PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES Mum on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ |

I .
_J
ﬂ

MSU Is An Affirmative Action/Equal Opportunity lnditution

 

 

 

_——__ __V.- _ .. _.

INTESTINAL PROSTANOIDS IN SHOCK

AND

POSTTRANDIAL INTESTINAL HYPEREMIA

BY

ADAMU ALEMAYEHU, MD.

A DISSERTATION __

Submitted to
Michigan State University

in partial fulﬁllment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Department of Physiology
1989

ABSTRACT

INTESTINAL PROST AN OIDS IN SHOCK
AND
POSTPRANDIAL INTESTINAL HYPEREMIA
BY
ADAMU ALEMAYEHU, MD.

The gastrointestinal tract produces and releases a number of prostanoids which
affect many of its functions, such as secretion, absorption, motility, blood ﬂow, and
metabolism. Recent studies suggest that endogenous intestinal prostanoids may play
a role in the regulation of blood flow and oxygen uptake during nutrient absorption.
Evidence has also accumulated which suggests a pathophysiological role for prosta-
noids in several diseases of the gastrointestinal tract. The purpose of this study was
to elucidate which of the endogenous prostanoids play a role during physiological
conditions (i.e. at rest and during nutrient absorption) and pathological
conditions (i.e. hemorrhagic shock). With this in mind, the first part of this
study tested the effects of intraluminal placement of normal saline (NS), digested
food (F) and F plus archidonate (AA) on jejunal blood flow, oxygen uptake and
prostanoid production. After it was found that prostanoid production was indeed
increased by luminal placement of F or F + AA, the vascular effects of local i.a
infusions of prostanoids at concentration similar to those observed in response to
luminal placement of F or F + AA was studied. In an attempt to determine the
stimulus for food- induced increases in jejunal prostanoid production the eﬂects of
NS, bile , intestinal distension and pump-induced hyperemia were individually

tested for their effect on prostanoid production.

Also in an attempt to determine which prostanoid may be involved, the effects

mimic changing luminal content from N8 to F did not significantly alter the JVR.
However, infusions of a mixture of the four or two prostanoids mimicking a situation
of changing luminal content from N8 to F + AA or from F to F + AA, respectively,
increased the JVR in a dose dependent fashion and abolished the F induced de-
crease in JVR.

Luminal placement of NS, 10% bile, or mechanical distension by air did not
alter jejunal blood flow and prostanoid release. This indicates NS and bile, which
were used to prepare the food mixture, and luminal distension by air, to simulate
the jejunal volume change observed during food placement, did not directly con-
tribute to the food- induced increase in prostanoid prduction. However, a pump-
induced hyperemia which mimicked postprandial hyperemia signiﬁcantly increased

‘ the rate of P612, TXAZ, and POE/2 releases from the jejunum while it did not alter
the PGF2 release. This study indicates part of the stimulus for the food-induced
prostanoid release might be secondary to the food-induced increase in blood ﬂow.

The selective thromboxane receptor antagonist 80- 29548 had no effect on
postprandial intestinal hyperemia but significantly potentiated the increases in
oxygen uptake and capillary exchange capacity. The latter was assesed by the capil-
lary ﬁltration coefficient.

Local intra-arterial infusion of a TXA2 agonist (U- 44069) , in a free flow
intestinal preparation, decreased both mucosal and muscularis blood flow of the
jejunal wall. The decrease in mucosal blood flow was significantly greater than the
decrease in the muscularis. In constant flow rate conditions, infusion of U-44069
decreased blood flow the mucosal and increased the muscularis blood flow. Infusion
of 80-29548, in both conditions, restored the U-44069 induced blood ﬂow changes
to control levels. This study indicates greater sensitivity of the mucosal vasculature
to TXA2 than of the muscle vasculature, which might favor a redistribution of

blood flow within the intestinal layers.

of a thromboxane receptor antagonist and/or agonist were studied. Their effects on
resting and postprandial jejunal blood flow, oxygen uptake, capillary exchange
capacity and blood ﬂow distribution between mucosal and muscularis layers were
determined.

In the last part of this study the role of intestinal prostanoids in the develop-
ment and progress of irreversible hemorrhagic shock was studied in dogs subjected
to hemorrhagic hypotension. Arterial pressure was maintained between 30-40
mmHg for 3 hours followed by reinfusion of the remaining shed blood. Intestinal
production of 6-keto-PGF1 and TXB2 were measured every hour. In separate dogs
subjected to the same hemorrhagic shock protocol, a glucose oleic acid (GOA)
solution was perfused continously into the lumen of an isolated intestinal segment
while intestinal prostanoid production was measured every hour. At the end of the
experiment intestinal biopsy was taken to determine if there are morphological
changes associated with severe hemorrhagic shock. The main findings of the study
are as follows:

Placement of a digested food into the jejunum significantly increased local
blood flow, oxygen uptake and jejunal prostanoid release. Prostanoids were released
in the order of PGE2 > P612 > TXAZ and > PGFZ . Arachidonate placed into the
lumen with the food solution abolished the vascular and metabolic changes associ-
ated with food placement. Accompanying the arachidonate action was a signiﬁcant
increase in jejunal venous concentration and release of TXA2 and PGF2 compared
to food placement alone. A TXA2 receptor antagonist (SQ-29548) reversed the
arachidonate vascular and metabolic action, indicating TXAZ is partly responsible
for the vascular and metabolic actions of arachidonate added to the food.

Local intra-arterial infusion of the TXA2 receptor agonist U-44069, and
PGF2 increased, while P612 and PGE2 decreased jejunal vascular resistance

(JVR) in a dose dependent fashion. Infusions of a mixture of the four prostanoids to

In dogs subjected to hemorrhagic hypotension for 3h followed by reinfusion of
the remaining shed blood, arterial and intestinal venous TXB2 concentrations signif-
icantly increased during hypotensive and post-transfusion periods. However, the 6-
keto-PGFl concenterations increased during hypotension but decreased during
post-transfusion periods. The intestinal TX82 release increased progressively,
whereas 6-keto-PGF1 release decreased. There was no significant change in the
control dogs.

Perfusion of a GOA solution intraluminally in dogs subjected to a similar
hemorrhagic shock protocol abolished the shock-induced increase in intestinal
venous concentration and release of TXAZ, whereas venous concentrations and
release of 6-keto-PGF1 increased progressively during hypotension and post-trans-
fusion periods. Accompanying the changes in prostanoid release were a signiﬁcant
decrease in the magnitude of intestinal vascular resistance and mucosal damage.
This study shows an imbalance in intestinal production and release of TXA2 and
P612, in favor of TXAz, during severe hemorrhgic hypotension and after blood
transfusion. The imbalance may contribute to the development of irreversible
hemorrhagic shock and reperfusion injury. Part of the mechanism for the benefical
effect of GOA perfusion during severe hemorrhagic shock seems to be related to its

effect on reversing the shock-induced imbalance in prostanoid production.

This dissertation is dedicated with love to my parents, and family for their love and

support.

ii

ACKNOWLEDGEMENTS

I would like to acknowledge the advice, support, and scientfic expertise of my
mentor, Dr. C.C. Chou, and the other members of my guidance comittee, Dr’s.
Chimoskey, Smith, Spielman, and Stephenson. Also, special thanks to COM (MSU),
Department of Physiology (MSU), Medical faculty (AAU), and WHO for their
financial support during my training. A special thanks to my parents and family for

their life long encouragement and endless support throughout my education.

iii

TABLE OF CONTENTS

 

 

 

 

 

 

PAGE
LIST OF TABLES ..................................... v
LIST OF FIGURES - - - ......... - - - vi-ix
INTRODUCTION - - - __ -- - 1-2
LITERATURE REVIEW .............................................................................. 3-19
MATERIALS AND METHODS .................................................................. 20-34
RESULTS ................................................... - - - -- ...... -- - 35-109
DISCUSSION -- -_ -- 110-127
CONCLUSION ................................................................................................. 128-131
BIBLIOGRAPHY.-- -- - ........................................... 132-142

iv

LIST OF TABLES

Table Page

 

Table I. Compositon of Glucose-Oleic Acid solution 23

Table II. Bleeding volumes in GOA treated and untreated hemorrhagic shock
dogs (mean 1 SEM) ..... 89

 

Table III. Heart rate values for GOA treated and untreated hemorrha 'c
shock and sham-operated dogs (mean 3; SEM)- - - 9

 

Table IV. Hematocrit values for GOA treated and untreated hemorrhagic
shock and sham-operated dogs (mean 1; SEM) -- _ 90

 

Table V. Intestinal TXBZ release for GOA treated and untreated hemorrha -
ic shock and sham-operated dogs (mean 1; SEM) - - _ - - 10

 

Table VI. Intestinal 6-keto-PGF1 release for GOA treated and untreated
hemorrhagic shock and sham-operated dogs (mean 1 SEM) .............. 107

LIST OF FIGURES

Figure Page

 

Figure 1. Biosynthetic pathway of the major eicosanoids 11.

Figure 2. Schematic representation of the preparation used in most
experiments - 21

 

Figure 3 Original poly raph chart recording of the data used to estimate
'e'unal capill filration coe ficient (Kfc). MAP = mean arterial blood pressure ;
= intesti venous pressure ; BF = intestinal blood flow ; I.Wt = intestinal rate
of weight change - . ..... - 29

 

Figure 4 Jejunal blood ﬂow and oxygen uptake (V02) during luminal place-
ment of normal saline (NS), food (F), normal saline plus arachidonate (NS+AA),
and food plus arachidonate (F+AA). n=9. SQ denotes the values after i.a. injec-
tion of 80-29548, a TXAz/endo eroxide receptor antagonist, when the lumen still
contained F+AA. n=6. " p<0.0 relative to NS values= - 37

 

Fi re 5 Arterial and jejunal venous plasma concentrations of 6-keto-PGF ,
TXB , GEZ, and PGF2 when the jejunal lumen contained normal saline (N15),
food IF , normal saline plus arachidonate (NS+AA) and food plus arachidonate
(F+AA . n=9. " and "‘ p<0.05 relative to venous concentration obtained during
placement of NS and F, respectively ........ _ - - ...... - 40

 

i%ure 6 Jejunal releases of 6-keto-PGF1 , TXBz, PGE2, and PGF2 (ng

" 1 Og'l) during luminal placement of normal $811118 (NS), food (F), normal
saline plus arachidonate (NS+AA), and food plus arachidonate (F+ AA). n=9. “
and " denote values signiﬁcantly different from those obtained during placement of
NS and F, respectively, p <0.05.-- - 42

Fi re 7 Effects of local intra-arterial infusions of U-44069, a TXAZ analog,
and PG 2 (A), P612 and PGE (B) on jejunal vascular resistance (JVR). Restin
vascular resistance were 3.57 i 0.73, 3.63 i 0.36, 2.77 i 0.53, 2.18 i 0.2
mmHg ml/min/ 100g for PGIZ, PGE2, PGF2 , and U-44069, rerspectively. n=5-9. ‘
p< 0.0 relative to control value. _ __ 45

Fi e 8 Effects of local i.a. infusions of mixtures of , PGF , PGI , U-44069
and PG 2 at concentrations 0.5X (times), 1X, 2X, 5X, 10X, an 20X t e value of
the increases in venous prostanoid concentrations shown in Fi . 5 when luminal
content was changed from normal saline to food ( [PGs] F-NS) fPGIz = 82pg/ml,
TXA = 72 pg/ml, PGF/2 = 259Ipé/ml and PGF2 = 47pg/ml in the mixture) (A ,
and izood lus arachidonate ( [ s FAA-NS (PGIZ = 82pg/ml, TXA = 2
pg/ml, P E2 = 259pg/ml and PG = 47p ml in the mixture) (B). Fillgd dia-
monds indicate the actual observed2 data, w ereas open diamonds indicate the
predicted data as calculated by the summation of the individual prostanoid vascular
effects shown in F i . 7. Resting vascular resistance was 3.41 _-l_-_ 0.38 and 3.63 1; 0.36
mmHg/ml/min/ 1 g. n=9 ‘ p<0.05 relative to control value 48

v1

 

 

 

Figure 9 Effects of luminal placement of food (F) (ﬁlled circle) and subse-
quent local i.a. infusions of a mixture of U-44069 and PGF on jejunal vascular
resistance. [PGs]FAA-F at 1X was TXAZ : 120pg ml and P 2 : 80pg/ml. Restln
vascular resistance was 3.13 _+_ 0.37 mmHg/ml min/ 100g. n=9. ‘ and ‘ p<0.0
relative to F and control value -- 51

 

Figure 10 Arterial and 'ejunal venous plasma concentrations of 6-keto PGF)~l
and when the jejunal umen is empty (unfilled bars) or filled with norma

 

saline ( ed bars) - 54
Fi e 11 Arterial and jejunal venous plasma concentrations of 6-keto PGF
3nd ‘ when the jejunal lumen is empty (unﬁlled bars) or ﬁlled with air (giGHed
arsl

 

Figure 12 Arterial and jejunal venous lasma concentrations of 6-keto PGF
when the jejunal lumen contained normal sa ine (unﬁlled bars) or 10% bile (Sﬁslled
bars, -

Figure 13 Jejunal blood flow and perfusion pressure before (control) and
after ump-induced jejunal h eremia (mechanical hyperemia). Asterisks indicate
signi cant change rom corresponding control value
p<0.05 - 61

Figure 14 Arterial and jejunal venous lasma concentrations of PGF/2, 6-keto
PGF , PGF and TXB before (control) an after pump-induced jejunal hyperemia
(mec anica hyperemig). Asterisks indicate signifcant change from corresponding
control valuep < 0.05---_ - ------- 63

Figure 15 'ejunal releases of PGIZ, TXAZ, PGF2 and PGE2
Engmin' .100gm'1) before (control) and after pump- induced Jejunal hyperemia
mechanical hyperemia). Asterisks indicate signifcant change from corresponding
control value p< 0.05- - -- 65

Figure 16 Original poly raph chart recording showing the effect of U-44069
on je'unal perfusion pressure efore and after adminstration of S0-29548 (a TXAZ
/ en oproxide receptor blocker) 68

Figure 17 The effect of intra-arterial bolus administration of graded dose: of
a TXA analog (U-44069) on percent changes in jejunal vascular resistance before
(unﬁlle circles) and after intra-arterial administration of 0.1 ug (ﬁlled circle) , 0.6
u (unfilled triangles) and 2.0 u (filled triangles) a TXAZ receptor blocker ($0-
2 548). Asterisks and crosses in icate signiﬁcant differences (p<0.05) from control
values and corresponding U-44069 values (unﬁlled circles) 70

Figure 18 Jejunal blood flow before and after administration of 80-29548
during normal saline (unfilled bars) and food plus bile (filled bars) in the lumen.
Aasiterisks indicate significant Increases (p < 0.05) above normal saline
V 1188 73

vii

 

Figure 19 J ejunal oxygen consumption before and after administration of 80-
29548 in the presence of normal saline (unfilled bars) and food plus bile (filled
bars). Single and double asterisks indicate signiﬁcant increases (p<0.05) compared
to normal saline and food plus bile values before adminstration of 80-29548, re-

spectively - 75

Fi re 20 Jejunal capillary ﬁltration coefﬁcient before and after administra-
tion of Sg01-29548 in the presence of normal saline (unﬁlled bars) and food plus bile
(ﬁlled bars). Single and double asterisks indicate significant increase (p<0.05)
above normal saline and food plus bile values before adminstration of S -29548,
respectively. - 77

 

 

Figure 21 Food-induced ercent changes in ﬂejunal oxygen u take, capillary
ﬁltration coefficient and bloo flow before (unfi led bars) and a ter (filled bars;
administration of 80-29548. Asterisks indicate significant differences (p<0.05
from corresponding values before 80-29548 79

 

Fi re 22 Mucosal plus submucosal (mucosa) and muscularis plus serosal
(muscle blood ﬂows durrng i.a. infusion of U-44069 before and after 80-29548
administration. Asterisks and crosses indicate signiﬁcant differences compared to
control and U-44069 values respectively (p<0.05). Jejunal segment was perfused at
free ﬂow rate -- - -- ..... 82

 

Fi re 23 Mucosal plus submucosal (mucosa) and muscularis plus serosal
(muscle blood flows during i.a. infusion of U-44069 before and after 80-29548
administration. J ejunal segment was perfused at constant flow rate. See text for
statistical analysis- - -- -

Figure 24 Mean arterial blood pressure of severe untreated hemorrhagic
shock dogs (triangles) (n=6), of sham- operated control dogs (ﬁlled circles) (n=5)
and GOA erfused dogs (unfilled circles) (n=8) over the time of experimental
procedure. 11 the hemorrhagic shock groups blood pressure was stablized between

0-40 mmHg for 3 hrs (0-3) by controlled bleeding. At the third hour all the shed
blood was transfused. About 30-45 minutes after bleeding, a glucose oleic acid
(GOA solution was perfused into the lumen of the segment at a rate of 1 ml/min in
the G A-treated group. Single and double asterisks denote signiﬁcant differences
relative to prehemorrhagic, and corresponding control values respectively p <0.05.
Each point represents mean 1 SEM --------- 88

Figure 25 Intestinal blood ﬂow of untreated hemorrhagic shock do (trian-

glelsg, sham-operated control dogs (filled circles) and GOA-treated s ock do 5
u lled circles) during the 5 hrs experimental period. single and double asteris
denote signiﬁcant differences relative to prehemorrhagic, and corresponding control
values, respectively p <0.05. Each point represents mean _t SEM 3

Figure 26 Intestinal vascular resistance of untreated hemorrhagic shock dogs
Etrian les), sham-operated control dogs (ﬁlled circles) and GOA-treated shock do
unﬁl ed circles) during the 5 hrs experimental period. single and double asteris

and crosses denote signiﬁcant differences relatlve to prehemorrhagic, correspond-

ing control and GOA-treated shock values respectively p <0.05 95

 

viii

 

Figure 27 Arterial blood lucose concentration in hemorrhagic shock dogs
treated with normal saline (unfi led circles) or GOA (trian les) perfused through
the entire small intestinal lumen (from previous study ( 1)) or GOA perfused
through a 5mm segment (ﬁlled circles) (present study) -- 96

 

Fi re 28 The arterial 6-keto PGF1 (stable metabolite of PGI ) concentra-
tion (11 71h) in the untreated hemorrhagic shock dogs (triangles), s am-operated
contro dogs (filled circles) and GOA-treated shock dogs (unfilled circles) durin
the 5 hrs experimental period. Single and double asterisks and crosses denote signig
icant differences relative to prehemorrhagic, corresponding control and untreated
shock values respectively p < .05. Each pornt represents mean _+_ SEM ............ 99

Figure 29 Intestinal venous 6-keto PGFl concentration (ng/ml) in the un-
treated hemorrhagic shock dogs, sham-operated control dogs and GOA-treated
shock dogs during the 5 hrs experimental period. Sin 1e and double asterisks and
crosses denote significant differences relative to pre emorrha ic, corresponding
control and untreated shock values respectively p <0.05. Eac point represents
mean _+_: SEM -- - - ..... 99

Fi re 30 The arterial TXBZ (stable metabolite of TXA2) concentration
(ng/ml 1n the untreated hemorrhagic shock dogs (trian les), sham-operated Control
dogs ( rlled circles) and GOA-treated shock dogs (unﬁ ed circles) uring the 5 hrs
experimental period. single and double asterisks denote signiﬁcant differences rela-
tive to prehemorrhagic and corresponding control values respectively p < 0.05. Each
point represents mean :1; SEM- - - - - - - -- 102

Figure 31 Intestinal venous TXB concentration (11 /ml) in the untreated
hemorrhagic shock dogs , sham— operate control do 5 and GOA-treated shock dogs
durin the 5 hrs experlmental period. Single and dou 1e asterisks and crosses denote
signi icant differences relative to prehemorrha ic, corresponding control and GOA-

treated shock values respectively p <0.0 . Each point represents mean 1;
SEM - -------- - -- - 102

Figure 32 Arterial TXBZ to 6-keto-PGF1 ratios in GOA-treated and untreat-
ed hemorrhagic shock dogs and sham- operated control do 5 during the 5 hrs e er-
imental eriod. Single and double asterisks and crosses genote sr nificant di fer-
ences re ative to prehemorrhagic, corresponding control and GO -treated shock
values respectively p <0.05. Each point represents mean i SEM---- 105

Figure 33 Intestinal venous TXBZ to 6-keto-PGF1 ratios in GOA-treated and
untreated hemorrhagic shock dogs and sham-operated control dogs during the 5 hrs
ex erimental eriod. Single and double asterisks and crosses denote significant
d1 ferences re ative to prehemorrhagic, corresponding control and GOA-treated
shock values respectively p <0.05. Each point represents mean i SEM ........... 105

INTRODUCTION

Recent studies suggest that endogenous intestinal prostanoids may play a
role in the regulation of blood flow and oxygen consumption during nutrient absorp-
tion, and in the pathogenesis of a variety of shock conditions. Inhibition of cycloox-
ygenase by indomethacin or mefenamic acid potentiates (52,117), whereas adminis-
tration of arachidonic acid attenuates (102), the food-or oleic acid-induced increases
in intestinal blood flow and oxygen uptake. These studies, however, raise more ques
tions because of complex arachidonic acid metabolism and the antagonistic vascular
effects of prostanoids. PGEZ and PG12 are vasodilators (22,43,49,56), whereas
PGF2 and TXAZ are vasoconstrictor (22,38,56) in the small intestine.

Supportive evidence for the involvement of prostanoids in the progression of
shock is accumulating. Cyclooxygenase inhibitors, such as aspirin and indomethacin,
are known to reduce thromboxane and prostacyclin production induced by endotox-
emia in dogs (40,70), rabbits (27) and rats (20), with an associated increase in sur-
vival. Administration of prostacyclin (93), thromboxane synthetase inhibitors (111),
and thromboxane receptor antagonists have been shown to improve the condition of
animals in severe shock conditions induced by trauma, splanchnic artery occlusion,
and hemorrhage, respectively. There is, however, no study that measures these two
prostanoids during the course of severe irreversible hemorrhagic shock. Further-
more, most studies measure their concentration in the systemic blood rather than in
the venous blood of an organ.

The purpose of this study was to determine intestinal prostanoid production
during nutrient absorption and during various stages of severe hemorrhagic shock
and after blood transfusion. After it was found that food indeed increased prosta-
noid production, the vasoactivity of PGI2, PGF/2, PGF2 and a TXAZ agonist were

determined individually or in a mixture. In an attempt to determine the constituents

of food responsible for intestinal prostanoid release, N8, bile, intestinal disten
sion, and pump-induced hyperernia were individually tested. Also, the effects of a
thrombortane receptor agonist and / or antagonist on resting and postprandial intes-
tinal blood ﬂow, oxygen uptake, capillary exchange capacity and intestinal blood
flow distribution between mucosal and muscularis layers were determined in an

attempt to determine which prostanoids may be involved.

LITERATURE REVIEW
Postprandial Intestinal Hyperemia

It has long been thought that, during the digestion of food, blood is diverted
from other tissues of the body to the viscera. The fact that the viscera appear
congested has been taken as evidence of an increased blood supply to these organs.
It was in 1910, Brodie et a1. (15) documented the first evidence of a postprandial
hyperemia in the canine small intestine by using a plethysmograph method. At the
same time suggestions were made that the drowsiness one feels after a meal may be
due to a redistribution of blood away from the brain to the intestine. In 1934, Her
rick et a1. (74) made the ﬁrst blood ﬂow measurement to quantify the cardiovascular
events during digestion utilizing the thermostromuhr method in conscious dogs.
This study indicated that during digestion of food the blood flow in the femoral,
carotid, mesenteric arteries and the external jugular vein markedly increased.
These findings were substantiated by Reininger et al. (119) who measured blood
ﬂow by the 42K fractional distribution technique in both fed and unfed rats. These
investigators reported that during digestion, there is a uniform increase in the blood
flow to all organs of the fed rats. The cardiac output also significantly increased
during digestion. The aforementioned studies questioned the prevailing concept
that digestion results in diversions of blood ﬂow from other organs to the digestive
tract.

However, more recent systematic studies have yielded different results. In
1968 Fronek et a1. (50) measured the systemic and regional hemodynamic changes
during food intake and digestion in dogs. During anticipation and food intake
(injestion) a signiﬁcant increase in cardiac output, heart rate, arterial blood pressure

and ﬂow in brachiocephalic artery were observed while the ﬂow in superior mesen-

 

teric and iliac artery dropped signiﬁcantly. During digestion (at 1 and 3 hours after
injestion), there were no signiﬁcant changes in cardiac output, heart rate, and mean
arterial blood pressure, whereas the ﬂow in the superior mesenteric artery increased
signiﬁcantly and flows in the brachiocephalic and iliac arteries decreased. These
ﬁndings were further substantiated by studies conducted by Vatner et al. in 1970
(145,146). In these studies, blood flow measurements were made utilizing pulsed
ultrasonic flow meters, Doppler ultrasonic telemetry ﬂow meters or Zepeda 400-
cycle square-wave electromagnetic ﬂow meters. Anticipation and injestion of food
were characterized by transient increase in cardiac output, heart rate, arterial pres-
sure, mesenteric resistance and renal resistance while iliac and coronary resistance
decreased transiently. Mesenteric ﬂow began to increase within 5 min after eating
and reached a peak within 30-90 min, gradually returning to preprandial control
levels 2-6 hrs later. Iliac flow was decreased slightly 30-60 min postprandially.
Within 10-30 min, renal and coronary ﬂow and resistance returned to preprandial
control levels and remained there during peak mesenteric vasodilation.

In 1970’s and the early 1980’s, studies were conducted which more carefully
characterized the cardiovascular responses to feeding. Two distinctly different
phases were noted, the anticipation and injestion of food phase, and the digestion
phase. In the former the cardiovas cular changes, (i.e., increased cardiac output,
heart rate, arterial pressure, mesenteric resistance and renal resis tance, and de-
creased ooronary resistance) were associated with a generalized sympathomimetic
stimulation and generally lasted for a very short time, (5-30'min). This was followed
by the digestion phase of cardiovascular response which is manifested by a signiﬁ-
cant increase in the mesenteric blood flow (30-90 min) and the return of cardiac
output, heart rate, arterial pressure, renal resistance and coronary resistance to
preprandial level (145,147). Even though these recent ﬁndings ruled out the possi-
bility of redistribution of blood flow away from other vascular beds to the intestine

 

there is still the unexplained fact of how the increases in mesenteric blood flow are
achieved. Vatner et al. in 1974 (147), and Gallavan et al. in 1980 (55) have suggest-
ed that small increases (5-10%) in cardiac output could account for the sustained
increase in mesenteric blood flow, these changes may be missed due to technical
error inherent in the current methods used to determine cardiac output.

