UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Morphological and functional studies of substance P and somatostatin in the small intestine Accili, Eric Anthony 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1992_fall_accili_eric_anthony.pdf [ 5.58MB ]
JSON: 831-1.0086500.json
JSON-LD: 831-1.0086500-ld.json
RDF/XML (Pretty): 831-1.0086500-rdf.xml
RDF/JSON: 831-1.0086500-rdf.json
Turtle: 831-1.0086500-turtle.txt
N-Triples: 831-1.0086500-rdf-ntriples.txt
Original Record: 831-1.0086500-source.json
Full Text

Full Text

MORPHOLOGICAL AND FUNCTIONAL STUDIES OF SUBSTANCE P ANDSOMATOSTATIN IN THE SMALL INTESTINEByEric Anthony Accili, B.Sc. (1982), M.Sc. (1987)The University of British ColumbiaA thesis submitted in partial fulfillment of the requirements for the degree ofDoctor of PhilosophyinThe Faculty of Graduate StudiesDepartment of PhysiologyWe accept this thesis as conforming to the required standardThe University of British ColumbiaJune 1992© Eric Anthony Accili, 1992Signature(s) removed to protect privacyIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or. her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)_______________________Department of /4’’-,yThe University of British ColumbiaVancouver, CanadaDate /977DE-6 (2/88)Signature(s) removed to protect privacyiiABSTRACTThe regulation of gastrointestinal function is partlydependent on the intrinsic neurones of the gut. Intrinsicneurones contain a large number of neurotransmitters,including neuropeptides, in different combinations.Neurones have been grouped according to the combination ofneurotransmitters they contain and this practice is calledchemical coding. The present studies were carried out toexamine differences in chemical coding and enteric neuronalmorphology in the human and canine small intestine,especially the submucosal plexus. The neuropeptides chosenfor examination were substance P, somatostatin andvasoactive intestinal peptide because of their knowninvolvement in both the physiology and pathophysiology ofthe small intestine. Further, primary cultures ofsubmucosal neurones from human and canine submucosal plexuswere utilized to determine whether differences in coding andmorphology paralleled differences in somatostatin secretion.Substance P inununoreactivity (SP—IR), somatostatinimmunoreactivity (SS-IR) and vasoactive intestinal peptideimmunoreactivity (VIP-1R) have been localized and theirdistributions have been compared in dog and human smallintestine using ilmrLunocytochemistry (ICC). An antibody toprotein gene product 9.5 (PGP) was used to localize allnerve cell bodies and fibres in the dog and human upperiiijejunum. In addition, the proportions of peptide-containingneurones were determined by double staining.Staining with PGP revealed neuronal cell bodies in thesubmucosal plexus (SMP) and the myenteric plexus (MYP) aswell as extensive innervation by fibres throughout allregions of the small intestine. Canine submucosal gangliacontained 7.7 ± 0.6 neurones per submucosal ganglion (184ganglia, n = 6 dogs), while the human ganglia contained 2.9± 0.3 (185 ganglia counted, n = 5 donors). Over 50 % of theganglia in the human sections contained 3 or less neuronesbut over 10 % of the ganglia in canine sections contained 15or more neurones. Finally, the canine circular muscle wasshown to possess a distinct deep neural plexus, in contrastto that of human circular muscle.The distribution of SPIR, SSIR and VIP’IR in canineand human jejunum was similar, confirming the results ofprevious studies. Double staining revealed that SP-IR andSS-IR were always co-localized in the human, but not canine,SMP and MYP. In both species VIPIR was present in apopulation of neurones distinct from those containing SP—IRand SS—IR. In canine ganglia, 30% of neurones per ganglionin the SMP contained SP-IR, 35% contained SSIR and 30%contained VIP-IR. In human ganglia, 42% of neurones perganglion contained SP-IR and SSIR while 40% contained VIPivIR. These results sugest different functions for SP and SSin canine and human enteric ganglia.The secretion of SSIR, from intact preparations ofsmall intestine, is difficult to interpret for two reasons0First, SS—IR has been demonstrated in vagal, submucosal andmyenteric neurones as well as endocrine cells, of the smallintestine. Second, enteric neurones have been shown to tocontain the 14 amino acid form of SS-IR (SS-14) whileendocrine cells have been shown to contain the 28 amino acidform (SS28). A dispersed culture of submucosal cells fromhuman small intestine was developed to examine thelocalization, release and molecular characteristics of SSIR. After 72 h, the neurones were shown to be viable and tosprout neurites containing varicosities suggesting that theyretained a morphologic phenotype similar to that observed insitu. Thirty percent of the submucosal neurones perganglion in tissue sections and 35 % of cells per cluster inculture contained SSoIR. Acetic acid extracts of culturescontained 1990 ± 809 pg SS-IR/well. Incubation of cultureswith phorbol 1213-myristate 13-acetate (13-PMA), an activatorof protein kinase C (PKC), at concentrations up to Mfor 120 mm increased the release of SSIR up to 23 timesthe basal level, and up to 27 times the basal level whenextracellular K+ was increased from 5 to 10 mM. Of thetotal SSIR released in response to 13-PMA (106 M, 10 mMK+), 59% was present in the medium after 30 iuin and 80%Vafter 60 mm. Basal release of SS—IR could be reliablymeasured only after 120 mm, therefore experiments whichexamined somatostatin secretion were carried out for thisamount of time. The release of SS-IR by 13-PMà was not dueto non—specific membrane effects since the inactive 4aphorbol at the same concentrations did not alter basalrelease. Greater than 90 % of SS—IR present in acidextracts of cultures and released by I3PMà eluted with thesame retention time as synthetic SS’14 on reverse phase highperformance liquid chromatography (HPLC)In summary, the results presented in this thesis haveshown that differences exist in the neuropeptidedistribution and neuronal.morphology between the canine andhuman small intestine. Moreover, SS—IR secretion from humansubmucosal neurones in response to SP and the phorbol esterwere found to be different from the secretion of SS—IR fromcanine neurones. The results suggest that differencesobserved in the pattern of secretion in submucosal neuronesprobably reflect the differences noted in neuropeptidedistribution and neuronal morphology. Furthermore, thepresent studies emphasize that the extrapolation ofexperimental data between species must be made with caution.viTABLE OF CONTENTSPageABSTRACT iiLIST OF TABLES xiLIST OF FIGURES xiiACKNOWLEDGEMENTS xviI. INTRODUCTION 1A. General Background 1B. Background on the Small Intestine 4C. Experimental Rationale 16D. Hypotheses 18E. Specific Objectives 19II. IMMUNOCYTOCHEMICAL METHODS 21A. Tissue Sections 211. Tissue Preparation 212. Protocol 21a. Primary antibodies 22b0 Secondary Layers 22i. Peroxidase 22ii. Inuuunofluorescence 23c. Double Stains 25d. Controls 25B. Tissue Culture 25C. Quantification of Peptide Containing Neurones 26viiIII. CHAPTER ONE. IMMUNOCYTOCHEMICAL STUDIES 29A. Introduction 29B. Results 331. Human tissue sections 33a. Protein gene product 9.5 33b. Autofluorescence 40c. Single Stains of SS—IR, SP—IR and VIP—IR. 40d. Double Stains 462. Canine Tissue Sections 58a. Protein Gene Product 9.5 58b. Single Stains of SS-1R9 SPIR and VIP—IR 67c. Double Stains 68C. Discussion 79D. Summary 86IV. CHAPTER TWO. ISOLATION OF CANINE AND HUMANSUBMUCOSAL NEURONES 87A. Introduction 87B. Methods 941. Human Donor Experiments. 94a. Procurement of Tissueb. Isolation of Submucosal Ganglia 94c. Elutriation Centrifugation 95d. Tissue Culture 96so Somatostatin Secretion 99i. General Protocol 99ii. High Potassium 100iii. Time-course 100viiif. Somatostatin Extraction 100g. Somatostatin Recovery 101h. Characterization of Primary Molecular Forms 101i, Sample Preparation 101ii. Reverse Phase HPLC 1012. Dog Experiments 102a. Procurement of Tissue 102b. Isolation of Submucosal Ganglia 102c. Elutriation Centrifugation 103d. Tissue Culture 103e. SSIR and SP—IR Secretion 104i. General Protocol 104ii. Acetic acid Extraction 104iii. Sep—Pak Extraction and Concentration 104iv. SP-IR Recovery 10530 Radioimmunoassay 105a. Soiuatostatini. Assay Buffer 106ii. Antibody 106iii. Standards 106iv. Preparation of 125-Soiuatostatin 106v. Separation 107vi. Assay Protocol 108vii. Calculation and Presentation of RIA Data 110viii. Inter- and Intra—assay Variation 110ix. Testing pH effects 111b. Substance P 116ixi. Assay Buffer 116ii. Antiserum 116iii. Assay Protocol 116iv. Calculations 1174. Data Analysis 117C. Results1. Isolation and Characterization of SubmucosalCultures 119a. Human 119Dog 1212. Somatostatin Secretion from HumanSubmucosal Neurones 144a. Effects of Secretagogues 144b. Somatostatin Content 165c, Somatostatin Recovery 165d. Characterization of Molecular Forms 1653. Somatostatin and Substance P Secretion fromCanine Submucosal Neurones 173D0 Discussion 174E. Summary 188V. GENERAL SUMMARY AND CONCLUSIONS 191A. Morphological Data from the Small Intestine 191B. Culture Studies 193VI, CONCLUSIONS AND FUTURE DIRECTIONS 197A. Conclusions and Significance 197B. Future Directions 1991. Morphological Studies 199x2. Functional Studies 199VII. REFERENCES 203APPENDIX I 233APPENDIX II 234xiLIST OF TABLESNumber Title Page10 Primary Antibodies 242. Biotinylated Antibodies 273. Fluorophore-conjugated Antibodies 274. Avidin Layers 285. Summary of Neuropeptide Distribution andCo—localization in the Human Small Intestine 766. Summary of Neuropeptide Distribution andCo—localization in the canine small intestine 777. Comparison of Quantification Data Between theCanine and Human SMP 788. Somatostatin Assay Protocol 1099. The Effect of the Calcium lonophoreon SS—IR Secretion 16410. Variations in Content of SoluatostatinImmunoreactivity 16811. Similarities and Differences inthe Secretion of Somatostatin Immunoreactivity 181from Canine and Human Submucosal NeuronesxiiLIST OF FIGURESNumber Page1. Diagram showing segments of enteric plexuses 72. PGP 9.5 staining of human mucosa 343. PGP 9.5 staining of human muscularis 364. VIP-IR and PGP 9.5 double stain 385. The relative size distribution of humansubmucosal ganglia of the small intestine 416. A section of human duodenum showing associationof autofluorescence and SP’IR 437. Human intestine representative inununostains 478. Human duodenal sections double stainedfor SS-IR and SP-IR 529. Human duodenum SMP stainedfor SS’IR and VIP-IR 5410, Human duodenum double stainedfor VIP-IR and SP-IR 5611. PGP 9.5 staining of canine duodenum 5912. Details of the PGP 9,5 staining inthe canine duodenal SN? 6113. Details of the PGP 9.5 staining in thecanine duodenum showing the deep muscular plexus 6314. The relative size distribution of caninesubmucosal ganglia of the small intestine 6515, Representative immunostains of the canine SMP 70xiii16. Double stains of canine submucosal gangliashowing SS-IR and SP-IR 7217. Double stains of canine submucosal gangliashowing SS-IR, SP—IR and VIP-IR 7418. Diagram of elutriator and chamber 9719. Somatostatin standard curves(% bound vs concentration of somatostatin) 11220. Somatostatin standard curves(logit % bound vs concentration of somatostatin) 11421. High magnification micrograph of humanduodenal neurones stained for SS-IR 12222. Cultures of human duodenal SMPstained for SS-IR 12423. Cultures of human duodenal SM?stained SP-IR and VIP’IR 12624. Cultures of human duodenal MY?stained SS—IR 12825 A time course of the attachment andprogression in short term cultures ofcanine SMP neurones 13126. Phase contrast micrograph of caninesubmucosal neurones 13527. Cultures of canine SM? stainedfor SS—IR, SP—IR and VIP—IR 13728. Cultures of canine SM? neüronesstained for VIP-IR 140xiv29. Human and canine SMP culturesdouble stained for SP-IR and SS-IR 14230. Release of soiuatostatin immunoreactivity(SS—IR) from submucosal neurones in response toincubation with B-PMA 14531. Release of somatostatin immunoreactivity(SS—IR) from submucosal neurones in response toincubation with 13-PMA in 5 mM and 10 mM KC1 14732. Release of SS—IR from submucosal neuronesin response to 13’-PMA (106 M, 10 mM KC1)after 30, 60 and 120 mm 14933. Release of SS—IR (% TCC) from submucosalneurones in response to incubation with 13—PMAand the inactive 4a-phorbo1 for 120 mm 15234. Release of SS-IR (pg/dish) from submucosalneurones in response to incubation withsubstance P for 120 mm 15435 Release of SSIR (pg/dish) from submucosalneurones in response to incubation withsubstance P for 120 mm in the presence orabsence of hexamethonium (hex) and atropine (atr) 15636. Release of SS-IR (pg/dish) from submucosalneurones in response to incubation withi3PMA for 120 mm in the presence or absenceof substance P 158xv37. Release of SS—IR (pg/dish) from submucosalneurones in response to incubation with 13—PMAfor 120 mm in the presence or absence of TTX 16038. Release of SS—IR (pg/dish) from submucosalneurones in response to incubation with CGRPand methacholine for 120 mm 16239. Release of SS—IR (pg/dish) from submucosalneurones in response to incubation 13PMAand carbachol or 120 mm 16640. HPLC profile of SS-IR released from submucosalneurones in response to BPMA 16941. HPLC profile of SS—IR contained in acetic acidextracts in response to 13PM 171xviACKNOWLEDGEMENTSI extend my thanks to Dr. Alison Buchan for her time andcominittment to my work, and for her perseverance during thecompletion of this thesis. The assistance of Dr. Chris McIntoshin providing training and expertise for the HPLC, and impartingcurrent information on everything known to man, was muchappreciated. Advice on the design of pharmacological studies,especially the importance of industrial doses of skepticism inscience, was graciously offered by Dr. Yin Nam Kwok andgraciously accepted by myself. As the graduate advisor, Dr. RayPederson has always put the concerns of students first and Ithank him for his enthusiastic interest in both my studies and myhobbies. I must thank Dr. John Brown because he would feel leftout otherwise and more importantly because he was responsible forcultivating my interest in regulatory peptides and forencouraging me to pursue this interest within the MRC group.Finally, I am especially grateful to my friends and colleagueswho have helped me with my curricular and extracurricularactivities.1I. INTRODUCTION.A. General Background.Neuropeptides have been recognized as potentialneurotransmitters since the discovery of the hypothalamichormones. These include vasopressin and oxytocin frommagnocellular neurones, sequenced by du Vigneaud (1953,1954), as well as luteinizing hormone releasing hormone,thyrotropin releasing hormone and somatostatin (SS) fromparvicellular neurones (Guillemin, 1978; Brazeau et al,1973; Schally et al, 1973)Information on their function is difficult to interpret,in part, because they are members of a larger group ofregulatory peptides that also have hormonal and paracrineactions. Peptides have been demonstrated to exist inneurones and endocrine cells, which suggests that they mayhave endocrine, paracrine, neuroendocrine andneurotransmitter functions (Brown et al, 1971; Feyrter,1953; Chang et al, 1971; Gullemin, 1978) For example, SShas been found in hypothalamic neurones and acts in aneuroendocrine manner to inhibit growth hormone release(Brazeau et al, 1973). Somatostatin has been localized tocerebral cortical neurones (Johansson and Ho]cfelt, 1980;Krisch, 1980) where it has been implicated as aneurotransmitter (Guillemin, 1978). Somatostatin has been2found in endocrine cells of the stomach, pancreas and smallintestine, indicating a possible role as a hormone (Luft etal, 1974; Dubois, 1975; Polak et al, 1975). Finally,paracrine actions have been proposed for SS, in addition toits endocrine actions, based on its presence in endocrinecells particularly those which have long cytoplasmicprocesses (Larsson, 1979; Yamada, 1987). Other examples ofpeptides with multicellular origins include substance P,cholecystokinin and neurotensin.Unlike classical neurotransmitters, the originalconsideration of neuropeptides as potentialneurotransmitters was the result of their localization inneural tissue by immunocytochemistry or radioimmunoassay(Hokfelt, 1991) and in most cases this evidence remains themost compelling. The function of the putative neuropeptideneurotransmitters has been difficult to elucidate. Thepeptides must fulfil several criteria in order to beconsidered as neurotransmitters including the following(Dockray, 1987):a. that the peptide is found in neurones.b. that it is concentrated in nerve terminals from which itcan be released by depolarizing stimuli with or without acalcium-dependent mechanism.c. application of the purified peptide exerts an effectthat is duplicated when the endogenous material is released.3d. a selective antagonist is able to block the actions ofboth endogenously released and exogenously applied peptide.e. there are mechanisms for the breakdown, reuptake or theremoval of the peptide.There are difficulties in the identification ofneuropeptides as neurotransmitters when compared toclassical neurotransmitters such as acetyicholine andnoradrenaline. Unlike classical neurotransmitters which aresynthesized exclusively in nerve terminals, neuropeptidesmust be synthesized in cell bodies and transported to thenerve terminal where they are stored prior to release.There is presently no evidence for the uptake and reuse ofneuropeptides by neurones.The cell types which contain peptides belong to a familyof cells with a common amine handling system present andhence were called APUD (Amine Precursor, Uptake andDecarboxylation) cells (Pearse, 1976). These cells includeneuropeptide containing endocrine cells which share commoncharacteristics with neuropeptide producing neurones.Peptides have been shown to be members of families codedby the same or similar genes, but become processed indifferent ways in different cell types. The processing canoccur at the level of mRNA, for example, by splicing such asin the case of the neuropeptide calcitonin gene-related4peptide (CGRP) and the hormone calcitonin. Peptides mayundergo post—translational modifications such as enzymaticcleavage and C—terminal amidation, as in the case of theneuropeptide CCK-8 and the hormone CCK-33.A further complication in the study of neuropeptides isthat more than one neurotransmitter can be present in oneneurone (Milihorn and Hokfelt, 1988) The transmittersubstances may be neuropeptides, amino acids or a classicaltransmitter (noradrenalin or acetylcholine) which wouldallow the release of three types of neurotransmitters toconvey fast, moderate or slow signalling from a singleterminal (Iversen and Goodman, 1986).B. Background on the Small Intestine.The small intestine is innervated by both extrinsic andintrinsic nerves. The vagus and sacral nerves as well asthe sympathetic ganglia supply the extrinsic parasympatheticand sympathetic innervation, respectively. However, thesmall intestine has an extensive intrinsic nervous componentwhich, together with the processes of the sympathetic,parasympathetic and sensory neurones, was named the entericnervous system. Langley (1921) considered this to be aportion of the autonomic nervous system separate from thesympathetic and parasympathetic branches.5Enteric neuronal axons are unmyelinated and have beenshown to have varicosities along their length, each of whichrepresents a nerve ending from which a neurotransmitter canbe released (Gabella, 1987).The enteric nervous system is made up of twoganglionated plexuses called, respectively, the submucosaland myenteric. The submucosal plexus (SMP), described byMeissner (1857), is located in the connective tissue of thesubmucosa while the myenteric plexus (MYP), lies between thelongitudinal and circular muscle of the muscularis externa(figure 1).The ganglia of the SMP have been shown to be smallerthan those of the MYP, and it has been suggested that theycontain fewer neurones (Furness and Costa, 1980; Gabella,1987; Gabella, 1990). The MYP has been shown to bestructurally uniform along the length and circumference ofthe small intestine, and the myenteric ganglia were shown tohave their long axis in the direction of the circular muscle(Gabella, 1987). The submucosal ganglia were shown to formsmaller meshes, they did not demonstrate polarization withrespect to the axes of the small intestine and they showedvariation with respect to cell number along the length ofthe small intestine.6The enteric ganglia also show substantial variationbetween species (Gabella, 1987; 1990). The ganglia of theguinea pig small intestine have- been shown to vary in size,but contain similar types of nerve cells and fibres (Wilsonet al, 1981; Furness et al, 1984; Bornstein and Furness,1988)7Figure 1. Diagram of a segment of intestine partlyseparated into layers showing the arrangement of entericplexuses. (1) MYP (2) longitudinal muscle (3) tertiaryplexus (4) circular muscle (5) deep muscular plexus (6)submucosal plexus (7) submucosal artery (8) mucosalplexus (9) muscularis mucosa (10) mucosa. (Adapted fromCosta et al, 1987).09In many species including the dog and human, the gangliaof the submucosa have been shown to form outer and innerplexuses (Schabadasch, 1930), the latter which would becomeknown as Schabadasch’s plexus and the former which would beregarded as conforming to Meissner’s original description ofa plexus in the submucosa (Furness et al, 1989; Gunn, 1968;Scheuerman et al, 1989; Christensen and Rick, 1987; Stach,1977). Submucosal ganglia of the outer plexus were shownto be more similar to the MYP, and the shapes, histochemicalstaining and patterns of innervation differ from the innerplexus (Gunn, 1968; Scheuerman et al, 1989; Stach, 1977).The outer submucosal plexus does not exist in allspecies. For example, in the guinea pig, submucosal nervecell bodies do not project to the circular muscle butexclusively to the mucosa where they have been shown toinfluence water and electrolyte absorption and secretion(Bornstein