Neuropeptides in the human appendix - Springer Link

7 downloads 0 Views 2MB Size Report
However, the coexistence of VIP/PHI and NP Y was unexpected but corroborates previous ..... tants was counted in a LKB 1260 gamma counter (80% counting ...
Digestive Diseases and Sciences, Vol. 34, No. 8 (August 1989), pp. 1217-1230

Neuropeptides in the Human Appendix Distribution and Motor Effects E. EKBLAD, E. ARNBJORNSSON, R. EKMAN, R. H A K A N S O N , and F. S U N D L E R

At present our knowledge of enteric peptide-containing neurons in man is limited. In this study we have used human appendices removed at surgery to examine the peptidergic innervation by immunocytochemistry, immunochemistry, and pharmacological in vitro experiments. Immunocytochemistry revealed a variety o f peptide-containing nerve fiber populations in the human appendix. VIP/PHI-, VIP/PHI/NP Y-, SP/NKA-, galanin-, and enkephalin-containing nerve fibers were numerous; CGRP- and GRP-containing nerve fibers were moderate in number, while only scattered NPY-, enkephalin/BAM-, and somatostatin-containing nerve fibers could be found. No CCK-, dynorphin A-, or dynorphin B-immunoreactive nerve fibers could be detected. The coexistence o f VIP/PHI, SP/NKA, and enkaphalin/BAM can be anticipated from the known sequence o f their respective precursors. However, the coexistence of VIP/PHI and N P Y was unexpected but corroborates previous observations in other species. Interestingly, SP and CGRP did not seem to coexist in nerve fibers of the human appendix. Immunochemistry (RIA and HPLC) confirmed the presence o f ViP , N P Y, SP, galanin, CGRP , GRP, enkephalin, and somatostatin. Motor activity studies suggest that acetylcholine plays a major role in the electrically evoked contractions, since atropine suppressed these contractions. Galanin (10-8-10 -6 M) and GRP ( 1 0 - 9 1 0 -7 M) caused concentration-dependent contractions that were unaffected by tetrodotoxin and thus probably reflect a direct action on smooth muscle receptors. GRP (10 -9 M) enhanced the electrically induced cholinergic contraction (to 193 +- 24%), while met-enkephalin ( 1 0 - 6 M ) reduced it (to 54 +- 6%). Both peptides failed to affect the contractile response to exogenous acetylcholine and probably act to modulate the release o f acetylcholine. NPY, VIP, CGRP, SP, and somatostatin failed to induce contraction or to affect the electrically evoked contractions. KEY WORDS: appendix; neuropeptides.

The mammalian gastrointestinal tract harbors a great number of peptide-containing neurons. Among the major neuropeptides are vasoactive intestinal peptide (VIP), substance P (SP), neuropeptide Y (NPY), gastrin-releasing peptide (GRP), galanin, calcitonin gene-related peptide (CGRP), and enkephalin (for reviews see 1-5). During the last few Manuscript received October 7, 1988; revised manuscript received January 31, 1989; accepted February 9, 1989. From the Departments of Medical Cell Research, Pediatric Surgery, Neurochemistry, and Pharmacology, University of Lund, Lund, Sweden. Address for reprint requests: Dr. Eva Ekblad, Department of Medical Cell Research, University of Lund, Biskopsgatan 5, S-223 62 Lund, Sweden.

years much information has accumulated on the occurrence and distribution of neuropeptides in the intestine of experimental animals. In contrast, our knowledge of the peptidergic innervation of the human intestine in general and of the appendix in particular is still limited. A number of observations suggest a relationship between the activity of the autonomic nervous system and appendicitis (cf. 6, 7); in neurogenic appendicopathy, characterized by spastic contractions, abnormal peristalsis, and hyperemia, a proliferation of nerve fibers storing SP and VIP in the appendix wall has been reported (8-10). The present study provides a detailed description of the peptide-containing innervation of the human

Digestive Diseases and Sciences, Vol. 34, No. 8 (August 1989)

0163-2116/89/0800-1217506.00/0 9 1989PlenumPublishingCorporation

1217

E K B L A D ET A L TABLE 1.

Antigen

Code

Directed against

Raised against

Working dilution immunofluoresec.

