In the muscle layer of the glandular portion of the rat stomach, in vivo capsaicin pretreatment mark- edly reduced calcitonin gene-related peptide-like.
GASTROENTEROLOGY
1991;101:1505-1511
Differential Effect on Neuropeptide Release of Different Concentrations of Hydrogen Ions on Afferent and Intrinsic Neurons of the Rat Stomach PIERANGELO GEPPETTI, MANUELA TRAMONTANA, STEFANO EVANGELISTA, DANlELA RENZI, CARLO A. MAGGI, BRUNO M. FUSCO, and ELENA DEL BLANC0 Institute of Internal Medicine IV and Gastroenterology Unit, University of Florence, and Department of Pharmacology, Menarini Pharmaceuticals, Florence, Italy
In the muscle layer of the glandular portion of the rat stomach, in vivo capsaicin pretreatment markedly reduced calcitonin gene-related peptide-like immunoreactivity (CGRP-LI) hut did not affect substance P-like immunoreactivity (SP-LI). Accordingly, in vitro superfusion of slices of this tissue with capsaicin (10 u.mol/L) released CGRP-LI but not SP-LI, whereas both neuropeptides were released by 80 mmol/L K’. Exposure to relatively low-pH (pH 8) physiological salt solution induced an increase in the CGKP-LI outflow that was reduced by 70% in a Ca’+-free medium and was completely abolished by a previous exposure to capsaicin. However, superfusion with pH-6 medium did not produce any detectable SP-LI release. After exposure to pH-6 medium, both capsaicin and high-K+ medium were still able to release a consistent quantity of CGKP-LI and SP-LI, respectively. Increased mucosal blood flow induced by acid back-diffusion is considered a protective mechanism against mucosal gastric lesion. The present findings suggest that hydrogen ions diffusing into the gastric wall may promote protective vasodilatation by activating the “efferent” function of capsaicin-sensitive nerves without affecting the secretory process of other intrinsic peptidergic neurons.
A
mong the various protective mechanisms that operate against an injury of the gastric mucosa, a major role is played by the increase in mucosal blood flow in response to the disruption of the gastric mucosal barrier and acid back-diffusion (l-3). However, the mechanism through which acid backdiffusion induces vasodilatation is still unknown. Peptide-containing sensory nerves at the visceral
level not only transmit interoceptive information to the brain, but can also exert a local “efferent” function determined by transmitter release from their peripheral endings (4,5). Knowledge in this field has been gained by the use of capsaicin, the pungent principle of the plants of the genus Capsicum, which in a time-related manner selectively excites and then induces functional impairment of a subset of primary sensory neurons (4,s). The stomach receives a dense peptidergic afferent innervation mainly from splanchnit nerves (6-9). The observation that functional ablation of capsaicin-sensitive sensory neurons exacerbates (lO,ll),while their stimulation protects (12) from, the formation of mucosal lesion in the stomach has led to the consideration of this subset of primary afferents as part of a relevant control mechanism for gastric mucosal protection (13). Furthermore, activation of peripheral endings of capsaicin-sensitive nerves in the stomach has proven to evoke the release of vasodilatatory peptides (14,15), such as substance P (SP) or calcitonin gene-related peptide (CGRP), the administration of which protects against gastric mucosal damage (16,17). There is some evidence that capsaicin-sensitive afferents are sensitive to acid (18,19).Exposure to low-pH medium increased conductance for cations in sensory neurons in culture (20) and particularly in those neurons sensitive to capsaicin (21). We have
Abbreviations used in thispaper:CGRP-LI, calcitonin generelated peptide-like immunoreactivity; SP-LI, substance P-like immunoreactivity. o 1991 by the American Gastroenterological Association 0016-5065/91/$3.00
1506
GEPPETTI ET AL.
