`Chemistry
`
`Proceedings
`of the First International Symposium
`and Pakistan-U.S. Binational Workshop,
`Karachi, Pakistan
`
`Edited by
`Atta-ur-Rahman
`
`Springer-Verlag
`Berlin Heidelberg NewYork
`London Paris Tokyo
`
`|
`
`MATA Exhibi 1033
`MAIA V. BRACCO
`IPR PETITION
`
`
`
`
`MAIA Exhibit 1023
`MAIA V. BRACCO
`IPR PETITION
`
`
`
`Professor Atra-ur-RAHMAN
`H.E.J. Research Institute of Chemistry
`University of Karachi
`Karachi-32, Pakistan
`
`With 142 Figures and Numerous Schemes
`
`ISBN-13: 978-3-642-71427-6
`DOI: 10.1007/978-3-642-71425-2
`
`e-ISBN-13: 978-3-642-71425-2
`
`Library of Congress Cataloging-in-Publication Data. International Symposium and Paki-
`stan-U.S. Binational Workshop on Natural Product Chemistry (ist : 1984 : Karachi, Paki-
`stan). Natural product chemistry. “The first International Symposium and Pakistan-U.S.
`Binational Workshop on Natural Product Chemistry ... February 1984”—Pref. Includes
`index.
`1. Natural products-Congresses.
`I. Rahman, Atta-ur-, 1942-
`.
`IL. Title.
`QD415.A1I47
`1984.
`547.7
`—-86-20420.
`
`This work is subject to copyright. All rights are reserved, whether the whole or part of the
`material is concerned, specifically those of translation, reprinting, re-use of illustrations,
`broadcasting, reproduction by photocopying machine or similar means, and storage in data
`banks. Under § 54 of the German Copyright Law, where copies are made for other than
`private use, a fee is payable to “Verwertungsgesellschaft Wort”, Munich.
`
`© Springer-Verlag Berlin Heidelberg 1986
`Softcover reprint of the hardcover Ist edition 1986.
`
`The use of registered names trademarks, etc. in this publication does not imply, even in the
`absence of a specific statement, that such names are exempt from the relevant protective
`laws and regulations and therefore free for generaluse.
`
`Offsetprinting: Druckhaus Beltz, Hemsbach/Bergstr.
`
`2131/3130-543210
`
`
`
`
`
`
`Atta-ur-Rahman (ed.), Natural Product Chemistry
`© Springer-Verlag Berlin Heidelberg 1986
`
`-255-
`
`
`
`
`
`
`296
`
`The main structural characteristics of the gastrin-cholecysto-
`
`kinin family of peptide hormones are:
`
`1)
`
`the common C-terminal
`
`tetrapeptide amide sequence Trp-Met-Asp-Phe-NHy which already
`in early structure-function studies on gastrin was recognized
`
`as the minimum fragment with physiological activity although
`
`at low potency, and thus as a kind of active center of the hor-
`
`mone (1);
`
`2)
`
`the tyrosine-O-sulfate residue which for chole-
`
`cystokinin (CCK) represents an essential structural feature for
`
`full hormonal expression, whereas for the gastrins its role has
`
`not yet been identified; they are isolated in the nonsulfated
`
`form I as well as in the sulfated form II (2,3) and no signifi-
`
`cant differences in their physiological actions have been de-
`
`3)
`
`the complex size-heterogeneity as resulting
`
`tected so far;
`
`from the posttranslation processing of their precursor mole-
`
`Fig. 1. Primary structures of A) human gastrins (HG) and B)
`
`porcine cholecystokinins. Arrows indicate enzymatic
`
`cleavages of the prohormonal molecules with production
`
`of gastrins and cholecystokinins of decreasing chain-
`
`length. The C-terminal amides result from enzymatic
`
`processing of glycine-extended forms
`
`-256-
`
`-256-
`
`
`
`
`
`
`257
`
`cules (Fig. 1)
`
`(4). To better understand the structural bases
`
`for the biological actions of this class of hormones and to pos-
`
`sibly identify the physiological significance of the observed
`
`heterogeneity both in respect to sulfation and extension at the
`
`N-termini, extensive synthetic studies have been performed in
`
`the last years in our laboratory (5-15). Aim of these inten-
`
`sive efforts was
`
`i)
`
`to obtain the various components of this
`
`class of hormones at the highest possible degree of purity for
`
`structure-function studies
`
`ii) to identify fully active ana-
`
`logs stable on storage and handling as needed for physiological
`
`and clinical studies, and finally iii) to synthetize deriva-
`
`tives well suited for an efficient improvement of the immuno-
`
`logical methods necessary for a reliable quantitative and selec-
`
`tive evaluation of the different components both when circula-
`
`ting in the plasma and in the tissue of their biosynthesis. Our
`
`contributions in the field are reviewed in the following:
`
`Synthesis of Gastrins
`
`The main difficulties encountered in the synthesis of gastrin
`
`and CCK related peptides derive from those structural elements
`which are essential for the biological activity and thus, di-
`
`rectly involved in the mechanism of action of this class of hor-
`
`mones, i.e.
`
`from the thioether function of methionine,
`
`the in-
`
`dole side-chain function of tryptophan and the tyrosine-O-sul-
`
`fate moiety. This sequence-dependent enhanced reactivity par-
`
`ticularly of the methionine and tryptophan residue towards
`
`oxidants and electrophiles as well as the instability of the
`
`sulfate-ester function has represented for years a challenge
`
`for the peptide chemist.