VIP PHI

7852 SII

natural porcine VIP pure porcine PHI

N-terminus

1:640 1:200

GRP*

6902

C-terminus

1:640

Substance PI

SP8

C-terminus

1:320

Neurokinin Ar

NKA2

protein-conj, synthetic porcine GRP protein-conj, synthetic bovine substance P synthetic bovine NKA

Met-enkephalinw

Met-enk

protein-conj, synthetic met-enkephalin

C-terminus

BAM-2282

8145

Dynorphin A** Dynorphin Btt

8023 8329

NPY

NPYY2

synthetic BAM-docosapeptide synthetic dynorphin A synthetic rimorphin (dynorphin B) natural porcine NPY

Galanin

8401 8416

Somatostatin

K18

Gastrin 2-17:~

4562

CGRP

8427

synthetic porcine synthetic porcine galanin protein-conj, synthetic ovine somatostatin protein-conj, human gastrin 2-17 synthetic rat CGRP

C-terminus

Ref

1:80

Milab, Malm6, Sweden S. Efendic, Stockholm, Sweden N. Yanaihara, Shizuoka, Japan P.C. Emson, Cambridge, England E. Brodin, Stockholm, Sweden R.J. Miller and K.J. Chang, Burroughs Wellcome, Research Triangle, Park, North Carolina Milab, Malm6, Sweden

1:80 1:80

Milab, Malm6, Sweden Milab, Malm6, Sweden

1:400

17

1:320 1:320

P.C. Emson, MRC, Cambridge, England; Milab, Maim6, Sweden Milab, Maim6, Sweden

1:160

Milab, Maim6, Sweden

16

1:640

J. Rehfeld, Copenhagen, Denmark Milab, Malm6, Sweden

20

1:320

Mid or C-terminus

Source

1:320

1:1280

11 11 12 13 14 15, 16

18 19

21

*Cross-reaction with bombesin. tNo cross-reaction with bombesin and GRP. :~Cross-reactions with neurokinin B, kassinin, and eledoisin. No cross-reaction with SP. w with leu-enkephalin. 82 cross-reaction with met- or leu-enkephalin, dynorphin, 13-endorphin, or rimorphin as tested on pig adrenal medulla. **No cross-reaction with met- or leu-enkephalin, BAM-22, 13-endorphin, or rimorphin as tested on pig adrenal medulla. ttNo cross-reaction with met- or leu-enkephalin, dynorphin, BAM-22, or ~-endorphin as tested on pig adrenal medulla. *$Cross-reaction with cholecystokinin. a p p e n d i x as w e l l as a n a n a l y s i s o f t h e n e u r o p e p t i d e s in e x t r a c t s o f h u m a n a p p e n d i c e s b y h i g h - p e r f o r mance liquid chromatography (HPLC) and radioimm u n o a s s a y ( R I A ) . F u r t h e r , this s t u d y g i v e s s o m e c l u e s a s to t h e p o s s i b l e f u n c t i o n a l r o l e o f t h e p e p t i d e - c o n t a i n i n g n e r v e s b y a n a l y z i n g t h e effects of various neuropeptides on appendiceal longitudinal s m o o t h m u s c l e in vitro. M A T E R I A L S A N D METHODS Immunoeytochemistry. T w e n t y human a p p e n d i c e s were removed at surgery and immediately immersed in cold Krebs solution. Within 1 hr whole wall specimens were d i s s e c t e d out and p r o c e s s e d for i m m u n o c y tochemistry. The specimens were fixed in a mixture of 2% formaldehyde and 15% of a saturated aqueous picric acid solution in 0.1 M phosphate buffer, pH 7.2, and kept in the fixative overnight. They were then rinsed in Tyrode solution, containing 10% sucrose, frozen on dry ice, and

1218

sectioned in a cryostat at 10 txm thickness. The sections were processed for the immunocytochemical demonstration of gastrin-releasing peptide (GRP), peptide with Nterminal histidine and C-terminal isoleucine amide (PHI), vasoactive intestinal peptide (VIP), substance P (SP), neurokinin A (NKA), enkephalin, galanin, calcitonin generelated peptide (CGRP), somatostatin, bovine adrenal medulla docosapeptide (BAM-22P), dynorphin A, dynorphin B (rimorphin), cholecystokinin (CCK), and neuropeptide Y (NPY) using previously characterized rabbit antibodies (Table 1). The site of the antigen-antibody reaction was visualized by the indirect immunofluorescence method of Coons et al (22). Possible coexistence of more than one peptide in the same nerve fiber was tested by doublestaining experiments. The section was first immunostained for one neuropeptide, examined, and photOgraphed. The antibodies were then eluted with acid potassium permanganate for 30 sec (23). The completeness of the elution was tested by application of fluoresceinated anti-rabbit IgG. Only sections devoid of immunofluorescence were processed for the immunocytochemDigestive Diseases and Sciences, Vol. 34, No. 8 (August 1989)

