JOURNAL
OF SURGICAL
RESEARCH
52,596-600
(19%)
Effect of Capsaicin on Gastric Mucosal Injury and Blood Flow Following Bile Acid Exposure THEODORE R. SULLIVAN, JR., M.D., RICHARD MILNER, B.S., DANIEL T. DEMPSEY, M.D., AND WALLACE P. RITCHIE, JR., M.D., PH.D. Department Temple Presented
at the Annual
Meeting
of Surgery University
and the Frederick A. Reichle Surgical Research School of Medicine, Philadelphia, Pennsylvania
of the Association
for Academic
Springs,
Colorado,
November
20-23,
1991
METHODS
Adult mongrel dogs weighing approximately 20 kg were fasted for 48 hr (water ad lib), anesthetized with ketamine (5 mg/kg im) and sodium pentobarbital (50 mg/kg iv), intubated, and ventilated. The animals were maintained on a warming blanket at 27°C and were monitored continuously using an EKG and a carotid artery pressure transducer. Bilateral femoral cutdowns were performed and PE240 catheters were placed in both femoral arteries and connected to variable speed withdrawal pumps (Harvard Instruments). The right femoral vein was infused with 0.9% NaCl. The right carotid artery was cannulated with PE240 tubing attached to a pressure transducer which was advanced until the typical left ventricular volume curve appeared. A midline laparotomy was performed and a full-thickness wedge of proximal gastric wall pedicled on the left gastroepiploic artery and vein was mounted between lucite rings as previously described [17, 181. The upper ring served as a reservoir for the gastric mucosa. An in situ splenectomy was also performed. Mucosae were then exposed to a neutral test solution (NTS = 160 mM NaCl, pH 7.0) for a 30-min equilibration period. Each experiment in every dog was divided into two consecutive 70-min periods with a 15-min NTS equilibration period in between (Table 1). The first 30-minute segment of period I consisted of two 15-minute topical pretreatments of one of the following combinations: saline followed by capsaicin vehicle (NSS/VEH, n = 5); 4% lidocaine followed by vehicle (LIDO/VEH, n = 4), saline followed by capsaicin 160 puM (NSS/CAP, n = 5), or lidocaine followed by capsaicin (LIDO/CAP, n = 5). The subsequent 30-minute treatment interval consisted
INTRODUCTION
Acidified bile acid in contact with the gastric mucosa triggers a cascade of physiologic events which appear to be designed to blunt the injurious effects of excessive acid “back diffusion” on the mucosa. Specifically, as hydrogen ions pass from lumen into mucosa, gastric mucosal blood flow (GMBF) increases subtantially, presumably to buffer or to clear intramucosal acid. This is a critical protective response because, when it and it alone is blunted, the mucosa develops severe gross injury [15, 596 Inc. reserved.
Colorado
17, 181. The mediators of this response are unknown. The purpose of the present study was to test the hypothesis that afferent sensory neurons (ASN) are involved. This was accomplished using the ASN stimulant capsaicin and the local anesthetic lidocaine as pharmacologic manipulators of the system.
Topical bile acid at low pH stimulates gastric mucosal blood flow (GMBF), thereby limiting injury to surface epithelial cells (SEC). Capsaicin-sensitive afferent neurons (ASN) are possible mediators of the GMBF response and, therefore, of mucosal protection. In order to investigate the effect of topical capsaicin (ASN stimulant) and topical lidocaine (ASN inhibitor) on SEC exfoliation and GMBF, vascularized wedges of canine gastric corpus were mounted in lucite chambers. Mucosae were pretreated for 15 min with saline (NSS), 160 pilfcapsaicin (CAP), 4% lidocaine (LIDO), or CAP and LIDO, followed by a 30-min exposure to acid test solution (ATS; pH 1.2). The same mucosae were then pretreated in an identical fashion followed by a second 30min exposure to 5 m&f taurocholate (5 TC; pH 1.2). Parameters evaluated during both ATS and 5 TC periods were the luminal accumulation of DNA (DNAE, a sensitive marker of SEC exfoliation) and GMBF measured using radiolabeled microspheres. It was found that, relative to NSS pretreatment, CAP pretreatment increased GMBF and decreased DNAE during exposure to both ATS and 5 TC. LID0 blocked the CAP effect on GMBF but not on DNAE. Thus, ASN stimulation by CAP enhances GMBF and is protective. ANS inhibition blocks CAP’s GMBF increase but not its protective capabilities. Therefore, augmentation of GMBF is not the only mechanism by which ASNs blunt SEC exfoliation. 0 1992 Academic Press, Inc.
