UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of a somatostatin monoclonal antibody on gastrin and insulin release Tang, Christine Louise 1988

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

Item Metadata

Download

Media
831-UBC_1988_A6_7 T37.pdf [ 9.02MB ]
Metadata
JSON: 831-1.0097854.json
JSON-LD: 831-1.0097854-ld.json
RDF/XML (Pretty): 831-1.0097854-rdf.xml
RDF/JSON: 831-1.0097854-rdf.json
Turtle: 831-1.0097854-turtle.txt
N-Triples: 831-1.0097854-rdf-ntriples.txt
Original Record: 831-1.0097854-source.json
Full Text
831-1.0097854-fulltext.txt
Citation
831-1.0097854.ris

Full Text

T H E E F F E C T S O F A S O M A T O S T A T I N M O N O C L O N A L A N T I B O D Y O N G A S T R I N A N D INSULIN R E L E A S E by C H R I S T I N E L O U I S E T A N G B . S c . (Hon.), University of British Co lumb ia , 1985 A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department of Phys io logy W e accept this thesis as conforming to the required s tandard T H E U N I V E R S I T Y O F BRIT ISH C O L U M B I A May 1988 © Christ ine Lou ise T a n g , 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ii A B S T R A C T The effects of endogenous somatostat in on gastr in and insul in re lease we re s tud i ed by pa s s i v e immun i za t i on with a somatos ta t i n monoc l ona l ant ibody, S O M A 10, in the i so la ted pe r fused rat s t omach and panc reas , respec t i ve l y . P r e v i ou s pa s s i v e immun i za t i on s tud ies with somatos ta t i n ant i serum in the iso lated per fused s tomach have y ie lded conf l ict ing results. The d i f fe rences in results cou ld be due to the d i f fe rences in the b inding character ist ics of the ant isera used, and the accessibi l i ty of the ant iserum to the t issue. In previous studies in vivo and in isolated islets, somatostat in ant iserum c a u s e d an i n c r e a s e in i n su l i n r e l e a s e . P r e v i o u s a t t e m p t s at immunoneu t r a l i z i ng somatos ta t i n in the i so l a ted pe r f u sed p a n c r e a s of m a m m a l s have y ie lded negat ive results. However , the i so la ted per fused p an c r e a s is an idea l mode l for pa s s i v e immun i za t i on s tud ies , b e c a u s e ho rmona l and cent ra l ne rvous s y s t em in f luences are e l im inated , and the microanatomy of the islet is preserved. Th is study differed from previous studies in that a monoc lona l antibody to somatostat in, which is more speci f ic than somatostat in ant iserum, was used in an attempt to neutral ize endogenous gastr ic and pancreat ic somatostat in. Fab f ragments of S O M A 10 were made by papa in d igest ion and purif icat ion on prote in A - s e p h a r o s e . T h e s e f r agmen t s are a d v a n t a g e o u s for p a s s i v e immunizat ion, s ince they contain the somatostat in binding site, and are much smal ler than the intact antibody. Therefore they should more readily penetrate into the interstitium and neutral ize endogenous somatostat in. S O M A 10 w a s pur i f ied by a m m o n i u m su l pha te p rec ip i ta t ion , in conjunct ion with hydroxylapat ite chromatography. Purity was checked by gel fi ltration and affinity H P L C and determined to be 93%. Sca t cha rd ana lys i s c a l cu l a t ed the b ind ing capac i t y of S O M A 10 to be 8.3 | ig/mg, and the d issoc iat ion constant to be 2.2 nM . Both S O M A 10 and the Fab fragments were shown to inhibit the effect of exogenous l y admin i s te red somatostat in in the isolated perfused s tomach and in gastric f istula rats. iii S ing le pa s sage of S O M A 10 in the iso lated per fused s t omach at 100 |ig/ml c au s ed a signif icant dec rease in basa l gastr in re lease. Rec i rcu lat ion of the antibody in the s tomach caused an increase in cumulated gastr in re lease in compar i son to controls in which perfusate without the antibody was recirculated. Infusion of the F ab fragment at 15 and 66 |ig/ml c au s ed an increase in basa l gastr in re l ease sugges t ing that somatostat in inhibits basa l gastr in re lease . Immunocytochemica l staining of the perfused s tomachs revea led that the Fab fragments but not the intact ant ibodies had penetrated into the interstitium. In the panc reas , infus ion of 45 j ig/ml S O M A 10 and 30 u,g/ml F ab f ragments inhibited insulin secret ion in response to 8.8 m M g lucose. There are severa l explanat ions for the unexpected dec rease in gastr in re lease and insulin re lease, obse rved when S O M A 10 was infused. The ant ibody and f ragment cou ld be b inding to the somatostat in receptor and mimick ing its effects on gastrin and insul in. A change in conformation of the somatostat in molecu le due to b ind ing of the ant ibody cou ld change the affinity for the receptor. The inc reased secret ion cou ld be a result of neutral izat ion of somatostat in , which inh ib i ted a p rev ious l y unknown inhibi tor of gast r in and insu l in re lease . Immunocytochemica l staining revealed that both S O M A 10 and its Fab fragment had entered into the interstitium, implying that endocr ine and paracr ine effects of S O M A 10 could not be dist inguished. In summary , the i n c rease in gast r in r e l ease ob s e r v ed when F ab f ragments were infused and when S O M A 10 was rec i rcu lated sugges ted that somatostat in exerts a cont inuous restraint on gastr in re lease. Both S O M A 10 and Fab fragments were found in the pancreat ic interstitium implying that large mo lecu l e s c an pa s s through the vascu l a r wal ls , and that e ndogenou s and paracr ine effects of the antibody cannot be dist inguished in this model . iv TABLE OF CONTENTS A B S T R A C T ii LIST O F T A B L E S x LIST O F F I G U R E S xi A C K N O W L E D G E M E N T S xiv DEDICATION xv INTRODUCTION 1 M E T H O D S 16 I. Antibody Purification Methods 16 A. Ammonium Sulphate Precipitation 16 1. Rationale 16 2. Procedure 16 B. Hydroxvlapatite Chromatography 17 1. Rationale 17 2. Procedure 17 C. Enzvme- l inked Immunosorbant A s say (ELISA) 17 D. High Per formance Liquid Chromatography (HPLC ) 18 1. Rationale 18 2. H P L C Equipment and Buffers 18 3. Affinity H P L C 20 a. Rationale 20 b. Derivitization of the Co lumn 20 c. H P L C Buffers 20 d. Sample Application 21 4. Ge l Filtration H P L C 21 a. Rationale 21 b. Procedure 21 c. Sample Application 21 V II. Antibody Characterization 21 A. Binding Region on Somatostat in Molecule 21 1. Rationale 21 2. Procedure 22 a. Assay Buffer 22 b. Standards 22 c. Antibodies 2 3 d. Label 24 i. lodination of Somatostatin 24 ii. Purification o f 1 ^ - Soma to s t a t i n 24 e. Assay Protocol 2 5 f. Calculations 26 B. Speci f ic Activity from Self D isp lacement Cu rves 26 1. Rationale 26 2. Procedure 26 C. Scatchard Plots 27 D. Immunodiffusion Gels. 27 1. Rationale 27 2. Procedure 28 III. S O M A 10 Fab Fragment Production 28 A. Production 28 1. Rationale 28 2. Procedure 29 B. S O M A 10 Fab Fragment Purification. 29 1. Rationale 29 2. Procedure 30 IV. Characterization of Fab Fragments 30 A. Sod ium Dodecv l Su lphate-Po lvacry lamide G e l E lectro- phoresis (SDS-PAGE) 30 1. Rationale 30 2. Materials 31 3. Preparation of Ge l 31 4. Samples and Standards 32 5. Running the Ge l 32 6. Staining 32 vi B. A s se s smen t of Binding of Ant ibody to 125l-Labelled Somatostat in by Chromatography on Sephadex G-100 33 1. Rationale 3 3 2. Preparation of Ge l 3 3 3. Samples 3 3 4. Chromatography 3 3 V. Protein Quantification 34 1. Rationale 34 2. Procedure 34 a. Standard Protocol 3 4 b. Micro Protocol 34 3. Protein Determination 3 5 IV. B ioassays 3 5 A. Acid Secretion 3 5 1. Rationale 3 5 2. Apparatus 3 5 3. Animals 37 4. Surgical Procedure 37 5. Experimental Protocol 3 7 B. Isolated Per fused Rat S tomach and Isolated Per fused Rat Pancreas 38 1. Rationale 38 2. Apparatus 38 3. Somatostat in, S O M A 10 and Fab Fragment Administration 40 4. Preparation of Perfusate 40 5. Animals 41 6. Surgical Procedure 41 a. Isolated Perfused Stomach 41 b. Isolated Perfused Pancreas 42 VII. Statistical Analysis, 4 2 vii VIII. Pept ide Quantif ication by Rad io immunoassay 42 A. Gastrin RIA 42 1. Assay Buffer 42 2. Standards 4 3 3. Antiserum 43 4. 125|-Gastrin 4 3 a. lodination of Gastrin 4 3 b. Purification of 1 2 5 l - G a s t r i n 4 3 5. Controls 44 6. Assay Protocol 44 7. Separation 44 8. Calculations 44 B. Insulin RIA 42 1. Assay Buffer 4 5 2. Standards 45 3. Antiserum 45 4. 125|.|risulin 4 5 a. lodination of Insulin 4 5 b. Purification of 1 2 5 l - l n s u l i n 46 5. Controls 46 6. Assay Protocol 47 7. Separation 47 8. Calculations 47 XI. Immunocytochemistry (ICC) 4 7 A. Rationale 47 B. Antibody Localization in T issue 48 1. Preparation of Tissue Sect ions 48 a. Procedure for Test ing Antibody and Fragment Immunoreactivity 4 8 b. Procedure for Locat ing Ant ibodies in Per fused Organs 49 C. Assessmen t of Nonspec i f ic Binding to Co l lagen 49 1. Rationale 49 2. Procedure 49 A P P E N D I X T O M E T H O D S 50 vii i R E S U L T S 51 I. Purification of S O M A 10 51 II. Assessment of Purity of S O M A 10 51 A. Affinity Chromatography 51 B. Ge l Filtration Chromatography 51 III. Characterization of Antibodies 59 IV. Fab Fragment Production 66 V. Identification of Fab Fragments 72 A. Sod ium Dodecy l Su lphate-Po lyacry lamide Ge l Electrophoresis 72 B. Chromatography of Incubation Mixtures of Ant ibody and 1 2 5 l - S o m a t o s t a t i n on Sephadex G-100 72 C. Immunocytochemica l Character izat ion of Fab Fragments 79 IV. Ac id Secretion 79 VII. Isolated Per fused S tomach Exper iments - Gastr in Re lease 84 i. Effect of S O M A 10 and Fab Fragments on Somatostat in- lnduced Inhibition of Gastr in Re l ease 84 ii. Effect of S O M A 10 and Fab Fragments on Basa l Gastrin Release 91 iii. Effect of Recirculat ion of S O M A 10 on Basa l Gastr in Release 97 VIII. Isolated Per fused Panc reas Exper iments - Insulin Re l ease 101 i. Effect of S O M A 10 and Fab Fragments on Insulin Responses to a Glucose-Gradient 101 ii. Effect of Fab Fragmetns on G lucose (16.5 mM)-Stimulated Insulin Re lease 101 iii. Effect of S O M A 10 and Fab Fragments on G lu cose (8.8 mM)-Stimulated Insulin Re lease 106 iv. Effect of Pre-surgical Immunoneutral ization of Somatostatin with S O M A 10 109 ix XI. Immunocytochemica l Identification of Fab Fragments and S O M A 10 in Perfused Organs 109 D ISCUSS ION 120 I. Purif ication and Character izat ion of S O M A 10 and Fab Fragments 120 II. Effect of S O M A 10 and Fab Fragments on Bio logical Activities 123 A. Pass i ve Immunization Studies in the Isolated Perfused Stomach 123 B. Pass i ve Immunization Studies in the Isolated Perfused Pancreas 128 III. Conclusions 130 R E F E R E N C E S 132 X LIST O F T A B L E S T A B L E P A G E 1 Relat ive Affinities of Somatostat in Ana logs to Somatostatin-14 62 2 Binding Capac i ty and Dissociat ion Constant (Kp) of S O M A Antibodies 71 3 E lectrophores is of S O M A 10 and Fragments on S D S -Polyacrylamide Ge l 75 xi LIST O F F I G U R E S F I G U R E P A G E 1 Digest ion of Ant ibody with Papa in and Peps in 13 2 Components of an H P L C System 19 3 Double-barrel led Gastr ic Cannu la 36 4 Apparatus for the Isolated Per fused Organ Preparation 39 5 Purif ication of Ammon ium Sulphate Precip i tated S O M A 10 on Hydroxylapatite 52 6 Affinity H P L C Elution Profile and E L ISA Activity of S O M A 10 at Var ious Purification Steps 53 7 G e l Filtration H P L C Elution Profi le and E L ISA Activity of S O M A 10 at Var ious Purif ication S teps 56 8 Compar is ion of S O M A 10 Activity on E L I SA 60 9 Amino Ac id Sequence of Somatostat in-14 61 10 Relat ive Affinity of Somatostat in Ana logs to Somatostatin-14 63 11 Elution Profi le of H A P S O M A 10 from Protein A 67 12 Compar i son of S O M A 10 RIA Standard Cu rves 68 13 Compar i son of Protein A-purif ied S O M A 10 Activity on ELISA 69 14 Elution Profi le of Papa in-d igested S O M A 10 from Protein A 70 15 SDS-Po l yac ry l am ide Ge l E lectrophores is of S O M A 10 and Fragments 73 16 Chromatography of Incubation Mixtures of Ant ibody and 1 2 5 l - l a b e l l e d Somatostat in on Sephadex G-100 76 17 Compar i son of Immunoreactivity of S O M A 10 Fragments by Immunocytochemistry 80 18 Effect of S O M A 10 Fab Fragments on Somatostat in-Induced Inhibition of Ac id Secret ion 83 xii 19 Effect of S O M A 10 on Somatostat in- induced (60 nM) Inhibition of Gastrin Re lease 85 20 Effect of S O M A 10 on Somatostat in- induced (6 nM) Inhibition of Gastrin Re lease 87 21 Effect of S O M A 10 on Somatostat in- induced (1.5 nM) Inhibition of Gastrin Re lease 88 22 Effect of S O M A 10 on Somatostat in- induced (0.6 nM) Inhibition of Gastrin Re lease 89 23 Effect of S O M A 10 on Somatostat in- induced (0.15 nM) Inhibition of Gastrin Re lease 90 24 Effect of S O M A 10 Fab Fragments on Somatostat in-induced (1.5 nM) Inhibition of Gastr in Re lease 92 2 5 Effect of S O M A 10 (20 ng/ml) on Basa l Gastr in Re l ease 94 26 Effect of S O M A 10 (100 |ig/ml) on Basa l Gastr in Re lease 96 27 Effect of S O M A 10 Fab Fragments (15 ng/ml) on Basa l Gastrin Release 98 28 Effect of S O M A 10 Fab Fragments (66 ng/ml) on Basa l Gastrin Release 100 29 Effect of Recirculat ion of S O M A 10 on Gastr in Re l ease 102 30 Effect of S O M A 10 (5 and 10 ng/ml) on G lucose Gradient-Stimulated Insulin Re lease 103 31 Effect of S O M A 10 Fab Fragments (5 \ig/m\) on G lucose Gradient-Stimulated Insulin Re lease 105 3 2 Effect of Fab Fragments on G lucose (16.5 mM)-Stimulated Insulin Re lease 107 3 3 Effect of S O M A 10 Fab Fragments on G lucose (8.8 mM)-Stimulated Insulin Re lease 108 34 Effect of S O M A 10 on G lucose (8.8 mM)-St imulated Insulin Release 110 35 Effect of Pre-Surg ica l Pass i ve Immunization with S O M A 10 on G lucose (8.8 mM)-St imulated Insulin Re l ease 111 xiii 36 Nonspec i f ic Binding of S O M A 10 and Fab Fragments to Col lagen 112 3 7 Immunocytochemica l Staining of Fab Fragments in the Perfused Stomach 116 38 Immunocytochemica l Staining of F ab Fragments in the Perfused Pancreas 118 xiv A C K N O W L E D G E M E N T S I would like to extend my thanks and appreciat ion to: Drs. Chr i s Mc in tosh and John Brown for giving me the opportunity to pursue graduate studies under their superv is ion. Their adv ice, crit ic ism and support in all my endeavours is greatly apprec iated. Dr. A l i son Buchan , Nar inder Dhatt and Sue Curt is for their help and gu idance in immunocytochemistry . Dr. Andrew Malco lm for providing the end less vo lumes of S O M A 10 I required. Drs . Kenny Kwok, E la ine Lui , and And rew S e a l for t each ing me surg ica l techn iques . B ruce A n d e r s o n , Drs. Rob Fa rghe r and Ken Ba inbr idge for their he lp in prepar ing e lectrophores is gels . J o e Tay and J ohn Sanke r for their invaluable techn ica l a ss i s tance dur ing the last three years . Mar ie Langton and Mary Forsythe for their cler ical help. Kerry Bokenfohr, Vick i Bak ish and Chery l Webb for being there. In particular, I would like to thank Dr. Ray Pede r son for bel ieving in me, for his concern , encouragement and fr iendship. W e are most fortunate to have him as our graduate student advisor. I wou ld a l so like to thank the M R C Regulatory Pept ide group and the members of the phys io logy department for making my stay here an enjoyable exper i ence . XV To my parents, Eugene and Murray. For all their love and understanding. 1 I NTROD UC T ION Somatostat in was iso lated in 1973 from ovine hypotha lamic extracts on the bas is of its ability to inhibit growth hormone secret ion (Brazeau et al., 1973). The isolat ion of somatostat in wa s a mi lestone in that it wa s the first hormone found to have an inhibitory act ion on the anterior pituitary. At the t ime only re leas ing hormones had been isolated. Fur ther ac t ions of somatosta t in were soon ident i f ied. Somatos ta t in administrat ion was shown to inhibit growth hormone secret ion induced by L-dopa in man (Si ler et al . , 1973) and in dogs (Lovinger et al., 1974), and mea l -st imulated gastrin secret ion in man (Bloom et al., 1974). In 1975, G o m e z - P a n et al . demonstrated that somatostat in inhibited gastr ic ac id and peps in secret ion in response to pentagastr in and food st imulat ion in cats . Koerke r et a l . (1974) found that somatos ta t in c a u s e d hypog l y cem ia , and inh ib i ted insul in and g lucagon re lease. Somatostat in is now known to have severa l other centra l and per iphera l act ions. In the gut, somatostat in s upp r e s se s motor activity, nutrient absorpt ion, b lood flow and the secret ion of all known pept ides. These pept ides inc lude gastr in, cho lecystok in in , vasoact ive intest inal pept ide (VIP), sec re t i n , insu l in a n d g l u cagon . It a l so inhibi ts the re l ease of exoc r i ne secret ions, including pancreat ic exocr ine and gastr ic ac id secret ion (Mc intosh, 1985). B io log ica l ly act ive somatosta t in ex is ts in three major forms, with 14 (somatostat in-14), 25 (somatostatin-25) and 28 (somatostat in-28) amino ac ids. Both of the larger forms are extended at the N-terminus. In mammals , the amino ac id s equence of somatostat in-14 is complete ly conse rved . Two forms with reg ions of homology are found in anglerf ish. Original ly, somatostat in-28 was thought to be merely a precursor of somatostat in-14. However , somatostat in-14 has been shown to be formed directly from prosomatostat in, a 116 amino ac id res idue protein, without the intermediate formation of somatostat in-28 in the rat panc reas (Patel , 1983). The relative proportions of somatostat in-14 and -28 are ce l l a n d o r g an spec i f i c . U s i n g r a d i o i m m u n o a s s a y with ge l f i l t rat ion ch r oma tog r aphy , P e n m a n et a l . (1983) found h igher concen t r a t i ons of somatostat in-14 than -28 in the antrum, duodenum and panc reas , but lower concentrat ions of somatostat in-14 than -28 in the corpus, jejunum , i leum and co lon in man. 2 The forms of somatostat in display dif ferences in act ion. Somatostat in-28 is more potent at inhibiting insul in secret ion than the 14 amino ac id res idue form (Mandar ino et a l , 1981), but only somatostat in-14 af fected mesenter i c c irculat ion, oxygen uptake and intestinal motility in dogs at the concentrat ions tested (Konturek et al., 1981). O n a molar bas is , somatostat in-14 and -28 were found to be equipotent in caus ing inhibit ion of peptone-mea l st imulated ac id secret ion (Sea l et al., 1982). Somatostat in-28 has a p l a sma half life of less than 3.6 minutes, whi le that of somatostat in-14 has a lways been found to be less than 2 minutes (Polonsky et al., 1982) depend ing on the spec ies . Neverthe less, the half life of somatostat in-28 has a lways been found to be longer than the half life of the 14 amino ac id residue peptide. The longer half life of somatostat in-28 may in f luence the po tency on a spec i f i c b io log ica l act iv i ty c o m p a r e d to somatostat in-14. 3 Somatostat in in the S tomach In the s tomach, somatostat in is located in endocr ine D cel ls in the corpus and in the antrum. In the gu inea pig, somatostat in is a l so located in gastr ic neurons . In the intest ines of al l an ima ls s tud ied, somatostat in is located in endocr ine ce l l s in the lamina propria, and in neurons in the myenter ic p lexus, and espec ia l ly the submucous plexus, where it innervates the c ircular musc le , muco sa and submucosa (Mcintosh, 1985). Somatos ta t in is thought to be a phys io log ica l regulator of both ac id secre t ion and gastr in re lease, be cause it is a strong inhibitor of both, and because of the proximity of the D ce l ls to parietal ce l l s in the corpus and to G ce l l s in the antrum. E lect ron mic roscopy (Larrson et a l . , 1979) and doub le immunocy tochemica l sta ining techn iques (Wolfe et a l . , 1984) have revea led long cy top l a sm i c p r o ce s se s from D ce l l s abutt ing on par ieta l and G ce l l s , sugges t ing paracr ine re lat ionships. The proport ion of D ce l l s to G ce l l s is reasonab le for such a relationship. The number of parietal ce l l s is much greater than the number of D cel ls, suggest ing that this type of relat ionship is unlikely. However , parietal ce l l s may be electr ical ly coup led or in subgroups , such that somatostat in cou ld inhibit them in a paracr ine manner. Somatostat in may also inhibit only a proport ion of parietal ce l l s by a paracr ine route. A n inhibition of ac id secret ion by blood-borne somatostat in has a lso been suggested. The re is substant ia l ev idence for a funct ional l inkage be tween gastr in and somatostat in . In earl ier studies, e xogenous somatostat in wa s shown to inhibit gastr in re lease . Haye s et a l . (1974) demons t ra ted that somatostat in cou ld e l iminate the b iphas i c r e sponse of gastr in to arg in ine st imulat ion in per fused p ieces of rat antrum. Infusion of somatostat in inhibited liver extract-st imulated gastrin re lease in dogs in vivo (Konturek et al., 1976), and basa l and ca rbamy l cho l i ne - i nduced gastr in re lease from the i so la ted per fused can ine s tomach (Lefebvre et al . , 1981). The regulation of gastrin and somatostat in secret ion is mediated by both cho l i ne rg i c and non-cho l inerg i c mechan i sms . E v i d ence for a cho l inerg i c component in the regulation of gastrin and somatostat in c omes from a variety of preparat ions. For example , in the anaesthet i zed pig (O leson et al . , 1987), rat 4 antral t i ssure cul ture (Wolfe et a l . , 1984; Lucey et a l . , 1985) and iso lated per fused rat s t omach (Saffouri et a l . , 1980 ,1984a; Mc i n to sh et a l . , 1981), s t imulat ion of the vagus or st imulat ion with cho l inerg i c agon i s t s p roduced inhibit ion of somatostat in re lease and the st imulat ion of gastr in re lease . The ga s t r i n r e s p o n s e to the n i co t i n i c r e c ep t o r a gon i s t 1 ,1 -d imethy l -4 -pheny lp iperaz in ium ( D M P P ) wa s complete ly abo l i shed by hexamethon ium, a nicotinic receptor antagonist, but only partially inhibited by atropine, suggest ing that the intramural neurons were of a chol inerg ic and a non-chol inergic nature (Schuber t a n d Mahk lou f , 1982) . In the i so la ted pe r f u sed rat s t omach te t rodo tox in , an a xona l b locker , c omp l e t e l y r e ve r s ed the inh ib i t ion of somatostat in secret ion and the stimulation of gastr in secret ion in response to a peptone solut ion, suggest ing that these responses were mediated by intramural neu rons . T h e s e r e s pon s e s we re only part ia l ly b l o c ked by a t rop ine , a muscar in ic receptor antagonist, support ing the involvement of both chol inergic and non-chol inergic neurons (Saffouri et al., 1984a). The most prominent cand idate for the st imulatory regulat ion of gastr in and somatostat in secret ion is gastr in-releasing peptide (GRP ) . Schubert et a l . (1983) found that bombes in ant i serum, the amph ib ian counterpart to G R P , inhibited neural ly mediated gastr in secret ion. St imulat ion of the vagus c au s ed an inc rease in G R P and gastr in re lease, but reduced somatostat in secret ion (Nishi et a l . , 1985). G R P is a neuropept ide present in gastr ic mucosa l nerve terminals, and has been shown to st imulate somatostat in and gastr in re lease. G R P infus ion dur ing vaga l b lockade e levated postprand ia l gastr in secret ion (Greenberg , 1987). T h e sec re t ion of somatos ta t in and gastr in are not a lways coup led . P e d e r s o n et a l . (1981) ob se r v ed that v ago t om i z ed rats exh ib i ted basa l hype rgas t r i naemia without concomi tant c hange s in somatos ta t in sec re t ion . Mart inda le et a l . (1982) found that a concentrat ion of bombes in required to st imulate somatostat in secret ion had no effect on gastrin re lease. Atropine and hexamethon ium inhibited bombes in-s t imulated somatostat in re lease , but not bombesin-st imulated gastrin secret ion, suggest ing that somatostat in and gastrin secret ion are not a lways functionally l inked. The secre t ion of gast r i c somatosta t in and gastr in is a l so in f luenced directly by luminal chemica l s . Microvil l i on the cel l membrane are exposed to 5 the lumen, and prov ide the means by wh ich c h ange s in the gut contents , espec ia l ly hydrogen ion concentrat ion, can be detected. Perfus ion of the lumen in the isolated perfused rat s tomach with an alkal ine perfusate increased gastrin re lease in the basa l state (Saffouri et a l . , 1984). Per fus ion with an ac id ic per fusate c a u s e d the re lease of less gastr in and more somatostat in in the unst imulated state (Saffouri et a l . , 1984) and in response to vaga l st imulat ion (Alino et al., 1986). Pep t ides such as vasoac t i ve intest inal pept ide (VIP), secret in , gastr ic inhibitory polypept ide (GIP), the opioid pept ides, subs tance P and ga lan in are known to affect gastr in and somatostat in secret ion . Infusion of g lucagon , secret in and V IP in the isolated perfused rat s tomach c au sed a concentrat ion-d e p e n d e n t i n c r e a s e in s oma t o s t a t i n s e c r e t i o n wi th a s i m u l t a n e o u s concent ra t ion-dependent d e c r ea se in gastr in secre t ion , sugges t ing that the suppress i on of gastr in secret ion induced by the ho rmones was mediated, at least in part, by somatostat in (Ch iba et al., 1980). A study by Wol fe et a l . (1983), us ing somatos ta t in an t i se rum on sec re t i n - i nduced inhibit ion of c a rbacho l -st imulated gastr in re lease in cultured rat antral muco sa concur red with Ch i ba et al . (1980) that secret in inhibition of gastrin re lease is mediated by somatostat in. In contrast, the infusion of V IP c aused a transient increase in gastr in secret ion, but a sus ta ined inc rease in somatostat in secret ion. Saffour i et a l . (1984b) conc luded from the s imul taneous infusion of V IP and somatostat in ant iserum that V I P par t i c ipates in the regulat ion of somatos ta t in , but not of gastr in secret ion. G I P increases somatostat in re lease and inhibits gastr in re lease only when the vagus is not st imulated (Mc intosh et al . , 1983), or when gastr ic ac id secret ion is high (Hoist et al . , 1983a). G A B A induced st imulat ion of gastr in re lease and inhibit ion of somatostat in re lease is thought to be through the st imulat ion of antral postgangl ion ic cho l inerg ic neurons (Harty and Frank l in , 1986). It has been sugges ted that opio id pept ides at least partial ly mediate vaga l ly- induced inhibition of somatostat in secret ion (Mcintosh et al . , 1983), and that they may be involved in the vaga l regulation of G R P and gastr in (Nishi et al . , 1985). Subs tance P inhibited both basa l somatostat in and somatostat in re lease st imulated by G I P or isoproterenol in the isolated per fused rat s tomach (Kwok et al . , 1985), but had no effect on gastr in re lease (Kwok, unpubl ished). Ga lan in supp ressed somatostat in and gastrin re lease dose-dependent ly in the i so la ted pe r fused rat s t omach , sugges t ing that it is invo lved in the neura l 6 regulation of gastr ic endocr ine secret ions (Kwok et al., 1988). The role of the sympathet ic nervous sys tem in the control of gastr in and somatostat in secret ion is still unclear. Koop et a l . (1982) found that infusion of i soprotereno l into the iso lated, per fused rat s tomach c a u s e d the re lease of somatostat in . However , Mc in tosh et a l . (1981) found that sp lanchn i c nerve st imulat ion inc reased somatostat in secret ion only when atropine was infused c o n c o m i t a n t l y . D i r e c t s t imu l a t i o n of s o m a t o s t a t i n s e c r e t i o n by sympathomimet i c s is unaffected by atropine, suggest ing that there may be a cho l inerg i c c omponen t within the sp l anchn i c nerve inhibit ing somatos ta t in secret ion. The infusion of atropine, but not a lpha and beta receptor b lockers in h u m a n s inh ib i ted the r ise in pos tp rand ia l somatos ta t i n , sugges t i ng that cho l inerg i c , but not ad rene rg i c me chan i sms are important modu la to rs of somatostat in (Lucey et al., 1985). 