GASTROENTEROLOGY
shown recently that exposure to low-pH physiological salt solutions induces a Ca’+-dependent release of CGRP from capsaicin-sensitive afferents in muscle slices of guinea pig urinary bladder (22). Because hydrogen ions may stimulate sensory nerves, and considering that acid back-diffusion increases gastric mucosal blood flow, the hypothesis might be advanced that proton-induced vasodilatation in the stomach is, at least in part, mediated by activation of the efferent function of capsaicinsensitive sensory nerves. Here, we have investigated whether exposure of the muscle layer of the rat stomach to low-pH media leads to release of sensory neuropeptide in a manner similar to that already observed in the guinea pig urinary bladder (22,23). The muscle layer was chosen for three reasons: (a) experimental support was needed for the hypothesis that after disruption of the gastric mucosal barrier, acid back-diffusion may activate sensory neuropeptide secretion in the deep layers of the gastric wall, in which case the muscle layer would be used as a paradigm of a tissue located beneath the gastric mucosal barrier; (b) because we have previously characterized the proton-evoked release in the mucosafree muscle slices of guinea pig urinary bladder (22,23), it seemed appropriate also to use a muscle preparation in this work; (c) the amount of CGRP releasable by capsaicin from the muscle layer of the rat stomach was 10 times higher than that observed from the mucosa (Renzi D., 1990, unpublished observation). Previous reports (8,9,24,25) indicate that in the rat stomach, CGRP and SP are suitable markers of distinct neuronal populations, e.g., capsaicin-sensitive afferents and intrinsic gastric neurons, respectively. Therefore, to ascertain whether protons, at least at certain concentrations, might exert a selective action on afferent nerves, release of both SP and CGRP by low-pH media was compared with those produced by capsaicin and by a nonspecific depolarizing agent, such as a high-K’ medium. Materials
and Methods
Neuropeptide
Tissue Measurement
Sprague-Dawley (Nossan strain] rats were used throughout the study. For in vivo treatment, capsaicin was dissolved in a vehicle containing 10% Tween 80, 10% ethanol, and 80% saline. Newborn rats received either the vehicle or 50 mg/kg of capsaicin SC on the second day of life. Two months later, rats were killed and the glandular portion of the stomach was removed and freed of the mucosa. Histological preparations of the samples (not shown) indicated that with the procedure adopted, the submucosal layer was dissected out with the mucosa, and consequently samples comprised only the muscle layer of the rat stomach. Tissue samples (80-100 mg) were homoge-
Vol. 101, No. 6
nized in 95°C 2N acetic acid (l/10 wt/vol) and kept in a boiling water bath for 10 minutes. After centrifugation was (20,OOOgat 4°C for 20 minutes), the supernatant freeze-dried. Release Experiments Different physiological phosphate buffers derived from normal Krebs’ solution, the composition of which is reported in Table I, were used. If needed, fine adjustment of pH was made using NaOH or HCl. Two-month-old rats of both sexes (280-350 g) that had fasted overnight were decapitated, and the stomachs were rapidly removed and placed in pH-7.4 physiological salt solution at 4”C, where all further steps were performed. The glandular portion of the stomach was excised and freed of the mucosa. Tissue slices (400-600 pm) were prepared with a tissue slicer (McIlwain, U.S.A.) and transferred to 1.5 perfusion chambers. Tissues (100-150 mg) were superfused at 0.4 mL/min with oxygenated (100% 0,) pH-7.4 solutions maintained at 37”C, containing bovine serum albumin (0.1%) and thiorphan (I pmol/L). A 30-minute equilibration period was allowed to elapse before the experiments were begun. Five-minutes (2 mL) fractions were collected before (n = 2), during (n = a), and after (n = 2) exposure to different stimuli. Eight chambers were run at the same time. Fractions were then added with acetic acid to a 2N final concentration and freeze-dried. At the end of the experiments, the tissues were blotted two or three times on a filter paper and weighed. In some experiments, in which samples were superfused with pH-5 or pH-6 media, the pH of each fraction was measured by a pH-meter (Titralab; Radiometer, Denmark) before the addition of acetic acid. Radioimmunoassays After reconstitution with assay buffer (pH 7.4, 0.1 mol/L phosphate buffer containing 0.1% bovine serum albumin, 0.01% NaN,, and 0.9% NaCl), CGRP-like immunoreactivity (CGRP-LI) was measured as reported previously (26) using rat a-CGRP as standard, RAS 6012 anti-CGRP antiserum, and [‘251]iodohistidyl CGRP. The coefficient of percent variation was < 10% for values between 20 and 300 pmol/L. The sensitivity of the assay was 2 fmol/tube. The
Table 1. Composition (mmollL) of the Various Physiological Salt Solutions Ca’+-free PH NaCl KC1 CaCl, W@O, Na,HPO, KH,P04 Glucose
7.4
6
5
7.4
6
5
134.5 5.4 2.5 1.5 5.45 1.21 11
39.2
139.9 2.5 1.5 0.06 6.6 11
137.9 5.4 1.5 5.45 1.21 11
142.6 0.74 1.5 0.8 5.66 11
143.3 1.5 0.06 6.6 11
0.74 2.5 1.5 0.8 5.86 11
NOTE. Ca”-free buffered solutions also contained 1 mmol/L EDTA. In the I30mmol/L K+ solution, NaCl and KC1 were 61.1 and 78.8 mmol/L, respectively.