NEUROPEPTIDES IN HUMAN APPENDIX ical demonstration of a second neuropeptide and again photographed. Alternatively, sections were immunostained for the demonstration of two peptides simultaneously. The following peptide antibodies were used in this set of experiments: monoclonal leu-enkephalin antibodies (raised in mouse) (code No. MAS 083c, Sera-lab, Sussex, England) in combination with rabbit antiserum against BAM-22P (proenkephalin A 182-203), dynorphin A (proenkephalin B 209-225) or dynorphin B (proenkephalin B 228-240) (Table 1). An SP antiserum raised in guinea pig (code No. JKI, kind gift from Dr. P.C. Emson, Cambridge, England) was used in combination with rabbit antiserum against NKA or CGRP (Table 1). As secondary antibodies we used fluorescein isothiocyanate (FITC)conjugated antibodies to mouse IgG or guinea pig IgG raised in goat (Sigma, St. Louis, Missouri) and tetramethylrhodamine isothiocyanate (TRITC) -conjugated antibodies to rabbit IgG raised in goat (Milab, Malmr, Sweden). Filters were shifted to enable selective demonstration of each of the two fluorophores. The monoclonal leu-enkephalin antibodies were used in dilution 1:320. These antibodies have previously been characterized by Cuello et al (24) and found to cross-react with metenkephalin but not with 13-endorphin or dynorphin A. Possible cross-reactivity with BAM-22P or dynorphin B was tested by addition of 100 ixg of synthetic peptide per milliliter of diluted antiserum. No such cross-reactivity was found. The guinea pig SP antiserum was used in dilution 1:640; it cross-reacts with physalaemin but not with NKA, kassinin, eledoisin, bombesin, or CGRP (see also 21). For controls we used antisera that had been inactivated by the addition of excess amounts of antigen (10-100 ixg of pure natural or synthetic peptide per milliliter of diluted antiserum). The same amount of peptides was used for tests of cross-reactivity. No crossreactions with the other peptides examined were found. Since cross-reaction with still other peptides or proteins sharing the immunoreactive sequence cannot be excluded, it would be appropriate to refer to the immunoreactive material as GRP-like, VIP-like, and so on. For brevity, however, the shorter terms are used in the following. Immunoehemistry. Whole wall specimens from 52 human appendices were, immediately after removal, frozen to the temperature of - 7 0 ~ C. The specimens were pooled and weighed (32.1 g) and extracted by boiling in 50 ml/0.9% saline for 10 min, followed by homogenization (Polytron 1-2 rain). The homogenates were centrifuged at +4 ~ C. The pellets were subjected to a second extraction, this time in 0.5 M acetic acid for 15 min. The supernatants from the first and second extractions were mixed and lyophilized. After reconstitution in 0.05 M phosphate buffer, pH 7.4, with 0.25% human serum albumin (Hoechst-Behring, Marburg, West Germany) and 20 IE aprotinin/ml (Bayer, Leverkursen, West Germany), they were subjected to high-performance liquid chromatography (HPLC) for characterization of the VIP-, SP-, enkephalin, galanin-, CGRP-, NPY-, and somatostatin-like immunoreactivities. The HPLC system consisted of a Waters model 204 liquid chromatograph equipped with a model U6K injector and an absorption detector 441, a model 6000 A pump, and an automated gradient control-

Digestive Diseases and Sciences, Vol. 34, No. 8 (August 1989)