7 Somatostat in in the Panc reas T h e interact ion be tween islet ho rmones is comp lex . Somatos ta t in inhibits both insul in and g lucagon secret ion. Insulin inhibits both somatostat in and g lucagon secret ion, and g lucagon st imulates both insul in and somatostat in re lease . In addi t ion, insu l in , g l u cagon and somatos ta t in inhibit their own secret ions (Weir and Bonner-Weir , 1985). However, stimulation of the pancret ic islet with different subs t ra tes doe s not y ie ld the s a m e r e sponse s of these ho rmones . S t imu la t ion with g l u cose c a u s e s an i n c r ease in insu l in and somatostat in secret ion, but a dec rease in g lucagon secret ion. St imulat ion of the panc r ea s with amino ac ids c a u s e s the re lease of al l three ho rmones . Pancreat i c polypeptide is present in the islets in specif ic ce l ls ( PP cel ls). It has been shown to affect insulin and g lucagon re lease, but only at pharmaco log ica l concentrat ions. T h e sec re t i ons of insu l in , g l u cagon and somatos ta t in are pulsat i le (Goodne r et a l . , 1982; S tagner et al . , 1982) with per iods of 10 minutes, 8.6 minutes and 10 minutes, respectively, in the can ine pancreas . Th is intermittent secret ion is thought to increase the eff iciency of g lucose d i sposa l (Bergman et al. , 1985) by preventing receptor desens i t izat ion. S tagner and Samo l s (1985 a , b) s u g g e s t e d that the p a c e m a k e r l ies in the panc r ea t i c gang l i a , a n d s imu l taneous l y innervates many is lets randomly d is t r ibuted throughout the panc reas . In man, the isolated perfused rat and canine pancreas , and in isolated rat is lets, e xogenous somatostat in administrat ion inhibited the first and s e cond pha se of insul in re lease st imulated by arg in ine and g lucose , and inhibited g lucagon re lease (Alberti et al . , 1973; Curry et a l . , 1974; E fend ic et a l . , 1974, 1976). S tud ies by Curry and Bennett (1976) indicated that the first phase of insul in re lease was 25 to 50 t imes more sens i t ive to somatostat in inhibition than the s e c ond phase . T h e s e s tud ies sugges ted that somatosta t in is a phys i o l og i ca l regu lator of insul in and g l u cagon sec re t i on . Somatos ta t i n modu la tes pos tprand ia l re lease of these ho rmones and of g l u co se leve ls (O 'Shaughnessy et al., 1985). The r e is i n c reas i ng e v i d en ce that somatos ta t i n a f fec ts g l u c agon 8 secret ion much more than it does insulin secret ion (Ger ich et al . , 1975; Itoh et a l . , 1980) . S i n c e the A ce l l is ~50 t imes more sens i t i ve to somatos ta t in inhibit ion than the B cel l , the inhibition of g lucagon may be a more important funct ion of somatostat in than the inhibition of insulin (Mandar ino et al . , 1981). Furthermore, there is parallel distribution of A and D cel ls. Fewer contacts exist be tween B and D ce l l s support ing the sugges t ion that somatos ta t in has a greater effect on g lucagon than on insulin secret ion (Orci and Unger, 1975). The sec re t i on of somatos ta t in is in f luenced not on ly by metabo l i c subs t r a te s and the sec re t i on of o ther panc rea t i c pep t ides , but a l so by neuropept ides and the autonomic nervous sys tem. VIP , secret in, subs tance P, c h o l e c y s t o k i n i n , gas t r i n , neu ro t ens i n a n d p r o s t ag l and i n E2 s t imu la te somatostat in secret ion. Endorph ins depress somatostat in secret ion (Reichl in, 1983) . V aga l s t imulat ion or st imulat ion of the rat and pig panc r ea s with cho l i ne rg i c agon i s t s inhib i ted somatos ta t in r e l ease ( Samo l s et a l . , 1978; Uvnas-Wa l lens ten et al . , 1980; Hoist et al . , 1983b, Ahren et a l . , 1986). In the can i ne panc reas , vaga l st imulat ion or acety lcho l ine i n c reased somatostat in secret ion (Ahren et al . , 1986). A l pha adrenerg ic agonists or adrenerg ic nerve st imulat ion inhibited somatostat in secret ion in s ome s tud ies (Samo l s et a l . , 1978, Hoist et al., 1983b), but had no effect in others (Lucey et al . , 1985; Nishi et al . , 1987). Be ta adrenerg ic agonists st imulate somatostat in secret ion (Samols et al . , 1978, Hoist et al , 1983b). Exocr ine secret ion of the panc reas is a lso inf luenced by islet secret ion. G l u c o s e st imulated islets receive a disproport ionate vo lume of the pancreat ic c i rcu lat ion in compa r i son to the exocr ine t i ssue ( Jansson and He l le rs t rom, 1983). Bonner-Wei r and Orc i (1982), us ing methacrylate corros ion casts , found that a proport ion of the efferent capi l lar ies from the islets se rve the exocr ine t issue before converg ing into venules. Therefore this t issue is subjected to high concentrat ions of islet hormones, espec ia l ly somatostat in (Kawai et al . , 1982), which inf luence exocr ine secret ion. The organizat ion of the islets of Langerhans var ies with the spec ies . In the rat, the Ch i ne se hamster and mouse , somatostat in-contain ing D cel ls and g lucagon-sec re t i ng A ce l l s lie in the per iphery of the islet sur round ing an insu l in-secret ing B cel l core . In man, A and D ce l l s a re g rouped together against capi l lary wal ls (Orci and Unger, 1975). 9 Us ing ant i sera against insul in, g lucagon , somatostat in and pancreat i c polypept ide to stain success i ve ser ia l sect ions of the pancreas , Bae tem et al . (1979) found that two different t ypes of is lets exist in rat. G lucagon- r i ch , pancreat ic po lypept ide-poor is lets are found mostly in the dorsa l lobe of the panc reas , whi le pancreat ic polypept ide-r ich, g lucagon poor is lets are found in the ventral lobe. The islet microanatomy appears to be important to the normal functioning of cel l secret ion. In d i spersed islet cel ls, in which the direct communicat ion and hormona l sec re t ion from other ce l l s is lost, the sec re t ion of insul in and g l u cagon were not a s respons i ve to g l u cose and arg in ine a s intact is lets (Dunbar and W a l s h , 1982; Hopcroft et a l . , 1985). Direct commun i ca t i on be tween B ce l l s o ccu r s by gap junct ions, wh ich were found to inc rease in number during B cel l secretory activity (Meda et al., 1979). Major islet secretory products are too large to pass through these junctions, however electr ic charge a n d sma l l e r mo l e cu l e s a re cont inua l l y e x c h a n g e d . G a p junc t ions may participate in the regulatory sys tem by which st imulated B ce l ls adjust their level of activity in relat ion to need . The compar tmenta l i za t ion of islet secretory products is part ial ly a ch i eved by tight junct ions be tween homo logous and hetero logous ce l ls . However , recent ev idence sugges t s that these junct ions oc cu r a s a result of the exper imenta l iso lat ion p rocedure , a s an adapt ive mechan i sm to protect the islet microdomains, and that they do not play a role in normal islet cel l function (Veld et al., 1984). Somatos ta t in has been p roposed to be a paracr ine inhibitor of insul in and g lucagon secret ion. There is much ev idence in support of this hypothes is . After somatostat in deplet ion by oral administrat ion of cys teamine , iso lated rat is lets re leased more insulin in response to g lucose st imulation than did control is lets (Kanatsuka et al . , 1984). Efendic et a l . (1980) found in fasted rats that the concen t r a t i on of somatos ta t i n requ i red to inhibit a rg in ine a n d g l u c o s e st imulated insul in and g lucagon secret ion inc reased the p l a sma somatostat in concent ra t ion to 1500 pg/ml. S ince p l a sma somatostat in is normal ly - 1 0 0 pg/ml, they sugges ted that such a high concentrat ion of somatostat in re lease wou ld only o c cu r local ly . Fur ther e v i dence for the pa rac r ine ac t ion of s oma tos t a t i n a r o s e f rom s tud i e s with s ta t i c i n cuba ted i s le ts in wh i ch somatostat in ant iserum increased insulin (Taniguchi et al . , 1977) and g lucagon 10 re lease (Barden et a l . , 1977; Itoh et a l . , 1980). Unde r these exper imenta l condit ions, it is poss ib le that the ant iserum is neutral izing the high concentrat ion of somatos ta t i n that a c cumu l a t ed in the incubat ion med i a , and not the somatostat in in the islet interstitium. Support for a local role of somatostat in was prov ided by Tabo r sky (1983) who in fused a non immunoreac t i ve ana log of somatos ta t in , and obse r ved a supp ress i on of somatosta t in re lease and an inc rease in g l u cagon and insul in re lease in anes the t i zed dogs . Ev i dence against an endocr ine act ion of somatostat in in the panc reas wa s prov ided by Bonner -We i r and Orc i (1982). Us ing ser ia l paraffin sect ions of islet per fused with India ink and immunocy to chem i ca l l y s ta ined a l ternate ly for insu l in , g lucagon, somatostat in and pancreat ic polypeptide, they found that the afferent arteriole entered the islet at discontinuit ies of the mantle of non-B cel ls . Upon entry to the B ce l l co re , the arter io le d i ve rged . There fo re somatos ta t in , g lucagon and pancreat ic polypept ide cou ld have very little inf luence on insulin secret ion v i a the circulat ion, un less the hormone wa s re leased ups t ream from the islet. Alternatively, somatostat in may affect insulin and g lucagon secret ion v ia an endoc r i ne route. S c h u s d z i a r r a et a l . (1980) f ound that in ject ion of soma tos t a t i n an t i s e rum in d o g s i n c r e a s ed pos tp rand ia l insu l in r e l e a se compared to control dogs in which non immune serum was injected, suggest ing that somatostat in is an endocr ine inhibitor. Ant ibod ies are large proteins and are a s s u m e d not to leave the c i rcu la t ion. In contrast to these f ind ings, administrat ion of somatostat in ant iserum in starved rats (Tannenbaum et al . , 1978), consc i ous baboons (Steiner et a l . , 1978) and in ear l ier dog studies by Schusdz i a r r a et a l . (1978) inc reased growth hormone, but not insul in re lease. T h e s e resul ts sugges t that somatostat in is an endocr ine inhibitor of growth hormone but not of insul in. The di f ferences in the results cou ld be due to the access ib i l i ty of the antibody to the pancreas , s ince an increase in the frequency of ant iserum administrat ion in the later dog studies (Schusdz ia r ra et al . , 1980) gave posit ive results. Other ev idence suggest ing that somatostat in is an endocr ine inhibitor is that administrat ion of exogenous somatostat in, at a concentrat ion of 10 to 20% of the pancreat i c ve in somatostat in concentrat ion, inhibited both insul in and g lucagon re lease in the per fused can ine pancreas . S i n ce such relatively low 11 concentrat ions of somatostat in affect the islet ce l ls , somatostat in probably acts through the circulat ion (Kawai et al., 1982). Immunocytochemica l staining of D ce l l s in the rat p an c r ea s s h o w e d that they p o s s e s s e d long cy top l a sm i c project ions ex tend ing to cap i l la r ies (Aponte et a l . , 1985). In unst imulated condit ions, 54% of the secretory granu les were distr ibuted in the capi l lary end of the cel l . Under st imulated condit ions the secretory granules polar ized in this region, suggest ing the re lease of somatostat in into the circulat ion. Pass i ve Immunization Studies in the S tomach and Panc reas P a s s i v e immun iza t ion or immunoneut ra l i za t ion is the con fe rment of res i s tance to an ant igen, by admin ist rat ion of ant ibod ies made in another indiv idual or an ima l . It is an important tool in invest igat ing the endogenous effect of a hormone . The results from pass i ve immuniza t ion s tud ies with somatostat in ant iserum are confl icting, although cons iderat ion must be g iven to the d i f ferences in the amount and the binding character i s t i cs of the ant i sera used , and the accessibi l i ty of the ant iserum to the t issue. Schusdz i a r r a et a l . (1980) found that repeated injection of somatostat in an t i se rum into dog s i n c reased postprand ia l gastr in, insul in and pancreat i c po lypept ide concent ra t ions , and conc luded that somatostat in ac ts a s a true hormone in the s tomach and pancreas . In anaes thet i zed rats, somatostat in an t i se rum d id not c hange p l a sma concent ra t ions of gast r in , but i n c reased growth hormone secret ion (Ch iba et al . , 1981). In the Isolated, per fused rat s tomach , infusion of the ant iserum had no effect on gastr in re lease in s ome studies (Ch iba et al . , 1981) and c au sed an increase in gastr in re lease in others (Saffouri et al . , 1979; Short et al . , 1985). Incubation of somatostat in ant iserum with rat antral muco s a in culture inc reased basa l gastr in re lease (Ch iba et al . , 1981). T he s e inc reases in gastr in re lease suggest that somatostat in exerts a cont inuous restraint on gastrin secret ion. The differences in the results could be due to the access ib i l i ty of the ant iserum to the t issue. If somatostat in inhibits gastr in by a paracr ine route, the ant ibody must enter the gastr ic interstit ium before it c an neutral ize somatostat in. S tud ies on the effect of pass ive immunizat ion on pancreat ic endocr ine 12 secret ion have been per formed in vivo, with iso lated per fused panc reas and i so la ted is lets . In v ivo s tud ies in the baboon (Ste iner et a l . , 1978) , rat ( Tannenbaum et a l . , 1978) and dog (S chusdz i a r r a et a l . , 1978) in wh ich endogenous somatostat in was neutral ized by somatostat in ant iserum revealed that growth hormone secret ion wa s inc reased , but there wa s no inc rease in insulin or g lucagon secret ion. The pass i ve immunizat ion of endogenous somatostat in in iso lated rat i s lets has cons i s tent l y y i e lded pos i t ive resul ts . Add i t ion of somatos ta t in ant iserum to the incubation med ia c aused an increase in insulin and g lucagon secret ion in response to g lucose stimulation (Barden et a l . , 1977; Tan iguch i et a l . , 1977; Itoh et a l . , 1980), and in r e sponse to am ino ac id s t imulat ion (Taniguchi et al . , 1979; Itoh et a l . , 1980). These exper iments provide ev idence that endogenous somatostat in is a phys io log ica l regulator of both insul in and g lucagon secret ion. However, as d i s cussed earlier, there is a strong argument against the use of this preparat ion to a s s e s s the paracr ine or endocr ine nature of somatostat in secret ion. The prepara t ion of cho i c e for pa s s i v e immun i za t i on s tud ies is the iso lated per fused pancreas , because it e l iminates the compl i ca t ing effect of circulat ing factors, yet maintains the integrity of the islet anatomy. Attempts at increas ing insul in re lease by pass ive ly immuniz ing against somatostat in in the iso lated per fused panc reas of mamma l s have so far y ie lded negat ive results (Sorenson et al., 1980). Poss ib ly , the ant isera concentrat ions were too low, or the ant ibod ies were inaccess ib le to endogenous somatostat in. In the ch icken pancreas , Honey et a l . (1981) found that somatostat in neutral izat ion caused a st imulation of both g lucose st imulated insulin and g lucagon re lease, suggest ing that somatostat in cont inuously inhibits A and B cel ls. It is quest ionable whether these conc lus ions can be extrapolated to the mammal ian pancreas . 13 F I G U R E 1 Digest ion of Ant ibody with Papa in and Peps in H = Heavy cha in L = Light cha in D iges t ion of the ant ibody with papa in y ie lds 2 F ab f ragments and an F c fragment. D igest ion of the ant ibody with peps in y ie lds an F(ab')2 and an Fc fragment. The Fab fragment conta ins the ant igen binding site of the ant ibody. The F c f ragment doe s not have any ant igen binding capac i ty , but can bind comp lement . 14 Monoc lona l Ant ibodies and Fab Fragments Monoc lona l ant ibod ies are ant ibod ies wh ich are der i ved from a s ingle ce l l , and are therefore d i rec ted aga ins t one ant igen ic site on a mo lecu le . B e cau se of this, they are much more speci f ic to that molecu le than ant isera, which are polyc lonal . Other advantages of monoc lona l ant ibodies are that they are relatively easy to purify, reasonably easy to produce, the cel l line is immortal a nd to p roduce them, pure ant igen is not required. The d i s advan tages of monoc lona l ant ibod ies are that they are expens i ve to produce, and the cel l l ines are relatively fragi le. Monoc lona l ant ibod ies have been u s ed as bas i c research tools in rad io immunoassays , enzyme- l inked immunosorbent a s s ay s ( E L I SA ) , a n d immunocy tochem i s t r y , a s r eagen t s in d i agnos t i c s , in the purification of ant igens by affinity chromatography and in therapeut ics. Four monoc lona l ant ibodies against somatostat in are avai lable: S O M A 3, 8, 10 and 20 (Buchan et al., 1985). S O M A 3 is u sed in rad io immunoassays . S O M A 8 and 10 are used in immunocytochemist ry . S O M A 20 has not been used for any specif ic purpose as of yet. Ant ibod ies are c omposed of 2 light cha ins (25 kd) and 2 heavy cha ins (50 kd) l inked by disulf ide bonds (fig. 1). The light cha in and half of the heavy cha in make up the F ab fragment (antigen binding). The Fab fragment is the portion of the ant ibody which conta ins its ant igen binding property, and there are 2 Fab fragments per antibody. The other portion of the antibody, c omposed of the two rema in ing ha lves of the heavy cha in is the F c (crysta l l i zable) fragment. It has the property of binding complement, for foreign cel l lysis. The ant ibody binding regions can be separa ted from the F c region by d iges t ion with papa in or with peps i n , wh i ch c l e ave at the h inge reg ion. Digest ion of the antibody with papain y ie lds two separate Fab fragments, while digest ion with pepsin g ives two attached antibody binding regions, F(ab')2- The advantages of the Fab and F(ab')2 f ragments over the intact ant ibody are that they have a lower nonspeci f ic binding due to the removal of the F c region, and they are a lso smal ler molecu les , ~50 kd and 100 kd respect ively, compared to the intact antibody which is - 1 5 0 kd. 15 Rat iona le The re is cont roversy as to whether somatostat in exerts a cont inuous restraint on gastrin re lease. In some situations, there is a functional relationship between the two hormones: inc reases in somatostat in paral le l a dec rease in gastr in re lease, and dec reases in somatostat in are assoc ia ted with increases in gastr in re lease. A s previously ment ioned, gastr in is thought to be a paracr ine effect of somatostat in. However , there is confl icting da ta a s to this relationship in the basa l state. Saffouri et al. (1979), infused somatostat in ant iserum into the i so la ted per fused rat s tomach and found an immed ia te i n c rease in gastr in re lease , support ing the hypothes is that somatos ta t in inhibits basa l gastr in re lease. Short et a l . (1985), us ing a simi lar mode l and procedure found only a signif icant increase after a 45 minute infusion of ant iserum, whi le Ch i b a et a l . (1981) did not find any effect of the ant iserum. The di f ferences in results could be due to the di f ferences in the binding capaci ty of the ant iserum used as wel l a s the accessibi l i ty of the ant isera to the t issue. In the p re sen t s tudy , m o n o c l o n a l an t i bod i e s d i r e c t ed a g a i n s t somatostat in were infused into the isolated perfused rat s tomach in an attempt to neutra l i ze endogenous somatosta t in , and the effects on gastr in re lease observed . Second ly , Fab fragment of the monoc lona l antibody, S O M A 1 0 , were made a n d in fused in s im i la r expe r imen t s . T h e s e f r a g m e n t s , b e i ng approx imate ly one-third the s i ze of the intact ant ibody shou ld more readi ly penetrate into the interstit ium of the s t omach and neutra l i ze e ndogenou s somatostat in. In the isolated perfused pancreat ic preparat ion, S O M A 10 and its Fab fragment were infused in an attempt to neutral ize endogenous pancreat ic somatostat in . The present exper iments differ from prev ious attempts in that monoc lona l ant ibodies and Fab fragments were used . For these studies, high purity ant ibodies and Fab fragments were prepared and character ized. 1 6 M E T H O D S I. Ant ibody Purif ication Methods A. Ammon ium Sulphate Precipitat ion 1. Rat ionale A m m o n i u m su lphate prec ip i tat ion is a method of c rude l y pur i fy ing ant ibodies. It is ba sed on the precipitation of proteins due to the interaction of the so lvent mo lecu les with the salt ions. A s this interact ion inc reases , the prote in-so lvent interact ion de c r ea se s . The interact ion of protein with other protein mo le cu l e s i nc reases , due to the ionic a tmosphe re sur round ing the protein, and precipitation of protein occurs a s the comp lexes inc rease in s ize. Ant ibodies precipitate between 25 and 50% saturation of ammon ium sulphate. 2. Procedure S O M A 10 (Buchan et al . , 1985), a mouse ant i-somatostat in monoc lona l ant ibody, w a s p roduced by conjugat ion of cyc l i c somatostat in-14 to keyho le limpet haemocyan in , and injection into B IO.BR S g S n mice. Sp leen cel ls from a mouse giv ing a good ant ibody response were fused with NS1 ce l l s and the hybr ids were grown. Posit ive c l ones were se lec ted and c loned out by limiting di lut ion. The cel l line wa s grown as an asc i tes tumour in irradiated outbred mice. A m m o n i u m su lpha te w a s a d d e d to a s c i t e s f lu id con ta i n i ng the monoc lona l antibody, S O M A 10, to a final concentrat ion of 25 %. Th is mixture wa s centr i fuged at 4000 rpm for 15 min, the pellet wa s d i s ca rded and the supernatant taken to 50% saturated ammon ium sulphate. The mixture was left overnight at 4 ° C and aga in centr i fuged. The precipitate w a s reconst ituted in 10 m M p h o s p h a t e , pH 6.5 , the s ta r t ing buf fer for hyd roxy l apa t i t e chromatography, and d ia lyzed overnight. 17 B. Hvdroxvlapat i te Chromatography 1. Ra t iona le Hydroxy lapat i te chromatography (Stanker et a l . , 1985) was cho sen as the method of purif ication of S O M A 10 because it is a s imple, rapid and cost-effective method of process ing a large vo lume of asc i tes fluid. Hydroxylapatite chromatography is thought to purify monoc lonal ant ibodies on the bas i s of light-cha in composi t ion var iat ions (Juarez-Sa l inas et al . , 1984). 