0022-4804/92 $4.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form
Surgery,
Laboratory, 19140
SULLIVAN
TABLE Period Pretreatment 15’/15’ NSS/VEH (n = 5) LIDO/VEH (n = 4) NSS/CAP (n = 5) LIDO/CAP (n = 5) Note. sequent lidocaine
ET
AL.:
GASTRIC
MUCOSAL
1
I
Period
II
Injury-Flux 30
Pretreatment 15’115’
Injury-Flux 30’
ATS
NSS/VEH
5TC
ATS
LIDO/VEH
ATS
NSS/CAP
ATS
LIDO/CAP
5TC 5TC 5TC
Outline of the experimental groups exposed to ATS and sub5 TC following pretreatment with either capsaicin (160 r&f), (4%), or both during two consecutive study periods.
of the topical application of an acid test solution (ATS = 100 mM HCl, 60 mM NaCl, 4 g carrier PEG, 5 j&i C14PEG, pH 1.2). At the conclusion of study period I the contents of the chamber were aspirated and assayed as described below. In addition, the mucosa was washed for 10 min with 30 ml of 10% N-acetylcysteine to facilitate recovery of exfoliated surface epithelial cells. During period II, each animal was pretreated in precisely the same manner as in period I, followed by a 30-min exposure to acidified 5 mM sodium taurocholate (5 TC = 5 mM NaTC, 100 mM HCl, 60 mM NaCl, 4 g PEG, 5 &i C14PEG, pH = 1.2). An N-acetylcysteine rinse was accomplished as before. Parameters evaluated during both periods I and II included net H+ ion flux (AH+, peq/30 min), luminal recovery of DNA (micrograms per 30 min), and GMBF (m/min/lOO g tissue), accomplished at the midpoint of each study period using cerium- and strontium-labeled microspheres. Details of all of these techniques have been described previously [20, 241. To assesswhether or not capsaicin per se had an effect on basal acid secretion (and therefore on AH+), four additional dogs were studied. After a 30-min equilibration period with NSS, 30 cc of either VEH or CAP were placed in the chamber for 15 min. Following aspiration of this solution from the chamber, 20 cc of NTS were instilled for an additional 15 min. This sample was then removed and titrated for H+ concentration. Calculations of H+ ion flux were then performed as previously described. Statistical analysis was performed using the unpaired Student t test. Data are expressed as mean value -t standard error of the mean. A P value equal to or less than 0.05 was considered to be statistically significant.
INJURY
AND
BLOOD
597
FLOW
CAP pretreatment significantly reduced luminal H+ ion loss as compared to NSS (CAP = -17 f 13 versus NSS = -66 ? 13 peq/30 min, P < 0.05). LID0 completely blocked this effect (LIDO/CAP = -71 -t 25 versus NSS = -66 f 13 peq/30 min). Interestingly, lidocaine alone caused a substantial increase in luminal H+ ion loss (-270 + 44 peg/30 min). During the topical application of 5 TC (period II), CAP pretreatment again significantly decreased luminal H+ ion loss (CAP = -527 k 38 versus NSS = -667 & 34 peq/30 min, P < 0.05). LID0 again completely blunted this response (LIDO/CAP -682 + 36 versus NSS -677 a 34 peq/30 min). H+ loss following pretreatment with LID0 alone appeared somewhat increased versus control (-764 +- 87 peq/30 min) but this difference was not statistically significant. Figure 2 illustrates the results with respect to DNA accumulation within the lumen. During exposure to ATS alone (period I) CAP significantly decreased luminal DNA appearance versus control (CAP = 24 + 9 versus NSS = 66 -+ 13 pg/30 min, P < 0.05). DNA accumulation following pretreatment with LID0 was not different from that of control. However, LID0 plus CAP resulted in significantly lower DNA accumulation as compared to that of control (LIDO/CAP = 15 f 5 versus NSS = 66 f 13 pg/30 min, P < 0.05). A completely similar pattern was observed during exposure to 5 TC (period II): pretreatment with either CAP or LID0 followed by CAP resulted in significantly lower luminal DNA accumulation than did either pretreatment with NSS or LID0 alone (CAP = 709 +- 136; LIDO/CAP = 602 f 70 versus NSS = 1180 f 135 or LID0 = 1352 + 126 Kg/30 min, P < 0.05).