`
`Oxidation of methionine: The pronounced tendency of the
`methionine residues in the gastrins and CCK-peptides to oxi-
`
`dation to the corresponding S-oxide derivatives with concomi-
`
`tant loss of activity (16-17)
`
`implies careful avoidance of
`
`traces of oxidants, e.g. peroxides and heavy metals along the
`
`various steps of synthesis and purification as well as of air-
`
`oxygen to prevent desactivation. Even if reduction of S-oxides
`
`with thiols is possible (18), handling of these peptides for
`
`in vitro and in vivo assays becomes delicate.
`
`-257-
`
`-257-
`
`
`
`
`
`
`258
`
`Additionally the lability of the tyrosine-O-sulfate ester
`
`prevents an "a posteriori reduction" of S-oxide derivatives
`
`by thiols (15). Basing on first indications of Morley (17),
`Kenner et al.
`(19) and Wtinsch et al.
`(5 ~- 6) both independently
`
`revealed the possibility to replace the methionine residue
`
`in the gastrins by the branched aliphatic side-chain of
`
`leucine with retainment of full biological activity. Subsequent
`
`substitutions of methionine by norleucine or its oxa-analogue
`methoxinine proved to be equally efficient
`(20).
`The related
`
`analogs were equipotent to the parent methionine- hormone in
`
`stimulating gastric acid secretion.
`
`Oxidative degradation of tryptophan: Since the first studies
`on tryptophan-containing peptides it is known that the indole
`
`function undergoes facile oxidation with production of
`
`usually colored complex mixtures consisting of components
`
`which are not yet studied in detail. This oxidative
`
`degradation occurs in pronounced manner in acid solutions,
`e.g. as needed in acidolytic deprotection steps.
`It has
`been, however, observed that addition of phenols to free
`the acid mixtures from halogens as well as operating in inert
`
`atmosphere such as under argon or adding antioxidants, e.g.
`diethylphosphite or preferentially thiols, constitute effi-
`
`cients methods to prevent or at least to minimize the oxi-
`
`dative destruction of tryptophan residues.
`
`Alkylation of tryptophan: A more serious side reaction at the
`level of tryptophan residues was detected in our laboratory
`during the synthetic work on gastrins, i.e.
`the alkylation
`
`of the indole ring as generated in the acidolytic removal
`
`of protecting groups. First indications of a possible chemical
`
`modification of the indole group derived from the synthetic
`
`studies on glucagon (21) and then, particularly on 15-leucine-
`human little gastrin-I (6).
`
`An exact identification of the side products and thus, of the
`
`nature of the side reaction which had occurred, was difficult
`
`because of the presence of two tryptophan residues in the
`
`little gastrin molecule. We
`
`then succeeded to identify this
`
`side reaction during the synthesis of a leucine-analogue of
`
`-258-
`
`-258-
`
`
`
`
`
`
`259
`
`minigastrin-I (10) initially proposed to correspond to the
`
`C-terminal tridecapeptide amide of little gastrin-I (22).
`
`Upon removal of the protecting groups on tert-butanol basis
`
`by exposure to trifluoroacetic acid and partition chromato-
`
`graphy of the resulting crude product on Sephadex G-25,
`main side fraction was identified as 10-N1"_tert-butyl-
`tryptophan-11-leucine-minigastrin-I by mass- and nmr-
`
`the
`
`spectroscopy of the C-terminal chymotryptic fragment.
`
`Subsequently in collaboration with Kisfaludy and associates,
`
`studies on model compounds allowed the characterization of
`
`the various alkylation products of the indolering formed in
`
`the acidolytic cleavage of protecting groups on tert-butanol
`basis. Besides the w?"_tert-butyl-derivative, additionally
`mono-C-tert-butyl-, poly-C-tert-butyl- and poly-C-, N-tert-
`
`butyl-derivatives are produced whereby the extent of such
`
`alkylations and the distribution pattern of the various
`
`substitutions at the indolefunction were found to strongly
`
`depend on the acids used, on time and temperature (23). To
`
`possibly minimize this deleterious side reaction which causes
`
`remarkable loss
`
`of product,
`
`the use of scavangers, e.g.
`
`indole derivatives, ethers and thioethers as well asa
`
`series of thiols and their mixtures, have carefully been
`
`examined by extensive comparative studies. All efforts to
`
`entirely suppress indole alkylation failed, even if
`
`1,2-ethanedithiole was found to efficiently reduce the rate
`
`of this side reaction as shown in Fig.
`
`2
`
`(11).
`
`lH-nmr, tle,
`Fortunately various analytical methods e.g.
`hplc, were found to be sufficiently sensitive to permit
`
`detection and quantitative monitoring of the extent of sucha
`
`side reaction,
`
`thus remarkably facilitating the purification
`
`processes.
`
`In this context noteworthy is the observation
`
`that the rate of alkylation of the indole function strongly
`
`depends on the chemical environment of the tryptophan
`
`residue. While during the synthesis of gastrins and gastrin
`
`analogs as well as gastrin-related peptides tert-butyl-
`
`tryptophan-derivatives have been detected in extents up to
`
`15 % even applying efficient scavangers (11), no alkylation
`
`-259-
`
`-259-
`
`
`
`
`
`
`260
`
`Areo 0
`
`A280
`
`8
`
`2
`
`16
`
`2
`
`2%
`
`28min
`
`Fig. 2.
`
`Hplc of crude i-des-tryptophan, 11-leucine-minigastrin-I
`as resulting from exposure of the fully protected
`tridecapeptide derivative Boe-Leu-[Glu (OBu>)] ,~Ala-
`Tyr (But ) -Gly-Trp-Leu-Asp (OBu“)-Phe-NH., to I)
`
`trifluoroacetic acid/anisole, 10:1 and II)
`
`trifluoro-
`
`acetic acid/anisole/1,2-ethanedithiol, 10:1:0.25.
