (from interstitium
to lumen), or vascular leukocytes. The data derived from
this study describe a new endogenous regulator of epithelial permeability,
whose inactivation
under pathophysiological
conditions may contribute to the impairment of mucosal barrier function.
METHODS
Experiments
were performed on 16 cats initially anesthetized
with ketamine HCl (75 mg im). The left jugular vein was cannulated, and anesthesia was maintained
with the use of pentobarbital sodium. A tracheotomy
was performed to maintain a
patent airway, and the animals were ventilated artificially.
The
experimental
procedure has been described previously in detail
(6). Briefly, a 45-75 g segment of small intestine was isolated
from the ligament of Treitz to the ileocecal valve; blood and
lymph vessels were maintained
intact. The remainder
of the
small and large intestine was extirpated. The ileal segment and
mesenteric pedicle were moistened with saline-soaked gauze and
covered with a clear plastic sheet to minimize evaporation
and
tissue dehydration.
The preparation
was maintained
at 38°C
using an infrared heat lamp.
The animals were given heparin (10,000 U) intravenously,
0 1992 the American
Physiological
Society
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Kubes, Paul. Nitric oxide modulates epithelial permeability
in the feline small intestine. Am. J. Physiol. 262 (Gustrointest.
Liver Physiol. 25): Gl138-Gll42,
1992.-The
objective of this
study was to assess whether inhibition
of nitric oxide production leads to increased epithelial
permeability
in feline small
intestine. Local intra-arterial
infusion of the nitric oxide synthesis inhibitor
NCI-nitro-L-arginine-methyl
ester (L-NAME;
0.025 pmol . ml-l. min-l)
was performed in autoperfused
segments of cat ileum for 90 min. An exogenous source of nitric
oxide, sodium
nitroprusside
(SNP)
was infused
(0.025
pmol. ml-l emin-I) for the last 30 min of the 90-min L-NAME
infusion. Epithelial
permeability
was quantitated
by measuring
blood-to-lumen
clearance of Wr-labeled
EDTA throughout
the
experiment.
An increase of approximately
sixfold in mucosal
permeability
was observed within 30 min of L-NAME
infusion
and this effect was completely
reversed by infusion of either
SNP or L-arginine
(0.125 pmol ml-l min-‘). NG-nitro-D-arginine-methyl
ester (D-NAME)
had no effect on mucosal
permeability.
The increase in epithelial permeability
was sufficiently large that rhodamine-dextran
(mol wt = 17,200) clearance from interstitium
to lumen was increased. Pretreatment
with IB4, a monoclonal
antibody directed against the leukocyte
adhesive glycoprotein
complex (CD 1 l/CD 18) did not prevent
the L-NAME-induced
increase in epithelial permeability.
These
data suggest that inhibition
of nitric oxide production
leads to
a reversible circulating
leukocyte-independent
increase in epithelial permeability.
intestinal
nine
permeability
NITRIC
OXIDE
AND
EPITHELIAL
l
adhesion glycoprotein
complex CDll/CDl&
This intervention
has previously been used to establish the contribution
of leukocyte adhesion
to platelet activating
factor- and ischemia/
reperfusion-induced
vascular protein leakage (10, 12). Moreover, this dose of MoAb IB, completely
abolishes L-NAMEinduced leukocyte adherence in feline mesenteric venules (10).
Control values for the aforementioned
parameters
were obtained, and then L-NAME
(0.025 prnol. ml-l. min-l)
was infused for 1 h, during which all parameters
were measured at
IO-min intervals. In a final series of animals, the effect of
blood flow on 51Cr-EDTA
clearance was examined. Control values were obtained, and then superior mesenteric arterial blood
flow was mechanically
reduced to 20% of control. The clearance of 51Cr-EDTA
was again measured at lo-min intervals
for 60 min.