2. P rocedure The d i a l y zed ammon i um sulphate-pur i f ied ant ibody w a s app l ied to a hydroxylapat i te co l umn (2.5 x 15 cm) , in 10 m M phosphate , 0 .02% sod ium az ide, pH 6.8. The antibody was deso rbed by increas ing the molarity to 0.5 M phosphate. The flow rate was 2 ml/min, and 1 min fract ions were col lected. A n approx imat ion of the protein content was made measur ing abso rbance at 280 nm (Pye Un i can S P 8 - 1 0 0 spect rophotometer) . F rac t ions with ab so rbances exceed ing 0.040 were pooled and tested for immunoreactivity with somatostat in by enzyme- l inked immunosorbant a s say (ELISA) . C. Enzyme- l inked Immunosorbant A s say (ELISA) The E L I S A is a sol id phase as say which prov ides a qu ick and sensi t ive determinat ion of immunoreact iv i ty and of quantif ication of protein (Voller et al . , 1976) . N inety-s ix we l l F a l c on 3912 microt itre p la tes (Bec ton D i ck i n son , Cal i fornia) were coated with 100 |il of somatostat in at a concentrat ion of 1 u.g/ml in carbonate-bicarbonate buffer, pH 9.6 (0.3 M N a H C 0 3 , 0.1 M N a 2 C 0 3 , 0.02% NaN3) . The plates were incubated overnight at 4 ° C . The plates were washed with phosphate buffered sal ine containing 0.5% Tween 20 (PBS-Tween) , pH 7.4. The P B S conta ined 137 m M sod ium chloride, 1.5 m M monobas i c potass ium phosphate, 7.2 m M d ibas ic sod ium phosphate, 2.7 m M potass ium chlor ide, and 0.02% sod ium az ide . A 100 |xl al iquot of test monoc lona l antibody, appropriately di luted in P B S - T w e e n , wa s added to each wel l , and the plates were incubated for 1.5 h at room temperature. A se cond mouse monoc lona l ant ibody, ant i -vasoact ive intestinal pept ide (VIP) ant ibody 18 (S ikora et a l . , 1986), at a di lution of 1:1000 in P B S - T w e e n w a s u sed as a control. Al l s tudies were performed in dupl icate. After wash ing with PBS -Tween , 100 \i\ alkal ine phosphatase-conjugated rabbit ant i -mouse immunoglobul in (RAMIg), diluted to 1:3000 was appl ied to the plate. After incubat ion for a further 1.5 h at room temperature, the plate was wa shed with P B S - T w e e n and deve loped with 100 nl of a lkal ine phosphatase substrate solut ion (S i gma 104-105) at a concentrat ion of 5 mg in 5 ml 10% d ie thano lamine buffer (10% d ie thano lamine, 0 .02% sod ium az ide and 0 .1% M g C l 2 - 6 H2O, pH 9.8). The react ion wa s a l l owed to p r o ceed at room temperature for 15 to 45 min, and the yel low colour deve lopment was monitored at 405 nm using a MR-580 Microe l i sa reader (Dynatech Laborator ies, Virginia). Immunoreact ive fract ions were then d ia l yzed us ing Spec t rapor 2 membrane tubing (Spect rum Med i ca l Industries Inc., Los Ange les ) aga inst P B S , without sod ium az ide . Th is tubing has a molecu lar weight cutoff between 12000 and 14000 d. D. High Per formance Liquid Chromatography (HPLC ) 1. Rat ionale H P L C is a technique for purifying and analyz ing the purity of proteins. Its a d v an t a ge s ove r conven t i ona l c h roma tog raph i c me thods a re rapidity of separat ion, reproducibil ity and resolution. Th is is a result of the high pressure that can be deve l oped and the pa s sage of the samp le through an extremely tight pack ing matrix. The components of an H P L C sys tem are a pump, sample injector, co lumn and detector (fig 2). 2. H P L C Equipment and Buffers S a m p l e s were app l ied us ing a Wa te r s U 6 K inject ion sys tem (Waters Assoc i a te s Inc., M iss i ssauga) . Protein outflow absorbance was measured by a Wa t e r s Mode l 450 Va r i ab l e Wave l eng t h detector . Fo r affinity and ge l ch romatog raphy , a b s o r b an ce was mea su r ed at 225 nm and at 280 nm, respect ive ly . The abso rbance profi les were recorded on a Hewlett Pa cka rd Integrator 3380A (Hewlett Packard , Cal i fornia). 1 9 Buffer/ Solvent B F I G U R E 2 Componen t s of an H P L C Sys tem Solvent and samp le are pumped under high pressure through the co lumn. The so lvent compos i t i on c an be va r i ed us ing a p r og r ammed grad ient maker . E lut ion of protein from the co lumn is detected by abso rbance with a var iable wave length detector. 2 0 Al l buffers were made up in HPLC -g r ade disti l led water (Mil l i-Q Reagent Wa te r sys tem, Mi l l ipore, M i s s i s sauga) , fi ltered and d e g a s s e d under vacuum. Grad ien ts were genera ted with a Wate r s 660 so lvent p rogrammer control l ing the speed of two pumps (Pump A - Waters Mode l M-45; Pump B - Waters Model 6000A) to mainta in a constant f low rate whi le so lvent concent ra t ions were var ied . 3. Affinity H P L C a. Rat ionale Affinity chromatography is based on the specif ic and reversible binding of a c ompound of interest to an immobi l i zed l igand. The l igand u s ed to b ind S O M A 10 wa s somatostat in. The compound can be deso rbed by lowering the pH of the e luent , us ing chao t rop i c agents , by denatu r ing agen t s or by increas ing the concentrat ion of the l igand in the elution buffer. T he s e agents w e a k e n the non-cova lent bonds be tween the immob i l i zed l igand and the protein of interest. b. Derivit ization of the Co lumn F ive mil l igrams of somatostat in (Peninsula) were d i sso lved in 10 ml of 1 M phospha te , pH 7. The solut ion was recyc led through an Ultraff in i ty-EP (Waters) co lumn (4.6 x 50 mm) overnight at a f low rate of 0.2 ml/min. The co lumn matrix was an epoxy base . The co lumn wa s then washed with 1 M phosphate buffer for 2 h. The effluent was tested for somatostat in content by r a d i o i m m u n o a s s a y and found to be unde tec t ab l e for somatos ta t i n - l i ke immunoreact iv i ty. c. H P L C Buffers The s amp l e w a s app l i ed to the co l umn in a so lu t ion of 0.02 M phosphate , 0.2 M sod ium chlor ide pH 7. The samp le wa s deso rbed from the co lumn with water, pH adjusted to 2 with hydrochlor ic ac id . The effluent was immediately neutral ized with sod ium hydroxide. The flow rate was 0.5 ml/min. 21 d. Samp l e Appl icat ion Al l samp les were d isso lved in P B S , without sod ium az ide. Not more than 50 |il of sample (0.3 mg) was appl ied to the co lumn at a t ime. The Ultraffinity-EP co lumn has a sample capacity of 10 to 20 mg of protein. 4. Ge l Filtration H P L C a. Rat ionale G e l filtration is a techn ique used to separate mo lecu les accord ing to molecular s ize. The co lumn contains a matrix of a particular pore s ize. Proteins larger than this s i ze cannot enter into the pores, and travel in the void vo lume of the co l umn , whe r ea s sma l l e r prote ins are reta ined within pores and their pa s sage through the matrix is impeded by the stationary phase . Prote ins are therefore eluted in decreas ing molecular s ize. b. P rocedure T h e c o l umn u s ed w a s the P ro te in P a k 3 0 0 S W (Wate r s ) . T h e fract ionation range of this co lumn is 10,000 to 500,000 d. Elut ion with 0.1 M phosphate pH 7 was performed at a flow rate of 1 ml/min. c. Samp l e Appl icat ion Al l samp les were d isso lved in P B S , without sod ium az ide. Not more than 50 (il (0.3 mg) of sample was appl ied to the co lumn at a t ime. II. Ant ibody Character izat ion A. Binding Reg ion on Somatostat in Molecu le 1. Rat ionale Somatostat in rad io immunoassays were u sed to determine the antibody binding site on the somatostat in molecu le. The rad io immunoassay (RIA) is a h igh ly sens i t i v e compet i t i ve b ind ing a s s a y for measu r i ng somatos ta t i n concen t ra t i on . It is b a s ed on the d i sp l a cemen t of rad ioac t i ve ly - labe l l ed somatostat in from its spec i f i c ant ibody by un labe l led somatostat in , either by s t anda rd s or s amp l e s . A s the concent ra t ion of un labe l l ed somatos ta t in 2 2 increases , the amount o f 1 2 5 | . S O m a t o s t a t i n binding to the ant ibody dec reases . Sepa ra t i on of the somatos ta t in -an t ibody or 1 2 5 | . S O m a t o s t a t i n - a n t i b o d y c omp l e xe s from unbound or free somatostat in and ant ibody is a ch i eved by charcoa l adsorpt ion of the unbound molecu les . S tandard curves of somatostat in-14, and its ana logues were compared . If an amino ac id posit ion is important for ant ibody binding, subst itut ion of the amino ac id would cause a dec rease in binding affinity of the antibody. 2. P rocedure The as say procedure was performed accord ing to the method descr ibed by Mc in tosh et a l . (1978) us ing monoc lona l ant ibodies to somatostat in (Buchan et a l . 1985), S O M A 3, 8 , 1 0 and 20. a. A s say Buffer A stock solut ion of somatostat in RIA buffer conta in ing 23.8 m M sod ium barbi ta l , 3.9 m M sod ium acetate, 43.6 m M sod ium ch lor ide and 0.24 m M merthiolate, pH 7.4, wa s prepared and stored at 4 ° C . For use in the assay , 0 .5% bovine se rum albumin (BSA , Mi les) and a proteolyt ic en zyme inhibitor, 1% aprotinin (Miles), were added. b. S tandards Synthet i c cyc l i c somatostat in-14 (Pen insu la) wa s d i s so l ved in 0.1 M acet ic ac id containing 0.05% BSA . Al iquots of 5 j ig were lyophi l ized and stored at -20 ° C . O n the day of the assay , an aliquot was d i sso lved in a s say buffer, and serial ly di luted to obtain the standards of 1000, 500, 250, 125, 62.5, 31.25, 15.625, 7.8 and 3.9 pg/ml for a s s ay s using S O M A 3, 8 and 20. For a s s ay s in which S O M A 10 was used as the antibody, ser ial di lutions were made to obtain the standards of 250, 125, 62.5, 31.25, 15.625, 7.8, 3.9 and 1.95 ng/ml. Synthet ic ana logues of somatostat in-14, syntheth ized by Drs. D.M. Coy a n d W . Mu rphy (Tu lane Un ivers i t y , Lou i s i ana ) , a n d soma tos t a t i n -28 (Peninsula) were d isso lved in 50 |il of 0.05 M acet ic ac id, made up to 1 ng/ml in a s s a y buffer, a nd then ser ia l ly d i lu ted to obta in the s t anda rd s a s for 23 somatostat in-14. The synthetic ana logs of somatostat in used in the RIA were as fo l lows: D E S - A L A G L Y Somatostat in-14 G L U - 4 Somatostat in-14 P H E - 4 Somatostat in-14 L E U - 4 Somatostat in-14 T H R - 5 Somatostat in-14 D -PHE-6 Somatostat in-14 D -PHE -7 Somatostat in-14 D-TRP-8 Somatostat in-14 L -5 -ME-TRP-8 Somatostat in-14 P H E - 9 Somatostat in-14 G L U - 9 Somatostat in-14 T H R - 9 Somatostat in-14 P H E - 1 0 Somatostat in-14 D-THR-10 Somatostat in-14 D-PHE-11 Somatostat in-14 D-THR-12 Somatostat in-14 L E U - 1 3 Somatostat in-14 D -CYS -14 Somatostat in-14 c. Ant ibod ies C rude asc i tes fluid contain ing S O M A 3 was thawed from storage at -20 ° C , fi ltered through 0.45 UJTI filters (Mil l ipore Corp. , Bedford, Massachusse t s ) , di luted 1 :1 in a solution of 0.9% sodium chloride, 0.5% sod ium az ide and 0 .1% B S A , and stored at 4 ° C . The antibody concentrat ion was approximately 0.4 mg/ml. On the day of the assay , this wa s di luted in a s s a y buffer to a final di lution of 1 : 1 0 6 - Th is dilution of ant ibodies was shown to be sens i t ive to a range of somatostat in-14 concentrat ions between 20 and 250 pg/ml. Fi l tered crude asc i tes fluid containing S O M A 8 (approximately 32 mg/ml) or S O M A 20 (approximately 29 mg/ml), stored at 4 ° C , were di luted 1 : 1 in a solut ion of 0.9% sod ium chlor ide, 0.5% sod ium az ide and 0 .1% B S A . On the day of the assay , S O M A 8 was diluted in assay buffer to a final dilution of 1 : 2 x 1 0 6 , and S O M A 20 to a dilution of 1 : 3 x 1 0 6 . The di lut ions of S O M A 8 and S O M A 20 we re s h o w n to be sens i t i ve to a range of somatos ta t i n -14 2 4 concentrat ions between 25 and 250 pg/ml and concentrat ions between 20 and 200 pg/ml, respectively. Hydroxy lapat i te pur i f ied S O M A 10 (1.7 mg/ml) w a s s to red in P B S containing 0.5% az ide and 0 .1% B S A . On the day of the assay , it wa s diluted to a final dilution of 1 : 1000. Th is dilution was shown to be sensit ive to a range of somatostat in-14 concentrat ions between 20 and 200 ng/ml. d . Labe l i. lodination of Somatostat in T en microl itres of 0.5 M phosphate, pH 7.5, 10 |il N a 1 2 5 l in N a O H (1 mC i , Amersham) and 10 |il ch loramine T (2 mg/ml in 0.05 M phosphate , pH 7.5) were added to 5 |ig synthetic tyr-1 somatostat in-14 (Serono) d i sso lved in 10 u.l of disti l led water. The reaction was a l lowed to proceed for 30 s e c before it was terminated with 10 |il sod ium metabisu lphi te (5 mg/ml in 0.05 M phosphate buffer, pH 7.5). One millilitre of hormone-free p l a sma and 20 mg of microfine s i l i ca ( Q U S O G 3 2 , Ph i l ade lph i a Quar t z Co . ) were a dded and the mixture centr i fuged. The supernatant was removed and the pellet resuspended in 1 ml water. Th is centrifugation and resuspens ion was repeated twice to remove the unreacted 1 2 5 | . The pellet was then resuspended in 1 ml of a mixture of acet ic ac id / acetone / water (0.1 : 3.9 : 4; v : v : v) and centr i fuged. The supernatant conta ined the 1 2 5 l - s o m a t o s t a t i n . Pe rcentage incorporat ion was measu red by measur ing the radioactivity in 10 uJ al iquots of the supernatants and the pellet. 1 2 5 l - s oma to s t a t i n (final supernant) counts % Incorporation = total counts Percent incorporat ion was usual ly about 30%. 1 2 5 | - s o m a t o s t a t i n was then di luted to 500,000 cpm/10 u.l with 0.1 M acet ic ac id conta in ing 0.5% B S A (RIA grade, S igma). Al iquots of 100 uJ were lyophi l ized and stored at -20 ° C . ii. Purif ication of l 2 5 l - S o m a t o s t a t i n On the day of the assay , lyophi l ized samp les of 1 2 5 l - s o m a t o s t a t i n were d i sso lved in 0.002 M ammon ium acetate, pH 4.6, and appl ied to a CM-ce l lu lose 25 (CM-52 , Wha tman Ltd, Ma ids tone, Eng land) co lumn (0.7 x 8 cm) previous ly equi l ibrated in the s ame buffer. 1 2 5 l - s o m a t o s t a t i n w a s de so r bed from the co lumn with 0.2 M ammon ium acetate, pH 4.6. The flow rate was 1 ml/min, two mill i l itre f ract ions were co l l ec ted and radioact iv i ty mea su r ed in a g a m m a -spectrometer . The peak fract ions and those fol lowing were neutra l i zed with s od i um hydrox ide, and d i luted with a s s a y buffer for use at 3000 to 3500 cpm/1 OOu.1. e. A s s a y Protocol O n e hundred microl i tres each of a s s ay buffer, s tandard , ant ibody and 1 2 5 | _ S O m a t o s t a t i n were a l lowed to incubate for 72 h at 4 ° C . Nonspec i f i c b inding (NSB) , the amount of 1 2 5 l - s o m a t o s t a t i n bound in the ab sence of the antibody, wa s determined for each set of s tandards. N S B and standard tubes were set up in triplicate. An t i b ody - bound a n d f ree 1 2 5 l - s o m a t o s t a t i n w e r e s e p a r a t e d by adsorpt ion to dext ran-coated cha rcoa l . One millilitre of a mixture of 0 .25% dextran T-70 (Pharmac ia) , 1.25% act ivated charcoa l (Fisher) and 1% charcoa l -extracted human p l a sma in 0.05 M phosphate buffer, pH 7.5 was added to each tube. Tube s were vortexed and a l lowed to sit for 15 min at 4 ° C . Fo l lowing centr i fugation at 3000 rpm for 30 min, the supernatant wa s decanted , and the pellet counted for 3 min in a g a m m a spectrometer (Sear le Mode l 1285). Charcoa l -ex t rac ted p l a sma was prepared from outdated human b lood (Red C ross , Vancouver , B.C.). After centrifugation of the b lood, the serum was fi ltered through sharksk in filter paper (Shle icher and Schue l l , Inc., Keene , N.H.). One percent act ivated charcoa l wa s added to the filtrate. After cont inuous stirring for 1 h at 4 ° C , the mixture was centr i fuged at 10,000 rpm for 30 min. The supernatant was f i ltered twice more, and this hormone-f ree p l a sma was separated into 10 ml fractions and stored at -20 ° C . 26 f. Ca lcu la t ions Percentage of bound 1 2 5 l - s o m a t o s t a t i n was ca lcu la ted by the fol lowing formula: T C - sample counts T C - N S B counts %B = T C T C where: %B = percent bound T C = total counts B. Spec i f ic Activity from Self D isp lacement Cu rves a. Rat ionale Spec i f i c activity is a measurement of the radioactivity in compar i son to mass . The spec i f i c activity of 1 2 5 | . s o r n a t o s t a t i n was u sed to ca lcu la te the amount of somatostat in in the radioactive tracer. Th is va lue was necessary for S c a t c ha r d ana l ys i s . To ca l cu la te spec i f i c activity, i nc reas ing amoun t s of labe l led ant igen are incubated with a constant amount of ant ibody under cond i t ions s imi lar to those emp loyed in the RIA. Spec i f i c activity is then de te rm ined by compa r i ng the ratio of bound to free t racer (B/F) at e a ch increment of tracer with B/F ratios for a normal RIA standard curve. The amount of s tandard having a simi lar B/F ratio at a g iven cpm of label g ives an est imate of the speci f ic radioactivity. b. P rocedure S imi lar materials and procedures for a somatostat in RIA standard curve were fo l lowed. In addit ion, increas ing amount of 1 2 5 | . s o m a t o s t a t i n ranging from 1000 cpm/100 jil to 100,000 cpm/100 \x\ were incubated with a constant amount of ant ibody. N S B was ca lcu lated for each concentrat ion of label, and all concentrat ions were measured in triplicate. 27 C. Scatchard Plots T h e ant igen-ant ibody interact ion c an be s tud ied by measu r i ng the binding of radioactive l igand to antibody. At equi l ibr ium: [Ab] [Ag] K D = [Ab-Ag] where: K D = dissociat ion constant Ab = free antibody Ag = free antigen Ab -Ag = antibody-ant igen comp lex Th is is represented as a hyperbol ic function when [Ab-Ag]/[Ab] is plotted against [Ag]; a s the concentrat ion of ant igen inc reases , the comp lexed to free ratio of ant ibody i n c r eases asymptot i ca l l y to app roach the total ant ibody concentrat ion. W h e n B/F is plotted against B, a l inear relat ionship is obta ined. The negative reciprocal of the s lope g ives the dissociat ion constant K D , and the X-intercept g ives the max imum bound ant igen. B wa s ca lcu lated by multiplying the %B with the total somatostat in, i.e. standard plus 1 2 5 l - l a b e l l e d somatostat in. D. Immunodiffusion Ge l s 1. Rat ionale T h e immunod i f fus ion or Ouch te lony ge l is a method of determin ing immunoreact iv i ty qualitatively. The interaction of soluble ant igen with antibody results in the formation of a lattice which precipitates when it g rows in s ize and be comes insoluble. These precipitation react ions can be carr ied out in var ious semi - so l i d med ia , eg . aga rose ge ls . We l l s , cut in the ge l , are f i l led with so lut ions of ant igen and ant ibody and these proteins diffuse through the gel surrounding the wel ls . Immunoreact ive componen t s form comp lexes , wh ich 28 prec ip i tate and present t h emse l v e s a s o paque s t reaks in the ge l . The immunodiffusion gel wa s used to classify ant ibodies. A l l ant ibody mo lecu les have a c o m m o n structure cons i s t i ng of four polypept ide cha ins , two large or heavy cha ins and two smal l or light cha ins (fig 1). An t ibod ies have been a s s i gned to f ive c l a s s e s b a s e d on the pr imary structure of their respect ive heavy chains. These c l a s ses are IgG, IgA, IgM, IgD, and IgE. IgG has severa l subc lasses . 2. P rocedure A g a r o s e (1%) w a s d i s s o l v ed in 0.01 M barb i tur i c a c i d , pH 8.2. Approx imate ly 3 ml was p laced on a sl ide and a l lowed to coo l . W h e n the gel wa s set, 7 wel ls (6 wel ls surrounding a centre well) were cut and removed. The surrounding wel l s were fi l led with 7 uJ of ant i -mouse l g G 1 , an t i - l gG2a , anti-l gG2b, ant i - lgG3, or anti-IgM (ICN Immunobiological). The centre wel l was filled with a di lution of either S O M A 3, 8, 10, or 20. A precipitation line deve loped be tween 24 and 48 h later. The t ime required for the deve lopment of the precipitat ion line depended on the concentrat ion of the proteins. W h e n the precipitation line had formed, the s l ides were dr ied. T h e s l i des were s ta ined in 0 .25% C o o m a s s i e B lue in a mixture of methanol/water/acetic ac id ( 5 : 5 : 5 , v:v:v) for 10 to 15 min. After a water rinse, the ge ls were desta ined in 10% glacial acet ic ac id for preservat ion of the gels . III. S O M A 10 Fab Fragment Product ion A. Product ion 1. Rat ionale A n antibody has two binding sites. The binding sites can be c leaved off by papain, yielding the two separate binding sites, and another fragment, the Fc (crystal l izable) fragment. Papa in c l eaves at the hinge region of the antibody, between the two sets of disulphide bonds. The antibody can a lso be c leaved by peps in , yielding two connec ted binding sites and the F c fragment (fig 1). Fab f ragments are approximately one-third the s i ze of the whole ant ibody (antibody: approximately 150,000 d; Fab: approximately 50,000 d). These antigen-specif ic 2 9 f ragments have low non-speci f ic binding character ist ics due to the removal of the F c region. 2. P rocedure S O M A 10 was found, by immunodiffusion gels, to be of c l ass l gG1 . This c l a s s of mouse antibody does not usual ly bind well to protein A, the purification co lumn. To ensure that during Fab fragment purif ication, all F c regions bound to the co l umn , S O M A 10 wa s initially app l ied to a protein A co l umn (Fab purification). The fraction of S O M A 10 that bound to the co lumn was desorbed, d ia l yzed against P B S , and concentrated us ing polyethylene g lyco l ( P E G ) to a concentrat ion greater than 1 mg/ml. The fract ion of S O M A 10 binding to protein A was d ia l y zed overnight aga inst 0.1 M phosphate , 0.004 M E D T A , pH 7.5. 2-mercaptoethano l (final concentrat ion 0.01 M) was added immediately before the addit ion of papa in. 2-mercaptoethanol was required to activate the papa in , a s wel l a s to reduce the d isu lphide bonds between the two heavy cha ins of the antibody. Mercur ipapain (18.2 mg/ml , 19 uni ts/mg prote in, S i gma ) in the form of a c r y s ta l l i z ed suspens ion in 70% ethanol, wa s added to the solution in an amount equivalent to 2% of the weight of S O M A 10. The tube was f lushed with nitrogen, c apped and p laced in a 37 ° C water bath. Th i s wa s mixed cont inuous ly dur ing the d iges t i on pe r iod . Af ter 4 h, the react ion w a s te rm ina ted with 0.3 M iodoace tamide (S igma) in 1.5 M glyc ine, 3 M sod ium chlor ide, pH 8.9. Fifty microl itres of this solution were added per millilitre of reaction mixture, bringing the final concentrat ion of iodoacetamide to 0.014 M. Be cau se this reaction is light sensit ive, it was maintained in the dark. The mixture was held on ice for 30 min, then d ia l yzed aga inst 1.5 M glyc ine, 3 M sod ium chlor ide, pH 8.9, the starting buffer for F ab purification. B. S O M A 10 Fab Fragment Purif ication 1. Rat ionale Purif icat ion of F ab fragments from the digest ion mixture was performed by affinity chromatography on protein A. Prote in A is a protein p roduced by s taphy lococc i with the unique property of binding the F c region of ant ibodies. 30 Thus , the F ab f ragments can be purif ied by adsorpt ion of the F c region of undigested and partially d igested (one binding site removed) S O M A 10, and of free F c fragments. 2. Procedure The purif ication procedure was s imi lar to that desc r i bed in Pha rmac i a Separat ion News (Vol. 13.5). The digest ion mixture was appl ied to a protein A-con juga ted S epha r o s e C L - 4 B (Pha rmac i a , Sweden ) co l umn (9 x 90 mm), which was previously equi l ibrated with a solution of 1.5 M glyc ine, 3 M sod ium ch lor ide , pH 8.9. The co lumn had a max imum capac i ty of 17 mg mouse antibody. The buffer was pumped at a rate of 0.8 ml/min. Two millilitre fractions were co l lected, and monitored for protein content by measur ing abso rbance at 280 nm in a spectrophotometer. The effluent that was not abso rbed onto the co lumn conta ined the Fab fragments. When absorbance dropped be low 0.040, the F c f ragments and undigested ant ibody were eluted us ing 0.1 M citric ac id, pH 6. Th is methodology was advantageous, a s it d id not expose the antibodies or f ragments to ext reme pHs , wh ich may be detr imenta l . The co lumn was regenerated with 0.1 M citric ac id , pH 3. W h e n not in use, the co lumn was stored in 0.5% sod ium az ide at 4 ° C . IV. Character izat ion of Fab Fragments A . S o d i u m Dode c v l Su l pha t e - Po l v a c r v l am i de G e l E l e c t r opho re s i s ( S D S - P A G E ) 1. Rat ionale SDS -Po l yac r y l am ide ge l e lectrophores is is a method of determining the pur i ty a n d the m o l e c u l a r we igh t of a d e n a t u r e d , r e d u c e d p ro te in . Po lyacry lamide se rves a s an inert matrix prevent ing turbulence and mixing of prote ins dur ing e lec t rophores i s , and a s a mo lecu la r s i eve . Ac r y l am ide mo lecu les cross- l ink forming pores. The greater the amount of cross- l ink ing, the sma l l e r are the pores of the matrix. Low cross- l ink ing of ac ry lamide is required for separat ion of high molecular weight proteins, and for stack ing gels. The purpose of the stacking gel is to al low the proteins to al ign before entering the resolv ing ge l . Greater cross- l inking is required for separat ion of proteins of lower molecular weight. S D S , an ionic detergent, denatures proteins. Proteins 3 1 which are highly charged or contain a large carbohydrate component , such as ant ibodies, may run s lower than proteins of a similar molecular weight. 2. Mater ia ls For Reso lv ing Ge l (12%): 16 ml acry lamide-b isacry lamide (30 : 0.8, Biorad) 13.4 ml disti l led water 10 ml 1.5 M tris base, pH 8.8 0.4 ml 10% S D S (Biorad) 0.2 ml 10% ammon ium persulphate (Biorad) freshly made in water 20 \i\ T E M E D (Biorad) For Stack ing Ge l (4%): 1.3 ml ac ry lamide-b i sacry lamide 6.1 ml dist i l led water 2.5 ml 0.5 M tris base, pH 6.5 0.1ml 10% S D S 0.05 ml 10% ammon ium persulphate freshly made in water 10 i l l T E M E D 3. Preparat ion of Ge l The B io rad ge l appara tus was set up us ing 0.75 mm sh ims . A 12% reso lv ing ge l wa s p repared . The ac ry lamide , water and tris buffer were comb ined , and d ega s s ed under va cuum. S D S , ammon ium persu lphate and T E M E D were added and the mixture gently stirred. Water-saturated isobutanol wa s gently layered on the ge l , the gel wa s cove red and a l lowed to polymer ize. The e x c e s s i sobutano l wa s poured off and the ge l top sur face was wa shed thoroughly with disti l led water before being blotted dry with filter paper. The 4% stack ing ge l wa s prepared similarly to the resolv ing ge l . Fifteen wel l c ombs were p laced in between the g lass plates and the stack ing ge l was poured. The ge l wa s a l lowed to set for over 1 h, the combs were removed and the ge l p laced in the reservoir tank. The tank was filled with coo led reservoir buffer, containing 0.3% tris base, 1.44 % glycine, 0 .1% S D S , pH adjusted to 8.3 with hydrochloric ac id . 