n q q
NSSNEH
(n=5)
NSSICAP
(n=5)
LIDONEH
(n=4)
LlDOlCAP
(n=5)
Period I CATS)
Period
II
(5 TC)
RESULTS l
The results with respect to net H+ ion flux are illustrated in Fig. 1. During exposure to ATS alone (period I),
FIG. 1. TC after
p
OF SURGICAL
RESEARCH
52,596-600
(19%)
Effect of Capsaicin on Gastric Mucosal Injury and Blood Flow Following Bile Acid Exposure THEODORE R. SULLIVAN, JR., M.D., RICHARD MILNER, B.S., DANIEL T. DEMPSEY, M.D., AND WALLACE P. RITCHIE, JR., M.D., PH.D. Department Temple Presented
at the Annual
Meeting
of Surgery University
and the Frederick A. Reichle Surgical Research School of Medicine, Philadelphia, Pennsylvania
of the Association
for Academic
Springs,
Colorado,
November
20-23,
1991
METHODS
Adult mongrel dogs weighing approximately 20 kg were fasted for 48 hr (water ad lib), anesthetized with ketamine (5 mg/kg im) and sodium pentobarbital (50 mg/kg iv), intubated, and ventilated. The animals were maintained on a warming blanket at 27°C and were monitored continuously using an EKG and a carotid artery pressure transducer. Bilateral femoral cutdowns were performed and PE240 catheters were placed in both femoral arteries and connected to variable speed withdrawal pumps (Harvard Instruments). The right femoral vein was infused with 0.9% NaCl. The right carotid artery was cannulated with PE240 tubing attached to a pressure transducer which was advanced until the typical left ventricular volume curve appeared. A midline laparotomy was performed and a full-thickness wedge of proximal gastric wall pedicled on the left gastroepiploic artery and vein was mounted between lucite rings as previously described [17, 181. The upper ring served as a reservoir for the gastric mucosa. An in situ splenectomy was also performed. Mucosae were then exposed to a neutral test solution (NTS = 160 mM NaCl, pH 7.0) for a 30-min equilibration period. Each experiment in every dog was divided into two consecutive 70-min periods with a 15-min NTS equilibration period in between (Table 1). The first 30-minute segment of period I consisted of two 15-minute topical pretreatments of one of the following combinations: saline followed by capsaicin vehicle (NSS/VEH, n = 5); 4% lidocaine followed by vehicle (LIDO/VEH, n = 4), saline followed by capsaicin 160 puM (NSS/CAP, n = 5), or lidocaine followed by capsaicin (LIDO/CAP, n = 5). The subsequent 30-minute treatment interval consisted
INTRODUCTION
Acidified bile acid in contact with the gastric mucosa triggers a cascade of physiologic events which appear to be designed to blunt the injurious effects of excessive acid “back diffusion” on the mucosa. Specifically, as hydrogen ions pass from lumen into mucosa, gastric mucosal blood flow (GMBF) increases subtantially, presumably to buffer or to clear intramucosal acid. This is a critical protective response because, when it and it alone is blunted, the mucosa develops severe gross injury [15, 596 Inc. reserved.
Colorado
17, 181. The mediators of this response are unknown. The purpose of the present study was to test the hypothesis that afferent sensory neurons (ASN) are involved. This was accomplished using the ASN stimulant capsaicin and the local anesthetic lidocaine as pharmacologic manipulators of the system.
Topical bile acid at low pH stimulates gastric mucosal blood flow (GMBF), thereby limiting injury to surface epithelial cells (SEC). Capsaicin-sensitive afferent neurons (ASN) are possible mediators of the GMBF response and, therefore, of mucosal protection. In order to investigate the effect of topical capsaicin (ASN stimulant) and topical lidocaine (ASN inhibitor) on SEC exfoliation and GMBF, vascularized wedges of canine gastric corpus were mounted in lucite chambers. Mucosae were pretreated for 15 min with saline (NSS), 160 pilfcapsaicin (CAP), 4% lidocaine (LIDO), or CAP and LIDO, followed by a 30-min exposure to acid test solution (ATS; pH 1.2). The same mucosae were then pretreated in an identical fashion followed by a second 30min exposure to 5 m&f taurocholate (5 TC; pH 1.2). Parameters evaluated during both ATS and 5 TC periods were the luminal accumulation of DNA (DNAE, a sensitive marker of SEC exfoliation) and GMBF measured using radiolabeled microspheres. It was found that, relative to NSS pretreatment, CAP pretreatment increased GMBF and decreased DNAE during exposure to both ATS and 5 TC. LID0 blocked the CAP effect on GMBF but not on DNAE. Thus, ASN stimulation by CAP enhances GMBF and is protective. ANS inhibition blocks CAP’s GMBF increase but not its protective capabilities. Therefore, augmentation of GMBF is not the only mechanism by which ASNs blunt SEC exfoliation. 0 1992 Academic Press, Inc.