PROTON RELEASES CGRP IX THE RAT STOMACH
December 1991
antiserum cross-reacted to 100% with rat f3-CGRP and human CGRP-I and -11. Assay for SP-LI was performed as
described elsewhere (25). Substance P as standard, 144 anti-SP antiserum, and [‘ZSI]Bolton and Hunter-conjugated SP were used. The coefficient of percent variation was < 10% for values between 10 and 300 pmol/L. The sensitivity of the assay was 1.1 fmol/tube. The antiserum crossreacted to 1% with neurokinin A, 0.5% with neurokinin B, and < 0.1% with physalaemin
Table 2. Effect of Superfusion With Capsaicin, High-K’ Solution, andpH-6 Solution on the Total Evoked Release of Calcitonin Gene-Related Peptide-Like Immunoreactivity and Substance P-Like Immunoreactivity From Muscle Slices of the Rat Stomach CGRP-LI n
and eledosin.
Statistical Analysis
(fmol.g~30min~‘)
SP-LI n
8 6
7481 k 524
8
ND
3602 ? 586
6
546 f 68
PHI
7
1626 k 271
9
ND
NOTE. Total evoked
sum of values observed during exposure to the stimulus (and during the next 10 minutes] subtracted by the mean basal value.
applied for 20 minutes. Data are mean ? SEM. ND, not detectable.
Peptides, thiorphan, and the 6012 anti-CGRP antiserum were from Peninsula (Belmont, CA]; capsaicin was from Sigma (St. Louis, MO); [‘251]iodohistidyl-CGRP and [“‘I]Bolton and Hunter-conjugated SP were purchased from Amersham (Buckinghamshire, England]; and 144 anti-SP antiserum was a kind gift of Dr. P. Pradelles (DRIPP-LERI; CEN-Saclay, Gif-sur-Yvette, France).
Results Calcitonin Gene-Related Peptide-Like Immunoreactivity and Substance P-Like Immunoreactivity Tissue Levels
The CGRP-LI found in the muscle layer of the glandular portion of the stomachs of vehicle-treated animals amounted to 44.2 ? 4.2 pmol/g wet weight (n = 5), whereas in capsaicin-pretreated animals it was reduced to 4.8 + 0.7 pmol/g (n = 5, P < 0.001). In contrast, the SP-LI content, which in vehicletreated rats averaged 91.7 + 12.7 pmol/g (n = 5), was unaffected by capsaicin administration (87.3 t 10.9 pmol/g; n = 5).