The postprandial increase in superior mesenteric blood flow is not shared
equally by all mesenteric organs. It is selective to those participating in digestion.
Even within the intestine itself, various regions are perfused to different degrees.
For example, the introduction of food into the stomach of the conscious dog in-
creases blood ﬂow only in the celiac artery (29). Likewise, fatty acids placed in the
colon of dogs elicit an organ-speciﬁc colonic hyperemia (84). Furthermore, with a
rise in superior mesenteric blood flow elicited by food in the intestine, little or none
is distributed to the stomach and colon. Within sequential segments of the small
intestine, all regions are not perfused equally at any one time (29,41). Chou et a1.
(29) demonstrated that intraduodenal placement of food increases blood flow in the
superior mesenteric artery, but does not increase celiac flow or blood ﬂow to an
isolated adjacent jejunal segment. These studies indicate that the postprandial
hyperemia is localized to that portion of the gastrointestinal tract that is exposed to
chyme.

The selectivity of postprandial hyperemia extends to the various layers of the
gut wall. Differences in blood flow distribution exist among the mucosa, submucosa,
and muscularis. Using radioactively labelled microspheres, Chou et al. (29,30)
showed the local distribution within jejunum favored the mucosa. This preferential
distribution to mucosa was also reported by Gallavan et a1. (55). On the other hand,
blood flow to the muscularis layers does not change over the ﬁrst 90 min after a
meal, during which time superior mesenteric blood flow increases over 100%. Post

prandial gastric and colonic hyperemia are also attributed to increased mucosal flow

(55). Thus it appears that mesenteric postprandial hyperemia is quite selective in its
distribution to regions that are closely tied to digestive and absorptive processes,
and that the mucosal region receives the largest share of the increase.

At the local intestinal level, the nature of the food material in the lumen is the
determinant of the vascular response. For example, Kvietys et al. (87) demonstrat-
ed that luminal osmolarity over the range 180-1000 osmole/ kg is not a signiﬁcant
factor in contributions to local hyperemia, but the concentration of the nutrient is.
When chyme was analyzed for its local hyperemia-stimulating activity (32), the
partially digested fats and carbohydrates were speciﬁcally effective. Undigested
food elicited no response (32 ). Additionally, bile apparently plays an important
role. In the jejunum, bile has no effect by itself, but enhances the hyperemic effect
of digested food. In contrast, bile salts alone elicit hyperemia in the ileum (82).

Chou et al. (32) and Kvietys et al. (82) conducted the ﬁrst systematic study of
the effects of the various constituents of chyme on intestinal blood flow using isolat-
ed loops of canine small intestine. These studies show that only glucose was capable
of increasing jejunal blood ﬂow when placed in the lumen at physiological concen-
tration (150 mM). The long-chain fatty acids, oleic acid and the monoglyceride
monolein, had no effect on intestinal blood flow when placed in the lumen in
aqueous solution. However, solubilization of these lipids, by increasing the pH of
the solution (pH = 9) (31) or by addition of either gallbladder bile (82) or sodium
taurocholate (32), resulted in their ability to increase intestinal blood ﬂow by 20-
30%. At physiological concentrations, it appears that few, if any, of the common
dietary amino acids have an effect on intestinal blood ﬂow. The by-products of
protein digestion are relatively potent vasodilators of the intestinal circulation
(Brodie et al. (15); Brandt et al. (14); Siregar et al. (140)). The peptides responsible
for vasoactivity are unknown, although it is possible that some fragments cleaved

from the parent protein may have amino acid sequences similar to those of the

active portions of vasoactive peptides of the gut.

The effect of the various nutrients on intestinal blood ﬂow might be mediated
by a variety of regulatory pathways. Some of the factors involved include tissue
oxygen tension (60,129,112) , intestinal peptides such as neuro tensin and vasoactive
intestinal polypeptide (51,57), paracrine substances such as prostanoids
(52,102,103,104,117) and histamine (33), and intrinsic nerves (27,30,146) .

The possible role of hormonal components in postprandial mesenteric hyper-
emia was suggested due to the vasodilator effect a variety of gastrointestinal pep-
tides showed during exogenous administration into the mesenteric vascular bed
(7,19,26,41,101,116). In 1967 Burns and Schenk (19) found intravenous infusion of
secretin and gastrin increased superior mesenteric blood ﬂow. Other investigators
(7,26,41,101) also found administration of CCK, gastrin, and secretin increased
intestinal blood ﬂow. However, a recent study (51) which measured the concentra-
tions of these peptides in the venous blood found it to be less than that required to
dilate the intestinal vasculature.

The involvement of extrinsic nerves in the postprandial hyperemia has been
ruled out, since vagal stimulation and vagotomy did not alter intestinal blood ﬂow
(78,85,144,145), and the postprandial hyperemia was not affected by alpha- or beta
adrenergic blockade (41,42,146). The role of intrinsic nerves remains controversial.
The hyperemia produced by mechanically stroking the intestinal mucosa and that
produced by transmural electrical ﬁeld stimulation have been shown to be mediated
by an intramural neural reﬂex (10). The reﬂex involves the release of 5- hydroxy-
tryptamine (5-HT or serotonin) and possibly vasoactive intestinal polypeptide (VIP)
(8,9,10). Luminal placement of a 50% glucose solution (3,000 mOsmol/kg) in-
creases intestinal blood ﬂow as well as motility, and both effects were blocked after
exposing the mucosal surface to a local anesthetic, dibucaine hydrochloride (110 ).

These studies seem to indicate that stimulation of mechano- or osmoreceptors on

 

the mucosal surface initiates a local neural reﬂex that increases intestinal blood ﬂow
and motility.

Recent studies by Gallavan et al. (51) found a close temporal relationship
between jejunal vasoactive intestinal polypeptide (VIP) release and the bile-oleate-
induced jejunal hyperemia. Furthermore, this increase in VIP release was not seen
when either bile or oleic acid was placed in the lumen alone (treatments which also
failed to produce hyperemia). Since VIP is found only in the neural tissue of the
small intestine (89) and is a vasodilator in the canine small intestine ( 39 ), a portion
of the bile- oleate-induced jejunal hyperemia may be mediated by a neural pathway
involving VIP-ergic neurons. Fahrenkrurg et al. (39) have also reported a correla-
tion between intestinal VIP release and the hyperemia induced by mechanical
stimulation of the mucosa.

Nyhof et al. (110) provided evidence against local neural mechanism for intes-
tinal postprandial hyperemia. Methysergide, hexamethonium, and tetrodotoxin
failed to alter either the vascular or metabolic responses to luminal placement of
glucose or oleic acid. Exposing the mucosa to dibucaine (a local anesthetic) blocked
or attenuated the increases in blood ﬂow and oxygen uptake. The inhibition of the
nutrient-induced intestinal hyperemia by dibucaine is due, at least in part, to its
effect on oxygen consumption and glucose transport of the mucosal epithelial cells.
These ﬁndings therefore indicate that local intestinal nerves may not play a signiﬁ-
cant role in the initiation and maintenance of nutrient-induced jejunal hyperemia.

Several studies have shown an increase in oxygen consumption as well as an
increase in blood ﬂow when foodstuffs are introduced into the gastrointestinal tract
(30,53,55). The mesenteric metabolic control theory has been based on the assump-
tion that delivery of oxygen to parenchymal cells is the controlled variable rather
than the blood ﬂow when foodstuffs are introduced into the gastrointestinal tract

(60,64,131,132,138). Of the possible mediators of the metabolic hyperemia only

adenosine and changes in tissue P02 have been examined in the mesenteric circula-
tion. Adenosine has been proposed as a mediator of metabolic hyperemia in a
variety of tissues and is a potent intestinal vasodilator (65). However, exogenously
administered adenosine decreases ileal oxygen consumption, the capillary ﬁltration
coefﬁcient, and PS product and redistributes blood ﬂow from the mucosa to the
muscularis (65,137). The role of endogenous adenosine in the regulation of intesti-
nal blood ﬂow during digestion and absorption requires further study.

Chou et al. (33) determined the role of histamine in postprandial intestinal
hyperemia utilizing H1- and/or H2- receptor blockers. Metamine, an H2 receptor
antagonist, had no effect on the food-induced increase in ﬂow and oxygen uptake,
whereas tripelennamine, an H1 receptor antagonist, significantly attenuated the
food-induced increases in blood ﬂow and blocked the increases in oxygen consump-
tion. A 30% increase in flow was reduced to 15% after administration of this agent.
They concluded that endogenous histamines may play a role in postprandial intesti-
nal hyperemia, and the effect is primarily mediated by the Hl-receptors.

Gallavan and Chou (52) and Procter et al. (117) have shown that inhibition of
prostanoid synthesis signiﬁcantly potentiated the jejunal vascular and metabolic re
sponse to food. Conversely, Mangino and Chou (102), and Procter et al. (117) have
reported that stimulation of prostanoid synthesis significantly inhibits the food-
induced increases in jejunal blood ﬂow and oxygen consumption. The inhibi-
tory effect of prostanoids on the jejunal response to food is complicated by the
antagonistic biological actions of prostanoids in the gastrointestinal tract. P612 and
PGF/2 are vasodilators (22,43,49,56) whereas PGF2 and TXA2 are vasoconstrictors
(22,38,56) in the small intestine.

Prostanoids
In 1930, Raphel Kurzork and Charles Lieb (81) discovered that human

seminal plasma contracted uterine smooth muscle. Shortly thereafter, broader

 

10

biological immi cations were realized from the research of Maurice Goldblatt and
Ulf 8. Von Euler (59) who demonstrated that the compounds in semen were active
on a number of smooth muscles and lowered blood pressure when injected in
animals. Von Euler called these factors prostaglandins (indicating the ﬁrst anatom-
ical region of origin).

Arachidonic acid and other prostaglandin precursors are not normally found
in the free state in the cell but occur bound in phospholipids, neutral lipids and with
cholesterol (11,114). Since esteriﬁed fatty acids are not substrates for the enzymes
which make prostaglandins, the ﬁrst step in prostaglandin biosynthesis is the release
of these fatty acids from their combination in lipids. The most important source of
fatty acids for prostaglandin formation is phospholipids.

The backbone of the phospholipid structure is a 3- carbon compound called
glycerol . The 1’ position is usually occupied by a saturated fatty acid (i.e., does not
have any double bonds such as palmitic acid ) and 2’ position is taken upiby an
unsaturated fatty acid (such as arachidonic acid). The 3’ position is taken up with a
base (usually choline, ethanalamine, serine or inosital) attached to the glycerol
backbone by phosphate linkage.

Release of arachidonic acid from membrane phospholipids can be achieved
enzymatically by one or another of several phospholipase enzymes (47,125). Phos-
pholipase A2 attacks at the 2’ position liberating the unsaturated fatty acid and
leaving a 2’ lysophosphatide. Phospholipase c removes the phosphorylated base
leaving a diglyceride which can then be broken down by a speciﬁc diglyceride lipase
enzyme to release both fatty acids in free form (76).

The most important enzyme normally responsible for liberating fatty acids in
prostanoid biosynthesis is phospholipase A2. In some cells, e.g., platelets, the
concerted action of phospholipase c and diglyceride lipase is necessary to release

arachidonic acid. The biosynthesis of all prostanoids (Fig 1 )is initiated by the same

 

11

FIGURE 1

 
  

' PLASMA MEMBRANE (Phosphel- pid mm")

.......................

WHO“. 12

xcifa” O- .\ AA ”‘1
Locum Cyclo-eaygmu onuA/W

 

MWNIC ACID

W ”‘so ProﬁquWin

60H
P603
1 ”Operetta“.

(countess.

....... W—
0" cww‘
HOOC o" syn. hell"
hatuyetin PM; 7 W (Tuzl Ot-t
Synthouse :

7WN"1\:W$1
3a: 231/ . .1..-

s-mo-Porg. P532 W2.

 

Figure 1. Biosynthetic pathway of the major eicosanoids

12

pair of enzymes: a cyclooxygenase, which catalyzes the oxygenation of arachidonic
acid to the cyclic endoperoxide prostaglandin G2 and a hydroperoxidase, which
converts prostaglandin 62 to prostaglandin H2 (Fig. 1 ) (125). After this point, most
cells are highly speciﬁc in their metabolism of prostaglandin Hz to the biologically
active products, a selectivity that usually is linked to the function of the cell and its
communication with cells that carry receptors for that speciﬁc prostaglandin or
thromboxane A2. As a group, these active metabolites of prostaglandin H2 have
been termed prostanoids.

The rate limiting step in prostanoid biosynthesis is the liberation of arachidon-
ic acid from phospholipids, making free fatty acid available for metabolism by
cyclooxygenase (114). In some cells, free arachidonic acid also may be metabolized
by lipoxygenase enzyme to products that include the leukotrienes, and by mixed-
function oxygenases to epoxyeicosatrienoic acids. As a group, these highly varied
oxygenated metabolites of arachidonic acid are called eicosanoids, reﬂecting their
origin from a 20-carbon (eicosa-) polyunsaturated fatty acid.

Prostanoids and the Gastrointestinal Tract

Ever since the discovery of prostanoids, use has been made of isolated gut
smooth muscle as bioassay preparation for prostanoids. Over and above the practi-
cal application many other aspects of gastrointestinal function may to some extent
depend upon the generation of endogenous prostanoids, or be inﬂuenced by the
administration of exogenous prostanoids (4,16,34,44,52,54,58,79,80,90,102-
104,117,128). The list of gut function so affected includes transport of water and
electrolytes in the colon and intes tine (16,35,44,79). Evidence has also accumulated
which sug gests a pathophysiological role for prostanoids in several diseases of the
gastrointestinal tract (5,36,93,122).

The effects of prostanoid synthesis inhibition on the postprandial jejunal

vascular and metabolic response was examined by Gallavan and Chou (52). Their

13

ﬁndings indicate that indomethacin and mefenamic acid, which are cyclooxygenase
inhibitors, reduced resting jejunal blood ﬂow and markedly enhanced the food-
induced jejunal hyperemia and oxygen consumption. Mangino and Chou (102) have
reported that stimulation of prostanoid production utilizing exogenously adminis-
tered arachidonic acid, signiﬁcantly attenuated the food-induced increases in jejunal
blood ﬂow and oxygen uptake. These ﬁndings suggest that endogenous prostanoids
may act to limit the magnitude of postprandial jejunal vascular and metabolic
response. Other studies performed by Proctor (117) showed that oleic acid, but not
glucose, induced functional hyperemia was potentiated by the local application of
cyclooxygenase inhibitors in rats.

The aforementioned studies performed utilizing cyclooxygenase inhibitors
and arachidonic acid raise many questions. One of the questions is which prosta-
noid(s) is/ are responsible for the biological effects observed during nutrient absorp-
tion. Mangino and Chou (103) reported that inhibition of thromboxane synthesis
utilizing imidazole and U-63557A (Upjohn) during nutrient absorption enhanced
jejunal oxygen uptake without altering the magnitude of the hyperemia. Additional
study by the same group (104) indicated that inhibition of thromboxane synthesis
potentiates the magnitude of postprandial capillary filtration coefficient. These
studies suggest thromboxane A2 might play a role in limiting 02 uptake and capil-
lary density during nutrient absorption. However, the observed results might also be
due to a diversion of the common substrate arachidonic acid to the formation of
other eicosanoids (118).

Prostanoids and Nutrient Transport and Metabolism
in the Gastrointestinal Tract
Gallavan and Chou (54) reported that inhibition of cyclooxygenase during
nutrient (enzymatically digested fat, carbohydrate and protein) absorption signiﬁ-

cantly enhanced the rate of glucose utilization and absorption in canine jejunum.

14
Previous studies by the same group have reported (52) that the postprandial hyper-
emia and oxygen consumption was potentiated by cyclooxygenase inhibitors. These
results indicate that endogenous prostanoids may act to limit glucose absorption and
hence limit the magnitude of postprandial jejunal hyperemia and oxygen uptake.
Several pieces of evidence support this hypothesis. A study by Sit et al. (141) showed
that the magnitude of the postprandial changes in blood ﬂow and oxygen consump-
tion were related to the rate of active glucose transport and its utilization by the
intestinal tissue. Exogenous administration of PGE2 and PGF2 into the canine
isolated jejunal loop decreased glucose absorption irrespective of their vascular
effects (4). Similar study in humans showed that administration of PGE1 decreased
the glucose absorption rate. Coupar and McColl (34) have shown that glucose
absorption was inhibited by exogenous administration of PGE1, PGF/2 and PGF2 in
the rat jejunum. The mechanism by which these prostanoids affect glucose transport
seems not to be related to their action on adenylcyclase since dibutryl 3’5’-cAMP
enhanced glucose absorption.
Prostanoid Effect on Jejunal Water and Electrolyte Transport
Prostanoids appear to play an important role in the physiologic regulation of

intestinal ﬂuid transport. Field et al. (44) have reported that addition of arachidonic
acid stimulates secretion, c AMP accumulation and PGE2 production in mm in
rabbit ileal mucosa. Brunson et al. (16) reported that administration of prostanoids
(PGE 1, PGFq) or their precursor arachidonic acid increased intestinal ﬂuid secre-
tion in the rat jejunum in $129 both when administered intraluminally or intra-arte-
rially. The mechanism by which prostanoids elicit this secretory action is dual. At
low doses, PGE2 (approximately 10'9M) acts mainly by local secretory nervous
reﬂexes as judged by the block with i.v. hexamethonium or locally applied Lidocaine
and not accompanied by any increase in tissue cAMP. At higher concentrations (10‘

8- 10'7) of PGF/2 there seems to be a non- neurogenic effect on the enterocytes

 

15

associated with an increase in tissue cAMP. Konturek et al. (79) have reported that
administration of PGF/z and 16.16.DMPGF/2 (2 ug/kg/h) into the mesenteric artery
signiﬁcantly enhanced net water secretion by the canine jejunum and ileum. Con-
versely, administration of indomethacin (5 mg/ kg/ hr) enhanced water absorption.
Administration of PGF2, P012 (2 ug/kg/h) and AA (10 ug/kg/hr) did not signiﬁ-
cantly alter intestinal net water movement. Stimulation of the receptor mediated
endogenous prostanoid production by whole rat ileal mucosa enhanced intestinal
prostaglandin and cAMP production, and a rise in mucosal short circuit current (an
indicator of chloride secretion) (148). Lawson et al. (90) showed that intestinal
epithelial tissue cAMP production is enhanced directly by PGF/2 or by stimulating
its synthesis by bradykinin. These studies suggest endogenous prostanoids might
regulate intestinal water and electrolyte transport under physiological conditions
and enhanced prostanoid production during pathological conditions might contrib-
ute to the pathophysiology of diarrhea] diseases.

Prostanoids are capable of stimulating alkaline secretion from the gastrod-
uodenal mucosa, whereas inhibitors of prostanoid biosynthesis inhibit this secretion,
causing mucosal damage (58). Bicarbonate secretion is involved in the protective
action of prostanoids on gastrointestinal mucosa and endogenous prostanoids are a
signiﬁcant factor in mucosal defense against various noxious agents. Konturek et al.
(80) have reported that natural PGEQ, PGF2, but not PGIZ were an effective stimu-
lant of duodenal alkaline secretion. Stable analogues such as 16-16DMPGEZ and
PGI2 were relatively more potent stimulants than their parent prostanoids. Con-
versely, inhibition of prostanoid synthesis by indomethacin suppresses duodenal
alkaline secretion.

Konturek et.al. (80) reported that infusion of PGE2 and PGF2 either i.v. or
directly into the artery supplying small intestine have an opposite effect on motility;
PGE2 inhibits while PGF2 stimulates intestinal motility. Infusion of PG12 also

16
inhibits intestinal motility. Conversely, inhibition of endogenous prostanoid synthe-
sis by indomethacin results in an increased intestinal motility. This study suggests
that the predominant action of prostanoids generated in the intestinal wall is the
inhibition of intestinal contractility and that this effect might be mainly due to re-
lease of PGIZ.
Intestinal Response During Hemorrhagic Shock

Shock is a condition in which acute circulatory failure occurs because of
derangement of circulatory control or loss of circulating ﬂuid. Shock can result from
many physiological disturbances. When acute circulatory metabolic dysfunction
occurs during infection with microorganisms, the condition is termed "septic shock."
Hypovolemic shock occurs when the volume of blood within the vascular compart-
ment decreases or shifts, resulting in inadequate perfusion of the body as a whole.
This form of shock can occur from severe hemorrhage, burns or dehydration.
Neurogenic shock results from a malfunction of the central nervous system caused
by drug intoxication or by injury leading to circulatory failure.

The hemodynamic responses that preced hemorrhagic shock lead to a reduc-
tion in the volume of venous blood returning to the heart. This results in stimula-
tion of the autonomic system which may maintain a fairly normal cardiac output
until the blood volume decreases by 15 %, at which time compensatory mechanisms
usually fail (123,127). In severe hemorrhagic shock, regional blood ﬂow to the
mesenteric, renal and femoral arteries is disproportionately decreased. Flow in
these vessels is apparently reduced in favor of maintaining circulation to the carotid,
coronary and hepatic arteries. Because of the high alpha-receptor activity in the
splanchnic area, shock can induce a pronounced vasoconstriction (120,154). Within
the intestinal wall, the mucosal blood ﬂow is selectively decreased. Thus, the tran-
sitory vasoconstriction induces ischemia in the mucosa of the gut.

Intestinal ischemia induces a characteristic damage to the small intestinal

17

mucosa within a short period of time (151). The lesions that are chracteristic of
ischemic injury are localized to the villi. Mucosal damage ranges from the devel-
opment of subepithelial space at the very tip of the villus to full scale damage
depending upon the severity and duration of ischemia. In pronounced phases of
ischemic mucosal damage, it is typical to ﬁnd the deeper layers of the intestinal wall
to be histologically normal (23). The pathogenesis of these lesions is controversial.
Three major pathophysiological factors have been implicated: tissue hypoxia (100),
superoxide radical (61) and pancreatic proteases (13).

Reactions of the small intestine during ischemia and shock have been report-
ed to be very important for survival. Lillehei (96) demonstrated that cross-perfusing
the small intestine of a dog during a period of hemorrhagic hypotension with blood
from a healthy donor dog prevented the intestinal mucosal damage and reduced
mortality from 80% to 10%. Other studies also have shown that prevention of intes-
tinal mucosal damage prevents the development of general cardiovascular de-
rangements accompanying shock. The degree of intestinal mucosal damage has
been directly correlated to the degree of cardiovascular derangements (97,99). The
ischemic intestine releases into blood material ”toxic" to the circulatory system,
probably from the ulcerated villi (96). In vitro experiments show that the intestinal
venous blood collected immediately after a period of regional intestinal ischemia
contains substances that exert a negative inotropic effect on the heart (94,115). This
substance ("the intestinal factor of irreversible hemorrhagic shock") was termed a
cardiotoxic or myocardial depressant factor (MDF) (94). The cardiotoxic material
released from the ischemic intestine has been characterized it has a molecular
weight of 500 to 1000 daltons and is water soluble. It is not a single myocardiac
depressant but several such substances that have been suggested by Haglund and
Lundgren (69).

The involvement of prostanoids in the pathogenesis and treatment of irrevers-

 

18
ible shock has been recently reported. The use of cyclooxygenase inhibitors in shock
was shown to increase survival of experimental animals (17,20,40,70). However,
cyclo-oxygenase inhibitors exert broad-ranging rather than selective effects on
prostanoid synthesis. Recently, prostacyclin (PGIz) has been found to possess a
variety of actions which are useful in the treatment of shock. This prostanoid is a
very potent inhibitor of platelet aggregation, a vasodilator, and is also a good
membrane stabilizer (1,21,66,92,107,108,109). It also prevents MDF formation in
shock (46,93,95). Conversely, administration of thromboxane synthetase inhibitors
has been found to prevent lysosomal disruption, and toxic factor formation during
shock (92,93). Prostacyclin, therfore, may be a potenial beneﬁcal factor, wheras
TXA2 is a deleterious factor in shock. No study has been done to determine the
origin of these prostanoids in shock. The splanchnic organs are known to play a
signiﬁcant role in the development and progress of irreversible hemorrhagic shock.
Therefore, the balance of these prostanoids (TXAZ and PGIZ) may play an impor-
tant role in the pathogenesis of irreversible hemorrhagic shock.
Nutritional requirements in shock
In a variety of shock conditions an early hyperglycemia followed by hypogly-

cemia has been documented. The hypoglycemia observed in shock condition is
associated with mortality (67,71,72,75,105,106,139,143). Strawitz and Hiff (143) in
1960 reported that the average survival time in fasted rats was 36 min as opposed to
an average of 65 min in a fed group subjected to standard hemorrhagic shock.
Moffat et al. (106) reported intravenous administration of glucose to dogs has signif-
icantly delayed physiologic decompensation and severity of metabolic changes when
compared to equally treated severe hemorrhagic shock dogs receiving no treatment.
These and other studies concluded that administration of glucose may have a merit
in the therapy of prolonged hypovolemic shock. Hinshaw et al. (75) reported that

glucose adminstration has a similar salutary effect in endotoxin shock.

19

In 1970 Chiu et al. (24) reported that during the development of intestinal
mucosal lesions in low-ﬂow states, a signiﬁcant reduction in mucosal ATP content
and oxygen uptake occurs. This correlates with the histological ﬁndings that epithe-
lial damage starts at the tips of the villi, extending gradually toward the crypts.
Intraluminal glucose administration in the presence of intestinal ischemia protected
the intestinal mucosa from ischemic damage.

Gump et al. (67) reported that radiolabelled glucose administered in post-
hemorrhagic shock following reperfusion was found to be incorporated into C14
lipids in higher quantity when compared to control values. Fat metabolism ap-
pears to be less sensitive to hypovolemic insult and may represent a more effective
energy substrate. Lang et al. (88) also reported that the free fatty acid oxidation by
the gut was not altered by endotoxin.

In 1982, Halper et al. (71,72) perfused a glucose- oleic acid solution through
the lumen of the small intestine during severe hemorrhagic shock. The perfusion
prevented the terminal hypoglycemia, histopathological damage of the intestinal
mucosa, the release of cardiodepressants from the splanchnic region and prolonged
survival rate, compared to untreated and saline-perfused dogs. These investigators
concluded that part of the mechanism by which luminal perfusion of glucose-oleic
acid solution protects against severe hemorrhagic shock might be related to the

prevention of hypoglycemia.

 

W AND METHODS
Surgical Winn
All experiments were conducted on mongrel dogs of both sexes (15-25 Kg),
fasted for 24 h. The animals were anesthetized with pentobarbital sodium (30
mg/kg, i.v.) and supplemented as needed to maintain a surgical plane of anesthesia.
All animals were ventilated with a positive pressure respirator (Harvard Apparatus,
Millis, MA) that was adjusted to achieve normal blood pH, 02 tension, and C02
tension before each experiment. Systemic arterial pressure was continuously moni-
tored through a cannula in the femoral artery.

