BERND GUTTE
`
`
`aemctete
`and Applications
`
`—
`
`Edited by
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`

`

`
`Peptides
`
`Synthesis, Structures,
`and Applications
`
`Edited by
`
`Bernd Gutte
`
`BiochemischesInstitut der Universitat Ztirich
`CH-8057 Ziirich, Switzerland
`
`aereeypuneyinesnanenenettinGititstkamemageeeneren=©~AtimmiiineniAtsimunnmnymengeeigot
`
`
`
`
`
`
`SoseemeeeneenopmnessraieinettonyempiHoe+:°eawe
`
`Academic Press
`San Diego
`NewYork
`
`Boston
`
`London
`
`Sydney
`
`Tokyo
`
`Toronto
`
`MPI EXHIBIT 1059 PAGE 2
`
`

`

`
`
`—ee,
`nnearnanemnenneetnnsa-—--hemmnememsentinesteeefsternum
`
`
`
`Cover art: Conformation of cyclosporin A observedin the crystal.
`See Chapter 5 by Marshall egal.
`
`This book is printed on acid-free paper.
`
`Copyright © 1995 by ACADEMIC PRESS,INC.
`
`All Rights Reserved.
`Nopart of this publication may be reproduced or transmitted in any form or by any
`means, electronic or mechanical, including photocopy, recording, or any information
`storage andretrieval system, without permission in writing from the publisher.
`
`Academic Press, Inc.
`A Division of Harcourt Brace & Company
`525 B Street, Suite 1900, San Diego, California 92101-4495
`
`United Kingdom Edition published by
`Academic Press Limited
`24-28 Oval Road, London NWI 7DX
`
`utte.
`
`Library of Congress Cataloging-in-Publication Data
`Peptides : synthesis, structures, and applications / edited by Bernd
`P.
`cm.
`Includes bibliographical references and index.
`ISBN 0-12-310920-5 (alk. paper)
`1. Peptides,
`2. Peptides--Synthesis.
`1. Gutte, Bernd,
`QP552.P4P487
`1995
`574.19'2456--de20
`
`3. Peptides--Structure.
`
`95-15417
`
`PRINTEDIN THE UNITED STATE
`S OF AMERICA
`95
`96
`97
`eh
`99
`00 BBO9
`8 7654321
`
`CIP
`
`MPI EXHIBIT 1059 PAGE3
`
`

`

`
`
`Contents
`
`Contributors
`
`Preface
`
`1
`The History of Peptide Chemistry
`Theodor Wieland
`
`Introduction
`I.
`Il. Early Peptide Syntheses
`III. A NewEra in Peptide Chemistry
`IV. Classic Peptide Syntheses
`V. Conclusion
`
`References
`
`2
`
`Amide Formation, Deprotection, and Disulfide Formation
`in Peptide Synthesis
`Yoshiaki Kiso and Haruaki Yajima
`
`I. Amide-Forming Reactions
`I]. Deprotection
`Ill. Disulfide Bond Formation
`
`Abbreviations
`
`References
`
`3
`Solid-Phase Peptide Synthesis
`Bruce Merrifield
`
`I.
`Introduction
`Il. The Solid Support
`
`xi
`
`xili
`
`26
`
`34
`
`34
`
`40
`
`53
`
`61
`
`82
`
`84
`
`94
`
`94
`
`v
`
`MPI EXHIBIT 1059 PAGE 4
`
`

`

`™e
`
`103
`114
`127
`137
`143
`146
`156
`159
`
`17]
`
`172
`
`175
`
`178
`
`183
`
`187
`
`188
`
`191
`
`193
`
`195
`
`209
`
`218
`
`225
`
`235
`
`Cc
`vi
`went
`Handles, Linkers, and Spacers
`Protecting Group Strategy
`The Coupling Reaction
`Cause and Correction of Incomplete Coupling Reactions
`The Cleavage Reaction
`Solid-Phase Synthesis of Peptides and Proteins
`Abbreviations
`
`Hl.
`
`IV.
`
`V.
`
`VI.
`
`VIL.
`
`Vill.
`
`References
`
`4
`a-Helix Formation by Peptides in Water
`J. Martin Scholtz and Robert L. Baldwin
`
`I,
`
`I.
`
`Ill.
`
`IV.
`
`V.
`
`VI.
`
`VII.
`
`Introduction
`Methodsof Studying Helix Formation
`Models for Helix—Coil Transition
`Helix Propensities
`Side-Chain Interactions
`Capping Interactions
`Protein Helices
`
`References
`
`5
`Peptide Conformation: Stability and Dynamics
`Garland R. Marshall, Denise D. Beusen, and Gregory V Nikiforovich
`
`l.
`
`II.
`
`III.
`
`IV.
`
`V.
`
`Introduction
`
`Theoretical Studies
`Experimental Analysis
`Goal-Directed Modifications of Conformational Flexibility
`Examples of Experimental and Computational Techniques
`Applied to the Study of Peptide Conformational Stability
`and Dynamics
`References
`
`MPI EXHIBIT 1059 PAGE 5
`
`