`
`Chromatographic conditions: yw-Bondapak C 18
`
`(30 x 0.4 cm); eluents: A) 0.1 M sodium phosphate,
`pH 5.4;
`B) acetonitrile/eluent A, 45:55;
`linear
`
`gradient of 35 % B to 80 % B in 30 min;
`
`flow rate:
`
`1.7 ml/min
`
`of the tryptophan residue was observed upon acidolytic
`
`removal of tert-butanol and 1-adamantanol derived protecting
`
`groups from fully protected somatostatin (24). Consequently,
`
`this sequence-dependent reactivity to electrophiles strongly
`
`questions the general efficacy of acidolytic deprotection
`
`procedures devised on the basis of model studies with simple
`
`tryptophan~-compounds.
`
`Racemization:
`
`An additional serious difficulty encountered
`
`in the synthesis of gastrins is related to the last fragment
`
`condensation step, i.e. assembly of the N-terminal pentapeptide
`Pyr-Gly-Pro-Trp-Leu-OH with H- [Glu (OBu*) ] ,-Ala-Tyr (Bu*) -Gly-
`
`-260-
`
`-260-
`
`
`
`
`
`
`261
`
`Trp-X-Asp (OBu‘) -Phe-NH, via DCC/HONSu even using a 3-to 4-fold
`excess of the carboxyl component (5). The
`low acylation rate
`
`induced us to a detailed investigation (11) of various known
`
`coupling methods (Table I).
`
`With benzotriazole as additive the acylation rate was sensibly
`enhanced; concomitantly quantitative racemization of the
`C-terminal
`leucine residue occurred (25). This coupling procedure
`proposed by Kénig and Geiger
`(26) has widely been used and it is
`surprising that its negative side effects have not been recognized
`earlier. The danger of related remarkable racemization rates
`has been confirmed by Sieber et al.
`in the synthesis of insulin
`
`(27) and by us in the synthesis of secretin (28) and PHI
`
`(29).
`
`The experiments listed in Table I indicate as only
`
`alternative to the DCC/HONSu method (30)
`
`the use of
`
`2-morpholinoethylisocyanide in presence of N-hydroxysuccinimide
`
`(31).
`
`These above summarized meticulous studies on the observed side
`
`reactions as well as the experiences gained in parallel
`
`synthetic studies on other gastrointestinal hormones allowed
`
`an efficient feedback for an optimization of the synthetic routes
`
`to gastrins-I (11)
`
`(Fig. 3) and gastrin-I-related peptides (12)
`
`The desired peptides became accessible ata
`(Fig. 4).
`high degree of purity as judged by thin-layer chromatography
`under various conditions, high pressure liquid chromatography
`
`in optimized conditions, electrophoresis, uv- and nmr-
`spectroscopy, amino acid analyses of the acid and enzymatic
`
`hydrolysates as well as by chiral analysis according to the
`method of Bayer and associates (32).
`
`Synthesis of tyrosine-O-sulfate peptides
`As mentioned initially the closely related gut hormones CCK
`and gastrin-II share as common characteristic a tyrosine-O-sulfate
`residue, whereby for latter hormone the nonsulfated form-I has
`also been isolated. The question arises of whether circulating
`
`in plasma the gastrins are sulfated and desulfation takes
`
`place during isolation or of whether both the sulfated and
`
`nonsulfated gastrins constitute native components. To shed
`
`light on this undoubtedly important question synthetic
`
`-261-
`
`-261-
`
`
`
`
`
`
`
`
`IeTqed
`
`
`
`
`
`
`
`
`
`[2t-9]CunN-oud-(nd30)dsy-etn-daz-At9-(nga)tAn-etw-"[(ng30)ND]-H+[S-L]HO-neq-dayp-o1g-AT9-17Ag-H
`
`
`
`
`
`
`
`
`
`
`
`[LL-L]“uN-oua-(ng30)dsy-eTN-dan-AT9-(naz)ZAL-eTW-"[(na3O)NTN]-neT-day-o1g-AT9-24a-H
`
`
`
`
`
`
`
`
`
`:JOZonpoidapnazsutj4uay\UOD
`
`
`
`
`
`%neq-d%[LL-9]SUOTFIPUODUCTzoReTUOTFEATIORpoyjeu
`
`
`
`
`
`262
`
`
`
`(SjucuTisdxe7)
`
`
`
`0S{-OLpp44ysutuo6NSNOH/IWW
`
`
`
`€+0SSl-Olpe“a:upzOoob-340H/D0a
`
`(SqucuTTedxe¢)
`
`@~0SZ-StpeSatupzoob-NSNOH/90d
`
`(SqueUTrdsdxe¢)
`
`(L
`
`(Z
`
`
`
`{Tow/Tou]ofzer
`
`{Z£t-9]/[s-t]
`
`-262-
`
`-262-
`
`
`
`
`€
`€
`€
`
`
`
`
`
`
`
`
`cHN-2uUd-(ng30)dsy-3ew-daz-AT9O-(ng3)TAL-eTw-%(na30)nTD)-neI-dag-o7rd-A[9-1ha
`
`
`
`
`
`
`
`NSNOH/Odd
`
`
`
`IT-uTiajsehaTARTTueumyofaqnoxzoTRaeyqUAS*E*bTtg
`
`
`
`
`
`
`
`ToTyatpeueyze-z‘t/atostue;Hooa*a5
`
`
`
`
`
`AyderbojeuoryouotytTqaed
`
`
`
`CuN-oug-dsy-teW-daL-A19-24L-eTW-2[no]-net-dap-o1g-At9-aha
`
`
`
`
`
`Zz
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`HN-eud-(n@30)dsy-jaw-day-ATD-(nq)TAL-etw-*{(30)nT9)-H+HO-neT-day-o1rg-AT9-1hg
`
`Zz
`
`
`
`
`
`
`
`
`
`
`
`HN-2Ud-(N30)dsy-3ew-daz-ATO-(ng@})ZAL-eTW-SdN.