Release by Capsaicin
and High-K+ Medium
Exposure of mucosa-free muscle slices of the glandular portion of the rat stomach to 10 Fmol/L capsaicin produced a remarkable total evoked release of CGRP-LI (Table 2) that was no longer observed at a second challenge with the drug (not shown). Capsaitin vehicle (0.001% ethanol) did not release any CGRP-LI (not shown). Administration of 10 kmol/L capsaicin failed to produce a measurable release of SP-LI (Table 2). Superfusion of the stomach slices with 80 mmol/L K’ produced a marked release of both CGRP-LI and SP-LI (Table 2). Exposure to capsaicin
(fmol.g.30mine’)
Capsaicin (10 pmoliL) K’ (80 mmol/L)
All values in the text and figures are mean + SEM. Statistical analysis was performed using Student’s t test for unpaired data. Total evoked release was calculated as the
Drugs
1507
observed minutes)
release was calculated by the sum of values during exposure to the stimulus (and during the next 10 subtracted by the mean basal value. Each stimulus was
(10 Fmol/L for 20 minutes) abolished the high-K’evoked CGRP-LI release and left that of SP-LI unchanged (not shown). Release by Low-pH Media
Basal outflow of CGRP-LI from muscle slices of the glandular portion of the rat stomach was 89 -t 11 fmol . g-’ . fraction-’ (n = 6) and 82 k 12 fmol . g-’ . fraction-’ (n = 6) in control preparation and in preparations incubated (30 minutes) in a Ca’+-free medium acid containing 1 mmol/L ethylenediaminetetraacetic (EDTA), respectively (Figure 1). Exposure to pH-6 medium produced a significant increase in the CGRP-LI outflow (total evoked release being 1590 -t 210 fmol . g-’ . 30 min-‘; n = 6) (Figure l), whereas it did not evoke any detectable SP-LI release (Table 2). Superfusion with pH-6 CaZi-free 1 mmol/L EDTA containing medium produced a slight increase in the CGRP-LI outflow (577 k 122 fmol . g-’ . 30 min-‘; n = 6) that was 35% of that obtained in parallel experiments from control preparations (Figure 1). To ascertain whether 1 mmol/L EDTA might produce any neurotoxic action, thus affecting the secretory function of sensory nerves, some samples (n = 6) were exposed (for 40 minutes) to the Ca*+-free medium containing 1 mmol/L EDTA while control samples (n = 6) were run (for 40 minutes) with the physiological salt solution. After all the samples were washed for 60 minutes with the physiological salt solution, they were exposed to 10 pmol/L capsaicin. The capsaicin-evoked release of CGRP-LI from the samples preexposed to 1 mmol/L EDTA (7796 rf: 477 fmol . g-’ . 30 min.‘) was not different from that observed from control preparations (7221 & 561 fmol . g-’ . 30 min-‘). In other experiments, the basal CGRP-LI outflow was 79 2 12 fmol . g. ’ . fraction-’ (n = 6) and 76 -cfmol . g-’ . fraction-’ (n = 6) in control preparations
1508
GEPPETTI ET AL.
GASTROENTEROLOGY
6
7.4
7.4
I-+ _L.
Vol. 101, No. 6
crease in the SP-LI outflow in 4 of the 10 preparations tested. In these 4 experiments, total evoked release was 338 + 49 fmol . g-’ . 30 min-‘. The pH in the effluent of the last fraction of superfusion with pH-5 medium was not significantly different in samples pretreated with capsaicin (5.36 + 0.1; n = 5) or superfused with a Ca’+-free medium (5.30 + 0.09; n = 5) from that of control samples (5.32 + 0.08; n = 5). After exposure to pH-6 physiological salt solution (20 minutes), 10 kmol/L capsaicin still evoked a significant increase in the CGRP-LI outflow that was 32% of that observed in control samples (Table 3). Likewise, after superfusion with pH-6 medium, 80 mmol/L K’ still produced a significant increase in both CGRP-LI and SP-LI outflow that amounted to 46% and 34% of those observed in control preparations (Table 3).