32 4. Samp l e s and Standards S amp l e s were di luted to approximately 0.3 mg/ml with water. To this was a dded a vo lume of samp le buffer (2x) equiva lent to half the vo lume of the samp le . Samp l e buffer conta ined 0.25 M tr is-HCI, 2 0% glycerol , 4% S D S , 0.2 M dithiothreital (DTT) and 0 .1% bromopheno l blue. D T T reduces d isu lph ide bonds . The s amp l e s were boi led in a waterbath for 10 min to denature the prote ins . A b lank conta in ing P B S and s amp l e buffer w a s g iven s imi lar treatment. Pha rmac i a Low Molecu lar Weight S tandards were used: Phosphory lase B 94,000 d B S A 67,000 d Ova lbum in 43,000 d Ca rbon i c anhydrase 30,000 d Soybean Tryps in Inhibitor 20,100 d Lysozyme 14,400 d The standards, approximately 100 u.g of each , were d i s so l ved in 250 uJ water and 250 \i\ samp le buffer (2x) and boi led for 10 min. 5. Running the G e l The s amp l e s and s tandards were app l ied to separa te we l l s us ing a Hami l ton microsyr inge. Cons tant vo l tage at 80 mV was run through the ge l . C o l d water cont inuous ly f lowed through the tanks to keep the reservoir buffer co ld and the reservoi r buffer was cont inuous ly st irred to mainta in the buffer temperature constant in all a reas . When the gel front reached the bottom, the current was turned off and the gel was removed from the apparatus. 6. Sta in ing The ge l wa s immediate ly s ta ined in 0 .25% C o o m a s s i e blue stain in methanol/water/acetic ac id ( 5 : 5 : 5 ; v:v:v) for 1 h at 60 ° C . After a brief rinse in water, the gel was desta ined at room temperature with 10% glac ia l acet ic ac id. 3 3 B. A s se s smen t of Binding of Antibody to 1 2 5 l - S o m a t o s t a t i n bv  Chromatography on Sephadex G-100 1. Rat ionale S e p h a d e x G - 100 ( Pha rmac i a , Sweden ) ha s a f rac t ionat ion range between 4,000 and 100,000 d. 1 ^ - s o m a t o s t a t i n and S O M A 10 or S O M A 10 f r agmen t s we re i n cuba ted a n d ch roma tog raphy of the c o m p l e x e s w a s per formed on S e p h a d e x G-100 . S O M A 10 c omp l e xe s are larger than Fab comp lexes and should elute earlier. 2. Preparat ion of Ge l The appropriate weight of gel was suspended in exces s somatostat in RIA buffer, without B S A or aprotinin, at room temperature for severa l hours. The f ines were decanted , and the ge l wa s dega s s ed under vacuum. The gel was poured into a 0.9 x 30 cm co lumn, and connected to a pump. 3. S amp l e s 1 2 5 l - s o m a t o s t a t i n was purified on CM -52 as descr ibed earlier. The peak fraction was neutra l ized and di luted to approximate ly 50,000 cpm/100 One hundred microlitres of S O M A 10 (25 u.g/ml), S O M A 10 Fab fragments (17 ng/ml) or of the F c fraction (25 ng/ml) were added to 100 n' of 1 2 5 l - s oma to s t a t i n , and the vo lume made up to 1 ml in somatostat in RIA buffer, without B S A or aprotinin. O the r s a m p l e s conta in ing 1 2 5 l - s o m a t o s t a t i n , ant ibody or f ragments , and somatostat in (3.5 ng or 7 ng) were prepared to check the d isp lacement o f 1 2 5 l -somatos ta t in with somatos ta t in . The s amp l e s were a l l owed to equi l ibrate overnight at 4 ° C before chromatography. 4. Chromatography Elut ion wa s per formed with! somatosta t in RIA buffer, without B S A or aprot inin, at a f low rate of - 0 . 4 mlknin. Half millilitre fract ions were co l lected and counted on a gamma-spectrometer . 34 V. Protein Quantif ication 1. Rat ionale B C A Prote in A s s a y Reagent (P ierce Chem i ca l Company , Illinois) was u s e d to de te rm ine prote in concent ra t ion , W h e n prote in is reac ted with c o p p e r 2 + in alkal ine condit ions, it p roduces coppe r 1 +. Th is is reacted with the key component of the B C A Protein A s s a y reagent, b ic inchoninic ac id , to form a water -so lub le , purple product. Detect ion of this product and therefore the protein quantity can be made by spectrophotometr ic measurement . 2. P rocedure a. S tandard Protocol To determine protein concentrat ion in the range of 100 ng/ml to 1200 fig/ml, the fol lowing protocol wa s used . Bov ine se rum a lbumin (5%, S igma) was d i sso lved in disti l led water and stored at -20 ° C . On the day of the assay, it was thawed and diluted in P B S , without sod ium az ide, to concentrat ions of 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 mg/ml. Samp l e s were di luted in the s ame buffer to a concentrat ion between 0.4 and 1.0 mg/ml. One hundred microlitres of s tandard or sample , in dupl icate, were pipetted into a test tube. B C A reagent A conta ined sod ium carbonate , sod ium bicarbonate, b ic inchonin ic ac id and sod ium tartrate in 0.1 N sod ium hydroxide. B C A reagent B conta ined coppe r su lphate. The work ing reagent was prepared by mixing reagent A and B in a ratio of 50 : 1. Two millil itres of this working reagent were added to each tube, vor texed and incubated at 37 ° C in a shaking waterbath for 30 min. b. Micro Protocol Th is micro protocol determined protein concentrat ion in the range of 5 u.g/ml to 250 jig/ml. On the day of the assay, the 5% B S A stock was thawed and diluted in P B S , without sod ium az ide, to concentrat ions of 0.05, 0.10, 0.15, 0.20, and 0.25 mg/ml. S amp l e s were di luted in the s ame buffer to a concentrat ion between 0.1 and 0.2 mg/ml. One hundred microlitres of s tandard or sample , in dupl icate, were pipetted into a test tube. Two millilitres of the working reagent 35 were added to ea ch tube, vor texed and incubated at 60 ° C in a shak ing waterbath for 30 min. 3. Protein Determination Tube s were coo led to room temperature after incubation. Abso rbance of the co lour deve loped was read at 562 nm against P B S , without sod ium az ide. The backg round abso rbance of B C A work ing reagent wa s subt rac ted from the absorbance , and a standard curve was prepared by plotting net absorbance at 562 nm against protein concentrat ion. VI. B ioassavs A. Ac id Secret ion 1. Rat ionale A n in vivo preparat ion for measur ing gastr ic ac id secret ion was used to test the biological activity of S O M A 10 Fab fragments. Its biological activity was ba sed on its ability to reverse the inhibition of l iver extract-st imulated gastr ic ac id secret ion by exogenous ly-admin is tered somatostat in. 2. Apparatus Doub le-bar re l led gastr ic cannu lae were made in wh ich fluid cou ld be infused and dra ined from the same open ing. A tubercul in syr inge, with f lared ends se rved as the dra inage tube. The flaring prevented the removal of the cannu l a once it wa s secured . The inner tube was removable, so that particles b lock ing fluid f low cou ld be removed easi ly (fig 3). Cannu l ae were secured in p lace by stabi l iz ing them on the table top. The cannu la inflow was connected to a tube secured on a bar to prevent movement of the cannu la during f lushing of the s t o m a c h . A c i d secret ion was measu red by titration of e a ch samp le with 0.1 M sod ium hydroxide using an automatic titrator (Radiometer, Copenhagen) . Infusion of drugs was performed by means of Harvard infusion pumps (Harvard Apparatus Co. , South Natick, Mass . ) . 3 6 F I G U R E 3 Double-barre l led Gast r i c Cannu l a The cannu la was constructed such that fluid could be infused and dra ined from the s a m e open ing . The open ing was inserted into an inc is ion made in the fundus for measurement of gastr ic ac id secret ion. The f lared end of the cannu la was removeable , so that particles blocking fluid flow cou ld be easi ly d is lodged. 37 3. An ima l s Ma l e S p r a g u e - D a w l e y rats (ob ta ined f rom A n i m a l C a r e , U .B .C . ) weigh ing between 150-200 g were used in all exper iments. The an imals were housed in metal c age s (5 - 6 rats/cage) in a light control led room (12 h cycle) with free a c c e s s to laboratory rat chow and water. The rats were fasted 24 h prior to exper imentat ion. 4. Surg ica l P rocedure An ima l s we re anaes the t i z ed with 2 5 % urethane at 1.25 g/kg body weight . A neck inc is ion wa s made , a t r acheo tomy per fo rmed and the oesophagus l igated. The right' jugular ve in was cannu la ted with PE - 50 tubing (Clay Adams , Pars ippany, N.J.) fi l led with hepar in i zed sa l ine such that there was easy withdrawal of b lood. Through a smal l midline abdomina l inc is ion, the fundus was located, a gastr ic cannu la inserted and secu red . The duodenum w a s l igated 3-4 mm distal to the pylorus, and the abdomen w a s sutured to prevent e x ce s s moisture loss. 5. Exper imenta l Protocol The s tomachs were f lushed with 30 ml sa l ine to remove the remain ing contents . The rats were a l lowed to equi l ibrate for a min imum of 1.5 h post-surgery. After equil ibration, the s tomachs were f lushed with 10 ml sal ine, fol lowed by 10 ml air. At this point, exper imenta l an ima l s were admin is tered 100 |ig S O M A 10 F a b f ragments , fo l lowed by an inject ion of a sma l l vo lume of hepar in ized sa l ine through the jugular cannu la . F lush ing of the s tomach was repeated every 10 min for 30 min for col lect ion of basa l gastr ic secret ion. In all an imals , 2 j ig k g " 1 f r 1 somatostat in-14 (Pen insu la Laborator ies, Belmont, Calif.) prepared in sa l ine and 0 .01% B S A were infused for the remaining exper imental per iod. The s tomach was st imulated by a 5% liver extract (T .E .C. Chem i ca l , Inc., Myers town, Pa.) mea l in sa l ine, pH adjusted to 5.5 with 0.1 M sod ium hydroxide. Every 10 min for 90 min, the s tomach was f lushed with 10 ml liver extract, of wh ich 3 ml were retained to form the next mea l . S amp l e s were co l l ec ted and ac id content was de te rmined by titrating with 0.1 M sod ium 38 hydrox ide to pH 5.5. The concent ra t ion of S O M A 10 F a b f ragment and somatos ta t in were cho sen ba sed on prev ious exper iments us ing the intact ant ibody (Sea l et a l . 1987). B. Isolated Per fused Rat S tomach and Isolated Per fused Rat Panc reas 1. Rat ionale T h e i so la ted , vascu la r l y -pe r fused s t omach and i so la ted , vascu la r l y -per fused panc reas were used to test the effects of pass ive immunizat ion with S O M A 10 ant ibody a n d F a b f r agment s on gas t r in a n d insu l in r e l ease , respec t i ve l y . T h e s e p repara t i ons have a d v an t a ge s ove r who l e an ima l preparat ions and isolated cel l preparat ions. They preserve the anatomy of the organs, i.e. the intrinsic innervation, microvasculature, and cel l to cel l contacts. However , they remove many of the compl icat ing factors present in an intact an imal , such as the inf luence of the centra l nervous sys tem, and of circulat ing factors. In the s tomach preparat ion, perfusate entered through the coe l iac artery, and was co l lected from the portal vein. In the pancreas preparat ion, perfusate entered through the coe l iac and super ior mesenter ic arteries and was co l lected from the portal ve in. 2. Apparatus The apparatus used in these studies is shown in figure 4. Perfusate was cont inuous ly stirred and ga s s ed with water-vapour saturated 9 5 % oxygen/5% carbon dioxide to maintain the pH at 7.4. Perfusate was pumped (Masterf lex, C o l e Pa lmer ) at 2 ml/min for s t omach preparat ions and at 3 ml/min for pancreat i c preparat ions through a servo-contro l led heated water bath, which mainta ined perfusate temperature at 37 ° C , and into a bubble trap in which the temperature probe was inserted. The per fused organ was mainta ined at a constant temperature of 37 ° C by a heating pad and a 60 watt desk lamp. The preparat ion was cove red to prevent moisture loss. V e n o u s effluent samp le s we re co l l e c t ed in ch i l l ed test t ubes in a f ract ion co l l e c to r ( F R A C - 1 0 0 , Pha rmac i a , Sweden) . 3 9 Drainage F I G U R E 4 Appara tus for the Isolated Per fused Organ Preparat ion Pe r fusa te w a s con t inuous ly g a s s e d with 9 5 % 0 2 / 5 % C 0 2 - It w a s pumped through a heated water bath, which maintained the organ temperature at 37 ° C , into a bubble trap to wh ich the temperature probe and p ressure monitor were a t tached . Per fusa te entered the organ v ia the coe l i ac artery, in the s tomach p repara t ion , a n d v ia the c o e l i a c a n d mesen te r i c ar ter ies , in the pancrea t i c prepara t ion . T h e v e n o u s effluent w a s co l lec ted from the portal ve in . Drugs were in fused v ia a s idearm at the level of the aortic cannu la . 40 3. Somatostat in . S O M A 10 and the Fab Fragment Administrat ion Somatostat in (Peninsula), S O M A 10 and the F ab fragments were infused through a s idearm infusion pump (model 940; Harvard Appa ra tus Co . , Inc., Mil l is, Mass . ) . They were made up in 10 ml syr inges, and del ivered v ia PE -90 polyethylene tubing (Clay Adams , Pars ippany, N.J.) into the rubber bulb on the aort ic cannu la . For s tomach preparat ions, subs tances were made up at 20 t imes the f inal concent ra t ion , and in fused at 0.206 ml/min c aus i ng a 10% change in final vo lume. For pancreat ic preparat ions, subs tances were made up at 30 t imes the final concentrat ion, and infused at 0.103 ml/min caus ing a 3% change in final vo lume. 4. Preparat ion of Perfusate The perfusing solution conta ined: KCI 4.4 m M C a C l 2 2.5 m M M g S 0 4 1.2 m M KH2PO4 1.5 m M NaCI 120 m M N a H C 0 3 25 m M B S A (RIA grade, S igma) 0.2% Dextran (Sigma) 3 % G lucose : dextrose (F isher Scientif ic) 4.4 m M liquid glucose 4.4 mM, 8.8 m M (160 mg%), 16.6 m M (300 mg%) S tock Krebs ' solut ion, conta in ing potass ium chlor ide, ca l c ium chlor ide, magnes ium sulphate and potass ium phosphate was mixed and stored at 4 ° C . The night before the exper iment, B S A and dextran were d i s so l ved in 0.9% sal ine. For s tomach perfusions, dextrose was a lso d isso lved. On the day of the exper iment , K rebs ' solut ion and b icarbonate were added . Fo r pancreat i c perfus ions, l iquid g lucose was added to give the required concentrat ion. The solut ion was made up to volume with sal ine. The final g lucose concentrat ion of 41 the perfusate wa s measu red us ing a Be c kman G l u co s e Ana l y ze r (Beckman Instruments, Inc., Fullerton, Calif.). 5. An ima l s Ma le Wistar rats (250 - 350 g) were housed in metal c age s (5 - 6 rats) in a l ight-controlled room. They were given free a c ce s s to laboratory rat chow and water. Rats were overnight fasted (16-18 h) prior to exper imentat ion. 6. Surg ica l P rocedure a. Isolated Per fused S tomach The an ima l s were anaes the t i zed with an int raper i tonea l inject ion of sod ium pentobarbital (60 mg/kg). A midline abdomina l incis ion was made, and the abdomina l aorta located. The left renal artery and vein were doubly l igated and sec t ioned in between. The super ior mesenter ic artery was doubly l igated and cut in between. A loose ligature was p laced around the portal vein and one p l a ced more rostral ly a round the portal ve in , hepat ic artery and bi le duct. Connec t i ve t i ssue connect ing the sp leen and s tomach was c l ea red , and the pancreat ic t i ssue at tached to the s tomach was l igated and de tached ensur ing that the gastroep ip lo ic artery was left intact on the s tomach s ide. A duodena l dra inage tube wa s inserted 1 to 1.5 cm distal to the pyloric sphincter. The co lon wa s sec t i oned at the rectum to a l low remova l of the intest ines, sp l een and panc reas . The aor ta was cannu la ted with hepar in i zed sal ine-f i l led PE - 160 tubing, such that the tip was adjacent to the coe l iac artery branch from the aorta. Ca re w a s taken to prevent introduct ion of air into the an ima l . The cannu l a was secu red , and the l igature above the coe l i ac artery was t ightened to interrupt b lood flow to the s tomach . Three millilitres of hepar in ized sal ine were injected v ia the aort ic cannu la , caus ing a b lanching of the s tomach . The an imal was hemi-sect ioned. The right renal artery and the inferior v ena c a va were l igated, and a cannu la was inserted into the portal vein for col lect ion of venous effluent. The aort ic cannu l a was connec ted to the pump and the an imal was per fused with oxygenated perfusate. The per iod of anox ia was less than 60 sec . Any leaks in the preparat ion were immediately halted by c l amps or l igatures. b. Isolated Per fused Panc reas The surg ica l preparat ion of the isolated, per fused panc reas wa s simi lar to the prepara t ion of the i so la ted , pe r fused s t omach with the fo l lowing except ions: l igatures were t ied between the sp leen and the pancreas , spar ing the pancreat ic t i ssue as much as poss ib le , and the sp leen was removed. A duodena l cannu la was p laced distal to the level of the pancreas adjacent to the l igament of Treitz for dra inage. A ligature was t ied around the intest ines from the dra inage tube to the ce cum, and this iso lated p iece of gut wa s removed. The s tomach and panc reas were separa ted by ligation and sect ion ing on the s tomach s ide, ensur ing that the gastroepip lo ic artery was preserved. A ligature w a s p l aced a round the pylor ic sphincter and the antrum w a s seve red . The oesophagus , including the gastr ic artery, was doubly l igated and cut to remove the s tomach. The aorta was cannu lated with hepar in ized sal ine-f i l led PE -160 tubing, such that the tip was adjacent to the super ior mesenter ic artery branch from the aorta. The cannu la was secured, and the ligature above the coe l iac artery was t ightened to interrupt blood flow to the pancreas. VII. Statist ical Ana lys i s In all perfus ion exper iments, s ign i f icance wa s tested us ing Student 's T test for either pa i red or unpaired groups. S ign i f icance was set at p < 0.05. In line g raphs , e a ch point represented the a c cumu la t ed da ta dur ing the t ime per iod prev ious to the point. In co lumn graphs, each co lumn represented the accumula ted data during 6 or 8 min. VIII. Pept ide Quantif ication bv Rad io immunoassay A. Gastr in RIA 1. A s say Buffer The a s say buffer was made by d isso lv ing sod ium barbital (0.02 M) and B S A (5 %, RIA grade, S igma) in disti l led water, and adjusting the pH to 8.4 with hydrochlor ic ac id . 43 2. S tandards Synthet ic human gastr in (Peninsula) was stored in 200 u.1 fract ions at a concentrat ion of 100 ng/ml. For use in the assay , one aliquot was di luted with a s say buffer to 25 ml. Ser ia l di lutions were made ach iev ing concentrat ions of 400, 200, 100, 50, 25, 12.5, and 6.25 pg/ml. These standards were stored at -20 ° C in 1 ml f ract ions. O n the day of the a s say , one al iquot of each concentrat ion was thawed and used. 3. Ant iserum Ant i se rum, L-2 (a gift from Dr. G . Dockray , L iverpoo l Univers i ty), was di luted 1 : 5000 in a s say buffer, and stored in 1 ml al iquots at -20 ° C . O n the day of the assay , one aliquot was thawed and further di luted in a s say buffer to ach i eve a f inal concentrat ion of 1 : 200 ,000. L-2 b inds gastr in-17 and 34 equal ly wel l . 4. 125i-Gastr in a. lodination of Gastr in F ive microl itres of synthetic human gastr in I (Peninsula), d i s so l ved in 10 u.l 0.4 M phosphate , pH 7.4 were incubated with 0.2 mC i N a 1 2 5 l in sod ium hydroxide, and 10 u.1 ch loramine T (0.5 mg/ml in 0.04 M phosphate buffer, pH 7.4) for 1 min at room temperature. The react ion wa s terminated by the addit ion of 10 u.l sod ium metabisulphite (0.5 mg/ml in 0.04 M phosphate buffer, pH 7.4). One millilitre of 0.05 M imidazole buffer, pH 7.5 was added . b. Purif ication of 1 2 5 I - G a s t r i n The iodinat ion mixture was purif ied on a D E A E - S e p h a d e x A 2 5 co lumn (0.9 x 13 cm), wh ich was previously equi l ibrated with 0.05 M imidazo le buffer, pH 7.5. The iodination mixture was eluted at a flow rate of 2.0 ml/min with a gradient of 1 M sod ium chlor ide in the imidazole buffer (total vo lume = 55 ml). One millilitre fractions were col lected and the radioactivity measured . Standard cu rves and contro ls were prepared us ing the 1 2 5 | - g a s t r i n fract ions. Fract ions y ie ld ing the best b ind ing and d i sp lacement were d i luted in a s s a y buffer to approx imate ly 1,000,000 cpm/ml, separated into 2 ml fract ions, and stored at 44 -20 ° C for use in the assay . On the day of the assay , an aliquot was di luted to give approximately 2000 cpm/100 |i l . 5. Contro ls Poo l ed venous effluent from an iso lated per fused s tomach exper iment was di luted with assay buffer to a gastrin concentrat ion of 100 pg/ml, and stored at -20 ° C as 1 ml fractions. These were used to monitor intra- and inter-assay variabil ity. To ensure that the p resence of S O M A 10 or its F ab f ragment in the samp le did not interfere with the assay, 100 \i\ of 100 Lig/ml S O M A 10 or Fab in per fusate were added to the s tandards . No change in the s tandard curve occur red . 6. A s s a y Protocol O n e hundred microl itres each of s tandard, sample or control, ant iserum and 1 2 5 l - g a s t r i n were added to each tube, and the vo lume made up to 1 ml with a s s ay buffer. Non-spec i f i c binding tubes were made up similarly, but without ant iserum. The a s says were incubated for 48 h at 4 ° C . 7. Separa t ion Ant ibody-bound and free label led-gastr in were separated us ing dextran-coa ted charcoa l . Dextran T-70 (0.25%) and act ivated charcoa l (1.25%) were made up in 0.04 M phosphate buffer, pH 7.5, and 6.5% charcoa l -ex t rac ted p l a sma added . Two hundred microl itres were added to ea ch test tube. The a s s a y s were vortexed, a l lowed to sit for 15 min, centr i fuged at 3000 rpm for 30 min at 4 ° C , decanted and counted. 8. Ca l cu la t ions The percentage of 1 2 5 l - g a s t r i n bound (%B) in each tube was calculated, by the formula used in the somatostat in RIA. A computer program wa s used to p roduce a log-logit plot of %B ve rsus s tandard concentrat ion in pg/ml. The gast r in concent ra t ion of the unknown s amp l e s wa s de te rm ined us ing the ca l cu la ted %B. Concent ra t ions per minute were cor rec ted for vo lume and exp ressed in pg/min. 45 B. Insulin RIA 1. A s say Buffer Insulin a s say buffer was made by adding 5% charcoa l -extracted p l a sma to 0.04 M phosphate buffer, pH 7.5. 2. S tandards Rat insulin (21.3 U/mg, NOVO ) was d isso lved in a s say buffer to ach ieve a concentrat ion of 4260 u.U/ml. One millilitre al iquots were stored at -20 ° C . One aliquot was further diluted (1 : 26.6) in assay buffer to ach ieve a concentrat ion of 160 uJJ/ml, d iv ided into 2 ml fractions, and stored at -20 ° C . On the day of the a s say , the 160 u.U/ml s tandard wa s ser ia l ly di luted in a s s a y buffer to g ive standard concentrat ions of 160, 80, 40, 2 0 , 1 0 , 5, and 2.5 u.U/ml. c. Ant iserum Insulin ant iserum was ra ised in gu inea pigs by injection of unconjugated porc ine insul in emuls i f ied in F reund ' s adjuvent. The ant i serum (GP01 ) was di luted (1:10), separated into 100 u,l fractions, lyophi l ized and stored at -20 ° C . T h e s e were reconst i tuted in a s s a y buffer to ach i eve a di lut ion of 1: 5000 , separated into 1 ml fractions and stored at -20 ° C . On the day of the assay , the ant iserum wa s further di luted to ach ieve a final dilution of 1 : 10^. Th is dilution of antibody was sensit ive to insulin in the range of 5 to 160 u.U/ml. 4. 1 2 5 l - l n s u l i n a. Iodination of Insulin Porc ine insul in (Novo) was d i sso lved in 10 uJ 0.1 N hydrochlor ic ac id, and then di luted to a concentrat ion of 5 u.g/10 u,l with 0.2 M phosphate buffer, pH 7.5, in a s i l i con ized test tube. Ten microlitres of N a 1 2 5 l (1 mCi) in sod ium hydroxide, 10 nl of 0.2 M phosphate buffer, pH 7.5, and 25 u.l of ch loramine T (4 mg/ml in 0.2 M phosphate buffer) were added , and the oxidat ive reaction was a l lowed to proceed for 10 sec . Sod ium metabisulphite (100 u.l at 2.4 mg/ml in 0.2 M phosphate buffer) was added and gently mixed for 45 s e c to stop the react ion. 4 6 b. Purif ication o f 1 25| . |nsu l i n O n e percent potass ium iodide (50 uJ at 10 mg/ml in 0.2 M phosphate buffer) was added to the iodination mixture, and 0.04 M phosphate buffer was added to make up to a vo lume of 2 ml. Ten mil l igrams of microfine si l ica ( Q U S O G-32) were added to adsorb the iodinated insul in. The mixture was vortexed and centr i fuged for 15 min at 3000 rpm. The supernatant was decanted and the pel let r e su spended in 3 ml dist i l led water. The mixture w a s vor texed and cen t r i f uged a s be fore . The superna tan t w a s d e c a n t e d and the pel let r e suspended in 3 ml ac id ethanol (1500 ml 9 5 % ethanol , 500 ml water, 30 ml concentrated HCI, stored at 4 ° C ) , vortexed, and centr i fuged. The supernatant, conta in ing the 1 25|_ j n su l i n , was decanted into a container, 1.5 ml water and 2 ml ac id ethanol were added , and the label was stored at -20 ° C . Incorporation of the 1 2 5 l wa s determined by count ing al iquots of the supernatants and the s i l i ca and correct ing for vo lume. Percent incorporat ion wa s ca l cu la ted as fo l lows: label counts + pellet counts % incorporation = total counts For use in the a s say , the 1 2 5 | . j n s u | j n w a s d i lu ted to approx imate ly 10,000 cpm/100j i l in assay buffer. 5. Contro ls Insulin controls were made from perfusate from a rat pancreas st imulated with 17.8 m M g lucose and 10 m M arginine for 30 min. The total venous effluent was poo led and the insul in content determined by RIA. The effluent was then di luted in a s s ay buffer to ach ieve a concentrat ion of 60 |iU/ml, and stored in 1 ml al iquots at -20 ° C . To ensu re that S O M A 10 or the F ab f ragments in the samp le did not interfere with the determinat ion of insul in, 50 jig/ml of the ant ibodies di luted in perfusate, were added to the s tandard curve test tubes . No change in the standard curve occurred. 4 7 6. A s say Protocol S amp l e s were di luted in a s say buffer to concentrat ions between 25 and 100 u,U/ml. One hundred microl itres of each standard, control or samp le and ant ibody were added to a test tube. A s s a y buffer was added to make up a vo lume of 0.9 ml, and the mixture was incubated at 4 ° C for 24 h. The following day, 1 2 5 l - i n s u l i n w a s p repared by di lut ion with a s s a y buffer to ~10,000 cpm/100 uJ, and 100 |il were added to each tube. The a s say was a l lowed to incubate for a further 24 h at 4 ° C . Nonspec i f i c binding (NSB) was determined in the absence of antibody. 