51Cr-EDTA
activity in plasma and in 2-ml aliquots of perfusate was measured in an LKB CompuGamma
spectrometer
(model 1282; LKB instruments,
Gaithersburg,
MD). Additionally, the amount of luminal perfusate was measured at the times
described above to determine alterations in net intestinal water
absorption
during ischemia and after reperfusion. This calculation was made by subtracting
the amount of fluid entering the
bowel from the amount collected from the distal end over the
lo-min perfusion period. At the end of the experiment, the loops
were removed, rinsed, and weighed. Loops that had a 51CrEDTA
clearance >O.l ml mine1 100 g-l during the control
period were excluded from this study.
The plasma-to-lumen
clearance of 51Cr-EDTA
and interstitium-to-lumen
clearance of the dextran molecules were calculated as follows
l
clearance
l
l
l
= cpm, X pr X lOO/cpm,,
X wt
l
where clearance of 51Cr-EDTA
is given in millimeters
per
minute per 100 g, cpm, is counts per minute per milliliter
of
perfusate, pr is the perfusion rate, cpmpl is counts per minute
per milliliter
of plasma (or lymph for dextran calculations),
and
wt is weight of the intestinal segment in grams.
Statistical
analysis was performed using standard methods,
i.e., one-way analysis of variance, and Student’s t test with a
Bonferroni
correction
for multiple
comparisons
where
necessary. All values are given as means t SE, and statistical
significance was set at P < 0.05.
RESULTS
Figure 1 demonstrates the time-dependent
effect of LNAME (0.025 pmol ml-l min-l) on intestinal blood-tolumen 51Cr-EDTA clearance. Clearance values increased
significantly
within 20 min, reached peak permeability
l
l
0.5
(.025
0
10
20
30
L-NAME
Fig. 1. Effect of 90
infusion;
0.025 pmol
Sodium nitroprusside
infusion.
Values did
side was not infused
to control;
+P < 0.05
40
50
INFUSION
60
70
pmole/ml/min)
80
90
(min)
min NG-nitro-L-arginine-methyl
ester (L-NAME
. ml-l
min-l)
on 51Cr-EDTA
clearance
(n = 7).
was infused for the last 30 min of the L-NAME
not decrease below the 60-min value if nitroprusfor the last 30 min (not shown). * P c 0.05 relative
relative
to 60 min of L-NAME.
l
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
and then an arterial circuit was established between the superior mesenteric (SM) and left femoral arteries. The SM arterial
pressure was measured via a T tube interposed within the arterial circuit, with the use of a Statham P23A transducer. Systemic arterial pressure was monitored
via a cannula inserted
into the femoral artery. All pressure cannulas were positioned at
the level of the heart. Blood pressures and intestinal blood flow
were continually
recorded with the use of a Grass physiological
recorder (Grass Instruments,
Quincy, MA). In all animals, both
renal pedicles were ligated to prevent excretion of Wr-EDTA
in the urine.
One or two loops of small intestine (- 15 cm in length) were
fitted with rubber plugs with inflow and outflow cannulas. The
intestine and abdominal contents were then covered with plastic wrap to avoid evaporative water loss. The gut loops were
perfused with warmed Tyrode solution at a rate of -1 ml/min.
51Cr-EDTA,
obtained
from New England
Nuclear (Boston,
MA), was injected intravenously
such that plasma counts per
minute were at least 25,00O/ml (100-150 &i/kg).