`

`
`
`
`
`Contents viiretts
`
`6
`Structure—Function Studies of Peptide Hormones:
`An Overview
`
`Victor J. Hruby and Dinesh Patel
`
`Introduction
`I.
`II. Mammalian Peptide Hormones
`Ill. Use of Classic Structure—Function Studies:
`General Considerations
`IV. Use of Conformational Constraints: General Considerations
`V. Selected ExamplestoIllustrate Approaches
`VI. Summaryand Conclusions
`References
`
`247
`248
`
`250
`252
`257
`277
`279
`
`7
`Neuropeptides: Peptide and Nonpeptide Analogs
`Andrzej W. Lipkowski and Damel B. Carr
`I. Biosynthesis of Neuropeptides
`II. Neuropeptide Receptors
`Ill. Opioids
`IV. Tachykinins
`V. Angiotensin II
`VI. Oxytocin and Vasopressin
`VII. Cholecystokinin
`VIH. Conclusions
`References
`
`8
`Reversible Inhibitors of Serine Proteinases:
`Naturally Occurring Miniproteins, Semisynthetic Variants,
`Recombinant Homologs, and Synthetic Peptides
`Herbert R. Wenzel and Harald Tschesche
`
`Introduction
`].
`Il. Serine Proteinase—Inhibitor Interactions
`III. Naturally Occurring Proteinase Inhibitors
`
`288
`289
`290
`297
`301
`305
`307
`310
`311
`
`322
`323
`325
`
`MPI EXHIBIT 1059 PAGE6
`
`
`
`is
`
`OELTTTTDIT
`
`aeopree
`“7G
`
`

`

`3
`IV. Semisynthetic Proteinase Inhibitors
`x
`V. Proteinase Inhibitors by Genetic Engineering
`VI. Chemically Synthesized Proteinase Inhibitors and Fragments *
`VII. Concluding Remarks and Outlook
`Ref
`ws
`eferences
`35
`
`9
`Design of Polypeptides
`Bernd Gutte and Stephan Klauser
`I.
`Introduction
`II. Recent Design Activities
`Ill. Polypeptide Design: Quo Vadis?
`References
`
`10
`
`Soluble Chemical Combinatorial Libraries:
`Current Capabilities and Future Possibilities
`Richard A. Houghten
`
`Introduction
`I.
`II. Classes and Applications of Soluble Combinatorial
`Libraries of Peptides
`Ill. Applications of Soluble Peptide Libraries
`IV. Conclusion
`References
`
`11
`Epitope Mapping with Peptides
`Hans RudolfBosshard
`
`Introduction
`I.
`II. The Nature of Epitopes
`Ill. Overview of Noncrystallographic Mapping Procedures
`IV. Cross-Reactivity Studies between Proteins and Peptides
`Using Antiprotein Antibodies
`V. Cross-Reactivity Studies between Proteins and Peptides
`Using Antipeptide Antibodies
`
`363
`366
`389
`399
`
`396
`
`401
`404
`414
`415
`
`419
`420
`425
`
`421
`
`4
`
`MPI EXHIBIT 1059 PAGE 7
`
`we
`
`

`

`
`
`
`
` Contents ix
`
`VI. Concluding Remarks
`References
`
`12
`Synthesis and Applications of BranchedPeptides
`in Immunological Methodsand Vaccines
`james P Tam
`
`Introduction
`I.
`Il. Types of Branched Peptides
`If]. Design and Physical Characteristics of Multiple
`Antigen Peptides
`IV. Multiple Antigen Peptide Immunogenicity
`V. Application of Multiple Antigen Peptides in Vaccines
`VI. Other Applications
`VII. Preparation of Multiple Antigen Peptides
`References
`
`Index
`
`450
`451
`
`456
`456
`
`458
`460
`469
`474
`477
`493
`
`501
`
`MPI EXHIBIT 1059 PAGE8
`
`
`
`