In all series, a midline abdominal incision was made, and a segment of the
jejunum (30—45 g) about 30 cm aboral to the ligament of Treitz was exteriorized (Fig.
2). After a rubber tube was placed into the lumen of the segment for placement and
withdrawal of solutions, both ends of the segment were tied and cut away from the
adjacent jejunum to exclude collateral ﬂow (32). Luminal pressure was continuously
recorded through the rubber tube by a pressure transducer (Statham P23Gb) during
the presence of a solution in the lumen. Before cannulation of blood vessels, hepa-
rin sodium (500U / kg) was administered intravenously. The segment was covered
with a plastic sheet and kept at 37° C with a heat lamp and a thermoregulator
(Yellow Springs Instruments, Model 63RC, Yellow Springs, OH). At the end of all
experi ments, the animals were euthanized by an overdose of pentobarbital sodium
or potassium chloride.

Ramadan at hummus
The food solution used contained equal parts by weight of fat, carbohydrate,
and protein. It was prepared by adding 30g high fat test diet, 15g high protein test
diet and 5g high carbohydrate test diet (U.S. Biochemical, Cleveland, OH) to 400 ml

20

 

21

FIGURE 2

 

_J ( ' )’ \lutlong 6»va

~ \ i.a-Ina
: ‘5 13:3
O .-
‘ .

 

 

 

Figure 2. Schematic representation of the preparation used ln most
experiments

22

of 0.1 N NaHCO3 containing 750 mg of a pancreatic enzyme preparation (Viokase,
VioBin, Monticello IL) (52,102). The mixture was then gently mixed with a magnetic
stirrer at room temperature for 5 h to permit digestion. Before the experiment nine
parts of digested food were mixed with one part of gallbladder bile, which was
extracted from the animal. The pH and osmolality of the mixture were adjusted to
about 7.4 and 300 mosmol/Kg, respectively, with either NaHCO3 or HCl and NaCl.
Both the digested food plus bile and the normal saline were kept at 37° C during
the experiment. For simplicity, the term "food”, instead of "digested food plus bile",
is used throughout the text. The composition of glucose oleic acid (GOA) solution
is shown in Table 1.

In Series I (n= 16), the single vein draining the jejunal segment was cannulat-
ed for measurement of venous outﬂow by timed collection with a stopwatch and
graduated cylinder. Arteriovenous oxygen content difference [(A-V)02] (Fig. 2)
was determined continuously by perfusing femoral arterial blood and a portion of
the venous outﬂow at 6 ml/ min through separate cuvettes of an arteriovenous
oxygen content difference analyzer (A-Vox Systems, San Antonio, TX) (133) with a
Gilson pump (Minipuls 2, Gilson Medical Electronics, Middleton, WI). The venous
outﬂow and outﬂows from the cuvettes were directed to a reservoir. The blood in
the reservoir was pumped back into the animal via a femoral vein at a rate equal to
the total outﬂows. Oxygen uptake was calculated as the product of blood ﬂow and
(a- v)02.

When blood ﬂow reached a steady state, the lumen content was changed to
food solution. The data obtained during the last 3 min of 15-min normal saline or
food placement period were used for analysis, because the blood ﬂow and (A-
V)02 at that time were in a steady state (32,52,83,102).

The objective of the ﬁrst series of experiments was to determine jejunal productions

of PGF/2, PGIZ, TXA2 , and PGF2 when the jejunal lumen contained normal saline,

 

23

TABLE 1

Composition of Glucose-Oleic Acid Solution

 

 

Chemicals mM Grams/L
Glucose 150.0 27.0
NaOH 0.18 0.007
Taurocolic Acidl 6.01 3.1
NaCl 77.6 4.5

KCl 1.3 0.1
Ncho3 6.0 0.5
Mgc12 6H20 0.53 0.107
NaHZPo4 320 0.21 0.029
CaC122 0.45 0.05
Oleic Acid3 40.0 11.0 (12.5 ml/L)

 

1 Sodium salt, practical grade.

Cleveland,0H.

2 The solution was stirring when CaClz was added.

ICN Nutritional Biochemicals,

Solution was

then sonified and pH adjusted to 7.3.

3 90% grade, United States Biochemical Corp., Cleveland,OH.

2 4

food, normal saline plus arachidonate and food plus arachidonate. The rationale for
adding arachidonic acid to the food is to determine the effects of enhanced prosta-
noid synthesis on blood ﬂow, oxygen uptake and the relative synthesis of vasodilator
and vasoconstrictor prostanoids during nutrient absorption. Depending upon the
cell types and the metabolic activity under study, substrate loading with arachido-
nate magnify the cellular responses to a particular stimulus, in this case nutrient
absorption. A previous study has shown that arachidonate added to food attenuates
the food-induced vascular and metabolic changes (102). However, which prosta-
noids are responsible for arachidonate action are unknown.

After reaching a steady state by repeated luminal placement of normal saline,
luminal contents were changed sequentially and randomly to digested food, normal
saline plus arachidonate and food plus arachidonate. Each solution remained in the
lumen for 15 min. The arachidonate concentration in normal saline or food solution
was 6.5 x 10' 4M (200 ug/ml). This intraluminal dose of arachidonate was selected
from a study performed in a similar in vivo jejunal preparation (102). In another
intestinal in vivo study, the minimum concentration of arachidonate in the lumen,
which signiﬁcantly increased intestinal prostanoid production and secretion, was 6 X
10‘3 M (16). These intraluminal doses of arachidonate were higher than the doses
required to induce a similar effect during intra-arterial administration (5 X 10'3 -
2.3 X 10'4 M) (102) or direct application to in vitro cell preparations (2 X 10 '5 M)
(44,90). The reasons for the dose differences in the two routes of administration are
not known. However, this is also true for other vasoactive agents exogenously
administered in the intestine.

In six animals, 80-29548 (a TXA2 / endoperoxide receptor blocker, a gift from
Squibb, Princeton, NJ , ) 1.6 ug i.a. bolus was injected into a side branch of the single
artery of the segment at the end of the experiment when the lumen still contained

food plus arachidonate. The objective was to determine the effect of this

25
TXAZ receptor blocker on the food plus arachidonate induced changes in blood ﬂow
and oxygen uptake.

For the determination of prostanoid concentration blood samples were col-
lected in plastic test tubes containing the cyclooxygenase inhibitor mefenamic acid
(Sigma Chemical, St. Louis, MO) (5 ug/ml of blood). The samples were taken
simultaneously from a femoral artery and the venous outﬂow of the jejunal segment
9 and 15 min after luminal placement of each solution. The venous outﬂow was
measured at the time of blood sampling. Blood samples were centrifuged immedi-
ately for 10 min. The plasma was removed and stored at -20 ° C until radioimmu-
noassay.

The objective of Series II and III (11 = 9) of experiments was to determine the
effect of intra-arterial infusions of PGIZ, PGE2, PGF2 and U-44069 (a TXAZ
analog) (a kind gift from Upjohn Co., Kalamazoo, MI) on jejunal blood ﬂow. The
single artery perfusing the gut segment was cannulated, and the segment was per-
fused at a constant rate with the aortic blood by interposing a Masterﬂex pump
(Cole- Parmer Instruments, Chicago, IL) between the single artery and a femoral
artery. The perfusion pressure was monitored via a catheter in the perfusion line
attached to a pressure transducer (Statham P23Gb). Pump ﬂow rate was adjusted
until perfusion pressure was 10 mmHg below the systemic arterial pressure and then
maintained constant throughout the experiment. The venous outﬂow was directed
to a reservoir and pumped back into the animal via a femoral vein at a rate equal to
the total outﬂows. Solutions containing one or a mixture of more than two of the
following prostanoids were infused upstream from the Masterﬂex pump while
perfusion pressure was continously recorded. The vasoactivity of individual prosta-
noids was determined ﬁrst in a series of experiments. Following this, the vasoactivity
of local arterial infusion of a mixture of prostanoids at concentrations similar to

those produced by luminal placement of food or food plus arachidonate was deter-

2 6

mined. The local blood concentration of prostanoids to be achieved were selected
from the data obtained from Series 1 and were calculated from the pump blood ﬂow
rate, hematocrit, and prostanoid infusion rate. All prostanoids except prostacyclin
were kept as a stock solution in ethanol (10 mg/ml). Fresh P012 solution was
prepared in 50 mM Tris buffer (pH 9.37 at 0° C) immediately before infusion. The
infusates of U-44069 and PGF2 were prepared by diluting the stock solution in 0.9%
normal saline and that of PGE2 was diluted with 0.1M phosphate buffer prior to
each experiment.

The objectives of Series IV, V, VI, and VII of experiments were to determine
the stimuli for food-induced increases in jejunal prostanoid production. Normal
saline (NS), 10% bile (B), intestinal distension and an increase in jejunal blood ﬂow
were examined for their individual contribution to the food-induced increases in
jejunal prostanoid production.

The experimental preparation used for studying the effects of normal saline,
10% bile, and luminal distension by air was similar to Series 1. However, the effects
of an increase in jejunal blood ﬂow were studied under constant ﬂow conditions in a
similar experimental setup to Series 11. When blood ﬂow reached a steady state,
control arterial and intestinal venous blood samples were taken. Following the
collection of blood samples, jejunal blood ﬂow was increased to twice the control
blood ﬂow value by increasing the pump speed. Arterial and jejunal venous blood
samples were taken 9 and 15 minutes after the blood ﬂow was increased. The blood
samples were centrifuged immediately for 10 min. The plasma was removed and

stored at -20°C until radioimmunoassay.

The objective of the eighth series was to determine the jejunal vascular

response to a thromboxane A2 analog and receptor blocker. In this series (n= 10),
the experiment was conducted under constant ﬂow condition. The jejunal prepara-

tion was similar to Series II. The perfusion pressure was continuously monitored via

 

27
a catheter in the perfusion line attached to a pressure transducer. When blood ﬂow
reached a steady state, a graded dose of U-44069 (a thromboxane A2 analog,
Upjohn) was administered by intra- arterial bolus injection before and after admin-
istration of graded bolus doses of 80-29548 (A TXA2 receptor antagonist, Squibb) .

The objectives of the ninth series of experiments was to determine the role of
endogenous thromboxane A2 on food-induced changes in jejunal blood ﬂow and
oxygen uptake. In series IX (n=8), the effects of 80-29548 on food-induced changes
in jejunal blood ﬂow and oxygen consumption were determined. The jejunal prepa-
ration was similar to that of Series 1. Fifteen ml of normal saline were placed into
the lumen for 15 min. This procedure was repeated three to four times until blood
ﬂow reached a steady state, at which time the carrier solution for 80-29548 (etha-
nol) was administered intra-arterially as a control.

Following this, luminal contents were changed to digested food. After
determining the response to food, 80- 29548 (2.0 ug i.a. bolus) was administered in
a side branch of the single artery of the segment. This dose has been determined to
inhibit the action of the thromboxane analog (Fig. 17). When blood ﬂow reached
steady state the response due to food was determined. Oxygen uptake was calculat-
ed as the product of blood ﬂow and the arteriovenous oxygen content difference.
Both blood ﬂow and oxygen uptake were expressed as ml/min/ 100g.

The objective of the tenth and eleventh series of experiments (n=11) was to
determine the role of endogenous thromboxane A2 on food-induced changes in
jejunal capillary exchange capacity. The single vein draining the segment was cannu-
lated so that venous outﬂows were measured as described above. Blood ﬂow also
was monitored continuously with an extracorporeal electromagnetic ﬂow transducer
(BL-2048-E 04, Biotronex Laboratory, Silver Springs, MD) placed in the venous
outﬂow line and connected to an electromagnetic ﬂowmeter (BL610, Biotronex

Laboratory). The ﬂowmeter was calibrated periodically by measuring blood ﬂow by

2 8
timed collection of venous outﬂow. An arterial side branch was cannulated for
administration of drugs. After steady state was reached, measurement of capillary
ﬁltration coefﬁcient started.

Capillary ﬁltration coefﬁcients (Kfc) were determined gravimetrically. Brief-
ly, the jejunal segment was placed on a platform attached vertically to a calibrated
force displacement transducer (F103B, Grass Instruments, Quincy, MA). Normal
saline or predigested food was placed in the lumen, and Kfc was determined by a
sudden elevation of venous pressure from 0 to about 10 mmHg, lasting 10-15 sec-
onds (86). When jejunal venous pressure is elevated, venous outﬂows from the
jejunum fall initially to a low value and subsequently reach a plateau lower than the
control level (Fig. 3 ). The point of attainment of this plateau ("ﬂow equilibration”)
represents the end of the blood volume shift (63), and measuring the rate of intesti-
nal weight increases from that point divided by the increment in capillary pressure
gives Kfc (Capillary pressure was estimated using the venous occlusion technique
(62,104)). Jejunal Kfc was measured before and after intraarterial administration of
the thromboxane receptor blocker 80-29548. For each luminal content, the Kfc was
determined four times at 4 min intervals, with the last value used to compare the
alterations during the experimental conditions.

The objective for the twelveth and thirteenth series of experiments was to
determine the effect of a thromboxane A2 analog on jejunal wall blood ﬂow distri-
bution to the mucosa plus submucosa and muscle plus serosa before and after intra-
arterial administration of 80-29548. In this series (n=19) the single artery perfus-
ing the gut segment was cannulated, and perfused at either a constant rate with the
aortic blood by interposing a Masterﬂex pump (Cole-Parmer Instruments, Chicago,
IL) between the single artery and a femoral artery, or free ﬂow rate by interposing a
three-way stopcock upstream from the pump. A mixing chamber was connected to

the arterial line proximal to the pump. During constant ﬂow, the perfusion pressure

29

FIGURE 3

 

MAHmml-lo)
3
O

 

 

 

 

so]
.. 20
O
E
5 to]
s o ,
so
2 I
2
E to]
II.
a
o
a
3 2
z:
3. l}
0

Figure 3 Original poly raph chart recording of the data used to estimate
jvj’unal cap unmaI' filration coe cient (Kfc). -mean arterial blood pressure ,

intes venous pressure; BF= intestinal blood ﬂow; I.Wt = intestinal rate
of weight change.

3 O
was monitored via a catheter in the perfusion line attached to a pressure transducer
(Statham P23Gb). Pump ﬂow rate was adjusted until perfusion pressure was 10
mmHg below the systemic arterial pressure and then maintained constant through-
out the experiment. The single vein draining the jejunal segment was cannulated for
measurement of venous outﬂow by timed collec tion with a stopwatch and graduat-
ed cylinder. Once a steady state had been reached, rnicrosphere injection upstream
from the mixing chamber was started.
Radiolabelled microspheres 1“Ce (diam = 13.75 _1- 0.74 um), 46Sc (diam =

13.56 _f 0.70 um) and 85Sr (diam = 13.75 :- um) were obtained from Medical
Products Division/3M, St. Paul, MN 55144. Stock microspheres (25 uCi, 1,125,000
spheres/ ml) plus 1 drop Tween 80 were vortexed to achieve uniform distribution
prior to injection so that settling and clumping of spheres is prevented. 0.2 ml of the
stock microsphere (5 uCi: 225,000 spheres) were injected with a Hamilton syringe
into the intestinal arterial line upstream from the mixing chamber. The three types
of microspheres were injected randomly during control (ﬁrst microsphere), during
intra-arterial infusion of 0.16 ug/rnin U-44069, a thromboxane A2 receptor analog
(second rnicrosphere), and after intra-arterial bolus injection of 2.0 ug 80-29548, a
thromboxane A2 receptor antagonist while U-44069 infusion continued (third
rnicrosphere). Jejunal blood ﬂow was measured before and after injection of each
rnicrosphere . The above experimental protocol was performed either at free ﬂow
(n= 12) or constant ﬂow (n=7) conditions. At the end of the experiment the whole
isolated jejunal segment was taken and pinned ﬂat in the mucosal surface. The
serosa plus muscularis was removed intact by dissecting from the submucosa with a
scalpel. Each tissue, serosa plus muscle and submucosa plus mucosa, was cut into
about 1 gm pieces and placed in a pre-weighed counting tube. The pre-weighed
counting tubes containing tissue samples were weighed again to determine tissue

weight. The radioactivity of 468C, 85Sr, and 141Ce in each tube was measured for

3 1
10 min in the Packard 5000 series gamma counter (Packard Instrument Company,
2200 Warrenville Rd., Downers Grove, IL 60515). The energy windows used were
465e, 650-2000 Kev; 353:, 200-650 KeV; and 141Ce, 15.200 Kev. A spillover of
one radioactive sphere count into the channel of another sphere count was correct-
ed. The blood ﬂow to each layer of the intestine was calculated using the following

formula:

TMcpm/gm x JBF nil/min

Blood ﬂow to tissue sample =

 

J Mcpm/ gm

Where TM = tissue radioactivity

J M = total radioactivity in the jejunal segment
JBF = jejunal blood ﬂow

The objectives of the fourteenth series of expriments were 1) to determine the
systemic and local intestinal production of TXA2 and PG12 during various stages of
severe hemorrhagic shock and following blood transfusion, and 2) to determine if
the beneﬁcal effect of glucose oleic acid solution (GOA) perfused through intesti-
nal lumen is related to its effect on shock-induced intestinal PG12 and TXA2
production . In this series (n= 19) mongrel dogs (10- 25kg) of either sex were anes-
thetized with pentobarbital sodium (30 mg/kg, i.v.), and ventilated with a positive
pressure Harvard respirator. After laparotomy, a loop of ileum (15-20 cm in length)
was exteriorized, and the single vein draining the loop was cannulated after intrave-
nous administration of heparin sodium (500 U/kg). The venous outﬂow from the
intestinal loop was directed to a reservoir, and the collected blood was returned by a

pump to the animal through a femoral vein at a rate equal to that of the venous

32

outﬂow. Both ends of the 100p were tied and cut away from the remainder of ileum.
The loop was covered with a plastic sheet and kept at 37°C with a heat lamp and a
thermo- regulator. Systemic arterial pressure was continuously monitored through a
right femoral arterial cannula connected to a Statham pressure transducer (p23 Gb).

Fourteen dogs were bled from the cannulated left femoral artery into a cali-
brated reservoir over a 30 minute period until the mean arterial blood pressure
decreased between 30-40 mmHg. Blood pressure was then maintained at this level
for 3 hours by adjusting the height of the blood reservoir. During the hypotensive
period, the blood reservoir was frequently shaken to prevent sedimentation of blood
cells. At the end of the three-hour hypotensive period, all blood in the reservoir was
reinfused back into the animal. The experiment was terminated when blood pres-
sure decreased below 60 mmHg.

In eight of the fourteen dogs, 30-45 min after bleeding, a GOA solution was
perfused into the lumen of the segment at a rate of 1 ml/min (GOA treated) . The
six remaining dogs were not perfused with GOA (GOA untreated).

Blood samples were collected in plastic test tubes containing the prostaglan-
din synthesis inhibitor mefenamic acid (5 ug/ ml of blood). Simultaneously samples
were obtained from a femoral artery and the venous outﬂow of the isolated ileal
segment, before hemorrhage and every hour during and after the hypotensive peri-
od. Intestinal blood ﬂow was measured by timed-collection of venous outﬂow at the
time of blood sampling. Blood samples were centrifuged immediately for 10
minutes. The plasma was removed and stored at -20°C until assay. The plasma was
analysed for TXBZ and 6-keto PGF1 (the stable breakdown products of thrombox-
ane A2 and prostacyclin, respectively), using radioimmunoassay techniques.

Five control dogs were given treatment identical to dogs in the hemorrhagic
shock series except that the dogs were not bled. Blood samples were collected at

time periods corresponding to those of hemorrhagic dogs. After termination of all

33

experiments, the intestinal loop was excised, its lumen content emptied, and
weighed. The loops weighed 27 1- 2 and 29 3- 2 gm in the control and hemorrhag-
ic shock dogs, respectively. Afterwards an intestinal tissue sample was taken from
the GOA perfused ileal segment and a non-perfused adjacent segment for morpho-
logical examination of the mucosa for damage. The following grading sequence was
used to quantify the morphological changes as described by Chiu et.al (23):

1. Grade 0 - Normal mucosal villi.

2. Grade 1 - Development of subepithelial Gruenhayen’s space, usually at the
apex of the villi; often with capillary congestion.

3. Grade 2 - Extension of the subepithelial space with moderate lifting of epithe-
lial layer from the lamina propria.

4. Grade 3 - Massive epithelial lifting down the sides of villi. A few tips may be
denuded.

5. Grade 4 - Denuded villi with lamina propria and dilated capillaries exposed.
Increased cellularity of lamina propria may be noted.

6. Grade 5 - Digestion and disintegration of lamina propria; hemorrhage and

ulceration.
B l' .

Radioimmunoassay was performed using standard techniques. TXBZ ,
PGE2, PGF2 , and 6-keto-PGF1 antisera (Seragen, Inc., Boston, MA) were appropri-
ately diluted in phosphate buffered saline (0.1 N phosphate buffer, 0.9% NaCl, 0.5%
gelatin: PBSG). Aliquots (100 microliters) of the appropriate antisera were mixed
with 100 ill aliquots of either standard (Seragen) or plasma samples. The standard
solutions were made by serial dilution of the prostanoid standard with the prosta-
noid-free canine plasma, which was prepared previously by stripping the plasma

collected from mefenamate-treated dogs with charcoal.The standard solutions and

34

plasma samples were then mixed with the appropriate radioactive tracer prostanoid
(Tritiated, New England Nuclear, N. Billerica, MA), 6,000 cpm in 100 ul of PBSG,
so that the total reaction volume was 300 ul. After incubation (0.5-1h at ambient
temperature, then 16-24 h at 4 C), the protein bound material was isolated by
adding 1.0 ml of dextran-coated charcoal suspension (5 mg/ml of charcoal, 0.5
mg/ml dextran in PBSG). Within 12 min the tubes were vortexed and began centrif-
ugation for 12 min at 1,500 rpm and 0 C. The supernatant of each tube was decant-
ed into a vial containing 15 ml of scintillation cocktail (Safety-Solve, Research
Products International, Mt. Prospect, IL), and the radioactivity determined in a beta
scintillation counter (Model A300C, Packard, Downers Grove, IL). Duplicate values
were averaged and compared with a standard curve performed with each assay. The
antisera cross reactivity was as follows: 6-keto-PGF1 with TXBZ, PGD2, PGAZ,
PGAl, PGBl, PGBZ < 0.01%, PGFl (7.8%), 6-keto-PGE1 (6.8%), PGEl (0.7%),
and PGF/2 (0.6%); TXB2 with 6-keto-PGF1 , PGFl , PGF2 , PGEl, PGE2, PGDZ,
PGAZ, PGAl, PGBI, and PGBZ < 0.1%; PGF/2 with TXBZ, 6-keto-PGF1 , PGFl , 6-
keto-PGEI, PGDZ, PGBl, PGBZ < 1.0%, PGF2 (1.3%), PGEI (50%), PGA2
(6.0%), and PGA1 (3.0%); PGF2 with 6-keto-PGE1, PGDZ, PGA2, PGAI, PGBI,
PGBZ < 0.1%, TXBZ (0.5%), 6-keto-PGF1 (1.1%), PGFl (100%), PGEl (1.1%),
and PGF/2 (0.3%). Inter- and intra-assay variations were less than 12%, respectively
(n=20).

WW

The data were analysed using analysis of variance and Student’s t-test for
comparison of two paired sample means. Dunnett’s multiple comparison procedure
was used to compare the signiﬁcance of the difference between and within groups.

Statistical signiﬁcance was set at p <0.05. All values in the text are means : SEM.

RESULTS

Series I experiments were designed to determine which prostanoid might play
a role in food- and arachidonate- induced jejunal response was determined. The
arterial and jejunal venous concentrations of 6-keto-PGF1 (PG12 stable
metabolite), TXBZ (TXAZ stable metabolite), PGF/2, and PGF2 were determined
when the jejunal lumen contained normal saline (NS), food (F), normal saline plus
arachidonate (NS+AA), or food plus arachidonate (F+AA). Fig. 4 shows jejunal
blood ﬂow and oxygen uptake during placement of these four solutions. Food
produced a signiﬁcantly higher blood ﬂow and oxygen uptake than normal saline in
the lumen. Although addition of arachidonate to normal saline did not signiﬁcantly
inﬂuence blood ﬂow and oxygen uptake, addition of arachidonate to food abolished
the food-induced increases in blood ﬂow (48 1 5 vs 37 3- 4 ml. min-l. 1003-1) and
oxygen uptake (2.32 _g. 0.23 vs 1.91 3 0.23 ml.min'1.1OOg' 1). 80-29548, a
TXAZ/endoperoxide receptor antagonist, was injected i.a. when the lumen still
contained food plus arachidonate in six of the nine dogs in this series. The injection
significantly increased blood ﬂow from 36 _+ 4 to 43 j- 5 m1.min'1.100g'1, and
oxygen uptake from 1.78 1» 0.31 to 2.49 i 0.46 ml.min'1.100g'1(p<0.05). Thus
the inhibitory action of arachidonate on food-induced changes appears to result

primarily from the action of its metabolites, TXAZ and / or endoperoxides.

Fig. 5 shows the arterial and j ejunal venous plasma concentrations of fou r

prostanoids when the lumen contained normal saline, food, normal saline plus
arachidonate, or food plus arachidonate. The values in Fig. 5 are the averages of
the samples values obtained at the 9th and 15th min after jejunal placement of the
four solutions in 9 dog experiments. Arterial concentrations of all four prostanoids
did not change signiﬁcantly by changing luminal contents throughout the experimen-

tal period, but venous concentrations changed with difference in luminal contents.

35

 

 

36

Figure 4 Jejunal blood ﬂow and oxygen uptake (V02) during luminal place-
ment of normal saline (N8), food (F), normal saline plus arachidonate (NS+AA),
and food lus arachidonate (F+AA). n=9. 80 denotes the values after i.a. injec-
tion of S -29548, a TXAz/endo eroxide rece tor antagonist, when the lumen still
contained F+AA. n=6. ‘ p<0.0 relative to 8 values.

13'7

FIGURE 4

OOOOOOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOOOO.
. e e

e 3%... co.eoeeeeeeeoeeeoeeeoe.eoeeeoeeoee.

o O O O O o O O O o o o O O O O o 0 o o o o O.