and Furness, 1988). The outer SMP is mostevident in the colon of larger species such as the human,pig and sheep (Gunn, 1968; Stach, 1977; Crowe et al, 1992),whereas this plexus has not been found in the smallintestine (Christensen and Rick, 1987), with the exceptionof the dog (Furness et al, 1989).Interestingly, the interstitial cells of Cajal havealso been found in the outer plexus of the mammalian10submucosa (Christensen and Rick, 1987). These cells werefirst described by Cajal (1893) who suggested they may bemodified neurones, and have since been found in the musclelayers of the small intestine of several mammals (Gabella,1987; Rumesson and Thuneberg, 1991) including the human.Their function has been the subject of considerable debatesince they share characteristics of both neurones and nonneural cells such as fibroblasts (Rogers and Burnstock,1966; Thuneberg, 1982). These cells have been implicated ingenerating the rhythmic activity of smooth muscle(Thuneberg, 1982; Huizinga, 1991). This has been supportedby experiments in the dog which demonstrated thatinterstital cells at the muscle/submucosal interface possesselectrical pacemaker activity (Barajas-Lopez et al, 1989).The submucosal plexus has been suggested to be involvedprimarily in the control of mucosal secretion and absorptionof electrolytes (Hubel, 1985; Keast et al, 1987; Cooke,1988). The SMP is tonically active and suppresses iontransport (Cooke, 1988). The SMP of the small intestine andcolon has been thought to help regulate motility byconveying sensory information to the MYP (Bulbring et al,1958; Crema, 1970; Kottegoda, 1970). Its innervation of theinner circular muscle in larger animals such as the dog(Furness et al, 1989) would suggest direct involvement ofthe SMP in control of motility. A portion of circularmuscle innervation of the rat also has been shown to come11from the SMP (Ekblad et al, 1987, 1988). To date, only thecircular muscle of the guinea pig has been shown not toreceive input from the SMP (Smith et al, 1988). Althoughimplicated in the control of motility in the colon, thefunction of submucosal innervation of circular muscle hasnot been examined in detail in the small intestine.However, motoneurones from the submucosal ganglia have beenshown to innervate and provide inhibitory and excitatoryinputs in the circular muscle of the canine colon (Sandersand Smith, 1986)The submucosal plexus receives inputs from themyenteric and extrinsic innervation in the guinea pig(Bornstein et al, 1988) but most mucosal fibres ariseprimarily from the SMP with the exception of some SPcontaining fibres (Furness and Costa, 1987). The majorityof fibres found in the mucosa of canine small intestine alsoappear to originate in the SMP while the submucosa receiveslittle input from extrinsic sources (Furness et al, 1989).In the guinea pig, inputs from the MY? may be responsiblefor inhibitory synaptic inputs to the SMP (Bornstein et al,1988), and may utilize SS as the neurotransmitter. Thefunction of the MY? in control of ion transport has not beendetermined.Substance P (SP), somatostatin (SS) and vasoactiveintestinal peptide (VIP) were the peptides examined in this12thesis and are the focus the following discussion. Theundecapeptide substance P was the first gut neuropeptide tobe discovered (von Euler and Gaddum, 1931), on the basis ofits ability to stimulate atropine resistant contractions ofthe rabbit ileum. Substance P was among the first knownneuropeptides to be isolated and sequenced (Chang et al,1971) and among the first whose presence in the gut wasdemonstrated by immunocytochemistry (Pearse and Polak,1975)Somatostatin is a tetradecapeptide originally isolatedfrom bovine hypothalamus as a factor which inhibited growthhormone release (Brazeau et al, 1973) and has since beenlocalized in many different regions including the endocrinecells (Polak et al, 1975) and neurones (Keast et al, 1985)of the gut.Vasoactive intestinal peptide was isolated from porcineduodenum on the basis of its vasodilatory ability (Said andMutt, 1970) and, unlike SP and SS, has been localizedexclusively to neurones including those of the gut (Larsson,1977)Somatostatin occurs in two major molecular forms, SS14which was demonstrated to predominate in enteric nerves, andSS28 which predominates in mucosal endocrine cells in thehuman small intestine (Penman et al, 1983; Keast et al,131984; Baldissera et al, 1985). Exogenous application ofeither form has been shown to inhibit both small intestinalmotility in the guinea pig (McIntosh et al, l987a) as wellas secretion of electrolytes in porcine small intestine(Brown et al, 1989). Inhibition of ion transport was shownto be partly mediated by enteric nerves in the guinea pig bySS—14 (Keast et al, 1987), or by both SS—14 and S—28 in thepig (Brown et al, 1989). Somatostatin has been shown toinhibit VIP—stimulated secretion in the dog but not thehuman, whereas basal secretion was not affected in eitherthe dog or human. (Keast, 1987; Krejs and and Fordtran,1980).Substance P and VIP have been shown to increasesecretion and decrease absorption of electrolytes in bothcanine and human small intestine in vivo (Kres et al, 1980;Hubel et al, 1984; McFadden et al, 1986). Using human smallintestine in situ, these effects were shown to be direct andnot mediated by intrinsic nerves (Hubel et al, 1984).Similar studies have not been carried out in the dog, andtherefore whether or not the secretory effects of VIP and SPwere mediated by intrinsic nerves in this animal was notdetermined. The effect of SP has been shown to be partlymediated by intrinsic nerves in several species (Hubel etal, 1984; Keast et al, 1985; Perdue et al, 1987) but VIP hasbeen demonstrated to act directly on the mucosa in all14species tested (Cassuto et al, 1983; Binder et al, 1984;Carey et al, 1985).Substance P and VIP have opposite effects ongastrointestinal motility. Vasoactive intestinal peptidehas been shown to inhibit motility in human colon (Coutureet al, 1981; Furness and Costa, 1982) and the smallintestine of the guinea pig (Furness and Costa, 1982).Substance P was shown to have excitatory effects on themotility of human gastrointestinal muscle (Zappia et al,1978; Couture et al, 1981)Intracellular recordings from enteric neurones haverevealed fast and slow excitatory postsynaptic potentials(EPSPs) and fast and slow inhibitory postsynaptic potentials(IPSPs) (Wood, 1987). Since membrane potentials or ionicconductance can be measured over a period of hours, putativeneurotransmitters can be applied and receptors can becharacterized using pharmacological methods (North, 1986).Substance P has been shown to mimic the non-’cholinergic slowEPSP evoked in enteric neurones by stimulation of axonbundles between enteric ganglia both in situ (Katayama etal, 1979; Bornstein et al, 1984; Surprenant, 1984) and inculture (Willard, 1990). The signal transduction mechanismof the SP-mediated EPSP did not involve cAMP (Palmer et al,1987). Somatostatin has been shown to produce two types ofslow IPSPs in both myenteric and subinucosal neurones (Mihara15et al, 1987). Interestingly, SS—14 and SS—28 have beenshown to exert opposite effects in rat neocortical neurones(Wang et al, 1989). Somatostatin—14 was shown to increasethe delayed rectifier K+ current while SS28 was shown todecrease this current. Both effects were mediated by GTP-binding proteins and mediated by distinct receptors.Different effects of SS-l4 and SS—28 have not beendemonstrated in enteric neurones.These neuropeptides have been shown to have effects onneurotransmitter release from enteric neurones. Substance Phas been shown to stimulate the release of Ach from guineapig myenteric neurones (Yau et al, 1986) and neurotensinrelease from canine submucosal neurones (Barber et al,1989). Substance P has diverse effects on SS—IR releasedepending on the cell type. It has been shown to inhibitSS—IR release from canine submucosal neurones (Buchan et al,1990) and gastric somatostatin cells (Kwok et al, 1985) butto stimulate SS—IR release from the hypothalamus andpancreas (Reichlin, 1981). Vasoactive intestinal peptidehas also been shown to stimulate the release of SS-IR fromenteric neurones (Grider, 1989). Somatostatin has beenshown to inhibit the release of neurotransmitters, such asAch, from enteric neurones (Wiley and Owang, 1987).Receptors for neuropeptides have also been studied usingreceptor binding assays (Quirion and Gaudreau, 1985). The16distributions of SP and VIP binding sites in human (Gates etal, 1989; Korman et al, 1989) and canine small intestine(Mantyh et al, 1988; Zimmerman et al, 1989) have been shownto be similar. In both species, binding of VIP was foundpredominantly in the epithelial layer while binding of SPwas found mostly in the smooth muscle layers but also in thesubmucosal arterioles and venules, and in the epithelium.Only few studies have detected receptors for anyneuropeptides on canine enteric ganglia and these includereceptors for opioids and bombesin (Allescher et al, 1989;Ahmad et al, 1989; Vigna et al, 1987). Substance P bindingsites have been demonstrated in guinea pig enteric ganglia(Bornstein and Burcher, 1987). Although useful for thedetermination of high and low density binding sites, thesetechniques have been shown to be insensitive, probablybecause of a lack of specific ligands to date.C. Experimental Rationale.The actions and secretion of SS and SP in the submucosalplexus of the small intestine has been difficult tointerpret for several reasons. Peptides such as SS arepresent in multiple locations and in multiple molecularforms. Therefore, studies which utilize in vivoexperimental models to examine the release or actions of SSand SP become confusing. Also, the distribution and thepattern of co—localization of peptides varies between17species. Differences in peptide distribution betweenspecies may result in differences in their function.The studies presented in this thesis compare thedistribution of SP-IR, VIP-IR and SS-IR in the submucosalplexus of the human and canine small intestine. Inparticular, the possibility of co—localization of theseneuropeptides is examined. Further, the studies describethe development of dispersed cultures of submucosal gangliafrom the human small intestine and utilization of canine andhuman cultures to study the release of SS-IR and SP-IR.Finally, the studies compare the secretion of SS-IR and theeffect of SP on SS-IR secretion between the dog and human,and relate these to differences in their distribution.18D. Eypotheses1 That interspecies variations in neuropeptide localizationand enteric neuronal morphology exist between the human andcanine small intestine.2. That differences in neuronal chemical coding andmorphology will be reflected in neuronal function.3. That shortterm cultures of human and canine SMP willprovide a model system in which to examine whetherdifferences in neuronal chemical coding and morphology willbe reflected in neuronal function.19E. Specific Objectives1. To develop dispersed cultures of submucosal neuronesfrom human small intestine and compare these to cultures ofcanine small intestine.2. To compare the morphology of canine and human smallintestinal innervation using the antibody to protein geneproduct 9.5.3. To compare the distribution of SP, SS and VIP in canineand human submucosal neurones using single and doublestains, and to compare the morphology of these neurones.4. To compare the total number of neurones and total numberof SPIR, SS-IR and VIP-IR containing neurones per canineand human submucosal ganglion using PGP/peptide doublestains.5. To compare the distribution of SP, SS and VIP in canineand human submucosa in culture.6. To examine and compare, the content and secretion of SS—IR from canine and human submucosal neurones in culture.7. To examine the content and secretion of SP—IR fromcanine submucosal neurones in culture.208. To compare the effects of SP on secretion of SS—IR fromhuman and canine submucosal cultures.21II. IMMUNOCYTOCHEMICAL METHODSA. Tissue Sections1. Tissue PreparationCross—sections of intact intestine (1 cm thick), andstripped submucosa (6 cm2 ) were fixed in Bouin’s solutionfor morphological study. The tissues were fixed (in Bouin’ssolution) for 2 h, washed and stored in 70% alcohol prior toprocessing. The tissue was dehydrated through gradedalcohols and xylene, and embedded in wax. Seven micronsections were air dried on glass slides at 37°C and the waxremoved with xylene followed by clearing through petroleumether2. ProtocolAll antibodies were diluted in PBS containing 10% bovineor swine serum while avidin layers were diluted in PBSalone0 Primary antibodies were incubated at 4°C, and allother procedures were carried out at room temperature.22a. Primary Antibodies (Table 1).The tissue sections were incubated for 48—72 h inprimary antibody diluted with PBS with 10% horse serum. Thebound antibodies were localized using peroxidase orimmunofluorescence techniques.b. Secondary Layersi0 PeroxidaseSections which were utilized for peroxidase stainingwere previously incubated with a 0.01% solution of hydrogenperoxide to block endogenous peroxidase activity beforeincubation with the primary antibody. The endogenous enzymebecomes blocked because the peroxide oxidizes peroxidase.Incubation with a biotinylated (biotin conjugated) secondaryantibody (Table 2) was then carried out for 1 h (caninetissue) or 2 h (human tissue, canine and human cultures). Afurther incubation was carried out for 1—2 h with a solutioncontaining avidin and biotin-peroxidase (ABC) previouslyincubated for 10—30 mm. Two methods were used to localizethe resulting complex. The first method utilized a solutionof diaiuinobenzidene (DAB, 4 mg/ml) and hydrogen peroxide(0.03 %), in 0.1 M Tris buffer, which was filtered and addeddropwise on slides and cultures to develop the peroxidasereaction. The second method utilized 100 ml of 0.1 M Tris23buffer containing 200 mg dextrose, glucose oxidase (0.3 mg),aimnonium chloride (40 mg) and DAB (4 ing) in which thesections or cultures were incubated for 1—1.5 h Aftercounterstaining with heiuatoxylin, the staining was observedwith a Zeiss Axiophot microscope equipped with phasecontrast optics.ii. linmunofluorescence,Primary antibodies were visualized by immunofluorescencein two ways. An indirect method utilized a 12 h incubationwith a fluorophore-conjugated second antibody (Table 3). Asecond method was carried out by incubation with abiotinylated second antibody for 1 h (canine tissue) or 2 h(human tissue, human and canine cultures) followed by a 45—60 mm incubation with an avidin-fluorophore third layer(Table 4). The fluorophores used were fluoro—isothiocyanate(FITC) and tetramethyl-rhodamine-isothiocyanate (rhodamine).The FITC and rhodamine staining was observed with a ZeissAxiophot microscope equipped with epifluorescence, usingbarrier filters of band width 465-490 nm (green) and 510-560nm (red), respectively.24Table 1. Primary Antibodies.Antigen Source Dilution Species TypeVIP CURE 1:1000 mouse serumVIP Peninsula 1:1000 rabbit serumSoma CURE 1:10,000 mouse ascitesSoma RPG 1:1000 mouse ascitesSP SL 1:2000 rabbit serumSP JP 1:2000 rabbit serumSP RPG 1:1000 guinea serumpigPGP Immunonuclear 1:2000 rabbit serumCURE Professor J. Walsh, Centre for Ulcer Research andEducationRPG Regulatory Peptide Group.SL— Professor S.A. Leeman.JP— Professor J.M. Polak.25c. Double Stains.Double staining was carried out using the techniquesdescribed for iimuunofluorescence. It was necessary tochoose antibodies to peptides raised in species such thatcross—reactivity between primary and secondary antibodiesdid not occur. Second layers have been affinity purified toremove potential cross—reacting globulins, and have beendeveloped specifically for double staining. For example,primary monoclönal antibodies were localized with secondaryantisera raised in goat or donkey rather than the more usualrabbit antimouse IgG to eliminate cross-reactivity withrabbit primary antibodies.d. Controls.Extensive characterization of the antisera/antibodieshas been carried out previously to eliminate the possibilityof cross-reactivity with other related peptides. Inaddition, incubations using PBS in 10 % serum layers werecarried out to determine the extent of non—specific stainingdue to the second and third layers.Be Tissue Culture.The cultures were fixed in Bouin’s solution for 15—30mm, washed with distilled water followed by phosphate26buffered saline. The localization of primary antibodies wascarried out as for tissue sections except for the followingchanges. Some cultures were frozen with liquid nitrogenbefore washing and/or incubated in first, second and thirdlayers containing Triton X-100 (0.1%), in order to lyse theplasma membrane and allow the optimal penetration ofantibodies into the cell. The durations of first, secondand third layer incubations were up to 50 % longer thanthose for tissue sections. The concentrations of primaryand secondary antibodies, and of avidin third layers weredouble those used in tissue sections.C. Quantification of Peptide—containing Neurones.In tissue sections, the percentage of the submucosalplexus occupied by ganglia was determined by planimetry insections stained with PGP 9.5. Neurones iituuunoreactive foreach peptide were counted in six ganglia per section in 6sections to a total of 180 qanglia and compared to the totalnumber of neurones in those ganglia. The total number ofneurones was determined using the PGP 95 antiserum. In thecultures, cells immunoreactive for somatostatin were countedand compared to the total number of neurones in that groupor cluster.27Table 2. Biotinylated AntibodiesAntigen Source Dilution SpeciesRabbit IgG Vector Co. 1:300 goatRabbit IgG Jackson Co. 1:1000 donkeyMouse IgG Vector Co. 1:300 horseMouse IgG Jackson Co. 1:1000 goatTable 3. Fluorophore-conjugated AntibodiesAntigen Source Dilution SpeciesRhodamineRabbit IgG Jackson Co. 1:1000 donkeyMouse IgG Jackson Co. 1:1000 goatFITCRabbit IgG Jackson Co. 1:1000 donkeyMouse IgG Jackson Co. 1:1000 donky28Table 4. Avidin LayersLayers. Source DilutionAvidin FITC Vector Co. 1:1000Avidin rhodaiuine Vector Co. 1:1000Avidin peroxidase Vector Co. 1:100029III. CHAPTER ONE. IMMUNOCYTOCHEMICAL STUDIESA. IntroductionEnteric neurones in several mammalian species have beenshown to contain a plethora of neuropeptides (Schultzberg etal, 1980; Furness and Costa, 1982) The enteric nervoussystem of the guinea pig has been the most thoroughlycharacterized with respect to the pathways of individualpeptide containing neurones (Costa et al, 1987; Furness etal 1990). Differences in the distribution of peptideswithin nerve cell bodies of the MYP and external musculatureof the dog and guinea pig have been shown (Tange, 1983;Daniel et al, 1985; Daniel et al, 1987; Furness et al,1989). In particular, differences in innervation of thecircular muscle of the guinea pig and dog may lie in thesubmucosa (Furness et al, 1990). Canine circular muscle hasbeen shown to contain SPIR and VIP-IR fibres whose originwas the outer submucosal ganglia, or Schabadasch’s plexus.Guinea pig circular muscle receives SP-IR and VIP-IR fibressolely from the MYP.Similarities exist between the dog and human withrespect to enteric submucosal/mucosal neuropeptidedistribution. Single staining of VIP, SP and SS wascomparable between canine and human submucosa/mucosa. Eachpeptide has been found in both submucosal and myenteric30neuronal cell bodies and fibres in canine and humaniuucosa/subxnucosa (Fern et al; 1982; Tange, 1983; Keast etal, 1984; Daniel et al, 1985; Keast. et al, 1985; Wattchowet al, 1988; Furness et al, 1990). These studies have shownthat fibres containing VIP iimuunoreactivity (VIP-IR) weremore abundant than those containing SP-IR, which in turnwere more abundant than those containing SS—IR.Comprehensive examinations of neurotransmittercombinations in enteric neurones have been carried out inthe guinea pig and rat and have revealed distinctdifferences (Furness et al, 1989; Pataky et al, 1990). Thisis often referred to as chemical coding of neurones (Furnesset al, 1989). In the SMP of the guinea pig, two major subgroups of neurones were distinguished. These weredynorphin/galanin/VIP containing and CCK/CGRP/choline acetyltransferase (Chat) /galanin/neuropeptide Y (NPY) /SScontaining neurones (Costa et al, 1987; Furness et al,1989). In the rat, the major subgroups were VIP/NPYcontaining and SS/SP/CGRP containing neurones (Pataky et al,1990). These studies used multiple staining techniques withhighly specific antisera/antibodies to characterize theneuronal types. However, no information was obtained aboutwhether functional differences could be correlated withdifferential peptide coding.31Earlier studies with canine intestine demonstrated thepresence of large ganglia containing a high proportion ofneurotensin immunoreactive (NT—IR) neurones (Buchan andBarber, 1987). These NT-IR positive neurones had not beendemonstrated in either the rat or the guinea pig. Thechemical coding of canine SMP neurones probably differs fromboth the guinea pig and the rat. The basis for such adifference may be the diet and size of the animal (Powell,1987; Gabella, 1990)Immunocytochemical studies of peptide co-’localizationand chemical coding of human and canine neurones of thesmall intestine are few (Wattchow et al, 1988; Furness etal, 1990). In the human small intestine SP and enkephalinhave been shown to co—exist in fibres which were suggestedto be excitatory to the external muscle (Wattchow et al,1988). These authors also showed that NPY and VIP coexisted in a separate population of fibres and weresuggested to be inhibitory to the external muscle.In the present study, immunocytochemical staining wascarried out to examine the hypothesis that there isinterspecies variation in the localization of SS-IR, SP—IRand VIP—IR between the human and canine small intestine.The availability of highly specific antibodies/antiseraraised in different species (mouse, rabbit and guinea pig)32combined with affinity-purified secondary antisera wasessential in these experiments.In order to quantify the number of neurones within aganglion, a