7. Separa t ion An t i body -bound and free 1 2 5 | . j n s u | j n w e r e s epa ra t ed by means of dext ran-coated charcoa l . Two hundred microl itres of 0.5% dextran T-70 and 5% act ivated charcoa l in 0.04 M phosphate buffer, pH 7.5, were added to each test tube. The mixture was vortexed, a l lowed to sit for 15 min and centr i fuged at 3000 rpm for 30 min. The supernatant was decanted, and the pellet, containing the free 1 2 5 l - i n s u l i n was counted for 2 min. 8. Ca l cu la t ions The percentage of 1 25 | - j n su l i n bound (%B) in each tube was ca lcu lated, using the formula used in the somatostat in RIA. A computer program was used to p roduce a log-logit plot of %B ve r sus s tandard concent ra t ion in u.U/ml. Insulin concent ra t ion of the unknown s amp l e s wa s de te rm ined us ing the ca l cu l a ted % B . Concen t ra t i ons per minute were e x p r e s s ed in ( iU/min by multiplication by vo lume, to account for f luctuations in perfusion flow rate. IX. Immunocvtochemistrv (ICC) A. Rat ionale I CC wa s used for two purposes : 1) to ensure that the S O M A 10 Fab f ragments were immunoreact ive and; 2) to locate the distribution of S O M A 10 and its Fab fragment in the s tomach and pancreas of perfused animals. 48 B. Ant ibody Local izat ion in T i ssue 1. Preparat ion of T i s sue sect ions T i s sue was exc i sed and f ixed in Bouin 's solut ion (25% formaldehyde, 75% picric acid) for 1 - 2 h. The solution was replaced with 70% ethanol for 1 h. The t issue was stored in fresh 70% ethanol. The f ixed t issue was cut into smal ler p ieces , and p laced in casset tes for au tomat i c p r o ce s s i ng (H is tomat i c T i s s u e P r o c e s s o r Mode l 166, F i sher ) . P rocess ing involved dehydrat ion of the t issue in 80%, 90%, and 100% ethanol for 30 min each , defatting the t issue with two washes of xy lene for 1 h each and finally, infiltration of the t issue with paraffin wax. The t i ssue wa s embedded in wax. The wax b locks were f rozen overnight, 5 j im sec t ions were cut with a microtome (1130/Biocut, Re icher t - Jung) , mounted, and p l aced on a heat ing block (37 - 40 ° C ) overnight to fix them onto the sl ide. a. P rocedure for Test ing Antibody and Fragment Immunoreactivity S l i de sec t i ons of rat panc reas were p l aced in xy lene for 10 min to remove the wax, and in petroleum ether for 2 min to remove the xylene. Normal h o r s e s e r u m ( D i m e n s i o n L abo r a t o r i e s , M i s s i s s a u g a , On ta r i o ) , at a concentrat ion of 1 : 100 was appl ied to dec rease non-spec i f ic b inding of the ant ibodies. After 30 min, this was removed with 3 washes in P B S with 0.004% az ide (PBS-az ide ) The ant ibodies to be tested, S O M A 10, the F ab fragments and the F c fraction were di luted in PBS - a z i d e ( P B S with 0.004% azide) to 10 u.g/ml, a nd p l a ced onto the s l ide t o ' c o v e r the entire t i s sue sec t ion . The ant ibod ies were a l l owed to incubate overnight at 4 ° C . The s l ides were washed 3 t imes in PBS -a z i de , and normal horse serum (1 : 100 in PBS-az i de ) w a s app l i ed for 30 min to de c r ea se nonspec i f i c b ind ing. The s l ides were wa shed aga in . Biotin-spacer-aff inity purif ied goat ant i -mouse immunoglobul in ( J a ck son Immunoresearch Lab. , Wes tg rove , PA ) wa s di luted 1 : 3000 and appl ied onto the sect ion. Th is second antibody was a l lowed to incubate for 1 h before it wa s w a s h e d off. F luo resce in av id in D (Vector Laborator ies , Inc., Bur l ingame, Calif.) was di luted 1 : 2000 and 1 : 3000 in PBS - a z i d e and appl ied to the sect ions. The third layer was a l lowed to incubate for 1 h. After wash ing 3 t imes, the s l ides were protected by covers l ip, and v iewed under a f luorescence microscope (Axiophot, Ze i s s , W . Germany) . Photographs of the t i ssues were taken us ing Fuj ichrome film. b. Procedure for Locat ing Ant ibodies in Per fused Organs T i s s u e sec t i ons of pe r fused o rgans ( s tomach and panc reas ) were subjected to the s ame treatment a s non-perfused organs (sect ion VIII.B.I.a.) with the excep t i on that the first ant ibody ( S O M A 10 or F a b f ragments) appl icat ion w a s omitted, and the s e cond ant ibody, the b io t in -spacer -AP goat ant i- immunoglobul in, was a l lowed to incubate for 2 h. C . A s se s smen t of Nonspec i f i c Binding to Co l l agen 1. Rat ionale It wa s found that there was a large amount of F ab fragment binding to structures in the per fused t i ssue that conta ined co l lagen. The possibi l i ty that Fab f ragments bind nonspecif ical ly to co l lagen was a s s e s s ed . 2. P rocedure Two mil l i l i tres of rat tail co l l agen (approx imate ly 5 g/l, pH 3) were incubated in a 35 mm culture plate for 3 h at room temperature. The fluid was decanted , and 1 drop of 1 M sod ium hydroxide was p laced on the cover of the inverted plate to neutral ize the co l lagen. The plate was a l lowed to dry. S O M A 10 and Fab f ragments at 1 u,g/ml were incubated on the plates overnight at 4 ° C . Mou s e ant i -VIP (1 u.g/ml) and P B S - a z i d e were u s ed a s contro ls . The plates were deve loped by the method descr ibed in sect ion IX.B.2., protected by covers l ip and v iewed under a f luorescence microscope. 50 A P P E N D I X T O M E T H O D S R E A G E N T G R A D E S U P P L I E R Acet ic ac id Aristar B D H Ace tone B D H Aga r o s e B io rad Ammon ium acetate B D H Ammon i um sulphate F i sher Barbituric ac id B D H Bov ine serum albumin RIA S i g m a Ca l c i um chlor ide B D H Ch lo ram ine -T Eas tman Coomas s i e blue stain B io rad Dextran T-70 Bake r D ie thano lam ine Bake r Dithiothreitol F i sher Ethano l (95%) S t an chem G l u co se (powder) B D H G l u co se (liquid) C l i n i ca l Abbott G l yc ine F i sher Hepa r i n F i sher Hydrochlor ic ac id Aristar B D H Imidazole S i g m a lodoace tamide S i g m a Magnes i um chlor ide F i sher Magnes i um sulphate F i sher 2-Mercaptoethano l S i g m a Merthio late Eas tman Methano l F i sher Petro leum ether B D H Po lyethy lene g lyco l Bake r Po tass ium chlor ide B D H Po tassu im phosphate Bake r Sod ium acetate F i sher Sod ium barbital RIA B D H Sod ium bicaronate Bake r Sod ium chlor ide B D H Sod i um hydroxide F i sher Sod ium metabisulphi te F i sher Sod i um pentobarbitol (Somnitol) M T C Sod ium phosphate , d ibas ic Bake r Sod ium phosphate, monobas i c B D H Tris base S i g m a Tris HCI S i g m a Tween -20 S i g m a Xy l ene F i sher N.B. Un less otherwise indicated, chemica l s were reagent grade. 5 1 R E S U L T S I. Purif ication of S O M A 10 The elut ion profile of ammon ium su lphate prec ip i tated S O M A 10 from hydroxylapat i te is shown in f igure 5. S O M A 10 e luted in the s e cond peak. Poo l ed fract ions from each peak were tested by EL ISA , and only the se cond peak was found to contain immunoreactivity to somatostat in. F rom one millilitre of asc i tes, 2 to 3 mg of purified S O M A 10 was extracted. II. A s ses smen t of Purity of S O M A 10 A. Affinity Chromatography Purity of S O M A 10 at var ious s tages of purification, a s crude ascites, after s a t u r a t e d a m m o n i u m s u l p h a t e p r e c i p i t a t i o n ( S A S ) , a n d f o l l ow i ng hydroxy lapat i te pur i f icat ion (HAP) , wa s a s s e s s e d by affinity H P L C on an Ultraff inity-EP co lumn. The sample was appl ied to the affinity co lumn with 0.02 M phosphate, 0.2 M NaCI, pH 7, and the ant ibody was deso rbed us ing water, pH adjusted to 2 by hydrochlor ic ac id (fig 6 a - c). Chromatography of 30 (il (0.09 mg) H A P S O M A 10, 15 nJ (0.09 mg) S A S S O M A 10, and 9 u.l (0.26 mg) crude asc i tes was performed on the Ultraffinity-EP co lumn. The integrated a rea under the peaks gave an indication of the proportion of protein in the peaks. An approximation of the S O M A 10 content of the crude asc i t e s , S A S S O M A 10, and H A P S O M A 10 wa s 52%, 75%, and 9 3 % respect ively. The immunoreact iv i ty of half millilitre fract ions of the effluent from the Ultraff inity-EP co lumn was tested by E L I SA on plates coated with somatostat in. Activity was found in the second peak eluting from the co lumn, while no activity was found in the first peak. B. Ge l Filtration Chromatography S im i l a r S O M A 10 s amp l e s were app l i ed to a Prote in P a k 3 0 0 S W co lumn. P e a k s eluted at 9.3, 12, 14.5 and 22 min. The peak conta in ing the S O M A 10 immunoreact iv i ty, determined by EL ISA , e luted at 14.5 min (fig 7 5 2 F I G U R E 5 Purif icat ion of Ammon ium Sulphate Precip i tated S O M A 10 on Hydroxylapat i te Fifteen millilitres (~100 mg) of ammon ium sulphate precipitated S O M A 10 was app l i ed to a hydroxy lapat i te c o l umn (2.5 x 15 cm) with 10 m M sod i um phosphate , 0 .02% sod ium az ide, pH 6.8. Pur i f ied S O M A 10 was deso rbed by increas ing the molarity to 0.5 M phosphate. The protein content in the second peak is not truly rep resen ted by the a b so r ban ce at 280 nm, b e c au se the max imum absorbance was reached. ^-F I G U R E 6 Aff inity H P L C E lut ion Prof i le and E L I S A Act iv i ty of S O M A 10 at Va r i ous Purif icat ion S teps S O M A 10 was appl ied to the Ultraff inity-EP co lumn iff 0.02 M phosphate, 0.2 M NaCI , pH 7-and deso rbed from the co lumn with water in wh ich the pH was d e c r e a s e to 2 by HCI. The f low rate wa s 0.5 ml/min. Prote in elut ion was detected at 225 nm. The effluent was co l lected as 0.5 ml fract ions, neutral ized with N a O H and app l i ed to E L I S A p lates coa ted with somatos ta t in . E L I S A readings were measured at 405 nm and are represented by squares . a. N ine microlitres of asc i tes fluid containing S O M A 10 (0.26 mg) were appl ied to the co lumn. b. Thirteen microl itres of S A S S O M A 10 (0.09 mg) were appl ied to the co lumn. c. Fifty microlitres of H A P S O M A 10 (0.09 mg) in P B S were appl ied to the co lumn. 5 4 1 0 20 30 TIME (min) b. TIME (min) \ TIME (min) } 5 6 F I G U R E 7 G e l Filtration H P L C Elut ion Profi le and E L I S A Activity of S O M A 10 at Var ious Purif icat ion S teps S O M A 10 wa s appl ied onto the 300 S W co lumn in 0.1 M phosphate, pH 7, and eluted with the s ame buffer. The flow rate wa s 1.0 ml/min. Protein elution was detected at 280 nm. The effluent was co l lected as 1.0 ml fract ions and appl ied to E L I S A plates coa ted with somatostat in. E L I SA read ings were measu red at 405 nm and are represented by squares . a. N ine microlitres of asc i tes fluid containing S O M A 10 (0.26 mg) were appl ied to the co lumn. b. Thirteen microl itres of S A S S O M A 10 (0.09 mg) were appl ied to the co lumn. c. Fifty microlitres of H A P S O M A 10 (0.09 mg) in P B S were appl ied to the co lumn . 5 7 0 1 0 20 30 T I M E (min) TIME (min) 59 a - c). The negat ive def lect ion in peaks elut ing at 22 min may be due to changes in the refractive index, a s a result of the salts in the sample . A n approx imat ion of the S O M A 10 content of the crude asc i tes , S A S S O M A 10, and H A P S O M A 10 was 38%, 89%, and 92% respect ive ly. The activity of S O M A 10 at var ious purification s tages was measured by EL ISA . The concentrat ion giving half max imum reading for crude asc i tes was 25 u.g/ml, for S A S S O M A 10 and H A P S O M A 10 it was - 1 0 ng/ml (figure 8). III. Character izat ion of Ant ibodies S O M A 3, 8 , 1 0 , and 20 were character ized accord ing to the region of the somatostat in to which they bind, the antibody classi f icat ion, and by determining their d issoc iat ion constants and max imum binding capac i t ies . F igure 9 shows the amino ac id sequence of somatostat in-14. Ana logs of somatostat in-14 with amino ac id subst i tut ions were used as s tandards in the somatostat in rad io immunoassay. The affinities of somatostat in ana logs to the ant ibody re lat ive to somatos ta t i n -14 were ca l cu l a t ed by c ompa r i ng the concen t r a t i on of somatos ta t i n ana l ogue g iv ing 5 0 % d i s p l a c emen t in a somatosta t in RIA s tandard curve to that of somatostat in-14 (table 1). The important pos i t ions of the molecu le , i.e. the pos i t ions at wh ich chang ing the am ino ac id r educed b ind ing, represent the ant ibody b ind ing reg ion on the somatostat in molecu le. For S O M A 3 and S O M A 8 the important amino ac ids are in posit ion 4 and, 6 to 12 (fig 10a and b). For S O M A 10, the binding region is at amino ac id position 5 to 12 (fig 10c), and for S O M A 20, it is at position 4 to 12 (fig 10d). S O M A 3, 8, 10 and 20 were c lass i f ied using an immunodiffusion gel , and ant ibodies against the var ious antibody c l asses , i.e. ant i - lgG1, ant i - lgG2a, anti-lgG2b, ant i - lgG3 and ant i- lgM. Al l the S O M A ant ibodies were of the c l ass l gG1 . The impurit ies of the asc i tes fluid made it difficult to c lassi fy S O M A 3, 8 and 20. However, the strongest band formed against ant i - lgG1. The spec i f i c activity of the 1 2 5 | - s o m a t o s t a t i n , de te rm ined from self-d isp lacement curves, was found to be 880 pCi/pg or 1.52 x 10^ pCi/pmole. The theoret ical 1 2 5 l - s o m a t o s t a t i n speci f ic activity, assuming 1 2 5 l speci f ic activity to be 13 u.Ci/u.g was 939 pCi/pg or 1.62 x 1 0 6 pC i /pmole . The exper imenta l 6 0 F I G U R E 8 Compar i son of S O M A 10 Activity on E L I SA 1 2 3 4 5 6 7 8 9 10 11 12 13 14 ALA - GLY - CYS - LYS - ASN - PHE - PHE - TRP - LYS - THR - PHE - THR - SER - CYS I . : I F I G U R E 9 Amino Ac id Sequence of Somatostat in-14 6 2 T A B L E 1 Relat ive Affinities of Somatostat in Ana logs to Somatostat in-14 (average of at least 2 exper iments) S O M A 3 S O M A 8 S O M A 10 S O M A 20 Somatostat in-14 100% 100% 1 0 0 % 1 0 0 % D E S - A L A G L Y Somatostat in-14 4 0 % 3 4 % 4 6 % 2 9 % G L U - 4 Somatostat in-14 15% 5% 4 2 % 8% P H E - 4 Somatostat in-14 . 0 % 3 2 % 4 7 % 0% L E U - 4 Somatbstat in-14 7% 1 5 % 3 3 % 5% T H R - 5 Somatostat in-14 4 7 % 6 4 % 0% 3 4 % D-PHE-6 Somatostat in-14 0% 0% 0% ^% D-PHE-7 Somatostat in-14 0% 0% 0% 0% D-TRP-8 Somatostat in-14 0% 0% 0% 0% L -5 -ME -TRP -8 Somatostat in-14 6 1 % 4 8 % 1 3 % 8 6 % P H E - 9 Somatostat in-14 0% 0% . 0 % 0% G L U - 9 Somatostat in-14 4% 1 0 % 0% 0% T H R - 9 Somatostat in-14 0% 0% 0% 0% P H E - 1 0 Somatostat in-14 0% 0% 2 2 % 0% D-THR-10 Somatostat in-14 0% 0% 1 8 % ' 0% D-PHE-11 Somatostat in-14 0% 0% 0% 0% D-THR-12 Somatostat in-14 2 8 % 2 3 % 1 9 % 0% L E U - 1 3 Somatostat in-14 112% 138% 6 2 % 154% D -CYS - 14 Somatostat in-14 6 6 % 4 5 % 44% 9 5 % Somatos ta t in -28 3 7 % 6 2 % 106% 3 9 % 63 F I G U R E 10 Relat ive Affinity of Somatostat in Ana logs to Somatostat in-14 Somatos ta t in r ad i o immunoassays were per formed us ing somatostat in-14 and somatostat in ana logs as s tandards . The molar va lues of the ana logs giv ing 5 0 % d i sp l a cemen t were c ompa r ed , and e xp r e s s ed as pe r cen tages of the somatostat in-14 molar value giving 50% disp lacement . CT RELATIVE AFFINITY (%) S S - 1 4 DES -ALA G L Y G L U 4 P H E 4 LEU 4 T H R 5 D-PHE 6 D - P H E 7 D-TRP 8 L -5 -ME-TRP 8 P H E 9 G L U 9 T H R 9 P H E 1 0 D-THR 10 D-PHE 11 D-THR 12 LEU 13 D-CYS 14 S S - 2 8 RELATIVE AFFINITY (%) S S - 1 4 DES -ALA GLY G L U 4 P H E 4 LEU 4 T H R 5 D-PHE 6 D-PHE 7 D-TRP 8 L -5 -ME-TRP 8 P H E 9 GLU 9 T H R 9 P H E 10 D-THR 10 D-PHE 11 D-THR 12 LEU 13 D-CYS 14 S S - 2 8 a. RELATIVE AFFINITY (%) SS-14 DES-ALA GLY GLU4 PHE 4 LEU 4 THR 5 D-PHE 6 D-PHE 7 D-TRP 8 L-5-ME-TRP 8 PHE 9 GLU9 THR 9 PHE 10 D-THR 10 D-PHE 11 D-THR 12 LEU 13 D-CYS 14 \ SS-28 p RELATIVE AFFINITY (%) L-5 SS-14 DES-ALA GLY GLU4 PHE 4 LEU 4 THR 5 D-PHE 6 D-PHE 7 D-TRP 8 ME-TRP 8 PHE 9 GLU9 THR 9 PHE 10 D-THR 10 D-PHE 11 D-THR 12 LEU 13 D-CYS 14 SS-28 o 6 6 spec i f i c act iv i ty is equ iva lent to 2442 cpm/pg , a s s u m i n g 8 0 % count ing eff ic iency. Th i s va lue was used to ca lculate the amount of somatostat in in the 1 2 5 l - s o m a t o s t a t i n for rad io immunoassays. The total counts for the somatostat in r a d i o i m m u n o a s s a y we re an a ve r age of 3200 c p m . T h e somatos ta t i n contr ibuted by the label was 1.3 pg. Th is va lue was used for calculat ion of the total amount of somatostat in, necessary for Scatchard analys is . The results from the Sca t cha rd plots are shown in table 2. was ca l cu la ted by tak ing the negat ive rec iproca l of the s lope of B/F ve r sus B. M a x i m u m bind ing capac i t y w a s ca l cu l a ted by tak ing the X- in tercept and dividing by the antibody concentrat ion. IV. Fab Product ion Hydroxy lapat i te purif ied S O M A 10 was initially app l ied to protein A to se lect ant ibod ies with the ability to bind to protein A (fig 11). Fract ions with abso rbance at 280 nm greater than 0.040 were pooled, d ia l yzed against P B S and the protein concentrat ion measured . The first portion that d id not bind to the co lumn, peak A, was saved . Between 47 and 5 5 % of the protein bound to the co lumn. Th is portion, peak B, was d igested with papain. P e a k s A and B were compa red to S O M A 10 by RIA (fig 12). S imi lar concentrat ions of samp le ant ibody were used . S O M A 10, peak A and peak B gave zero b indings of 33%, 13% and 32%, respect ively. The standard curve us ing peak B as the antibody was simi lar to that using H A P S O M A 10. P eak A gave a standard curve which was relatively flat. B ind ing of S O M A 10, prote in A-pur i f ied peak A and B were a l so compa red by E L I S A (fig 13). The concentrat ions required for half max imum abso rbance at 405 nm for S O M A 10, peak A and peak B were 10 ng/ml, 25 |ig/ml and 5 jig/ml, respectively. Papa in -d iges ted S O M A 10 peak B was purif ied on a protein A co lumn (fig 14). The first peak conta ined the Fab fragments. The second conta ined Fc fragments, und igested and partially d igested S O M A 10. Between 58 and 65% of the protein wa s recovered as F ab f ragments . Therefore , there was an average of 26% yield of Fab fragments from the original H A P S O M A 10. 6 7 E c o CO CM LU o -z. < CQ CC o CO CQ < •i 1.5 M glycine 1.4- 3 M NaCI, pH 8.9 0.1 M citric acid Y P H 6 0.1 M citric acid pH3 ^ 4 0 V O L U M E (ml) 8 0 F I G U R E 11 . Elution Profi le of H A P S O M A 10 from Protein A H A P S O M A 10 wa s app l ied to a Prote in A co lumr f with a solut ion of 1.5 M glyc ine, 3 M NaCI, pH 8.9, and deso rbed with 0.1 M citric ac id , pH 6. Peak A conta ined 4 4% of the original protein. P eak B conta ined 5 1 % of the original protein. F ive percent of the original protein was not recovered. 6 8 Q ID 40 -I 30 -20 -10 -S O M A 10 Peak A Peak B 10 100 [ S O M A T O S T A T I N ] ng/ml 1 0 0 0 F I G U R E 12 Compar i son of S O M A 10 RIA Standard Cu rves Somatos ta t i n s t anda rd cu rves were p roduced us ing H A P S O M A 10 (open square) , Prote in A-pur i f ied S O M A 10 peak A (fil led d i amond) and B (filled square) as ant ibodies. 69 [ S O M A 10] |ig/ml F I G U R E 13 Compar i son of Protein A-purif ied S O M A 10 Activity on E L I SA 70 V O L U M E (ml) F I G U R E 14 Elution Profi le of Papa in-d igested S O M A 10 from Protein A H A P S O M A 10, peak B was d iges ted with mercur ipapa in for 4 hours. The d igest ion mixture was app l ied to a Prote in A co lumn with a solut ion of 1.5 M glycine, 3 M NaCI, pH 8.9, and deso rbed with 0.1 M citric ac id, pH 6. The peaks conta in ing the Fab f ragments and the F c fraction conta ined 5 8 % and 2 1 % of the or ig inal protein, respect ive ly. Twenty-one percent of the or ig inal protein was not recovered. T A B L E 2 Binding Capac i ty and Dissoc iat ion Constant (KD) of S O M A Ant ibod ies Binding Capac i ty KD S O M A 3 2.58 ng/mg 3.2 x 1 0 " 1 2 mol/l S O M A 8 1.78 ng/mg 1.7 x 1 0 " 1 2 mol/l S O M A 10 8.3u.g/mg 2.2 x 1 0 " 9 mol/l S O M A 20 1.98 ng/mg 3.1 x 1 0 ' 1 2 mol/l 72 V. Identification of Fab Fragments A. Sod ium Dodecv l Su lphate-Po lvacrv lamide G e l E lect rophores is S amp l e s of S O M A 10, protein A-pur i f ied S O M A 10 peak A and B, the papa in d igest post-react ion, and the purif ied Fab f ragments and F c fraction were denatured and reduced, and e lec t rophores is on a SDS-po l yac ry l am ide ge l wa s performed (fig 15). The gel wa s sta ined with C o o m a s s i e B lue stain, and des ta ined in a mixture of acet ic ac id and methano l . Bands were found corresponding to the molecular weights shown in table 3. B. Chromatography of Incubation Mixtures of Ant ibody and 1 2 5 l - S o m a t o s t a t i n  on Sephadex G-100 Ch roma tog r aphy of an incubat ion mixture of S O M A 10 and 1 2 5 | . somatostat in wa s performed on Sephadex G-100 . The elution profile is shown in f igure 16a. The first peak, conta in ing the comp l exe s of S O M A 1 0 - 1 2 5 l -somatostat in, was maximal at 9 ml, and the second , contain ing unbound 125|. somatostat in was maximal at 25 ml. l 2 5 | - soma to s t a t i n was partially d i sp laced by 3.5 u\g/ml somatos ta t in . Comp le t e d i sp l acement of 1 2 5 l - s o m a t o s t a t i n required 7 u.g/ml somatostat in. The elut ion profile of the incubat ion mixture conta in ing Fab f ragments and 1 2 5 l - s o m a t o s t a t i n is s hown in f igure 16b. T h e peak e lut ion of the comp lexes was at 15 ml, the unbound l 2 5 l - s o m a t o s t a t i n eluted at 25 ml, and 3.5 u,g/ml of somatostat in completely d i sp laced the 1 2 5 l - s o m a t o s t a t i n . F igure 16c shows the elution profile of the incubat ion mixture containing the F c fraction and 1 2 5 l - s o m a t o s t a t i n . Th is Fc fraction conta ined F c fragments, und iges ted and partially d igested (one binding site removed) S O M A 10, and eluted between 9 and 20 ml. The s e cond peak, unbound 1 2 5 | - s o m a t o s t a t i n , p e a k e d at 2 5 ml . 3.5 |ig/ml of somatos ta t in comp le te l y d i s p l a c ed the radioact ively- label led somatostat in. 1 2 5 l - s o m a t o s t a t i n a lone was appl ied to the co lumn and eluted at 25 ml. O n the same co lumn, chromatography of immunoglobul in and B S A was per formed as s tandards to ensure that the ant ibody comp lexes were not being F I G U R E 15 SDS-Po l yac ry l am ide Ge l E lectrophores is of S O M A 10 and F ragments 12% Reso lv ing gel L a n e 1 H A P S O M A 10 2 Protein A-purif ied S O M A 10, peak A 3 Protein A-purif ied S O M A 10, peak B 4 P apa i n 5 Low molecu lar weight s tandards (94000 d, 67000 d, 43000 d, 30000 d, 20100 d, 14400 d) 6 Papa in-d iges ted S O M A 10, post-digest ion 7 Fab fragments 8 F c regions 9 Low molecu lar weight s tandards Lanes T A B L E 3 E lec t rophores is of S O M A 10 and Sample H A P S O M A 10 Protein A-pur i f ied H A P S O M A 10 P e a k A Protein A-pur i f ied H A P S O M A 10 P e a k B Papa in -d iges ted S O M A 10 Post - react ion F a b f ragments F c fraction ments on S D S - P o l y a c r y l a m i d e G e l Mo lecu la r Weight 69 kd , 57 kd , 53 kd , 31 kd , 26 kd 75 kd , 69 kd , 57 kd , 53 kd 57 kd, 26 kd 31 kd, 25 kd 25 kd 31 kd, 26 kd F I G U R E 16 C h r o m a t o g r a p h y of Incubat ion M ix tu res of An t i body and 1 ^ l - l a b e l l e d Somatostat in on Sephadex G-100 1 2 5 l - l a b e l l e d somatos ta t in was incubated with e i ther S O M A 10 (a), F a b f ragments (b) and the F c fraction (c) and chromatography of the mixture was per formed on S ephadex G-100 (open squares ) . Labe l l ed somatostat in was d i s p l a c e d f rom the an t i body or f r agmen t s with c o l d s oma t o s t a t i n at concentrat ions of 3.5 u.g/ml (filled d iamonds) and 7 |ig/ml (filled squares) . 7 7 0 1 0 20 30 40 V O L U M E (mis) b. 8000 -i V O L U M E (mis) 78 V O L U M E (mis) 7 9 retained by the co lumn. Immunoglobul in (150 kd) and B S A (67 kd) eluted at 10 and 12 ml, respect ively. S O M A 10 and the Fab comp lexes eluted at 9 and 15 ml , respect ive ly . Th i s ind i ca ted that the S O M A 10 c o m p l e x e s and F ab comp lexes were not retained by the co lumn. C. Immunocytochemica l Character izat ion of Fab Fragments F igure 17a - c shows the local ization of somatostat in with H A P S O M A 10, Fab fragments and the F c fraction on rat pancreas, respectively. These pictures were taken in consecut ive sect ions of the pancreas . The app le-green co lour represents speci f ic staining of D cel ls in the periphery of the islet. H A P S O M A 10 gave the brightest f luorescence. The F c fraction gave the se cond brightest f luorescence. Th is fraction conta ined F c fragments, a s wel l a s undigested and partially d igested S O M A 10. Fab fragments gave the least f luorescence. VI. Ac id Secret ion S O M A 10 was previously tested for biological activity on the bas is of its ability to neutral ize exogenous ly -admin i s te red somatostat in (Sea l et a l .1987). S O M A 10 dose-dependent ly reversed the somatostat in (2 u.g/kg/hr)-induced inhibition of liver extract-st imulated ac id secret ion. Concentrat ions of S O M A 10 greater than 100 u,g completely reversed the inhibition. F igure 18 shows the effect of F a b f ragments on somatostat in- induced inhibition of gastr ic ac id secret ion, st imulated by 5% liver extract meal in gastr ic f istula rats. Ac id secret ion from Fab fragment infused rats (6 u.Eq/10 min) was consistant ly higher than from control rats (4.4 p.Eq/10 min). The se exper iments we re pe r f o rmed in co l l abora t ion with Dr. A . S e a l . In p rev i ous contro l exper iments in which ac id secret ion was measured in response to liver extract a lone (9 u.Eq/10 min), ac id secret ion was much higher than when somatostat in was a lso infused. 8 0 F I G U R E 17 Compar i son of immunoreact iv ity of S O M A 10 Fab Fragment by Immunocytochemistry 250 x F i v e m i c r ome t r e se r i a l s e c t i o n s t h rough p a n c r e a t i c rat i s l e t s w e r e immunocytochemica l l y sta ined with H A P S O M A 10 (a), Fab f ragments (b) and the F c fraction (c) to test for immunoreact iv i ty of the ant ibody and f ragments . The app l e - g r een co l ou r r ep re sen t s spec i f i c s ta in ing of soma tos t a t i n in pancreat ic D cel ls . 