DODODODODODODODODODODODODODODODODODODODODODOD

 

 

 

 

60 -r

o O

o o o
o o o o o o o

O O O o o o o o o 0 o o o
o. ODODWDODO OOODODODOOOOODODODOo

 

 

 

 

 

 

 

 

 

1. PI
1‘ Q
w 0 05
2 3

38

The venous concentration of each prostanoid, with one exception, was always higher
than the corresponding arterial concentration in all four conditions, indicating that
all four prostanoids were produced and released from the jejunum whether the
luminal contents were normal saline, food, or arachidonate in combination with
saline or food. The one exception is that jejunal venous and arterial PGF2 concen-
trations during normal saline placement were not statistically different. As com-
pared to normal saline in the lumen, food tended to increase venous 6-keto-PGF1
and signiﬁcantly increased venous TXBZ, PGFq, and PGF2 concentrations. Addi-
tion of arachidonate to normal saline signiﬁcantly increased only venous PGF2 , but
addition of arachidonate to food significantly increased venous PGE2, TXBZ and
PGF2 concentrations. The venous concentrations of two vasoconstrictor prosta-
noids, TXB2 (stable metabolite of TXAZ) and PGF2 were signiﬁcantly higher
during luminal placement of food plus arachidonate than food alone. Thus, the
inhibition of food-induced hyperemia by luminal arachidonate (Fig. 4) might be, at
least in part, due to an increase in jejunal production of these two vasoconstrictor
prostanoids.

Fig. 6 compares the rate of individual prostanoid releases under the four
experimental conditions, calculated as the product of blood ﬂow and veno-arterial
concentration difference. In the presence of normal saline in the lumen, PGIz (6-
keto-PGFI precursor), PGEZ and TXAZ (TXB2 precursor) were released signiﬁ-
cantly from the jejunum (p<0.05), but PGF2 release was minimal. The relative
magnitude of releases (in 11g min'1 100g'1) was PGIZ (24 _+_- 5) > PGEZ (16 j- 12)
> TXA2 (11 _j- 2) > PGF2 (0.93 3- 2.0). When the relative magnitude of the
release was expressed as percent fraction of the sum of all four prostanoid releases,
vasodilator prostanoids, PG12 (44.4%) and PGF/2 (31.9%), comprised 76% of the
total release during luminal presence of normal saline. Changing the luminal

content from salin to food signiﬁcantly increased the release of all four prostanoids.

39

Fi re 5 Arterial and jejunal venous plasma concentrations of 6-keto-PGF ,
T , GEZ, and PGF2 when the jejunal lumen contained normal saline (N18),
food F , normal saline plus arachidonate (NS+AA) and food plus arachidonate

(F+AA . n=9. ‘ and " p<0.05 relative to venous concentration obtained during
placement of NS and F, respectively.

4O

”Bans/ml) O-Itete-PGFI (pa/ml)

. 3 3 31%??? 3 3. 3 33 333 3 3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 ,—"I _Z‘I-
-__1
IIIl-- 2
3 2222+ EEEEI‘ 33
i§§§§§+e Essss- E}
3; EI- 3
ing
is 33310 93-3
m: 3; f 3
wrung/ml) PCEZ(P9/ﬂ'“)
333L33.1333 3 3 3 3
IIIIIIIII———-
’/////
\\\\\\\\\ °
3 IlIIl——
BBIU 4540’/‘ BEID
25‘s \ 3313
I; '3'

 

41

Fi re16 Jejunal releases of 6-keto-PGF1 , TXB2, PGE2, and PGF2 (ng
min“1 1 g' ) during luminal placement of normal saline (NS), food (F), normal
saline plus arachidonate (NS+AA), and food plus arachidonate (F+AA). n=9. ‘
and ” denote values signiﬁcantly different from those obtained during placement of
N S and F, respectively, p <0.05.

1x32(ng/min/1009)

42

G-kote-PGFI . (ml /min /100 s)

 

 

 

POI-'2“ elm/1009)
. s 8 s a

 

 

 

W+JIZI
W+SN E]

 

 

 

 

 

86
4
j

 

 

 

eaglggsgsaseasssssas
O
3313

Pcez(n9/mIn/1009)
2; 3 s a5 s s 3- a

J

9 HHﬂOL-I

43

PGIZ and TXA2 releases almost doubled, while PGEZ and PGF2 releases increased
4-fold and 8-fold, respectively. The relative magnitude of release (in ng.min'1.100g'
1) became PGE2 (60 3- 12) > P012 (40 3- 13) > TXA2 (17 3- 3) > PGF2 (s 3
2). However, vasodilator prostanoids, PGEQ and PGIZ, still comprised 77% of the
total release. Addition of arachidonate to normal saline did not signiﬁcantly alter
the releasing rates or relative magnitude of releases of the four prostanoids com-
pared to normal saline alone in the lumen. Addition of arachidonate to food,
however, decreased the release of two vasodilators, PGIz and PGF/2 , and increased
the release of the two vasoconstrictors TXA2 and PGF2 , as compared to food alone
in the lumen. Now TXA2 and PGF2 comprised 21.4% and 9.7% of the total re-
lease, respectively, as compared to 16.3% and 6.4%, with food alone in the lumen.
The relative magnitude of releases (ng.min‘1. 100g‘1) were PGE2 (37 _-_l- 13) >
PGIZ (33 j- 8) > TXAZ (20 i- 4) > PGF2 (12 ;l- 5).

The above experiments indicated that venous concentrations and jejunal
productions of the four prostanoids increased when luminal content was changed
either from normal saline (NS) to food (F) or from N8 plus arachidonate (AA) to F
plus AA. The next series of experiments was therefore conducted to determine
whether the increase in these prostanoid concentration could directly act on the
vasculature to alter vascular resistance. At first the vasoactivity of the individual
prostanoid was determined. Infusions of the carriers at the same rates as those of
prostanoids did not signiﬁcantly alter jejunal vascular resistance. As shown in Fig.
7A, U-44069, a TXA2 analog, and PGF2 increased jejunal vascular resistance (JVR)
in a dose-dependent fashion. The minimum blood concentration required to in-
crease JVR signiﬁcantly (p<0.05) was 1000 pg/ml, and 3 x 104 pg/ml blood for U-
44069 and PGF2 , respectively. Both PG12 and PGF/z (Fig 7B) caused a dose-
dependent decrease in jejunal vascular resistance. The minimum blood concentra-

tion required to cause a signiﬁcant decrease in JVR was 1000 pg/ ml for PG12 and

44

Fi re 7 Effects of local intra-arterial infusions of U-44069, a TXA2 analog,
and PG 2 (A), PGIZ and PGE (B) on jegunal vascular resistance (JVR). Resting
vascular resistance were 3.57 1; 0.73, .63 _+_ 0.36, 2.77 i 0.53, 2.18 i 0.2
mmHg/ml/min/lOOg for PGIZ, PGE2, PGF2 , and U-44069, rerspectively. n=5-9. ‘
p < 0.0 relative to control value.

XCHANGEJVR

45

FIGURE 7

100- A *
‘ . U-44069 6
804 O PGFZa /‘L ¢

0)
O

 

 

 

 

 

 

t\ I
'50'1 A P612 1 K * t
A P052 * \ s_ A
-801
-1ooI
102 103 104 105 A 105

pg/ml of blood

46
6 x 104 pg/ml for PGE2.

Systemic arterial and jejunal venous blood samples (n = 7) were taken during
intra-arterial infusion of PGIZ to determine the actual blood concentrations. The
systemic arterial concentrations of 6-keto-PGF1 were not altered by local infusion
of PGIZ, indicating that there was no recirculation of these prostanoids during infu-
sions. However, the jejunal venous 6-keto-PGF1 concentrations measured by ra-
dioimmunoassay were 31 i 7% lower than the calculated infused blood concentra-
tions. This might reﬂect dilution or tissue uptake of the prostanoid within the intes-
tinal wall. Thus the calculated blood concentrations shown in Figs. 7, 8 and 9 might
be over-estimating the actual blood concentrations by 31.7%, and the venous con-
centrations shown in Fig. 5 during luminal placement of various solutions might
underestimate the actual tissue concentrations by the same magnitudes.

Figure 8 shows the effect of intra-arterial infusions of mixtures of PGIZ,
PGFQ, PGF2 and U-44069 (a TXAZ analog) on jejunal vascular resistance. Based
on the data from Fig. 5 (venous concentration during placement of NS, F, N8 + AA
and F + AA) blood ﬂow rate, and hematocrit at the time of the infusion, the infu-
sion rates were pre- calculated to give local arterial concentrations equivalent to 0.5,
1, 2, 5, 10 and 20 times the value of the change in venous concentration when lumen
content was changed from normal saline to food, [PGs]F-NS.

As shown in Fig. 8a, i.a. infusions of the four prostanoids, to mimic changing
luminal content from normal saline to food (ﬁlled diamonds), did not signiﬁcantly
alter the jejunal vascular resistance at all infusion rates. The open diamonds repre-
senting the predicted changes in jejunal vascular resistance, were calculated from
the summation of the individual prostanoid actions shown in Fig. 7. The equations
for the regression lines for the observed and predicted data were Y = -3.310gX - 3.7,
r = 0.77, and Y = 8.9logX - 0.8, r = 0.92 (where Y = % JVR, and X = 0.5, 1.0...

20), respectively, and their slopes were not significantly different. These studies

47

Pl re 8 Effects of local i.a. infusions of mixtures of, PGF , PGI , U-44069
and PG 2 at concentrations 0.5x (times), 1X, 2X, 5X, 10X, an 20X t e value of
the increases in venous prostanoid concentrations shown in Fi . 5 when luminal
content was changed from normal saline to food ( [PGs] F-NS) G12 = 82pg/ml,
TXA = 72 pg/ml, PGE2 = 252p /ml and PGF2 = 47pg/ml in the mixture) (A),
and tzood lus arachidonate ( [ (Is FAA-NS (PGI = 82pg/ml, TXA = 72
pg/ml, P E2 = 259pg/ml and PG = 47p ml in t e mixture) (B). Fil ed dia-
monds indicate the actual observe data, w ereas open diamonds indicate the
predicted data as calculated by the summation of the individual prostanoid vascular
effects shown in Fi . 7. Resting vascular resistance was 3.41 ,+_- 0.38 and 3.63 3; 0.36
mmHg/ml/min/l g. n=9.

 

48

 

 

 

FIGURE 8
.60-
m1
0: 9—0 OBSERVED
3, o- -o PREDICTED
8 20.
3 -°
’0’
n .— o—
°L\°;.:9=—‘8‘ .
. I. . ‘ \.\
s O
-201 1
' 03x Ix 2x 5x 10x 201:
Local blood [P03] 0: multiple of
60. a A [908] F-NS
4.0. /°
:2 a
a , 1 1
Lu .0 "'
20" ’ / . ‘
,2 i
o 92/
A
-201

 

q

ﬁ ﬁ

0.5x 1X 2X 5X 10X 20X
Locol blood [P63] 0: multiple of

A [P691 FAA-NSAA

 

49

indicate that the vascular effects of these prostanoids were additive over the range
of the doses studied. When the mixture was infused at the concentrations that
simulated changing luminal content from normal saline to food plus arachidonate
(Fig. 7B ﬁlled diamonds), the mixture produced a dose-dependent increase in jeju-
nal vascular resistance. The minimal concentration required to produce a signiﬁcant
rise in resistance was five times that of [PGs]FAA-NSAA. The open diamonds in
Fig 7B show the predicted response. The equations for the observed and the pre-
dicted data were Y = 21.510gX + 6.7, r = 0.98 and Y = 26.5X + 7.9, r = 0.99,
respectively, and their slopes were not signifcantly different. The close agreement in
predicted and observed responses was also found in data shown in Fig. 9 ( Y =
23.7logX - 2.2, r = 0.98, and Y = 32.7logX - 55, r= 0.94 for observed and predicted
values). These ﬁndings further indicate that the vascular effects of these prostanoids
were additive over the range of the doses studied.

The infusion experiments described above were performed when the lumen
contained normal saline. In the next series of experiments, food was placed into the
lumen, then a mixture containing PGF2 and U-44069 was infused intra- arterially, to
mimic the change observed when luminal contents were changed from food to food
plus AA. As shown in Fig. 5, changing lumen content from food to food plus ara-
chidonate significantly increased only TXB2 and PGF2 concentrations. Figure 9
shows that placement of food (F) p51 35, signiﬁcantly (p<0.05) decreased jejunal
vascular resistance. Infusion of the the two prostanoids mixture increased the jejunal
vascular resistance in a dose dependent manner. At concentrations similar to and
two times that of [PGs]FAA-F, the infusate completely abolished the food- induced
decreases in jejunal vascular resistance. The increased resistance at concentrations
above ﬁve times that of [PGs]FAA-F were signficantly greater than the value ob-
tained during food placement alone, and the increases in vascular resistance at

concentrations above ten times were signiﬁcantly greater than the resting value.

50

Figure 9 Effects of luminal placement of food (nggilled circle) and subse-
quent local i.a. infusions of a mixture of U-44069 and F on jejunal vascular
resistance. [PGs]FAA-F at 1X was TXA : 12 g/ml and P 2 : 80pg/ ml. Resting
vascular reSlstance was 3.13 i 0.37 mmHg/ml nun/100g. n=9. ‘ and ‘ p,0.05 rela-
tive to F and control value.

FIGURE 9

 

51

 

60 1
40-
0: O
->. 20- 0/I
L35! 0/ I " 3
g 065 Q’O/i a:
—20d 1 l
t
-40.
F 0.5x 1X 2X 5X 10X 20X
.. Lcool blood [P69] as multiple of _.

A [PCS] FAA-F

52

In the next Series of experiments an attempt was made to determine the
stimulus for the food-induced increases in jejunal prostanoid production. In order to
study this, the jejunal prostanoid release was determined during luminal placement
of normal saline, 10% bile, air distension, and during a mechanical hyperemia.

Intestinal blood ﬂow (ml/min/ 100gm) during luminal placement of normal
saline (54 ;l-_ 6), 10% bile (52 i 7), or an equivalent volume of air (56 _+_; 8) was not
signiﬁcantly different compared to an empty lumen (55 i, 6).

As shown in Fig. 10, the arterial as well as jejunal venous concentration of 6-
keto-PGFI and TXB2 was not signiﬁcantly altered by luminal placement of normal
saline. This finding might indicate that normal saline added to food might not
contribute to the food-induced increases in jejunal prostanoid releases.

Fig. 11 shows the effects of luminal distension with air on arterial and jejunal
venous prostanoid concentrations. The arterial and jejunal venous concentrations of
6-keto-PGF1 (n = 6) and TxB2 (n = 4) were not significantly altered by luminal
placement of air. No effect of distension on intestinal prostanoid production might
indicate that physical placement of food in the lumen pg: 5; is not a contributing
factor for the food-induced increase in jejunal prostanoid release.

Fig. 12 shows the effects of a 10% bile solution on jejunal prostanoid produc-
tion. The arterial and jejunal venous concentrations 6-keto-PGF1 and TXB2
during luminal placement of 10% bile were not signiﬁcantly different compared to
the values obtained during NS placement. This ﬁnding indicates that the 10% bile
used in preparing the food solution might not be a direct stimulus for food-induced
increases in prostanoid production. However the bile in the food mixture might
have an indirect effect on food-induced jejunal prostanoid production. Previously it
has been shown that bile added to food enhanced the food-induced vascular and
metabolic changes, possibly through facilitation of nutrient absorption, whereas bile

alone did not (82).

53

and T when the jejunal umen is empty (unfilled bars) or ﬁlled with normal

ﬁre 10 Arterial and ljejunal venous plasma concentrations of 6-keto PGFl
saline ( ed bars).

Paosrmom cone. ( pg/ml )

54

FIGURE 10

1000-I
800 _ 1:1 EMPTY LUMEN

1 - NORMAL SALINE IN LUMEN
600--

400-

o-—}—o
Fil-t

200 4

. l‘lL_ - 1

 

 

 

 

 

 

 

ARTERIAL VENOUS ARTERIAL VENOUS

m . W

6-keto-PGF1a TX82

55

Fi re 11 Arterial and jejunal venous plasma concentrations of 6-keto PGF1
3nd gl'XE; when the jejunal lumen is empty (unﬁlled bars) or ﬁlled with air (ﬁlled
ars .

PROSTANOID CONC. ( pg/ml )

56

FIGURE 11

 

1000 .I-
800" E EMPTY LUMEN
- AIR IN LUMEN
600"
400“ II 'II l
200" H
0d — .—

 

 

 

 

 

 

 

 

 

 

ARTERIAL VENOUS ARTERIAL VENOUS

W W

6—keto—PGF1 a TXBZ

57

Figure 12 Arterial and jejunal venous lasma concentrations of 6-keto PGFl
Ethel; the jejunal lumen contained normal s ine (unﬁlled bars) or 10% bile (ﬁlled
ars .

PROSTANOID CONC. ( pg/ ml )

58

FIGURE 12

2000--

1600-

1200‘

800-

4003

 

E NS IN LUMEN
- BILE IN LUMEN

 

 

 

mL

 

 

 

 

 

 

 

 

O

ARTERIAL VENOUS
W

6-keto—PGF1 a

ARTERIAL VENOUS
W

TXBZ

59

During nutrient absorption one of the fundamental changes reported to occur
is an increase in intestinal blood ﬂow. The effects of a mechanically-induced in-
crease in intestinal blood ﬂow on jejunal prostanoid production was studied in series
VII. The objective was to determine whether the food-induced increases in jejunal
blood ﬂow is the stimulus for enhanced prostanoid production.

Figure 13 shows jejunal blood ﬂow and perfusion pressure before (control)
and after pump-induced hyperemia (mechanical hyperemia). The control jejunal
blood ﬂow was 37 i 5 ml/min/ 100gm. Blood ﬂow was then increased to 83 _+_ 7
ml/min/ 100gm utilizing a pump. Accompanying the increase in jejunal blood ﬂow
was a significant increase in perfusion pressure from 95 1; 5 to 176 i 10 mmHg.
The jejunal vascular resistance decreased from 2.56 mmI-IG/ml/min/ 100gm (con-
trol) to 2.12 mmHg/ml/min/ 100gm (mechanical hyperemia). This demonstrates a
ﬂow dependent vasodilation in the intestine.

The arterial and jejunal venous concentration of PGE2 6-keto-PGF1 ,
TXB2, and PGF2 before and after the mechanically-induced hyperemia are shown
in Fig. 14. The arterial concentrations of the four prostanoids were not signiﬁcantly
different during control and mechanical hyperemia. Though not statistically signiﬁ-
cant, the 6- keto-PGFl and TXB2 concentrations tended to decrease while the
PGF/2 concentrations tended to increase. The intestinal venous PGF2 concentration
was signiﬁcantly (P<0.05) decreased. The two-fold increase in jejunal blood ﬂow is
expected to decrease the jejunal venous concentrations of all the four prostanoids
with the same magnitude. However the ﬁndings indicate that the increase in jejunal
blood ﬂow affected the venous concentrations of each prostanoid differently. This
might indicate that the jejunal synthesis of PGE2, PGI2 and TXA2 is increased
while PGF2 is not altered.
The jejunal rate of prostanoid production as calculated from the products of

blood ﬂow and veno-arterial concentration differences is shown in Fig. 15 . The rate

60

Figure 13 Jejunal blood ﬂow and perfusion pressure before (control) and
after pump-induced jejunal hyperemia (mechanical hyperemia). Asterisks indicate
signifcant change from corresponding control value p < 0.05.

BLOOD FLOW (ml/min/l 009m)

FIGURE 13

 

 

 

61

- MECHANICAL HYPEREMIA

 

 

 

 

200?

160v 1:100NTR0L
120+

sou .

» WI

0 ,

BLOOD FLOW

- 200

«160

«3120

w 80

 

 

PERFUSION PRESSURE

PERFUSION PRESSURE (mm. Hg)

62

Figure 14 Arterial and jejunal venous C{alasma concentrations of PGE2, 6-keto
PGF , PGF and TXsz)before (control) an after pump-induced jejunal hyperemia
(mec am' hyperemia . Asterisks indicate signifcant change from corresponding
control value p < 0.05.

63

FIGURE14

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

as). u 6.80 3.8%.:

 

64

Figure 15 'ejunal releases of P612, TXAZ, PGF2 and PGE2
Engmin' .IOOgrn‘lg before (control) and after pump- induced Jejunal hyperemia
mechanical hyperemia). Asterisks indicate sigmfcant change from corresponding
control value p < 0.05.

eremia
ondmg

65

FIGURE 15

O
O
l
I

80"L a CONTROL
- MECHANICAL HYPEREMIA

( numb/1009"!)

N
O

INTESTINAL PROSTANOID RELEASES

 

 

 

 

 

 

Pong m2 PGF2, PGE2

66

of jejunal releases of P612, TXAZ and PGF/2 were significantly increased while the
PGF2 release was not altered. These ﬁndings indicate that part of the food-induced
increases in jejunal prostanoid production might be due to the food- induced in-
creases in jejunal blood ﬂow.

The next series of experiments were performed to determine the role of
TXA2 on food-induced vascular and metabolic changes. To study this, U-44069 and
8029548 (a TXAZ receptor agonist and antagonist),respectively were used. At
ﬁrst the jejunal vascular response to the exogenously administered thromboxane
A2 analog and receptor blocker was determined. Fig 16 is a representative tracing
showing the effect of U-44069 on jejunal perfusion pressure before and after admin-
stration of SQ-29548. U-44069 caused a significant increase in jejunal perfusion
pressure. Adminstration of SQ- 29548 (0.8 ug) significantly attenuated the U—44069-
induced increases in jejunal vascular resistance.

Fig. 17 shows the effect of graded doses (0.05-2.0 ug) of a thromboxane analog
(U-44069) on % changes in jejunal vascular resistance before and after intra-arterial
administration of 0.1, 0.6, and 2.0 ug of a thromboxane receptor blocker (SQ-29548).
U-44069 produced an increase in jejunal vascular resistance, that was rapid in onset,
in a dose-related fashion. U-44069 in the minimum dose (0.05 ug) used caused a
significant (P<0.05) increase in jejunal vascular resistance. The maximum increase
in vascular resistance was observed at a dose of 1.5 ug; thereafter any increase in the
dose of U-44069 administered did not further enhance the jejunal vascular resist-
ance. The U-44069 vehicle, ethanol, did not alter the jejunal vascular resistance.
Administration of 0.1 and 0.6 ug of SQ-29548 prior to U-44069 significantly
(P< 0.05) attenuated and at a dose of 2.0ug completely abolished the effect of U-
44069. This blocking action lasts 40-60 min. SQ-29548 p521: 5; had no effect on
resting intestinal vascular resistance.

In Series IX, the effects of thromboxane receptor blocker on food-induced

 

67

Figure 16 Original polygraph chart recording showing the effect of U-44069
on je'unal perfusion pressure e ore and after adminstration of SQ-29548 (a TXAZ
/ en oproxide receptor blocker).

68

Eocmm Am

 

e: «an; IV

       

4.
$225.0: «agate
vqommeqo k.

+ w

.

 

Ctuaoac mbtncmao Cubaoao
. .95 0.92.. . .95
2.9.60.3mv 2.9.60.6wv 2.9.Gozav
F L

 

69

Figure 17 The effect of intra-arterial bolus administration of graded dose: of
a TXA analog (U44069) on percent changes in jejunal vascular resistance before
(unﬁllé circles) and after intra-arterial administration of 0.1 ug (ﬁlled circle) , 0.6
u (unfilled triangles) and 2.0 u ; (filled triangles) a TXAZ recegtor blocker (SO-
2 548). Asterisks and crosses in icate signiﬁcant differences (p< .05) from control
values and corresponding U-44069 values (unfilled circles).

 

7O

96 CHANGE IN VASCULAR RESISTANCE

 

395m AV

OIO Clugomo

 

 

 

 

 

 

.mo. eIec+mo 9:5
DIbc+mO Pate *
I i
##O .. V P C+m0 thb 0/0
.w \— —
‘00.
*+
mo- ...+ I
lethtIIIt »
h+ ...+
NO 1
D F
Inc A. 4
d m Nb

 

Clkgcmo

71

changes in jejunal blood ﬂow and oxygen uptake were determined. Fig. 18 shows
jejunal blood ﬂow during luminal placement of normal saline and food plus bile
before and after intra-arterial administration of 2.0 ug SQ-29548 . Before adminis-
tration of SQ-29548, placement of food plus bile in the jejunal lumen significantly
increased (p<0.05) jejunal blood ﬂow by 34 j- 6%. After administration of SO-
29548 jejunal blood ﬂow measured during placement of normal saline was not sig-
niﬁcantly different compared to the corresponding value obtained before $029548
(46 _+ 4 before and 45 _f 6 after). After administration of SCI-29548, food plus
bile significantly increased (p<0.05) jejunal blood flow by 26.7%. This increase was
not significantly different from the corresponding value obtained before SQ-29548.

J ejunal segment oxygen consumption during nutrient absorption before and
after administration of 8029548 is shown in Fig. 19. The magnitude of oxygen
consumption during normal saline before and after SQ-29548 was not significantly
different. Before SQ-29548, food increased jejunal oxygen consumption by 28 _+
7%. After SQ-29548 food increased jejunal oxygen consumption by 51 j- 11%.
The increase after SQ-29548 was signiﬁcantly (P< 0.05) greater than before.

In Series X, the effect of a thromboxane A2/endoperoxide receptor blocker
(SQ-29548) on food-induced changes in capillary filtration coefficient was deter-
mined. Figure 20 shows the effects of luminal placement of digested food on jejunal
capillary filtration coefficient (Kfc) before and after administration of 8029548,
(2.0 ug i.a. bolus). Placement of food in the lumen significantly increased (p<0.05)
jejunal Kfc from 0.386 1- 0.041 to 0.508 1- 0.047 ml/min/mm.Hg/100gm. Admin-
istration of 5029548 per: as: did not alter resting jejunal Kfc, but significantly poten-
tiated the food-induced increases in jejunal Kfc from 0.416 3- 0.043 to 0.633 3-
0.107 ml/min/mm.Hg/100gm. During nutrient absorption, the absolute and the
percent changes in jejunal Kfc due to SQ-29548 (0.262 _+ 0.089
ml/min/mm.Hg/100gm, 63.3 3- 22.5%), respectively, were signiﬁcantly greater

72

Figure 18 Jejunal blood flow before and after administration of SQ-29548
durin normal saline (unfilled bars) and food plus bile (filled bars) in the lumen.
Astensks indicate signiﬁcant increases (p < 0.05) above normal saline values.

73

-Blood Flow ( ml/min/IOOgm. )

mo-

we-

mot

mol

so-

me.

No

I

If

I

I

 

locum mm

D 2035. 3:3
I moon + mzm

.w

 

bﬁ

 

 

 

 

€an

 

 

 

 

 

@633 mOINmuém

 

>29. mOlmmmam

74

Figure 19 J ejunal oxygen consumption before and after administration of $0-
29548 in the presence of normal saline (unfilled bars) and food plus bile (ﬁlled
bars). Single and double asterisks indicate signiﬁcant increases (p<0.05) compared

to normal saline and food plus bile values before adminstration of 8029548, re-
spectively.