One hour was
permitted for tissue equilibration
of the 51Cr-EDTA
after which
luminal
perfusate was collected over three IO-min control
periods. The nitric oxide synthesis inhibitor
NG-nitro-L-arginine-methyl
ester (L-NAME;
Sigma, St. Louis, MO) was then
infused at 0.025 prnol. ml-l. min-’ for 90 min. Increasing the
infusion rate of L-NAME
did not cause a further increase in
arterial blood pressure. The aforementioned
hemodynamic
and
mucosal parameters were recorded at lo-min intervals. In some
animals, at 60 min of L-NAME
infusion, sodium nitroprusside
was infused at 0.025 pmol ml-l. min-l for the last 30 min of the
experiment. In a second series of experiments,
the enantiomer
NG-nitro-D-arginine-methyl
ester (D-NAME;
0.025 pmol ml-l
min-l)
or L-NAME
plus L-arginine
(0.125 pmol . ml-l min-l)
were infused, and intestinal
blood flow, mucosal permeability,
and the different blood pressures were monitored for 1 h. In the
latter experiments,
L-arginine
infusion was initiated
30 min
before the L-NAME
infusion. A fivefold higher dose of L-arginine than L-NAME
was used, because this has previously been
reported to be necessary to reverse the L-NAME
effects both in
our laboratory
(13) and in others (9).
To further quantify the restrictive properties of the mucosal
membrane, in some animals a large lymphatic
vessel emerging
from the mesenteric pedicle was cannulated,
and rhodaminedextran (mol wt = 17,200, radius = 29.5 A) was administered
intravenously
2 h before the start of the experiment.
This was
sufficient time to achieve interstitial
equilibration
as assessed
by the lymph to plasma concentration
of rhodamine-dextran.
Interstitial-to-lumen
clearance rather than blood-to-lumen
of
the dextrans was determined
before and during 60 min of LNAME
infusion. Dextran concentrations
in fluid were determined fluorometrically
using an SPF-500C spectrofluorometer
(SLM-Aminco,
Urbana, IL). Rhodamine
was stimulated to fluorescence with an excitatory wavelength
of 560 nm, and fluorescence was observed at 584 nm. Slit widths were kept at 1 nm.
A standard concentration
curve for rhodamine
conjugated dextran was prepared, and total fluorescence from samples was
obtained as a function of concentration.
Over the range of O-O.1
mg/ml this was a linear function, and all unknowns were diluted
so that the final concentration
fell within the standard range.
Lymph rather than plasma samples was used, because the microvasculature
partly contributes
to restricting
the movement
of macromolecules
(dextrans) from blood to lumen. The use of
lymph rather than blood samples (interstitial-to-lumen
clearance) circumvented
the complication
associated with the properties of the microvascular
barrier.
In another set of experiments,
the contribution
of polymorphonuclear
leukocytes
to the L-NAME-induced
mucosal
dysfunction was assessed. Animals were pretreated with monoclonal antibody (MoAb) IB, (1 mg/kg iv) a monoclonal antibody
directed against the common subunit (CD18) of the leukocyte
G1139
PERMEABILITY
G1140
NITRIC
OXIDE
AND
EPITHELIAL
l
Table 1. L-NAME - induced intestinal hemodynamics
and transmucosal fluid flux
Pre-L-NAME
L-NAME
(30 min)
41.2t6.3
27.1&5.5*
21.5t3.4*
3.4t0.5
6.2t1.8"
7.3tl.9”
l
l
Control
L-NAME
D-NAME
L-NAME
L-arginine
+
Fig. 2. Peak (30 min) 51Cr-EDTA
clearance
during
L-NAME
(0.025
pm01 ml-l . min-l)
infusion
(n = 7)) NG-nitro-D-arginine-methyl
ester
(D-NAME;
0.025 prnol. ml-l
mix+)
infusion
(n = 4) or L-NAME
(0.025 pm01 . ml-l. min-l)
+ L-arginine
(0.125 pm01 . ml-l
mix+)
infusion (n = 5). * P < 0.05 relative
to control;
+P < 0.05 relative to 30 min
of L-NAME.
l
l
l
SMA blood flow,
ml*min-l*lOO
g-l
Precapillary
resistance,
mmHg
ml-l. min. 100 g
Net secretory
fluid flux
ml.min-l*lOO
g-l
L-NAME
(60 min)
l
-0.6kO.2
-0.5kO.3
-1.2t0.3
Values are means t SE; n = 7 cats. Superior
mesenteric
(SMA)
blood flow, resistance,
and net secretory
flux at 0,30, and
of L-NAME
infusion
(0.25 ~molml-’
amin-l).