general neuronal marker was required. In thesestudies PGP 9.5 was used. Protein gene product 9.5 is asoluble cytoplasmic protein originally detected in proteinextracts of human organs by high resolution two dimensionalpolyacrylamide electrophoresis (Doran et al, 1983). Thedistribution of PGP is similar to neurone-’specific enolase,since both have been found in neurones and neuroendocrinecells (Doran et al, 1983). Polyclonal antibodies to PGPhave demonstrated peripheral nerve cell bodies and nervefibres with clarity and intensity (Gulbenkian et al, 1987;Lauweryns and Van Ranst, 1988; Wilson et al, 1988).33B. Results1. Human Tissue Sectionsa. Protein Gene Product 9.5.The antibody to PGP strongly stained human neuronal cellbodies and fibres throughout the small intestine. Thisstaining revealed the nerve cell bodies of the SMP and nervefibres throughout the mucosa (figure 2). Submucosal gangliawere anatomically separated into those adjacent to theinterface with the mucosa and those situated by the circularmuscle. The myenteric ganglia were strongly stained andthere was extensive innervation of the circular andlongitudinal muscle (figure 3).Quantification of the relative proportion of peptidecontaining neurones was carried out using double stains ofPGP 9.5 in combination with the particular neuropeptide (foran example see figure 4). The intensity staining was muchgreater for PGP than for SP—IR, SS-IR or VIP-IR and revealedmore abundant nerve fibres found throughout the ganglia.The volume of the myenteric ganglia was larger than that ofthe submucosal ganglia because of larger amounts of fibresand other neuronal processes (figure 4).34Figure 2. PGP 9.5 staining of human mucosa. Note cellbodies (large arrows) immediately below muscularis mucosae.Immunoreactive fibres were present throughout the mucosallayer (small arrows).x 200.36Figure 3. PGP 9.5 staining of human muscularis (CM =circular muscle, LM = longitudinal muscle). Note fibres inboth muscle layers (small arrows) and intense staining ofthe neurones in the NYP (large arrow).x 100.,38Figure 4a) VIP-IR neurone (large arrow) and fibres in the humanMYP.,x 400.b) The same ganglion in a double exposure showing VIP-IRneurone double stained by PGP 9.5 and a single neuronestained by PGP-IR 9.5 (small arrow)x 400.6E40The average number of neurones per ganglion in the SMP was2.9 ± 0.3 (184 ganglia, n = 5 donors) (figure 5) and made uponly 5% of the submucosa while collagen made up over 85% ofthis layer.b. Autofluorescence.Certain neurones demonstrated autofluorescence whenviewed under ultraviolet light which was often associatedwith neuropeptide containing neurones (figure 6).Autofluorescence was indicative of the high intrinsic aminecontent of these neurones and fibres. The autofluorescencewas visible at 3 wavelengths (380, 480 and 570nm) but couldbe differentiated from Rhodamine (570 nm) arid FITC (480 nm)because these were not observed at the 340 nm wavelength.c. Single Staining of SSIR, SPIR and VIP-IR.Single staining of human small intestine with antibodiesto SS, SP and VIP revealed a pattern similar to that foundpreviously in other laboratories. Somatostatin—IR was foundin endocrine cells, submucosal ganglia and myenteric ganglia(figure 7 a, b, c). Endocrine cells containing SS—IR wereabundant in the mucosal epithelium, stained more intenselythan neurones containing SS—IR and were concentrated at thebase of the crypts (see also figure 8 a).41Figure 5. The relative size distribution of humansubmucosal ganglia of the small intestine. Note that over50 % of ganglia contained 3 or less neurones.42aCa9-00EDC7060504030201000 5 10 15 20 25 30number of neurones per ganglion43Figure 6.a) a section of human duodenum photographed underultraviolet light (370 rim). Note the presence of intenseautofluorescence (arrows) in the SNP ganglion.x 200.b) The same area photographed under long wavelength (570nm) to show SP—IR neurones. Note the close association ofthe autofluorescent areas (arrows) with the positivelystained neuronesx 200.444445Submucosal and myenteric ganglia also contained SS—IRnerve fibres but these were less abundant than thosecontaining SP-IR or VIP-IR. Forty two percent of neuronalcell bodies per human submucosal ganglion contained SS—IR.Cell bodies containing VIP-IR were again found inmyenteric (separate from neurones containing SS—IR) andsubmucosal ganglia (figure 7 d, e). Occasional nerve cellbodies were observed in the smooth muscle and submucosa.Fibres containing VIP-IR were found within the gangliarunning between other neurones. Fibres were also foundthroughout the MYP, SMP and the smooth muscle layers.Fibres containing VIP—IR were also found in the mucosa(figure 7 f), and in general were found in similar amountsto SP—IR fibres, and greater than SS—IR fibres. Fortypercent of neurones per human submucosal ganglion containedVIP-IR. Endocrine cells containing VIP-IR were notobserved.Cell bodies containing SP-IR were found in humanmyenteric (not shown) and submucosal ganglia (figure 7 g).Fibres containing SP-IR were found within the myenteric andsubmucosal ganglia and seemed to form pericellular basketsaround other neurones. Fibres were also found in the musclelayers, the SMP, the mucosa and the mucosal villi.Substance P fibres were more plentiful than fibrescontaining SS—IR and were present in amounts similar to46those containing VIP—IR. Forty two percent of neuronal cellbodies per human ganglion contained SP-IR while endocrinecells containing SP—IR were not found.d. Double Stains.Substance P—IR and SS-IR were found to completely coexist in the neurones of human submucosal (figure 8 a, b, c)and myenteric ganglia. Large numbers of fibres weredemonstrated in all layers of the gut which contained onlySP—IR while endocrine cells of the mucosa were found tocontain only SS-IR.Neurones containing VIP—IR were demonstrated to be apopulation completely distinct from those neuronescontaining SS-IR/SP-IR in both the myenteric and submucosalneurones (figure 9, and 10 a, b). Double stains revealedneuropeptide containing fibres and varicosities of eachneuropeptide distributed around other neuropeptidecontaining neurones within the ganglia. Some fibres formedpericellular baskets around neuropeptide containingneurones. Certain small ganglia were made up of neuroneswhich were exclusively of the VIP-IR type or SS-IR/SP-IRtype.47Figure 7. Human intestine representative inununostains.a) SS—IR endocrine cell (arrow) x 200.b) SS—IR SMP neurones (arrow) x 200.c) SSIR neurones (arrows) and nerve fibres in the MYPx 200.d) The same ganglion as in stained for VIP-IR. Notethe separation of neurones. SS-IR (arrow heads), VIP-IR(long arrows) x 200.e) VIP-IR neurones (arrows) in a ganglion of the SMPx 200.f) VIP—IR fibres (arrows) in the lamina propria of a villusin the duodenal mucosa x 200.g) SP-IR neurones (arrows) and fibres in a ganglion in theduodenal SMP x 200.6t’50.1Ln52Figure 8. Human duodenal sections double stained for SS—IR(FITC) and SP-IR (Rhodamine).a) A low magnification overview showing SS-IR epithelialendocrine cells (open arrows) which do not contain SP—IR;SP—IR fibres in the lamina propria and muscularis mucosae(small arrows) which do not contain SS-IR; a ganglion in theSMP (large black arrow) containing a neurone double stainedfor both SS-IR and SP-IR (large white arrow)x 50.b) A higher magnification micrograph of SP-IR in a SMPneurones co—localized with c) SS—IRx 500C,,54Figure 9. Human duodenal SMP stained for SS-IR (Rhodamine,large arrows) and VIP-IR (FITC, small arrows). Note thecomplete separation of the two neurone types and thepresence of SS—IR varicosities around the VIP—IR neurone(small arrow heads) x 500.Lfl56Figure 10. Human duodenum double stained for VIP-IR (FITC)and SP-IR (Rhodamine).a) A ganglion in the SNP with a single VIPIR neurone(small arrow) x 200.b) A double exposure of the same ganglion showing the VIPIR neurone (small arrow) and two SP—IR neurones (largerarrows) x 200.Note the lack of co—localization.In582. Canine Tissue Sectionsa. Protein Gene Product 9.5.The antibody to PGP strongly stained canine neuronalcell bodies and fibres throughout the small intestine. Thisstaining revealed the nerve cell bodies of the SMP and nervefibres throughout the mucosa (figure 11). Canine submucosalganglia were also present in two distinct groups, one closerto the mucosa and one near the interface with the circularmuscle (figure 11, 12, 13). Unlike the human smallintestine, PGP 9.5 staining revealed the presence of adiscrete deep muscular plexus, situated near the outer layerof the submucosa (figure 13). Canine myenteric ganglia werestrongly stained and there was extensive innervation of thecircular and longitudinal muscle, The intensity of stainingwas much greater for PGP than for SP”IR, SS—IR or VIP-IR andmore abundant nerve fibres were apparent throughout theganglia. The volume of the myenteric ganglia was largerthan that of the submucosal ganglia again due to largeamounts of neuropil i.e. fibres and other neuronalprocesses, and glia. The average number of neurones perganglion in the SMP was 7.7 ± 0.6 (185 ganglia counted, n=6dog) and made up 10* of the submucosa while collagen madeup over 80% of this layer (figure 14).59Figure 11. PGP 9.5 staining of canine duodenum (MUC =mucosa, CM = circular muscle). Note the presence of twodistinct sets of ganglia in the SMP, one close to theinterface with the mucosa (upper dotted line), the otherapposed to the circular muscle (lower dotted line)x 50.6061Figure 12. Details of the PGP 9.5 staining in the canineduodenal SMP. Note the close association of the largeganglion with the muscularis mucosae (in = inuscularismucosae, CM = circular muscle, M = inucosa).x1OO.6263Figure 13. Details of the PGP 9.5 staining in the canineduodenum. Note the band of inununoreactive fibres (smallarrows) at the inner surface of the circular muscle formingthe deep muscular plexus (dmp)x200.6465Figure 14. The relative size distribution of caninesubmucosal ganglia of the small intestine. Note that themajority of ganglia contained between 4 and 7 neurones.Large ganglia (those which contained > 15 neurones) made up10 % of the population0660C,C0C,‘4-0L.a)-aEDC70504030201000 5 10 15 20 25 30number of neurones per ganglion67b. Single Stains of SS—IR, SP—IR and VIP—IR.The pattern of single staining of canine small intestinewith antibodies to SS, SP and VIP was again similar to thatfound in previous investigations. Cell bodies containingSS—IR were found in submucosal ganglia (figure 15 a),endocrine cells and myenteric ganglia. Submucosal andmyenteric ganglia also contained SS—IR nerve fibres but wereless abundant than those containing SP-IR or VIP-IR. TheSS—IR neurones were often grouped into clusters within theganglia whose fibres would exit the ganglia in the samedirection. Thirty five percent of neurones per caninesubmucosal ganglion contained SS—IR, Endocrine cellscontaining SS-IR were predominant in the region of thecrypts of the mucosal epithelium and were stained moreintensely than neurones containing SSIR.Cell bodies containing VIPIR were again found inmyenteric and submucosal ganglia. Fibres were also foundthroughout the MYP, SMP and the smooth muscle layers.Fibres containing VIP-IR were also found in mucosa villiand, as in the human, were found in numbers similar to SP-IRfibres and greater than SS-IR fibres. Thirty one percent ofneurones per canine submucosal ganglion contained VIP-IR.Endocrine cells containing VIP-IR were not observed.68Cell bodies containing SP—IR were found in caninesubmucosal (figure 15 b) and myenteric ganglia. NumerousSP—IR fibres containing varicosities were found within themyenteric and submucosal ganglia which were distributedamongst other neuronal cell bodies. As with the otherpeptide containing neurones, SP—IR neurones were segregatedwithin the ganglia and their fibres exited the ganglia inthe same direction. Fibres were also found in the mucosa,the SMP, MYP and muscle layers. Substance P fibres weremore plentiful than fibres containing SS-IR and were presentin amounts similar to those containing VIP-IR. Thirtypercent of neurones per canine submucosal ganglion containedSP—IR while endocrine cells containing SP-IR were not found.c. Double Stains.The double stains revealed that, unlike the human, SS-IRand SP—IR were never present in the same neurone (figure 16a, b). Neurones containing VIPIR were demonstrated to be apopulation completely separate from those neuronescontaining SS—IR or SP-IR (figure 17 a, b). Double stainsalso demonstrated neuropeptide containing fibres andvaricosities of each neuropeptide distributed around otherneuropeptide containing neurones within the ganglia. Somefibres formed pericellular baskets around neuropeptidecontaining neurones, Finally, double stains provided69further evidence for the spatial segregation of differentneuronal types within the ganglia.The data obtained from the iimuunocytochemistry ofsections of canine and human small intestine are summarizedin tables 5, 6 and 7.70Figure 15. Representative iinmunostains of the canine SMP.Note that the neurons are grouped into small clusters withinthe larger ganglion (larger arrows) and that the fibres of acluster appear to exit the ganglion as a unit (smallarrows).a) SS—IR x 200.b) SP-IR x 200.IL72Figure 16, a and b. Double stains of canine SNP ganglia forSS—IR (FITC, small arrows) and SP—IR (Rhodamine, longerarrows). Once again note the grouping of the neurones intoclusters of a single type and the merging of exiting fibres(see ‘a’, medium sized arrow)x500,74Figure 17. Double stains of canine SMP ganglia. a) VIP-IRneurones (Rhodamine, large arrows) and SS—IR (FITC, smallarrows) x 500. Note the absence of co—localization.b) VIP—IR neurones (Rhodamine, large arrows) and SP—IR(FITC, small arrows) x 500. Note the absence of co—localization.cL76Table 5. Summary of Neuropeptide Distribution and Colocalization in the Human Small IntestineSS—IR SP-IR VIP—IRMucosa.endocrine cells +++nerve fibres + ++Ganglia(SMP and MYP)neurones +++ +++ +++nerve fibres + ++ +++Co—localization.SS—IR yes noSP-IR yes noVIP—IR no no+++ most abundant0++ less abundant.+ least abundant.77Table 6. Summary of Neuropeptide Distribution and Colocalization in the Canine Small IntestineSS—IR SP—IR VIP-IRMucosa.endocrine cells +++nerve fibres + ++Ganglia(SMP and MYP)neurones +++ +++ +++nerve fibres + ++ +++Co—localization.SS-IR no noSP—IR no noVIP-IR no noDeep Muscular — ++ +++Plexus+++ most abundant. ++ less abundant. + least abundant.CDCl)Cl)iCDHCl)iiII0DIIHHI-’HCDCDU)CDCDCDCD1‘1bIi00CD0qCDCD0CD(0U)0U)QII0I1+0DI Cl.DI w CD rIlDCD0CD1+ 0 (JCD79C. DiscussionStaining of tissue sections of canine and human smallintestine with PGP 9.5, a general marker for peripheralnerves (Gulbenkian et al, 1987; Lauweryns and Van Ranst,1988; Wilson et al, 1988), was used to localize allmyenteric and submucosal neuronal cell bodies and revealedan extensive network of nerve fibres throughout the smallintestine. The distribution of nerve fibres was similarbetween the dog and the human, but canine submucosal gangliacontained, on average, a greater number of neurones. Also,the density of neurones was greater in the canine SMP thanhuman small intestine based on the smaller amount ofcollagen in the canine submucosa. It has been suggestedthat the size of the intestine determines the number ofneurones per unit of serosal surface and the total number ofneurones, with each of these values being parallel to thesize (weight) of the animal (Gabella, 1987; 1990).The average size of each submucosal ganglion was largerin canine small intestine than in the human small intestine.These data suggest that the size of submucosal ganglia doesnot parallel the size of the animal. The number of neuronesper submucosal ganglion in the human was closer to that ofthe rat which has 3-5 per ganglion (Pataky et al, 1990),than to that of the dog, Sheep probably represent a thirdgroup of animals, based on the low spatial density of80neurones in the small intestine and the large size of theenteric ganglia. The small intestine of these ruminants hasbeen shown to have a low spatial density of neurones buthave neuronal ganglia which contain large numbers ofneurones (Gabella, 1987).It is generally accepted that mammalian myentericganglia contain a greater number of neurones than thesubmucosal ganglia (Gabella, 1987; Furness et al, 1987),although preliminary càunts have indicated that this is notthe case in the dog, while in the human the numbers ofneurones in submucosal and myenteric ganglia seem to beequal (data not shown). A major portion of the volume of.the iuyenteric ganglia in both the dog and human was made upof neuropil i.e. neuronal fibres and other neuronalprocesses, and glia. It could be argued that 2 dimensionalrenditions of 3—D structures do not accurately reflectneuronal number and volume, requiring the use of whole mountpreparations. However, analysis of sections cutperpendicular and parallel to the direction of the circularmuscle provided a reasonable construction of both plexi.The myenteric ganglia were in an orderly array withtheir long axis travelling in the same direction as thecircular muscle as noted in previous studies (Gabella,1987). Whether the submucosal ganglia possessed a similarorientation could not be discerned from tissue sections but81most likely these ganglia were less structured, as has beendemonstrated in previous studies (Gabella, 1990).The presence of SS—IR in submucosal and myentericneurones, and endocrine cells of the mucosa, and the smallnumbers of SS—IR containing nerve fibres in canine and humansmall intestine tissue sections was in agreement with theresults of previous studies (Daniel et al, 1985; Keast etal, 1986). Nerve fibres throughout all layers of the smallintestine and neurones of the myenteric and submucosalplexus contained SP-IR and VIP-IR, as demonstratedpreviously (Tange, 1983; Daniel et al, 1985). Mucosalendocrine cells did not contain SP-IR, although this hasbeen reported in both the human and canine small intestine(Daniel et al, 1985; Keast et al, l985) In studies carriedout by these authors, endocrine cells containing SP—IR werefew in number and stained less intensely than thoseendocrine cells containing SSIR and SPIR. Only the smallintestine of the marmoset has been shown to contain largenumbers of endocrine cells which stained brightly for SP-IR(Keast et al, 1985).The double staining experiments revealed the coexistence of SP—IR and SS—IR in neurones of human, but notcanine, small intestine which has not been previouslydemonstrated. Neurones containing VIPIR were distinct fromthose containing SP-IR and SS-IR in both species, which is82in agreement with previous studies which have demonstratedno co-existence between SP-IR and VIP-IR fibres in themuscularis externa throughout the canine and humangastrointestinal tract (Wattchow et al, 1988; Furness et al,1989)Two separate populations of motoneurones have beenproposed by these authors to innervate the external muscleof the human gastrointestinal tract, one containing VIP-IRand neuropeptide Y-IR and the other containing SP-IR andenkephalin—IR (Wattchow et al, 1988). Fibres containingVIP—IR and NPY—IR were called inhibitory by Wattchow et al,(1988), since both peptides have been shown to inhibitmotility of human gastrointestinal muscle (Couture et al,1981; Furness et al, 1982; Allen et al, 1987). The presencein the human small intestine of nerves which are excusivelyinhibitory is consistent with the results obtained in thisstudy.Fibres containing SP—IR and enkIR were calledexcitatory by Wattchow et al, (1988), since these peptideshave been shown to be excitatory on human gastrointestinalsmooth muscle (Zappia et al, 1978; Couture et al, 1981).The presence of nerves which are exclusively excitatory inthe human small intestine is not supported by the presentstudy since neural SS-IR, which has been shown to inhibitmotility in the small intestine (McIntosh et al, 1987a), was83found to co-exist in all SP-IR containing neurones in boththe myenteric and submucosal ganglia. It may be argued thatSS—IR may only act as an interneuronal neurotransmitter andthus not affect motility by acting directly on the muscle,in a physiological setting. An indirect action of SS onmuscle, by inhibiting enteric neurones, has not been provenand furthermore each peptide may potentially affect adjacentneurones containing excitatory and inhibitoryneurotransmitters making the terms “excitatory” and“inhibitory” neurone inappropriate (Cooke, 1989).The interspecies alteration in peptide profilesdemonstrated in this study parallels that seen between theguinea pig and the rat (Costa et al, 1987; Pataky et al,1990). All neurones containing SS—IR were found to co—existwith SP—IR in the rat small intestine which was similar tothe human small intestine (Pataky et al, 1990). Theproportion of neurones containing SP-IR and VIP-IR in therat (46 and 42 %, respectively) and human (42 and 39 %) wasalso similar. There were two major differences between therat and the human in the distribution of peptides examinedin this study. Not all rat SP-IR neurones contained SS-IR,and SS—IR neurones made up only 18 % of submucosal gangliarather than 40 % as in the human. The similarities inpeptide distribution between the rat and human smallintestine were in addition to the similarity in total numberof neurones per ganglion.84In the guinea pig, SP-IR and SS-IR have been shown toexist in separate neurones similar to canine submucosalganglia (Costa et al, 1987). The proportion of neuroneswhich contain SS-IR was similar between these species (29 %in the guinea pig and 32 % in the dog), whereas theproportion containing SP-IR and VIPIR was different. Theproportions of canine submucosal neurones containing SP—IRand VIP—IR were 32 % and 30 % respectively, but were 11 %and 45 % respectively in the guinea pig. Thus, even inthose species in which there were similarities in chemicalcoding of the neurones, significant differences inganglionic composition (i.e. the proportions of peptide—containing neurones) occurred. There is general agreementthat interspecies variation in peptide localization of