`

`
`
`
`
`
`
`
`Amide Formation, Deprotection, and
`Disulfide Formation in Peptide Synthesis
`
`Yoshiaki Kiso
`Department of Medicinal Chemistry
`Kyoto Pharmaceutical University
`Yamashina-ku, Kyoto 607, Japan
`
`Haruaki Yajima
`Niigata College of Pharmacy
`Niigata 950-21, Japan
`
`
`
`L Amide-Forming Reactions
`A. Acid Chloride and Fluoride Methods
`B. Active Ester Methods
`1. Phenyl Esters
`2. N-HydroxylamineEsters
`3. Bifunctional Active Esters
`C. Unsymmetrical Anhydride Methods
`1. Mixed Anhydride Method
`2. Phosphoric Mixed Anhydride Methods
`D. Carbodiimide Methods
`
`1. DCC-Additive Methods
`2. Fragment Condensation at Glycine or Proline
`Residues
`
`E. Azide Methods
`1. Protected Hydrazide Methods
`2. Azide Method Using Diphenylphosphorylazide
`F. Phosphonium and Uronium Salt Type Coupling
`Reagents
`G. Other Amide-Forming Reactions
`II. Deprotection
`A. Deprotection of a-Amino Protecting Groups
`1. Mild Acids
`
`2. Alkaline Conditions
`B. Final Deprotection
`1.
`tert-Butoxycarbonyl/ Benzyl]
`2. 9-Fluorenylmethoxycarbonyl /tert-Buty|
`3. OtherStrategies
`
`Pepuades:Synthesis, Structures, andApplications
`Copyright © 1995 by Academic Press, Inc. All rights of reproductionin any form reserved.
`
`39
`
`MPI EXHIBIT 1059 PAGE 9
`
`

`

`40
`
`Yoshiaki Kiso and Har
`
`lll. Disulfide Bond Formation
`A. Reduction—Reoxidation of Disulfide Bond-
`Containing Peptides and Proteins
`1. Examination of Oxidation Conditions
`2. Formation of Isomers
`B. Disulfide Formation by Air Oxidation
`1. Disulfide Formation by One-Step Deprotection
`2. Disulfide Formation by Two-Step Deprotection
`C. Disulfide Formation by lodine Oxidation
`D. Stepwise Disulfide Formation
`1. Combination of 4-Methylbenzyl and Acetamido-
`methyl Groups
`2. Combination of Trityl and Acetamidomethyl
`Groups
`3. Combination of 3-Nitro-2-pyridosulfenyl and
`Acetamidomethyl Groups
`E. Disulfide Formation by Silyl Chloride—Sulfoxide
`System
`1. Disulfide Bond Formation Using Sily! Chloride—
`Sulfoxide System
`2. Possible Mechanism
`3. Synthesis of Human Brain Natriuretic Peptide
`(hBNP)
`4. Directed Formation of Double Intrachain
`
`Disulfides
`
`5. Directed Formation of Double Interchain
`
`Disulfides
`6. Regioselective Formation of the Three Disulfide
`Bonds of Human Insulin
`Abbreviations
`
`References
`
`
`I. Amide-Forming Reactions
`The amide-formingreaction, or coupling reaction, is one of the most im-
`portant steps in peptide synthesis. Because harsh chemical conditions can dam-
`age complex peptides, appropriate activation is required to perform the coupling
`under mild conditions. Theoretically,
`two types of reactions are available.
`namely, aminoactivation and carboxylactivation. The former methods, howevel,
`have rarely been applied to practical peptide synthesis because of the drastic fe-
`action conditions and racemization. Carboxylactivation has been used in almost
`all the more recent peptide couplings. The general mechanism ofcarboxyl acti-
`vation is illustrated in Scheme 1, where X represents any kind of activating
`group or atom.
`
`MPI EXHIBIT 1059 PAGE 10
`
`