7S

Eocmm do

 

D 2035. 8:3
I moon + gm

9an

 

 

 

 

\/ Pm --
m
g
m we .ﬁ
w
.m Nm %
V
/m\ No ..
2
O
V fm ..
#0

+

 

 

 

 

 

wmﬁoqo melmmmsam

>23 melmmmbm

76

Fi re 20 Jejunal capillary ﬁltration coefﬁcient before and after administra-
tion of 029548 in the presence of normal saline (unﬁlled bars) and food plus bile
(filled bars). Single and double asterisks indicate significant increase <0.05)
above normal saline and food plus bile values before adminstration of S -29548,
respectively.

77

Eocmm N0

 

 

0.0004 D2035. mezzo In
.t . 'ﬁooa +026
m \/ Aaﬂmv
m m
H Aw. 0.80:
e O *
O I
C /
n g
o H
w m 030-- 0 * f
.v m _
m w

m

“WV oboe:
.m. m
o (
C

0.000

 

 
 

 

 

 

 

 

 

 

0638 melmmmﬁw >29. melmmmém

78

Figure 21 Food-induced ercent changes in Ilejunal oxygen u take, capillary
filtration coefficient and bloo flow before (unfi led bars) and a er (ﬁlled bars;
administration of SQ-29548. Asterisks indicate signiﬁcant differences (p<0.05
from corresponding values before SQ- 29548.

 

79

300mm N.

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

 

.00 -
. D0208 monmmmsm ..
'28.. melmmmam ..
s
m. m0: 9..an
n
nu
.m 4
% 00.:
.nu
.
d #0--
.m
d H
m we...
0
05.00: 503.8 000:5 3.00:0: 0.000 :0:

0002.203

apilary
.ed bats)
p<OwI

80
(p<0.05) than the corresponding values before SQ-29548 ( + 0.086 _j- 0.021
ml/min/mm.Hg/100gm, +228 _1- 6.4%; P<0.05).

In series XI (n=3) U-44069 (thromboxane A2 analog), norepinephrine (NE),
isoproteranol (I) and adenosine (A) were infused i.a. when luminal food had already
raised jejunal Kfc. U-44069, NE and A decreased Kfc from 0.44-0.49 to 0.25- 0.31
ml/min/mm.Hg/100gm, while I enhanced jejunal Kfc from 0.48 to 0.77
ml/min/mm.Hg/100gm. Thus a food induced increase in jejunal Kfc could be
altered by chemicals.

Figure 21 shows the effects of 8029548 on food- induced percent changes in
jejunal oxygen uptake, capillary filtration coefficient and blood flow in a summa-
rized form. After treatment with SQ-29548 food-induced increases in jejunal oxygen
uptake and capillary ﬁltration coefﬁcient were signiﬁcantly potentiated (p<0.05);
however, there was no change in blood flow.

An attempt was made to elucidate the mechanism by which the TXA2 receptor
blocker potentiated the food-induced increases in jejunal oxygen consumption and
capillary exchange capacity without a change in jejunal blood flow. One likely
mechanism is a redistribution of blood between mucosal and muscle layers of the
jejunal wall by endogenous TXAZ. In order to study this possibility the jejunal wall
blood ﬂow, distribution to each layer was measured utilizing a rnicrosphere tech-
nique during infusion of U-44069 and 80- 29548. The studies were conducted under
free-ﬂow (series XII) or constant-ﬂow (series XIII) conditions. Systemic arterial
blood pressure (123 j- 4 mmHg) was unaffected by local infusions of U-44069, $0-
29548 and microspheres. Figure 22 shows mucosal plus submucosal (mucosa) and
muscularis plus serosal (muscle) blood ﬂows at rest (control), and during intra-
arterial infusion of U-44069 (0.16 ug/min) with and without prior administration of
SQ- 29548 (2.0 ug i.a. bolus) under a free flow condition. At rest, mucosal and

submucosal blood flow was 0.72 1» 0.10 ml/min/gm and muscularis plus serosal

81

Fi re 22 Mucosal plus submucosal mucosa) and muscularis plus serosal
(musclef‘blood flows during i.a. infusion 0 044069 before and after SQ-29548
administration. Asterisks and crosses indicate signiﬁcant differences compared to
‘cfontrfpl and U-44069 values respectively (p<0.05). Jejunal segment was perfused at
ree ow rate.

82

FIGURE 22

EControl
1.00 -_ - U—44069
E + EXZlU-44069+SQ-29548
saI °‘ 0.80--
$53543 E .
to .-
33.. a +
\ 0.60"
_E.
‘3’ 0.40--
O
_1
L1. is
8 0.201-
O
_I
m
0.00

 

 

 

 

 

 

 

 

 

 

MUCOSA MUSCLE

83

Figure 23 Mucosal lus submucosal (mucosa) and muscularis plus serosal
(muscle) blood flows during i.a. infusion of U-44069 before and after SQ-29548
administration. Jejunal segment was perfused at constant flow rate. See text for
statistical analysis.

84

FIGURE 23

BLOOD FLOW ( ml / min / gm )

 

 

 

 

 

 

 

 

1.00 T
G Control
- U-44069
0.80 -- m U-44069+SQ-29548
0.60 -- FI-
0.40 '4- +
0.20 *-
I100

MUCOSA MUSCLE

85
blood flow was 0.64 3- 0.09 ml/min/ gm. Infusion of U-44069 signiﬁcantly (p<0.05)
decreased both mucosal and muscularis blood ﬂow to 0.20 1- 0.03 and 0.28 1- 0.05
ml/min/gm respectively. The percent decrease in blood ﬂow to the mucosa (~67 i“
5%) was signiﬁcantly higher than the muscle (-53 j- 6%) (p<0.05). Administration
of 8029548 (a thromboxane A2 receptor blocker) blocked the U-44069-induced
decreases in blood ﬂow. SQ-29548 signiﬁcantly increased (p<0.05) mucosal blood
ﬂow to 0.79 1- 0.07 ml/min/gm and muscle blood ﬂow to 0.52 i- 0.05 ml/min/gm.
The percent increase in blood flow to the mucosa (+329 1- 60%) was signiﬁcantly
greater than to the muscle (+165 _j- 75%) (p<0.05). Even though the mucosal
plus submucosal and muscularis plus serosal vascular bed of the jejunal layers are
affected by the thromboxane A2 analog, the response of the mucosal plus submu-
cosal layer is signiﬁcantly (P< 0.05) greater than the muscularis plus serosal layer.
Figure 23 shows, under constant flow conditions, mucosal plus submucosal

(mucosa) and muscularis plus serosal (muscle) blood ﬂow at rest (control) and
during U-44069 infusion before and after administration of SQ-29548. Administra-
tion of U-44069 during constant ﬂow conditions decreased mucosal and increased
muscularis blood ﬂow. The percent decrease in mucosal blood ﬂow (-19 j- 9%)
was signiﬁcantly (P<0.05) different from the percent increase in muscularis blood
flow (+52 _+ 22%) (p<0.05). Administration of 8029548, while U-44069 was
being infused, increased mucosal and decreased muscularis blood flow. The percent
changes in mucosal plus submucosal blood flow (+34 i- 17) was signiﬁcantly dif-
ferent from the corresponding values of the muscularis plus serosa (-24 1- 14) after
SQ-29548 (p<0.05). These actions of U-44069 and SQ-29548 under free and con-
stant ﬂow conditions might indicate that endogenous thromboxane A2 induces a
greater vasoconstriction in the mucosal plus submucosal vascular bed than the
muscularis plus serosal vascular bed.

The next Series of experiments were designed 1) to determine relative rates of

86

production of TXAZ and PG12 during various stages of severe hemorrhagic shock
and after blood transfusion, 2) to determine whether the small intestine is a signiﬁ-
cant site for increased prostanoid production and 3) to determine whether the
possible mechanism of the benefical effects of diet, particularly a Glucose Oleic
Acid (GOA) solution, in severe hemorrhagic shock is related to its effect on prosta-
noid production. Previous studies have shown that placement of GOA into the intes-
tinal lumen prevent intestinal mucosal damage and increase survival of animals in
severe hemorrhagic shock (71,72). Intestinal TXAZ and PGIZ production during
various stages of severe hemorrhagic shock and following blood transfusion were
determined to study this.

Three groups of dogs (n= 18), control (sham- operated), GOA treated hemor-
rhagic shock dogs (GOA + shock), and GOA untreated hemorrhagic shock dogs
(shock) were used in Series XIV. The mean arterial blood pressures (MABP) of
untreated hemorrhagic shock (n=6), sham-operated control dogs (n=5), and GOA-
perfused shock dogs (n=8) are illustrated in Figure 24. The resting blood pressures
in all groups were comparable (126-132 mmHg). The MABP in the sham-operated
control dogs did not change signiﬁcantly during the 5 hour experimental period. In
both hemorrhagic shock groups, the blood pressure was maintained between 30—40
mmHg for 3 hours by controlled bleeding. In untreated and GOA-treated shock
groups , after all the remaining shed blood was reinfused at the end of three hours,
the MABP rose and was 87.5 1» 15 and 73 _+_ 10 at hour (hr) 4 and 74 1- 15 and 65
_j- 8 mmHg at hr 5, respectively. These values were significantly lower than the
pre-hemorrhagic and the corresponding sham-operated control values (P<0.05).
The bleeding volumes in both hemorrhagic shock dogs are shown in Table II. The
maximum bleeding volume was 43 _+ 4 and 34 j- 6 ml/kg at 1.5 hours in the
untreated and treated shock dogs respectively. In order to maintain the same

hypotensive state at hr 2 and 3, 3% and 17% of the shed blood was reinfused

87

Figure 24 Mean arterial blood pressure of severe untreated hemorrhagic
shock do (triangles) (n=6), of sham- operated control dogs (ﬁlled circles) (n=5)
and GO perfused dogs (unfilled circles) (n=8) over the time of experimental
procedure. It the hemorrhagic shock groups blood pressure was stablized between

040 mmHg for 3 hrs (03) by controlled bleeding. At the third hour all the shed
blood was transfused. About 3045 minutes after bleeding, a glucose oleic acid
(GOA) solution was perfused into the lumen of the segment at a rate of 1 ml/ min in
the GOA-treated group. Single and double asterisks denote signiﬁcant differences
relative to prehemorrhagic, and corresponding control values respectively p <0.05.
Each point represents mean i SEM.

88

FIGURE 24

 

 

 

 

K L
C
O
H
S
M...“ .F
AAO
HOH
SGS
GOA
._ .F
. H
GOA -
T w n u +
O O O 0 O O
O 5 2 8 4
2 1 1|
OISE

mmDmmumn. 00040 .2355? 2(02

1

-0.5 0

TIME ( hrs. )

Bleeding Volumes in GOA treated and Untreated Hemorrhagic
Shock Dogs

Time (h) (after MABP was maintained between ;_ and 4_ mm HG)

‘Volume 0 0.5 l 1.5 2 2.5 3

m_1. --------------------------------------------------------
GOA 419 631 706 728 667 729 610
:70 :125 :122 :138 :142 :128 :129

SHOCK 395 530 646 661 641 620 551
:51 :58 :58 :78 :90 :90 :85
mlgxg ----------------------------------------------------
GOA 19 28 33 34 32 35 29
:2 i5 :5 :6 :5 :5 :5

SHOCK 26 34 42 43 41 40 35
+4 +4 +3 :4 +5 :5 :4

All values are means : SEM

90

TABLE III

Heart Rate Values for GOA treated and untreated Hemorrhagic
Shock and Sham-Operated Dogs ( mean : SEM )

Heart rate 1 beats/min 1

Time (h) GOA Shock Sham

-0.5 165 i 14 160 i 6 157 : 6

1 171 i 17 160 i 12 163 i 9

2 216 i 15 152 i 13 152 i 13
3 195 : 19 146 : 14 153 : 14
4 194 i 15 139 i 8 158 1 10
5 207 : 10 147 : 8 164 : 13

TABLE IV

Hematocrit Values for GOA treated and untreated
Hemorrhagic Shock and Sham-Operated Dogs ( mean : SEM )

Hematocrit 1 vol. 3 1

Time :h: GOA Shock Sham
-o.5 43 : 1 36 : 3 35 : 2
1 41 : 2 39 : 3 33 : 3
2 41 : 2 37 : 3 36 : 2
3 41 : 2 40 : 3* 37 : 2
4 44 : 3 46 : 3* 37 : 2
5 45:3 47:3 37:2

Values significantly different from time -0.5 value

91
(decompensation stage of shock). The heart rate and hematocrit values for both
groups are shown in Table III and IV. The only signiﬁcant change from control (-0.5
hr) in the three groups was an increase in hematocrit at hr 4 and 5 in untreated
hemorrhagic dogs.

Fig. 25 shows intestinal blood ﬂow during the experimental period in control
and GOA-treated and untreated hemorrhagic shock dogs. The blood ﬂows in all
groups at the initiation of the experiment were comparable (62.5 1- 11.8 vs 63.6 1
5.5 vs 74 :- 12 ml/min/ 100 gm tissue). In the untreated hemorrhagic shock group,
blood ﬂow fell significantly to 9.1 _+ 1.2 ml/min/ 100 gm tissue during the three-
hour hemorrhagic period. When the remaining shed blood was reinfused, blood ﬂow
increased signiﬁcantly one hour after the reinfusion to a level (51.8 :- 14.6) which
was not signiﬁcantly different from the prehemorrhagic value (hr 4, Fig. 25). Flow
then decreased signiﬁcantly to 29.5 i- 5.4 ml/min/ 100 gm tissue one hour later. In
the GOA-treated shock group, blood flow fell significantly to 14.6 _+ 2.2
ml/min/ 100gm during the hemorrhagic period. One hour after reinfusion of the
remaining shed blood , flow increased to 33 j 5 ml / min/ 100gm, which was signif-
icantly lower than the value at -0.5 hr. Flow then decreased to 26 ml/min/ 100gm
one hour later. Intestinal blood ﬂow of the control group did not change signiﬁcant-
ly until hr 4 and 5. These two values were signiﬁcantly lower than the -0.5 hr value,
but were not signiﬁcantly different from the corresponding hr 4 and 5 values of the
hemorrhagic dogs.

Fig. 26 shows the intestinal vascular resistance (IVR) in the GOA treated and
untreated hemorrhagic shock dogs and sham-operated control dogs during the
experimental periods. The inital IVR values were comparable in all groups. In the
GOA-treated and sham-operated dogs the IVR did not signiﬁcantly change until hr.
5, at which time the IVR signiﬁcantly increased in the sham-operated control dogs

compared to pre-hemorrhagic values. In the untreated-shock dogs, the IVR

92

Figure 25 Intestinal blood flow of untreated hemorrhagic shock dogs (trian-
les , sham-operated control dogs (filled circles) and GOA-treated shock dog:
I led circles) during the 5 hrs experimental period. single and double asteris
denote signiﬁcant differences relative to prehemorrhagic, and corresponding control
values, respectively p < 0.05. Each point represents mean 1; SEM.

93

FIGURE 25

.._ a.
.681 :5
‘6
.....
. -
IA.§.NUI LIA.
..._.
. O ... \I/
. .30
a. .tnvnet mm
K ..
by x ’m /\
mm . e... .. r We
SK v. .I 8H- 2 a.“
mum . H
HOH \ m
5.65 4 2.. ”0
.OA .. 0 0A #1.
. u . ”is
. m . ”
GOA . a... .4

 

 

 

 

100?

4.. mu m. 0
7 5 2
E000 — \:_E\.E

30...”. 000.5 ..<Z_._.mm:z_

94

Figure 26 Intestinal vascular resistance of untreated hemorrhagic shock dogs
(triangles), sham-operated control dogs (ﬁlled circles) and GOA-treated shock do
unﬁl ed circles) during the 5 hrs experimental period. single and double asteris
and crosses denote signiﬁcant differences relative to prehemorrhagic, corresponding
control and GOA-treated shock values respectively p < 0.05.

INTESTINAL VASCULAR RESISTANCE
PRU

95

FIGURE 26

0—0 Shock

 

 

 

1o“ 0- -O Sham
Am A GOA + Shock +43
5..
61-
4"
2d.
0 F r 'e t s
-1 0 1 2 3 4

96

 

 

 

FIGURE 27
0—0 NS long seq.
1‘ 300~~ A—AGOA long seg.
Q O—O GOA short seq.
0"
E
I5” 2004-
0
o
D
_l
0 VA.
3 100:
0: 0. ..
10
h ‘
g " .1
0 e .L e e l :
-1 0 1 2 3 4 5
TIME ( hrs. )

Figure 27 Arterial blood lucose concentration in hemorrhagic shock dogs
treated with normal saline (unfi led circles) or GOA (trian les) perfused through
the entire small intestinal lumen (from previous study ( 1)) or GOA perfused
through a 5mm segment (ﬁlled circles) (present study).

 

97
progressively increased during the hypotensive period. The values obtained at hrs 1,
2 and 3 were significantly higher than the pre-hemorrhagic values and the corre-
sponding sham-operated and GOA-treated values. Upon transfusion of the remain-
ing shed blood the IVR returned back to pre-hemorrhagic values (hrs. 4 and 5).

The time course of changes in arterial blood glucose concentration of the
GOA-treated shock dogs perfused through an isolated intestinal segment (GOA
short seg.) is shown in Fig. 27. Plotted on the same ﬁgure is the data obtained from
a previous hemorrhagic shock study (71,72) in which normal saline (NS long seg.),
or GOA (GOA long seg.) was perfused through the entire small intestinal lumen.
In all groups, the prehemmorhagic blood glucose levels were comparable. During
the hypotensive period a signiﬁcant increase in blood glucose levels was observed in
all groups (hr.1). After the inital hyperglycemia peak, the blood glucose levels
progressively decreased in all groups. The normal saline-treated groups developed
severe hypoglycemia at hrs. 4 and 5. GOA perfusion through an entire or short small
intestinal segment prevented the severe hypoglycemia.

The concentrations of the stable metabolite of prostacyclin, 6-keto-PGF1 , in
the arterial and intestinal venous plasma are illustrated in Fig. 28 and Fig. 29. In the
sham-operated control, the concentration did not change significantly throughout
the experimental period, while in the untreated hemorrhagic shock group, the
concentrations tended to increase at hr 3 and 4 in the arterial plasma (Fig 28) and
at hr 2 and 3 in the venous plasma (Fig 29) compared to the pre-hemorrhagic values
(-0.5 hour). However, the transient increase in the venous concentration was fol-
lowed by a return to prehemorrhagic values during the post transfusion period (hr 4
and 5). In the GOA-treated shock group, the arterial concentration increased signif-
icantly at hrs 3, 4, and 5 compared to the prehemorrhagic concentration. The values
at hrs 4 and 5 were significantly higher compared to untreated shock and sham-

operated control dogs. Intestinal venous 6-keto PGFl concentration signiﬁcantly

98

Fi re 28 The arterial 6-keto PGF1 (stable metabolite of P61 ) concentra-
tion (11 ml) in the untreated hemorrhaglc shock dogs (triangles), sham-operated
contro dogs (filled circles) and GOA-treated shock dogs (unfilled circles) durin
the 5 hrs experimental period. single and double asterisks and crosses denote signi -
icant differences relative to prehemorrhagic, corresponding control and untreated
shock values respectively p < 0.05. Each pomt represents mean 1; SEM.

Figure 29 Intestinal venous 6-keto PGFI concentration (ng/ml) in the un-
treated emorrhagic shock dogs, sham-operated control dogs and GOA-treated
shock dogs during the 5 hrs experimental period. Sin 1e and double asterisks and
crosses denote significant differences relative to pre emorrhagic, corresponding
control asrgl 1\étntreated shock values respectively p <0.05. Eac point represents
mean i .

:he 1':-
treazec
@003

resesﬁ

99

FIGURE 28

8.0 T‘

7.0 4..
6.0»
_ 5.04-
E
\ 4.0--
2’
3.0--
2.0--

1,0...

 

ARTERIAL 6-KEI'0 POI-'1“ CONCENTRATION

O- -OSHAM
O—O GOA + SHOCK

Am- ASHOCK

 

 

0.0 -—-r .

-1

 

FIGURE 29 TIME ( hrs.)

 

 

 

8.0T VENOUS O-KEIO PGF-1“ CONCENTRATION
7'0" e- -OSHAM +‘3
' AH" ASHOCK

_. 5.0 --

E

\ 4.0 --

2’
3.0--
2.0“. ........
1,‘)-"' 2“ ......... lal::
0.0 i i I r .L I

-1 ‘I 2 3 4 5

100
increased at hrs 2, 3, 4, and 5, compared to prehemorrhagic values. The concentra-
tions at hrs 3, 4, and S were also signiﬁcantly higher than sham-operated control and
untreated shock values. In all groups the 6-keto-PGF1 venous concentrations were
higher than the corresponding arterial concentrations.

Fig. 30 and Fig. 31, respectively, shows arterial and intestinal venous plasma
concentrations of thromboxane B2, a stable metabolite of thromboxane A2, in the
treated shock, untreated shock, and sham-operated control groups. Arterial and
venous TXB2 concentration in the sham-operated control dogs did not show a sig-
niﬁcant change during the entire experimental period. In the untreated hemorrhagic
shock dogs, however, both arterial and venous plasma TXB2 concentrations signiﬁ-
cantly increased at 1, 2, 3, 4, and 5 hours (P<0.05). The peak values measured were
5.15 and 3.26 ng/ml in the intestinal venous and arterial plasma, respectively, at hr
5, (i.e., 2 hours after blood transfusion). The TXB2 concentrations in venous sam-
ples obtained between hr 2 and 5 were signiﬁcantly higher than those of the corre-
sponding arterial samples obtained at the same time. This indicates that TXA2 was
being produced and released from the intestine during that period. In the GOA-
treated animals the arterial TXB2 concentrations increased at 1,2,3,4,and 5hrs
compared to the prehemorrhagic and sham- operated dogs. The magnitude of
increases in the arterial TXB2 concentrations were similar to the corresponding
untreated shock dogs. Intestinal venous TXB2 concentrations in the GOA-treated
groups were signiﬁcantly increased at hrs. 2,3,4 and 5 compared to prehemorrhagic
value. The values at hrs. 4 and 5 were signiﬁcantly greater than the corresponding
sham-operated values. However, the magnitude of increases at 2,3 and 5 hrs were
signiﬁcantly lower than the untreated shock dogs.

In two untreated hemorrhagic shock dogs, the experiment extended for
another 3 hours. Arterial TXB2 concentration increased from 2.6 and 4.6 at hr 5 to

5.1 and 9.1 ng/ml at hr 8, respectively, while the intestinal venous concentration

101

Figure 30 The arterial TXB2 (stable metabolite of TXAZ) concentration
(ng/mlﬁ) 1n the untreated hemorrhagic shock dogs (trian les), sham-o rated control
dogs ( lled circles) and GOA-treated shock dogs (unﬁ ed circles) uring the 5 hrs
expenmental period. single and double asterisks denote signiﬁcant differences rela-
tive to prehemorrhagic and corresponding control values respectively p <0.05. Each
point represents mean i SEM.

Figure 31 Intestinal venous TXB concentration (n / ml) in the untreated
hemorrhagic shock do , sham- o erate control do 5 and OA-treated shock dogs
durin the 5 hrs expenmental period. Single and dou le asterisks and crosses denote
signi rcant differences relative to prehemorrhagic, corresponding control and GOA-
treated shock values respectively p < 0.05. Each point represents mean _4; SEM.

Icentraﬁrzz
ated 0::
mg the 5':
:rences re:
< 0.05. E;

untreazsi
I shock 9.3
ISSCS daze“:
I and 00‘?
, SEM.

102

FIGURE 30
8.0 - ARTERIAL TXBZ CONCENTRATION

7.0 ~-

*I

O- -OSHAM

6.0" 0—0 GOA + SHOCK
A---- ASHOCK

.. 5.0--

E

\ 4.0:

g1 .
3.0.
2.0.
1.0 4
0.0 1

 

 

 

TIME ( hrs. )

FIGURE 31

8 O VENOUS 1X82 CONCENTRATION
. ar
7 0 O- -OSHAM

' "' O—OGOA + SHOCK

A. . . .
6.0 .. ASHOCK 4... .

E 3.1 ......... 1

\ 4.0--

.“ .N 9‘.
0 o o
A l l
l I I

 

 

 

 

103
increased from 6.4 and 7.5 at hr 5 to 9.5 and 13.2 at hr 8, respectively. Arterial 6-
keto-PGFI changed from 0.14 and 0.22 at hr 5 to 0.30 and 0.27 at hr 8, while venous
concentration changed from 0.84 and 0.30 at hr 5 to 1.20 and 0.84 ng/ ml at hr 8.
This indicates that the TXB2 concentration continued to increase while the 6-keto-
PGFl concentration remained relatively steady for another 3 hours.

In three experiments, blood samples were obtained from the arterial reser-
voir at hour 3 before reinfusion started. The TXB2 concentration in the reservoir
was 0.99 _-_l- 0.2 ng/ ml, whereas the corresponding arterial plasma TXB2 concentra-
tion was 3.25 :l- 0.87 ng/ml. Concentrations of 6- keto-PGFl in the arterial reser-
voir and arterial plasma were 0.22 i- 0.06 and 0.67 ;i- 0.3 ng/ml, respectively.
This indicates that the increased arterial TXB2 and 6-ketO-PGF1 concentration
after blood reinfusion was not due to transfusion of the reservoir blood.

Thromboxane A2 and prostacyclin have antagonistic effects, and their result-
ant effects depend on the balance of the changes in their concentrations. The ratio
of TXBz to 6- keto-PGFI in arterial and intestinal venous plasmas are illustrated
in Fig. 32 and Fig. 33, respectively. In the sham-operated controls, there was no
signiﬁcant change in the ratio throughout the experimental period in either blood
samples. In the untreated hemorrhagic shock group, the arterial blood ratio in-
creased progressively during the hypotensive period, and the values at hr 1, 2 and 3
were significantly increased in the arterial blood compared to prehemorrhagic
value. The increases at hrs. 2 and 3 were signiﬁcantly greater than the corresponding
sham-operated values. The value at hr. 3 was also significantly greater than the
corresponding GOA value (Fig. 32). After blood transfusion the ratio in arterial
blood showed a biphasic change: a decrease one hour after transfusion (hr 4) and
then a marked increase one hour later, (i.e., at the terminal stage). In the venous
blood (Fig. 33), the ratio progressively increased during the hypotensive period and

following the transfusion in the untreated shock dogs. The values at hrs 1, 2, 3, 4,

 

104

Figure 32 Arterial TXBZ to 6-keto-PGF1 ratios in GOA-treated and untreat-
ed hemorrhagic shock dogs and sham- operated control dogs during the 5 hrs e r-
imental eriod. Single and double asterisks and crosses enote $1 'ﬁcant di er-
ences re ative to prehemorrhagic, corresponding control and GO -treated shock
values respectively p < 0.05. Each point represents mean i SEM.