ler and gradient former pump M-45. A reverse column, ix-Bondapak C18 (Waters) was used. The samples were eluted with acetonitrile (CH3CN)/0.08% trifluoroacetic acid (v/v), pH 2.5, using various linear gradients of CH3CN (10-22.5%) during 60 min followed by isocratic elution at 30% CH3CN (flow rate 1.0 ml/min). Fractions of 0.5 ml were collected, lyophilized, and assayed for VIP-, SP-, enkephalin-, galanin-, CGRP-, NPY-, and somatostatin-like immunoreactivity after reconstitution in 0.05 M phosphate buffer, pH 7.4, containing 0.25% human serum albumin and aprotinin 200 IE/ml. For the radioimmunoassay of the various peptides, the samples were analyzed in serial dilutions (duplicate sampies) and with individual blanks. VIP was determined using a rabbit antiserum (code 7852, Milab). The antiserum was used in dilution 1:60,000 and does not crossreact with PHI, secretin, or glucagon. The detection limit was 5 pmol/liter (25). The interassay variation was 8.5% in the range 10-100 pmol/liter. Standard was synthetic porcine VIP. Immunoreactive SP was determined using a rabbit antiserum raised against synthetic SP (code No. SP2, a kind gift from Dr. E. Brodin, Stockholm, Sweden). The SP RIA has previously been described in detail (14). The detection limit was 10 pmol/liter and the interassay variation was 9% in the range 20-90 pmol/liter. The antiserum was used in dilution 1:50,000 and does not detect any known tachykinin besides SP. In the metenkephalin RIA we used a rabbit antiserum (R-26, a kind gift from Prof. K. Voigt, Freie Universit~it, Berlin) in a final dilution of 1:20,000 (26). The detection limit was 10 pmol/liter. The interassay variation was below 10% in the range 50-250 pmol/liter. The antibody recognizes the C-terminus of met-enkephalin and does not cross-react with leu-enkephalin, 13-endorphin (1-31), 13-endorphin (117), 13-endorphin (1-9), met-enkephalin-Arg6-Phe 7, dynorphin (1-8), BAM 12P, or peptide E (26). For RIA of CGRP we used a rabbit antiserum (code No. A13, Milab) raised against albumin-conjugated synthetic rat CGRP. The antiserum was used in a final dilution of 1:30,000. The detection limit was 20 pmol/liter. There was no crossreaction with human calcitonin, katacalcin, or calcitonin C-terminal adjacent peptide (27). The interassay variation was below 11% in the range 100-300 pmol/liter. Immunoreactive NPY was determined using a rabbit antiserum raised against pure porcine NPY conjugated to bovine serum albumin (a kind gift from Dr. P.C. Emson, Cambridge, England). The antiserum was used in dilution 1: 40,000 and cross-reacts with PYY (peptide with N- and C-terminal tyrosine) to 100% but does not cross-react with PP (pancreatic polypeptide). The detection limit was 25 pmol/liter, and the interassay variation in the range 50100 pmol/liter was below 12%. For further details see reference 28. For RIA of somatostatin we used a rabbit antiserum (code No. KI8, Milab) in a final dilution of 1: 25,000. The antibodies cross-react with cyclic somatostatin (100%), linear somatostatin (50%), (tyrl)-somatostatin (100%), and (tyrl~)-somatostatin (38%). The detection limit was 5 pmol/liter and the interassay variation was 12% in the range 10-80 pmol/liter. For further details see reference 29. Immunoreactive galanin was measured using a specific and sensitive radioimmunoassay not published previously. A rabbit antiserum (code No. 8419,