8 1 83 c "E o cr LU 3 . -z. O h-LU DC O LU CO Q O < 14 12 -10 -8 -6 4 -2 -2 ug/kg/hr S S 5% liver extract V 100 u.g Fab CONTROL • — + S O M A 10 Fab i • i • i • i • i • i • i • I • I • I • I ' I • B1 B2 B 3 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 T I M E (min) F I G U R E 18 Effect of S O M A 10 F a b F ragments on Somatos ta t in - induced Inhibition of A c i d S e c r e t i o n n = 4 In exper imenta l an imals , 100 ixg of S O M A 10 F a b f ragments were injected as a bo lus into the jugular ve in (filled d i amonds ) . In contro l an ima ls (n = 4) , no ant ibody w a s injected (open squares) . B a s a l ac id secret ion w a s measu red for 30 min. A f ive percent l iver extract mea l w a s f lushed into the s t o m a c h to st imulate ac id secre t ion . A n infusion of 2.0 }ig/kg/hr of somatosta t in inhibited ac id secre t ion . 84 VII. Isolated Per fused S tomach Exper iments  Gastr in Re l ease i. Effect of S O M A 10 and Fab Fragments on Somatostat in- induced Inhibition  of Gastr in Re l ease T h e s e exper iments were per fo rmed in order to invest igate whether S O M A 10 and Fab fragments of S O M A 10 cou ld reverse inhibition of gastr in re lease by somatostat in . After an equi l ibrat ion per iod of 20 - 30 min, basa l s amp le s were co l lected for 10 min. Somatostat in was infused v ia a s idearm for 26 min at the following concentrat ions (nM): 60, 6 ,1 .5 , 0.6, and 0.15. At 20 min, 10 min into the somatostat in infusion, an infusion of S O M A 10 (20 ng/ml) or S O M A 10 Fab fragments (15 u.g/ml) was introduced. The se concentrat ions of S O M A 10 and of F a b f ragments were approx imate ly 130 n M and 300 nM, respect ive ly . S ince the intact ant ibody is divalent, the concent ra t ions were s imi lar with respect to the number of binding s ites. Contro l exper iments were performed in which no peptide or antibody were infused. Infusion of 60 nM somatostat in c aused an 85% dec rease in basa l gastrin re lease (fig 19). Infusion of 20 u.g/ml S O M A 10 had no effect on the inhibition. W h e n the somatos ta t in infus ion wa s terminated, gastr in re lease i n c reased again to a level not significantly different from basa l (time = 2 - 1 0 min). Infusion of 6 n M somatostat in produced a 5 2% dec rease in basa l gastrin re lease (fig 20). Infusion of 20 (xg/ml S O M A 10 caused a smal l , but insignificant inc rease in gastr in re lease. The onset of the reversa l of somatostat in- induced inhibition of gastr in re lease was de layed. W h e n the somatostat in infusion was terminated, gastr in re lease increased again to a level not signif icantly different from basa l (time = 2 - 1 0 min). Infusion of 1.5 n M somatostat in p roduced a 3 2 % dec r ea se in basa l gastr in re lease (fig 21). Infusion of 20 u.g/ml S O M A 10 inc reased gastr in to a level not s ignif icantly different from basa l . W h e n the S O M A 10 infusion was terminated, gastr in re lease dec reased back to p r e - SOMA 10 levels. When the somatostat in infusion wa s terminated, gastr in re lease i nc reased to levels not significantly different from basa l . 8 5 F I G U R E 1 9 - 2 3 Effect of S O M A 10 on Somatostat in- induced Inhibition of Gastr in Re l ease n = 6 For f igures 19 to 23, the fol lowing protocol was fol lowed. a. Ba sa l samp les were co l lected for 10 min. Somatostat in was infused for 26 min (filled d iamonds) . Ten minutes into this infus ion, an infus ion of 20 ug/ml S O M A 10 was super imposed for 6 min. Effluent was co l lected for a further 10 min, after the termination of the somatostat in infusion. In control exper iments (n = 6), buffer a lone was per fused for 46 min (open squares ) . Immunoreact ive gastrin (IRG) was measured by RIA as descr ibed in methods. b. The integrated gastr in re lease during per iod 4 to 10 min, 14 to 20 min, 22 to 28 min, 30 to 36 min, and 40 to 46 min is shown here. The aster isks represent s igni f icance (p <_0.05) compared to basa l levels (2 to 10 min). 8 6 F I G U R E 19 Effect of S O M A 10 on Somatos ta t i n - i nduced (60 nM) Inhibit ion of Gas t r in Re l e a s e 8 7 a. CT) C L CD DC 200 - i 150 -100 -50 -SOMATOSTATIN (6 nM) ^ 20u .g /m lSOMA10 W -n CONTROL -• + 6 n M S S 10 2 0 3 0 T I M E (min) 4 0 5 0 co o . (3 800 n 600 400 -200 -4 - 10 14 - 20 22 - 28 30 - 36 40 - 46 TIME PERIODS (min) 6 nM SS F I G U R E 20 Effect of S O M A 10 on Somatos ta t i n - i nduced (6 nM) Inhibit ion of Gas t r i n Re l e a s e n = 6 8 8 200 - i 150 -c 'E g 100 -o cr 50 H SOMATOSTATIN (1.5 nM) A A 20 ug/ml S O M A 10 CONTROL +1.5 nM S S —i 1 1 1 1 1 1 ' 1 1 0 2 0 3 0 4 0 5 0 T I M E (min) c 'E CO O l Q . o CC 1000 -i 800 -600 -400 -200 4 - 10 14 - 20 22 - 28 30 - 36 40 - 46 TIME PERIODS (min) 1.5 nM SS F I G U R E 21 Effect of S O M A 10 on Somatos ta t i n - i nduced (1.5 nM) Inhibition of Gast r in Re l e a s e n = 6 89 F I G U R E 22 Effect of S O M A 10 on Somatos ta t i n - i nduced (0.6 nM) Inhibition of Gas t r in Re l e a s e 9 0 c ' E CL u DC 200 n 150 -100 -50 -0 S O M A T O S T A T I N (0.15 nM) 20 ug /ml S O M A 10 10 2 0 3 0 T I M E (min) 4 0 —i 5 0 CONTROL + 0.15 nM S S 1200 -1000 -800 -E CO • O) 600 -Q . • o 400 -cc 200 -0 -0.15 nM SS 4 - 10 14 - 20 22 - 28 30 - 36 40 - 46 TIME P E R I O D S (min) F I G U R E 23 Effect of S O M A 10 on Somatos ta t i n - i nduced (0.15 nM) Inhibition of Gas t r in R e l e a s e n = 6 9 1 Infusion of 0.6 n M somatostat in p roduced an 18% dec rea se in gastr in re lease (fig 22). Infusion of 20 M.g/ml S O M A 10 c au s ed a reversa l of gastr in r e l ease to a leve l s imi lar to ba sa l gast r in re l ease . Te rm ina t i on of the somatos ta t in infus ion returned gastr in re l ease to a level not s igni f icant ly different from basa l . Infusion of 0.15 n M produced an 18% dec rease in gastr in re lease (fig 23) . Infusion of 20 u,g/ml S O M A 10 produced a smal l , but insignificant reversal of gastrin re lease to a level similar to basa l gastrin re lease. In s tud ies on the effect of Fab f ragments on the inhibit ion of gastr in re lease by 1.5 n M somatostat in, somatostat in p roduced a 3 7 % dec rease (fig 24) . Infusion of 15 ug/ml Fab fragments reversed the inhibition to a level similar to basa l gastr in re lease. After termination of the Fab fragment infusion, gastrin re lease returned to leve ls wh ich were not s ignif icant ly different from basa l gastr in re lease. A s when S O M A 10 was infused, the onset of the reversal of somatostat in- induced inhibition of gastrin re lease was de layed. ii. Effect of S O M A 10 and Fab Fragments on Basa l Gastr in Re lease The effect of a s ingle pas sage of the ant ibody on basa l gastr in re lease was s tud ied. After a 20 to 30 min equi l ibrat ion per iod, basa l s amp l e s were co l lected for 20 min. 100 u.g/ml and 20 |ig/ml final concentrat ions of S O M A 10 were infused v ia a s idearm for 6 min, and samp les were again co l lected for 20 min. In control exper iments, no antibody was infused. S O M A 10 at 20 jig/ml c au s ed no signif icant change in gastr in re lease c ompa red to control exper iments (fig 25). S O M A 10 at 100 u\g/ml c au s ed a s igni f icant d e c r ea se in gastr in re lease 2 min into the infus ion, c ompa r ed to control va lues at the s ame time, and compared to p r e - SOMA 10 gastr in re lease (fig 26). Basa l gastr in re lease was approximately 170 pg/min and S O M A 10 c au sed it to dec rease to 110 pg/ml. After termination of the S O M A 10 infusion, gastr in re lease returned to basa l levels within 2 min. There was a smal l , but insignificant, rebound in gastrin re lease fol lowing the termination of S O M A 10. 9 2 F I G U R E 24 Effect of S O M A 10 Fab Fragments on Somatostat in- induced (1.5 nM) Inhibition of Gastr in Re lease n = 6 a. Ba sa l samp les were col lected for 10 min. Somatostat in at a concentrat ion of 1.5 nM was infused for 26 min (filled d iamonds) . Ten minutes into this infusion, an infusion of 15 jig/ml S O M A 10 Fab fragments was supe r imposed for 6 min. Eff luent w a s co l l e c ted for a further 10 min, after the te rm ina t i on of the somatos ta t in infus ion. In control exper iments (n = 6), perfusate a l one was pe r f u sed for 46 min (open squa res ) . Immunoreac t i ve gast r in ( IRG) wa s measured by RIA as descr ibed in methods. b. The integrated gastrin re lease during per iod 4 to 10 min, 14 to 20 min, 22 to 28 min, 30 to 36 min, and 40 to 46 min is shown here. The aster isks represent s igni f icance (p < 0.05) compared to basa l levels (2 to 10 min). 93 c 'E C L O CC 2 0 0 n 15 ng/ml S O M A 10 Fab F R A G M E N T S 150 -100 -50 -SOMATOSTATIN (1.5 nM) —1 r— 0 1 0 —J— 2 0 —T— 3 0 TIME (min) — r — 40 -t3 CONTROL -• + 1 . 5 n M S S 50 b. CL O DC 1000 800 600 400 -200 -1.5 nM SS + Fab 4 - 10 14 - 20 22 - 28 30 - 36 40 - 46 TIME PER IODS (min) 94 F I G U R E 25 and 26 Effect of S O M A 10 on Basa l Gastr in Re lease n = 6 For f igures 25 and 26, the fol lowing protocol was fol lowed. B a s a l s amp l e s were co l l ec ted for 20 min before the infus ion of S O M A 10. S O M A 10 wa s in fused at 20 min for 6 min (fil led d i amonds ) . After the terminat ion of the infus ion, s amp l e s were co l lec ted for a further 20 min. In control exper iments (n = 6), perfusate a lone wa s per fused for 46 min (open squares) . Immunoreact ive gastrin (IRG) was measu red by RIA as desc r ibed in methods. 95 F I G U R E 25 Effect of S O M A 10 (20 ug/ml) on Basa l Gastr in Re lease n = 5 9 6 F I G U R E 26 Effect of S O M A 10 (100 ug/ml) on Basa l Gastr in Re lease n = 6 T h e as te r i s k s represent a s ign i f icant d i f ference (p<0.05) b e tween gast r in re lease (IRG) in control and S O M A 10 infused animals. 97 The exper iment was repeated us ing S O M A 10 Fab fragments instead of the whole antibody. The concentrat ions of Fab fragments u sed were 66 tag/ml and 15 u.g/ml. These concentrat ions are approximately equivalent to 100 fig/ml and 20 (ig/ml S O M A 10 with respect to the number of b inding sites. Contro l exper iments were performed in which no fragments were infused. In f igure 27, basa l gast r in re lease wa s approx imate ly 150 pg/min. Infusion of 15 u.g/ml Fab f ragments c au sed an immediate inc rease in gastr in re lease to approx imate ly 270 pg/min. Fo l lowing terminat ion of the infusion, gastr in re lease returned to basa l levels within 6 min. The inc rease in gastr in re lease was not significant. Infusion of 66 j ig/ml F ab f ragments (fig 28) c a u s e d an immed ia te i n c r ea se in gas t r in r e l ease that wa s s ign i f i cant ly di f ferent f rom contro l exper iments , and from basa l gastr in re lease at 20 min. Gas t r i n re lease rema ined e levated throughout the F ab fragment infus ion. W h e n the infusion was terminated, gastrin re lease slowly returned to basa l within 6 min. iii. Effect of Recirculat ion of S O M A 10 on Basa l Gastr in Re l ease Norma l l y in the i so la ted pe r fused s t omach prepara t ion , an in fused ant ibody wou ld have one pa s sage through the s tomach vascu la ture . To g ive the ant ibody more opportunity of neutra l iz ing endogenous somatos ta t in , a sys tem was set up such that perfusate containing antibody could be recirculated in the s t omach vascu la ture . Ba sa l s amp l e s were co l lec ted for 10 min, and gas t r i n c on cen t r a t i o n s we re m e a s u r e d . P e r f u s a t e con ta i n i ng a f ina l concentrat ion of 20 |ig/ml S O M A 10 was infused for 14 min. Th is was sufficient t ime to f lush all perfusate without ant ibodies from both the tubing and s tomach vasculature. For the last 4 of the 14 min, the venous effluent was co l lected and reoxygenated . Th is was used a s a reservoir for recirculat ion of the ant ibody. The perfusate conta in ing the ant ibody was rec i rcu lated for 20 min, and then non-c i r cu la ted per fusa te was pe r fused for a further 10 min. In contro l exper iments, perfusate without the antibody was recirculated. 98 F I G U R E 27 and 28 Effect of S O M A 10 Fab Fragments on Basa l Gastr in Re l ease For f igures 27 and 28, the fol lowing protocol was fol lowed. a. Ba sa l s amp le s were co l lected for 20 min before the infusion of S O M A 10 Fab f ragments . F ab f ragments were in fused at 20 min for 6 min (fi l led d iamonds) . After the terminat ion of the infusion, s amp l e s were co l lected for a further 20 min. In control exper iments (n = 6), perfusate a lone was infused for 46 min (open squares) . Immunoreact ive gastrin (IRG) wa s measu red by RIA as desc r ibed in methods. b. The integrated gastrin re lease during period 16 to 20 min, 22 to 26 min, 28 to 32 min, and 34 to 38 min is shown here. The aster isks represent s igni f icance (p < 0.05) compared to controls. 99 c 'E a. (D DC 300 H 200 H 100 H CONTROL + S O M A 10 Fab 15 ug/ml S O M A 10 Fab F R A G M E N T S 1 0 i 3 0 2 0 T I M E (min) 40 i 5 0 b. 1000 -i c "E CD CL U DC 15 u.g/ml Fab fragments CONTROL + 15 u.g/ml Fab 16 - 20 22 - 26 28 - 32 34 - 38 T I M E P E R I O D S (min) F I G U R E 27 Effect of S O M A 10 Fab Fragments (15 ug/ml) on Basa l Gastr in Re lease n = 6 100 CT) C L CD DC 300 -i 250 -200 -150 -100 -50 -0--* * 0 66 p.g/ml S O M A 10 Fab FRAGMENTS T r T r-10 2 0 3 0 T I M E (min) 4 0 1 5 0 CONTROL + S O M A 10 Fab CD C L o C C 66p.g/ml Fab fragments 16 - 20 22 - 26 28 - 32 T I M E P E R I O D S (min) 34 - 38 CONTROL + 66 u.g/ml Fab F I G U R E 28 Effect of S O M A 10 Fab Fragments (66 ng/ml) on Basa l Gastr in Re lease n = 6 T h e as te r i sks represent a s igni f icant d i f ference (p<0.05) b e tween gas t r in re lease (IRG) in control and Fab fragment infused animals . 1 0 1 Per fus ion of the s tomach with 20 u.g/ml S O M A 10 c au s ed no significant c hange in gastr in re lease c ompa red to contro ls . After rec i rculat ion of the per fusate for 20 min, cumu la ted gastr in re lease f rom S O M A 10-per fused an ima l s was signif icant ly h igher than from contro l an ima l s (fig 29). Gas t r in re lease in both an imals dec reased to a level that was not signif icantly different from basa l . VIII. Isolated Per fused Panc reas Exper iments  Insulin Re l ease i. Effect of S O M A 10 and Fab Fragments on Insulin Responses to a  G l u co se Gradient Per fusa te conta in ing 4.4 m M g lucose wa s admin is tered for 5 min. A gradient of g lucose ranging from 4.4 to 16.5 m M was then infused over the next 30 to 35 min. In exper iments using the whole antibody, 16.5 m M g lucose was perfused after termination of the g lucose gradient. The effect of the antibody on basa l insulin re lease was tested by s idearm infusion to a final concentrat ion of either 5 u,g/ml or 10 u,g/ml S O M A 10, or 5 u.g/ml S O M A 10 Fab fragments at time 5 min and cont inued throughout the g lucose gradient st imulat ion. In control exper iments, the pancreas was st imulated with a glucose-gradient, but was not subjected to the antibody. The gradient was generated by a gradient maker in which the chamber s were filled with perfusate at concentrat ions of 4.4 m M and 16.5 m M g lucose. S O M A 10 and Fab fragments at the concentrat ions used (fig 30 and 31) had no effect on basa l insulin re lease. Insulin re lease increased in response to the g lucose gradient, reaching between 800 u.U/min and 1000 u.U7min at 16.5 m M g lucose . Neither S O M A 10 nor the F ab fragments had a signif icant effect on g lucose-st imulated insulin re lease in compar i son to controls. ii. Effect of Fab Fragments on G lucose (16.5 mM)-St imulated Insulin Re lease Per fusate containing 4.4 m M g lucose was administered for 5 min. Insulin r e l ease w a s s t imu la ted with 16.5 m M g l u co se for the rema inde r of the 1 0 2 F I G U R E 29 ' Effect of Recirculat ion of S O M A 10 on Gastr in Re lease n = 7 Basa l immunoreact ive gastrin (IRG) was measured for 10 min. In exper imenta l an imals , S O M A 10 at a final concentrat ion of 20 jig/ml was infused for 14 min (filled d iamonds) . For the last 4 of the 14 min, the venous effluent was col lected, oxygena ted , and u s ed as a reservo i r for rec i rcu lat ion. Rec i r cu la t ion of the ant ibody cont inued for 20 min, upon which the an imals were per fused with non-recyc led perfusate for a further 20 min. In control an imals (n = 9), the perfusate did not conta in antibody (open squares) . T h e as te r i s k s represent a s igni f icant d i f ference (p<0.05) b e tween gas t r in re lease (IRG) from exper imental an imals and control an imals . 1 0 3 F I G U R E 30 Effect of S O M A 10 (5 and 10 jig/ml) on G l u c o s e Grad ient -S t imu la ted Insulin Re l e a s e Per fusate was co l lected for 10 min, after which a gradient of g lucose from 4.4 m M to 16.5 m M was admin istered (total vo lume = 120 ml). After the gradient, g lucose was mainta ined at 16.5 m M for 10 min. In control an imals (n = 6, open squares) , no ant ibody was infused. Immunoreact ive insulin (IRI) was measured by RIA as descr ibed in methods. In f igure a , exper imenta l an ima ls (n = 6, fi l led tr iangles) were in fused with 5 jig/ml S O M A 10 from the 5th minute to the end of the gradient. In f igure b, exper imenta l an ima ls (n = 6, fil led tr iangles) were in fused with 10 u,g/ml S O M A 10. 104 a CONTROL * — + 10 u.g/ml S O M A 10 10 ng/ml S O M A 10 b. 2000 - i T I M E (min) 1 0 5 c 'E 1000 800 -600 -400 200 -5 u.g/mL S O M A 10 Fab fragments - 16.5 2 0 T I M E (min) r 3 0 T 4 0 CONTROL + 5 ug/mL Fab - 4.4 0 F I G U R E 31 Effect of S O M A 10 Fab F ragments (5 jig/ml) on G l u c o s e Grad ient -St imu la ted Insulin Re l ease n = 6 Ba sa l immunoreac t i ve insul in (IRI) was co l lec ted for 10 min, upon wh ich the panc reas wa s st imulated with a gradient of g lucose from 4.4 m M to 16.5 mM (total vo lume = 120 ml). In exper imental an imals (filled tr iangles) Fab fragments were infused from the 5th minute to the end of the gradient. In control an imals (n= 6, open squares) no fragments were infused. 106 perfusion. In exper imental animals, 10 u.g/ml Fab fragments were infused from the 16th to the 25th min. Insulin was co l lected for a further 10 min. In control exper iments, no fragments were infused (fig 32). T h e f irst insu l in peak in r e s p o n s e to the i n c r e a s e in g l u c o s e concen t ra t i on in cont ro l p a n c r e a s e s d id not c o r r e s pond to that of the exper imenta l an ima ls . However , the shape and height of this peak is quite var iab le . The f ragments c au s ed a dec rease in insul in re lease . S ign i f i cance from control va lues was ach ieved only during the 20th to the 22nd min. After the termination of the F ab infusion, insulin re lease was not significantly different from controls. iii. Effect of S O M A 10 and Fab f ragments on G l u c o s e (8.8 mM>St imu la ted  Insulin Re l ease A s imi lar protocol as above (section ii) was fo l lowed, with the except ion that the pancreas was st imulated with 8.8 m M g lucose . A lower concentrat ion of g lucose wa s used s ince the st imulat ion wa s thought to be too strong. A g l u co se concent ra t ion of 8.8 m M wa s c h o s e n ba sed on p rev ious s tud ies ind icat ing that in the p r e sence of a weak insul in s t imulus , the effect of somatos ta t in on insul in re lease w a s greatest . Conve r se l y , the effect of somatosta t in wa s weake r when insul in re lease wa s strongly s t imulated. In exper imenta l an imals , either 30 u.g/ml Fab f ragments or 45 u.g/ml S O M A 10 were infused. The concentrat ion of Fab f ragments was chosen based on the results of the infusion of 10 [ig/ml Fab fragments on 16.5 m M glucose-st imulated insu l in r e l ea se . T h e concen t ra t i on of S O M A 10 w a s c h o s e n a s it is approximately equivalent to 30 |a.g/ml Fab fragments with respect to the number of binding sites. St imulat ion of the panc reas with 8.8 m M g lucose c a u s e d a first phase increase in insul in re lease reaching 320 |iU/min at 10 min (fig 33). Infusion of the F ab fragment (30 ug/ml) had no significant effect, except at the 22 min time per iod where insul in re lease dec reased . Immediately after the terminat ion of the infusion of the Fab fragments, insulin re lease significantly inc reased to 222 l iU/min, and then slowly returned to control va lues within 3 min. 107 1200 4.4 mM glucose 16.5 mM glucose CONTROL + 10 ug/mL S O M A 10 Fab fragments T I M E (min) F I G U R E 32 Effect of S O M A 10 Fab F ragments on G l u co se (16.5 mM)-St imu la ted Insulin Re l e a s e n = 6 Basa l immunoreact ive insulin (IRI) was measured for 5 min. The panc reas was then st imulated with 16.5 m M g lucose for the remainder of the exper imenta l per iod. In exper imenta l an ima ls (filled d i amonds) , 10 u.g/ml S O M A 10 Fab f ragments were infused during the 16th to 25th minute. In control an imals (n = 6, open squa res ) , no f ragments were in fused. The as te r i s k s represent a signif icant difference (p<0.05) between the control and exper imenta l va lues. 108 E 4.4 mM glucose 8.8 mM glucose 30 u-g/ml Fab fragments CONTROL + Fab fragments T I M E (min) F I G U R E 33 Effect o f - S O M A 10 Fab F ragments on G l u c o s e (8.8 mM)-S t imu la ted Insulin Re l e a s e n = 7 Basa l immunoreact ive insulin (IRI) was measured for 5 min. The pancreas was then s t imulated with 8.8 m M g lucose for the rema inder of the exper imenta l per iod . In exper imenta l an ima l s (filled d iamonds) , 30 u.g/ml S O M A 10 Fab f ragments were infused during the 16th to 23th minute. In control an imals (n = 9, open squa res ) , no f ragments were in fused. The as te r i sks represent a signif icant difference (p < 0.05) between control and exper imenta l va lues . 109 In the exper iments with the higher concentrat ion of S O M A 10 (45 |ig/ml), st imulat ion of the panc reas with 8.8 m M g lucose c au sed an increase in insulin re lease between 190 and 270 u.U/min at 10 min (fig 34). Infusion of S O M A 10 c a u s e d a s ignif icant d e c r ea se in insul in re lease 3 min into the infus ion, in compar i son to controls. The dec rease remained significant up to 1 min after the terminat ion of the S O M A 10 infusion. Insulin re lease then returned to va lues similar to controls. iv. Effect of Pre-sura ica l Immunoneutral ization of Somatostat in with S O M A 10 Exper imenta l an ima ls were injected with 0.5 mg of S O M A 10 into the v ena c a v a just before the surgical procedure. Th is a l lowed for a 30 to 40 min per iod in wh ich the ant ibody cou ld neutral ize somatostat in . The preparat ion was equi l ibrated for 10 min with perfusate (4.4 mM) not containing antibody. In e xpe r imen ta l an ima l s , 20 j ig/ml S O M A 10 w a s i n fused for the ent i re expe r imen ta l per iod . No ant ibody w a s i n fused in cont ro l expe r imen t s . Per fusate was co l lected for 5 min, and the pancreas was then st imulated with 8.8 m M g lucose for the remainder of the exper imental period. The ant ibody c au s ed a s ignif icant de c r ea se in the first insul in peak; cont ro l s r eached 276 i iU/min , whi le in S O M A 10 in fused an ima l s , insul in re lease only reached 126 u,U/min (fig 35). In the remainder of the exper imental period, control and exper imental insulin re lease was not significantly different. IX. Immunocytochemica l Identification of Fab Fragments and S O M A 10 in  Pe r fused Organs In vitro studies in which antibody and Fab f ragments were incubated on co l l agen-coa ted p lates (fig 36) showed that F ab f ragments, and to a lesser extent S O M A 10 bound non-speci f ica l ly to co l lagen. Contro ls , a monoc lona l ant ibody to V IP and P B S , did not cause as much non-speci f ic f luorescence as S O M A 10. During process ing of the perfused t issues, in which S O M A 10 or Fab f ragments were infused, the antibody would have been washed out if it had not bound to the t issue, i.e. to somatostat in in or on cel ls, or to co l lagen. 1 10 4.4 mM glucose 8.8 mM glucose CONTROL + 45 ug/ml S O M A 10 T I M E (min) F I G U R E 34 Effect of S O M A 10 on G lucose (8.8 mM)-St imulated Insulin Re l ease n = 6 Basa l immunoreact ive insulin (IRI) was measured for 5 min. The pancreas was then s t imulated with 8.8 m M g lucose for the remainder of the exper imenta l per iod. In exper imenta l an ima ls (filled d i amonds) , 45 [ig/ml S O M A 10 was infused during the 16th to 23th minute. In control an imals (n = 9, open squares) , no ant ibody was in fused. The aster isks represent a signif icant di f ference (p < 0.05) between control and exper imenta l va lues. 1 1 1 4.4 mM glucose 8.8 mM glucose 0 T I M E (min) F I G U R E 35 Effect of P re -Su rg i ca l P a s s i v e Immunizat ion with S O M A 10 on G l u co se (8.8 mM)-St imulated Insulin Re l ease n = 6 In exper imenta l an ima ls , the v e n a c a v a wa s e xpo s ed and a 0.5 mg bo lus injection of S O M A 10 in P B S was administered. The an ima ls were equi l ibrated for 10 min after the preparat ion for the iso lated-vascu lar ly per fused panc reas . Basa l immunoreact ive insulin (IRl) was measured for 5 min. The panc reas was then s t imu la ted with 8.8 m M g lucose for the rema inder of the exper imenta l per iod. In exper imenta l an ima l s (filled d i amonds) , 20 |ig/ml S O M A 10 was infused dur ing the entire exper imenta l per iod. In control an ima ls (n = 9, open squa res ) , no ant ibody was in fused. The as te r i sks represent a s ign i f icant difference (p < 0.05) between the control and exper imental va lues. 1 1 2 F I G U R E 36 Nonspec i f i c B inding of S O M A 10 and Fab Fragments to Co l l agen P B S (a), a nd 1.0 u.g/ml of ant i -VIP (b), S O M A 10 (c), a nd S O M A 10 Fab f ragments (d) were incubated with co l lagen bound to Petri d i shes to test for nonspec i f i c binding of ant ibodies to co l lagen. The p lates were then incubated with b io t in - spacer -AP goat ant i - immunoglob in , fo l lowed by av id in -FITC. The plates were s c reened using epi - f luorescence and photographed. 