`
`
`
`NSNOH/3dda
`
`{
`
`
`
`@HN-28uUd-(g90)dsy-3eW-H+HO-day-ATO-(ng3)aAL-eTW-sdn
`
`
`
`
`
`
`
`
`
`@ToputTAyzeU~-Z/TOH
`
`
`
`
`
`ZHN-eud-(ng@30)dsy-qew-dap-AT9-(nqq)AAL-CTW-H+HO-*[(n@30)NT]-sdn
`
`
`
`
`
`
`
`
`
`USNOH/30d
`
`Zz
`
`
`
`
`
`
`
`
`
`
`
`
`
`HN-9ud-(n@30)dsy-3eW-daL-AT9~(ng)TAL-eTw-9[(ng30)NTO]-sdn
`
`263
`
`
`
`
`
`eToputTAyjeu-—z/TOH
`
`-263-
`
`-263-
`
`
`
`
`
`
`264
`
`Z-Glu(OBu")-Ala-Tyr(Bu‘)-Gly-OH[10~13] + H-Trp-Nle-Asp(OBu*)-Phe-NH} (Nle!5}[14-17b]
`
`DCC/HONSu
`
`Z-Glu(OBu")-Ala-Tyr(Bu’)-Gly-Trp-Nle-Asp(OBu‘)-Phe-NH[Nle!5|{10-—17a]
`
`H2/Pd
`
`Z-{Glu(OBu*)}4-OH[6~—9] + H-Glu(OBu4)-Ala-Tyr(Bu‘)-Gly-Trp-Nle-Asp(OBu‘)-Phe-NH» [Nle!5}{10-17b]
`
`DCC/HONSu
`
`Z-[Glu(OBu!)}5-Ala-Tyr(Bu‘)-Gly-Trp-Nle-Asp(OBu!)-Phe-NH> [Nle! 5 }{6—17a]
`
`H2/Pd
`
`CC1u-Gly-Pro-Trp-Leu-OH{1—5} + H-(Glu(OBu‘)]g-Ala-Tyr(Bu‘)-Gly-Trp-Nle-Asp(OBu‘)-Phe-NH[Nle!5][6-17b]
`
`DCC/HONSu
`
`CGIu-Gly-Pro-Trp-Leu-[|Glu(OBu!)|s-Ala-Tyr(Bu’)-Gly-Trp-Nle-Asp(OBu‘)-Phe-NH> [Nle!54[1-17a}
`
`CF,CO,H/anisole/ 1,2-ethanedithiol
`partition-chromatography
`
`CGlu-Gly-Pro-Trp-Leu-Glu-Glu-Glu-Glu-Glu-Ala-Tyr-Gly-Trp-Nle-Asp-Phe-NII>
`
`{Niel S|I{L-17b]
`
`Fig. 4. Synthetic route to 15-norleucine-human little
`
`gastrin-I. The identical synthetic scheme was
`
`followed for the preparation of the 15-leucine-
`
`and 15-methoxinine-analog
`
`sulfated gastrins should be accessible.So far,
`
`the synthesis
`
`of sulfated gastrins has not been achieved; previous attempts
`
`in Kenner's laboratory (33) based on direct sulfation of
`gastrin using pyridine/so, failed. Surprisingly more success
`was obtained with this method for the preparation of CCK-
`
`peptides upon optimization of the reaction conditions and
`
`extensive purification of the final compounds (34 - 35).
`
`Nonetheless several side reactions such as nuclear sulfonation
`
`of the tyrosine side-chain could not be prevented;
`
`the extent
`
`of sulfonation was found to strongly depend on sulfating
`agents used, on the operating conditions and on the chemical
`
`-264-
`
`-264-
`
`
`
`
`
`
`265
`
`environment of the tyrosyl residue (36). The mentioned
`
`difficulties compelled us to devise a new synthetic route
`
`te tyrosSine-O-sulfate-peptides based on the direct use of
`
`suitable derivatives of tyrosine-O~-sulfate. This key inter-
`mediate was obtained by sulfation of 2Z-Tyr-OH with pyridine/
`S03 followed by its isolation as barium salt. The homogeneity
`of this starting material
`is easily assessed by chromato-
`graphic and spectroscopic methods: free tyrosine or sulfonated
`
`derivatives show red-shifted uv-maxima of strong intensities
`
`in basic solution. Equally sensitive is the detection of such
`contaminants by ‘yenmr
`(13). The usefulness of N*-acyl-
`tyrosine-O-sulfate as starting material in the synthesis of
`
`peptides was then investigated on the C-terminal sequence
`24-33 of CCK (Fig. 5). For this purpose Fmoc-Tyr (SO0,Ba, ,>)-OH
`was prepared and coupled to the C-terminal hexapeptide
`
`derivative as outlined in the synthetic scheme; subsequent
`removal of the N*-protecting group with piperidine proceeded
`smoothly as well as the N-terminal elongation to the deca-
`peptide derivative 24-33, which upon N°-deprotection by
`treatment with piperidine represents a suitable key-
`
`fragment for the synthesis of higher molecular weight
`
`CCK-peptides. Removal of the residual acid-labile protecting
`
`groups by exposure to 90 % trifluoroacetic acid followed
`
`by chromatographic purification of the crude product led
`
`to the CCK-decapeptide in over 60 % yield as homogeneous
`
`material as judged by all the analytical assays performed
`
`(14). The sulfate-ester moiety was found to be surprisingly
`
`stable towards the acidolytic deprotection step, most
`probably because of the sequential proximity of the strong
`basic function of the arginine side-chain.