theenteric plexi will be present in almost all cases (Furnesset al, 1989).In human tissues, VIP—IR fibres innervated SP-IR/SS-IRcontaining neurones and SP—IR fibres innervated VIP—IRneurones. In addition, SP—IR containing fibres wereassociated with SP—IR containing neurones, providinganatomical evidence for interganglionic regulation.In canine tissues, SP-IR fibres were distributedthroughout the larger ganglia surrounding neuronescontaining both VIP—IR and SS-IR. The results provided85anatomical evidence for the action of SP on SS—IR secretion.It should be cautioned that anatomical evidence cannot betaken as conclusive, since data presented in this study donot support an action of SP on SS—IR secretion in the human,even though SP-IR fibres were found apposed to SS-IRcontaining neurones.Fibres located in all regions were shown to contain onlySP—IR in both canine and human tissue. In human tissue, theco-existence of SP—IR and SS-IR in neuronal cell bodies, butnot fibres, could have been due to low levels of SS—IR.Another possibility for the lack of co-existence in fibreswas that those containing SP—IR were extrinsic fibres, suchas sensory afferents of the vagus. Substance P has beenshown to commonly be present in vagal afferents in the smallintestine and several other organs of many species where itdoes not co-exist with SS—IR (Costa et al, 1987).The source of SP-IR and VIP-IR fibres in the outer layerof circular muscle in the canine small intestine has beenshown to be in the MYP, whereas fibres in the inner layerand the innervation of the muscularis mucosae and mucosallayer originated in the submucosal plexus (Furness et al,1989)In both species, there was a spatial difference in thelocalization of neuronal types. In the canine sections, the86SS-IR, VIP-IR and SP—IR were clearly segregated within theganglion. In the human sections, segregation was moreextreme, with small ganglia (2-4 cells) being made upexclusively of VIP-IR or SS-IR/SP-IR neurones.D. SummaryThe data collected have supported the existence of majorinterspecies variations in chemical coding of entericneurones between human and canine small intestine. Theyhave confirmed the existence of two distinct plexuses in thesubmucosa of the canine small intestine. The presence ofneuropeptide containing nerve fibres and varicosities withinenteric ganglia has provided a morphological basis forneuropeptide actions on other neurones. Finally, it wasdemonstrated that different neuronal types are segregatedwithin enteric ganglia, and send processes in the samedirection. The factors which determine differences ingastrointestinal gross anatomy and function have beensuggested to be related to diet (Powell, 1987) and size ofthe animal (Gabella, 1990). Thus, the gastrointestinaltract of omnivores such as rats and humans are more similarthan the gastrointestinal tract of carnivores such as thedog, or ruminants such as sheep. In other words, theinterspecies differences observed in this study may be due,in part, to differences in diet.87IV. CHAPTER TWO. ISOLATION OF CANINE AND HUMAN SUBMUCOSALNEURONESA. IntroductionThe release of SS—IR and SP-IR from the mammalian smallintestine using in situ and in vivo experimental models hasbeen demonstrated (Andersson et al, 1982; Donnerer et al,1984; Manaka et al, 1989) but the results have beendifficult to interpret for two reasons. First, SS-IR andSP-IR have been found in three different groups of cells,namely neurones of the submucosal and myenteric plexi, andendocrine cells of the canine and human small intestine(Keast et al, 1986; Keast et al, 1985; Daniel et al, 1985).Thus, a peptide released from endocrine and/or neural cellscould play a role in the physiological homeostasis of thesmall intestine. Second, SSIR is present in the smallintestine in two primary molecular forms, namely SS—14 andSS-28. Substance P is processed from the tachykinin genewhich can express three different peptides, and it is notknown which are expressed in different cell types (Dockray,1987). Therefore, it has become necessary to establishmodels that circumvent difficulties associated with in vivoexperiments in order to understand differential release ofpeptides from neurones or endocrine cells.88Organotypic cultures from small intestine have beendeveloped using segments of the gut wall which contain theenteric plexuses in association with muscle layers andconnective tissue (Gershon et al, 1980). The separation ofthe myenteric and submucosal plexuses from the gut wallusing a combination of enzymatic treatment andmicrodissection, and the maintenance of explants in tissueculture was later carried out (Jessen et al, 1978; 1983).Dispersed cultures of rat small intestine have also beendeveloped and have been extensively characterized withrespect to their morphological, pharmacological andelectrophysiological properties (Nishi and Willard, 1985;Willard and Nishi, 1985 a,b). These authors found thatenteric neurones maintained in tissue culture possessedmorphological, pharmacological and electrophysiologicalproperties similar to those possessed by neurones in situ.More recently, methods to examine the release ofpeptides from isolated cultured cells obtained from mucosal(Barber et al, 1986) and submucosal layers (Barber et al,1989) of the small intestine have been developed and theseoffer certain advantages over in vivo studies. Acutelydissociated ganglia from myenteric neurones of guinea pigs(Grider, 1989) and a newly developed short-term culture ofsubmucosal neurones from canine small intestine (Buchan etal, 1989) have been used specifically to study the releaseof SS—IR from enteric nerves, The use of acutely89dissociated neurones for secretion studies is limitedbecause nerve fibres are not present after isolation. Thus,the release of neurotransmitter would have to be from thecell soma for which there is presently no evidence in vivo.The control of peptide release in vivo is mediated viareceptor regulation and/or a membrane voltage dependentmechanism. In culture studies, the role of receptors can bemimicked by the addition of specific pharmacological agents.These drugs activate second messenger pathways normallyassociated with receptor activation (Berridge, 1985). Threesuch drugs are commonly utilized to activate differentintracellular pathways. The phorbol esters activate proteinkinase C, the calcium ionophore (A23187) increasesintracellular calcium and forskolin increases intracellularcAMP.The effects of these drugs have been examined incultures of canine SMP neurones. Of particular interest wasthe action of phorbol 12-myristate 13-acetate (i3—PMA) which,in the presence of the calcium ionophore A23187, stimulatedthe release of SS—IR from canine submucosal neurones inculture (Buchan et al, 1990). Phorbol esters, such as B—PMA, have been shown to activate protein kinase C (PKC)(Blumberg, 1981). Protein kinase C is normally activated bydiacyiglycerol formed from phospholipase C-mediatedcleavage of membrane phospholipids, which also produces90inositol triphosphate (Nishizuka, 1986). The activation ofPKC has been shown to stimulate the release of hormones andneurotransmitters (Kaczmarek, 1987). The release of SS—IRby phorbol esters has also been demonstrated using dispersedcultures of fetal rat hypothalamus and cortex (Peterfreundand Vale, 1983).The addition of the calcium ionophore or the phorbolester alone was shown not to increase the secretion of SS—IRfrom canine submucosal neurones (Buchan et al, 1990). Thisindicated that increases in calcium per se or activation ofPKC were not sufficient to stimulate SSIR secretion.Interestingly, the addition of forskolin did notstimulate the secretion of SS—IR from canine cultures,although it was able to stimulate the release of neurotensinfrom similar cultures (Barber et al, 1989; Buchan et al,1990). Previous experiments carried out using gliafreecultures of rat cerebral neurones have shown that adenylatecyclase activation by forskolin was not sufficient tostimulate SSIR secretion (Tapia-Arancibia et al, 1988).Substance P has been shown to inhibit SS—IR release fromcanine submucosal neurones (Buchan et al, 1990) and gastricsomatostatin cells (Kwok et al, 1988) and to stimulate SS—IRrelease from the hypothalamus and pancreas (Reichlin, 1981).The effect of SP on SS—IR secretion from canine neurones was91unexpected, since it has been shown to stimulate the releaseof NT from similar cultures (Barber et al, 1989). Theinhibitory actions of SP on SS—IR release in canine cultureswere likely indirect since this peptide has also been shownto release Ach from myenteric neurones (Wiley and Owang,1987), increase intracellular calcium in dorsal hornneurones (Womack et al, 1988) and produce EPSPs insubmucosal neurones (Surprenant, 1984). Its action on SSIRsecretion from human submucosal neurones was examined in thepresent study and compared to its effects on caninesubmucosal neurones.The experiments outlined in the following chapterdescribe the development of a system to isolate humanneurones from the upper small intestine and to maintainthese neurones in tissue culture. These experiments testthe hypothesis that the difference in chemical codingbetween canine and human neurones is reflected in theirfunction. Specifically, the effects of the phorbol ester,the calcium ionophore and SP on the secretion of SS-IR fromhuman submucosal neurones were examined. Preliminaryexperiments which examine the release of SPIR from canineneurones are also described.In addition to substance P, a variety of agents weretested for their ability to modify SS-IR secretion.Receptor independent secretagogues, the calcium ionophore92A23187 and the phorbol ester i3-PMA, were utilized toincrease intracellular Ca2+ and PKC activity, respectively.It was expected that A23187 would require the conconunitantaddition of 13—PMA to stimulate secretion , since increasingintracellular calcium was insufficient in canine neurones.The phorbol, 4a-phorbol, is similar in structure to 13-PMAbut does not stimulate PKC activity in vitro (Blumberg,1981). Therefore this was used as a control for nonspecific membrane effects, since phorbols are highlylipophilic and may cause non-specific destabilization oflipid membranes.Experiments were carried out using 10 mM KC1, alone andconcommitantly with the phorbol ester. This level ofpotassium would produce a depolarization of 20 my, asdetermined by the Goldman equation using “normal” values forintracellular and extracellular ions. This was done for tworeasons. First, this determined whether a smalldepolarization would stimulate the secretion of SS—IR fromhuman submucosal neurones, Second, to determine whether asmall depolarization would render the neurones moresensitive to stimulation by the phorbol ester.It has been shown that cholinergic neurones are presentin the submucosal plexus and may be involved in the releaseof other neurotransmitters (Barber et al, 1989). In orderto determine the role of cholinergic neurones in the93secretion of SS—IR two agonists and two antagonists wereused. Carbachol and methacholine are non—specificcholinergic agonists. Methacholine is more potent, has lessnicotinic activity and has been shown to be more effectivein stimulating the release of antral SS—IR (Buchan et al,1991). Hexamethonium and atropine are nicotinic andmuscarinic antagonists and were used to block both exogenousand endogenous cholinergic substances. Cholinergic agonistshave been shown to produce both excitatory and inhibitoryeffects in enteric neurones (North et al, 1985) thereforethe overall effects of the antagonists on SS—IR secretionwere not predictable. These antagonists were also used todetermine whether the effects of SP were direct or indirectvia endogenous cholinergic action.Calcitonin gene-related peptide (CGRP) has been shown toco—exist with SP in primary sensory neurones, and has beenshown to stimulate the release of SS—IR in other systems(Dockray, 1987). Therefore, its effects on SS—IR secretionfrom human submucosal neurones were examined.94B. Methods1. Human Donor Experimentsa. Procurement of TissueTwelve to fifteen inches of upper small intestine wasobtained from multiple organ donors in association with thePacific Organ Retrieval for Transplantation (PORT) program.Previous to surgical removal, the small intestine wasperfused with Eurocollins (see appendix 2) at 4°C to removeall red blood cells. The donors were pre—screened fortransplantation and therefore had no knownpathophysiological conditions.b. Isolation of Submucosal GangliaThe duodenal bulb was not taken nor was the initialportion of duodenum which contained Brunner’s glands sincethe submucosa could not be properly separated in theseareas. The mucosa and muscle layers were removed by bluntdissection, the remaining submucosa was washed in Hanks’Balanced Salt Solution (HBSS) containing 0.1% bovine serumalbumin (BSA) and 20 mM N-2-hydroxyethyl piperazine-N-2-ethane sulphonic acid (HEPES), and chopped finely. TheHank’s buffer was used throughout the isolation for washingtissue between periods of incubation. The medium used for95incubation consisted of Basal Medium Eagle (BME) containing0.1% BSA, 20 mM HEPES and collagenase, and was gassed with5% CO2 in 02. The pH of the incubation and washing mediawas strictly maintained between 7.0 and 7.4, low enough toinhibit collagen reassembly yet remaining withinphysiological limits. Each 10 g of tissue was incubatedwith 50 ml of the incubation medium in 200 ml flasks,continually shaken in a water bath at 200 Hz and maintainedat 37°C throughout each incubation period. The isolationwas carried out in three stages. First, the tissue wasincubated for 30 mm in incubation medium containing 600U/ml collagenase (type XI) and 4 inN Cad2. The partlydigested tissue was washed and centrifuged for 5-10 mm at200 x g to remove collagen, fat and debris. The tissue wasfurther digested for two periods of 60 mm, each followed bywashing and centrifugation. The suspension was thenfiltered through a 240 u Nytex mesh and resuspended in HBSS.c. Elutriation Centrifugation (figure 18)Elutriation centrifugation utilizes centrifugal forceand flow, which act in opposing directions, to separatecells on the basis of their volume. Different fractions ofcells can be removed by altering the flow rate of fluidpassing through the elutriation chamber by altering the pumprate or by changing the speed of the centrifuge.Appropriate flow rates and centrifugation speeds were96determined empirically. The cell suspension was loaded intoan elutriator rotor (Beckman) at 2500 rpm at a flow rate of25 mi/mm. Flow was supplied by a pump (Cole Parmer,Masterfiex Model 7520-20) with a quick-loading head (ColeParmer, model 7021—20) fitted with silicon tubing (ColeParmer, type 6411-16), equipped with a pressure gauge. Thepump flow rates were calibrated before each experiment at2500 rpm. A fraction was collected at 2200 rpm at a flowrate of 35 mi/mm which contained fibroblasts, red bloodcells and cell fragments. A second fraction was collectedat 800 rpm and a flow rate of 100 mi/mm which containedsingle ganglia and clusters of two or three cells as well asundigested collagen fibres and fragments of blood vessels.Approximately 90% of the neurones were viable, asdemonstrated by trypan blue exclusion.d. Tissue CultureThe final fraction was again centrifuged at 200 x g for5—10 mm to concentrate the cells and resuspended in growthmedium at a density of approximately 1x106 cells/mi andplated on 12 well plates coated with rat tail collagen atimi/well. Single cells were counted and numbers wereestimated for clusters of cells in order to obtain totalcounts. The growth medium consisted of Dulbecco’s ModifiedEagl&s Medium (DMEM) containing 5.5 mM glucose andsupplemented with 20 mM HEPES, 2 mM glutamine, 200 mM97Figure 18. Diagram of elutriator rotor and chamber.Variations in flow rates and rpm allow cells to be separatedon the basis of their volume.98Vewh’g PortISrobetampSamplelnjCCtiOnPressureGaugcBubble TrapBuffer PumpReservoirStandard Elutriacion Chamber‘eb elutriation Iboundaryeb -andrmmCentrlugalForceAxis of Ret3tionII99cytosine i3—D-arabinofuranoside, 8 tLg/ml insulin, 20 ng/mlnerve growth factor CS-7S, 100 Jhg/ml gentamicin, 1 jg/m1hydrocortisone, 4 ug/ml fungizone and 5% fetal calf serum.The cells were maintained in culture for 72 h at 37°C afterwhich immunocytochemistry (ICC), release experiments,extraction and HPLC were carried out.e. Somatostatin Secretioni. General Protocol.The cells were washed with 1 ml of release medium whichconsisted of DMEM containing 5,5 inN glucose, 1.0% aprotininand 0.1% BSA. A 585 jl aliquot of release medium and 15 ,tlof drug or peptide solution was added to each well (totalvolume = 600 J.Ll). The drug or peptide solutions wereprepared at 40 times the desired final concentration indimethyl sulfoxide (DMSO) and release medium. The highestratio of DMSO to release medium was 1:400 and at thisconcentration the release of SS—IR was not affected. Thecells were incubated for 120 mm, after which the releasemedium was removed, centrifuged to remove any particulatematter, and stored at —70°C for radioimmunoassay (RIA).This procedure was used for 13—PMA, 4a—phorbol and allpeptides. Comparisons were carried out using the samepreparation i.e. in paired experiments.100ii. High potassium.Experiments were carried out using release mediumaugmented with potassium chloride (KC1) in order to obtain apotassium concentration of 10 mM. This resulted in adepolarization of approximately 20 my, as determined by theGoldman equation. The osmolality of the release medium was346 mOsm/kg and 337 mOsm/kg, with and without the additionalKC1, respectively, and therefore no adjustment to thecomposition of the medium was made.iii. Time—courseExperiments were carried out to determine the timecourse of SS—IR secretion in response to i3—PMA (10—6 M, 10mM KCL). 13-’PMA was incubated in separate sets of wells for15, 30, 60 and 120 mm,f. Somatostatin Extraction.Each well was extracted to determine the content of SS—IR per well and the variability of SS—IR between wells(n=8). The isolated ganglia were dislodged from the bottomof each well using a rubber spatula after the addition of600 j.l of 2N acetic acid and boiled for 15 mm. Theextracts were centrifuged and the supernatant was frozen at-70°C for RIA and HPLC analysis.101g. Somatostatin RecoveryThe recovery of SS—IR was determined in two ways.Firstly, SS (500 pg) was added to release medium andincubated with submucosal cultures for 2 h. Secondly, SS(500 pg) was dissolved in 2 N acetic acid, boiled andlyophilized. SS-IR was then measured by RIA,h. Characterization of Primary Molecular Forms.i Sample PreparationExtracts and release samples were added to a WatersSep-pak C18 column that had been primed by the addition of 5mls of 100% acetonitrile (ACN) containing 0.1%trifluoroacetic acid (TFA) followed by 5 ml of 100% dH2Owith 0.1% TFA and dried with 5 ml of air. The sample andextracts were added to the column, washed with 1.5 ml ofdH2O and eluted with 1.5 ml of 70% ACN containing 0.1% TFA,The samples were gassed with 100% nitrogen to remove theACN, frozen at -70°C and lyophilized.ii, Reverse Phase HPLC.The samples were reconstituted in distilled water forsubsequent reverse phase HPLC. Reverse phase HPLC wascarried out on Waters equipment consisting of a 3.9 x 30 mm102-Bondapak C18 column, a model 512 WISP, two model 510 pumpsand a model 441 absorbance detector. A gradient of ACN (28-35%) in 0.1% TFA run over a 10 minute period was used at aflow rate of 1 ml/min and 0.5 ml fractions were collected.The fractions were lyophilized, reconstituted in assaybuffer and SS-IR was measured by RIA. Synthetic SS-14 andSS—28 were used for calibration of the column.2. Dog Experimentsa. Procurement of TissueMature mongrel dogs were sedated with phentanoltriperidol (0.1 mi/kg) given with atropine (0.05 mg/kg),anaesthetized with sodium pentobarbital ( 30 mg/kg) Lv, andprepared for abdominal surgery. The animal was bled fromthe abdominal vena cava, the upper small intestine wasremoved and immediately placed in a container of ice—coldHank’s buffer.b. Isolation of Submucosal GangliaThe initial preparation of the tissue was similar tothat which was carried out with the human small intestine.The isolation was carried out in two 1 h stages. First, thetissue was incubated for 60 mm in incubation mediumcontaining 300 U/mi collagenase (type XI, Sigma). The103partly digested tissue was washed and centrifuged for 5-10mm at 200 x g to remove collagen, fat and debris. Thetissue was further digested for 60 mm followed by washingand centrifugation. The suspension was then filteredthrough a 240 p. Nytex mesh and resuspended in HBSS.c. Elutriation CentrifugationThe cell suspension was loaded into an elutriator rotor(Beckman) at 2500 rpm at a flow rate of 25 ml/min. Afraction was collected at 2200 rpm at a flow rate of 35mi/mm which contained fibroblasts, red blood cells and cellfragments. A second fraction was collected at 1600 rpm anda flow rate of 100 ml/min which contained single ganglia andclusters of two or three cells as well as undigestedcollagen fibres and fragments of blood vessels.Approximately 90% of the neurones were viable asdemonstrated by trypan blue exclusion. In severalexperiments, a third fraction which contained large groupsof cells, fragments of blood vessels and collagen fibres wascollected at 800 rpm and a flow rate of > 100 mi/mm.d0 Tissue CultureThe procedure for tissue culture of canine submucosalneurones was identical to that for human neurones.104e. SS—IR and SP-IR Secretioni. General ProtocolThe cells were washed with 1 ml of release medium whichconsisted of DMEM containing 5.5 itiN glucose, 1.0% aprotininand 0.1% BSA. A 1 ml aliquot of release medium and 10 /hl ofdrug or peptide were added to each well. Drugs or peptideswere prepared at 100 times the desired final concentrationin distilled water. The cells were incubated for 45 mmafter which the release medium was removed, pipetted intoEppendorf tubes containing 110 jl concentrated HC1,centrifuged to remove any particulate matter and stored at70°C for RIA. Comparisons were carried out using the samepreparation