`

`pe
`
`
`
`2. Amide Formation, Deprotection, and Disulfide Formation 41es
`
`ie
`
`Ri—cx + NHp——R?0— a + HX
`sen
`i
`Scheme 1__Peptide bond formation by carboxylactivation.
`There are two possible synthetic strategies: stepwise elongation and frag-
`ment condensation. A synthetic scheme which employs either or both methods
`is elaborated accordingto the size and nature of the peptide to be synthesized. In
`stepwise elongation, each amino acid bearing an N*-amino protecting group is
`activated and coupled to the carboxyl-terminal amino acid one by one. This
`methodis usually applied to the synthesis ofrelatively small peptides or peptide
`fragments of a larger peptide. Fragment condensation is used to construct a
`larger peptide from two peptide fragments. The method, however, is often ac-
`companied by racemization unless appropriate care is taken. Antonovics and
`Young (1967) studied in detail
`the coupling reactions which proceeded via
`azlactone (oxazolone) formation after carboxyl activation (Scheme 2). From this
`result, it is clear that the carboxyl activation on an aminoacid residue other than
`glycine or proline produces a racemized peptide unless means to inhibit race-
`mization are provided.
`
`R3
`
`a) nucleophilic
`
`RI.
`
`asic
`
`H
`N
`
`°
`
`R?
`
`-
`
`0
`
`fast” Wye |
`sykedhe“ES
`attack
`“te
`b) b
`NH)—R*ot
`rm
`AhMA,Uf “Mhy"on
`
`Scheme:2 Racemizationmechanismduringhagmencondensation.
`
`Stepwise elongation and fragment condensation methods have been devel-
`oped to circumvent the inevitable racemization. The latter methods, including
`the useful azide method, require distinctive practical considerations.
`
`A. Acid Chloride and Fluoride Methods
`
`The acid chloride method wasfirst introduced by E. Fischer (1903). Most
`acid chlorides are prepared by treating amino acids with PCI,, PCl,, SOCL,
`(Pizey, 1974; Matsuda et al., 1985), or (COCI),. Any side-chain hydroxyl,thiol,
`a8 well as amino groups should be protected during coupling. The carboxy]
`
`MPI EXHIBIT 1059 PAGE11
`
`

`

`Yoshiaki Kiso and Haruaki Yajin,
`42
`‘ggem
`
`<=
`
`group, however,is activated so strongly that some side reactionslimitthe appli.
`cation of this method.
`Acid chloride methods have been usedefficiently in coupling of aming
`acids bearing N*-protecting groups such as 9-fluorenylmethoxycarbony| (Fmoc:
`Carpino ef al., 1986) and 2-(triphenylphosphino)ethyloxycarbonyl (Peoc; Kunz
`and Bechtolsheimer, 1982) which are stable under acidic conditionsandcleay.
`able under alkaline conditions. The stability of the protecting groups to acidic
`conditions madeit possible to survive conversion to an acid chloride.
`Aminoacid fluorides used for solution and solid-phase synthesesare ob.
`tained by treatment with cyanuric fluoride (Bertho et al., 1991; Carpinoetal,
`1991) under mild conditions.
`
`B. Active Ester Methods
`
`The characteristics of the active ester method are as follows: (i) Twopath-
`ways of aminolysis are possible in the mixed anhydride method, whereas only
`one is found in the active ester method. (ii) The active ester can be stored.(iii)
`The ester can be coupled with an amine component that has free carboxy!
`groups. (iv) The method is especially useful for the synthesis of glutamine-
`and/or asparagine-containing peptides (Bodanszky and du Vigneaud, 1959). The
`side product generated on carboxy! activation, a nitrile derivative, can be elimi-
`nated by purification.
`Active esters are usually prepared by the dicyclohexylcarbodiimide (DCC)
`or mixed anhydride method from N-protected amino acids and alcohol. Trans-
`esterification methods using trifluoroacetate (Sakakibara and Inukai, 1965; Rze-
`szotarska and Vlasov, 1967) and trichloroacetate (Fujino and Hatanaka, 1968)
`have also been reported.
`
`I. Phenyl Esters
`Generally, phenols which have electron-withdrawing substituent in the
`ortho or para position can be used as active esters (Wieland and Jaenicke, 1956;
`Wieland et al., 1962). In practice, however, p-nitrophenol (Bodanszky, 1955;
`Bodanszkyet al., 1957), 2,4,5-trichlorophenol (Pless and Boissonnas, 1963a,b,c;
`Guttmannet al., 1965; Strumer etal., 1965; Huguenin, 1964), pentachlorophe-
`nol
`(Kovacs and Ceprini, 1965; Kovacs et al., 1967a; Kupryszewski and
`Formela, 1961), and pentafluorophenol (Kovacs et al., 1967b) are widely used
`becauseoftheir easy availability.
`After the coupling reaction with a phenol-type active ester, the liberated
`acidic phenol must be removed by an appropriate method such as washing
`with aqueous sodium carbonate or recrystallization. If the desired peptide
`productis an oil, the phenol cannot be removed completely. Water-soluble ac-
`tive esters, such as 4-hydroxyphenyldimethylsulfonium methylsulfate (Kougé
`et al., 1986; 1987), overcomethese difficulties not only by allowing easy fe
`moval of the by-products, but also by undergoing the reaction in aqueous sol-
`
`MPI EXHIBIT 1059 PAGE 12
`
`