`In fact the sulfate
`ester function in tyrosine-O-sulfate and related peptides
`is hydrolysed under identical conditions usually to about
`20 %. As expected the alkylation of the tryptophan
`residue and the high tendency of the methionine residues
`to oxidation were again observed. Unfortunately reduction
`ot
`the S-oxides with thiols as well as partial suppression
`of indole-alkylation by thiols as scavangers is precluded
`by the facile hydrolysis of the sulfate ester under acidic
`conditions in the presence of thiols (15).
`
`-265-
`
`-265-
`
`
`
`
`
`
`
`
`
`
`[a¢¢—87)THN-2Ud-GngO)dsy-1eW-dIL-AID-PW-H+[ITLZ)@/Tego-(@/Teg€gg)iAj-00ury
`
`
`
`[e97~-$7]HO-jngo)dsy-(%Z)81y-Z
`
`IGOH/90d|paltHy
`
`HO?W/IDH
`
`Zz
`
`OCH:3uantTe!92-HT
`
`
`
`
`
`HO-(@/legEQsy4L-H+[9-00ug
`
`[97]HO-Gndo)dsy-H+
`(szASNO-(2Z)3I1y-Z
`
`
`
`
`
`
`
`(118¢€-2£7]@HN-aud-(,nGO)dsy-19W-dIL-A1D-20W-(7/TegEQg)14|-D0uL,y{997—-$7]HO-GdO)asy-(lOH)aIV-H6+)«[p7]NSNO-GndO)dsy-coury
`
`
`
`
`
`
`
`
`
`
`
`[€€-pz~]-MDDFOstsayQudAsjFosuayos
`
`°¢g*bta
`
`
`
`
`
`
`
`
`
`[€€-#7}Zd-NOD‘loE€-#7)THN-aug-dsy-1aW-diL-A19-20W-(HEOS)IAL-dsy-81y-dsy-H
`
`
`
`
`
`e1oputTAyyeu-z/HOOD*43D&06
`
`(t
`
`
`
`Aydeizbo3euorzysashueyoxa-uor
`
`(€
`
`9dTabotguouoTzeIIITITES(z
`
`
`
`
`
` autptazaditd|[ege—%7)SHN-24d-(,NGO)dsy-20W-4IL-419-29W-CHEOS)IAL2G0)dsy-31y-(,ngO)dsy-souly
`
`
`
`lace—b7}CHN-2Ud-(,9gO)dsy-eW-d1L-419-19W-(HEOS)IAL(gO)dsy-31y-(,ngo)dsy-H
`
`
`
`autptiadrd
`
`266
`
`ASNOH/O0d
`
`[qee-27)THN-2ud-(,ngo)dsy-r9W-daL-AID-12W-(Z/Teg€QS)IAL-H
`
`
`
`+[97-67]HO-(,9€0)dsW-(IOH)31V-nao)dsy-oouly
`
`-266-
`
`-266-
`
`
`
`
`
`
`267
`
`Thus, as for the gastrins the search for stable
`
`analogs
`
`of CCK-peptides became important leading us to the
`
`design of two nonapeptides corresponding to the C-terminal
`CCK-sequence 25-33 in which methionine-28 was replaced by
`threonine - in analogy to caerulein,
`the CCK-active peptide
`
`by leucine and
`and methionine-31
`of amphibian skin -
`norleucine, respectively. The syntheses of the two analogs
`were again successfully performed using N*-acyl-tyrosine-0-
`sulfate as intermediate and the desired nonapeptides were
`
`6 as
`obtained according to the scheme outlined in Fig.
`analytically homogeneous products in satisfactory yields (15).
`
`In vivo and in vitro assays revealed the norleucine-analog
`
`as active as CCK~8 and caerulein, whereas replacement of
`
`methionine by leucine was found to lower the hormonal
`
`potency by a factor of 10. These findings surprisingly differ
`
`from those related to the gastrins and may be explained by
`
`the need of a well balanced amphiphilicity which is
`better achieved with the combination threonine-norleucine
`
`according to the hydrophobic/hydrophilic increments of the
`
`various residues using the factors of Tanford (37). The observed
`high stability of thr?®, nte?!-ccK-[25-33] both in solution
`and on storage as lyophilized powder, resulting from the salt
`
`bridge between the O-sulfate-and guanido- function (15,16), its
`
`fast conversion in vitro to the corresponding CCK-octapeptide form
`(38) as well as its equipotency to CCK-8 and caerulein (15),
`
`revealec this analog as perfectly suited for physiological and
`
`clinical research,
`
`The experiences gained in these synthetic studies on CCK-
`
`peptides have opened the way to the preparation of sulfated
`
`gastrin-peptides.
`
`Immunochemical studies
`
`The pronounced tendency of the methionine residue to oxidation
`and of tryptophan to oxidative degradation may lead to
`deleterious chemical modifications of the gastrin molecule
`
`during its labelling with radioactive iodine unter oxidative
`
`conditions as needed for the preparation of tracers.