1219

EKBLAD ET AL Milab) raised against synthetic unconjugated porcine galanin was used in a final dilution of 1:25,000. [125I]Galanin was used as tracer and purified by highperformance liquid chromatography using a C18 column (Waters) with a linear gradient 28-30% acetonitrile and 0.1 M phosphate buffer, pH 3.5. The specific activity was calculated to be 470-510 Ci/g. Antiserum, 200 ~1 diluted 1:10,000, was incubated first with 100 tzl of standard (synthetic galanin, Peninsula, Belmont, California) or extract for 24 hr at 4~ C and with 200 ixl (5000 cpm) of the HPLC-purified tracer for another 24 hr. Bound and free [125I]galanin were separated using electron-coated charcoal (0.5% activated charcoal, 0.05% Dextran T-70 in phosphate buffer 0.05 M, pH 7.5, containing 0.25% human serum albumin). The radioactivity of the supernatants was counted in a LKB 1260 gamma counter (80% counting efficiency for ~25I). The detection limit was 10 pmol/liter (1 fmol/tube) with 95% confidence. The interassay variation was below 10% in the 15-150 pmol/liter range. There was no cross-reactivity of the antibodies with substance P, neurokinin A, or physalaemin up to 1000 pmol/liter. In Vitro Studies of Motor Activity. Motor activity was studied using 35 appendices removed at surgery. After removal, the appendices were immediately placed in cold Krebs solution. Only specimens without overt signs of inflammation were used. Strips (5 x 20 mm) of longitudinal smooth muscle with adherent myenteric plexus were dissected out (within 5-10 hr after collection) and placed in a modified Krebs solution with the following composition (mM): NaC1 133; NaHCO 3 16.3; KC1 4.7; MgCI2 1.0; NaH2PO 4 1.4; CaCI2 2.5, and glucose 7.8. The solution was bubbled with 7% CO2 in 02 giving a pH of 7.2-7.3. The preparations were mounted vertically on a Perspex holder in a 7-ml organ bath thermostated at 37~ C. One end was attached to a rigid support and the free end to a lever connected via a spring to a Grass FT 03 force displacement transducer for isotonic registration of mechanical activity (30, 31). The load on the muscle was set at 0.2 g. Platinum ring electrodes were placed around the muscle with a constant electrode distance of 5 mm. The electrodes were connected to a Grass $4C stimulator for field stimulation with square wave pulses (10-15 V over the electrodes, 1 msec duration, frequency 2-20 Hz, pulse trains lasting for 3 sec). The polarity of the electrodes was shifted after each stimulation period in order to avoid polarization. The mechanical activity was recorded continuously throughout the experiment on a Grass model 7 polygraph. The preparations were allowed to equilibrate in the bath for 30-60 min before experimentation started. The following drugs were used: atropine sulfate from ACO, Stockholm, Sweden; guanethidine from CibaGeigy, Basel, Switzerland; tetrodotoxin and acetylcholine from Sigma, St. Louis, Missouri; synthetic GRP, NPY, SP, NKA, VIP, PHI, CGRP, galanin, BAM-22P, dynorphin A, dynorphin B, and met-enkephalin from Peninsula, Belmont, California. Synthetic galanin 1-10 and a tachykinin antagonist, Spantide, (D-Arg 1, D-Trp7'9, Leu11)-SP, were provided by Ferring Pharmaceuticals, Malm6, Sweden (courtesy of Dr. J. Trojnar). Synthetic cyclic ovine somatostatin was from Ferring and sulfated

1220

CCK-8 (Sincalid) from Squibb, New Brunswick, New Jersey).

RESULTS Peptide-Containing Nerves in the Appendix All the antisera used d e m o n s t r a t e d neuronal elements in the appendix (Figures 1 and 2 and Table 2) with the exception of antisera to gastrin/CCK, dynorphin A, and dynorphin B. T h e s e three latter antisera h a v e previously b e e n found to demonstrate neuronal elements in the rat gastrointestinal tract (16; Ekblad et al, unpublished observations). VIP and P H I were found to coexist in the same nerve fibers in all layers of the appendix wall (Figure 3A and B). T h e population of V I P / P H I - c o n t a i n i n g nerve fibers occurring in the s m o o t h muscle contained, in addition, i m m u n o r e a c t i v e N P Y (VIP/ P H I / N P Y fibers) (Figure 3 C - F ) . In c o n t r a s t , perivascular NPY-containing fibers did not contain V I P / P H I (Figure 3C and D). SP and N K A coexisted in n e r v e fibers in all layers of the appendix (Figure 4A and B). S o m e scattered enkephalin-containing fibers also h a r b o r e d i m m u n o r e a c t i v e BAM-22P (Figure 4C and D). The smooth muscle layer and myenteric ganglia h a r b o r e d n u m e r o u s V I P / P H I / NPY-, S P / N K A - , and enkephalin-containing nerve fibers; a m o d e r a t e n u m b e r of galanin- and CGRPcontaining fibers; and a few storing G R P , enkephaIin/BAM-22P or somatostatin. V I P / P H I / N P Y - , SP/ N K A - , GRP-, CGRP-, and enkephalin-immunoreactive nerve cell bodies were regularly seen in the myenteric ganglia. N e r v e fibers storing V I P / P H I / N P Y , enkephalin, enkephalin/BAM-22P, or somatostatin were present in the muscle coat exclusively. N e r v e fibers containing V I P / P H I or S P / N K A were n u m e r o u s in the m u c o s a and s u b m u c o s a . Scattered CGRP-, GRP-, N P Y - , and galanin-containing fibers could also be detected in this location. The submucous ganglia contained V I P / P H I immunoreactive nerve cell bodies. V I P / P H I - , C G R P - , NPY-, SP/ N K A - , and galanin-containing fibers w e r e seen also around blood vessels, particularly in the submucosa. C G R P did not s e e m to coexist with S P / N K A in perivascular fibers or in neurons or fibers in any layer of the appendix wall. The findings are summarized in Figure 1 and Table 2.