`

`
`
`2, Amide Formation, Deprotection, and Disulfide Formation 43rrr
`
`vents. It is noteworthy that an active ester acylates the phenolic hydroxyl
`group on tyrosine in the presence of a base (Agarwaler al., 1969; Ramachan-
`dran and Li, 1963).
`
`2. N-HydroxylamineEsters
`A series of N-hydroxylamine-type active esters represented by N-hydroxy-
`succinimide (HOSu) ester (Anderson et al., 1963; 1964) have been described.
`These types of active esters are regarded as bifunctional active esters which
`rarely induce racemization of the a-carbon atom. In contrast to the above phenol
`esters, these esters eliminate N-hydroxylamine after coupling, which is easily
`soluble in water and thus removable by simple washing.
`HOSu esters are prepared by coupling N-protected amino acids and
`HOSu by DCC. An undesired side reaction was reported to take place during
`the preparation (Gross and Bilk, 1968; Weygandet al., 1968): the binding of
`DCC and HOSuproduces B-alanine via Lossen rearrangement (Scheme3). N-
`Hydroxy-5-norbornene-2,3-dicarboximide (HONB,1 in Fig. 1) introduced by
`Fujino et al. (1974), rarely undergoes suchside reactions because ofits rigid
`structure.
`
`d bcc
`
`N-OH
`°
`
`okdso
`bon LO
`(0
`(NI
`NQ “Aee
`
`H
`
`AT
`
`oo
`
`n:—U-cH,-CH,-0-0~Su
`
`|
`
`oOll
`
`=C=N-CH,-CH,-C-O—Su
`
`ll
`H
`|
`R-NH2
`il
`Su~O-C—8-Ala-NH-R—_————_SuO—C—N-CH,.-CH,-C-O—Su
`
`Scheme3_Side reaction on reacting HOSu with DCC.
`
`B-Ala
`
`| HOSu
`
`N-Hydroxybenzotriazole (HOBt, 2 in Fig. 1; K6nig and Geiger, 1970a,b)
`esters are widely used in peptide synthesis and are prepared using DCC. The
`racemization-suppressing effect and coupling accelerating effect of HOBtare
`excellent. Many groups have reported the reaction mechanism involving HOBt
`ester (Horiki, 1977; Konig and Geiger, 1973). The effects of HOBt as an addi-
`tive are described in a later section (D.1.). 3-Hydroxy-4-oxo-3,4-dihydro-1 ,2,3-
`benzotriazine (HOOBt, 3 in Fig. 1; Konig and Geiger, 1970c) is a reagent supe-
`nor to HOBt with regard to racemization. However, HOOBt esters have the
`disadvantage of undergoing Lossen rearrangement.
`
`MPI EXHIBIT 1059 PAGE 13
`
`

`

`44
`
`HO
`N
`N
`\
`
`Q
`
`HO—N
`
`0
`
`(1)
`
`~
`
`Yoshiaki Kiso and Haruakiy,.ij
`j
`
`HO
`
`N
`Nay
`
`(2)
`
`(3)
`
`Figure 1 Structures of HONB(1), HOBt(2), and HOObt(3).
`
`3. Bifunctional Active Esters
`Various bifunctional active esters such as hydroxyquinoline (HOQ,4 in F;
`2; Jakubke, 1965; Jakubke and Voigt, 1966a,b), catechol (5 in Fig. 2, Lloyd ang
`Young, 1968, 1971), and hydroxypyridine (6 in Fig. 2; Taschner et al., 1965ap)
`have been developed as racemization-free reagents for peptide fragment coupling
`They can, of course, be applied to coupling by stepwise elongation and are readily
`removable after the reaction, taking advantage oftheir bifunctionality.
`
`C. Unsymmetrical Anhydride Methods
`1, Mixed Anhydride Method
`The mixed anhydride method of peptide bond formation involves aminoly.
`sis of an anhydride consisting of an N-protected amino acid and another acid
`(Albertson, 1962; Meienhofer, 1979). The carboxylic—carboxylic mixed anhy-
`dride method and carbonic-carboxylic mixed anhydride methods were devel-
`oped. Isobutychloroformate (Vaughan and Osato, 1951) is commonlyused for
`the preparation of carbonic—carboxylic mixed anhydrides (Scheme4).
`
`Oo
`.
`NHR?
`4
`RO-COC
`R'—COO"EN’ —e RI ou R'--CO-NH—R? + ROH + CO,
`0
`tees.
`R—o-—c ™—R—0-CO-NH—A? Soyproduc)
`
`Xo
`
`R = isodulyl
`Scheme 4 Mixed anhydride method using isobutylchloroformate.
`
`N |
`
`HO
`
`HO
`
`HO
`
`N
`
`OH
`
`Figure 2
`
`(6)
`(5)
`(4)
`Structures of HOQ (4), catechol(5), and hydroxypyridine (6).
`
`MPI EXHIBIT 1059 PAGE 14
`
`‘
`
`