`
`These structural modifications
`
`of
`
`the C-terminal
`
`-267-
`
`-267-
`
`
`
`
`
`
`
`
`atoputrAyzeu-z/HO0DEaD$06(i
`
`
`
`
`
`
`
`
`
`[€€-Sz]-MOO-[,-eTN“gztun]JOstseyqudsjosueyos
`
`"9*bTa
`
`
`
`f€€-Sz]-MOO-[)NET‘gztuk]pure
`
`
`
`
`
`Ayde2z603eworzyoabueyoxs-uor
`
`(¢
`
`
`
`9dTabotguouoTZeARITIJIaBHEz
`
`
`
`loge—S71CHN-9Ud-dsy-X-dy1-51)-14,L-(HEOS)14L-dsy-81y-H
`
`
`
`
`
`
`
`
`
`lee-zelSHN-eud-(ndO)dsy-H+[f€-0€!HO-X-d1L-2
`
`TOF
`
`
`
`[1€1HO-xX-H+[o€lNSNO-diL-Z
`
`NSNOH/ODd
`
`
`
`
`
`(ac¢e-0€]*HN-oud-(NAO)dsy-X-diL-H+[67-87]HO-41ID~,9A)IUL-Z
`
`pd/tH
`
`NSNOH/O0d
`
`leee—O€]THN-OUd-(,9d0)dsy-X-di]-Z
`
`
`
`(621HO-AID-H+[871RSNO-(jNE)UL-Z
`
`
`
`
`
`lece—87]7HN-2Y4d-(yngO)dsy-xX-dIL-A19-(,24LUL-Z
`
`Pd/?H
`
`(I)Ray=X
`
`(ID9IN=X
`
`
`
`(q¢€-87]HN-ayd-(;nGO)dsy-X-daL-A1D-GNG)IUL-H+[£7]HO-(2/leg€Qs)si41-7
`
`
`
`
`
`
`
`[qg€-S7]CHN-eud-(9dO)dsy-X-d4ay-A12)-(,2g)HICHEOS)IAL-GndO)dsv-31y-H
`
`
`
`
`
`1OH/Pd/CH
`
`
`
`
`
`lege—S7)SHN-A4d-(NGO)dSW-X-diL-AID-(,ngsyL-(2/Teg€Qs141-(ngo)dsy-(2z)31y-Z
`
`NSNOH/90d
`
`
`
`(ag€-27]*HN-2Ud-(;9G0)dsy-X-dip-A19-(ngUL-(2/Teg€QSAL-H+[97-$Z]HO-(nao)dsw-(2Z)1y-z
`
`
`
`
`
`268
`
`
`
`[ege-27]CHN-2Ud-(,nGO)dsy-K-d1p-A19-(,NGYL-(7/TegEQSyAL-Z1OHJ9PONSNOH/OId
`
`Pd/tH
`
`-268-
`
`-268-
`
`
`
`
`
`
`
`269
`
`sequence of the gastrins undoubtedly arise severe questions
`
`at least in respect to the quantitative interpretation of
`
`data derived from conventional radioimmunoassays. Regarding
`
`the problem methionine a comparative analysis of various
`related gastrin analogs revealed the 15-methoxinine-little
`
`gastrin-I - an analog stable to oxidants - as immunoreactive
`
`as the parent methionine peptide towards a series of anti-
`
`gastrin-antisera (20) and thus, suited for the preparation of
`
`tracers. The main difficulty, however, derives from the
`
`tryptophan residue. A systematic investigation of the
`
`oxidative destruction of tryptophan during radioiodination
`
`of tyrosine via the various proposed procedures revealed
`
`that under all conditions employed tryptophan is degraded
`
`to large extents (39 - 40). Formylation of the indole function
`
`was found to prevent the side reaction (40); but even if
`promising in this respect,
`the w”_¢ormyl group is certainly
`not in all cases the most adequate bypass, since its removal
`
`occurs under alcaline conditions where not all peptide
`
`sequences are sufficiently stable.
`
`These shortcomings of the preparation of radioiodinated
`
`tracers,
`
`the molecular heterogeneity of the gastrin-CCK family
`
`and the relatively high immunogenicity of the common C-terminal
`
`sequence as resulting from immunization with conventionally
`
`produced antigens, make the characterization of gastrin-and
`
`cCK-immunoreactivity by the usual
`insufficient.
`
`immunclogical methods
`
`Our approach to improve selectivity and specificity of these
`
`methods is outlined in Fig. 7. It is based on the incorporation
`
`of a reactive anchor-group via a spacer, at a definite position
`
`of the peptide-chain,
`
`to be used for the specific conjugation
`
`of the peptide factor to high molecular weight carriers as
`
`needed for immunization experiments, as well as for the
`
`incorporation of the tracer group. For this purpose the
`maleimido group was selected since it is known to react
`rapidly and rather specifically with thiols.
`In a first set of experiments maleimido-B-alanine-N-hydroxy-
`
`succinimido ester (41) was used to acylate human little
`
`gastrin-I-[2-17]
`
`(Fig. 8). The resulting gastrin derivative
`
`-269-
`
`-269-
`
`
`
`
`
`
`|
`(J.J
`
`apnded
`
`AogYyoue
`
`
`
`apizdedpazeniz9e
`
`270
`
`
`
`puesza0ez}FOSstsayRUASSATROSeETSSebAOFowayos—
`
`“L
`
`“bTa
`
`-270-
`
`
`
`
`
`siojoeyeptjdadjosuahtjue
`
`-270-
`
`
`
`
`
`
`271
`
`Mal-8-Ala-ONSu + H-Gly-Pro-Trp-Leu-(Glu) 5Ala-Tyr-Gly-Trp-Met-Asp~Phe-N H 9
`
`DMF / ACN
`
`Mal-8-Ala-Gly-Pro-Trp-Leu-(Glu) 5Ala-Tyr-Gly-T rp-Met-Asp-Phe-NH 2
`
`peptide content (M.. = 2138.4)
`
`ot
`: 90 %
`
`hpic (u-Bondapak C18):
`
`isocratic elution with acetonitrile/0.1.M sodium phosphate
`
`(pH 5.0), 28:72, flow rate 1,4 ml/min, 280 nm;
`
`tlc: n~BuOH/AcOH/Pyridine/H,0; 60:6:20:24
`
`n-BuOH /AcOH/H,0; 60:20:20
`
`Fig. 8. Synthesis of maleoyl-8-alanyl-human little
`
`gastrin-I-[2-17]
`
`was found to be well suited for a smooth and clean preparation
`of radiolabelled tracers via its reaction with p25 I] -desamino-
`tyrosyl-cysteine,
`[7H] -acetylcysteine or
`[75s] -cysteine;
`the
`related derivatives were found to retain full immunoreactivity,
`e.g. Fig. 9. With this approach the problem related to the
`methionine and tryptophan residue would be resolved.