1’
Fractions Figure 1. Effect of superfusion with pH-6 physiological salt solution on the oufflow of CGRP-LI from mucosa-free slices of the rat stomach. Dashed line indicates the CGRP-LI outilow by pH 6 in a Cazf-free medium containing 1 mm01 EDTA. Bars are mean 4 SEM of six experiments. *P < 0.01, Student’s t test.
and in preparations preexposed to capsaicin (10 p,mol/L for 20 minutes), respectively (Figure 2). After exposure to capsaicin, basal values were not different from those observed before administration of the drug (not shown). Preexposure to capsaicin abolished the release of CGRP-LI by pH-6 medium (Figure 2). The lowest pH level in the effluent was observed at the last fraction of superfusion with the pH-6 solution. The pH value in this fraction was not significantly different in samples preexposed to capsaicin (6.16 f 0.07; n = 5) and samples superfused with a Ca’+-free medium (6.12 ?Z 0.08; n = 5) from that in control samples (6.15 ? 0.09; n = 5). Likewise, no statistical difference of the pH value was observed in all the other fractions of capsaicin-pretreated and Ca’+-free superfused preparations as compared with control samples (not shown). Superfusion of the tissues with pH-5 physiological salt solution produced a remarkable increase in the CGRP-LI outflow (total evoked release, 2832 f 380 fmol . g-’ . 30 min-‘; n = 8). The pH 5-evoked release was reduced to 717 + 89 fmol . g-’ . 30 min-’ (n = 7; P < 0.001) when in the medium Ca*+ was omitted and EDTA (1 mmol/L) was added. After exposure to 10 kmol/L capsaicin (20 minutes) pH-5 medium failed to significantly increase the CGRP-LI outflow (210 fmol . g-’ . 30 min-‘; n = 6). Superfusion with pH-5 solution produced a slight although significant in-
Discussion Calcitonin gene-related peptide in the rat stomach is almost entirely stored in capsaicin-sensitive nerves (8,24), whereas SP, although present in a small proportion in these nerves, is mostly contained within intrinsic neurons of the gastric wall (6,7,25). The present findings dealing with tissue levels of the two
PH
7.4
1
6
1
7.4
]
11 +++** J___ J_,
J-r
‘-,
J-_--I-
Fractions Figure 2. Effect of superfusion with pH-6 physiological salt solution on the CGRP-LI outIlow from mucosa-free slices of the rat stomach preexposed in vitro to capsaicin (10 CmohL, 20 minutes: dashed line) or capsaicin-vehicle (0.001% ethanol, 20 minutes; continous line). Bars are mean k SEM of six experiments. *P < 0.001, Student’s t test.
December
PROTON
1991
RELEASES
CGRP IN THE RAT STOMACH
1509
Table 3. Total Evoked Release of Calcitonin Gene-Related Peptide-Like immunoreactivity and Substance P-Like Immunoreactivity by Capsaicin or High-K’ Medium From Muscle Slices of the Rat Stomach in Control Preparations and in Preparations Preexposed to pH-6 Medium for 20 Minutes K+
Capsaicin (IO pmol/L)
(fmol
g 30 g-’
7290
mine’)
30 min-‘)
NOTE. Total evoked ND, not detectable.
1
542
2334
ND release
was calculated
as indicated
Control (n = 5)
Preexposed (n = 6)
Control (n = 6) CGRP-LI (fmol SP-LI
(80
3237
k 645
610 ? 98
ND in Table 2. Capsaicin
neuropeptides agree with these previous data. In fact, in vivo capsaicin pretreatment produced a large reduction of CGRP-LI, but not of SP-LI, in the muscle layer of the glandular portion of the rat stomach. Release experiments further support this view. High K’ increased the outflow of both CGRP-LI and SP-LI. However, capsaicin, which was more effective than high K’ in releasing CGRP-LI failed to affect the SP-LI outflow. Although it is likely that capsaicin might release a small amount of sensory SP, these results clearly indicate that in the present experimental conditions we were able to detect secretion of CGRP-LI from capsaicin-sensitive fibers (8,24) and of SP-LI only from capsaicin-insensitive nerves, possibly intrinsic neurons of the stomach (6,7,25). The finding that the increased CGRP-LI outflow produced by low-pH physiological salt solutions was markedly reduced when Ca*+ ions were omitted in the medium or abolished if tissue slices were preexposed to capsaicin supports the view that hydrogen ions induced an exocytotic process of the neuropeptide from capsaicin-sensitive fibers. Protons may induce profound modification of neuronal activity both at the membrane level and in the cytosol(2 7-29). Therefore, the possibility that neuropeptide release relies on nonexocytotic process cannot be completely excluded at present. Neuropeptide secretion by low pH observed in the rat stomach appears to be identical to that described recently from the muscle slices of the guinea pig urinary bladder (22,23). This finding suggests that proton-induced activation of the “efferent” function of capsaicin-sensitive afferents, e.g., local neuropeptide release from their peripheral endings (4,5), may represent a general phenomenon irrespective of the tissue or animal species analyzed. The observation that pH-6 medium released a remarkable amount of CGRP-LI without affecting the SP-LI outflow speaks in favor for a selective action of hydrogen ions on capsaicin-sensitive nerves, providing that their concentration is only moderately increased. Selectivity by hydrogen ions in affecting the secretory function of sensory nerves seems dependent
k 586
and high-K+
medium
were superfused
mmoliL) Preexposed (n = 5)
1507 213
+ 284 2 45
for 20 minutes.