i.e. in paired experiments.ii. Acetic Acid ExtractionCanine submucosal cells were extracted to determine thetotal cell content of SS-IR and SP-IR in the same manner asSSIR from human submucosa.iii. Sep—Pak Extraction and ConcentrationSamples were extracted as previously described (Kwok andMcIntosh, 1990). Release and acetic acid extract sampleswere applied to Sep Pak C18 cartridges which had been primed105with 10 ml of acetonitrile containing 0.1% trifluoroaceticacid (TFA), followed by 10 ml dH2O and 10 ml 1% BSA. Thecartridge was then washed with 10 ml dH2O and 1 ml 20%acetonitrile containing 0.1% TFA. Two ml of 50%acetonitrile containing 0.1% TFA were added to the column toelute the peptides. The samples were lyophilized using aspeed vac, and stored at —70°C for RIA. Each Sep Pakcartridge was used no more than twice, a procedure which hasbeen shown not to alter the recovery of substance P andsomatostatin.iv. SP-IR RecoveryKnown amounts of SP-IR (500 or 1000 pg) were added torelease medium and incubated with canine submucosal cellsfor 1 h in wells which either did or did not contain theenzyme inhibitors captopril (20 jM) and bacitracin (20 ,.LN).3. RadioinuuunoassayRadioimmunoassay (RIA) was used to measure SS—IR and SP—IR in release medium and cell extracts of submucosalcultures, and the techniques for each have been previouslydescribed (Kwok and McIntosh, 1990; McIntosh et al, 1987b).106a. Somatostatini. Assay bufferSodium barbital (4.90 g), sodium acetate (0.32 g) andethylmercurithiosalicyclic acid sodium salt (inerthiolate;0.10 g) were dissolved in 700 ml dH2O. The pH was adjustedto 7.4 with HC1, and this stock buffer was stored at 4°C.Aprotinin (trasylol; 100 K.I.U.) and BSA (Pentex; 5.0 g/l)were added to the stock buffer for preparation of assaybuffer.ii. AntibodyThe SS-IR was assayed in duplicate using a monoclonalantibody (SOMA 3) to somatostatin (Buchan et al, 1985) whichdetects both SS-14 and SS28 (McIntosh, 1987). The antibodywas prepared from crude mouse ascites as previouslydescribed and kept at 4°C until use. This stock solutionwas diluted with assay buffer to obtain a final titre of4x106 for the assay. The antibody has been shown to notcross—react with GIP, motilin, gastrin or substance P.ui0 StandardsSynthetic cyclic somatostatin was dissolved in 0.1 Macetic acid, diluted to 100 J.Lg/ml using dH2O containing1070.05% BSA (Pentex) and aliquots of 50 jl (5 JLg) werelyophilized and stored at -20°C. An aliquot was dissolvedin 100 ji of dH2O, and serially diluted in assay buffer onthe day of the assay to obtain standards ranging from 3.9 to500 pg/mi.iv. Preparation of 125-Soiuatostatin.Synthetic Tyr1-somatostatin was iodinated using thechloramine-T method, purified initially by adsorption tosilica and lyophilized in aliquots of 1x106 cpm. Aliquotswere further purified on the day of the assay using a CM-52Sephadex column previously equilibrated with 0.002 Mammonium acetate. An aliquot of‘251—somatostatin wasdissolved in 0.002 M ammonium acetate, applied to the columnand eluted using 0.2 N ammonium acetate at a flow rate of 1ml/min. One or two peak fractions were counted, neutralizedwith 2 N NaOH and diluted to 30003500 cpm/100 Jhl in assaybuffer.v. SeparationActivated charcoal (1.25%) and dextran (0.25%) weredissolved in 005 N phosphate buffer and this mixture wasstirred for at least one h after the addition of 0.1%charcoal—extracted plasma (CEP).108vi. Assay Protocol (Table 8)Assays were carried out on a refrigerated Table,maintained at approximately 4°C, using 12 x 75 borosilicateglass tubes. Total count, non—specific binding (NSB), zerobinding and standard tubes were assayed in triplicate, whilesamples were assayed in dup1icate If necessary, sampleswere diluted with assay buffer so that concentrations fellwithin the most sensitive portion of the standard curve.For release experiments, 100 jl of release medium was addedto NSB, zero and standard tubes in place of assay bufferAfter a 72 h period of equilibration at 4°C, separationof bound and free peptide was carried out by adding 1 ml ofcharcoal slurry to all except “total count” tubes. Tubeswere vortexed and centrifuged at 3000 rpm for 30 mm at 4°C.Tubes were decanted, dried over absorbant paper and countedon a gamma counter. The remaining pellet contained freeiodinated and cold peptide.109Table 8. Somatostatin Assay Protocolsample totalTUBES NSB zero orstandard count___________________________F____________________________buffer 300 j.Ll 200 l 100 j.Llstandardor — 100 Llsampleantibody——— 100 j.Ll 100 jllabel 100 J.Ll 100 jl 100 jl 100 1total 400 J.Ll 400 l 400 JA1 100 JhlvolumeV 110vii. Calculation and Presentation of RIA Data% bound = 100 X (C NSB - C sample)C totalwhere C = cpm% NSB = 100 X (C total C NSB)C totalA standard curve was plotted of % bound versusEsomatostatin] on semilogarithmic paper and/or wastransformed into a logit—log plot using a RIA softwareprogram (RIA Analysis v 1.0). Sample concentrations weredetermined from the original standard curve or by the RIAprogram using the mean of duplicate counts.viii. Inter— and Intra-Assay VariationSamples containing known amounts of somatostatin couldnot be kept for long periods of time because of a loss inimmunoreactivity and therefore were not used as controls.Inter—assay variation was determined by comparing multiplestandard curves while intra—assay variation was determinedby comparing 55—IR standards randomly placed within theassay. Both inter—assay and intra—assay were less than 10%,111ix. Testing pH Effects.Acetic acid (2N) was normally used to extract SS-IR fromtissue and cultures and samples were diluted 1:20 and 1:40for measurement in RIA. Therefore, the effects of addingacetic acid alone (2N, diluted 1:20 and 1:40) to the SS-IRstandard curve were examined. The plots of % bound versus[somatostatin] showed that the “zero binding” was reducedfrom 50% to 36%, and the linear portion of the curve wasreduced when the 1:40 dilution was used in the standardcurve (figure 19). In addition, the nonspecific binding ofthe label was increased when the 1:40 dilution was used.The standard curve could be corrected when the pH of the1:20 and 1:40 dilutions of 2N acetic acid was adjusted to7.4 with sodium hydroxide. It should be noted that thestandard curve produced using the 1:40 dilution without thepH adjustment could be utilized to determine sample SS-IRconcentrations. Interestingly, linear logit—log plots ofthe 1:40 undiluted, and the ‘normal’ standard curves weremore similar than the plots of % bound vs. concentration(figure 20). This illustrates the inability of lineartransformations of assay data to assess the usefulness of astandard curve.112Figure 19. Somatostatin standard curves (% bound vsconcentration of somatostatin), without acetic acid, withacetic acid (1:40 dilution) and with acetic acid (1:40 and1:20 dilution) where pH was adjusted with NaOH.—S00x0113o Std. Curveo 1/40 dilution1/20 dilution pH adjusted with NaOH1/40 dilution pH adjusted with NoCH50403020100—3.9 500[somatostatin] (pg/mi)114Figure 20. Somatostatin standard curves (logit % bound vsconcentration of somatostatin), without acetic acid, withacetic acid (1:40 dilution) and with acetic acid (1:40 and1:20 dilution) pH adjusted with NaOH0[somatostatin] (pg/mi)11a6o Std. Curve• 1/40 dilution1/20 dilution pH adjusted with NaCH1/40 dilution pH adjusted with NaOH765,__4,02‘ 10—1•—3,039I I I I500116b. Substance PThe substance P RIA has been described previously (Kwokand McIntosh, 1990) and was identical to that forsomatostatin with the following differences.i. Assay bufferGelatin was used in place of BSA in the assay buffer0ii. AntiserumThe antiserum KGPO5 was raised in guinea pig usinghaemocyanin-conjugated SP and was used at a final dilutionof 1:180,000. The antiserum has been shown to cross—react100% with SP, SP-(3--l1) and physalaemin, < 0.3% withkassinin, < 0.07% with SP methyl ester, < 0.04% with SP—(5—11), SP—(7—11) and eledoisin, < 0.009% with SP-(1—4), SP—(l-7), SP—free acid, neurokinin A, neurokinin B, bombesin andsomatostatin.iii. Assay protocol125-Substance P was added to each RIA tube 2-3 h afterthe addition of the antiserum. Also, polypropylene(plastic) tubes were used rather than borosilicate tubes117since the latter were shown to result in a loss ofmeasurable SP-IR due to the adsorption of peptides to glass.iv. CalculationsThe RIA data were analyzed in the same way as forsomatostatin.4. Data AnalysisThe total amount of SSIR or SPIR per well wasdetermined by adding the amounts extracted and released.Peptide release was calculated as a percent of total cellcontent (%TCC) in the following way:% TCC = [r/(e + r]) x 100where r = amount of SS-IR releasede = amount of SS—IR extracted(e+r) = total SS—IR per wellAll values are given as means ± SE and n always refersto the number of donors/dogs. The Mann Whitney U analysiswas used for statistical comparisons of secretion data anddifferences were considered significant at the p<0.05 level.To test human extract data, a one way ANOVA wasutilized.118119C. Results1. Isolation and Characterization of Submucosal Culturesa. HumanThe isolation protocol described produced the highestyield of viable cells. Using less collagenase (300 U/ml)and increasing the time of digestion always resulted in moreundigested tissue and a smaller yield of viable cells.Increasing the collagenase (1000 U/nil) and decreasing thetime also resulted in a lower yield of viable cells.Maintaining the pH below 7.4 reduced the reaggregation andgelation of collagen. Likewise, several washing andcentrifugation steps were required to remove digestedcollagen and reduce gelation. The post-collagenase digestconsisted of single cells, clusters of cells, fragments ofblood vessels, red blood cells (RBCs) and satellite cells.Following filtration and elutriation, the suspensioncontained no RBC5, fewer fragments, fewer single cells andless debris but individual ganglia could be seen clearlyunder the microscopeCells adhered to the collagen substrate overnight andtheir viability in culture remained > 90% for up to 5 daysas shown by trypan blue exclusion conducted throughout thisperiod. After 5 days, the viability decreased and the cells120became detached from the collagen. Cells aggregated aroundindividual ganglia and there was abundant neurite outgrowthafter 72 h in culture. The individual clusters of cellswere linked by neurite extensions and resembled thesubmucosal plexus in situ (see canine isolation).Therefore, ICC, release experiments and HPLC extractionswere carried out at this time point. An initial platingdensity of 1-2x106 cells/mi/plate was chosen and was foundto be optimal for the survival of the neurones for 72 h. Atdensities of 5—8 x io cells/mi, the majority of cells wouldnot adhere while at densities of 3—5 x io cells/mi thecells would detach from the collagen substrate after 1-3days.The mitotic inhibitor cytosine arabinoside was includedin the growth medium and effectively prevented theovergrowth of fibroblasts. A sheet of fibroblasts tosupport the attachment of neurones was not required with theuse of plates coated with rat tail collagen.Hematoxylin staining of sections of stripped submucosaconfirmed that all mucosal and muscle tissue had beenremoved (not shown). Human neurones were phase bright,sprouted neurites which contained varicosities along theirlength and made anatomical connections to other cells after72h.121Human cultures contained neurones which stained for SSIR (figure 21 a, b). In human cultures, 35 % of all cellsper cluster contained SS—IR in culture, which was similar tothe amount found in tissue sections Positively stainedfibres were abundant and varicosities present along thelength of the fibres also contained SS—IR (figure 22 a, b,c). Submucosal neurones and fibres also contained SPIR andVIP—IR (figure 23 a, b).Myenteric cultures also contained neurones which stainedfor SS—IR (figure 24), as well as SPIR and VIPIR neurones0These neurones sprouted extensive neurites with varicositieswhich contained SS-IR, SP-IR and VIPIR..b DogThe isolation and culture of canine SMP differed fromthat of human tissue in several ways. The amount ofcollagenase and overall incubation time required to dispersethe submucosa were less than for the human submucosa. Thiswas due to a lesser amount of collagen contained in thecanine submucosa.122Figure 21 a and b High magnification micrograph ofcultures of human duodenal neurones, grown on coverslips,stained for SS—IR.x 5OO124Figure 22. Cultures of human duodenal SMP stained for SS—IR. Note the presence of IR neurones (large arrows) andvaricosities (small arrows).a) x 100b) x 200c) x 200.It-I-I*efr,.frA-fr44.a126Figure 23. limaunostained cultures of human duodenal SMP.a) SP—IR neurone (arrow).x 200b) VIP’IR neurones (arrow) and varicosities (small arrows)x200,-*•i‘4.128Figure 24. Cultures of human duodenal MYP stained for SSIR. Note the presence of IR—neurones (large arrows) andvaricosities (small arrows).a) x 100b) x 200,—‘4A4VCDI.0130Three different protocols were tested for elutriation ofcanine tissue. Fraction 1 contained primarily single cells,including neurones and red blood cells. Fraction 2contained mainly ganglia and groups of 2—5 cells. Fraction3 was eluted according to the method used for humansubmucosa, and contained large fragments and large groups ofcells. The viability of all fractions was > 95% immediatelyafter elutriation, but fractions 2 and 3 contained the cellsof interest. The elutriation procedure for fraction 2 waschosen for two reasons. First, only single ganglia werepresent in fraction 2, similar to the situation in the humancultures. Second, the cells in fraction 3 would flatten andsprout neurites but became detached from the collagensubstrate after 2-3 days.An examination of the time—course and progression ofshort—term cultures is shown in figure 25. Clusters werephase bright, and attached to the plates after 24 h inculture but no sprouting was observed (figure 25 a). After72 h in culture, ganglionic structures formed and madeinterconnections (figure 25 b). After 120 h, many of thecells in the ganglionic structures were dead (figure 25 c).Therefore, 72 hwas selected as the time when ICC andsecretion experiments were carried out131Figure 25. A time—course of the attachment and progressionin short—term cultures of canine SMP neurones,a) Day 1, 24 h after plating neurones clusters attached(arrow) but no sprouting of neurones was observed x 200.b) Day 3, 72 h after plating, neuronal clusters formedganglionic type structures (arrow) with interconnections x2OO. c) Day 5, 120 h after plating, the majority of thecells in the ganglion—like structures were deadx 200.132‘33£I.V 134Phase contrast microscopy revealed that the canineneurones were phase bright, sprouted neurites whichcontained varicosities along their length and madeanatomical connections to other cells after 72 h (figure26). Canine cultures contained neurones which stained forSS—IR (figure 27 a, b), SP—IR (figure 27 c) and VIPIR(figure 27 d, figure 28). Positively stained fibres wereobserved and varicosities present along the length of thefibres also contained SSIR and SPIR.Double stains of human cultures demonstrated colocalization of SS—IR and SP-IR (figure 29 a,b). Doublestains of the canine cultures revealed separate populationsof SS-IR and SP-IR neurones (figure 29 c,d) Thus, theneurones in culture displayed the same phenotype as neuronesin situ with respect to SPIR and SSIR. Interestingly, thecanine neurones also demonstrated the segregation ofneuronal types normally observed in neuronal ganglia insitu.135Figure 26. Phase contrast micrograph of canine submucosalneurones. Note the presence of phase bright neuronal cellbodies (large arrows) and extensive sprouting of neuritescontaining varicosities along their length (small arrows)x 200.Cl137Figure 27. Immunostained cultures of canine SMP.a) SS—IR neurones x 200b) SS-IR neurones x 200C) SP—IR neurones x 200d) VIP-IR neurones x 200.Note the extensive sprouting of neurites.138w‘ab139‘C140Figure 28. Cultures of canine SMP neurones stained for VIPIR (arrows), which are located within ganglion-like clustersof cellsx 100142Figure 29. Double stains of human and canine cultures ofSMP for SP—IR (Rhodamine) and SS—IR (FITC).a) Human SP—IR neurones and b) SS—IR neurones (largearrows) x 2OO, Note the co—localization of SS—IR and SP-IRc) Canine SP—IR neurones (large arrows) and d) SS—IRneurones (small arrows) x 100. Note the lack of co—localization. Also, note the grouping of each type ofneurone reminiscent of what is observed in situ.1442. Somatostatin Secretion from Human Submucosal Neurones.a. Effects of SecretagoguesThe addition of B—PMa at concentrations of io8, ioand M caused significant increases in the release ofSS—IR (figure 30). Increasing the potassium concentrationfrom 5 to 10 mM resulted in a further increase in the meanvalue of SS—IR released in response to i3—PMA, but this wasnot statistically significant (figure 31). The basal levelof SS-IR over the two hour time period was 16 ± 6 (n=6) and24 ± 4 (n=1l) pg/600l using release medium with and withoutadded KC1, respectively. Basal release of SS-IR could bemeasured only after 120 minutes. The variability of SS—IRrelease in response to 13—PMA and 10 mM KC1 was determined in3 wells and was found to be < 5% (n=3). Variation in basalrelease levels was found to be < 3% between wells (n=6).Of the total SS-IR released in response to 13’PMA (106M, 10 mM KC1) after 120 mm, 59% was present in the mediumafter 30 mm, and 80% was present after 60 mm (figure 32)(n=3)145Figure 30. Release of somatostatin immunoreactivity (SSIR)as a percent of total cell content (% TCC) from submucosalneurones in response to incubation with i3-PMA for 120 mm (n= 6 donors). “C” is the basal level of SSIR after 120 mm.Values are means ± S.E.C)C)j)146***302520151050C —8 —7log [fiPMA]V 147Figure 31. Release of SS—IR (% TCC) from submucosalneurones in response to incubation with 8—PHA for 120 mm in5 mM and 10 mM KC1 (n=6 donors). Values are means ± S.E.*Significantly different from basal release (p O.05)C)C))LI)148— 5mM K+10mM K+** ****151050C —8 —7 —6log [PMA]149Figure 32. Release of SS—IR (% TCC) from submucosalneurones in response to 8—PMA (io’6 M, 10 mM KC1) after 30,60 and 120 mm (n = 3 donors). Control values represent theamounts of SS—IR released in the absence of B—PMA over 120mm. Values are ± means S.E.1500C),(I)•=. flPMA (106 M, 10 mM KCL)0—0 control5040100 30 60 90 120 150Time (mm)151The addition of the inactive 4a-phorbol atconcentrations of io8, lO and io6 M did not causesignificant increases in the release of SS—IR (figure 33)(n=3)The addition of substance P did not cause any increasein SS—IR release in comparison to basal (figure 34). Theaddition of SP in the presence of hexamethonium (106 M) andatropine (106 M) did not cause any change in the SS-IRsecretion, but did result in more variation of the basallevels (figure 35). The addition of SP (1O N) did notaffect 13-PMA stimulated SS—IR release (figure 36).The effects of tetrodotoxin (TTX, l0’6 M) on 13-PMA (106M) —stimulated SS—IR release were examined in one experiment(figure 3 (figure 37). The secretion of SS—IR in responsewas partly attenuated by TTX.The effects of CGRP io6 N and methacholine 1o6 N onSS—IR release were examined in one experiment (figure 38).The secretion of SS-IR was stimulated by CGRP, but notaffected by methacholine. Methacholine attenuated CGRPinduced secretion of SS-IR.The calcium ionophore (10—6 N, 5 x 10-6 N) was tested intwo donors and it was found to not have any effect on SS—IRsecretion (Table 9).152Figure 33. Release of SS-IR (% TCC) from submucosalneurones in response to incubation with 13—PMA and theinactive 4a—phorbol for 120 mm (n = 3 donors). Values aremeans ± S.E.003(I)10153— 4phorboI50403020f3PMAC—8—7—6log [drug]154Figure 34 Release of SSIR (pg/dish) from submucosalneurones in response to incubation with substance P for 120mm (n = 4 donors) Values are means ± S,E.U,ri)U,15550403010C —9 —8 —7 —6Log [Substance P]156Figure 35. Release of SS—IR (pg/dish) from submucosalneurones in response to incubation with substance P for 120mm in the presence or absence of hexamethonium (hex) andatropine (atr) (n = 4 donors) Values are means ± S.E.UICLC’,175040302010Chex6+otr6—9 —8 —7 —6Log [Substance P]÷hex6 + atr6158Figure 36. Release of SS-IR (pg/dish) from submucosalneurones in response to incubation with i3-PMA for 120 mm inthe presence or absence of substance P (n = 3 donors).Values are means ± S.E,159350• 30025020015010050— PMAPMA + SP (1Q—7M)400U)(1)0C —9 —8 —7Log [PMA]160Figure 37. Release of SS—IR (pg/dish) from submucosalneurones in response to incubation with 13—PMA for 120 mm inthe presence or absence of TTX (n = 1 donor).16].900800700600500003002001000•-6 +TrX—6log [fiPMA]c -nx--6 —6Figure 38. Release of SS-IR (pg/dish) from submucosalneurones in response to incubation with CGRP andmethacholirie for 120 mm (n = 1 donor).162163cn+Methchoflne61110080604020C CGRP6 Methcholine6 CGRP6I()M0.I..)1Prtb.oI-C%0CDI-ILYISx PP00ICDo.CD I-h CD a 0OI-hcn 0Hc1 CDw0CDC)CDC)IIII-J—.CDC)‘t1.I-’.QI-a.I-a.-0wIJ.D‘JPO-CDCDCD0II0CD‘a 00I-C CD w I-a 0165The effects of carbachol (1o 10 M) were tested inone donor (figure 39) and it appeared to increase thesecretion of SS-IR.b. Somatostatin Content (Table 9)The amount of SS—IR per well was found to be 1990 ± 809per well; however the variability of SS-IR content betweenwells within each donor was found to be < 3% (n = 6 donors),(Table 10). Also, incubation with different concentrationsof 13-PMA, with or without increased KCL, did notsignificantly alter SS-IR content (n=6)c. Somatostatin Recovery.The recovery of SS—IR from release medium afterincubation with the cultures was greater than 95% and afterextraction with 2 N acetic acid was greater than 90%.d. Characterization of molecular forms.The majority (> 90%) of SS—IR in acid extracts of theneurones in culture and released in response to 13—PMA elutedwith the same retention time as synthetic SS-l4 on HPLC(figure 40, 41).166Figure 39. Release of SSIR (pg/dish) from submucosalneurones in response to incubation with 13—PHA and carbacholfor 120 mm (n = 1 donor)167400-PMACarbcchc. 