`However,
`to bypass also
`radiolysis of the gastrin molecule
`and thus, the storage problem as well as waste-problems, an
`alternative fluorescence-immunoassay based on a fluorogenic
`enzyme substrate linked to the hormone molecule (Fig. 10) was
`elaborated. The fluorogenic tracer was used to develop a double-
`antibody enzyme-immunoassay for gastrin which showed to be
`
`sufficiently sensitive to determine normal gastrin levels in
`healthy humans
`(41).
`
`The conventional conjugation procedures used in the preparation
`of antigens employ reagents, e.g. carbodiimides, which undoubtedly
`involve the most reactive functions along the hapten molecule,
`e.g.
`in the case of gastrins the various carboxyl functions in
`the central region of the molecule. This fact may be responsible
`for the observed high immunogenicity of the C-terminal
`
`-271-
`
`-271-
`
`
`
`
`
`
`272
`
`100
`
`90
`
`80
`
`70
`
`60
`
`50
`
`40
`
`BiB,x100
`
`0
`
`10
`
`pmol/I
`
`107
`
`Fig. 9. Comparative immunoreactivities of HG~17 (o—o—o)
`
`and of the conjugate cysteine/Mal-8-Ala-HG- [2-17]
`1255-HG-[1-17]
`using antigastrin-antiserum No. 2127 of Rehfeld
`
`in cross reactions versus
`
`(e—-e—-e)
`
`tetrapeptide and thus, for the difficult production of
`
`gastrin- or CCK-specific antisera (42). On the assumption that
`
`selective linkage of the gastrin at its N-terminus to carriers
`
`with concomitant full exposure of the unmodified hapten
`
`molecule to the antigen recognition may lead to more selective
`
`and specific antisera, maleimido-8-alanyl-human little gastrin-
`
`{2-17] was conjugated to reduced RNase A. The antisera raised
`
`in rabbits with this antigen were used to titrate the immuno-
`
`reactivity of the gastrin sequence. As shown in Fig. 11, a
`
`sharp cooperative transition was observed upon incorporation
`
`of the third glutamic acid residue, while the C-terminal
`
`tetrapeptide is recognized to very low extents. A similar
`
`sigmoidal curve in function of chain-length of the gastrin
`
`was determined for the acid secretion potency (12) as
`
`shown in Fig. 12 and for the onset of the proposed "active"
`
`B-structure as determined by circular dichroism measurements
`
`-272-
`
`-272-
`
`
`
`
`
`
`VOW-2Ud-09-*(£H9)-O9-HN-“(SHO)-SH+=[ZL-7]-9H-eIV-g-1eW
`
`Zzdle6og
`
`
`
`eayzerysqnsutsdAzzowAyo
`
`
`
`[£t-21-9H-09-°(2HD)-N
`
`“OL*btg
`
`
`
`("HD)a¢,9.2~|___HNOD-(°H*9-"HD)EHD6&(%HD)<——_EuD62||NSNOH/D0G
`
`
`
`VOW-2ud-09-*(£HD)-OD-HN-<(fHO)-SNSNOOD
`
`
`WOW-2Ud-09-©(£HD)-09-HN-“(£HD)-S
`se2e0ezqutaq4sebotusebozontyzefostsayquds
`
`
`
`
`|vow-eud-09-*(2H3D)-09-HN-“(2HD)
`an-*(0)-5-8-2(fH)HNme
`
`
`
`
`
`
`
`joueu}sOUON]$14}
`
`
`
`/euudsoydAndy
`
`273
`
`o=Oo
`
`HD-HN-O9
`
`\
`
`S
`
`HNOD-(£H29-°HD)HD-HN-OD
`
`VOW-94d-09
`
`HOOD
`
`-273-
`
`-273-
`
`
`
`
`
`
`100
`
`75
`
`ICeg(HGI-17)/1Cgpx100 25
`
`50
`
`7
`5
`13
`"1
`9
`7
`5
`3
`1
`Pyr-Gly-Pro-Trp-Leu-Glu-Glu-Glu-Glu-Glu-Ala-Tyr-Gly-Trp-Met-Asp-Phe-NH>
`PyMeteet
`H-Gl y—_——__________________—______ Met —— Phe-NH2
`Pyr —__—_—____——__ Nile ——Phe-NH
`
`Pyr
`Nle ——— Phe-NH2
`Pyr-——___—~_—~—- Nle —— Phe-NH2
`Pyr--——_—__—- Nle
`Phe-NH>
`Pyr—————_§—— Nle-——~ Phe-NH2
`
`Fig. 11. Titration of the immunoreactivity of the human
`
`little-gastrin-I sequence
`
`foAktivitat
`
`o@oO
`
`aDoO
`
`40
`
`20
`
`* a v
`
` &
` @
`0
`
`4 °
`
`4% 6 6 17
`23
`12
`0 1
`9
`8
`7
`6
`5
`4
`3
`2
`1
`(tu-Gly -Pro -Trp - Leu-Glu - Glu- Glu - Glu - Glu- Ala - Tyr- Gly - Trp-Nle-Asp- Phe-NH
`
`H-Ley ————____________________Phe-NH
`C6ty$$$Phre-NH
`(6{yj ————____________________Phe-NHp
`iy—______________________Phe-NH)
`Cty ——________________Phe-NH)
`(Gtu—_——____________Phe-NH2
`
`H-Trp —————Phe- NH
`
`Fig. 12. Stimulation of gastric acid secretion in rats by
`gastrin-I-peptides expressed as percentage of the
`response to Nie!°-HG-17
`
`-274-
`
`-274-
`
`
`
`
`
`
`275
`
`of the gastrin peptides in trifluoroethanol
`
`(43 - 44).