on their concentration. In fact, exposure to pH 5, which released a large amount of CGRP-LI, also produced a slight SP-LI release in 40% of the preparations tested. Therefore, this latter evidence indicates that the more acidic the pH of the bathing medium, the less selective the secretory process activated by protons, which may involve different neuronal populations. Regarding the origin of the SP-LI release at pH 5, it is difficult to assume that it is released from capsaicin-sensitive afferents. In fact, capsaicin, a more powerful stimulus than pH 5 in secretion of sensory neuropeptides, failed to produce the secretion of significant amount of SP-LI. It must be also noted that a relatively slight increase in proton concentration (pH 6) did not produce a total impairment of both capsaicin-sensitive sensory nerves and SP-containing neurons of the rat stomach; after exposure to pH 6 for 20 minutes, capsaicin and high-K+ medium were still able to evoke significant release of CGRP-LI and SP-LI, respectively. Maintenance and facilitation of mucosal blood flow is an important mechanism for gastric mucosal protection, and increased mucosal blood flow represents a protective mechanism activated by acid back-diffusion (l-3). Calcitonin gene-related peptide has been suggested to protect against mucosal injury by its vasodilatatory property (13,16). Recent evidence (16,30,31) that CGRP infusion increased the blood flow to the stomach seems to support this hypothesis. There is a great deal of evidence indirectly suggesting that capsaicin-sensitive sensory nerves distributed around gastric arterial vessels play an important role as protective structures against mucosal lesion, possibly by releasing vasodilatatory peptides (10-12). Defunctionalization of capsaicin-sensitive sensory neurons obtained by in vivo or in vitro administration of capsaicin has represented an important pharmacological instrument for unraveling the different roles played by these nerves in various tissues (4,5). It is also well known that capsaicin is also a powerful exogenous stimulatory compound of this subset of primary sensory neurons at the gastric level (14,15).
1510
GEPPETTI
ET AL.
However, little is known about endogenous agents able to stimulate the dual sensory-efferent function of capsaicin-sensitive nerves and about the possible relevance of such stimulation in tissue-specific pathophysiological conditions. In the case of the stomach, the possibility that elevated hydrogen ion concentration entering submucosal layers leads to release of potent vasodilatatory transmitter from sensory nerves possesses a peculiar interest for the pathogenesis of gastric mucosal lesion. The present data indicate that the hydrogen ions themselves may activate the efferent function of capsaicin-sensitive sensory neurons in the rat stomach. Neuropeptide secretion may occur even when the pH of the bathing medium is relatively low (pH 6) and certainly much higher than the usual pH of the intraluminal gastric fluid. At least at this pH value, the action of hydrogen ions in promoting neurosecretion appears to involve preferentially capsaicinsensitive fibers and not other peptidergic intrinsic neurons of the stomach. The hypothesis may be advanced that after mucosal weakening or damage, even a small quantity of hydrogen ions diffusing into the submucosal tissue, which per se might not represent a threatening factor for tissue integrity, is sufficient for the prompt activation of a vasodilatatory protective mechanism directed to the disposal of (13). Finally, harmful higher proton concentrations the present findings suggest that vasodilatatory response to low concentrations of hydrogen ions might be mostly caused by stimulation of capsaicin-sensitive sensory nerves, although participation of enteric neurons cannot be excluded when the process is activated by higher proton concentrations.