Distribution of Individual Nerve Fiber Populations VIP/PI-II/NPY. The smooth muscle layers and the myenteric ganglia received a dense innervation of coarse and strongly i m m u n o r e a c t i v e V I P / P H I / N P Y Digestive Diseases and Sciences, Vol. 34, No. 8 (August 1989)

NEUROPEPTIDES IN HUMAN APPENDIX

SG cm

MG Im

A

B

C

D

E

~

~ ,~"

~

~ ~

~

~

~

I

J

"~

F

G

H

SG be"

cm

MG Im

Fig. 1. S c h e m a t i c outline of the t o p o g r a p h y and relative de ns i t y of p e p t i d e - c o n t a i n i n g n e r v e fibers (o-e,) a nd n e r v e cell bodi es

(9 in the human appendix. (A) VIP/PHI/NPY-containing fibers, (B) VIP/PHI-containing fibers, (D) SP/NKA-containing fibers, (C) NPY-containing fibers, (D) SP/NKA-containingfibers, (E) galanin-containing fibers, (F) CGRP-containingfibers, (G) GRP-containing fibers, (H) enkephalin-containing fibers, (I) enkephalin/BAM-22P-containingfibers, (J) somatostatincontaining fibers. SG, submucous ganglia; cm, circular muscle; MG, myenteric ganglia; lm, longitudinal muscle. fibers (Figures 1A and 2). Moderately immunoreactive nerve cell bodies were regularly seen in the myenteric ganglia (Figure 2). VIP/PHI. The mucosa and submucosa harbored numerous fine varicose, strongly immunoreactive VIP/PHI fibers. These fibers were devoid of NPY. Many fibers reached high up between the crypts (Figures 1B and 3A). Blood vessels received a moderate supply of such fibers. N P u Under this title are described those NPYimmunoreactive fibers which did not contain VIP/ PHI. F r o m previous studies it is quite likely that these fibers are adrenergic (11, 16, 17). A few, weakly immunoreactive NPY fibers were found in the basal portion of the mucosa and in the submucosa (Figure 1C). Perivascular NPY-immunoreactive fibers were numerous (Figure 3E). SP/NKA. Numerous fine varicose fibers displaying intense S P / N K A immunoreactivity occurred in the mucosa and submucosa (Figures 1D and 2B). Such fibers were often seen to run between the crypts all the way up to the surface epithelium. SP/ NKA-containing fibers were numerous also in the smooth muscle and myenteric ganglia (Figures 2C and 4). H e r e the fibers were generally more coarse than in the mucosa. Moderately immunoreactive Digestive Diseases and Sciences, Vol. 34, No. 8 (August 1989)

nerve cell bodies were occasionally seen in the myenteric ganglia. SP/NKA-immunoreactive nerve fibers were regularly seen around blood vessels. Galanin. A few galanin-immunoreactive fibers occurred in the m u c o s a and submucosa. T h e y predominated basally in the m u c o s a but some fibers extended high up between the crypts. The smooth muscle layers and myenteric ganglia harbored a moderate supply of galanin-immunoreactive fibers (Figure 1E and 2D). These fibers were generally more coarse and more intensely immunoreactive than those in the mucosa. Moderately immunoreactive nerve cell bodies were regularly seen in the myenteric ganglia (Figure 2D). Scattered galaninimmunoreactive fibers were observed around blood vessels. CGRP. A few moderately immunoreactive CGRP fibers occurred in the submucosa but none could be detected in the mucosa. The myenteric ganglia harbored numerous intensely immunoreactive fibers, but v e r y few CGRP-immunoreactive fibers could be detected in the smooth muscle, A few scattered, weakly immunoreactive nerve cell bodies were seen in the myenteric ganglia (Figure 1F). Around blood vessels, CGRP-immunoreactive fibers were few in number.

1221

EKBLAD ET AL

Fig. 2. (A) VIP- and (B) SP- immunoreactive nerve fibers are numerous in the mucosa where they reach high up between the crypts. In the smooth muscle and myenteric ganglia numerous (C) SP- and (D) galanin- immunoreactive nerve fibers as well as nerve cell bodies can be seen. A and B, • C and D, • 175.