1 1 3 1 1 4 115 Immunocytochemica l staining of the F ab f ragments and S O M A 10 was per formed on the s tomach and panc reas of per fused o rgans . Sec t i ons of s tomach perfused with 100 u.g/ml S O M A 10 or with recirculated antibody did not exhibit any f luorescence. However, sect ions of s tomach perfused with 66 u.g/ml F ab . f ragments demons t ra ted the infi ltration of the F ab f ragment into the interstit ium. App le -green f luorescence represents the spec i f ic staining of Fab fragments. In figure 37 (right), f luorescence is seen over the co l lagen matrix in the s ubmuco sa of Fab fragment infused (66 ng/ml) s tomachs . On the left s ide, in which no ant ibody was infused, only intrinsic f luorescence from chief ce l l s is s e en (bright ye l low). In the panc reas , both S O M A 10 and espec ia l l y F ab f ragments (fig 38) were found in the interstitium, binding to col lagen-conta in ing structures. Note that a halo surrounds the duct in the Fab fragment infused (30 u.g/ml) pancreas , but not in the control pancreas . Intrinsic f luorescence (bright yel low) from z ymogen granu les is found in ac inar t i ssue in both control and fragment infused pancreases . 1 16 F I G U R E 37 Immunocytochemica l staining of Fab Fragments in the Per fused S tomach 125 x F ive micrometre sect ions through the corpus of the rat s tomach. On the left is a control sect ion showing only the intrinsic f luorescence from the chief cel ls. On the right is a sect ion after perfusion with 66 u.g/ml S O M A 10 Fab fragments for 6 minutes. Note the p resence of spec i f i c (apple-green) f l uo rescence over the co l lagen matrix in the submucosa . 1 18 F I G U R E 38 Immunocytochemica l Staining of Fab Fragments in the Per fused Panc reas 250 x a) A five micrometre sect ion through a control pancreas showing exocr ine and endocr ine e lements . Note that the ductular t issue to the left of the islet is not sur rounded by a halo of f luorescence. b) A s imi lar sect ion from a panc reas per fused with S O M A 10 Fab fragments (30 u.g/ml). Note the halo of posit ive immunof luo rescence a round the duct in the centre of the figure. N.B. The bright ye l low f luo rescence on the ac inar t i ssue is intr insic to the z ymogen granu les . 119 120 D I S C U S S I O N S i n ce high purity ant ibod ies are important in an ima l s tud ies , asc i tes contain ing S O M A 10 was purified, a c ces sed , and character ized. Fab fragments of S O M A 10 were produced and character ized. S O M A 10 and its Fab fragment were then used in in vitro studies of the effect of somatostat in on gastr in and insulin re lease in the s tomach and pancreas , respectively. I. Purif ication and Character izat ion of S O M A 10 and Fab Fragments The purif icat ion of S O M A 10 by ammon i um su lphate prec ip i tat ion, in conjunct ion with hydroxylapatite chromatography resulted in an ant ibody with purity of 93%. Th i s indicates that hydroxylapat ite chromatography is a good method of purifying mouse monoc lona l ant ibodies. The yie ld of 2 to 3 mg of pure S O M A 10 per millilitre of asc i tes fluid is low compa red to the y ie ld from other monoc l ona l ant ibody p roduc ing ce l l s . Th i s is a result of the low production of antibody by S O M A 10 cel ls, and not of the purification procedure. The purity of S O M A 10 at var ious s tages of purif ication was determined by affinity and ge l filtration H P L C . In the elution profile of S O M A 10 on the Ultraff inity-EP co lumn, the first peak, which d id not bind to the co lumn, had no immunoreact iv i ty to somatostat in, as determined by EL ISA . On ly the second peak, deso rbed with water in which the pH was dec r ea sed to 2 by HCI, was immunoreact ive to somatostat in. The increase in purity of S O M A 10 after each purif ication step is determined by the dec rease in s ize of the first peak. On the gel filtration co lumn, the S O M A 10 peak eluted at 14.5 min. A s the purification of S O M A 10 p roceeded , this peak be came larger and more homogeneous . A l though there were smal l d i f ferences in the results of the percentage purity of asc i tes and S A S S O M A 10 a s s e s s ed by affinity and gel filtration H P L C , both methods were in agreement that H A P S O M A 10 was of very high purity. W h e n S O M A 10 samp les , at var ious s tages of purif ication, were tested for immunoreact iv i ty to somatostat in on EL ISA , asc i tes , ammon ium sulphate prec ip i ta ted , a nd hydroxy lapat i te pur i f ied S O M A 10 gave half max imum readings of 25 ug/ml, 10 u.g/ml and 10 ug/ml, respectively. The dec rease in the concentrat ion giving half max imum reading for ammon ium sulphate precipitated 121 S O M A 10 and H A P S O M A in compar i son to that for asc i t es ref lects the increase in purity of S O M A 10. By immunod i f fus i on ge l s , it w a s de te rm ined that the four S O M A an t i bod ies were of c l a s s l g G 1 . Unde r the u sua l cond i t i ons ut i l i zed for purif ication of ant ibodies, this c l a s s of antibody does not bind wel l to protein A (God i n g , 1978) . To a s s e s s the b ind ing of S O M A 10 to prote in A, hydroxylapat i te S O M A 10 was appl ied to a protein A co lumn. A n average of 5 2 % bound to the co lumn. The fraction that did not bind to the protein A, peak A, was reappl ied to a fresh protein A co lumn, to confirm that the antibody did not bind to protein A, and not that the co lumn was over loaded. Aga in this fraction did not bind to protein A, suggest ing that this fraction did not conta in S O M A 10, or that the antibody did not have the ability to bind to protein A . S O M A 10, protein A purified peak A and the fraction binding to protein A (peak B) were compared by RIA, E L I S A and on SDS-po l yac ryam ide ge ls . By RIA, the activity of S O M A 10 and peak B were similar. P eak A gave 39% of the max imum binding of S O M A 10 and peak B. The resul ts f rom the E L I S A ind ica ted that peak A wa s 2.5 t imes and 5 t imes l e s s immunoreac t i ve to somatostat in than S O M A 10 and peak B, respectively. That peak A did not bind to protein A, does not imply that peak A did not contain S O M A 10, s ince it was slightly immunoreact ive to somatostat in. Futhermore, hydroxylapat i te purif ied S O M A 10 was determined by H P L C on the affinity co lumn to be 9 3 % pure. The 7% n o n - S O M A 10 cou ld not account for the difference in the activity between peak A and peak B. Monoc lona l ant ibodies, such as S O M A 10, shou ld be homogeneous . There is speculat ion that there is differential g lycosylat ion of the ant ibod ies caus ing some to bind to protein A, and s ome to lose this ability. A l te rna t i ve ly , p eak A an t ibod ies may have been d a m a g e d dur ing the hydroxylapatite purification, and lost the ability to bind to protein A. E lec t rophores is of hydroxylapatite S O M A 10 on an SDS-po lyac ry lam ide gel revea led strong bands corresponding to molecular weights of 57 and 26 kd, which are the heavy and light cha ins of S O M A 10, respectively. Weake r bands appea red at 69, 53 and 31 kd. The 69 kd band is thought to conta in a lbumin, the 53 kd band cou ld be another contaminat ing protein, and the 31 kd band cou l d be part of the heavy cha i n . The s t ronges t b and appea r i ng on e lec t rophores i s of protein A puri f ied S O M A 10 peak A co r r e sponded to a 122 molecu lar weight of 69 kd, with weaker bands at 75, 57 and 53 kd. The 75 kd band is thought to be a d imer or trimer of parts of the heavy cha in and the light cha in . E lectrophores is of peak B revealed only two bands at 57 and 26 kd. Incubation of protein A purif ied S O M A 10 peak B with papa in for 4 h ach ieved almost complete digest ion of the antibody. Affinity chromatography on protein A was chosen as the method of purifying the F ab fragment from the d igest ion mixture, because Fab fragments purif ied by other mean , such as by ion e x c h a n g e ch roma tog raphy , may sti l l con ta in und i ge s t ed or part ial ly d iges ted S O M A 10, that have all or s ome of the F c portion of the ant ibody attached. T h e F a b f r agmen t s we re ident i f ied on a reduc ing ge l by S D S -po lyac ry l am ide ge l e lec t rophores i s . The papa in -d iges ted S O M A 10 post-digest ion sample gave bands corresponding to molecular weights of 31 kd and more prominent ly at 25 kd. F a b f ragments e luted at a mo lecu la r weight corresponding to 25 kd. The fraction of the digest ion mixture absorbed onto the protein A co lumn, the F c fraction, may have conta ined undigested and partially d iges ted S O M A 10 and F c fragments. E lec t rophores i s of this samp le gave a strong band eluting at a molecu lar weight of 31 kd, and a weaker band at 26 kd. They represent parts of the heavy cha in in the F c fragment and the light cha in . The lack of a 57 kd heavy chain g ives support that S O M A 10 was almost complete ly d igested by papain, although treatment of this samp le with S D S and dithiothreitol may have damaged weakened bonds, so that intact heavy cha ins were broken down. To a s s e s s the abi l i ty of the F ab f ragments to b ind somatos ta t in , ch romatography of an t i body - 1 2 5| - somatos ta t i n comp l exes was per formed on Sephadex G-100 . The comp lexes were prepared by incubating S O M A 10, Fab fragments and the F c fraction with 1 2 5 l - s o m a t o s t a t i n overnight. S O M A 1 0 - 1 2 5 l -somatos ta t i n c o m p l e x e s e lu ted first, fo l l owed by F a b - 1 2 5 l - s o m a t o s t a t i n comp lexes . F c fract ion- 1 ^ - s o m a t o s t a t i n comp lexes eluted over a range, s ince this fract ion may have conta ined intact and partially d iges ted ant ibodies. F c fragments, wh ich have no somatostat in binding ability would not be recognized by moni tor ing radioact iv i ty. Ch roma tog raphy of 1 2 5 | - s o m a t o s t a t i n a l one showed that it e luted last. When somatostat in was incubated with S O M A 10 and 1 2 5 l - s o m a t o s t a t i n , and the chromatography of the comp lexes performed, 123 the rad ioact ive ly- labe l led somatostat in was d i sp l a ced from the ant ibody, as indicated by the dec rease in the complex peak, and the increase in the 125| . somatostat in peak. 1 2 5 l - s o m a t o s t a t i n was completely d i sp laced from S O M A 10 by 7 tig/ml somatostat in . With the F ab comp lexes , 3.5 u.g/ml of un labe l led somatostat in was required to d isp lace the 1 2 5 | - s o m a t o s t a t i n . The immunoreact iv i ty of the F ab f ragments w a s a l so a s s e s s e d by immunocytochemist ry . S O M A 10, the F c fraction and then the F ab fragments were found to cause f luorescence in decreas ing order of intensity. It is tempting to suggest that the activity of the F ab fragments is less than that of the intact ant ibody. However the second antibody, goat ant i -mouse immunoglobul in, may not b ind equal ly to the Fab fragment and the intact antibody. The f luorescence caused by the F c fractions suggests that this fraction still conta ined somatostat in binding ability. II. Effect of S O M A 10 and Fab Fragments on Biological Activit ies The reversa l of the somatos ta t in - induced inhibit ion of l iver extract-st imulated gastr ic ac id secret ion by the Fab fragments was the bas i s by which the biological activity was a s s e s sed . Ove r a 90 min per iod, the average control ac id secret ion was 40 + 4 u.Eq. Fab fragment injected an imals secreted 55 + 8 ( i Eq , sugges t i ng that the F a b f ragments cou l d neut ra l i ze e xogenous l y -admin is tered somatostat in. Ove r a 60 min period, the ac id secret ion from Fab f ragment injected an ima l s (100 u.g) wa s 37 ixEq, whi le that from S O M A 10 injected an ima ls (100 u,g) was 46 u.Eq (Sea l et al., 1987). Aga in this suggests that the activity of the fragment is less than the intact antibody. A. Pass i ve Immunization Studies in the Isolated Per fused S tomach The b io logica l act ivit ies of S O M A 10 and the F ab fragment were a l so a s s e s s e d by the i r abi l i ty to reve rse the inhib i t ion of gast r in r e l ease by somatostat in . Somatostat in c au s ed a dec rease in gastr in re lease at the four highest concentrat ions used , 60 nM , 6 nM, 1.5 n M and 0.6 nM (fig 19-22), but fai led to signif icantly inhibit gastrin re lease when 0.15 nM somatostat in (fig 23) was in fused. A n infusion of 20 |xg/ml of S O M A 10 reversed the inhibition of gastr in re lease, except at 60 n M somatostat in. Gastr in re lease inhibited by 1.5 nM somatostat in was completely b locked by 20 u.g/ml of S O M A 10. 1 2 4 Va l e et a l . (1978) determined that the binding site to the somatostat in receptor was located in the amino ac id res idues 6 to 11. Modif icat ion of these pos i t ions p roduced a drast ic dec rease in b io logica l potency. Subst i tut ion of res idues 7, 8 and 9 c a u s ed a comple te loss of activity. The binding site of S O M A 10 was determined to be between res idues 5 and 12. The binding site of the other S O M A ant ibodies was found to be between pos i t ions 4 and 12. Th i s impl ies that the S O M A antibodies bind to the active site of the somatostat in molecule, and can therefore potentially neutralize the act ion of somatostat in. The somatostat in receptor on iso lated rat panc reas membrane s was shown to have a K D of 0.1 nM , with a densi ty of 1800 s i tes per ce l l (Rey l -Desmar s and Lewin, 1982). Two somatostat in receptors have been found to be present on iso lated rat gastr ic cel ls. The high and low affinity receptors had a K D of 0.08 nM with a density of 560 sites/cell, and 4.5 nM with a density of 4600 s i tes/ce l l , respect ive ly (Rey l et a l . , 1979) . T h e s e resu l ts sugges t that somatos ta t in wou ld preferential ly b ind to pancreat i c somatosta t in receptors rather than to S O M A 10, but that somatos ta t in wou ld b ind S O M A 10 preferential ly to gastr ic somatostat in receptors because of the high densi ty of low affinity somatostat in receptors. The max imum binding capac i ty of S O M A 10, determined by Sca t cha rd analys is , wa s found to be 8.3 u.g/mg. As sum ing the molecu lar weight of S O M A 10 to be 150 kd, the binding capaci ty in molar terms is 760 mmoles/mole of S O M A 10, i.e. each ant ibody molecu le b inds one somatostat in molecu le . The max imum binding capac i t ies of S O M A 3, 8 and 20 were 2.58, 1.78, and 1.98 ng/ml, ~1000 less than that of S O M A 10. The d issoc iat ion constants of these ant ibodies were 3.2, 1.7 and 3.1 pM, respectively, and that of S O M A 10 was 2.2 nM. A s sum ing that the molecular weight of S O M A 10 is 150 kd, 20 u.g/ml is equivalent to 130 nM; 130 nM S O M A 10 neutral ized 1.5 n M somatostat in in the iso lated per fused s tomach . The reasons for the requirement of such a high c o n c e n t r a t i o n of S O M A 10 to neu t r a l i z e e x o g e n o u s l y - a d m i n i s t e r e d somatostat in in the isolated, perfused s tomach, in compar i son to the one to one binding of somatostat in and S O M A 10 as determined by Sca t cha rd ana lys i s cou ld be 1) the greater affinity of somatostat in to the high affinity somatostat in receptor than to S O M A 10; 2) the neutra l izat ion of both endogenous and 1 2 5 exogenous somatostat in in this preparat ion and; 3) the l imited a c c e s s of the antibody to somatostat in. F ab f ragments at a concentrat ion of 15 u.g/ml were ab le to complete ly reverse the inhibition of gastr in re lease by 1.5 n M somatostat in in the isolated per fused s tomach (fig 24). In compar i son to the concentrat ion of S O M A 10 requi red for comple te b lockage of inhibit ion, the F ab f ragments have slightly less activity. Whether these fragments have a lower affinity to somatostat in or a lower binding capaci ty than the intact antibody was not determined. S O M A 10 and its F ab f ragments were then u s ed in an attempt to neutral ize endogenous somatostat in. A n infusion of 20 u.g/ml S O M A 10 had no effect on basa l gastr in re lease (fig 25). W h e n the concentrat ion of S O M A 10 was inc reased to 100 u.g/ml, a signif icant dec rease in gastr in re lease occurred (fig 26). W h e n the S O M A 10 infusion was terminated, there wa s a smal l but insignificant rebound in gastr in re lease. These results are contrary to what was expec ted , which wa s an increase in gastr in re lease. There are a number of poss ib le explanat ions for these results. The dec rease in gastr in re lease cou ld be due to S O M A 10 neutral iz ing somatostat in, which in turn inhibited an inhibitor of gastr in. Th is disinhibit ion of a p rev i ous l y u n r e c o g n i z e d inh ib i tory m e c h a n i s m has b e e n p rev ious l y sugges ted . Sea l et a l . (1987) found the ac id secret ion dec rea sed in response to both 5% liver extract and sa l ine, when gastr ic f istula rats were pass ive ly immunized with 200 u.g S O M A 10. A s e cond explanat ion for the dec reased gastrin re lease cou ld be that the ant ibody is binding to somatostat in, and chang ing its conformat ion, such that it enhances its effects on the receptor. That is, S O M A 10 may increase the affinity or inc rease the binding time of somatostat in to its receptor. Severa l ant ibodies have been ob s e r v ed to enhance the effect of their an t igens . Monoc l ona l ant ibodies to thyrotropin (Holder et al . , 1987) and to growth hormone (Aston et a l . , 1986) have been shown to potentiate the product ion of thyroxine, and to potent ia te the s oma t ogen i c and l ac togen i c act iv i ty of g rowth ho rmone , respect ive ly . The enhancemen t of activity of somatostat in by S O M A 10 is un l ike ly s i n ce in fus ion of S O M A 10 r eve r sed the soma tos t a t i n - i nduced inhibition of gastr in re lease. 126 The de c r ea se in basa l gastr in re lease obse rved when S O M A 10 was in fused cou ld be a nonspec i f i c effect of the ant ibody, s i nce su ch a large concentrat ion was used . To test this possibil ity, mouse monoc lona l ant ibodies not d irected against a gastr ic peptide, or purified ascit ic fluid from mice injected with mye loma cel ls cou ld be used as control ant ibodies. A final poss ib le explanat ion for the observed dec rease in gastr in re lease is that S O M A 10 is binding to the somatostat in receptor, and mimick ing the effect of somatostat in. It shou ld be noted that S O M A 10 b inds to amino ac id res idues 5 to 12. The somatostat in receptor binds to amino ac id res idues 6 to 11. In cont ras t to the p resen t resu l ts , Sa f four i et a l . (1982) , u s i ng somatostat in ant iserum with a binding capacity of 19.2 ng/ml, found that infusion of the ant iserum at both a two fold and a tenfold dilution, c au sed an immediate and signif icant inc rease in basa l gastr in re lease in the iso lated, per fused rat s tomach. Short et al. (1985) infusing somatostat in ant iserum (binding capaci ty = 2.86 u,g/ml) at a 2000 fold dilution inc reased gastr in re lease in the isolated, per fused rat s tomach , but only after 45 min of infusion. C h i b a et a l . (1981), us ing a s imi lar preparat ion and an ant i serum with a binding capac i ty of 8.7 |ig/ml, found no increase in gastrin re lease with infusion of the ant iserum at a twofold and hundredfold dilution. In contrast to the results with the intact ant ibody, F a b f ragments at concentrat ions of 15 |ig/ml and 66 |ig/ml (fig 27 and 28), cor respond ing to 20 u.g/ml and 100 ixg/ml S O M A 10, respect ively, c au sed an immediate increase in gastr in re lease to 270 pg/min. Gastr in re lease remained greater than control for at least 5 min fol lowing the termination of the fragment infusion, indicating that the F a b f ragments were not immediate ly removed f rom the s tomach . The i n c r e a s e in gas t r in r e l ea se s u g g e s t s that S O M A 10 c ou l d neut ra l i ze endogenous somatostat in, and that somatostat in inhibits basa l gastr in re lease. Immunocy tochemica l sta in ing per fo rmed on the S O M A 10 per fused s tomachs revea led no ant ibody in the interstitium of the s tomach . Sta in ing of the F a b f ragments in the pe r fused organ (fig 37) r evea led that the F ab f ragments had entered into the interstitium, but were binding nonspecif ical ly to co l lagen. In vitro, it wa s determined that the S O M A 10 and espec ia l ly the Fab 127 f ragments bound nonspecif ical ly to co l lagen (fig 36), much more than another ant i -mouse monoc lona l antibody (anti-VIP; V31) . Both ant ibodies and the Fab f ragments p roduced more f l uo rescence than the backg round f luo rescence produced by incubating P B S on the co l lagen-coated plates. Ant ibod ies or Fab f ragments in the interstit ium cou ld not be detected by immunocytochemis t ry un less they bound to someth ing, because the v igorous process ing and multiple wa she s of the t issue would have removed any unbound molecu les . G r ange r (unpubl ished) found that when rad ioact ive ly- labe l led gastr in antibody was perfused into the intestinal circulation, after one hour, the antibody cou ld be detected in the lymphatic sys tem. Th i s demonstrat ion sugges ts that the b lood vesse l s are permeable to molecu les even as large as 150 kd. The di f ference in the gastr in re lease obta ined when S O M A 10 and its F a b fragment were infused cou ld be due to the permeabi l i ty of the ant ibody. The intact ant ibody is ~150 kd, whi le the F ab fragment is only - 5 0 kd. The sma l le r s i ze of the fragment wou ld inc rease the probabi l i ty of its penetrat ion through the b lood ve s se l into the interst i t ium. The detect ion of the F ab fragment, but not the intact antibody in the gastr ic interstitium may be due to the permeabi l i ty di f ferences of the molecules. Alternatively, the lack of detect ion of the intact ant ibody cou ld be due to the lower nonspec i f ic binding of S O M A 10 with respect to the fragment, or a combinat ion of both the reduced permeabil ity and de c r ea sed nonspec i f ic binding to co l lagen. The ant ibody may have been detec ted in the interstit ium if it were per fused for a longer per iod of t ime. If, however, S O M A 10 does not leave the b lood vesse l , the difference in the effect on gastr in re lease of S O M A 10 and the F ab fragment infusion cou ld be the result of an endocr ine and a paracrine effect of somatostat in, respectively. In the above exper iments , the ant ibody had only a s ing le p a s s age through the vascu lature. In order to inc rease the opportunity of the ant ibody enter ing into the interstitium, a sys tem was set up such that the ant ibody was recirculated (fig 29). After recirculation of the antibody for 20 min, the cumulated gastr in re lease from the S O M A 10 perfused s tomachs was much greater than from the controls, suggest ing that somatostat in tonical ly inhibits gastrin re lease. Th is is in agreement with results from the Fab infusion. A n argument against the validity of these results is that metabol i tes which would normally be el iminated, 128 are co l lec ted and recirculated. However , the bui ldup of metabol i tes would be present in both the control and the S O M A 10 perfused animals. Immunocy tochemica l sta in ing of the S O M A 10 rec i rcu lated s tomachs revea led no S O M A 10 in the interstitial space . The lack of detect ion may be due to the concentrat ion of S O M A 10 binding nonspecif ical ly to co l lagen, or that the ant ibody may have been broken down. B. Pas s i ve Immunization Studies in the Isolated Per fused Panc reas Most pass ive immunizat ion exper iments using somatostat in ant iserum in per fused pancreas (Sorenson et al . , 1980) and in vivo studies (Schusdz ia r ra et al . , 1978, Ste iner et al., 1978, Tannenbaum et a l , 1978) have y ie lded negative resu l t s . Howeve r , p a s s i v e immun i za t i on s tud ies in i so la ted i s le ts have consistent ly been success fu l . On ly Schusdz i a r ra et a l . (1980) showed that in v i vo in jec t ions of somatos ta t i n an t i s e rum into d o g s e n h a n c e d p l a s m a concentrat ion of insulin following a meal . Honey et al. (1981) demonstrated that the neutral izat ion of endogenous somatostat in us ing somatostat in ant iserum st imulated g lucagon and insul in re lease from the i so la ted per fused ch i cken pancreas . They stress that these results were obta ined in ch i cken and that the local effects of islet somatostat in in mamma l s may be different. In isolated rat i s le ts , i ncubat ion with somatos ta t in an t i se rum i n c r e a s ed insu l in r e l ease st imulated by leuc ine (Taniguchi et al . , 1979), arginine (Itoh et al . , 1980) and g lucose (Taniguchi et al., 1979, Itoh et al . , 1980). G lucagon re lease was a lso increased by somatostat in ant iserum in isolated islets (Barden et al., 1977, Itoh e t a l . , 1980). S O M A 10 has prev ious ly been shown to b lock the b io logica l effect of e xogenous somatostat in on GIP-s t imulated insul in re lease (Takemura et al . , unpubl ished). S imul taneous infusion of a g lucose gradient from 4.4 to 16.5 m M g lucose and of S O M A 10 at 5 or 10 u.g/ml or Fab fragments at 5 jig/ml into the iso lated per fused panc reas c au s ed no change in insul in re lease compared to contro ls . The lack of response cou ld be due to the low concentrat ion of the ant ibody and fragments used. Infusion of 10 (ig/ml Fab f ragments c au s ed a slight dec rease in insulin re lease st imulated by 16.5 m M g lucose (fig 32). The minor effect of the F ab 129 f r agment on insu l in r e l ease cou l d be attr ibuted to the sma l l effect of somatostat in on strongly st imulated insulin re lease (Taniguchi et al., 1977). Reduct ion of the insul in st imulus to 8.8 m M g lucose , and infusion of 30 ng/ml Fab f ragments c au sed insulin re lease to be dep r e s sed slightly (fig 33). However , there wa s a large inc rease in insul in re lease obse r ved fol lowing termination of the infusion. W h e n insulin re lease was st imulated with 8.8 m M g lucose and a S O M A 10 (45 p,g/ml) infusion, equivalent to 30 u.Q/ml F ab fragments, wa s appl ied, a more distinct dec rease in insulin re lease was observed (fig 34). Terminat ion of the infusion did not c au se the prominent increase in insul in re lease observed when the Fab fragment infusion was terminated. During infusion of S O M A 10, and increase in insulin re lease may not have been observed, because of the 22 fold greater d i ssoc ia t ion constant of the ant ibody in compa r i s on to the rat pancreat ic somatostat in receptor. The results from the latter three exper iments are contrary to what was expec ted , wh i ch w a s an inc rease in insul in re lease . Exp lana t i ons for the inhibit ion of insul in re lease are s imi lar to those for the inhibit ion of gastr in re lease by the infusion of S O M A 10. The dec rease in insulin re lease cou ld be a nonspec i f ic effect of the antibody. S O M A 10 cou ld be binding to somatostat in chang ing its conformation so that it becomes more suitable to its receptor, i.e. by increas ing its affinity or increas ing its binding t ime. Somatostat in neutral ization cou ld cause a disinhibit ion of a previously unknown inhibitor of insulin re lease. The dec rease in insulin re lease may a lso be c au sed by S O M A 10 binding to the somatostat in receptor, mimicking the effects of somatostat in. Immunocytochemica l sta ining of the pancrea ta per fused with 30 ng/ml Fab fragments and 45 ng/ml S O M A 10 revealed that the fragment and antibody had pe rmeated into the interstitial s pa ce s of the panc reas , and had bound to co l l agen conta in ing structures (fig 38). S ince both ant ibody and fragment had left the circulat ion, endocr ine and paracr ine effects of somatostat in cou ld not be d i s t ingu ished. Kvietys et a l . (1983) studied the permeabil ity of capi l lar ies in the isolated, b lood-per fused can ine panc reas with the double- indicator di lution technique. 