300-Q200ci)10:og [dose]168Table 10. Variations in content of somatostatinimmunoreactivitySomatostatin content of extracts(pg/dish)control 3359 ± 144110 mM KC1 3131 ± 125313—PMA io6 M 3483 ± 1429M 3221 ± 1259io’8 M 3276 ± 13321o6 M, 10 inN KC1 3475 ± 1367H 3286 ± 1327io8 H 3307 ± 1444169Figure 4O HPLC profile of SS—IR released from subinucosalneurones in response to B—PMA (106 M). Sample representspooled medium from 6 wells SS—28 and SS-l4 markersindicate the elution position of synthetic peptide170SS—141500 351000ci)=C0C/) 500 28 -‘-C)- I00 10 15 20 25Elution Volume (ml)‘1’SS—28171Figure 41. HPLC profile of SS—IR contained in acetic acidextracts in response to B—PMA (106 M). Sample representspooled extracts from 2 control wells. SS—28 and SS-l4markers indicate the elution position of synthetic peptide.172SS—141500 351000(I) 500 28 S(J)Li ___•II_•i.r_I.•_II..I .Ii_IIi.10 5 10 15 20 25Elution Volume (ml)‘1’SS—28‘if1733. Somatostatin and Substance P Secretion from CanineSubmucosal Neurones.SS-IR and SP-IR were extracted according to the protocolused for SP-IR previously (Kwok and McIntosh, 1980) and weremeasurable by RIA. The total cell content of SSIR was 1200± 210 pg/well and for SP—IR was 810 ± 107 pg/well (n=3).The basal release of SP—IR was 4.5 ± 0.4 ( % TCC) and thatfor SS—IR was 8.1 ± 1.5 (% TCC) (n=3). The recovery of SSIR was found to be > 90% using this procedure. The recoveryof SPIR was > 95 % with or without the presence ofbacitracin and captopril. Depolarization by 55 mM K didnot increase SP—IR secretion (n=3).V 174D. DiscussionIn the present study, submucosal neurones from human andcanine small intestine were isolated and maintained intissue culture and the localization and release of SS—IRwere examined. A vigorous digestion procedure was requiredto obtain a high yield of neuronal ganglia from humantissue. Gelation and reaggregation of collagen occurred dueto the large amount of collagen present in the humansubmucosa. Dimeric collagen has been shown to undergo areversible gelation as temperature and pH increase (withsignificant changes in turbidity at pH=7.4 and T=28°C), theprocess also being dependent on the concentration of freecollagen (Yurchenko and Furthmayr, 1984). Therefore, the pHof the washing and incubation medium was kept below 7.4, andthe cells were washed and centrifuged to reduce theconcentration of collagen. If these procedures were notfollowed, the yield of viable neurones was insufficient toperform secretion studies. The method of short—term culturedescribed was adapted from previous work carried out usingcanine small intestine (Barber et al, 1986; Buchan et al,1989). In order to obtain neuronal ganglia, human tissuerequired a more vigourous digestion than canine tissue. Athicker submucosa in the human small intestine was theprimary reason for this difference.175The average yield of neurones was two 12 well plates ata concentration of 1-2 x 106 cells/well. This meant thatonly 12 conditions could be tested in duplicate for eachexperiment.The human and canine submucosal neurones attachedovernight, were phase bright as demonstrated by phasecontrast microscopy, sprouted processes which containedvaricosities along their length and made anatomicalconnections with other cells. This indicated that afterisolation and culture the neurones were viable and did notalter their normal morphologic phenotype. Time—coursestudies of canine cultures demonstrated that the cells inculture formed aggregates after 72 h in culture. This typeof aggregate formation has been observed previously inexplants of enteric neurones maintained in tissue culture(Jessen et al, 1983). Electron microscopic examination ofthe aggregates examined by these authors showed that thecellular contacts inside of them were similar to those seenin enteric ganglia in situ.The staining pattern of SS-IR in human culturedsubmucosal neurones was similar to that observed in tissuesections, suggesting that peptide localization was unchangedafter isolation and culture. Nerve fibres and varicositiespositively staining for SS-IR and SP-IR were commonlyobserved in the cultures. The more intense staining of SS—176IR fibres in culture probably relates to the lack ofobscuring collagen which surrounds the neurones in situ.Somatostatin and SP—IR were co—localized in human, butnot canine, submucosal neurones which parallels the locationof these neuropeptides in tissue sections0 Thus, theisolation of the ganglia and maintenance of submucosalneurones in tissue culture did not alter the expression ofSP—IR and SS—IR.The results presented have shown for the first time thatactivation of PKC by 13-PMA markedly stimulates the releaseof SS—IR from human enteric neurones0 Release of SS—IR byactivation of PKC using phorbol esters has been reportedwith dispersed cultures of fetal rat hypothalamus and cortex(Peterfreund and Vale, 1983) and canine jejunal submucosa(Buchan et al, 1989)The predominant molecular form of SS—IR present in acidextracts of the cultures was SS-14. This supports the workof other groups that SS—14 is the the major molecular formin acid extracts of human enteric nerves (Baldissera et al,1985; Keast et al, 1986; Penman et al, 1983) Furthermore,the predominant form of SS-IR released in response to 13—PMAwas also found to elute with the same retention time as SS—14 using HPLC. Central neurones have been shown to containand release SS-14 (Bonanno et al, 1988). Somatostatin177endocrine cells of the human gut have been shown to containpredominantly SS-28 (Baskin and Ensinck, 1984). The lackof measurable SS—28 in the human submucosal cultures furtherindicated the absence of endocrine cell contamination.The release of SSIR by B-PMA was not due to nonspecific membrane effects since the inactive 4a-phorbol didnot significantly alter basal release. The secretion of SS—IR was probably due to activation of protein kinase C(Nishizuka, 1986) although effects other than P1CC activationcannot be ruled out (Castagna et al, 1982).The release of SS-IR in response to i3PMA (106 M with10 mM KC1) reached a plateau over the incubation period sothat 59% of the SS-IR released after 120 minutes was presentafter 30 mm and 80% of the total SS—IR released was presentafter one hours Therefore, the rate of release of SS—IRdecreased after the first 30 minutes of the incubationperiod. The decrease in SS—IR release may have been due todepletion of SS-IR from the cells, down-regulation ofprotein kinase C or autocrine regulation by SS. Autocrineregulation of SS has been demonstrated in.the pancreas andstomach (McIntosh, 1985). Phorbol esters have been shown todown-regulate PKC activity involved in norepinephrinerelease from rat brain synaptosomes (Oda et al, 1991). Therelease period was extended to 120 mm to allow measurement178of basal SS—IR levels which were not detectable at earlierperiods.To examine whether neuronal depolarization would augmentbasal or phorbol ester-stimulated SS—IR secretion, theextracellular concentration of potassium was doubled to10mM, which should have caused a small but sustaineddepolarization of neurones, Basal release of SSIR was notincreased, implying that attenuation of membrane polarity isnot sufficient to generate release. However, there was anindication that the neurones were more sensitive tostimulation by 13-PMA after potassium depolarization althoughthis effect was not statistically significant. Furthermore,preliminary experiments have demonstrated that TTX onlypartly attenuated BPMA-stimulated release of SS-IR, whichsuggests a mechanism only partly dependent on membranedepolarization. Previous studies demonstrated that evenhigh levels of potassium (> 50 mM) did not evoke the releaseof vasoactive intestinal peptide or calcitonin gene—relatedpeptide from enteric nerves (Belai et al, 1987; Belai andBurnstock, 1988; Besson et al, 1983). The results presentedin this thesis imply that exocytosis of peptide-containingvesicles requires the activation of second messengercascades (e.g. protein kinase C activation) in addition tomembrane depolarization.179The release of SS—IR by B—PMA from human neurones wasnot affected by the presence of the calcium ionophoreA23187, unlike canine neurones in which stimulation of SSIRsecretion by 13—PMA occurred only in the presence of theionophore (Buchan et al, 1989) (see Table 11).Interestingly, the basal secretion of SS-IR in the dog was23 fold higher than that in the human cultures suggestingthat there was a tonic basal stimulation of SS-IR secretion.It is probable that due to this background stimulation, thecanine neurones required both activation of PKC (by iS-PHA)and influx of calcium (by A23187) to further stimulate SSIRrelease.The calcium ionophore on its own had no effect on SS--IRsecretion in human cultures as was the case in canineneurones (Buchan et al, 1990). The ability of the calciumionophore to stimulate the release of NT from caninecultures has suggested that different mechanisms must beinvolved in the release of different peptides,It is possible that I3-PMA affected the release of otherneurotransmitters present in the cultures which in turn mayhave altered the release of SS-IR. For example, SP has beenshown to release Ni from canine submucosal neurones (Barberet al, 1989). The release of SP by 13—PMA presumably wouldnot have affected the release of SS—IR, since the presentresults have shown that exogenous SP had no effect. The180release of other neuropeptides from similar cultures inresponse to 13—PMA remains to be determined.While SP—IR neurones were present in human submucosalcultures, the addition of exogenous SP did not inhibit SS-IRrelease. Substance P has been found to inhibit SS-IRrelease from neurones of canine submucosal cultures andendocrine cells of the perfused rat stomach (Buchan et al,1990; Kwok et al, 1988) (see Table 11). Conversely,substance P has been shown to stimulate the release of SS—IRfrom the hypothalamus and pancreas (Reichlin, 1981). Thevariation in SP effect and co—localization of SP—IR and SSIR observed in the human but not dog may reflect anunderlying difference in the regulation of SS-IR secretion.Different effects of SP, on canine versus humansubmucosal neurones, were probably due to a combination ofdirect and indirect effects on SS—IR containing neurones.Support for direct SP effects on submucosal neurones comesfrom experiments using isolated mucosa/submucosapreparations which have shown that SP—mediated increases inthe secretion of C1 ions were TTX—sensitive (Keast et al,1985; Perdue et al, 1987). In addition, substance P hasbeen shown to release neurotransmitters such asacetylcholine from guinea pig myenteric neurones (Wiley andOwang et al, 1987) and181Table 11. Similarities and differences in the secretion ofsomatostatin immunoreactivity from canine and humansubmucosal neurones* Dog Human13-PMA no effect increase13-PMA + A23187 increase # increaseSp decrease no effectSP + PMA no effect* Buchan et al, 1990# Data not shown182neurotensin from canine submucosal neurones (Barber et al,1989)The signal transduction mechanism for the SPmediatedslow EPSP in myenteric neurones was shown not to involvecAMP (Palmer et al, 1987) but substance P has been shown toincrease levels of intracellular calcium in dorsal hornneurones and pancreatic acinar cells (Womack et al, 1988;Gallacher et al, 1990). In both cell types, SP wassuggested to increase cytoplasmic calcium by activatingprotein kinase C. The activation of PKC has been shown tostimulate SS—IR secretion from a variety of neuronesincluding submucosal neurones in this study.Substance P has been shown to produce a slow EPSP inmyenteric nerves (Katayama et al, 1979; Willard, 1989) andsubmucosal nerves (Surprenant, 1984; Mihara et al, 1985).The electrophysiological effects of SP in combination withits stimulatory effects on PKC activation andneurotransmitter secretion suggest that a direct action ofSP on SS-IR containing neurones would have causedstimulation of SS-IR secretion, The data obtained in thepresent human study and the results previously shown in thedog do not support a direct action of SP on SSIR containingneurones.183An indirect action of SP caused by the concoimnitantrelease of another neurotransmitter, for example Ach, is amore probable explanation for the results observed in bothcanine and human neuronal cultures. The release of Ach fromenteric neurones by SP has been demonstrated (Keast et al,1985) and also provides a possible explanation fordifferences in the effects of SP on SS—IR secretion betweencanine and human cultures. The release of varioustransmitters suggests that SP may have both excitatory andinhibitory effects (see section below on effects of Ach)which could be why no SPeffeet was observed in the humancultures.Acetylcholine has been shown to excite enteric neurones(Wood, 1970; North and Nishi, 1976). These neuronesexhibited a postsynaptic fast nicotinic EPSP (Nishi andNorth, 1973; Hirst et al, 1974; Surprenant, 1984) and a slowEPSP by activation of a N1 receptor (North et al, 1985). Inaddition, Ach was shown to produce a presynaptic IPSP byactivation of a N2 receptor (North and Tokimasa, 1982) whichalso inhibits the release of Ach and non—cholinergicneurotransmitters (Morita et al, 1982). The effects of Achrelease on SS—IR secretion would then depend on the numberand distribution of different cholinoceptor types. Further,cholinergic effects would depend on the anatomicalrelationship between SS—IR and Ach containing neurones. Theresults of this study do not provide evidence for a184difference in cholinergic receptor number and distribution,or in Ach distribution between the canine and humanneurones. However, they have clearly demonstrated adifference in the anatomical relationship of SS—IR neuronesbetween these species.There was large variability in the basal SS-IR secretionof human submucosal neurones in the presence ofhexamethonium and atropine. Preliminary experiments havedemonstrated the release of SS—IR in response to themuscarinic agonist carbachol. However, the muscarinicagonist methacholine had no effect on its own and attenuatedi3-PMA stimulated release of SS-IR. These experiments serveto illustrate that the effect of endogenous Ach andexogenous muscarinic agonists was variable. Furtherexperiments using cholinergic agonists and antagonists arerequired to determine the precise role of cholinergicneurones in the secretion of SS—IR from submucosal neurones.Immunohistochemical identification of neuronescontaining choline acetyl transferase (Chat) has not beencarried out in human or canine enteric neurones but studiesin the guinea pig indicated that at least 50 % of entericneurones were cholinergic (Furness et al, 1984; Steele etal, 1991). The prevalence of fast postsynaptic EPSPs(mediated by Ach) during.stimulation of presynaptic fibres(Wood, 1987) also argues in favor of a large population of185cholinergic enteric neurones. The lack of a suitable Chatantibody or antiserum has made it impractical to carry outstudies to examine the localization of this enzyme insubmucosal cultures.The levels of added SP or SS, or secreted SP—IR or SS-IRin the present study may have been altered as a consequenceof degradation by proteolytic enzymes, e.g. the lack of SPaction on SS—IR secretion from human submucosal neuronescould be attributed to the degradation of either added SPand/or secreted SS—IR. The recovery of > 95 % of exogenousSS—IR from release medium after incubation with human orcanine submucosal cells suggested that significantimiuunoreactive SS was not lost during the 2 h incubationperiod. Substance P recovery was also found to be completewith or without the addition of bacitracin and captopril.These results indicate that minimal levels of proteolyticdegradation occurred in the submucosal cultures. Thepeptidases responsible for degradation of SP and SS havebeen found in or on glia (Bunnett, 1987; Lentzen et al,1983) and endothelial cells (Defendini et al, 1983; Takadaet al, 1982). Endothelial cells were not present in thecultures while the amount of qua was small due to thepresence of the mitotic inhibitor cytosine arabinoside. Thebasolateral membrane of enterocytes has been suggested to bea primary region for the degradation of both SS-28 and SS-14(Weber et al, 1986) and it is probable that particular186capillary beds and circulatory enzymes contribute to thedegradation of neuropeptides since their half-life in bloodis short (Bunnett, 1987). These results have suggested thatthe use of dispersed neuronal cultures significantly reducedthe levels of proteolytic enzymes Previous studies havereported a lack of degradation of SS-IR in cultured ratbrain neurones (Lucius and Mentlein, 1991) and neurotensinin cultured canine submucosal neurones (Barber et al, 1989).Isolation methods utilizing collagenase have been shownto specifically damage muscarinic receptors on neurones andvarious cell separation methods using hypertonic solutionshave been implicated in general receptor damage (Guarnieriet al, 1975). However, the lack of SP effect on SS-IRrelease from human cultures was not likely due to receptordamage for several reasons. Receptor damage due tohyperosmolality was avoided since cell separation in theseexperiments was carried out using elutriation which permitsthe use of isotonic solutions (Meinstrich, 1983; Guarnieriet al, 1975). Once the cells have been isolated, thereceptors would have regenerated in culture, as has beenshown for the nicotinic acetylcholine receptor (Hartzell etal, 1973). Further support comes from experiments usingcanine submucosal neurones which have been isolated andmaintained in tissue culture in a similar fashion and wereable to respond to receptor dependent secretagogues such asSP (Barber et al, 1989; Buchan et al, 1989). Finally,187preliminary experiments using CGRP and carbachol have shownincreases in SS—IR secretion suggesting, but not confirming,the presence of functional receptors on the neurones.As mentioned, the neurones in tissue culture sproutedprocesses which were associated with varicosities alongtheir length. In addition, the proportion of human neuronescontaining SS—IR was similar to that in tissue sections.These results have suggested that the culture and isolationprocedures did not alter their normal morphologic andimmunocytochemical phenotype. Neurones dissociated from ratMYP have also been shown to retain morphological andimmunocytochemical properties after having grown in culturefor 4—8 weeks (Nishi and Willard, 1985). These authors havealso demonstrated that dissociated myenteric neurones invitro retain normal pharmacological and electrophysiologicalproperties (Willard and Nishi, 1985a; 1985b).‘-p188E. SummaryThese data have demonstrated that neurones of the SMP ofthe human small intestine can be isolated and maintained intissue culture for 72 h. The neurones were viable andsprouted neurites which contained varicosities along theirlength and which made anatomical connections with othercells. This indicated that the neurones retained a ‘normal’morphologic phenotype in culture. The cultures containedSS—IR, SP—IR and VIP—IR which were present in cell bodies,neurites and varicosities. The proportion of SS—IR neuroneswas similar in culture and in situ and these neuronescontained SP-IR which reflected the co—localization observedin situ. This indicated that the isolation and cultureconditions did not alter the expression of theseneuropeptides. A major advantage of these cultures was thatthere were no SS—IR endocrine cells present and thereforeexperiments were carried out which examined SS—IR secretionfrom neurones exclusively. The presence of varicositiessuggested that the release of SS-IR was regulated, sinceregulated peptide secretion from unmyelinated nerve fibreshas been shown to occur at this level. The secretion of SS—IR from these neurones was stimulated by 13—PMA activation ofPKC in a time-dependent fashion. Somatostatin—14 was thepredominant form of SS—IR present in the neurones andreleased by 13-PMA. Potassium depolarization had no effecton SS—IR secretion but seemed to make the neurones more189sensitive to stimulation by 13-PMA, although this was notstatistically significant. Preliminary experimentssuggested that TTX could attenuate the 8-PMA-stimulatedsecretion of SS-IR. This would further support thecontention that membrane depolarization was only partlyresponsible for the secretion of SS—IR from these neuronesin response to activation of PKC. Further experiments arerequired to determine the role of membrane depolarization inthe secretion of SS-IR. Unlike the situation in the dog,the calcium ionophore A23187 was not required to elicit thesecretion of SS—IR in the human but, like the dog, theionophore on its own did not release SSIR. The inhibitoryeffect of SP on basal SS—IR secretion observed in caninecultures (Buchan et al, 1990) did not occur in the humancultures. In addition, SP did not have an effect on 13-PHA--stimulated SS—IR release. The difference in SP effectbetween the canine and human cultures is correlated to thedifference in the co-localization of SP-IR and SS-IR betweencanine and human submucosal neurones . Preliminaryexperiments suggested that cholinergic agonists have mixedactions on the secretion of SS—IR from these neurones. Thestimulation of SS-IR secretion by CGRP and carbacholindicated that the neurones were responsive to receptordependent stimulation in addition to receptor independentstimulation by i3-PMA.190Human myenteric cultures exhibited characteristicssimilar to those observed in submucosal cultures. Neurones,neurites and varicosities contained SS-IR, SP-IR and VIP’IR0The canine cultures exhibited similar characteristics tohuman cultures. These cultures contained SP-IR and SS—IR asdemonstrated by RIA, and were present in distinct neuronesas demonstrated by ICC, which reflected their localizationin situ. The release of SPIR was measurable but was notstimulated by potassium depolarization0 The recovery of SPIR incubated with canine cultures was close to 100 % and wasnot affected by the enzyme inhibitors bacitracin orcaptopril.191V. GENERAL SUMMARY AND CONCLUSIONSHypotheses 1.. That interspecies