`

`
`
`a
`
`
`2, Amide Formation, Deprotection, andDisulfide Formation 452,Amide
`
`
`
`Formation,Deprotection,andDisulfideFormation49
`
`the amide-forming reaction is complete
`In mixed anhydride methods,
`within several minutes in an ice bath. The side-chain functional groups in the
`carboxyl component should be protected. The carboxyl groups in the amine
`component could be either in ester form orprotected tentatively with a tertiary
`amine. In the latter case, the reaction can be performed in water-containing sol-
`vents, but the product should be carefully separated becauseofits acidity.
`Asa side reaction in mixed anhydride methods, urethane formation occurs
`in the amine component (Albertson, 1962; Wieland et al., 1962; Wieland and
`Stimming, 1953), but it can be neglected under usual experimental conditions.
`Addition of HOBtto the reaction mixture before the amine componentis added
`reduces urethane formation (Prasad et al., 1985). Also, when Z-Gly (where Z
`represents the benzyloxycarbonyl group) is used as the carboxyl component,at-
`tention must be paid to the spontaneous formation of acylimide (Kopple and
`Renik, 1958; Schellenberg and Ullrich, 1959).
`Mixed anhydridesare also obtained by the reaction of amino acids and di-
`hydroquinoline derivatives. On reaction with a carboxyl component, 1-ethoxy-
`carbony]-2-ethoxy-1,2-dihydroquinoline (EEDQ; Belleau and Malek, 1968) or
`1-isobutyloxycarbonyl-2-isobutyloxy-1,2-dihydroquinoline (IIDQ; Kiso and Ya-
`jima, 1972; Kisoet al., 1973) is converted via a six-membered ring intermediate
`to quinoline and a carbonic—carboxylic mixed anhydride which immediately re-
`acts with an amine componentto give a peptide in good yield (Scheme 5). The
`by-products are carbon dioxide and quinoline as long as undesired urethane for-
`mation is negligible.
`
`R'-COOH
`
`R'-CO-NH-R?
`
`COOR
`oD OSo-F _NHyR? CO, ROH
`orar a
`n2
`
`=ethyl or isobutyl
`
`Scheme 5 Mixed anhydride method using EEDQ or IIDQ.
`
`2. Phosphoric Mixed Anhydride Methods
`The mixed anhydrides of amino acid and phosphoric acid are also useful
`for formation of the peptide bond. Manykinds of phosphoric reagents, for exam-
`ple, (PhO),P(O)CI (Albertson, 1962; Meienhofer, 1979) and Me,P(S)CI (Uekier
`al., 1979; 1988; Ueki and Inazu, 1982), have been developed. In this cate-
`gory, 3,3’-(chlorophosphoryl)bis(1,3-oxazolidin-2-one) (BOP-Cl, 7 in Fig. 3;
`Mesegueref al., 1980; Cabre and Palomo, 1984; Omodei-Sale et al., 1984) is
`particularly useful for coupling of imino acids (Van Der Auwera and Anteunis,
`1987; Tung et al., 1986) such as N-methyl amino acids (Tung and Rich, 1985)
`because of the strong reactivity and selectivity toward the amine component.
`Norborn-5-ene-2,3-dicarboximido diphenyl phosphate (NDPP,8 in Fig. 3)
`is a reagent for the so-called active ester-type mixed anhydride method (Kiso et
`
`MPI EXHIBIT 1059 PAGE 15
`
`