0022-4804/92 $4.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form
Surgery,
Laboratory, 19140
SULLIVAN
TABLE Period Pretreatment 15’/15’ NSS/VEH (n = 5) LIDO/VEH (n = 4) NSS/CAP (n = 5) LIDO/CAP (n = 5) Note. sequent lidocaine
ET
AL.:
GASTRIC
MUCOSAL
1
I
Period
II
Injury-Flux 30
Pretreatment 15’115’
Injury-Flux 30’
ATS
NSS/VEH
5TC
ATS
LIDO/VEH
ATS
NSS/CAP
ATS
LIDO/CAP
5TC 5TC 5TC
Outline of the experimental groups exposed to ATS and sub5 TC following pretreatment with either capsaicin (160 r&f), (4%), or both during two consecutive study periods.
of the topical application of an acid test solution (ATS = 100 mM HCl, 60 mM NaCl, 4 g carrier PEG, 5 j&i C14PEG, pH 1.2). At the conclusion of study period I the contents of the chamber were aspirated and assayed as described below. In addition, the mucosa was washed for 10 min with 30 ml of 10% N-acetylcysteine to facilitate recovery of exfoliated surface epithelial cells. During period II, each animal was pretreated in precisely the same manner as in period I, followed by a 30-min exposure to acidified 5 mM sodium taurocholate (5 TC = 5 mM NaTC, 100 mM HCl, 60 mM NaCl, 4 g PEG, 5 &i C14PEG, pH = 1.2). An N-acetylcysteine rinse was accomplished as before. Parameters evaluated during both periods I and II included net H+ ion flux (AH+, peq/30 min), luminal recovery of DNA (micrograms per 30 min), and GMBF (m/min/lOO g tissue), accomplished at the midpoint of each study period using cerium- and strontium-labeled microspheres. Details of all of these techniques have been described previously [20, 241. To assesswhether or not capsaicin per se had an effect on basal acid secretion (and therefore on AH+), four additional dogs were studied. After a 30-min equilibration period with NSS, 30 cc of either VEH or CAP were placed in the chamber for 15 min. Following aspiration of this solution from the chamber, 20 cc of NTS were instilled for an additional 15 min. This sample was then removed and titrated for H+ concentration. Calculations of H+ ion flux were then performed as previously described. Statistical analysis was performed using the unpaired Student t test. Data are expressed as mean value -t standard error of the mean. A P value equal to or less than 0.05 was considered to be statistically significant.
INJURY
AND
BLOOD
597
FLOW
CAP pretreatment significantly reduced luminal H+ ion loss as compared to NSS (CAP = -17 f 13 versus NSS = -66 ? 13 peq/30 min, P < 0.05). LID0 completely blocked this effect (LIDO/CAP = -71 -t 25 versus NSS = -66 f 13 peq/30 min). Interestingly, lidocaine alone caused a substantial increase in luminal H+ ion loss (-270 + 44 peg/30 min). During the topical application of 5 TC (period II), CAP pretreatment again significantly decreased luminal H+ ion loss (CAP = -527 k 38 versus NSS = -667 & 34 peq/30 min, P < 0.05). LID0 again completely blunted this response (LIDO/CAP -682 + 36 versus NSS -677 a 34 peq/30 min). H+ loss following pretreatment with LID0 alone appeared somewhat increased versus control (-764 +- 87 peq/30 min) but this difference was not statistically significant. Figure 2 illustrates the results with respect to DNA accumulation within the lumen. During exposure to ATS alone (period I) CAP significantly decreased luminal DNA appearance versus control (CAP = 24 + 9 versus NSS = 66 -+ 13 pg/30 min, P < 0.05). DNA accumulation following pretreatment with LID0 was not different from that of control. However, LID0 plus CAP resulted in significantly lower DNA accumulation as compared to that of control (LIDO/CAP = 15 f 5 versus NSS = 66 f 13 pg/30 min, P < 0.05). A completely similar pattern was observed during exposure to 5 TC (period II): pretreatment with either CAP or LID0 followed by CAP resulted in significantly lower luminal DNA accumulation than did either pretreatment with NSS or LID0 alone (CAP = 709 +- 136; LIDO/CAP = 602 f 70 versus NSS = 1180 f 135 or LID0 = 1352 + 126 Kg/30 min, P < 0.05).
n q q
NSSNEH
(n=5)
NSSICAP
(n=5)
LIDONEH
(n=4)
LlDOlCAP
(n=5)
Period I CATS)
Period
II
(5 TC)
RESULTS l
The results with respect to net H+ ion flux are illustrated in Fig. 1. During exposure to ATS alone (period I),
FIG. 1. TC after
p