GRP. A moderate number of weakly fluorescent, fine varicose GRP-containing fibers were seen in the mucosa and submucosa. Scattered, moderately immunoreactive GRP fibers were observed in the smooth muscle and myenteric ganglia. A few GRP immunoreactive nerve cell bodies occurred in the myenteric ganglia (Figure 1G). No GRP-containing fibers could be detected around blood vessels. Enkephalin. Numerous intensely immunoreactive enkephalin fibers could be detected in the smooth 1222

muscle and myenteric ganglia (Figures 1H and 5A and B). Immunoreactive nerve cell bodies were regularly seen in the myenteric ganglia. No enkephalin-immunoreactive fibers could be detected in the mucosa or submucosa or around blood vessels. Enkephalin/BAM-22P. This heading deals with those neuronal elements that contain both enkephalin and BAM-22P, ie, the entire BAM 22Pimmunoreactive nerve fiber population and a minor population of enkephalin-containing nerve fibers. Digestive Diseases and Sciences, Vol. 34, No: 8 (August 1989)

NEUROPEPTIDES IN HUMAN APPENDIX TABLE 2. TOPOGRAPHICALDISTRIBUTION AND RELATIVE FREQUENCY OF PEP'rIDE-CONTAINING NERVE FIBERS IN HUMAN APPENDIX*

Immunoreactive peptide

Mucosa and subrnucosa

Smooth muscle

Myenteric ganglia

VIP/PHI/NPY VIP/PHI NPY SP/NKA Galanin CGRP GRP Enkephalin Enkephalin/BAM-22P Somatostatin CCK Dynorphin A Dynorphin B

0 +++ + +++ + + + 0 0 0 0 0 0

+++ 0 0 ++ ++ + + +++ + + 0 0 0

+++ 0 0 +++ ++ ++ + +++ + + 0 0 0

* + + + , n u m e r o u s ; + + , m o d e r a t e n u m b e r of fibers; + , few fibers; 0, no fibers detected. A s s e s s m e n t s were m a d e by t w o independent o b s e r v e r s and based on examination o f 5-10 s p e c i m e n s f r o m eight appendices (at least two sections from each specimen).

Scattered nerve fibers displaying enkephalin/BAM22P immunoreactivity could be detected in the smooth muscle and myenteric ganglia (Figures 1I and 4C and D). Immunoreactive nerve cell bodies were observed occasionally in the myenteric ganglia. No fibers could be detected in the mucosa, submucosa, or around blood vessels. Somatostatin. A few fine varicose, moderately immunoreactive fibers occurred in the smooth muscle layer and myenteric ganglia; they were not found in the mucosa, submucosa or around blood vessels (Figure 1J). No immunoreactive nerve cell bodies could be detected in the intramural ganglia.

Immunochemistry The concentrations of VIP, NPY, SP, galanin, CGRP, GRP, met-enkephalin, and somatostatin in human appendix (whole wall specimens) are shown in Table 3. HPLC of extracts revealed four major VIP-immunoreactive peaks, one with the same elution position as synthetic porcine VIP; two major NPY-immunoreactive peaks, one with the same elution position as synthetic porcine NPY~ and one major SP immunoreactive peak coeluting with synthetic SP. The met-enkephalin-immunoreactive material eluted in two major peaks, one with the same elution position as synthetic met-enkephalin. One major somatostatin-immunoreactive peak with the same elution position as synthetic ovine somatostatin 15-28 was found (Figure 5). It was not possible to obtain an acceptable resolution of the galaninDigestive Diseases and Sciences, Vot. 34, No. 8 (August 1989)

and CGRP-immunoreactive material that eluted in several poorly resolved peaks.