Negative
values
secretory
flux represent
net absorption.
* P < 0.05 relative
to
Control
L- NAME
1--NAME
MoAb IB,
arterial
60 min
for net
control.
Ischemia
Fig. 3. Peak (30 min) 51Cr-EDTA
clearance
during
L-NAME
(0.025
pmol . ml-l
min-l)
infusion
in the presence (n = 4) and absence (n = 7)
of MoAb
IB, pretreatment
(1 mg/kg)
as well as 51Cr-EDTA
clearance
during 80% reduction
in blood flow (ischemia,
n = 5). * P < 0.05 relative
to control;
+P < 0.05 relative to 30 min of L-NAME.
l
(lymph flow did increase significantly
infusion).
after L-NAME
DISCUSSION
The results of this study suggest that inhibition of
nitric oxide synthesis with the L-arginine analogue LNAME causesa rapid increase in mucosal permeability to
51Cr-EDTA. D-NAME, the biologically inactive enantiomer, did not affect epithelial permeability to these
probes, whereas L-arginine almost entirely prevented the
increase in mucosal permeability induced by L-NAME.
These data indicate that inhibition of endogenous release
of nitric oxide from either epithelial cells, enteric nerves,
mast cells, endothelial cells, or some other cell type increased mucosal permeability to 51Cr-EDTA as well as
macromolecules. This hypothesis is further supported by
the observation that nitroprusside, a nitrogen oxide-containing compound that spontaneously releases nitric oxide when it interacts with plasma, completely reversed
the increase in clearance of 51Cr-EDTA from interstitium
to lumen. Although nitroprusside may have nonspecific
effects, these data are consistent with the view that the
continuous release of nitric oxide plays an instrumental
role in maintaining and perhaps modulating the integrity
of the mucosal barrier.
51Cr-EDTA was used in this study to quantitatively
assess subtle alterations in the mucosal barrier of the
intestine. EDTA is a small molecule that is quickly distributed throughout the extracelluar compartment after
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
values (-6-fold)
at 30 min and then leveled at approximately four times the L-NAME
preinfusion
values for the
remainder of the L-NAME
infusion. A similar result was
obtained (0.04-0.20 ml min-l
100 g-l) when 51Cr-EDTA was measured from lumen-to-blood
before and after
L-NAME
infusion. Administration
of nitroprusside
at 60
min of L-NAME
infusion reversed the L-NAME-induced
mucosal permeability
increase (Fig. 1). In the absence of
nitroprusside,
the L-NAME-induced
clearance of 51CrEDTA was not different between 60 and 90 min (not
shown). Pretreatment
of animals with L-arginine greatly
reduced the L-NAME-induced
increase in mucosal permeability, whereas D-NAME
alone had no effect on this
parameter (Fig. 2).
Associated
with the peak (30 min) increase in 51CrEDTA clearance was a substantial
reduction
(35%) in
intestinal blood flow and a 100% increase in SM arterial
resistance that lasted for the duration of the L-NAME
infusion (Table 1). L-NAME,
however, had no effect on
transmucosal
fluid flux in the feline intestine; net fluid
absorption
was maintained
throughout
the experiment.
Higher concentrations
of L-NAME
did not further alter
any of these or any other parameters
(data not shown).
Figure 3 illustrates
51Cr-EDTA
clearance values obtained by 1) mechanically
reducing superior mesenteric
arterial blood flow to 20% of control for 60 min and 2)
pretreating
animals with MoAb IB4 the monoclonal antibody that completely prevents L-NAME-induced
leukocyte adhesion (13). Physically decreasing blood flow to as
little as 20% of control had no effect on 51Cr-EDTA
clearance. Figure 3 also demonstrates
that leukocyte adhesion did not contribute to the increase in mucosal permeability; the increased movement of 51Cr-EDTA
from
the mucosal interstitium
to the bowel lumen was unaffected by MoAb IB4.