Figure 33 Intestinal venous to 6-keto-PGF1 ratios in GOA-treated and
untreated hemorrhagic shock dogs an sham-operated control do during the 5 hrs
e erimental eriod. Single and double asterisks and crosses enote significant
di erences re ative to prehemorrhagic, corresponding control and GOA-treated
shock values respectively p <0.05. Each point represents mean _1; SEM.

ARTERIAL [ TX82 ] / [ 6-keto- PGF1a]

l. Venous [1X82] / [6-k—PGF1 a]

105

 

 

 

 

 

FIGURE 32 ..
'4".
1OT ‘4":
0—0 GOA +SHOCK .
_ O- -OSHAM e
5.. A---- ASHOCK ”
5..
4..
2..
0 4
-1 6
FIGURE 33
10v-
0- -. Sham “4
Am. A Shock I
5.. 0—0 Shock+GOA ,
+99 .A
5.. +0 '1" I .,
I l ,.A'
O
4“ '1'}. ”A ........ A
_A'" I I
2" . Y ,Q- — -O- - -l, ‘ -3
O 2 1 : i . : :
-1 O 1 2 3 4 5 6

 

106

TABLE V: Intestinal TXB release for GOA treated and untreated Hemorrhagic
Shock and Sham-Operated ogs ( mean 1 SEM )

TABLE VI: Intestinal 6-keto-PGF1 release for GOA treated and untreated
Hemorrhagic Shock and Sham-Operated Dogs ( mean 1 SEM)

Time Ihl 928 i £0923 EDQQK
-O.5 11 i 3 14 j; 11
1 -7 i 8 11 i 7
2 -5 : 4#+ 24 1 3
3 -10 : 7#+ 17 i 5
#+ *+
4 13 j; 1% 69 i 36*
5 10 i 7 + 86 i 44 +
* p<0.05 relative to -0.5 value
# relative to untreated shock
+ relative to sham operated control
IABLE VI
Intestinal M o- 1 neleaee 01323111190201
Time Ihl 998 _ ﬁhQEB SDQEE
-O.5 13 i 3 34 1 7 * 34
1 11 i 7 * 6 i 2 *+ 36
2 40 i 12 f 9 i 3 *+ 27
3 16 i 9 5 i 1 + 27
*# *4-
4 47 i 18 * 2 1: 6 * 27
5 37 i 13 i 3 i 4 + 23
* p<0.05 relative to -O.5 value
# relative to untreated shock
+ relative to sham operated control

107

TABLE V

Intestinal TXB; Release lnnAninAlennnl

21
38
22
47
24
15

H4+P+H4+I+

|+ H H I+ I+ H-

11
11

10

19
18
13
19
13
19

 

108
and 5 were significantly greater than the pre-hemorrhagic, corresponding sham
operated, and GOA-treated shock values (P < 0.05).

After a transient increase (1 hr) in the arterial blood ratio, the values in
GOA-treated hemorrhagic shock group returned back to the pre-hemorrhagic value.
However, the ratio in venous blood did not increase during the whole experimental
period and the values at 1, 2, 3, 4, and 5 hrs were signiﬁcantly lower than the corre-
sponding untreated shock values.

Table V and Table V1, respectively, show the rate of TXB2 and 6-keto-PGF1
release from the intestine, as calculated from the product of blood ﬂow and arterio-
venous concentration difference. The intestinal release of TXBZ and 6-keto-PGF1
did not significantly change throughout the experimental period in control sham-
operated dogs. In untreated hemorrhagic shock dogs, TXBZ release did not change
signiﬁcantly until after blood transfusion when a marked and signiﬁcant increase in
release occurred. The release of 6- keto-PGFI , however, significantly decreased
throughout the hemorrhagic hypotensive period and after blood transfusion. In
GOA-treated hemorrhagic shock dogs, a signiﬁcant decrease in TXB2 release at 2,
3, 4, and 5 hrs was noticed compared to the untreated group. The values at 2 and 3
hrs were signiﬁcantly lower than the corresponding sham-operated values. Howev-
er, GOA treatment has signiﬁcantly increased 6-keto-PGF1 release at 2, 4, and 5 hrs
as compared to prehemorrhagic values and corresponding untreated shock values.
If all TXA2 and prostacyclin produced in the intestinal tissue is released into the
venous effluent, the rate of release can be considered as the rate of intestinal pro-
duction.

At the end of the experiment, intestinal tissue samples were collected to
determine if any morphological changes had occured during the experimental
procedure from GOA perfused ileal segment and non-perfused adjacent gut (n=8).

According to the classiﬁcation of the degree of intestinal lesions used by Chiu et al.

 

109

(page 33), the non- perfused segment showed mucosal lesions ranging from Grade 3
to Grade 5 in all eight dogs. The major abnormalities detected were epithelial cell
ulceration and necrosis, mixed inflamatory infiltrates with severe congestion, and
hemorrhage extending into the submucosa. The GOA-treated segment showed no
apparent mucosal loss (Grade 0 to Grade 2 ). The morphological findings were
some mild vascular congestion with focal minor surface blebbing and small petechi-
al hemorrhages. The ﬁndings in the untreated segment are consistent with necrotiz-
ing and hemorrhagic enteropathy of small intestine . In comparison to the untreated
group , the group with GOA-treated segment had signiﬁcantly fewer severe mucos-
al abnormalities.

 

 

DISCUSSION

Recent studies have shown that metabolites of arachidonic acid may play a
role in regulation of intestinal blood ﬂow, capillary exchange capacity, and oxygen
uptake under resting and postprandial conditions (52, 102,103,104, 117) as well as in
shock-induced intestinal ischemia (12,111). However, there are still many unan-
swered questions. The present study addressed some of these questions.

In the ﬁrst series of experiments, jejunal production of PGE2, PGIZ, PGF2 ,
and TXAZ was determined under four experimental conditions. These four prosta-
noids are most commonly found in the gastrointestinal tract (2,6,91,124,150,152) and
have well-deﬁned vasoactivity (1,22,38,49,56).

As shown in Figs. 5 and 6 , the jejunum produces and releases PG12 (meas-
ured as its metabolite 6-keto-PGF1 ), PGFq, and TXAZ (measured as its metabolite
TXB2), and an insigniﬁcant amount of PGF2 under resting conditions (i.e., normal
saline in the lumen). This finding is in agreement with other studies (91,152). The
relative magnitude of the production (release) in the present study is PG12
(44.4%)> PGE2 (31.9%) > TXA2 (21.2%) > PGF2 (2.5%) . Bennett et al. (6)
found in the human small intestinal mucosa or muscle that these four prostanoids
are formed in amounts greater than 80 ng.g’1, tissue wt, while only trace amounts of
the lipooxygenase product lZ-HETE, eicosatrienoic acid metabolites, and eicosa-
pentaenoic acid metabolites are formed. Le Duc et al. (91) found that PGIZ is the
major product (81%) of the muscularis tissue. In mucosal tissue, they found TXB2
(15%), PGIZ (14%), PGE2 (10%), PGF2 (5%) and PGDZ (4%) are the major
prostanoids (with 51% of unspeciﬁc hydroxy-fatty acids) . In canine jejunal muscle
the relative magnitude of prostanoid synthesis is: PG12 (43%) > PGF2 (20%) =
PGF/2 (17%) > PGA2 (10%) = PGD2 (7%) > TXB2 (trace) (124). The produc-

tion rate measured in the present study is from the whole gut wall and is compara-

110

J

 

111

ble with the above studies. Series 1 experiments shows that under resting conditions,
the vasodilator prostanoids, PGIZ and PGE2, comprise 76% of the four common
vasoactive prostanoids produced in the jejunum. Thus, in the resting gut, prostanoid
synthesis favors vasodilators, and the mechanism by which cyclooxygenase inhibitors
decrease blood flow in resting gut (22,38,52,56,98) may be due to inhibition of the
synthesis of these vasodilators.

Introduction of food containing 10% gallbladder bile increased the venous
concentrations and releases of all four prostanoids without significantly altering
their arterial concentrations (Fig. 5 and 6). If all the prostanoids produced in the
gut were released into the venous outﬂow, one can assume that release rate is the
same as production rate. Thus, the presence of digested food in the jejunal lumen
indeed stimulates synthesis of all four prostanoids, particularly PGE2 and PGF2 ,
which increase 4- fold and 8-fold, respectively. Could the food-induced increases in
prostanoid concenterations be responsible for the changes in blood flow by their
direct vascular actions ? In order to assess this possibility, the effects of local i.a.
infusions of these prostanoids, individually and in combination, were studied in the
next series of experiments.

As shown in Fig. 8A, infusions of the four prostanoids, to simulate the increase
in blood concentrations over a range 0.5 to 20 times that occurring when luminal
content was changed from saline to food, did not signiﬁcantly alter jejunal vascular
resistance. However, as other investigators have reported (52,38,43,49,56), when
these prostanoids were infused individually, they produced either vasoconstriction
or vasodilation (Fig. 7). At concentrations above 1000 pg/ ml, both PGI2 and U-
44069 produced signiﬁcant vasodilation and vasoconstriction, respectively, whereas
PGF/2 and PGF2 required concentrations greater than 104 pg/ ml to produce signiﬁ-
cant vasoactivity (Fig. 7). The local blood PGIZ, U-44069, PGE2, and PGF2

concentrations achieved during infusion of the mixture at 10 to 20 times that of

mini"

 

112

[PGs]F-NS were 823 to 1646, 717 to 1434, 2593 to 5186, and 470 to 941 pg/ml,
respectively. The failure of the mixture to produce significant vasoactivity is most
likely due to the antagonistic vasodilator action of PGIZ and vasoconstrictor action
of U-44069. The close agreement between the predicted and observed data shown
in Fig. 8A supports the above possibility.

However, a similar antagonism between PG12 and TXAZ might not occur
during food placement for the following reasons. The jejunum is composed of three
tissue layers and the major prostanoids produced in each layer are different. The
two potent vasoactive prostanoids, TXA2 and PGIZ, are produced primarily in the
mucosa and muscularis, respectively (91,124), and that villus epithelial cells produce
signiﬁcant amounts of TXAZ and PGF2 (3). These prostanoids, which are released
during food placement, act only on the vasculature in the vicinity of the site of their
releases. As a result, vasoconstriction could occur in the mucosa, while vasodilation
could occur in the muscularis. In contrast, i.a. infusion delivers these prostanoids to
both the mucosa and muscularis evenly, resulting in no significant change in total
vascular resistance, as shown in Fig. 8A. In the interpretation of Fig. 8A, one must,
therefore, consider the difference in the events occuring in the mucosa during food
placement and during i.a. infusion of the prostanoid mixture. This is important
because postprandial intestinal hyperemia occurs mainly in the mucosa.

It must be concluded from the data of Fig. 8A that the increases in prostanoid
release and concentration during food placement do not act directly on the vascula-
ture to alter blood flow. The inhibitory effect of endogenous prostanoids as indicat-
ed by the studies utilizing cyclooxygenase inhibitors (52,117), therefore, must be
mediated by other mechanisms. The most likely mechanism is the action of prosta-
noids on intestinal functions and oxidative metabolism. Several pieces of evidence
support this hypothesis. Cyclooxygenase inhibitors affect food-induced changes in
both blood ﬂow and oxygen uptake in parallel fashion. Some prostanoids have been

113

shown to alter intestinal absorption, secretion, and motility (34,44,152). In our
experimental setting, mefenamic acid increases carbohydrate absorption and
metabolism (54). Alteration in these intestinal functions by endogenous prostanoids
could indirectly alter blood flow via changes in oxidative metabolism and motility
(25,53) In the present study, an attempt was made to measure oxygen uptake during
i.a. infusions of prostanoids. Unfortunately, the infusions alter the reading of the A-
V oxygen content difference analyzer, making it impossible to analyze accurately the
oxygen uptake data.

Addition of arachidonate to normal saline did not signiﬁcantly alter jejunal
blood ﬂow nor oxygen uptake, and increased only the venous PGF2 concentrations
as compared to normal saline in the lumen (Figs. 4-5). In contrast, addition of ara-
chidonate to food attenuated the food-induced increases in blood ﬂow and oxygen
consumption, and produced higher venous TX82 and PGF2 than did food alone
(Figs. 45). Thus, food in the presence of arachidonate appears to shift arachidonate
metabolism in favor of TXA2 synthesis. The reasons for the effects of arachidonate
on jejunal blood flow, oxygen uptake, and the relative synthesis of vasodilator and
vasoconstrictor prostanoids during nutrient absorption, but not during luminal
placement of normal saline, are not known. The metabolic activities accompanying
nutrient absorption seem to play a signiﬁcant role in the observed arachidonate
actions. Could the alterations in TXA2 and PGF2 act directly on vasculature to
inhibit food-induced hyperemia? In order to assess this possibilty NS or food was
placed in the lumen; then prostanoids were infused at concentrations which mim-
icked the situation when lumen content was changed from N8 or food to food plus
AA.

As shown in Fig. SB and 9, i.a. infusion of the prostanoid mixture at concen-
trations 0.5 to 20 times those occurring when luminal content is changed from NS or

food to food plus AA increased jejunal vascular resisitance in a dose-dependent

114

fashion. At concentrations 0.5 to 2 times those occurring when luminal content was
changed from food to food plus AA, abolished the food induced decrease in vascu-
lar resistance (Fig. 9). This ﬁnding is similar to that shown in Fig. 4, when arachido-
nate was added to food in the lumen. This study therefore suggests that the inhibito-
ry action of arachidonate on food-induced hyperemia shown in Fig. 4 and observed
in a previous study (103) was at least in part due to the direct vasoconstrictor action
of endogenous TXAZ and PGF2. The parallel change in blood flow and oxygen
uptake shown in Fig. 4, however, may suggest involvement of the metabolic mecha-
nism described in the previous paragraph. Regardless of what the actual mechanism
is, the effect of luminal arachidonate has its physiological signiﬁcance. Since ara-
chidonate is a natural food constituent, the amount of arachidonate ingested might
inﬂuence postprandial intestinal blood ﬂow and oxygen uptake.

TXA2 is a much more potent vasoconstrictor than PGF2 (56,150) and quanti-
tatively venous TXA2 concentration and production during luminal placement of
food plus arachidonate were 2- to 3-fold greater than those of PGF2 (Figs. 5 and 6).
In order to determine the contribution of TXAZ in vascular and metabolic re-
sponses, we infused a 'I'XAZ/endoperoxide receptor antagonist, SQ-29548, when the
lumen still contained food plus arachidonate. As shown in Fig. 4, this infusion
increased blood ﬂow and oxygen uptake to levels similar to those observed with
food alone in the lumen (Fig. 4). Thus, the inhibitory action of arachidonate on
food-induced change in blood ﬂow and oxygen uptake appears to be, at least in part,
due to the action of TXA2 /endoperoxide which decreased intestinal blood flow as
well as capillary ﬁltration coefﬁcient (56,103).

In conclusion, the present study in Series I, II and III has addressed some of
the many questions in regard to the involvement of the cyclooxygenase products,
prostanoids, in the regulation of blood flow and oxygen uptake of the jejunum,

particularly the regulation during nutrient absorption. By utilizing radioimmunoas-

 

115

say the arterial and venous concentrations of four commonly found prostanoids
(PGIz, PGEQ, TXA2 and PGF2 ) were measured and the ﬁndings are as follows:
Under resting conditions, the jejunum produces and releases the ﬁrst three prosta-
noids, and digested products of food in the gut lumen enhance the production of all
four. The relative magnitude of releases are PGF/z (46%) > PGIZ (32%) > TXAZ
(16%) > PGF2 (6.4%). Addition of arachidonate into the food inhibits food-in-
duced increases in blood ﬂow and oxygen uptake, decreases the release of two
vasodilators PGIZ and PGE2, and increases the release of two vasoconstrictors,
TXA2 and PGF2 . SQ-29S48, a TXAz/endoperoxide receptor antagonist, reverses
the inhibitory action of arachidonate on blood ﬂow and oxygen uptake. Intra-arterial
infusion of the mixture of the above four prostanoids at blood concentrations similar
to the increases observed when lumen content was changed from saline to food did
not signﬁcantly alter jejunal vascular resistance (JVR). However, at concentration
similar to the increases observed when arachidonate was added to the luminal food,
the mixture increased JVR and abolished the food- induced decrease in JVR. The
vascular effects of a prostanoid mixture was similar to the algebraic sum of each
individual prostanoid effect. These studies indicate that jejunal productions of P612,
PGFQ, TXA2, and PGF2 increase during nutrient absorption. Addition of arachido-
nate to food attenuates the former two and enhances the latter two releases, which
act to attenuate food-induced jejunal hyperemia and oxygenation.

In Series 1 it has been shown that placement of food into the jejunal lumen
enhanced prostanoid release. However, which of the constituents of food, and/ or
events accompanying food absorption increased prostanoid release is/ are unknown.
In Series IV, V, VI, and VII jejunal prostanoid production was measured during
luminal placement of normal saline, 10% bile and air and during pump-induced
mechanical hyperemia. Normal saline and 10% bile were used in preparing the food

mixture, air was used to simulate the intestinal wall distension during food place-

 

116

ment, and the pump induced hyperemia was used to simulate the food-induced
hyperemia. As shown in Figures 10 , 11 and 12, luminal placement of normal
saline, air and 10% bile, respectively, did not significantly alter jejunal venous
prostanoid concenteration. However, previous study (82) have shown that bile per
se in the jejunal lumen, does not alter local blood ﬂow. However, it signiﬁcantly
enhanced the hyperemic and metabolic effect of digested food. The augmenting
effect of bile is probably due to its facilitating effect on digestion and absorption of
lipids. This facilitatory effect of bile on nutrient absorption might also have an indi-
rect effect on prostanoid production. The distension of the intestinal wall has been
known to induce various response patterns from the visceral musculature at intra-
luminal pressure greater than 25mm.Hg (25,28). However, repeated placement of
solutions in the lumen have been shown to induce receptive relaxation.
(25,32,83,102). In the present study, the intraluminal pressure was less than 5
mmHg. Therefore, the absence of changes on blood flow and prostanoid production
with luminal placement of air suggest that intestinal distension per 5; observed
during luminal placement of solutions might not be responsible for the food-induced
increases in prostanoid production. This is further supported by the data obtained
during NS and bile placement.

The effect of pump induced hyperemia on jejunal prostanoid release is shown
in Fig 15. The jejunal releases of PGE2, PGI2 and TXA2 were enhanced while
PGF2 release was not altered by the hyperemia. During nutrient absorption, an
increase in jejunal blood ﬂow is a well-documented observation (30,53,55). It is
possible that part of the food-induced increases in jejunal prostanoid releases might
be mediated through the food-induced increases in jejunal blood ﬂow.

In conclusion the studies in Series IV-VII showed that normal saline and bile
added to the food mixture did not directly contribute to the food-induced increases

in jejunal prostanoid production. Similarly, distension of the intestinal lumen with

117

an equivalent volume of air as the food placement, did not alter jejunal prostanoid
release. However, pump-induced jejunal hyperemia to simulate the postprandial
hyperemia has signiﬁcantly increased the jejunal releases of PGE2, PG12 and
TXA2 while PGF2 release was not altered. Part of the food-induced jejunal prosta-
noid release might be secondary to the food-induced hyperemia.

In the next Series of experiments an attempt was made to determine the
physiological role of jejunal TXAZ during resting conditions and nutrient absorp-
tion. Previous studies performed utilizing cyclooxygenase inhibitors suggested that
vasoconstrictor prostanoids might play a predominant role during nutrient absorp-
tion (52,117). TXA2 is the most potent vasoconstrictor prostanoid and is primarly
synthesized by the jejunal mucosal layer (3,6,91). It is possible that it plays a signiﬁ-
cant role in the regulation of jejunal blood ﬂow, oxygen consumption, and capillary
exchange capacity. As shown in Fig. 18, 19, and 20 administration of a
TXAz/endoperoxide receptor blocker did not alter resting jejunal blood flow,
oxygen consumption, and capillary exchange capacity. These results agree with
previous studies performed utilizing thromboxane synthetase inhibitors (imidazole
and U-63557A). In the resting gut, TXA2 seems not to play a signiﬁcant regulatory
role in the jejunum. However, in the presence of nutrients in the lumen the
TXAZ/endoperoxide receptor blocker signiﬁcantly enhanced the food induced
increases in oxygen consumption (Fig. 19) and capillary exchange capacity (Fig 20)
without altering the magnitude of the hyperemia (Fig 18). A similar finding was
observed utilizing TXAZ synthetase inhibitors (103,104). Therefore, TXAZ seems to
regulate effective jejunal capillary exchange capacity and oxygen extraction during
nutrient absorption.

The mechanism by which endogenous TXA2 limits the magnitude of food-
induced increases in jejunal oxygen extraction is unknown. One possible way might

be through direct action on oxidative metabolism. In general, vasoactive agents may

118

affect oxygen uptake in accord with their action on oxidative metabolism
(83,112,113,114). Gallavan and Chou (54) reported that inhibition of cyclooxyge-
nase during nutrient absorption signiﬁcantly enhanced the rate of glucose utilization
and absorption in canine jejunum. However, there are very few studies performed
on the effect of endogenous TXA2 on jejunal oxidative metabolism. These studies
showed that TXA2 did not alter jejunal glucose absorption (103), and colonic chlo-
ride (35) and duodenal bicarbonate (58) secretion. Since the effect of thromboxane
A2 on other energy-dependent intestinal functions is not well known, one cannot
rule out the possible role that thromboxane A2 might have on jejunal oxidative
metabolism.

In the present study, infusion of vasoactive chemicals during nutrient absorp-
tion has altered the magnitude of food induced changes in jejunal capillary ﬁltration
coefficient (Series XI) in parallel to the blood ﬂow changes. Other studies also have
shown that all vasoconstrictor chemicals attenuate and vasodilator chemicals
(except adenosine) enhance capillary filtration coefficient by altering intestinal
blood ﬂow (121,134). However, the findings with thromboxane receptor blocker
and synthtase inhibitors during nutrient absorption seem to lack the parallelism
between blood ﬂow changes and capillary exchange capacity. A similar ﬁnding was
previously observed during infusion of cholecystokinin, secretin and SHT where an
increase in capillary filtration coefficient was noticed at dosages lower than those
needed to produce changes in intestinal blood ﬂow (8,48). These investigators
concluded that capillary ﬁltration coefﬁcient reﬂects in most cases the tone of the
pre-capillary sphincter section. The contribution by pre-capillary sphincters to the
total jejunal resistance to blood ﬂow is slight, so net changes in resistance or blood
ﬂow do not necessarily correspond to any capillary filtration coefficient changes
(48). Therefore, it is possible that the predominant effect of TXA2 might be on the
precapillary sphincter tone which could possibly limit the capillary density without

 

119

altering jejunal blood ﬂow through a redistribution of blood between different layers
of the jejunum.

In conclusion, the present study in Series VIII, IX, and X indicated that a
thromoxane A2 analog (U-44069) is a potent vasoconstrictor in the jejunal vascular
bed. During nutrient absorption thromboxane A2 receptor antagonist (SQ- 29548)
potentiated the food-induced increases in jejunal oxygen consumption and capillary
ﬁltration rate. There was no change in blood ﬂow. Endogenous TXA2 might limit
the food-induced changes in jejunal capillary density and oxygen extraction.

In the next Series of experiments the effect of TXA2 on blood ﬂow distribu-
tion between mucosal and muscularis layers of the jejunal wall was determined uti-
lizing radioactive microspheres. As shown in Fig. 22 with a free-ﬂow jejunal prepa-
ration, local intra-arterial infusion of a thromboxane A2 analog decreased blood
ﬂow to both layers of the jejunal wall. However, the magnitude of decrease in
mucosal plus submucosal blood ﬂow was significantly greater than the muscularis
plus serosal. This ﬁnding might indicate the presence of a difference in response
between the two layers of the jejunal wall to the action of thromboxane A2 analog.
The data obtained in a constant ﬂow jejunal preparation (Fig 23) presents addition-
al evidence indicating a greater sensitivity of the mucosal-submucosal vascular bed
to the actions of the thromboxane A2 analog. Furthermore, Figs. 22 and 23 showed
that a thromboxane A2 receptor antagonist inhibited the TXA2 analog-induced
decreases in jejunal blood ﬂow in both layers of the intestinal wall. However, the
percent increase in mucosal plus submucosal blood ﬂow was signiﬁcantly greater
than the muscularis plus serosal. These findings might indicate that endogenous
thromboxane A2 could possibly elicit a variable degree of vascular response in the
mucosal- submucosal and muscularis-serosal vascular bed.

In the intestine where the mucosal-submucosal and muscularis-serosal blood

vessels are arranged in parallel (77), a redistribution of blood between the two

 

 

 

120

layers has been previously reported during sympathetic stimulation (48), adenosine
infusion (65,137), reactive hyperemia (136) and isoproterenol infusion (135,137).
The present study suggests that thromboxane A2 redistributes intestinal blood ﬂow.
This might indicate that the enhancement of the food-induced increases in jejunal
capillary exchange capacity by thromboxane receptor blocker shown in Fig. 20 and
by thromboxane synthetase inhibitors observed in the previous study (104) was at
least in part due to a redistribution of blood away from the muscularis plus serosal
layer to the mucosal plus submucosal layer. Since, thromboxane A2 is primarly
produced by the intestinal mucosal layer (3,91), it is likely that a true redistribution
of ﬂow takes place in such a way that the blood supply to the mucosal-submucosal
layer becomes limited. One possible physiological advantage of this TXA2 action
might be to regulate the rate of nutrient absorption.

The hyperreactivity of the mucosal layer to thromboxane A2 may have seri-
ous consequences in certain pathophysiological situations associated with increased
thromboxane A2 production. During intestinal ischemia secondary to hemorrhagic
shock (Fig. 30, Fig. 31 and Table V), superior mesenteric artery occlusion (111) and
endotoxemia (45), an increase in TXAz production and severe intestinal mucosal
damage have been reported. The intestinal muscle layer is minimally affected
under these conditions. One of the pathophysiological mechanisms contributing to
mucosal damage might be a higher sensitivity of the intestinal mucosa to thrombox-
ane A2. This may constrict the vessels and further aggravate the degree of mucosal
ischemia to the extent of producing necrotic lesions.