variations in neuropeptidelocalization and enteric neuronal morphology exist betweenthe human and canine small intestine.A. Morphological Data from the Small Intestine1. The human and canine small intestine contain twodifferent sub—groups of submucosal ganglia based onmorphological examination. One group was nearer the mucosaand the other was nearer the circular muscle. There were nodifferences noted in their morphology or chemical coding.2. Canine and human submucosal ganglia contained neuronaltypes which were segregated into clusters of a predominanttype. In canine submucosa, large ganglia were composed ofseveral such clusters of different neuronal types. In humansubmucosa, the situation was more dramatic in that somesmall ganglia contained one neurone type exclusively. Theseresults have both developmental and functional implications.From a developmental standpoint, the existence of suchclusters suggests that neurones migrating into the entericplexi could be pre—prograimued to differentiate, for example,into SS—IR neurones rather than receive signals todifferentiate after arrival. The embryonic neurone afterarrival at a specific location would resume division to192produce clusters of a single type within the plexus.Alternatively, the existence of the clusters could resultfrom the parallel tracking of axons growing out of theembryonic ganglia. In this case, the phenotype would bedependent on the target innervated but all neurones withaxons in the same bundle would differentiate into the sametype.3. The canine, but not the human, small intestine has acharacteristic deep muscular plexus that was separate fromother sets of fibres in the circular muscle layer0 Thisimplies that neural control of circular muscle differsbetween the dog and human.4. Immunocytochemical studies have demonstrated, for thefirst time, that SP-IR and SS-IR are co-localized in thehuman, but not canine, enteric neurones of the smallintestine. Furthermore, canine. ganglia were shown to besubstantially larger than their human counterparts. Thedifference in chemical coding of neurones and the differencein ganglion size may be indicative of differences in thefunction of submucosal neurones.Conclusions.These data support the hypothesis that interspeciesvariations in neuropeptide localization and enteric neuronal193morphology exist between the human and canine smallintestine. In addition, these studies have shown that thereare differences between canine and human submucosal neuroneswith respect to ganglion size, chemical coding of submucosalneurones and the innervation of circular muscle. It hasbeen suggested that interspecies differences in gutmorphology are related to diet (Powell,1987; Gross, 1986)and size (Gabella, 1990). The interspecies differencesobserved in the present study support the contention thatanimals which have similar diets have similar entericganglia with respect to neuropeptide distribution andneuronal number. Moreover, the results have emphasized thatcomparisons of the results of physiological experimentscarried out in different species should be made withcaution.Rypothesis 2. That differences in neuronal chemical codingand morphology will be reflected in neuronal function0Hypothesis 3. That short-term cultures of human and canineSMP will provide a model system in which to examine suchdifferences0B. Culture Studies1. These experiments have demonstrated, for the first time,that adult human submucosal neurones can be isolated and194maintained in tissue culture. Cell cultures eliminate theproblems associated with peptides present in both submucosaland myenteric neurones, and enteroendocrine cells.2. The neurones sprouted neurites which containedvaricosities, suggesting that the secretion of neuropeptidesis from varicosities and can be regulated. This provides anadvantage over secretion studies carried out with acutelydissociated neurones which do not have neuronal fibres.Secretion from neurones which do not possess nerve fibressuggests that secretion of neuropeptides occurs from thecell body for which there is presently no evidence using invivo experimental techniques. Previous studies and thepresent evidence have suggested that SSIR secretion wasindependent of the cell body. Thus, the regulation of SSIRsecretion is suggested to be at the level of the varicosity.These data have also suggested that a small depolarizationrenders the varicosities more sensitive to a secretorystimulus.3. The difference in neuronal localization of SSIR andSP—IR between the canine and human small intestine alsosuggested a difference in function. Substance P did notinhibit the secretion of SSIR from human neurones as wasthe case in canine neurones (Buchan et al, 1990). Thissupports the hypothesis that the differences observed in SP-195IR and SS—IR localization are related to differences in SSIR secretion and SP actions.4. The phorbol ester, 13—PMA, was able to stimulate thesecretion of SS—IR from human neurones in the absence of theA23l87, unlike canine neurones which required the presenceof the calcium ionophore (Buchan et al, 1990). This furthersupports the hypothesis that canine and human submucosalneurones differ with respect to the regulation of SS-IRsecretion.5. The effects of CGRP and carbachol suggested thatreceptor dependent secretagogues were able to stimulate SSIR secretion. This suggested that functional receptors werepresent on the neurones after isolation and tissue culturebut further experiments are required to confirm thispossibilty.Conclusions.These results support the hypothesis that dispersedcultures of submucosal neurones are useful models to examineneuronal function. The cultures provided advantages over invicth experiments for studying the secretion of neural SS-IR.The regulation of SS-IR was shown to be different in canineand human submucosal neurones and this difference was196related to the difference in the localization of SSIR andSP-IR i.e. a difference in neuropeptide phenotype.197VI. CONCLUSIONS AND FUTURE DIRECTIONSA. Conclusions and SignificanceThe data presented and previous studies suggest that themorphology of the small intestine reflects diet i.e.structure and function are closely linked. The corollary tothis statement is that animals having similar diets havesimilar gastrointestinal tracts. The similarity instructure is illustrated by the correspondence in the smallintestine of omnivores (rat and human) compared to othergroups (e.g. ruminants). Genetic similarity does not seemto be a prerequisite except where it confers a preference indiet. An example of this is seen in two new world monkeyssimilar in size, the howler and spider monkeys (Milton,1986). Both species are plant eaters, but howler monkeyspass food through their digestive system at half the rate ofspider monkeys. This reflects the larger colon of thehowler monkey. Although the diets of both species are plantbased, the spider monkey eats mostly fruit and meets itsnutritional requirements by ingesting large volumes of food.The howler monkey eats less and is able to fermentquantities of plants present in the colon. There are manyexamples which show that species can rapidly respond tochanges in dietary quality by altering the features of thegut (Gross et al, 1986). Alterations in the gut no doubt198include changes in the morphology of enteric neurones andthe neurotransmitters they contain.The experiments presented in this thesis havedemonstrated that differences exist between canine and humansmall intestine with regard to the morphology of the SMP,the localization of SP, SS and VIP in enteric nerves and theactions of SP and P1CC—activation on enteric nerves.Differences in the secretion and absorption of electrolytesand in the control of motility between the canine and humansmall intestine (outlined in the general introduction) areprobably related to the differences observed in the presentexperiments. For example, SP increases the secretion ofelectrolytes in the small intestine by a mechanism which isTTX sensitive in some mammals but not the human (Hubel etal, 1984; Keast et al, 1987). It is possible that SP exertsits TTX sensitive effects by the indirect inhibition ofneural SS—IR release similar to that which is observed inthe canine small intestine.The necessity and usefulness of combining morphologicaland physiological experimental techniques to study thefunction of the small intestine was exemplified by theresults obtained in the present studies. Moreover, theyhave suggested possibilities for further examination of themorphology of the small intestine, characterization of theneuronal cultures themselves, differences in SP action and199SS—IR release from the cultures and regulation of otherneurotransmitters present in the cultures. Finally,functional studies of the small intestine can be carried outwhich examine whether the differences in morphologycorrelate with neurotransmitter regulationB. Future Directions1. Morphological StudiesThe differences in neuropeptide distribution and plexusmorphology between the human and canine enteric nervoussystem will probably be more extensive than reported in thisthesis. Further characterization of the distribution ofother known enteric neuropeptides will allow the neuronetypes to be identified in a manner similar to that which hasbeen carried out previously in the rat and guinea pig(Furness et al, 1989; Pataky et al, 1990).2. Functional StudiesWith the use of dispersed neuronal cultures of the SMP,it has been possible to study the secretion of a peptidewhich is produced by several cell types. It is possiblethat the regulation of SS-IR secretion from neurones isdifferent from that in endocrine cells0 Furthermore,differences in the modulation of secretion could reflect200different physiological roles for endocrine and neuronal SS-IR. Parallel studies of SS—IR secretion from endocrine celland neuronal cultures would provide an excellent modelsystem in which to define different release patterns.Although the data collected demonstrated bothstimulation and inhibition, a role for cholinergictransmission in neural SS—IR release from human SMP neuroneswas indicated. Further experiments are required toelucidate the precise actions of cholinergic agonists andantagonists on SSIR secretion. Testing the actions of thecholinergic agonists and antagonists on canine entericneurones would be useful for comparison to human neurones.Access to an antibody for Chat would permit thedetermination of the proportion of cholinergic neuronespresent enteric neurones in culture and in situ.The overall actions of SP on SSIR secretion weresuggested to be indirect, based on canine and humanexperiments. This hypothesis can be tested in the followingway. First, the effects of SP on cellular events, such asionic currents and calcium transients, can be examined inneurones known to contain SS-IR. Second, the effects of SPon the release of other neurotransmitters, such as Ach, canbe determined. Neurotransmitters can then be tested fortheir ability to affect the cellular events of SS-IRneurones.201The present study and previous studies (Buchan et al,1990) have suggested that somatostatin secretion from canineand human submucosal neurones is modulated differently bythe phorbol ester. The secretion of SS-IR can be furthercharacterized by examining the response to depolarization bylevels of potassium higher than those used in the presentexperiments. It should be noted that depolarization ofenteric neurones by potassium may not result in thesecretion of SS-IR. The lack of neurotransmitter release inresponse to high potassium has been noted in otherpreparations where it has been suggested that calcium entryis mediated by receptor-operated calcium channels (e.g.Belai et al, 1987).The cultures utilized in the present experiments wereshown to contain neurones which possessed a morphologicphenotype similar to that observed in situ, It would beuseful to compare ionic currents and pharmacologicalresponses of neurones to known neurotransmitters in order toconfirm the similarity of neurones maintained in tissueculture and in situ.The calcium dependence of SS-’IR secretion fromsubmucosal neurones is necessary to support its role as aneurotransmitter, The examination of calcium transients inSSIR neurones by the use of dyes such as Fura2 would be202necessary to confirm the calcium dependence of SS—IRsecretion.The actions of PKC stimulation on SS-IR secretionsuggests that neuropeptides which activate this pathway alsostimulate SS-IR secretion. Gastrin-releasing peptide is anexample of a neuropeptide, found in enteric neurones, whichactivates this pathway in a variety of cell types. Thepresence of this peptide in the human enteric nerves (Priceet al, 1984) suggests that it influences the secretion ofneural SS-IR.The complexity of the enteric nervous system is suchthat the studies outlined will provide information in only asmall fraction of the control systems. However,improvements in technology, such as calcium imaging systems,and experimental techniques, such as isolated cell culturesystems, would allow concepts of enteric nervous function tobe tested rigorously at the cellular level Together withexperiments carried out in the whole animal (e.g. thedietary habits of new world monkeys), the studies outlinedwill provide insight into the function of the smallintestine under different conditions.203VII. REFERENCESAhmad, 5., Allescher, H.D., Manaka, H., Manaka, Y., andDaniel, E.E. Biochemical studies on opioid and a2adrenergic receptors in canine submucosal neurones. Am. J.Physiol. 256: G957—965, 1989.Allen, J.M., Hughes, J.and Bloom, S.R. Neuropeptide Y inthe gastrointestinal tracts: distribution, chromatographicanalysis and pharmacology. Can. J. Physiol. Pharmacol.Gastrointestinal Hormones Symposium, Vancouver, Canada0 pp.101—102, 1986.Allescher, H.D., Ahmad, S.., Kostka, P., Kwan, C.Y. andDaniel, E.E. Distribution of opioid receptors in caninesmall intestine: implications for function, Am J. Physiol.256: G966—974, 1989.Andersson, P.—O., Blàom, S.R. and Jarhult, J. Colonic motorand vascular responses to pelvic nerve stimulation and theirrelation to local peptide release in the cat. J. Physiol.334: 293—307, 1983.Baldfssera, F.G.A., Holst, J.J., S.L. Jensen and T. Krarup.Distribution and molecular forms of peptides containingsomatostatin immunodeterminants in extracts from the entire204gastrointestinal tract of man and pig. Biochim. Biophys.Acta 838: 132—143, 1985.Bannerman, P.G., Mirsky, R. and Jessen, K.R. Analysis ofenteric neurones, glia and their interactions using explantcultures of the MYP. Dev. Neurosci. 9: 201—227, 1987.Barajos—Lopez, C., Berezin, I., Daniel, E.E. and Huizinga,J.D. Pacemaker activity recorded in interstitial cells ofCajal of the gastrointestinal tract. Am, J.. Physiol. 257:C830—835, 1989.Barber, D.L., A.M.J. Buchan, J.H. Walsh and A.H. Soll.Isolated canine ileal mucosal cells in short—term culture: amodel for study of neurotensin release. Am. J. Physiol.250: G374—G384, 1986.Barber, D.L., A.M.J. Buchan, S,E. Leeman and A.H. Soil.Canine enteric submucosal cultures: transmitter release fromneurotensin iixununoreactive neurons. Neuroscience 32: 245—253, 1989.Baskin, D.G. and Ensinck, J.W. Somatostatin in epithelialcells of intestinal mucosa is present primarily assomatostatin 28. Peptides 5: 615—621, 1984.205Belai, A. and G. Burnstock. Release of calcitonin gene—related peptide from rat enteric nerves is Ca2+ —dependentbut is not induced by K+ depolarization. Req. Pept. 23:227—235, 1988.Belai, A., V. Ralevic and G. Burnstock. VIP release fromenteric nerves is independent of extracellular calcium.Reg. Pept. 19: 79—89, 1987.Bell, J.D., I.L.O. Buxton, and L.L. Brunton. Enhancement ofadenylate cyclase activity in S49 lymphoma cells by phorbolesters. J. Biol. Chem, 260: 2625—2628, 1985.Berridge, N.J. The molecular basis of coimnunication withinthe cell. Scientific American, 1985.Besson, J., W. Rostene, A—N. Lhiaubet and G. Rosselin.Release of vasoactive intestinal peptide from rat jejuno—ileum in vitro. Effect of various depolarizing agents.Experentia. 39: 732—733, 1983.Binder, H.J., Laurensen, J.P, and Dobbins, J.W. Role ofopiate receptors in the regulation of enkephalin stimulationof active sodium chloride. Am. J. Physiol. 247: G432—436,1984.206Blumberg, P. M. Protein kinase C as the receptor for thephorbol ester tumour promoters: sixth Rhoads Memorial awardlecture. Cancer Res. 48: 1—8, 1981.Bonanno, G., Raiteri, N. and Emson, P.C. In vitro releaseof somatostatin from cerebral cortical slices:characterization of electrically evoked release, Brain Rest447: 92—97, 1988.Bornstein, J.C. Substance P binding sites in guinea pigenteric ganglia. In: Substance P and Neurokinins, eds0J.L. Henry, R. Couture, A.C. Cuello, G. Pelletier, R.Quirion, pp. 59—70, 1987.Bornstein, J.C. and Furness, JB. Correlatedelectrophysiological and histochemical studies ofsubmucosal neurones and their contribution to understandingneural circuits. J. Autonom. Nerv. Syst. 25: 1—13, 1988Brazeau, P., Vale, W., Burgus, R., Ling, N., Butcher, N.,Rivier, J. and Guillemin, R. Hypothalamic polypeptide thatinhibits the secretion of immunoreactive pituitary growthhormone. Science 179: 77—79, 1973.Brown, D . R., N. F. Overend and B. G. Treder. Neurohormonalregulation of ion transport in the porcine distal jejunum.207Actions of somatostatin—14 and its natural and synthetichomologs. J. PharmacoL Exp. Therapeut. 252: 126—134, 1989.Brown, J.C., Mutt, V. and Dryburgh, J. The furtherpurification of motilin, a gastric motor activitystimulating polypeptide from the mucosa of the smallintestine of hogs. Can. J. Physiol. Pharmacol. 49: 399-405, 1971.Buchan, A.M.J., MacLeod, M.D., Meloche, R.M. and Kwok, Y.N.Muscarininc regulation of somatostatin release from primarycultures of human antral epithelial cells. Pharmacology44: 33—40, 1992.Buchan, A.M.J., A.D. Doyle and E.A. Accili. Canine jejunalsubmucosa cultures: characterization and release of neuralsomatostatin. Can, J. Physiol. Pharmacol. 68: 705’710,1990.Buchan, A.M.J. Neurofilament-M and Calbindin D28K arepresent in two discrete populations of submucosal neurons.Brain Res. 538: 171—175, 1991.Buchan, A,M.J. and Barber, D.L. Neurotensin—containingneurones in the canine enteric innervation. Neurosci. Lett.76: 1317, 1987208Buibring, E., Lin, R.C.Y. and Schofield, G. Aninvestigation of the peristaltic reflex in relation toanatomical observations. Q.J. Exp. Physiol. 43: 26-37,1958.Bunnett, N. Post Secretory metabolism of peptides. Am. Rev.Respir. Dis, 136: S27—34, 1987.Cajal, S.R. Sur les ganglions et plexus nerveux del’intestin. C.R. Soc. Biol. (Paris), 45: 217—223, 1893.Carey, H.V., Cooke, H.J. and Zafirova, M. Mucosal responsesevoked by stimulation of ganglion cell somas in the SMP ofthe guinea pig ileum. J. Physiol. 364: 69—80, 1985.Castagna, M.Y. Takai, K. Kaibuci, K. Sano, U. Kikkawa and Y.Nishizuka. Direct activation of calcium—activated,phospholipid dependent protein kinase by tumor-promotingphorbol esters. J. Biol. Chem. 13:7847—7851, 1982,Cassuto, J., Siewert, A., Jodal, M. and Lundgren, 0. Theinvolvement of intramurals in cholera toxin—inducedintestinal secretion. Acta Physiol. Scand. 115: 157-158,1983.Chang, M.M.., Leeman, S.L., and Niall, H.D, Amino acidsequence of substance P. Nature New. Biol. 232:86-87, 1971209Christensen, J. and Rick, G.A. Intrinsic nerves in themammalian colon: confirmation of a plexus at the circularmuscle—submucosal interface. J. Autonom. Nerv. Syst. 21:223—231, 1987.Cooke, H.J. Role of the “little brain” in the gut in waterand electrolyte homeostasis. FASEB J. 3: 127—138, 1989.Costa, M., Furness, J,B. and Liewellyn—Smith, I.Histochemistry of the enteric nervous system. In:Physiology of the Gastrointestinal Tract, second edition,edited by Leonard R Johnson. Raven Press, New York, pp. 140, 1987.Couture, R., Mizrahi, J., Regoli, D. and Devroede, G.Peptides and the human colon.: an in vitro pharmacologicalapproach. Can J, Physiol. Pharmacol, 59: 957—64, 1981.Crema, A. On the polarity of the peristaltic reflex in thecolon. In: Smooth Muscle ed. Bulbring, E., Brading, A.F.,Jones, A.W. and Tomita, T. pp 542—548, 1970.Crowe, R., Kamm, M.A., Burnstock, G. and Lennard—Jones, J.E.Peptide—containing neurones in different regions of the SMPof human sigmoid colon, Gastroenterology 102: 461-467,1992.210Daniel, E.E., , Furness, J.B., Costa, M. and Belbeck, L.The projections of chemically identified nerve fibres incanine ileum. Cell Tiss. Res. 247: 377—384, 1987.Daniel, E,E., Costa, M., Furness, J.B. and Keast, J.R.Peptide neurons in the canine small intestine. J. Comp.Neurol. 237: 227—238, 1985.Defendini, R., Zimmerman, E.A., Weare, J.A., Aihene—Gelas,F. and Erdos, E.G. Angiotensin converting enzyme inepithelial and non—epithelial cells. Neuroendocrinology37: 32—40, 1983.Dockray, G. Physiology of enteric neuropeptides. In:Physiology of the Gastrointestinal Tract, Leonard Johnsoned. vol. 1, pp. 41—66, 1987.Donnerer, J. P. Hoizer and F. Lembeck. Release ofdynorphin, somatostatin and substance P from the vascularlyperfused small intestine of the guinea pig duringperistalsis. Br. J. Pharmac. 83: 919—925, 1984.Doran, J.F., Jackson, P, Kynoch, P.A.M. and Thompson, R.J.Isolation of PGP 9,5, a new human neurone—specific proteindetected by high-resolution two-dimensional electrophoresis.J. Neurochem. 40: 1542—1548, 1983.211Dubois, M.P. Inununoreactive so!natostatin is present indiscrete cells of the pancreas0 Proc0 Nati. Acad. Sci., USA72: 1340, 1975.du Vigneaud, V., Gish, D,T. and Katsoyannis, P.G. Asynthetic preparation possessing biological propertiesassociated with arginine—vasopressin. J. Amer. Chem. Soc.75: 4751—4752, 1953.du Vigneaud, V., Ressler, C., Swan, J.M., Roberts, C.W.,Katsoyannis, P,G. and Gordon, S. The synthesis of anoctapeptide amide with the hormonal activity of oxytocin.J. Amer. Chem. Soc. 76: 4879—4880, 1954.Ekblad, E., Ekman, R., Hakanson, R. and Sundler, F.Proj ections of peptide—containing neurones in rat largeintestine. Neuroscience 27: 655—674, 1988.Ekblad, E., Ekman, R., Hakanson, R. and Sundler, FProjections of peptide—containing neurones in rat smallintestine. Neuroscience 27: 655674, 1987.Fern, G.