In some animals, the interstitial-to-lumen
clearance of
rhodamine-dextran
(mol wt = 17,200) was examined before and during L-NAME
infusion. Rhodamine-dextran
clearance increased
approximately
twofold
(0.028 t
0.01-0.058
t 0.01 mlmin-l
100 g-l) after L-NAME
infusion. The increased clearance of rhodmaine-dextran
occurred without
an increase in the concentration
of
rhodamine-dextran
in lymph throughout
the experiment
PERMEABILITY
NITRIC
OXIDE
AND
EPITHELIAL
mucosal permeability
associated with the inhibition
of
nitric oxide production. We cannot rule out the possibility that other inflammatory
cells may increase mucosal
permeability
by releasing cytotoxic agents after nitric
oxide synthesis inhibition.
For example, inhibition
of
nitric oxide production
causes mast cell degranulation
(20), which may play a role in the increased epithelial
permeability.
This
mechanism
warrants
further
attention.
We have previously reported that L-NAME
infusion
into the intestinal
circulation
increases microvascular
permeability
(11). It is conceivable that the microvascular dysfunction coupled to L-NAME
infusion could be
responsible for the increased epithelial permeability.
For
example, a rise in capillary fluid filtration associated with
increased capillary permeability
might augment mucosal
interstitial
fluid pressure sufficiently to provide the driving force for fluid and solute filtration across a disrupted
mucosal membrane (filtration secretion) (5). However, in
this study, there appears to be little evidence for filtration
secretion as net absorption was observed throughout the
experiment. This is further supported by the observation
that L-NAME produces 51Cr-EDTA clearance of similar
magnitude when measured in the opposite direction, i.e.,
from lumen-to-blood.
In conclusion, the results of this study indicate for the
first time that nitric oxide may be an important endogenous modulator of intestinal epithelial permeability.
This
effect appears to be independent of alterations in intestinal blood flow or adhesion of leukocytes to vascular
endothelium.
Based on our observations, one would prediet that decreased nitric oxide production
and/or increased nitric oxide inactivation,
a potential feature of
inflammation,
would likely enhance permeability
of the
epithelium.
Because superoxide 1) is known to inactivate
nitric oxide and 2) is produced in large quantities under
certain inflammatory
conditions including inflammatory
bowel disease and ischemia/reperfusion
of the intestine,
these data raise the possibility that nitric oxide inactivation may be an important event contributing
to a compromised mucosal barrier associated with the aforementioned disease states. Clearly, this mechanism
now
warrants attention in various inflammatory
models.
We thank Dr. Karl Arfors for the
Jon Meddings
for help in measuring
lymph and superfusate.
This study was supported
by a
Foundation
for Medical
Research.
Address
for reprint
requests:
P.
Group,
Dept. of Medical
Physiology,
Calgary,
Calgary,
Alberta
TZN 4N1,
Received
12 February
REFERENCES
l* Aoki, N., G.
2.
3.
1992; accepted
generous
levels
grant
gift of MoAB
IB, and Dr.
of rhodamine-dextran
in
from
the
Alberta
Kubes, Gastrointestinal
Faculty
of Medicine,
Canada.
in final
form
23 March
Heritage
Research
Univ.
of
1992.
Johnson
III, and A. M. Lefer.
Beneficial
effects
of two forms of NO administration
in feline splanchnic
artery
occlusion
shock. Am. J. Physiol.
258 (Gastrointest.
Liver Physiol.
21): G275G281,
1990.
Boeckxstaens,
G. E., P. A. Pelckmans,
J. G. de Man,
H. Bult,
A. G. Herman,
and Y. M. Van Maercke.
Release of
nitric
oxide upon stimulation
of non-adrenergic
non-cholinergic
nerves in the gut (Abstract).
Gastroenterology
100: A244, 1991.