In conclusion, the study in Series XII and XIII showed that a thromboxane A2
analog (U-44069) decreased mucosal plus submucosal and muscularis plus serosal
blood ﬂow. The percent decrease in blood ﬂow was signiﬁcantly higher in the
mucosal plus submucosal layer than in muscularis plus serosal. The U-44069-in-

duced decreases in blood ﬂow were reversed by administration of a thromboxane

 

121

A2 receptor antagonist (SQ-29548). These actions of U—44069 and SQ-29548 sug-
gest that the intestinal mucosal vascular bed has a greater reactivity to the action of
thromboxane A2 than the muscle. The greater reactivity of the mucosal vascular
bed to the actions of thromboxane A2 may play a signiﬁcant role in various physio-
logical (nutrient absorption) and pathological (shock) conditions.

The next Series of experiments was designed to determine the role of endoge-
nous prostanoids, particularly TXA2 and PG12 released from the small intestine on
the progress and development of irreversible hemorrhagic shock and intestinal
mucosal damage. Furthermore, the effects of intraluminal instillation of a diet,
particularly a glucose oleic acid (GOA) solution, on the magnitude and proportion
of hemorrhagic shock-induced intestinal PG12 and T'XA2 releases and the degree of
intestinal mucosal damage were determined. ‘

Previous studies have shown that prostanoids, particularly thromboxane A2
and prostacyclin, may play a role in the development and progress of a variety of
shock conditions (12,17,18,20,37,40,45,70,73,111,149). Evidences for their in-
volvement is derived from experiments utilizing synthetase or receptor inhibitors
(12,17,20,40,70,111) and those measuring the concentration of these two prostanoids
in systemic blood (17,18,20,37,45,70, 73,111, 149). The studies which measure pros-
tanoids show that both TX82 and 6-keto-PGF1 increase in septic shock (20,153),
endotoxic shock (73,45), and after splanchnic artery occlusion (111). There is,
however, no study which measures these two eicosanoids during the course of severe
irreversible hemorrhagic shock. Furthermore, most studies have measured their
concentrations in systemic blood rather than in venous blood draining specific
organs. In order to determine the relative contribution of TXA2 and prostacyclin
and the contribution of the intestine in the development of irreversible hemorrhagic
shock, arterial and intestinal venous blood concentrations of TXB2 and 6-keto-

PGFl and the rate of release of these two prostanoids from the intestine were

122

measured during severe hemorrhagic hypotension and after blood transfusion.
Although there are species differences in the blood concentration of TX82 and 6-
keto-PGFI in shock conditions (153), the resting and experimental values obtained
in the present study are comparable to those obtained in anesthetized dogs subject-
ed to splanchnic arterial ischemia (12).

The result shows that prostacyclin and TXA2 productions are affected differ-
ently at different stages of hemorrhagic shock. Intestinal venous 6-keto-PGF1 in
creased at hrs. 2 and 3 during the hypotensive period, then fell sharply after retrans-
fusion of the shed blood in untreated shock group ( Fig. 29 ). The increases in
venous concentrations during the hypotensive period were not signiﬁcantly different
from the corresponding values of the sham-operated control dogs, but the decreased
venous concentration values after reinfusion were signiﬁcantly lower than the corre-
sponding control values (Fig. 29). The three-fold decrease in intestinal venous 6-
keto-PGFI concentration after reinfusion (Fig. 29) may be misleading because
there was a concurrent six-fold increase in blood ﬂow. Indeed, 6-keto-PGF1 output
(venous concentration x blood ﬂow) signiﬁcantly increased from 8.5 _+_- 2.0 ng/ min
at hr 3 to 31.4 1- 7.9 ng/min (one hour after reinfusion (hr 4)) (p<0.05). The
increase, however, was primarily due to an increase in arterial '6-keto-PGF1 con-
centration. As shown in Table VI, intestinal 6-keto-PGF1 production remained low
at hr 4 and 5. Other investigators have shown that arterial or systemic venous
6-ketooPGF1 concentration increases in septic shock in rats and pigs (20,153,142),
in endotoxin shock in piglets (126), and following 3-hour splanchnic arterial ische-
mia in dogs (111). However, septic shock in dogs (153) and endotoxin shock in
piglets (149) did not signiﬁcantly alter arterial and systemic venous 6-keto- PGFl .
The response of prostacyclin production to stress therefore appears to vary with
species and types of shock.

Conversely, perfusion of a GOA solution into an isolated intestinal segment

 

 

123

signiﬁcantly increased intestinal venous 6-keto-PGF1 concentration when compared
to that observed in the untreated hemorrhagic shock and sham- operated control
groups (Fig. 29). As shown in table V1, intestinal 6-keto-PGF1 release in the GOA
perfused segment signiﬁcantly increased at 2, 4, and 5 hrs compared to prehemor-
rhagic values and corresponding untreated shock values. This indicates that GOA
treatment during hemorrhagic shock stimulates prostacyclin production from the
perfused ileal segment.

In contrast to no change or a decrease in intestinal venous 6-keto-PGF1 (as
compared to sham-operated control and GOA-treated hemorrhagic shock dogs)
(Fig. 28 and Fig. 29 ), TXB2 concentration signiﬁcantly and progresively increased
during hypotensive (hr 0-3) and post-transfusion (hr 4 and 5) periods in the untreat-
ed shock groups (Fig. 30, and Fig. 31). Furthermore, TXB2 concentration in the
intestinal venous blood samples was significantly higher than that in the arterial
samples obtained at the corresponding time. This indicates that thromboxane
production increases in the intestine, and the increased production may contribute
at least in part, to the increased arterial thromboxane concentration during these
periods. A recent study shows that the gastrointestinal tract contains high activities
of thromboxane synthetase (130). In the GOA-treated shock dogs, the arterial
TXA2 concentration signiﬁcantly increases in a similar way to the untreated shock
dogs. Other investigators have also found that arterial TXB2 increases in septic
shock in pig (153,142), rat, and dog ( 153) and that the systemic venous TXB2
concentration increases in endotoxic shock in piglets (149, 126) and splanchnic
arterial ischemia in dogs (111). The increased arterial and systemic venous TXB2
levels under these conditions may also, at least in part, be due to the increased
thromboxane production in the intestine, particularly in the case of splanchnic
ischemia. As shown in Fig. 26 and Fig. 27, respectively , intestinal blood ﬂow de-

creases and intestinal vascular resistance increases markedly during hemorrhagic

 

T

 

124

hypotension.

However, GOA treatment signiﬁcantly lowered the shock-induced increases
in intestinal venous TXB2 concentrations at 2, 3, and 5 hrs compared to the corre-
sponding untreated shock values. Thus, GOA treatment increases prostacyclin and
decreases thromboxane A2 production from the perfused ileum during hemorrhagic
shock.

Because of their opposite actions, it has been proposed that TXA2 and PG12
play opposite roles in their contribution to the progress of shock states. TXA2 is
generally considered to be destructive (12,92,93,111), whereas PG12 is considered to
be beneﬁcial (45,93,95). However, PG12 has also been implicated as responsible for
the terminal hypotension in endotoxic shock (17,18,73). One way to evaluate their
relative contribution is to calculate the relative concentration in the blood and rela-
tive amount of production in the intestine of these two eicosanoids. It must be
noted here that although various actions of PG12 are in opposition to those of
TXA2, the concentrations at which the former (e.g. vasodilation) will effectively
cancel the latter action (e.g. vasoconstriction) are unknown. The concentration
ratio, therefore, can provide only a general idea of the balance in their actions.

As shown in Fig. 32, the TXB2/6-keto-PGF1 ratio in the arterial blood was
increased, particularly at hr 3 and 5 in the untreated shock groups. A comparison of
the present result was made with the changes in arterial TXB2/6-keto- PGFl ratio
as reported in septic shock ( 153) and as calculated based on the data presented in
endotoxic (126) and septic (142) shock. In all of the four conditions the ratio in-
creased. However, the time of onset and the magnitude of increase in the ratio is
variable. The sources for the variations might be the difference in species and the
type, severity, and duration of shock under study. This data seems to suggest that
the imbalance in the change of arterial thromboxane and prostacyclin concentration,

and thus the production, may be a contributing factor for the progression of these

125

shock states. In the untreated shock group, the intestinal venous concentration ratio
of TXB2 and 6-keto- PGFl increased progressively ( i.e. four-fold at hr 3 and
seven-fold at 2 hours after blood transfusion (hr 5) (Fig. 33)). Table V and Table
VI further show that while 6-keto- PGFl release from the intestine decreased sig-
niﬁcantly, TXB2 release increased significantly, five- to six-fold 1-2 hours after
blood transfusion. If the rate of release reﬂects fairly the rate of production in the
intestine, the data indicates that the inﬂuence of thromboxane in the intestine in-
creases while that of prostacyclin decreases.

In the GOA-treated segment the intestinal venous concentration ratio of
TXB2 and 6-keto- PGFl tended to decrease during the hypotensive period (Fig.
33). Table V and V1 further show that while intestinal TXB2 release significantly
decreased, 6-keto-PGF1 increased, particularly during the post-transfusion period.
The data indicates that the hemorrhagic shock-induced increased inﬂuence of
thromobxane in the intestine is signiﬁcantly abolished by luminal perfusion of GOA.

The imbalance of inﬂuence of these two prostanoids may contribute to the
development of irreversible shock as well as reperfusion injury of the intestine.
Though reperfusion of ischemic intestine has the beneﬁcial effect of salvaging the
organ, it is also associated with tissue damage (61). The finding in the untreated
shock group that thromboxane production in the intestine markedly increases after
reperfusion, seems to suggest that thromboxane may be a factor contributing to this
damage. In support of this thesis, GOA perfusion has signiﬁcantly attenuated the
degree of shock-induced intestinal mucosal damage and TXA2 production. Recent
studies have also shown that administration of a TXA2 receptor blocker or synthe-
tase inhibitor attenuates the rise in a proteolytic enzyme activity that occurs after
reperfusion in splanchnic artery occlusion (111) or hemorrhagic shock (12).

The intestine has been shown to play a signiﬁcant role in the development of

irreversible shock. One possible involvement is the release of myocardial depres-

 

 

126

sant factor (MDF) from the intestine (69,94,115). In a previous study (115), a signif-
icant release of MDF occurred two hours after blood transfusion in dogs subjected
to 3-hour severe hemorrhagic hypotension. The time of MDF release coincides with
the maximum increase in thromboxane release shown in Fig. 31 and Table V in
untreated shock dogs. This appears to suggest that in addition to other harmful
actions, such as platelet aggregation, an increase in intestinal TXA2 production may
be related to intestinal MDF production and release. Bitterman et al. (12) have
shown that a TXA2 receptor antagonist reduced the rise in plasma cathepsin D and
MDF activities and amino-nitrogen concentration in cats shbjected to severe hemor-
rhagic shock. In another study (111), the rise in the beta-glucuronidase activity that
occurred after the release of splanchnic artery occlusion was attenuated by a throm-
boxane synthetase inhibitor. Instillation of GOA into the intestinal lumen of dogs
subjected to irreversible hemorrhagic shock has been shown to prevent the release
of cardiodepressants from the splanchnic region (71,72). In the present study (Fig.
31 and Table V), it has been shown that GOA treatment significantly attenuated
the hemorrhagic shock-induced increases in intestinal TXA2 production and
mucosal damage. Furthermore, the degree of mucosal damage was significantly
correlated with mortality (68,69). Part of the mechanism by which GOA treatment
prevented irreversible hemorrhagic shock and increased survival rate (67% for 26 -
41 hrs) (71,72) might be through enhancement of a beneﬁcial prostanoid (prostacy-
clin) and inhibition of a destructive prostanoid (thromboxane A2) from the intes-
tine.

In 1982, Halper et al. (71,72) perfused GOA throughout the lumen of the
small intestine during severe hemorrhagic shock. The perfusion prevented the
terminal hypoglycemia and prolonged survival. In the present study, the GOA
treatment through a very small segment of the small intestine has also prevented the

terminal hypoglycemia (Fig. 25). An additional mechanism by which GOA prevent-

127

ed the severity of irreversible hemorrhagic shock is by making available an ade-
quate level of substrate for oxidative metabolism. Strawitz and Hiff (143), Moffat et
al. (106), and Hinshaw et al. (75) have reported a similar salutatory effect of exoge-
nous glucose administration in different types of shock. Moreover, Gump et al. (67)
and Lang et a1. (88) reported that fat metabolism may represent a more effective
energy substrate than glucose in hypovolemic or endotoxic shock.

In conclusion, the study in series XIV shows that arterial and intestinal venous
TXB2 concentration increased progressively during severe hemorrhagic hypotension
and after blood transfusion. The 6-keto-PGF1 concentration in the blood increased
during hypotension and decreased during the post-transfusion period. Thrombox-
ane production in the intestine increased, whereas prostacyclin production de-
creased during the course of hypotension and reperfusion. GOA treatment during
the development and progress of shock inhibits the shock-induced increases in intes-
tinal TXA2 production and enhances the shock-induced decreases in PG12 produc-
tion. This GOA effect on prostanoid production might be related to the saluatory
effect that GOA perfusion has been shown to have during shock. The imbalance in
the production and release from the intestine of these two eicosanoids, which act in
opposition to each other, may be a factor contributing to the development of irre-

versible hemorrhagic shock and reperfusion injury of the intestine.

 

SUMMARY AND CONCLUSION

Recent studies suggest that endogenous intestinal prostanoids play a role in
the regulation of jejunal blood ﬂow and oxygen uptake at rest, during nutrient
absorption, and during the progress of severe hemorrhagic shock. In the present
study which of the intestinal prostanoids that play a regulatory role was determined.
The main ﬁndings of the study are:
1. During luminal placement of normal saline (NS), the jejunum releases PGI2,
PGF/2 and TXA2.
2. During nutrient absorption:

A. Food (F) produced a signiﬁcantly higher blood ﬂow and oxygen uptake.

B. Food signiﬁcantly increased the jejunal production and releases of PGF/2, PG12,
TXA2 and PGF2. Infusions of a mixture of the four prostanoids to mimic a situation
of changing luminal content from NS to F did not signiﬁcantly alter the jejunal
vascuar resistance (JVR). However, local intra-arterial infusion of U-44069 (a
TXA2 analog), and PGF2 increased, while PGI2 and PGE2 decreased JVR in a
dose dependent fashion.

3. Effect of substrate loading with arachidonate:

A. Addition of arachidonate to food (F +AA) abolished food- induced increases in
blood ﬂow and oxygen uptake.

B. Addition of arachidonate to food signficantly increased PGE2, TXB2 and
PGF2 releases.

C. As compared to F in the lumen, F+ AA signiﬁcantly increased venous TXA2
and PGF2 releases.

D. Infusion of the four prostanoids, to mimic a situation of changing luminal

content from NS to F+ AA increased the JVR in a dose dependent fashion.

E. Placement of F per se signiﬁcantly decreased J VR, which was abolished and

reversed by prostanoid mixture, infused to mimic a situation of changing luminal

128

 

 

 

 

129

content from F to F+AA.

F. Local intra-arterial infusion of SQ-29548, a TXA2/ Endoperoxide receptor
antagonist abolished the arachidonate inhibition on blood ﬂow and oxygen uptake
during food placement.

4. Effects of individual constituent of food and events associated with nutrient
absorption on jejunal prostanoid production.

A. Jejunal blood ﬂow was not altered by luminal placement of NS, air, or 10% bile
compared to an empty lumen.

B. Jejunal venous concentrations of 6-keto-PGF1 and TXB2 were not signiﬁcantly
altered by luminal placement of NS, air, or 10% bile.

C. Jejunal venous concentrations of 6-keto-PGF1 and TXB2 tended to decrease
and PGE2 tended to increase during pump-induced jejunal hyperemia, whereas, the
PGF2 concentration signiﬁcantly decreased.

D. Pump-induced jejunal hyperemia signiﬁcantly increased the rate of PG12,
TXA2 and PGE2 releases from the jejunum. However, PGF2 release did not
change.

5. Role of enodgenous TXA2 on the regulation of intestinal vascular and metabolic
changes.

A. Local intra-arterial infusion of U-44069, a TXA2/ Endoperoxide agonist,
increased JVR in a dose dependent fashion. SQ-29548, a TXA2/Endoperoxide
receptor blocker, reversed the U-44069-induced JVR changes to control levels.

B. Local intra-arterial infusion of SQ-29548 potentiated the magnitude of post-
prandial jejunal oxygen uptake and capillary filtration coefﬁcient without altering
the magnitude of hyperemia.

C. Local intra-arterial infusion of U-44069 decreased jejunal mucosal-submucosal
and muscularis-serosal blood ﬂow in a free-ﬂow jejunal preparation. However, the

magnitude of decrease in blood ﬂow was significantly greater in the mucosal-

 

130

submucosal layer than the muscularis-serosal layer. SQ-29548 reversed the U-44069-
induced blood ﬂow changes to control levels.

D. Local intra-arterial infusion of U-44069 decreased jejunal mucosal-submu-
cosal and increased muscularis- serosal blood ﬂow in a jejunal preparation perfused
at a constant ﬂow rate. SQ-29548 restored the U-44069-induced blood ﬂow changes
to control levels.

6. Hemorrhagic shock and intestinal prostanoids:

A. Intestinal vascular resistance (IVR) increases during the hypotensive period of
severe hemorrhagic shock. Perfusion of a glucose oleic acid (GOA) solution through
the intestinal lumen abolished the shock-induced increase in IVR.

B. GOA perfusion prevented the hemorrhagic shock-induced terminal hypogly-
cemia .

C. Intestinal venous concentrations of TXB2 progressively increased during
hypotensive and reperfusion periods of severe hemorrhagic shock, while the 6-keto-
PGFl concentration decreased. Conversely, GOA perfusion abolished the shock-
induced increases in intestinal venous TXB2 concentration, whereas the 6-keto-
PGFl concentrations signiﬁcantly increased.

D. Intestinal rate of TXA2 release signiﬁcantly increased following reperfusion of
the shed blood in severe hemorrhagic shock dogs. However, the PG12 release signif-
icantly decreased during hypotensive and reperfusion periods. Conversely, GOA
perfusion abolished the shock- induced increases in intestinal TXA2 release,
whereas, the PG12 release was signiﬁcantly increased.

E. Perfusion of GOA solution during severe hemorrhagic shock reduced the
magnitude of intestinal damage.

The data indicates resting gut produces a signiﬁcant amount of PG12, PGF/2,
and TXA2. Nutrient absorption signiﬁcantly increased the release of PG12, PGEZ,
TXA2 and PGF2 from the gut. Substrate loading with arachidonate signiﬁcantly

131

enhanced the formation of vasoconstrictor prostanoids, which in turn abolished the
postprandial intestinal hyperemia and oxygen consumption. Intra-arterial infusion
studies suggest that prostanoid action on the gut might be a direct or an indirect
vascular action through alteration of the parenchymal cell metabolism. Part of the
stimulus for the food-induced increase in intestinal prostanoid release might be to
the food-induced increase in blood ﬂow.

Endogenous TXA2 limits intestinal oxygen consumption and capillary ex-
change capacity during nutrient absorption. The TXA2 action seems to be primarly
localized to pre-capillary sphincter. Among the intestinal layers, the mucosal vascu-
lar bed seems to have a significantly greater senstivity to TXA2 than the muscle.
This enhanced mucosal senstivity to TXA2 might be the mechanism by which
endogenous TXA2 limits the rate of oxygen uptake and capillary exchange capacity
without altering the total intestinal blood ﬂow during nutrient absorption. Further-
more, the greater mucosal sensitivity to TXA2 might play a significant role in
pathophysiological conditions associated with intestinal mucosal damage.

During severe hemorrhagic shock, intestinal production of TXA2 progressive-
ly increases, whereas PG12 production decreases, indicating a possible pathophysio-
logic role TXA2 might play in the progress and development of irreversible hemor-
rhagic shock, particularly in the development of intestinal mucosal damage. The
deleterious effects of TXA2 during shock conditions were signiﬁcantly attenuated
by perfusion of GOA solution through the intestinal lumen. GOA significantly
decreased the shock- induced increase in intestinal TXA2 release, enhanced intesti-

nal PG12 release, and reduced the magnitude of intestinal mucosal damage.

BIBILIOGRAPHY

1. Akiba, T, M. Miyazaki and N. Toda. Vasodilator actions of TRK-lOO, a new
prostaglandin I2 analogue. Br. J. Pharmac. 89: 703-711, 1986.

2. Aly, A., C. Johnsson, P. Slezak and K. Green. Bioconversion of arachidonic acid
in the human gastrointestinaltract. Biochem. Med. 31: 319-332, 1984.

3. Balaa, M.A. and D.W. Powell. Prosta landin synthesis by enterocyte microsomes
of rabbit small intestine. Prostaglandins 1: 609-624, 1986.

4. Balent, G.A., Z.F. Kess, T. Varkonyl, T. Wittman, and V. Varro. Effects of pros-
taglandin E2 and F2 on the absorption and ortal trans/Eon of sugar and on the
local intestinal circulation. Prostaglandins 18: 65- 268, 19 .

5. Balent, G.A. and V. Vaero. Cytoprotective effect of prostacyclin and protein
synthesis. Acta Physiolog. Hunganca 61: 15-122, 1983.

6. Bennett, A., C.N. Hensby, G.J. Sanger and IF. Stamford. Metabolites of arachi-
donic acid formed by human astrointestinal tissues and their actions on the muscle
layers. Br. J. Pharmac. 74: 43 -444, 1981.

7. Biber, B. Vasodilator mechansim in the small intestine. Acta Physiolog. Scand.
89: 449-458, 1973.

8. Biber, B., J. Fara and O. Lund ren. Vascular reactions in the small intestine
during vasodilatation. Acta Physio . Scand. 89: 449-456, 1973.

9. Biber, B., J. Fara, and O. Lundgren. A pharmacological stud of intestinal
vasodilator mechanisms in the cat. Acta Physiol. Scand. 90:673-683, 1 74.

10. Biber, B., O. Lund ren, and J. Svanvik. Studies on the intestinal vasodilation
observed after mec anical stimulation of the mucosa of the gut. Acta
Physiol.Scand. 82:177-190. 1971.

11. Bills, T.K., J .B. Smith and MJ. Silver. Selective release of arachidonic acid from

{She1 %ospholipids of human platelets in response to thrombin. J. Clin. Invest. 60: 1-

12. Bitterman, H., A. Yanagisawa, and AM. Lefer. Beneﬁcial actions of thrombox-
ane receptor antagonism in hemorrhagic shock. Circ. Shock. 20:1-11, 1986.

13. Bounous G. Role of intestinal contents in the pathophysiology of acute intesti-
nal ischemia. Am. J. Surg. 114: 368, 1967.

14. Brandt, J., L. Castleman, H. Ruskin, J. Greenwold and J. Kelly. The effect of
oral protein and glucose feeding on splanchinic blood ﬂow and oxygen utilization in
normal and cirrhotic subjects. J. Clin. Invest. 34: 1017-1025, 1955.

15. Brodie, T.G., W.C. Cullis and W.D. Halliburton. The gasseous metabolism of

the small intestine. II. The asseous exchanges during the absorption of witte’s
peptone. J.Physiol. 40: 173-18 , 1910.

132

133

16. Brunsson, I., A. Sjoqvisit, M. Jodal and O. Lundgren. Mechanism underlyin
the small intestinal ﬂuid secretion caused by arachidonic acid, rostaglandin E1 an
prostaglandin E2 in the rat in vivo. Acta Physiol. Scand. 130: 6 3-642, 1987.

17. Bult, H., J. Beetens, and AG. Herman. Blood levels of 6-oxo-prostaglandin F61
during endotoxin-induced hypotension in rabbits. Eur. J. Pharmacol. 63:47-5 ,
1980.

18. Bult, H., J. Beetens, P. Vercruysse, and AG. Herman. Endotoxin-induced
hypotension and blood levels of 6-keto-prostaglandin F1 . In: Samuelsson, B.,
Ramwell, P.W., Paoletti, R. (eds): "Advances in Prostaglandin and Thromboxane
Research." New York: Raven Press, 1980, p 839.

19. Burns,G.P., and W.G. Schenk. Effects of di estion and exercise on intestinal
blood ﬂow and cardiac output. Arch. Surg., 98: 7 0—794, 1969.

20. Butler, R.R., Jr., WC. Wise, P.V. Halushka, and J .A.Cook. Thromboxane and
prostacyclin production during shock. Adv. Shock Res. 7: 133- 145, 1982.

21. Campion, T.W., J .C. Kerr, T.G. Lynch and R.W. Hobson. Comparison of
canine femoral and s lanchnic hemodynamic during intravenous infusions of pros-
tacyclin. Curr. Surg. 2:214-215, 1985.

22. Chapnick, B.M., L.P. Feigen, A.L. Hyman, and RI. Kadowitz. Differential ef-
fects of prosta landins in the mesenteric vascular bed. Am. J. Physiol. 235 (Heart
Circ. Physiol. 4 :H326-H332, 1978.

23. Chiu, C.J., A.H. Mcardle, R.Brown, H.J.Scott and F.N. Gurd. Experimental
Sggery: Intestinal mucosal lesions in low-ﬂow states. Arch. Surg. 101: 478-483,
1 .

24. Chiu, C.J., M.J. Scott, and F.N. Gurd. The protective effect of intraluminal
glucose as energy substrate. Arch. Surg. 101:484-488, 1970.

25. Chou, C.C. Relationship between intestinal blood ﬂow and motilit . In: Ann.
Rev. Physiol. 42: 29-42. edited by LS. Edelman palo alto : Ann Rev Inc, 1 82.

26. Chou, C.C., C.P. Hsieh, and J .M. Dabney. Com arison of vascular effets of
gastrointestinal hormones on various organs. Am. J. P ysiol.,232(2): H103- H109.

27. Chou, C.C., T.D. Burns, C.P. Hsieh, and J .M. Dabney. Mechanisms of local
vasodilation with hypertonic glucose in the jejunum. Surgery 71: 380-387, 1972.

28. Chou, CC, and J .M. Dabney. Interrelation of ileal wall compliance and vascu-
lar resistance. Am. J. Dig. Dis. 12:1198-1208, 1967.

29. Chou, CO, CR Hsieh, Y.M. Yu, P. Kvietys, L.C. Yu, R. Pittman, and J .M.
Dabney. Localization of mesenteric hyperemia during digestion in dogs.
Am.J.P ysiol. 230: 583-589, 1976.

30. Chou, CC. and PR. Kvietyﬁ; Physiological and harmacological alterations in
astrointestinal blood ﬂow. In: e Measurement of planchnic Blood Flow. edited
47565.30 Bulkley and D.N. Granger. Baltimore MD: Williams and Williams, 1981, pp.

134

31. Chou, C.C., P.R. Kvietys, R. Gallavan and R. Nyhof. Blood ﬂow, oxggen con-
sumption and abso tion of glucose and oleic acid in the canine jejunum. rastroen-
trology 76: 1114, 19 6.