References 1. Ritchie WP. Acute gastric damage induced by bile salts, acid and ischemia. Gastroenterology 1975;68:699-707. 2. Bruggeman TM, Wood JG, Davenport MW. Local control of blood flow in the dog’s stomach: vasodilatation caused by acid back-diffusion following topical application of salicylic acid. Gastroenterology 1979;77:736-744. 3. Stalinger M, Schiessel R, Hung CR, Silen W. H’ back diffusion stimulating gastric mucosal blood flow in the rabbit fundus. Surgery 1981:89:232-236. sen4. Holzer P. Local effector functions of capsaicin-sensitive sory nerve endings: involvement of tachykinins, calcitonin gene-related peptide and other neuropeptides. Neuroscience 1988;24:739-768. 5. Maggi CA, Meli A. The sensory-efferent function of capsaicinsensitive sensory neurons. Gen Pharmacol1988;19:1-43. 6, Shultzberg M, Hokfelt T, Nilsson G, Terenius L, Rehfeld L, Brown M, Elde R, Goldstein M, Said S. Distribution of peptideand catecholamine-containing neurons in the gastrointestinal tract of rat and guinea-pig: immunohistochemical studies with antisera to substance P, vasoactive intestinal polypeptide enkephalins, somatostatin, gastrin/cholecystokinin, neurotensin and dopamine B-hyroxyalse. Neuroscience 1980;5:689744.
GASTROENTEROLOGY
Vol. 101, No. 6
7. Sharkey KA, Williams RG, Dockray GJ. Sensory substance P innervation of the stomach and pancreas. Gastroenterology 1984;87:914-921. 8. Sternini C, Reeve JR, Brecha N. Distribution and characterization of calcitonin gene-related peptide immunoreactivity in the digestive system of normal and capsaicin treated rats. Gastroenterology 1987;93:852-862. 9. Green T, Dockray GJ. Characterization of the peptidergic afferent innervation of the stomach in the rat, mouse and guinea-pig. Neuroscience 1988;25:181-193. 10. Szolcsanyi J, Barth0 L. Impaired defense mechanism to peptic ulcer in the capsaicin desensitized rat. In: Mozsik G, Hanninen 0, Javor T, eds. Gastrointestinal defense mechanisms. Oxford and Budapest: Pergamon Press and Akademiai Kiado, 1981;3951. 11. Holzer P, Sametz W. Gastric mucosal protection against ulcerogenie factors in the rat mediated by capsaicin-sensitive afferent neurons. Gastroenterology 1986;91:975-981. 12. Holzer P, Lippe IT. Stimulation of afferent nerve endings by intragastric capsaicin protects against ethanol-induced damage of gastric mucosa. Neuroscience 1988;27:981-987. 13. Holzer P. Afferent nerve-mediated control of gastric mucosal blood flow and protection. In: Costa M, Gorini S, Maggi CA, Surrenti C, eds. Sensory nerves and neuropeptides in gastroeneterology: from basic science to clinical perspectives. New York: Plenum, 1991:3-16. 14. Renzi D, Santicioli P, Maggi CA, Surrenti C, Pradelles P, Meli A. Capsaicin-induced release of substance P-like immunoreactivity from guinea-pig stomach in vitro and in vivo. Neurosci Lett 1988;92:254-258. 15. Holzer P, Peskar BM, Peskar BA, Amann R. Release of calcitonin gene-related peptide induced by capsaicin in the vascularly perfused rat stomach. Neurosci Lett 1990;108:195-200. 16. Lippe IT, Lorbach M, Holzer P. Close arterial infusion of calcitonin gene-related peptide into the rat stomach inhibits aspirin and ethanol-induced hemorrhagic damage. Regul Pept 1989;26:35-46. 17. Evangelista S, Lippe IT, Rover0 P, Maggi CA, Meli A. Tachykinins protect against ethanol-induced gastric lesions in rats. Peptides 1989;10:79-81. afferents con18. Martling C-A, Lundberg JM. Capsaicin-sensitive tribute to the acute airway edema following tracheal instillation of hydrochloric acid or gastric juice in the rat. Anesthesiology 1988;68:350-356. 19. Cervero F, McRitchie MA. Neonatal capsaicin does not affect unmyelinated efferent fibres of the autonomic nervous system: functional evidence. Brain Res 1982;239:283-288. 20. Krishtal OA, Podiplikho VI. A receptor for proton in the
21.