Crissinger,
K. D., P. R. Kvietys,
and D. N. Granger.
Pathophysiology
of gastrointestinal
mucosal
permeability.
J. Intern.
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
intravenous administration.
The limiting
barrier to the
movement of this molecule from blood to bowel lumen is
the epithelial cell layer of the mucosa and is completely
independent of alterations in the endothelial cell layer of
the vasculature (3, 15). The sixfold increase in mucosal
permeability to 51Cr-EDTA in addition to the appearance
of macromolecules
(rhodamine-dextran
= 29.5 A radius)
in the lumen during the inhibition
of nitric oxide synthesis suggests a large increase in mucosal permeability.
The
lack of obvious mucosal damage in this study (neither
gross lesions nor blood on the mucosal surface) coupled
with a complete reversal of the increased 51Cr-EDTA
clearance with nitroprusside
would support the contention that inhibition
of nitric oxide leads to an increase in
mucosal permeability
that is unrelated to tissue injury. In
fact Hutcheson et al. (8) have reported that inhibition
of
nitric oxide production with L-arginine analogues (50 mg/
kg) did not produce any histological alterations in the
mucosa.
These data may also provide an explanation
for the
observation that administration
of nitric oxide is beneficial in animals exposed to splanchnic artery occlusion (1)
or ethanol (16) as well as the finding that nitric oxide
inhibitors enhance mucosal injury after such insults as
endotoxic shock (8) and ethanol administration
(18). In
the aforementioned
studies, the mechanism underlying
beneficial effects of nitric oxide was not elucidated; however, the investigators
proposed that the observations
may be a result of altered intestinal blood flow. We now
present data that endogenous production of nitric oxide
plays an important role in the modulation of permeability
across the epithelial barrier, and because superoxide is
known to inactivate nitric oxide, in inflammatory
conditions associated with enhanced superoxide production,
increased epithelial permeability
may contribute to intestinal dysfunction.
Although we propose that nitric oxide has a direct effect on the epithelial cell barrier there are several other
possibilities that warrant attention. First, inhibition
of
nitric oxide production causes vasoconstriction
and a reduction in superior mesenteric blood flow of 40%. It is
conceivable that the inhibitor of nitric oxide production
caused increased mucosal permeability
as a consequence
of reduced blood flow. To assess this possibility, we compared the mucosal permeability
elicited by partial occlusion of the superior mesenteric artery with that observed
with infusion of L-NAME.
The results of this analysis
reveal that a reduction in blood flow to as much as 20% of
control does not increase 51Cr-EDTA clearance, suggesting that the increased epithelial permeability
induced by
L-NAME is not blood flow related.
Inhibition
of nitric oxide has previously been reported
to increase leukocyte adhesion to postcapillary
venules
(13), an event that could be completely reversed by administration
of MoAb IB4 directed against the common
subunit (CD18) of the leukocyte adhesion glycoprotein
complex CDll/CDl8.
In the present study, pretreatment
with the same dose of MoAb IB4 as in the aforementioned
study had no effect on mucosal permeability
increase induced by L-NAME.
These data suggest that leukocyte
adhesion per se is not responsible for the increase in
G1141
PERMEABILITY
G1142
NITRIC
OXIDE
AND
EPITHELIAL
12. Kubes,
P., M. Suzuki,
and D. N. Granger.
Modulation
of
PAF-induced
leukocyte
adherence
and increased
microvascular
permeability.
Am. J. Physiol.
259 (Gastrointest.
Liver Physiol.
22):
G859-G864,
1990.
13. Kubes,
P., M. Suzuki,
and D. N. Granger.
Nitric
oxide: an
endogenous
modulator
of leukocyte
adhesion.
Proc. N&Z. Acad.
Sci. USA 88: 4651-4655,
1991.
14. Kvietys,
P. R., W. G. Patterson,
J. M. Russell,
J. A.
Barrowman,
and D. N. Granger.