`

`
`
`46
`
`Yoshiaki Kiso and Haruak;Yajma
`
`
`
`(7)
`
`(8)
`
`Figure 3 Structures of BOP-CI (7) and NDPP(8).
`
`al., 1980a). In the reaction of NDPP and an amino acid, the HONBesterjs
`formed via a carboxylic-phosphoric mixed anhydride intermediate (Fig.3),
`
`D. Carbodiimide Methods
`
`N,N’-Dicyclohexylcarbodiimide (DCC) has been widely used (Rich and
`Singh, 1979; Williams and Ibrahim, 1981) since Sheehan and Hess (1955)first
`applied it to peptide synthesis. The reaction mechanism involving DCCisillus-
`trated in Scheme 6 (Schiissler and Zahn, 1962; Detar et al., 1966). Briefly, an
`activated intermediate, O-acylisourea, which is produced by reaction of a pro-
`tected amino acid and DCC, undergoes aminolysis to form a peptide anddicy-
`clohexylurea (DCU). The reaction of an N-protected amino acid and DCCwith-
`out an amine component gives another activated species, a symmetrical
`anhydride, which also reacts with amine components. The DCU produced in
`this reaction is insoluble in almost all commonorganic solvents andis thuseas-
`ily removed from the reaction mixture by filtration.
`
`R'-CO-0
`R-—N=C—NH-R
`,” (acylisourea) \
`
`————> R’-C0-NH-R?
`NH>-R2
`
`R'.COOH
`
`
`R—-N==C==N—R
`(DCC)
`
`
`
`
`
`
`,
`
`R'-COOH
`
`
`
`1R
`
`ICO.
`7°
`R'-CO
`
`R'CcO
`R-NH-CO—N-R
`
`(acylurea)
`
`
`
`recycled
`
`Scheme 6_Reaction mechanism of DCC.
`
`The noticeable side reaction is the formation of an acylurea by intramolec-
`ular rearrangement of the acylisourea. The extent of the side reaction is ef
`hanced when the proton concentration of N-protected aminoacid is decreased
`by excess triethylamine (which is usually used for neutralization of chloride ions
`boundto the amine component). The side reaction is profoundly influenced by
`
`MPI EXHIBIT 1059 PAGE 16
`
`

`

`
`
`2, Amide Formation, Deprotection, and Disulfide Formation 47ar
`
`the choice of solvent as well as the base (Sheehan et al., 1956). In the fragment
`condensation by DCC,this side reaction has to be considered seriously, but in
`the coupling of N-protected aminoacids,it can be ignored because of its easy
`removalby recrystallization. Dehydration of the side-chain amidegroup ofglut-
`amine or asparagine also occurs during DCCactivation.
`To simplify the removal of DCU and unreacted carbodiimide, water-sol-
`uble carbodiimides such as 1-ethyl-3-(3’-dimethylaminopropyl)carbodiimide
`were developed; in such systems,both the reagentand the resulting urea deriv-
`ative are soluble and easily removable by washing (Sheehan ef al., 1961).
`Diisopropylcarbodiimide (DIPCDI)is a useful reagent in solid-phase peptide
`synthesis (SPPS) because the diimide and the corresponding urea are soluble
`in organic solvents (Sheehan and Hess, 1955; Sheehan and Yang, 1958).
`
`1, DCC-Additive Methods
`Wiinsch’s group found that fragment condensation using DCC and HOSu
`under strictly controlled conditions preserves the chiral purity almost com-
`pletely, with less than 1% racemization being reported (Weygand et al., 1966;
`Wiinsch and Drees, 1966). The method wasapplied to the synthesis of glucagon
`to obtain an active peptide (Wiinsch, 1967; Wiinsch and Wendlberger, 1968;
`Wiinsch et al., 1968). Strictly speaking, however, the synthesis was carried out
`with suppression of racemization, and an accurate racemization test is required
`to judge whether the product is absolutely optically pure.
`The disadvantage of the DCC-HOSu method is the side reaction, Lossen re-
`arrangement, which gives a B-alanine derivative by reaction with amine compo-
`nents (Gross and Bilk, 1968; Weygand et al., 1968). To remedy this, Kénig and
`Geiger (1970a,b,c) examined various analogous reagents and reported DCC-HOBt
`as the most suitable. This reagent suppresses not only racemization but also acy-
`lurea formation in fragment condensation. In these syntheses, use of water-soluble
`carbodiimide instead of DCC was recommended (Kimura et al., 1981, 1988), as the
`urea or acylureas formed are more easily removed from the reaction products.
`These methodsarealso useful for the coupling of amino acids with respect
`to increases in yield and suppression of side reactions. The suppression of
`racemization by this methodology has been reported by many researchers using
`various additives (Yajimaet al., 1973a,b; Fujino, 1976). However, the racemiza-
`tion test is carried out in a model system, and the results do not always agree
`with practical synthesis.
`
`2. Fragment Condensationat Glycine or Proline Residues
`Activation of glycine residues does not cause racemization because
`glycine has no chiral center, nor does activation of proline residues because pro-
`line has no imino hydrogen which would participate in azlactone formation.
`Therefore, whatever kind of coupling method is used, fragment condensation by
`activating the carboxy! group ofglycine or proline neverfails to preserve chiral
`
`MPI EXHIBIT 1059 PAGE 17
`
`
`
`