1 3 0 They suggest that the fenestrat ions of the capi l lar ies restrict the movement of so lutes greater than 27 A diameter, therefore a l lowing polypept ide hormones (less than 15 A) to move freely ac ross vesse l wal ls, while restricting the leakage of p l a sma proteins (greater than 37 A) from the vesse l . However , the Krebs-bicarbonate buffer used in the present study to perfuse the pancreas conta ined dextran, which increases the s ize of the capil lary pores, al lowing the entrance of molecu les larger than would be a l lowed through in a phys io logica l situation. Immunoneutra l i zat ion of somatostat in with 0.5 mg S O M A 10 prior to surgery for the iso lated per fused pancreas , fo l lowed by infus ion of 20 |ig/ml S O M A 10 c au sed the first phase of insulin re lease to dec rease significantly (fig 35). No effect was observed on the second phase of insul in. The se results are contrary to what was expected . However , s ince the first, but not the se cond phase of insulin was affected, somatostat in may have more of an effect on the first phase of insulin re lease than on the second . Th is concurs with f indings by Gurry and Bennett (1976), in which they found the first insulin phase to be 25 to 50 t imes more sensit ive to somatostat in inhibition than the second phase. III. Conc lus ions Pur i f i ca t ion of mouse monoc l ona l ant ibody by a m m o n i u m su lpha te precipitat ion, in conjunct ion with hydroxylapatite chromatography is a fast and reproduceable method of purification. A s se s smen t of H A P S O M A 10 by affinity and ge l filtration H P L C revea led that it wa s over 9 0 % pure. S O M A 10 was found to b ind to the somatos ta t in mo lecu le at the s a m e reg ion as the somatostat in receptor, therefore neutral izat ion of somatostat in activity can be ach ieved. S O M A 10 has a large binding capacity. These character ist ics make S O M A 10 a va luable antibody for pass ive immunizat ion. F rom stud ies in which S O M A 10 recirculat ion and Fab infusion c au sed an inc rease in basa l gastr in re lease, it c an be conc l uded that somatostat in exer ts a con t i nuous restraint on gastr in re lease . S i n c e in the pe r fused panc reas exper iments, and possib ly in the perfused s tomach exper iments, both S O M A 10 and the F ab f ragment were enter ing into the interst i t ium, it is impo s s i b l e to d i s t i ngu i sh be tween endoc r i ne and pa rac r i ne e f fec ts of somatostat in in these preparat ions. Resu l t s from studies in which S O M A 10 131 was infused into the iso lated per fused s tomach , and pancreat ic studies were contrary to expected results, and further exper iments need to be performed. Better controls are needed for these exper iments. A s d i s cus sed earl ier, purif ied asc i tes fluid from mice injected with mye loma ce l ls cou ld be used as contro l ant ibod ies . Rec i rcu lat ion studies, s imi lar to those per formed in the s tomach , need to be performed in the pancreas . Pass i ve immunizat ion studies in s tat ic i n cuba t ed i s le ts s hou l d a l s o be p e r f o rmed , s i n c e p r ev i ou s immunoneutra l i za t ion s tud ies us ing this preparat ion have cons is tent ly been success fu l . To a c c e s s the binding of S O M A 10 to somatos ta t in in the per fused organs, S O M A 10 cou ld be infused first to bind endogenous somatostat in and to saturate the sys tem with S O M A 10. The sys tem cou ld then be titrated with somatostat in. To test the hypothes is that S O M A 10 binds to the somatostat in receptor mimick ing the effect of somatostat in , an anti- idiotypic ant ibody to S O M A 10 cou ld be infused in similar gastr ic and pancreat ic exper iments. If an increase in insulin re lease occurs, this would support the hypothesis. Somatostat in appears to have more inf luence on g lucagon re lease than on insul in re lease (Itoh et al . , 1980; Honey et al . , 1981), partial ly because of anatomica l cons idera t ions and because of greater A than B ce l l sensit ivity to somatostat in (Mandar ino et al . , (1981). Future studies shou ld invest igate the effect of S O M A 10 on g lucagon re lease. Future studies should a lso inc lude the infusion of other S O M A ant ibodies, s ince S O M A 10 has a higher d issoc iat ion constant than the pancreat ic somatostat in receptor and the low affinity gastr ic somatostat in receptor. S O M A 3, 8, and 20 have a lower d issoc ia t ion constant than these receptors, so somatostat in should preferential ly bind to the antibody and not the receptor. 132 R E F E R E N C E S Ahren B, Taborsky G J , Porte D. Neuropept idergic versus chol inergic and adrenerg ic regulation of hormone secret ion. Diabeto log ia 29:827-836, 1986. Alberti K G M M , Chr i s tensen N, Chr is tensen S E , P range-Hansen A, Iversen J , Lundbaek K, Seye r -Hansen K, Orskov H. Inhibition of insulin secret ion by somatostat in. Lancet 2 :1299-1301,1973. Al ino S F , Ga r c i a D, Uvnas-Moberg K. Effect of intragastric pH , prostaglandins and prostaglandin synthes is inhibitors on the re lease of gastrin and somatostat in into the gastr ic lumen of anaesthet ized rats. Ac ta Phys io l S c and 126:1-8, 1986. Aponte G , G r o s s D, Y a m a d a T. Capi l lary orientation of rat pancreat ic D-cell p rocesses : Ev idence for endocr ine re lease of somatostat in. A m J Phys io l 249: G599-606 .1985 . As ton R, Holder AT, P reece MA, Ivanyi J . Potentiation of somatogen ic and lactogenic activity of human growth hormone with monoc lona l ant ibodies. J Endocr ino logy 110: 381-388,1986. Bae tem K, Ma la i sse-Laga i F, Perrelet A, Orc i L. Endocr ine pancreas: Three-d imens iona l reconstruction shows two types of islets of Langerhans. S c i ence 206: 1323-1324,1979. Barden N, Lavoie M, Dupont A, Cote J , Cote J P . Stimulation of g lucagon re lease by addit ion of antisomatostatin serum to islets of Langerhans in vitro. Endocr ino logy 101: 635-638, 1977. B loom SR , Mort imer C H , Thorner MO , Besse r G M , Hall R, Gomez - P an A, Roy V M , Russe l l R C G , Coy DH , Kast in A J , Schal ly AV . Inhibition of gastrin and gastr ic ac id secret ion by growth hormone release-inhibit ing hormone. Lancet 2: 1106-1109, 1974. Bonner-Wei r S and Orci L. New perspect ives on the microvasculature of the islets of Langerhans in the rat. Diabetes 31 :883-889 ,1982 . B razeau P, Va le WL , Burgus R, Ling N, Butcher M, Rivier J , Gui l lemin R. Hypotha lamic polypeptide that inhibits the secret ion of immunoreact ive pituitary growth hormone. Sc ience 179:77-79, 1973. Buchan A M J , S ikora LK J , Levy J G , Mcintosh C H S , Dyck I, Brown J C . An immunocytochemica l investigation with monoc lona l ant ibodies to somatostat in. Histochemistry 83:175-180,1985. 133 Ch i ba T, Taminato T, Kadowak i S, Abe H, Ch iha ra K, Se ino Y , Matsukara S, Fujita T. Effects of g lucagon, secret in, and vasoact ive intestinal polypeptide on gastric somatostat in and gastrin re lease from isolated perfused rat s tomach. Gastroenterology 79: 67-71, 1980. Ch i ba T, Kadowak i S.Taminato T, Kadowak i , Ch iha ra K, Se ino Y, Matsukara S, Fujita T. Effect of antisomatostatin gamma-globul in on gastrin re lease in rats. Gastroentero logy 81: 321-326, 1981. Curry DL, Bennett LL, Li C H . Direct inhibition of insulin secret ion by synthetic somatostat in. B iochem Biophys Re s C o m m 58:885-889, 1974. Curry DL and Bennett LL. Does somatostat in inhibition of insulin secret ion involve two mechan i sms of act ion? P roc Nat A c a d Sc i U S A 73: 248-251 , 1976. Dunbar J C and Wa l sh MF . G lucagon and insulin secret ion by d i spersed islet ce l ls: Poss ib le paracr ine relat ionships. Hormone Re s 16: 257-267, 1982. G o m e z - P a n A, Reed JD , A lb inus M, Shaw B, Hall R, Bes se r G M , C o y DH , Kast in A J , Scha l ly AV . Direct inhibition of gastr ic ac id and peps in secret ion by growth hormone release-inhibit ing hormone in cats. Lancet 2: 888-890, 1985. Efendic S, Luft R, Gril l V. Effect of somatostat in on g lucose induced insulin re lease in isolated perfused rat pancreas and isolated rat pancreat ic islets. F E B S Letters 42 :169-172 , 1974. Efendic S, C la ro A, Luft R. Studies on the mechan i sm of somatostat in action on insulin re lease. Ac ta Endocr ino log ica 81: 753-761, 1976. Efendic S, L ins P E , Luft R, Uvnas-Wal lensten, S z e cowka J . Somatostat in -paracr ine or endocr ine subs tance? Front Hormone Re s 7: 41-51, 1980. Ger i ch J , Lovinger R, Grodsky G . Pancreat ic effects of somatostat in. Endocr ino logy 96: 750-754, 1975. God ing JW . Use of s taphylococca l protein A as an immunologica l reagent. J Immunological Methods 20: 241-253, 1978. Goodne r C J , Horn F G , Koerker D J . Hepat ic g lucose production osci l lates in synchrony with the islet secretory cyc le in fasting Rhesus monkeys. S c i ence 215: 1257-1260, 1982. Greenburg G R . Influence of vagal integrity on gastrin and somatostat in re lease in dogs. Gastroenterology 93: 994-1001, 1987. 1 3 4 Harty R F and Frankl in PA . Chol inerg ic mediat ion of gamma-amino butyric ac id-induced gastrin and somatostat in re lease from rat antrum. Gastroentero logy 91: 1221-1226, 1986. Hayes JR , Johnson D G , Koerker D, Wi l l iams R H . Inhibition of gastrin re lease by somatostat in in vitro. Endocr inology 96 :1374 -1376 , 1975. Holder AT, As ton R, Rest JR , Hill D J , Patel N, Ivanyi J . Monoc lona l antibodies can enhance the biological activity of thyrotropin. Endocr ino logy 120: 567-573, 1987. Hoist J J , J en sen SL , Knuhtsen S, Nie lsen OV , Rehfeld J F . Effect of vagus, gastr ic inhibitory polypeptide, and HCI on gastrin and somatostat in re lease from perfused pig antrum. A m J Phys io l 244: G515 -G522 , 1983. Hoist J J , J en sen SL , Knuhtsen S, Nie lsen OV . Autonomic nervous control of pancreat ic somatostat in secret ion. A m J Phys io l 245: E542-E548 , 1983. Honey R N , A r imua A, We i r G C . Somatostat in neutralization st imulates g lucagon and insulin secret ion from the av ian pancreas . Endocr inology 109: 1971-1974, 1981. Hopcroft DW, Mason DR, Scott RS . Structure-function relat ionships in pancreat ic islets: Support for intraislet modulat ion of insulin secret ion. Endocr ino logy 117: 2073-2079, 1985. Itoh M, Mandar ino L, Ger i ch J E . Antisomatostat in g a m m a globulin augments secret ion of both insulin and g lucagon in vitro. D iabetes 29: 693-696, 1980. J an s s on L and Hel lerstrom C. Stimulation by g lucose of the blood flow to the pancreat ic islets of the rat. Diabetologia 25: 45-50, 1983. Juarez -Sa l i nas H, Engelhorn S C , B igbee WL , Lowry MA, Stanker LH . Ultrapurif ication of monoc lona l ant ibodies by high-performance hydroxylapatite (HPHT) chromatography. B iotechniques May/June: 164-169, 1984. Kana t suka A, Mak ino H, O s e g a w a M, Kasanuk i J , Suzuk i T, Yo sh i da S, Hoie H. Is somatostat in a true local inhibitory regulator of insulin secret ion. D iabetes 33: 510-515, 1984. Kawa i K, Ipp E, Orc i L, Perrelet A, Unger R H . Circulating somatostat in acts on the islets of Langerhans by way of a somatostat in-poor compartment. S c i ence 218: 477-478, 1982. Kawa i K, Orc i L, Unger R H . High somatostat in uptake by the isolated perfused dog panc reas consistent with an "insulo-acinar" axis. Endocr ino logy 110: 660-662, 1982. 135 Koerker D J , Ruch W, Ch idecker E, Pa lmer J , Goodne r C J , Ens inck J , Ga l s e C C . Somatostat in: Hypothalamic inhibitor of the endocr ine pancreas . Sc i ence 183: 482-484, 1974. Konturek S J , Tas ler J , C ieszkowsk i M, Coy DH , Schal ly AV . Effect of growth hormone release-inhibit ing hormone on gastr ic secret ion, mucosa l b lood flow, and serum gastrin. Gastroenterology 70: 737-741,1976. Konturek S J , Tas le r J , Jaworek J , Pawl ik W, Wa laus K M , Schusdz ia r ra V, Meyers CA , Coy DH, Schal ly AV . Gastrointest inal secretory, motor, circulatory and metabol ic effects of prosomatostat in. P roc Natl A c ad Sc i U S A 78: 1967-1971,1981. Koop H, Behrens I, Bothe E, Mc intosh C H S , Pederson RA, Arnold R, Creutzfeldt W. Adrenerg ic and chol inergic interactions in rat gastr ic somatostat in and gastrin re lease. Digest ion 25: 96-102, 1982. Kvietys PR , Perry MA, Granger DN . Permeabi l i ty of pancreat ic capi l lar ies to smal l molecu les. A m J Phys io l 245: G519-G524 .1983 . Kwok Y N , Mc intosh C H S , Pederson RA, Brown J C . Effect of substance P on somatostat in re lease from the isolated perfused rat s tomach. Gastroentero logy 88:90-95, 1985. Kwok Y N , Verchere C B , Mc intosh C H S , Brown J C . Effect of galanin on endocr ine secret ions from the isolated perfused rat s tomach and pancreas . European J Pharmaco logy 145:49-54 1988. Larrson L l , Go l termann N, DeMagistr is L. Somatostat in cel l p rocesses as pathways for paracr ine secret ion. Sc ience 205: 1393-1395, 1979. Lefebvre P J , Luyckx A S , Brass inne A H . Inhibition by somatostat in of carbamylcho l ine- induced gastrin and g lucagon re lease from the isolated perfused canine s tomach. Gut 22: 793-797, 1981. Lov inger R, Bo ryczka AT, Shackel ford R, Kap lan SL , Ganong W F , G rumback M M . Effect of synthetic somatotropin re lease inhibiting factor on the increase in p l asma growth hormone elicited by L-Dopa in the dog. Endocr ino logy 95: 943-946, 1974. Lucey MR, W a s s J A H , Fairc lough P, Webb J , Medbak S, Ree s LH . Autonomic regulation of postprandial p l asma somatostat in, gastr in, and insul in. Gut 26: 683-688,1985. Mc in tosh C H S , Arnold R, Bothe E, Becker H, Kobberl ing J , Creutzfeldt W. Gastrointest inal somatostat in: Extraction and rad io immunoassay in different spec ies . Gut 19: 655-663, 1978. 136 Mcintosh C H S , Pederson RA, Mul ler M, Brown J C . Autonomic nervous control of gastr ic somatostat in secret ion from the perfused rat s tomach. Life S c i en ce s 29: 1477-1483,1981. Mc in tosh C H S , Kwok Y N , Mordhorst T, N ish imura E, Pederson RA, Brown J C . Enkephal inerg ic control of somatostat in secret ion from the perfused rat s tomach. C a n J Phys io l Pharmaco l 61:657-663, 1983. Mc in tosh C H S . Gastrointest inal somatostat in: Distribution, secret ion and phys io logica l s ignif icance. Life Sc i ences 37: 2043-2058, 1985. Mc in tosh C H S , Bak ich V, Kwok Y N , Brown J C . A compar ison of the inhibitory effects of somatostat in-14, -25 and -28 on motility of the gu inea pig i leum. C a n J Phys io l Pharmaco l 64:303-306,1986. Mandar ino L, Stenner D, B lanchare W, N issen S Ger i ch J , Ling N, B razeau P, Boh len P, E s ch F, Gui l lemin R. Select ive effects of somatostat in-14, -25 and -28 on in vitro insulin and g lucagon secret ion. Nature 291: 76-77, 1981. Mart indale R, Kauffman GL , Levin S, Wa l sh J H , Y a m a d a T. Differential regulation of gastrin and somatostat in secret ion from isolated perfused rat s tomachs. Gastroenterology 83: 240-244, 1982. M e d a P, Perrelet A, Orc i L. Increase of gap junctions between pancreat ic B ce l ls during stimulation of insulin secret ion. J Ce l l B io l 82: 441-448, 1979. Nishi S, Se ino Y, Takemura J , Ishida H, Seno M, Ch i ba A, Yana iha ra C, Yana iha ra N, Imura H. Vaga l regulation of G R P , gastr ic somatostat in and gastrin secret ion in vitro. A m J Phys io l 248:E425-431, 1985. Nishi W , Se ino Y , Ishida H, Seno M, Taminato T, Sakura i H, Imua H. Vaga l regulation of insulin, g lucagon, and somatostat in secret ion in vitro in the rat. J C l in Invest 79: 1191-1196, 1987. Nishi S, Se ino Y, Kitano N, Seno M, Tsuji K, Kurose T, Taminato T, T suda K, Yana iha ra C, Yana iha ra N, Imura H. Effects of naloxone on basa l and vagus nerve- induced secret ions of G R P , gastrin and somatostat in from the isolated perfused rat s tomach. Life Sc i ences 41 : 1787-1793,1987. O 'Shaughnessy D J , Long R G , Adr ian TE , Chr istof ides ND, Ghate i MA, Sa r son DL, B loom SR . Somatostat in-14 modulates postprandial g lucose levels and re lease of gastrointestinal and pancreat ic hormones. Digest ion 31 : 234-242,1985. O l e sen M Hoist J J , Sott imano C, Nie lsen OV . Autonomic nervous control of fundic secret ion of somatostat in and antral secret ion of gastrin and somatostat in in pigs. Digest ion 36: 24 -35 ,1987 . 137 Orc i L and Linger R H . Funct ional subdiv is ion of islets of Langerhans and poss ib le role of D cel ls. Lancet 2 :1243-1244, 1975. Pate l Y C . A high molecular weight form of somatostatin-28(1-12)-l ike immunoreact ive substance without somatostat in-14 immunoreactivity in the rat pancreas. J C l in Invest 72: 2137-2143, 1983. Pede rson RA, Mc in tosh C H S , Muel ler MK, Brown J C . Ab sence of a relationship between immunoreact ive-gastr in and somatostat in- l ike immunoreact iv ity secret ion in the perfused rat s tomach. Regulatory Pept ides 2: 53-60, 1981 . Penman E, W a s s J A H , Butler M G , Penny ES , Pr ice J , W u P, Rees LH . Distribution and character izat ion of immunoreact ive somatostat in in human gastrointestinal tract. Regulatory Pept ides 7: 53-65,1983. Po lonsky K, J a span J , Berelowitz M, Pugh W, Moo s s a A, Ling N. The in vivo metabol ism of somatostat in 28: Poss ib le relationship between d imin ished metabol ism and enhanced biological act ion. Endocr ino logy 111: 1698-1703, 1982. Reichl in S. Somatostat in. New Eng land Journa l of Med ic ine 309:1556-1563, 1983. Rey l -Desmars F and Lewin M J M . Ev idence for an intracellular somatostat in receptor in pancreas: A comparat ive study with reference to gastric mucosa . B iochem Biophys Re s C o m m 109: 1324-1331, 1982. Saffouri B, We i r G , Bitar K, Makhlouf G . Stimulation of gastrin secret ion from the perfused rat s tomach by somatostat in ant iserum. Life Sc i ences 25: 1749-1754, 1979. Saffouri B, We i r G , Bitar K, Makhlouf G . Gastr in and somatostat in secret ion by perfused rat s tomach: Funct ional l inkage of antral pept ides. A m J Phys io l 238: G495-501 , 1980. Saffouri B, DuVa l J W , Makhlouf G M . Stimulation of gastrin secret ion in vitro by intraluminal chemica l s : Regulat ion by intramural chol inergic and nonchol inergic neurons. Gastroentero logy 87: 557-561, 1984. Saffouri B, DuVa l JW , Ar imura A, Makhlouf G . Effects of vasoact ive intestinal peptide and secret in on gastrin and somatostat in secret ion in the per fused rat s tomach. Gastroenterology 86(5): 839-842, 1984. Samo l s E, We i r G C , Ramseur R, Day JA , Patel Y C . Modulat ion of pancreat ic somatostat in by adrenergic and chol inergic agon ism and by hyper- and hypog lycemic su l fonamides. Metabo l i sm 27 suppl 1 :1219-1221, 1978. Samo l s E, S tagner J l , We i r G C . Autonomic function and control of pancreat ic somatostat in. Diabetologia 20: 388-392, 1981. 1 3 8 Samo l s E, Bonner-Wei r S, We i r G C . Intra-islet insul in-glucagon-somatostat in relationship. C l in ics in Endocr ino logy and Metabo l i sm 15:33-58, 1986. Schubert M L and Makhlouf G M . Regulat ion of gastrin and somatostat in secret ion by intramural neurons: Effect of nicotinic receptor stimulation with d imethyl-phenylp iperaz in ium. Gastroentero logy 83: 626-632, 1982. Schusdz ia r ra V, Roui l ler D, Ar imura A, Unger R H . Ant isomatostat in serum increases levels of hormones from the pituitary and gut, but not from the pancreas . Endocr ino logy 103: 1956-1959, 1978. Schusdz ia r ra V, Zyznar E, Roui l ler D, Boden G , Brown J C , A r imua A, Unger R H . Sp lanchn ic somatostat in: A hormonal regulator of nutrient homeostas is . S c i ence 207: 530-532, 1980. Sea l A, Y a m a d a T, Debas H, Hol l inshead J , Osadchey B, Aponte G, Wa l sh J . Somatostat in-14 and -28: C lea rance and potency on gastr ic function in dogs. A m J Phys io l 243: G97 -G102 , 1982. Sea l A M , Me loche R M , Lui Y Q E , Buchan A M J , Brown J C . Effects of monoc lonal ant ibodies to somatostat in on somatostat in- induced and intestinal fat-induced inhibition of gastric ac id secret ion in the rat. Gastroentero logy 92: 1187-1192, 1987. Schubert ML, Wa l sh J H , Makhlouf G M . Bombes in ant iserum inhibits neurally-st imulated gastrin secret ion in the isolated perfused s tomach. Gastroentero logy 84: 1302, 1983. (Abs) S ikora L K J , Buchan A M J , Levy J G , Mc intosh C H S , Brown J C . VIP- lnnervat ion of rat intestine: A deve lopmenta l sudy with monoc lonal ant ibodies. Pept ides 5:231-237, 1984. Shaf fer -Tannenbaum G , Epe lbaum J , Co l le E, B razeau P, Martin J B . Ant iserum to somatostat in reverses starvat ion- induced inhibition of growth hormone but not insulin secret ion. Endocr ino logy 102: 1909-1914, 1978. Short G M , Doyle W, Wolfe M M . Effect of ant ibodies to somatostat in on ac id secret ion and gastrin re lease by isolated perfused rat s tomach. Gastroentero logy 88: 984-988, 1985. S i ler T M , VandenBe rg G , Y e n S S C . Inhibition of growth hormone re lease in humans by somatostat in. J C l in Endocr inol Metab 37: 632-634,1973. So renson RL , L indel DV, E lde R P . G lucose stimulation of somatostat in and insulin re lease from the isolated perfused rat pancreas. Diabetes 29: 747-751 , 1980. S tagner J l , Samo l s E, We i r G C . Susta ined osci l lat ions of insulin, g lucagon and somatostat in from the isolated can ine pancreas during exposure to a constant g lucose concentrat ion. J Cl in Invest 65: 9 39 , 1980 . 139 Stagner J l and Samo l s E. Perturbation of insulin osci l lat ions by nerve b lockade in the in vitro can ine pancreas. A m J Phys io l 248: E516 -E521 , 1985. S tagner J l and Samo l s E. Role of intrapancreatic gangl ia in regulation of periodic insular secret ions. A m J Phys io l 248: E522-E530 , 1985. S tanker L H , Vander laan M, Juarez-Sa l inas H. One-step purification of mouse monoc lona l ant ibodies from asc i tes fluid by hydroxylapatite chromatography. J Immunol Methods 76: 157-169, 1985. Ste iner RA, Stewart JK , Barber J , Koerker D, Goodner C J , Brown A, Miner P, Ga l e C C . Somatostat in: A physiological role in the regulation of growth hormone secret ion in the adolescent male baboon. Endocr ino logy 102: 1587-1594, 1978. Taborsky G J . Ev idence of a paracrine role for pancreat ic somatostat in in vivo. A m J Phys io l 245: E598-E603 , 1983. Taniguchi H, Utsumi M, Hasegawa M, Kobayash i T, Watanabe Y , Murakami K, Sek i M, Tsutou A, Mak imua H, Sakoda M, B aba S. Phys io log ic role of somatostat in. Insulin re lease in rat islets treated by somatostat in ant iserum. Diabetes 26: 700-702, 1977. Taniguch i H, Ha segawa M, Kobayash i T, Watanabe Y , Murakami K, Sek i M, Tsutou A, Utsumi M, Mak imua H, S a koda M, B aba S. Enhancement of leuc ine- induced insulin secret ion by somatostat in ant iserum pretreatment. Horm Metab Re s 11: 23 -26 ,1979 . Uvnas-Wal lens ten K, Efendic S, Roovete A, Johansson C. Dec reased re lease of somatostat in into the portal vein fol lowing electr ical vaga l stimulation in the cat. A c t a Phys io l S c and 109: 393-398, 1979. Va le W, Rivier J , Ling N, Brown M. Biologic and immunologic activities and appl icat ion of somatostat in ana logs. Metabol i sm 27 suppl 1 :1391-1401, 1978. Ve ld PA , P ipe leers D G , Gepts W. Ev idence against the presence of tight junct ions in normal endocr ine pancreas. D iabetes 33 :101 -104 , 1984. Vol ler A, Bidwel l DE , Bartlett A. Enzyme immunoassays in d iagnost ic medic ine. Bull W H O 53 :55 -65 , 1976. Wei r G C and Bonner-Weir S. Pancreat ic Somatostat in. In: Y C Pate l , G S Tannenbaum (ed.) Somatostat in. pp403-423, P l enum Press , New York, N.Y., 1985. Wol fe M M , Ree l G M , McGu igan J E . Inhibition of gastrin re lease by secretin is . mediated by somatostatin in cultured rat antral mucosa . J Cl in Invest 72: 1586-1593, 1983. 140 Wolfe M M , Ja in DK, Ree l M, McGu igan J E . Effects of carbacho l on gastrin and somatostat in re lease in rat antral t issue culture. Gastroentero logy 87: 86-91 , 1984. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0097854/manifest

Comment

Related Items