Role of the microcirculation
in ethanol-induced
mucosal injury in the dog. Gastroenterology
87:
562-571,
1984.
15. Larsson,
M., L. Johnston,
G. Nylander,
and U. Ohman.
Plasma water
and Cr-EDTA
equilibration
volumes
in different
tissues in the rat. Acta Physiol.
Stand.
110: 53-57, 1980.
16. MacNaughton,
W. K., G. Cirino,
and J. L. Wallace.
Endothelium-derived
relaxing
factor
(nitric
oxide)
has protective
actions in the stomach.
Life Sci. 45: 1869-1876,
1989.
R. M. J., D. S. Ashton,
and S. Moncada.
Vascular
17. Palmer,
endothelial
cells synthesize
nitric
oxide from L-arginine.
Nature
Lond. 333: 664-666,
1988.
18. Peskar,
B. M., M. Respondek,
K. M. Muller,
and B. A.
Peskar.
A role
for
nitric
oxide
in
capsaicin-induced
gastroprotection.
Eur. J. PharmacoZ.
198: 113-114,
1991.
19. Rubanyi,
G. M., and P. M. Vanhoutte.
Superoxide
anions and
hyperoxia
inactivate
endothelium-derived
relaxing
factor. Am. J.
Physiol.
250 (Yeart
Circ. Physiol.
19): H822-H827,
1986.
20. Salvemini,
D,, E. Masini,
A. Pistelli,
P. F. Mannaioni,
and
J. Vane.
Nitric
oxide:
a regulatory
mediator
of mast cell
reactivity.
J. Cardiovasc.
Pharmacol.
198: 113-114,
1991.
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Med. 228, Suppl. 1: 145-154,
1990.
4. Granger,
D. N. Role of xanthine
oxidase
and granulocytes
in
ischemia-reperfusion
injury.
Am. J. Physiol.
255 (Heart
Circ.
Physiol.
24): H1269-H1275,
1988.
5. Granger,
D. N., P. R. Kvietys,
W. H. Wilborn,
N. A. Mortillaro,
and A. E. Taylor.
Mechanism
of glucagon-induced
intestinal
secretion.
Am. J. Physiol.
239 (Gastrointest.
Liver Physiol.
2): G30-G38,
1980.
6. Granger,
D. N., M. A. Perry,
P. R. Kvietys,
and A. E.
Taylor.
Interstitium-to-blood
movement
of macromolecules
in
the absorbing
small intestine.
Am. J. Physiol.
241 (Gastrointest.
Liver Physiol.
4): G31-G36,
1981.
7. Gryglewski,
R. J., R. M. J. Palmer,
and S. Moncada.
Superoxide
anion is involved
in the breakdown
of endothelium-derived vascular
relaxing
factor.
Nature
Lond. 320: 454-456,
1986.
B.
J.
R.
Whittle,
and
N.
K.
8. Hutcheson,
I.
R.,
Boughton-Smith.
Role of nitric
oxide in maintaining
vascular
integrity
in endotoxin-induced
acute intestinal
damage in the rat.
Br. J. Pharmacol.
101: 815-820,
1990.
9. Ignarro,
L. J. Biological
actions and properties
of endotheliumderived
nitric
oxide formed
and released
from artery
and vein.
Circ. Res. 65: 1-21, 1989.
10. Kubes,
P., and D. N. Granger.
Interaction
between circulating
granulocytes
and xanthine
oxidase-derived
oxidants
in the postischemic
intestine.
In: Clinical
Aspects
of O2 Transport
and
Tissue Oxygenation,
edited by K. Reinhart
and K. Eyrich.
Berlin:
Springer-Verlag,
1989, p. 133-147.
11. Kubes,
P., and D. N. Granger.
Nitric
oxide modulates
microvascular
permeability.
Am. J. Physiol.
262 (Heart Circ. Physiol.
31): H611-H615,
1992.
PERMEABILITY