`
`This parallel development of the conformational, biological
`
`and immunological properties in function of chain-length
`
`strongly suggests that the immunodeterminant expressed by
`
`this new approach may possibly be conformation-dependent.
`
`This should be equivalent to an enhancement of selectivity
`
`and specificity of the antisera. Further experiments related
`
`to both the gastrins and cholecystokinins are in progress
`
`to confirm these promising findings.
`
`References
`
`1.
`
`Tracy, H.J. and Gregory, R.A.
`
`(1964) Physiological
`
`properties of a series of synthetic peptides
`
`structurally related to gastrin I. Nature, 204, 935-938.
`
`Gregory, H., Hardy, P.M., Jones, P.S., Kenner, G.W.
`
`and Sheppard, R.C.
`
`(1964) The antral hormon gastrin;
`
`structure of gastrin. Nature, 204, 931-933.
`
`Gregory, R.A. and Tracy, H.d.
`
`(1966) Isolation of two
`
`gastrins from human antral mucosa. Nature, 209, 583.
`
`For review see Mutt, V.
`(1983) Chemistry of the
`Gastrointestinal Hormones and Hormone-like Peptides and
`
`a Sketch of Their Physiology and Pharmacology. Vitamins
`
`and Hormones. Vol. 39, 231-427.
`
`Wiinsch, E. and Deimer, K.H.
`
`(1972) Zur Synthese des
`
`{15-Leucin]Human-Gastrin I. II. Mitteilung: Herstellung
`
`der Gesamtsequenz. Hoppe-Seyler's Z. Physiol.Chem. 353,
`1255-1258.
`
`(1972)
`Wiinsch, E., Jaeger, E., Deffner, M. and Scharf, R.
`Zur Synthese des [15-Leucin]Human-Gastrins I. III. Mitteilung:
`
`Zur Reindarstellung des synthetischen Heptadecapeptidamids.
`Hoppe-Seyler's Z. Physiol.Chem. 353, 1716-1720.
`
`Wunsch, E., Wendlberger, G., Hallet, A., Jaeger, E.,
`
`Knof, S., Moroder, L., Scharf, R., Schmidt, I.,
`
`Thamm, P. and Wilschowitz, L.
`
`(1977)
`
`Zur Totalsynthese
`
`des Human-Big~-gastrins I und seines 32-Leucin-Analogons.
`
`Z. Naturforsch. 32c, 495-506.
`
`-275-
`
`-275-
`
`
`
`
`
`
`276
`
`Wunsch, E., Wendlberger, G., Mladenova-Orlinova, L.,
`
`GOhring, W., Jaeger, E., Scharf, R., Gregory, R.A.
`and Dockray, G.J.
`(1981) Totalsynthese des Human-
`
`Big-Gastrins I, Revidierte Primdrstruktur. Hoppe-Seyler's
`Z. Physiol. Chem. 362, 179-185.
`
`(1978)
`Moroder, L., Dress, F., Jaeger, E, and Wiinsch, E.
`Zur Synthese von [11-Leucin]Human-Minigastrin I. Darstellung
`
`der Gesamtsequenz. Hoppe-Seyler's Z. Physiol. Chem. 359,
`
`155-164,
`
`10.
`
`Jaeger, E., Thamm, P., Schmidt, J., Knof, S., Moroder, L.
`
`and Wiinsch, E.
`
`(1978)
`
`Zur Synthese von [11-Leucin]Human-
`
`Minigastrin I. II. Mitteilung: Reindarstellung des
`
`synthetischen Tridekapeptidamids sowie Isolierung und
`Strukturaufkladrung eines Synthese-Nebenprodukts.
`Hoppe-Seyler's Z. Physiol.Chem. 359, 155-164.
`
`11.
`
`Moroder, L., Gdhring, W., Nyfeler, R., Scharf, R.,
`
`Thamm, P. and Wendlberger, G.
`
`(1983)
`
`Zur Synthese von
`
`Human-Little-Gastrin I und dessen Leucin-15, Norleucin-15
`
`und Methoxinin-15-Analoga. Hoppe-Seyler's Z. Physiol.
`
`Chem. 364, 157-171.
`
`12.
`
`Go6hring, W., Moroder, L., Borin, G., Lobbia, A., Bali, J.P.
`
`and Wiinsch, E.
`
`(1984)
`
`Synthese von Gastrinaktiven
`
`Peptiden. Untersuchungen zur Struktur-Wirkungsbeziehungen
`
`des natlrlichen Hormons Human-Little-Gastrin I. Hoppe-
`Seyler's Z. Physiol. Chem. 365, 83-94.
`
`13.
`
`Moroder, L., Wilschowitz, L., Jaeger, E., Knof, S.,
`
`Thamm, P. and Wtinsch, E.
`
`(1979)
`
`Synthese von Tyrosin-0O-
`
`sulfat-haltigen Peptiden. Hoppe-Seyler's Z. Physiol. Chem.
`360, 787-790.
`
`14.
`
`Wiinsch, E., Moroder, L.,