-L., Botti, P,L., Vezzadini, P., Biliotti, G.,Bloom, S.R. and Polak, J.M. Peptide-containing innervation212of the human intestinal mucosa. Histochemistry 76 (1982)413—420.Feyrter, F. Uber die peripheren endokrinen (parakrinen)drusen des menschen, W. Maudrich, Wien—Dusseldorf pp. 1231, 1953Furness, J.B, Lloyd, K.C., Sternini, C. and Walsh, J,H.Projections of substance P, vasoactive intestinal peptideand tyrosine hydroxylase iinmunoreactive nerve fibres in thecanine intestine, with special reference to the innervationof the circular muscle. Arch. Hist. Cyt. 53: 129—140, 1990,Furness, J., Morris, J.L and Gibbins, I.L Costa, M.Chemical coding of neurones and plurichemical transmission.Annu. Rev. Phariuacol. Toxicol. 29: 289—306, 1989.Furness, J.B. and Costa, M. The enteric nervous system.Churchill- Livingstone. Edinburgh-London-Melbourne-New York,1987.Furness, J.B., Costa, M. and Keast, J.R. Choline—acetyltransferase and peptide immunoreactivity of submucosalneurones of the small intestine of the guinea pig. CellTiss. Res. 237: 329—336, 1984.213Furness, J.B. and Costa, M. VIP and enteric inhibitorynerves. In: Vasoactive Intestinal Peptide S. Said ed., NewYork:Raven pp. 391—406, 1982.Furness, J.B. and Costa, M. Types of nerves in the entericnervous system. Neuroscience 5: 1—20, 1980.Gabella, G. On the plasticity of form and structure ofenteric ganglia. J, Autonom. Nerv. Syst. 30: S59—66, 1990.Gabella, G. Structure of muscles and nerves in thegastrointestinal tract. In Physiology of theGastrointestinal Tract, second edition, edited by Leonard RJohnson. Raven Press, New York, 1987.Gallacher, D,V,, Hanley, N.R., Petersen, O.H,, Roberts,M.L., Squire-Pollard, L.G. and Yule, D.I. Substance P andbombesin elevate cytosolic Ca2+ by different molecularmechanisms in a rat pacreatic acinar cell line. J. Physiol.426: 193—207, 1990.Gates, T. S., Zimmerman, R. P., Mantyh, C.R., Vigna, S .R.,Maggio, J.E., Welton, M.L., Passaro Jr., E.P, and Mantyh,P.W. Substance P and substance K binding sites in the humangastrointestinal tract: localization by autoradiography.Peptides 9: 1207—1219, 1989.214Gershon, M.D. Erde, S.M. and Dreyfus, C.F, Long—termsurvival of serotoninergic neurones in tissue cultures ofmyenteric plexus dissected from guinea pig ileum andduodenum. Soc. Neurosci. Abstr. 6: 274, 1980.Grider, J.R. Somatostatin release from isolated ganglia ofthe MYP. Am. J0 Physiol. 257 (Gastrointest. Liver Physiol.):G3l3—G315, 1989.Gross, J., Wang, Z. and Wunder, B.A. Adaptations to foodquality and energy needs: changes ingut morphology andcapacity of Microtitus Ochrogaster. J. Mammal. 66: 661-669,1986.Guarnieri, M., Krell, L.S,, McKhann, G.M., Pasternak, G.W.and Yamamura, H,I. The effects of cell isolation techniqueson neuronal membrane receptors. Brain Res. 93: 337—342,1975.Guillemin, R. Biochemical and physiological correlates ofhypothalamic peptides. The endocrinology of the neuron.In: The Hypothalamus, S. Reichlin, R.J. Baldessarini andJ.B. Martin, eds, New York: Raven Press, pp. 155-194, 1978.Gulati, N., Mathison, R., Huggel, H., Regoli, D. and Beny,J. -L. Effects of neurokinins on the isolated pig coronaryartery. European J. Pharmacol. 137: 149—154, 1987.215Gulbenkian, S., Wharton, J. and Polak, JM. Thevisualization of cardiovascular innervation in the guineapig using an antiserum to protein gene product 9.5 (PGP9.5). J. Autonom. Nerv, Syst. 18: 235—247, 1987.Gunn, M. Histological and histochemical observations on themyenteric and submucous plexus of mammals. J. Anat. 102:223—239, 1968.Hartzell, H.C. and Fambrough, D.H. Acetylcholine receptorproduction and incorporation into membranes of developingmuscle fibres. Develop. Biol. 30: 153—165, 1973.Hirst, G,D.S., Holman, M.E. and Spence, I. Two types ofneurones in the MYP of duodenum in the guinea pig. J.Physiol. 236: 303—326, 1974,Hokfelt, T. Neuropeptides in perspective: the last tenyears. Neuron 7: 867879, 1991,Hubel, K.A. Intestinal nerves and ion transport: stimuli,reflexes and responses. Am. J. Physiol. 248: G261—271,1985216Hubel, K.A., K.S. Renquist and S. Shirazi0 Neural controlof ileal ion transport: role of substance P (SP) in rabbitand man. Gastroenterology 86: 1118P, 1984.Huizinga, J.D. Action potentials in gastrointestinal smoothmuscle. Can J. Physiol. Pharmacol. 69: 11331142, 1991.Inoue, K. and Nakazawa, K. ATP receptor operated Ca2+influx and catecholamine release from neuronal cells. NIPS7: 56-59, 1992.Iversen, L. and Goodman, E., eds. Fast and Slow ChemicalSignaling in the Nervous System. New York:Oxford UniversityPress, 1986.Jessen, K.R., Saffrey, M.J. and Burnstock, G. The entericnervous system in tissue culture I. Cell types and theirinteractions in explants of the myenteric and SMPes from theguinea pig, rabbit and rat. Brain Research 262: 17-35,1983.Jessen, K.R., McConnell, J.D., Purves, R.D., Burnstock, G.and Chamley—Campbell, J. Tissue culture of mammalianenteric neurones. Brain Research 152: 573—579, 1978.217Johansson, 0. and Hokfelt, T. TRH, somatostatin andenkephalin: Distribution studies using immunohistochemicaltechniques. J. Histochem. 28: 364—366, 1980.Kaczmarek, L. The role of protein kinase C in theregulation of ion channels and neurotransmitter release.Trends in Neurosciences 10: 30—34, 1987.Katayama, Y., North, R.A, and Williams, J.T. The action ofsubstance P on neurones of the MYP of the guinea pig smallintestine. Proc. R. Soc. Lond. B 206:191—208, 1979.Keast, J.R. Mucosal innervation and control of water andion transport in the intestine. Rev, Physiol. Biochem,Phariuacol. 109; 1—59, 1987.Keast, J.R., Furness, J.B. and Costa, M. Distribution ofcertain peptide—containing nerve fibres and endocrine cellsin the gastrointestinal mucosa in five mammalian species. JComp Neurology 236: 403—422, 1985Keast, J.R., J.B. Furness and .Costa. Somatostatin inhuman enteric nerves: distribution and characterization.Cell Tissue Res. 237: 299—308, 1984.Kimura, S., Okada, N., Sugita, Y., Kanagawa, K. andMunekato, E. Novel neuropeptides, neurokinin A and B,218isolated from porcine spinal cord0 Biochem. Biophys. Res.Commun. 114: 533, 1983.Korman, L.Y., Hassan, S., Bass, B., Moody, T.W. and Harmon,J.W. Distribution of vasoactive intestinal polypeptide andsubstance P receptors in human colon and small intestine.Dig. Dis. Sci. 34: 1100—1108, 1989.Kottegoda, S.R. Peristalsis of the small intestine. In:Smooth Muscle ed. Bulbring, E., Brading, A.F., Jones, A.W.and Tomita, T. pp. 525541, 1970.Krejs, G. J. and Fordtran, J.S, Effect of VIP infusion onwater and ion transport in the human jejunum.Gastroenterology 78: 722-’777, 1980.Krisch, B. Differing immunoreactivities of somatostatin inthe cortex and the hypothalamus of the rat. Cell Tiss, Res.212: 457—464, 1980.Kulakowski, E.C., Lampson, W.G., Schafer, SOW, andLovenberg, W. Action of substance P on the working ratheart. Biochem. Pharmacol. 32: 1097—1100, 1983.Kwok, Y.N. and McIntosh, C.H.S. Release of substance P-likeinununoreactivity from the vascularly perfused rat stomach.Eur. J. Pharmacology. 180: 201—207, 1990.219Kwok, Y,N., McIntosh, C.H.S., Pederson, R.A. and Brown, J.C.Effect of substance P on somatostatin release from theisolated perfused stomach. Gastroenterology 88:90—95,1985.Langley, J.N. The Autonomic Nervous System, Part 1. W.Heffer and Sons, Cambridge, England, 1921.Larsson, L.I., Golteriuann, N., DeMagistris, L., Reh, J.F.and Schwarz, T.W. Somatostatin cell processes as pathwaysfor paracrine secretion. Science 205: 1393—1395, 1979.Larsson, L.I. Ultrastructural localization of a newneuronal peptide (VIP). Histochemistry 54: 173-176, 1977.Lauweryns, J.M. and Van Ranst, L. Protein gene product 9.5expression in the lungs of humans and other mammals.Immunocytochemical detection in neuroepithelial bodies,neuroendocrine cells and nerves. Neurocience Lett. 85:311—316, 1988.Lentzen, H. and Palenker, J. Localization of the thiorphan—sensitive endopeptidase, termed enkephalinase A, on glialcells. FEBS Lett. 153: 93—7, 1983.220Lucius, R. and R. Mentlein. Degradation of the neuropeptidesomatostatin by cultivated neuronal and glial cells. J.Biol. Chelu. 28: 18907—18913, 1991.Luft, R., Efendic, S., Hokfelt, T., Johansson, 0. andAriiuura, A. Iinmunohistocheiuical evidence for thelocalization of soiuatostatin—like immunoreactivity in a cellpopulation of the pancreatic islets, Med. Biol. 52: 428-430, 1974.McFadden, D., Zinner, M.J. and Jaffe, B.J. Substance P—induced intestinal secretion of water and electrolytes. Gut27:267—272, 1986.McIntosh, C.H.S., V. Bakich, Y.N. Kwok and J.C. Brown. Acomparison of the inhibitory effects of somatostatin-14, -25and -28 on motility of the guinea pig ileuiu. Can. J.Physiol. Pharmacol. 64: 303—306, 1987a.McIntosh, C.H.S,, Y.N. Kwok, C. Tang and J.C. Brown. Theuse of monoclonal antibodies in RIA for gastrointestinalhormones. J. din. Immunoassay 10: 79—84, 1987b.McIntosh, C. H. S. Gastrointestinal soinatostatin:distribution, secretion and physiological significance.Life Sciences 37: 2043—2058, 1985.221Maggi, C.A., Giuliani, S., Patacchini, R., Santicioli, P.,Theodorsson, Barbanti, G., Turini, D. and Giachetti, A.Tachykinin antagonists inhibit nerve-mediated contractionsin the circular muscle of the human ileuiu: Involvement ofneurokinin—2 receptors. Gastroenterology 102: 88—96, 1992.Manaka, H., Manaka, Y, Kostolanska, F., Fox, J,E,T, andDaniel, E.E. Release of VIP and substance P from isolatedperfused canine ileum. Am. J. Physiol. G182—190, 1989.Mantyh, P,W., Mantyh, C,R,, Gates, T., Vigna, S.R. andMaggio, J.E. Receptor binding sites for substance P andsubstance K in the canine gastrointestinal tract and theirpossible role in inflammatory bowel disease, Neuroscience25 (1988) 817—837.Meinstrich, I’LL. Experimental factors involved inseparation by centrifugal elutriation. In Cell SeparationMethods and Selected applications, vol 2. pp. 33—61, 1983.Meissner, G. Uber dieNerven der Darmwand. Z. Ration. Med.N.F. 8: 364—366, 1857.Milihorn, D.E. and Hokfelt, T. Chemical messengers andtheir co—existence in individual neurones. NIPS 3: 1—5,1988.222Mimer, P., Ralevic, V., Hopwood, A.M., Feher, E., Lincoln,J., Kirkpatrick, K.A. and Burnstock, G. Ultrastructurallocalization of substance P and choline acetyltransferase inendothelial cells of rat coronary artery and release ofsubstance P and acetyicholine during hypoxia. Experentia45: 121—125, 1989.Mihara, S., North, R.A. and Surprenant, A. Somatostatinincreases an inwardly rectifying potassium conductance inguinea pig SMP neurones. J. Physiol. 390: 335—355, 1987.Milton, K. Digestive physiology in primates. NIPS 1: 76—79, 1986.Morita, K., North, R,A, and Tokimasa, T. Muscarinincagonists inactivate potassium conductance of guinea pigmyenteric neurones. J. Physiol. 333: 125—139, 1982.Nishi, R. and Willard, A.L. Neurones dissociated from ratmyenteric plexus retain differentiated properties when grownin cell culture, I. Morphological properties andimmunocytochemical localization of transmitter candidates.Neuroscience 16: 187—199, 1985.Nishi, R. and North, R.A. Intracellular recording from theMYP of the guinea pig ileum. J. Physiol. 231: 471—491,1973.223Nishizuka, Y. Studies and perspectives of protein kinase C.Science 233: 305—312, 1986.North, R.A. Receptors on individual neurones. Neuroscience17: 899—907, 1986.North, R.A., Slack, B.E. and Surprenant, A. Muscarinic N1and N2 mediate depolarization and presynaptic inhibition inguinea pig enteric nervous system. J. Physiol. 368: 435452, 1985.North, R.A. and Tokimasa, T. Muscarininc synapticpotentials in guinea pig MYP neurones. J. Physiol. 333:151—156, 1982.Oda, T., M.S, Shearman and Y. Nishizuka, Synaptosomalprotein kinase C subspecies: B. Down—regulation promoted byphorbol ester and its effect on evoked norepinephrinerelease. J. Neurochemistry 56: 1263-°1269, 1991.Palmer, J.M,, Wood, J,D. and Zafirov, D.H. Transduction ofaminergic and peptidergic signals in enteric neurones of theguinea pig. J. Physiol. 387: 371383, 1987.Pataky, D.M., Curtis, S,B. and Buchan, A.M.J. The colocalization of neuropeptides in the submucosa of the small224intestine of normal Wistar and non—diabetic BB rats.Neuroscience 36: 247—254, 1990.Pearse, A.G.E. Peptides in the brain and intestine0 Nature262: 92—94, 1976.Pearse, A.G.E. and Polak, J.M. Immunocytochemicallocalization substance P in mammalian intestine.Histochemistry 41: 373375, 1975.Penman, E., J.A.H. Wass, ?4.G. Butler, E.S. Penny, J. Price,P. Wu. and L.H. Rees. Distribution and characterization ofiinmunoreactive somatostatin in human gastrointestinal tract.Reg. Peptides 7: 53—67, 1983.Perdue, 14.H., Gaibraith, R. and Davison, J. Evidence forsubstance P as a functional neurotransmitter in the guineapig small intestinal mucosa. Regul. Pept. 18: 63—74, 1987.Peterfreund, R.A. and W.W, Vale. Phorbol diesters stimulatesomatostatin secretion from cultured brain cells.Endocrinol. 113: 200—208, 1983.Polak, J., Pearse, A.G.E., Grimelius, L., Bloom, S.R. andArimura, A. Growth—hormone release—inhibiting hormone ingastrointestinal and pancreatic D cells. Lancet 1: 1220-1224, 1975.225Powell, D.W. Intestinal water and electrolyte absorption.In: Physiology of the Gastrointestinal Tract, secondedition, edited by Leonard R Johnson. Raven Press, NewYork, pp. 1267—1306, 1987.Price, J., Penman, E., Wass, J.A.H. and Rees, L,H.Bombesin—like iminunoreactivity in human gastrointestinaltract. Regulatory Peptides 9: 1-10, 1984.Quirion, R. and Gaudreau, . strategies in neuropeptidereceptor binding research. Neuroscience and biobehaviouralrev. 9: 413—420, 1985.Reichlin, S. Systems for the study of regulation ofneuropeptide secretion. In Advances in BiochemicalPsychopharmacology. Vol. 28, Neurosecretion and brainpeptides, implications for brain function and disease.Edited by J.B. Martin, S. Reichlin and K.L. Bick. RavenPress:New York, 1981.Rogers, D.C. and Burnstock, G. The interstitial cell andits place in the concept of the autonomic ground plexus. J.Comp. Neurol. 126: 255—286, 1966.Rumesson, J.J. and Thuneberg, L. Interstitial cells ofCaal in human small intestine. Ultrastructural226identification and organization between the main smoothmuscle layers. Gastroenterology 100: 1417—1431, 1991.Said, S. and Mutt, V. Potent peripheral and splanchnicvasodilator peptide from normal gut. Nature 225: 863-864,1970Sanders, K.M. and Smith, T.K. Motoneurones of thesubmucosal plexus regulate electrical activity of thecircular muscle of canine proximal colon, J. Physiol, 380:293’310, 1986.Schally, A.V., Arimura, A. and Kastin, A.J. Hypothalamicregulatory hormones. Science 179, 341—350, 1973.Scheuermann, D., Stach, W., Tinuuermans, J.—P., Adriansen, D.and De Groodt—Lasseel. Neuron—specific enolase and S—i—protein inuuunohistochemistry for defining the structure andtopographical relationship of the different enteric nerveplexuses in the small intestine of the pig. Cell Tiss. Res.256: 65—75, 1989.Schultzberg, M., Hokfelt, T., Nilsson, G, Terenius, L.,Rehfeld, J,F., Brown, M., Elde, R., Goldstein, M. and Said,S. Distribution of peptide— and catecholamine-containingneurones in the gastrointestinal tract of rat and guineapig: Immunohistochemical studies with antisera to substance227P, vasoactive intestinal peptide, enkephalins, somatostatin,gastrin/cholecystokinin, neurotensin and dopamine—i—hydroxylase. Neuroscience 5: 689—744, 1980.Smith, T.K., Furness, J.,B., Costa, 14. and Bornstein, J,C.An electrophysiological study of the projections of motorneurones that mediate non—cholinergic excitation in thecircular muscle of the guinea pig small intestine, J.Autonom. Nerv. Syst. 22: 115—128, 1988.Stach, W. Der plexus submucosus externus (Schabadasch) imDunndariu das Schweins. I. Form, Struktur und Verbindungender Ganglien und Nervenzellen. Z. Mikrosk.—Anat. Forsch 91:737—755, 1977.Surprenant, A. Slow excitatory synaptic potentials recordedfrom neurones of guinea pig submucous plexus. J. Physiol.351: 343—361, 1984.Takada, Y., Hiwada, K., Unno, 14. and Kokubu, T.Inuuunocytochemical localization of angiotensin convertingenzyme at the ultrastructural level in the human lung andkidney. Biomed. Res. 3: 169—74, 1982.Tange, A. Distribution of peptide containing endocrinecells and neurons in the gastrointestinal tract of the dog:immunohistochemical studies using antisera to somatostatin,228substance P1 vasoactive intestinal polypeptide, met—enkephalin and neurotensin. Biomed. Res. 4: 9—24, 1983.Tapia—Arancibia, L., Pares—Herbute, N., Astier, H.,Reichlin, S. and Nathanson, J. Adenylate cyclase activationis not sufficient to stimulate somatostatin release fromdispersed cerebral cortical and diencephalic cells in gliafree cultures. Brain Research 450: 101—110, 1988.Thuneberg, L. Interstitial cells of Cajal: intestinalpacemaker cells? Adv. Anat. Embryol. Cell, Biol0 71: 1’130, 1982Unverferth, D.V., WDorisio, T.M., Muir, W,W.III, White, J.,Miller, M.M., Hamlin, R.L. and Magorien, R.D. Effect ofvasoactive intestinal polypeptide on the caninecardiovascular system. J. Lab, din. Med. 106: 542—550,1985.Vigna, S.R., Mantyh, C.R., Giraud, A.S., Soil, A.H., Walsh,J.H. and Mantyh, P.W. Localization of specific bindingsites for bombesin in canine gastrointestinal tract.Gastroenterology 93: 1287—95, 1987.von Euler, U.S. and Pernow, B. An unidentified depressorsubstance in certain tissue extracts, J0 Physiol. 72: 74—87, 1931.229Wang, H.-L, Bogen, C., Reisine, T. and Dichter, M.Somatostatin—14 and somatostatin—28 induce opposite effectson potassium currents in rat neocortical neurones. Proc.Natl. Acad. Sci. USA 86: 9616—9620, 1989.Wattchow, D.A., Furness, J,B. and Costa, M, Distributionand co—existence of peptides in nerve fibres of the externalmuscle of human gastrointestinal tract. Gastroenterology95: 32—41, 1988.Weber, M. Cole, T. and Conlon, J.M. Am. J. Physiol. 250:G679—685, 1986.Wiley, J. and Owang, C. Soiuatostatin inhibits cAMP-mediatedcholinergic transmission in the MYP. Am. J. Physiol. 253:G607—612, 1987.Willard, A.L. Substance P mediates synaptic transmissionbetween rat myenteric neurones in cell culture, J. Physiol.426: 453—471, 1990.Willard, A,L and Nishi, R. Neurones dissociated from ratMYP retain differentiated properties when grown in cellculture. II. Electrophysiological properties and responsesto neurotransmitter candidates, Neuroscience 16: 201—211,1985a.VV 230Willard, A.L and Nishi, R. Neurones dissociated from ratMYP retain differentiated properties when grown in cellculture. III. Synaptic interactions and modulatory effectsof neurotransmitter candidates, Neuroscience 16: 2 13—221,1985b.Wilson, A.J., Liewellyn—Smith, I.J., Furness, J.B. andCosta, N. The fine structure of the SMP of the guinea pigileum. I. The ganglia, neurones, Schwann cells andneuropil. J Neurocytol. 10:759—784, 198lWilson, P.O.G., Barber, P.C., Hamid, Power, B.F.,Dhillon, A.P., Rode, J,, Day, I.N.M., Thompson, R.J0 andPolak, J.M. The immunolocalization of protein gene product9.5 using rabbit polyclonal and mouse monoclonal antibodies.Br. J, Exp. Path. 69: 91—104, 1988.Womack, M.D., MacDermott, A.B. and Jessell, TM. Sensoryneurotransmitters regulate intracellular calcium in dorsalhorn neurones. Nature 334: 351-353, 1988.Wong, H.C., J.H. Walsh, H. Yang, Y. Tache and A.MJ. Buchan.A monoclonal antibody with potent immunoneutralizingactivity Peptides 11: 707—712, 1990.231Wood, J.D. Physiology of the enteric nervous system. In:Physiology of the Gastrointestinal Tract, second edition,edited by Leonard R Johnson. Raven Press, New York. pp.67—110, 1987.Wood, J.D. Electrical activity from single neurones ofAuerbach’s plexus. Am. J. Physiol. 219: 159—169, 1970.Yamada, T. Local regulatory actions of gastrointestinalpeptides. In: Physiology of the Gastrointestinal Tract,second edition, edited by Leonard R Johnson. Raven Press,New York, 1987.Yau, W.M., Dorsett, J.A. and Youther, M.L, Inhibitorypetidergic neurones: functional difference betweensomatostatin and enkephalin in the NYP. Am. J. Physiol.250: G60—63, 1986,Yurchenko, P.D. and H. Furthmayr. Self-’assembly of basementmembrane collagen. Biochemistry 23: 1839—1850, 1984.Zappia, L., Molina, E., Sianesi, M. and Bertaccini, G.Effects of natural analogues of substance P on the motilityof the human gastrointestinal tract in vitro, J. Pharm.Pharmacol. 30: 593—594, 1978.Zinuueriuan, R.P., Gates, T..S., Mantyh, C.R., Vigna, S.R.,Boehiuer, C.G. and Mantyh, P,W. Vasoactive intestinalpeptide (VIP) receptors in the canine gastrointestinaltract. Peptides 9: 1241—1253, 1989.232233APPENDIX I. CHEMICAL SOURCESChemical SourceAcetic acid BDHAcetone BDHAmmonium acetate BakerAnuuonium chloride FisherAanphotericin B GibcoAprotinin MilesBasal medium eagle (powder) GibcoBovine serum albumin (fraction V) SigmaBovine serum albumin (RIA grade) Sigma, MilesCalcium chloride FisherCalcium ionophore (A23187) SigmaCarbon decolourizing neutral (Activated charcoal) FisherChioramine T SigmaCollagenase (type I, XI) SigmaCytosine B-D-arabinoside GibcoDextran T—70 PharmaciaDiaminobenz idine BDHDimethylsufoxide SigmaDulbecco’s modified Eagle Medium GibcoEthanol Commercial AlcoholFetal calf serum GibcoFormaldehyde (histology grade) FisherForskolin SigmaGelatin SigmaGentamycin sulphate SigmaGlucose (50 % commercial solution) AbbottGlucose oxidase SigmaGlutamine SigmaHank’s balanced salt solution (powder) GibcoHematoxylin FisherHEPES FisherHydrocortisone SigmaHydrogen peroxide FisherImidazole SigmaInsulin SigmaLithium carbonate FisherMagnesium sulphate FisherNerve growth factor Collaborative ResearchNormal swine serum GibcoParraff in (paraplast) MonojectPermount FisherPetroleum ether FisherPhenol red SigmaPhorbol esters (I3PMA, 4aphorbol) Sigma234Chemical SourcePicric acid BDHPotassium chloride FisherPotassium phosphate (monobasic) FisherSephadex CM—52 PharmaciaSodium acetate FisherSodium barbital BakerSodium bicarbonate FisherSodium chloride FisherSodium hroxide FisherSodium iodide AmershamSodium merthiolate Eastman KodakSodium metabisulphite FisherSodium pentobarbital GibcoSodium phosphate FisherSodium pyruvate GibcoSomatostatin PeninsulaSubstance P PeninsulaTris-HCL SigmaTriton X-100 FisherXylene FisherAPPENDIX II Eurocollins buffer (mmol/L)Na 10 , K 115, Cl 15, HCO3 10, P04 577, glucose 195Osmolarity = 330 mosm/kgpH = 7.0


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items