22.
23.
24.
membrane of sensory neurons may participate in nociception. Neuroscience 1981;6:2599-2601. Bevan C, Yeats JC. Protons activate a sustained inward current in a subpopulation of rat isolated dorsal root ganglia (DRG) neurons. J Physiol 1989;417:81P. Geppetti P, Tramontana M, Patacchini R, Del Bianco E, Santicioli P, Maggi CA. Neurochemical evidence for the activation of the “efferent” function of capsaicin-sensitive nerves by lowering of the pH. Neurosci Lett 1990;114:101-106. Geppetti P, Del Bianco E, Patacchini R, Santicioli P, Maggi CA, Tramontana M. Low pH-induced release of calcitonin generelated peptide from capsaicin-sensitive sensory nerves: mechanism of action and biological response. Neuroscience 1991; 41:295-301. Mulderry PK, Gathei MA, Spokes RA, Jones PM, Pierson AM, Hamid QA, Kanse S, Amara SG, Burrin JM, Legon S, Polak JM, Bloom SR. Differential expression of (Y-CGRP and B-CGRP by primary sensory neurons and enteric autonomic neurons of the rat. Neuroscience 1988;25:195-201.
December
1991
25. Sternini C, Anderson K, Gretchen F, Krause JE, Brecha N. Expression of sustance P/neurokinin A-encoding preprotachykinin messenger ribonucleic acids in the rat enteric nervous system. Gastroenterology 1989;97:348-356. 26. Geppetti P, Frilli S, Renzi D, Santicioli P, Maggi CA, Theodorsson E, Fanciullacci M. Distribution of calcitonin gene-related peptide immunoreactivity in various rat tissues: correlation with substance P and other tachykinins and sensitivity to capsaicin. Regul Pept 1988;123:289-293. 27. Wanke E, Carbone E, Testa PL. The sodium channel and intracellular H’ blockage in squid axons. Nature 1980;287:6263. 28. Konnerth A, Lux HD, Morad M. Proton-induced transformation of calcium channel in chick dorsal root ganglion cells. J Physiol 1987;386:603-633. 29. Gruel DL, Barker JL, Huang L-YM, Ferguson MacDonald J, Smith TG Jr. Hydrogen ions have multiple effects on the
PROTON
RELEASES
CGRP IN THE RAT STOMACH
1511
excitability of cultured mammalian neurons. Brain Res 1980; 183:247-252. 30. DiPette DJ, Schwarzenberger K, Kerr N, Holland OB. Systemic and regional hemodynamic effects of calcitonin gene-related peptide. Hypertension 1987;9(Suppl 3):142-146. 31. Bauerfeind P, Hof R, Cucala M, Siegrist S, von Ritter C, Fisher JA, Blum AL. Effects of hCGRP I and II on gastric blood flow and acid secretion in anesthetized rabbits. Am J Physiol 1989;256:G145-G149.
Received August 3,199O. Accepted March 20,199l. Address requests for reprints to: Pierangelo Geppetti, M.D., Institute of Internal Medicine IV, Viale Pieraccini 18, 50139 Florence, Italy. Supported in part by the Italian Research Ministry (60% funds). The authors thank M. K. Lokken for revision of the English text,