Motor Activity of Isolated Longitudinal Smooth Muscle No spontaneous motor activity could be recorded. Electrical stimulation evoked frequencydependent contractions. Atropine (10 -6 M) reduced the electrically induced contractile response to 18 -+ 4% (N = 8) (Figure 6). While tetrodotoxin (TTX) ( 1 0 - 6 M), a blocker of nervous conduction, abolished the electrically induced contractile response (N = 15) (Figure 6), neither guanethidine (5 x 10 - 6 M) (Figure 6) nor Spantide (10 -5 M), a tachykinin antagonist, had any effect (N = 6). NPY, VIP, CGRP, somatostatin, CCK-8, and SP (10-9-10 -6 M) failed to induce contraction, and none of them affected the electrically evoked contraction (N = 8). Galanin (10-8-10 -6 M) and GRP (10-9--10 -7 M) produced dose-dependent contractions (Figure 7) that were unaffected by TTX (10 -6 M), atropine (10 -6 M), and Spantide (10 -5 M) (N = 8-16). At 10 -9 M, GRP enhanced the electrically evoked response to 193 --+ 24% (N = 12) (Figure 8). Metenkephalin (10 -6 M) had no effect on the basal tension; however, it reduced the electrically evoked contraction to 54 --- 6% (N = 6) (Figure 8). Both GRP (10 -9 M) and met-enkephalin (10 -6 M) failed to affect the contractile response to exogenous acetylcholine (10-8-3 x 10-5 M) (Figure 9). DISCUSSION The enteric nervous system comprises several populations of peptide-containing neurons. Neuropeptides demonstrated in the human gut are SP, VIP (10, 32-39), enkephalin (32-35, 37-39), somatostatin (38-40), GRP (34), NPY (38, 40, 41), galanin, and CGRP (41). Except for the study of Hrfler et al (10), none of these reports included the appendix. Intestinal SP-, VIP-, somatostatin-, galanin-, and NPY-containing nerve fibers occur in the mucosa, smooth muscle, and the myenteric ganglia, while enkephalin-, GRP-, and CGRP-containing fibers occur mainly in the smooth muscle and myenteric ganglia. On the whole, the topography and distribution of the various peptide-containing neuronal populations found in the appendix agree with the innervation pattern previously described for other regions of the human intestine (10, 32-41). However, the density of peptide-containing nerve fibers appears to be higher in the wall of the appendix than 1223

EKBLAD ET AL

Fig. 3. Coexistence of(A) PHI and (B) VIP and of(C) NPY and (D) VIP in the same nerve fibers and nerve cell bodies in the smooth muscle and myenteric ganglia of human appendix. (E) NPY does not coexist with (F) VIP in perivascular nerve fibers as demonstrated in submucosal blood vessels (v) in the human appendix. The sections were first immunostained with the PHI or NPY antibodies and after elution of the antibodies immunostained for VIP. A-D, x 180; E and F, x 150.

in other parts of the human intestines (unpublished observation). It is worth noting that the NPYcontaining myenteric nerve fibers have been re-

1224

ported to be fewer than the VIP-containing ones in the human small and large intestine (39). However, in the appendix the NPY- and VIP-containing nerve Digestive Diseases and Sciences, Vol. 34, No. 8 (August 1989)

NEUROPEPTIDES

IN HUMAN APPENDIX

Fig, 4. Coexistence of (A) NKA and (B) SP in nerve fibers in both smooth muscle and myenteric ganglia. The section was double stained with NKA antibodies (TRITC) and SP antibodies (FITC). Coexistence of (C) enkephalin and of (D) BAM-22P in the same nerve fibers in smooth muscle of human appendix. Some scattered enkephalin nerve fibers (arrow head) contain in addition BAM-22P. The sections were double stained with enkephalin antibodies (FITC) and with BAM-22P antibodies (TRITC). A and B, x240; C and D, • 175.

fibers are equally numerous; in fact, the two peptides are colocalized in the same myenteric neurons. Further, the human appendix seems to lack enkephalin-containing nerve fibers in the mucosa/ submucosa in contrast to the small and large intestine, which contain scattered met-enkephalin immunoreactive fibers in this location (33). Digestive Diseases and Sciences, Vol. 34, No. 8 (August 1989)

The results of the present study extend previous observations on the distribution of peptidecontaining nerves in the human intestine, particularly with respect to the coexistence of peptides. VIP/PHI-, VIP/PHI/NPY-, SP/NKA-, galanin-, and enkephalin-immunoreactive nerve fibers were numerous in the appendix. The coexistence of VIP/

1225

r~'~

W~

~

~

~

~

~

z

~-

E 5

c

\\

u~ o

nn~

\\

m-

\

~

~

!1

\

~

~

i

I

~

~

z

I

m

2

m-

~g

=~

~m-

9

o

o

\ \

//

S

\

\

\

\\

~

\ L

~

~

z

i

c~

m

c

9-~

B

nn~

~o

~-

g

nn--

o~_

~-

.

f

i

i

~

\

i

oi

\ \

\\

\

\

\\

\

\

I

I I

?

0.~

i