 

32. Chou, C.C., P. Kvietys, J. Post, and SP. Sit. Constituents of chyme res nsible
for ostprandial intestinal hyperemia. Am. J. Physiol. 235 (Heart Circ. hyisol.
4): 677-H682, 1978.

33. Chou, CC. and H. Siregar. Role of histamine H1- and H2- receptors in ost-
prandial intestinal hyperemia. Am. J. Physiol. 243 (Gastrointest. Liver Phys10. 6):
G248-G252, 1982.

34. Con ar, I.M., and I. McCall. Inhibition of glucose absorption by PGEl, PGF/2
and PG 2 . J. Pharm Pharmacol. 24:254-255, 1972.

35. Cuthbert, A.W., P.V. Halushka, H.S. Margolius and J .A. S a e. Mediators of
the secretory response to kinins. Br. J. Pharmac. 82: 597-607, 9 .

36. Dembinski, A. and SJ. Konturek. Effects of E, F, and I series prostaglandins
and analogues on growth of gastroduodenal mucosa and pancreas. Am. J. Physiol.
248 (Gastrointest. 'ver Physrol. 11): G170-G175, 1985. ‘

37. Demling, R.H., M. Smith, R. Gunther, J.T. Flynn, and M.H. Gee. Pulmon
injury and rostaglandin production durin endotoxemia in conscious sheep. Am. .
P ysrol. ( eart. irc. Physiol. 9)240:H348- 353, 1981.

38. Dusting, G.J., S. Moncada, K.M. Mullane, and JR. Vane. Biotransformation of
arachidonic acid in the circulation of the dog. Br. J. Pharmacol. 63:359, 1978.

39. Fahrenkrug, J., U. Haguland, MJodal, O. Lundgren, L. Oibe and O. Schaffait-
zky De Mucka ll. Nervous release of vasoactive intestinal polypeptide in the gas-

trointestinal tract of cats: Possible physiological implications hysiol. London
284: 291-305,1978.

40. Famewo, C.E., W.H. Noble, and MB. Garvey. Use of aspirin in hemorrhagic
shock. Cand. Anaesth. Soc. J. 22:50-60, 1975.

41. Fara, J .W., E.H. Rubinstein, and RR. Sonnenschein Intestinal hormones in
naggintric vasodilation after intraduodenal agents. Am. J. Physiol. 223: 1058-1067,
1 .

42. Fara, J.W., E.H. Rubinstein, and RR. Sonnenschein Visceral and behavioural
response to intraduodenal fat. Science 166: 110-111, 1969.

43. Fei en, L.P. Actions of prostaglandins in peripheral vascular bed. Fed. Proc.
40: 198 -1990, 1981.

44.. Field, M., M.W. Musch and J .8. Staff. Role of prostaglandins in the regulation
of mtestinal electrolyte transport Prostaglandins 21: Suppl. p. 73-79, 1981.

45. Fletcher, J.R., Ramwell, P.W., and RH. Harris. Thromboxane, grostacyclin, and
1113§nlodynamic events in primate endotoxin shock. Adv. Shock. es. 5: 143-148,

 

 

 

135

46. Fletcher JR, Ramwell PW: The effects of prostacyclin PGI ) on endotoxin
shggk and endotoxin-induced platelet aggregation in dogs. irc ghock 7 :299-308,
19 .

47. Flower, RJ. and GJ. Blackwell. The importance of {hos holipase A2 in prostg-
landin biosynthesis. Biochemical Pharmacology 25: 285- 91, 976.

48. Folkow, 8., DH. Lewis, 0. Lundgren and I. Wallentin. The effect of the sympa-
thetic vasoconstrictor ﬁbres on the 'stribution of capillary blood ﬂow in the intes-
tine. Acta Physiol. Scand. 61: 458-466, 1964.

49. Fondacaro, J .D., KM. Walus, M. Schwaiger, and E.O.Jacobson. Vasodilation of
the normal and ischemic canine mesenteric circulation. Gastroenterology. 80: 1542-
1549, 1981.

50. Fronek, K., and LH. Stahlgren. Systemic and re 'onal hem 'c chan es
during god intake and digestion in non-anesthetized ogs. Circulation Res. 23:6 7-
692, 19 .

51. Gallavan, R.H., Jr., M.H. Chen, S.N. Joffe, and E.D.Jacobson. Vasoactive intes-
tinal polye tide, chole stokinin, glucagtjn, and bile-oleate-induced 'ejunal hypere-
mia. Am. . Physiol. 24 (Gastrointest. 'ver Physiol. 11):G208-G21 , 985.

52. Gallavan, R.H., Jr. and CC. Chou. Prostaglandin synthesis inhibition and
gosgrandial intestinal hyperemia. Am. J. Physio . 242 (Gastrointest. Liver Physiol.
): 140-G146, 1982.

53. Gallavan, R.H., Jr. and CC. Chou. Possible mechanisms for the initiation and
maintenance of the post randial intestinal lgyperemia. Am. J. Physiol. 249 (Gas-
trointest. Liver Physiol. 1 ): G301—G308, 198 .

54. Gallavan, R.H., Jr. and CC. Chou. The effects of mefenamic acid on postpran-
dial intestinal carbohydrate metabolism. Prostaglandins 31:1069- 1076, 1986.

55. Gallavan, R.H., J r., C.C. Chou, P.R. Kvie s, and SP. Sit. Regional blood ﬂow
during digstion in the conscious dog. Am. J. hysiol. 238 (Heart Circ. Physiol. 7):
H220-H2 , 1980.

56. Gallavan, R.H., Jr. and ED. Jacobson. Prostaglandins and the splanchnic circu-
lation. Proc. Soc. Exp. Biol. Med. 170: 391-397, 1982.

57. Gallavan, R.H., J r., C. Shaw, R.F. Murphy, S.V. Joffe, and ED. Jacobson. The
lipid-induced jejunal hyeremia and neurotensin release. Dig. Dis. Sci. 29:295, 1984.

58. Ganer, A. Stimulants of duodenal alkaline secretion. In: New

Ehannacelea ef
weer Disease: iﬁenmemalx ’ and 11m Therapeutic Agnraaghes. S. Szabo and
G.Y. Moszik, eds. ew York: Elsevier Sci. Pub. Co., 198 . p. 37-47.

59. Goldblatt, A.W. A depressor substance in seminal ﬂuid. J. Soc. Chem. Ind.
(Lond) 52: 1056, 1933.

 

136

60. Gran er, H.J., and CR Norris. Intrinsic re lation of intestinal o enation in
the anest etized dog. Am. J. Physiol. 238: ( eart Circ. Physiol. 7): 836-H843,
1980.

61. Granger DN, Hollworth MA, Parks DA: Ischemia re erfusion injurg: role of
oxygen-derived free radicals. Acta Physiol Scand 126 (Supp . 548):47-63, 1 86.

62. Granger, D.N., M.A. Perry, P.R. Kvietys and A.E.Taylor. A new method for
estimatin intestinal capillary pressure. Am. J. Physiol.244(Gastrointest. Liver
Physiol. 7 :GB41-G344, 1983.

63. Granger, D.N., P.D.I. Richardson, and AE. Taylor. Volumetric assessment .of
the capilla 7filtration coefficient in the cat small intestine. Pﬂl‘gers Al'ChlV.
381:25-33, 1 9.

64. Granger, H]. and R.A. Nyhof. Dynamics of intestinal oxygenation: interactions
between oxy en supply and uptake. Am. J. Physiol. 243 (Gastrointest. Liver Physi-
ol. 6): G-9l- 96, 1982.

65. Granger, D.N., J .D. Valleau, R.E. Parker, R.S. Lane, and AB. Taylor. Effects
of adenosine on intestinal hemodynamics, oxygen delivery, and capillary ﬂuid
exchange. Am. J. Physiol. 235 (Heart Circ. Physio . 4): H707-H719, 197 .

66. Gryglewski, RJ., S. Bunting, S. Moncada, RJ. Flower, and JR. Vane. Arterial
walls are protected against deposition of platelet thrombi by a substance (prostag-

landin X which they make from prostaglandin endoperoxides. Prostaglandins
12:685-713, 1976.

67. Gum , F.E., C.L. Long, M. Wong, and J .M. Kinney. Exogenous glucose as an
gillsrgyéslg strate in experimental hemorrhagic shock. Surg. Gynec. Obstet. 136:611-
, 1 .

68. Hagiland, E., U. Haglund, O. Lundgren, M. Romanos and T. Schersten. Graded
intestinal vascular obstruction. Circ. Shock 7: 83-91, 1980.

69. Haglund U, Lundgren O: Intestinal ischemia and shock factors. Fed Proc
37:2729-2733, 1978.

70. Hales, C.A., L. Sonne, M. Peterson, D. Kong, M.Miller, and W.D. Watkins.
Role of thromboxane and prostacyclin in pulmonary vasomotor changes after
endotoxin in dogs. J. Clin. Invest. 68:497-505, 1981.

71. Halper J.M. Protective mechanism of intestinal luminal glucose-oleic acid
installation in hemorrhagic shock. Thesis M.S. MSU, 1984. ~

72. Hallper, J.M., R.P. Pittman, and CC. Chou. Prevention of irreversible hemor-
rha 'c s ock by perfusion of glucose-oleic acid through the small intestinal lumen.
Fe . Proc. 42:322, 1982.

73. Harris, R.H., M. Zmudka, Y. Maddox, P.W. Ramwell, and LR. Fletcher. Rela-
tionship of TXB2 and 6- keto-PGF to the hemod amic changes during baboon
endotoxic shock. In: Samuelsson, ., Ramwell, P. ., Paoletti. R. (eds): "Advances
in Prostaglandin and Thromboxane Research." New York: Raven Press, 1980, p

 

 

 

137

74. Herrik, J .F., H.E. Essex, R.C. Mann and EJ. Baldes. The effects of di estion on
blood ﬂow in certain blood vessels of the dog. Am. J. Physiol. 108: 621-6 , 1934.

75. Hinshaw, L.B., M.D. Peyton, L.T. Archer, M.P. Block, J .J . Coalson, and LI.
Greenﬁeld. Prevention of death in endotxin shock by glucose adminstration. Surg.
Gynecol. and Obstet. 139: 851-859, 1974.

76. Irivine, RE How is the level of free arachidonic acid controlled in mammalian
cells ? Biochem. J. 204: 3-16, 1982.

77. Jacobson, E.D., R.H. Gallavan, Jr. and J .D. Fondacaro. A model of the mesen-
teric circulation. Am. J. Physiol. 242 (Gastrointest. Liver Physiol 5): GS41-G546,
1982.

78. Kewenter, J. The vagal control of the jejunal and ileal motiliy and blood ﬂow.
Acta Physiol. Scand. (Suppl.) 251: 1-68

79. Konturek, S.J. Gastrointestinal ph siolo and pharmacolo of rostaglan-
dins. In: NLWBhannaeolanfllleetlli’sease: ' and Eewfhmﬂs

S. Szabo and G.Y. Moszik, eds. ew York: Elsevier Sci. Publ.
Co.,1987. p. 304-321.

80. Konturek, 8.1., J. Tasler, J. Bilski and J. Kania. Prostaglandins and alkaline
secretion from oxyntic, antral, and duodenal mucosa of the dog. Am. J. Physiol. 245
(Gastrointest. Liver Physiol. 8): G539-G546, 1983.

81. Kurzrok, R. and CC Leib. Biochemical studies of human semen: II. The action
of semen on the human uterus. Proc. Soc. Exp. Biol. Med. 28: 268, 1930.

82. Kvietys, P.R., R.H. Gallavan, and CC. Chou. Contribution of bile to pos ran-
$218 i8nte5§161al hyperemia. Am. J. Physiol. 238 (Gastrointest. Liver Physiol. 1): 284-
, 1 .

83. Kvietys, RR. and D.N. Granger. Vasoactive agents and splanchnic oxygen
uptake. Am. J. Physiol. 243 (Gastrointest. Liver Physiol. 6): G1-G9, 1982.

84. Kvietys, RR. and D.N. Granger. Effects of volatile fatty acids on blood ﬂow
and oxygen uptake by the dog colon. Gastroentrology, 80: 926-969, 1981.

85. Kvietys, P.R., D.N. Granger, N.A. Mortillan and AE. Taylor. Effect of atrOpine
on fatty acid induced chan es in blood ﬂow, oxygen consumption and water trans-
port. Physiologist 22: 74, 19 9.

86. Kvietys, P.R., M.A. Perry, and D.N. Granger. Intestinal ca illary exchange
capaci and oxygen delivery-to-demand ratio. Am. J. Physiol. 2 5 (Gastrointest.
Liver P ysiol. 8): G635-G640, 1983.

87. Kvietys, P.R., R.P. Pitman and CC. Chou. Contribution of luminal concentra-
tion of nutrients and osmolali to postprandial intestinal hyperemia in dogs. Proc.
Soc. Exp. Biol. Med., 152: 659- 63, 1976.

 

 

 

138

88. Lang, HQ, and RJ. Alteveer. Effects of glucose-insulin-potassium on intestinal
hemodynamics and substrate utilization durin endotoxemia. Am. J. Physiol. 251
(Gastrointest. Liver Physiol.14):GB41-G348, l9 6.

89. Larsson, L., J. Fahrenkrug, O. Schaffalitzky De Muckadell, F. Sundler, R.
Hakanson, and J. Rehfeld. Localization of vasoactive intestinal polype tide (VIP)
to central and peripheral neurons. Proc.Nat. Acad.Sci. USA. 73: 19 -3 00, 1976.

90. Lawson, LD. and D.W. Powell. Bradykinin-stimulated eicosnaoid synthesis and
secretion by rabbit ileal components. Am. J. Physiol. 252 (Gastrointestinal Liver
Physiol. 15): G-783-790, 1987.

91. LeDuc, LE. and P. Needleman. Regional localization of prostacyclin and
abromsboxéage 79thesis in dog stomach and intestinal tract. J. Pharmacol. Exp. Ther.
11:1 1-1 , 1 .

92. Lefer, A.M. Role of prostaglandins and thromboxanes in shock states, In:
Altura, B.M., Lefer, A.M., Sc umer, W. (eds): ”Handbook of Shock and
Trauma."New York: Raven Press, 1983, p 355.

93. Lefer, AM. and H. Araki. Analysis of otential beneﬁcial actions of prosta an-
dins in traumatic shock. In: Lefer, A.M., chumer, W. (eds): "Molecular and ﬂu-
lar Aspects of Shock and Trauma." New York: Alan R. Liss, Inc., 1983, p 199.

94. Lefer, AM. and J. Martin. Origin of myocardial depressant factor in Shock. Am.
J. Physiol. 218(5): 1423-1427, 1970.

95. Levittt MA, Lefer AM: Anti-shock properties of the prostacyclin analog, ilo-
prost, in traumatic shock. Prostagl Leuko Med 25 :175-185, 1986.

93.57Lillehi, R.C. The intestinal factor in irreversible shock. Surgery 42: 1043-1054,

97. Lillehi, R.C., J.K. Longerbeam, J.H. Block and W.G. Manax. The nature of
ilrgrgzersible shock: Experimental and clinical observations. Ann. Surg. 160: 682.

98. Lippton, H.L., W.M. Arstead, A.L. Hyman and P.J.Kadowitz. Characterization
of the vasoconstrictor activity of indomethacin in the mesenteric vascular bed of
the cat. Prostaglandins Leuk. and Med. 27: 81-91, 1987.

99. Lund en, 0. Studies on blood ﬂow distribution and countercurrent exchange in
the small mtestine. Acta Physiol. Scand. 303: 5-42, 1967.

100. Lundgren, O. and U. Haglund. The pathophysiology of the intestinal counter-
current exchanger. Life Sci. 23: 1411-1422, 1978.

101. Mailman, D. Effects of penta astrin on intestinal absorption and blood ﬂow in
the anesthetized dogs. J. Physiol. 07: 429-442, 1980.

102. Mangino, MJ. and CC. Chou. Arachidonic acid and 05 randial intestinal
hyperemia. Am. J. Physiol. 246 (Gastrointest. Liver Physiol. 9 :G 21-GSZ7, 1984.

 

 

 

139

103. Mangino, M.J. and CC. Chou. Thromboxane synthesis inhibition andpost-
randial intestinal h eremia and oxygenation. Am. . Physiol. 250 (Gastromtest.
'ver Physiol. 13):G -G69, 1986.

104. Mangino, M.J. and CC. Chou. Thromboxane synthesis inhibitors and ost-
prandial jejunal capillary exchange capacity. Am. J. Physiol. 254: G695-G701, 988.

105. McCormick, J .R., W.M. Lien, A.M. Herman, and R.H.Egdahl. Glucose and
insulin metabolism during shock in the dog. Surg. Forum 20: 12-14, 1969.

106. Moffat, J .G., J .A.C. King, and W.R. Drucker. Tolerance to prolonged h vo-
lemic shock: effect of infusion of an energy substrate. Surg. Forum 19:5-8, 196 .

107. Moncada, S., E.A. Higgs, and J .R. Vane. Human arterial and venous tissues
enerate pgozstacygﬁlgn (prostaglandin X), a potent inhibitor of platelet aggregation.
cet 1:1 - 1, 1 .

108. Moncada, S. and JR. Vane. Pharmacology and endo enous roles of prostag-
landinggrgrdoperoxides, thromboxane A2, and prostacyclin. harmacol. Rev. 30:29 -
331, 1 .

109. Needleman, P, M. Minkes, and A. Raz. Thromboxanes: Selective biosynthesis
and distinct biological properties. Science 193:163-165, 1976.

110. Nyhof, R.A. and CC. Chou. Evidence against local neural mechanism for
intestinal postprandial hyperemia. Am. J. Physiol. 245 (Heart Circ. Physiol. 14):
H437-H446, 1983.

111. Oei, H.H.H., H.C. Zoganas, Y. Sakane, R.D. Robson, and T.M. Glenn. Inhibi-
tligg oggéromboxane biosynthesis in splanchnic ischemic shock. Circ. Shock. 18:95-
, 1 .

112. Pawlik, W.W., J .D. Fondacaro, and ED. Jacobson. Metabolic h eremia in the
canine gut. Am. J. Physiol. 239 (Gastrointest. Liver Physiol. 2): 612- 17, 1980.

113. Pawlik, W.W., AP. Shepherd, and ED. Jacobson. Effects of vasoactive agents
ggoinltgslginal oxygen consumption and blood ﬂow in dogs. J. Clin. Invest. 56: 484-

114. Piper, PJ. and J .R. Vane. Ann. NY. Acad. Sci. 180, 363 (1971)

115. Pittman RP, Senko JT, Nyhof R, Chou CC: Release of cardiodepressants from
the canine jejunum in irreversible hemorrhagic shock. Circ Shock 11:149-158, 1983.

116. Premen, A., C. Soika, J. Dabney, and D. Dobbins. Effects of gastrointestinal
hormones on ileal vascular and visceral smooth muscle. Am. J. Physiol. 246 (Gas-
trointest. Liver Physiol. 9):G1-G7, 1984.

117. Proctor, K.G. Differential effect of cyclooxy enase inhibitors on absorptive
hyperemia. Am. J. Physiol. 249 (Heart Circ. Physio . 18):H755-H762, 1985.

118.. Proctor, K.G., J .R. Falck and J. Capdevila. Intestinal vasodilation b epoxyeico-
satnenoic acids: Arachidonic acid metabolites produced by a cytoc rome p450
monooxygenase. Circ. Res. 60: 50-59. 1987.

140

119. Renninger, E.J. and LA. Sapiristein. Effect of digestion on distribution of
blood ﬂow in the rat. Science 126: 1176, 1957.

120. Re old, D., K.G. Swan. Intestinal micrvascular architecture in endotoxin
shock. astroentrology 63: 601, 1972.

121. Richardson, P.D.I., D.N. Granger and AE. Taylor. Capillary ﬁltration coefﬁ-
cient: the technique and its application to the small intestine. Cardiovasc. Res. 13:
547-561, 1979.

122. Robert, A. Cytoprotection and enterogoolin by prosta andins. In: Beri, F.,
Velo (eds): In; W System, New ork: lenum, 19 1. p. 393-400.

123. Rothe, C.F., J.R. Love and EB. Selkurt. Control of total vascular resistance in
hemorrhagic shock in the dog. Circ. Res. 12: 667-675. 1963.

124. Sanders, KM. and TE. Northrup. Prostaglndin s thesis bgﬁiicrosomes of
cirscglar and longitudinal gastrointestinal muscles. Am. . Physiol. : G442-G448,
19 .

125. Samuelsson, B. Biosynthesis of prostaglandin. Prog. Biochem. Pharmacol. 5:
109-128, 1969.

126. Schrauwen E, Houvenaghel A: Hemodynamic evaluation of endotoxic shock in
323358623618 piglets: antagonism of endogenous vasoactive substances. Circ Shock
1 :1 - , 1 .

127. Selkurt, E> E. and BA Brecher. Splanchinic hemodynamic and oxygen utiliza-
tion during hemorrhagic shock in the dog. Circ. Res. 4: 693-704, 1956.

128. Sharon, P., F. Karmeli and D. Rachmilewitz. PGE2 mediates the effect of
entagastrin on intestinal adenylate cyclase and Na-K-ATPase activities. Prostag-
andins 21: suppl. 25-32, 1981.

129. Shehadeh, A., W.E. Price, and ED. Jacobson. Effects of Vasoactive a cuts on
intestinal blood ﬂow and motiliy in the dog. Am. J. Physiol. 216: 386-392,19 9.

130. Shen, RP. and H.H. Tai. Monoclonal antibodies to thromboxane synthetase
from porcine lung. Production and apaﬂication to develo ment of a tandem immun-
oradiometric assay. J. Biol. Chem. 2 1: 11585-11591,19 6.

131. Shepherd, A.P. Intestinal blood ﬂow autore lation du ' food stuff absorp-
tion. Am. J. Physiol. 239 (Heart Circ. Physiol. 8): 156-H162, 1 80.

132. Shepherd, A.P. Role of capillary recruitment in the re lation of intestinal
oxygenation. Am. J. Physiol. 242 (Gastrointest. Liver Physiol. ): G435-G441, 1982.

133. Shepherd, AP. and CG. Bur er. A solid-state arteriovenous oxygen difference
analyzer for ﬂowing whole bloo . Am. J. Physiol. 232 (Heart Circ. Physiol. 1):
H43 -H440, 1977.

 

 

141

134. She herd, A.P., W. Pawlik, D. Mailman, T.F. Burks and ED. Jacobson.
Effects 0 vasoconstrictors on intestinal vascular resistance and oxygen extraction.
Am. J. Physiol. 230 (2): 298-305, 1976.

135. She herd, A.P., and G.L. Riedel. Continuous measurement of intestinal
mucosal lood ﬂow b laser-Doggler velocirnetry. Am. J. Physiol. 242 (Gastrointest.
Liver Physiol. 5): G6 -G672, 1 2.

136. Shepherd, A.P., and G.L. Riedel. Differences in reactive hyperemia between
the intestinal mucosa and muscularis. Am. J. Physiol. 247 (Gastrointestliver Physi-
ol. 10): G618-G622, 1984.

137. Shepherd, A.P., G.L. Riedel, L.C. Maxwell and J.W.Kiel. Selective vasodilators
redistribute intestinal blood ﬂow and de ress oxggen uptake. Am. J. Physiology 247
(Gastrointest. Liver Physiol. 10): G377- 384, 19 .

138. Shepherd, AP. and HJ. Granger. Autoregulatory escape in the gut. A system
analysis. Gastoentrology 65: 77-91, 1973.

$9. Sggmer, W. Localization of the energy pathway block in shock. Surgery 64:55-
, l .

140. Siregar, H., and C.C.Chou. Relative contribution of fat, rotien, carbohydrate,
and ethanol to intestinal hyperemia. Am. J. Physiol. 242: Gastrointest. Liver
Physiol. 5): G27-G31, 1982.

141. Sit, S.P., P. Nyhof, R. Gallavan, and CC. Chou. Mechanisms of glucose-in-
duced hyperemia in the jejunum. Proc. Soc. Exp. Biol. Med. 163: 273-277,1980.

142. Slotman GJ, Quinn JV, Burchard KW, Gann DS: Thromboxane, Prosta clin,
ﬁgsthe hemodynamic effects of graded bacteremic blood. Circ Shock 16:39 -404,

143. Strawitz, J.G., and H. Hift. Glucose and glycogen metabolism during hemor-
rhagic shock in the rat. Surg. Forum 11:112-114, 1960.

144. Tibblin, S., G. Burns, P.B. Habnloser and W.G. Schrenk,J r. The inﬂuence of

va otom on sugperior mesenteric artery blood ﬂow. Surg. Gyn. and Obstet. 129:
1 1- 1 4, 196 .

145. Vatner, S.F., D. Franklin, and R.L. VanCitters. Coronary and visceral vasoac-

tivity associated with eating and digestion in conscious dogs. Am. J. Physiol.
219:1380-1385, 1970.

146. Vatner, S.F., D. Franklin, and R.L. VanCitters. Mesenteric vasoactivity associ-

2119970 with eating and digestion in the conscious dog. Am. J. Physiol. 219: 170-174,

147. Vatner, S.F., T.A. Patrick, C.B. Higgins and D. Franklin. Regional circulatory
adjustments to eatin and digestion in conscious unrestricted primates. J. App .
Physiol. 36: 524-529,1 74

142

148. Warhurst, G., M. Lees, N.B. Higgs and LA. Turnber . Site and mechanisms of
action of kinins in rat ileal mucosa. Am. J. Physiol. 252 ( astrointest. Liver Physiol.
15): G293-G300, 1987.

149. Webb, PJ., J. Westwick, M.F. Scully, J. Zahavi, and V.V. Kakkar. Do rostacy-
clin and thomboxane play a role in endotoxic shock? Br. J. Surg. 68:720-72 , 1981.

150. Whittle, B.J.R. and JR. Vane. Prostanoid as regulators of gastrointestinal
function. In: L.R. Johnson (ed.) Physiology of the gastrointestinal tract. N.Y.: Raven
press. 1987. pp. 143-180.

151. Wiggers, CJ. Physiology of shock. Common Wealth Fund. New York, 1980.

152. Wu-Wang, C. and J. Neu. Microsomal arachidonate coversion in rat small
ilrétsestine during maturation. Prostaglandins Leukotrienes and Medicine 27: 71-80,

153. Yellin SA, Nguyen D, Quinn JV, Burchard KW, Crowley JP, Slotman GJ:
550233;; ggnlggg thromboxane A2 in septic shock: species differences. Circ Shock

154. Zwerfach, B.W., A. Fronek. The interplay of central and {Jersipheral factors in
irreversible hemorrhagic shock. Prog. Cardiovasc. dis. 18: 147, 97 .

 

 

‘11111111111111“