`

`48
`
`Yoshiaki Kiso and HaruakiYajin,
`
`purity of the peptide, and these tactics are commonly usedin synthetic Schemes
`for larger peptides. When DCC is used as a coupling reagent in fragment con.
`densation, however, acylurea formation becomes a problem. The removal ofacy.
`lurea is fairly difficult. The addition of racemization-suppressing reagents Such
`as N-hydroxysuccinimide also suppresses acylurea formation.
`More recently, coupling reagents such as benzotriazole-1-yloxytris(dimetp.
`ylamino)phosphonium hexafluorophosphate (BOP) have been employedinsteag
`of DCC (Castro et al., 1975, 1977). The strong reactivity of the coupling reagent,
`is advantageousfor fragment condensation, since the reaction rate is slowerthay
`that of amino acid coupling. In addition, 4-(dimethylamino)pyridinecan be useq
`as a catalyst because of the lack of the problem of racemization at Gly or Pro
`residues.
`
`E. Azide Methods
`
`is currently
`The azide method, first developed by Curtius (1902, 1904),
`used not only for the stepwise coupling of troublesome aminoacids suchas ser-
`ine orhistidine, but also for fragment condensation, where it showsreal merit. It
`is the most suitable method for fragment condensation because little racemization
`occurs, and side-chain protection can be kept at a minimum (only the e-amino
`group of lysine and the thiol group of cysteine have to be protected) because
`azideis less reactive with side-chain functional groups such as hydroxyl groups.
`First, the hydrazide is prepared by treating an N-protected aminoacid or
`peptide ester with excess hydrazine hydrate. The hydrazide is convertedto azide
`by tert-butyl or isoamyl nitrite (Rudinger’s variation; Honzl and Rudinger,
`1961), and the azide reacts with the amine component withoutextraction with
`organic solvent. Because the azide is unstable at room temperature, the reaction
`with the amine componentis carried out below 0°C for 40 hr after neutralization
`of hydrogen azide with a base such as triethylamine or N-methylmorpholine.
`The resistance to racemization in this method is explained by theelectron
`distribution of the N, group, which reducesthe tendency of the R-CO—NHgroup
`to participate in azlactone formation (Scheme 7). However, using an isotope dilu-
`tion assay, Kempetal. (1970) detected racemization evenin the azide method.
`
`
`
`® ©
`RONO
`R'CO-NHCHA?CO-NHNH, ———» RICO-NHCHR2CO-N==N==N
`
`0
`R?
`oO
`Re
`NH,-R?
`od
`le — Hi
`—», et!
`N—CH—C—N
`@N—CH—C—N
`1_f
`\.@
`4
`\
`R So
`© Pa
`Ri—C
`N
`\
`/
`0 R°—NH, N
`PR?—NH—NH
`
`——> R'CO-NHCHR?CO—NHR?
`
`Scheme 7
`
`Suppression of racemizationof azide.
`
`MPI EXHIBIT 1059 PAGE 18
`
`

`

`
`
`2, Amide Formation, Deprotection, and Disulfide Formation 49AALreneetnrenrryemnnetpyrerrerereveraasernannaenepayvtt
`
`
`
`Theside reactionsthat generally occurin the azide procedureare as follows:
`(i) Curtius rearrangementat higher temperature, producing a urea derivative or an
`amine (Hofmann etal., 1960) (Scheme8), and (ii) hydrolysis of azide, giving an
`amide and nitric oxide (Honzl and Rudinger, 1961). In addition, side reactions
`which involve particular amino acids have also been reported (Gregory etal.,
`1968). In conclusion, the azide procedure is an excellent coupling method; how-
`ever, some side reactions have been reported, and the procedure is not simple.
`
`$-
`oO
`heat
`5/
`RG st —- R—N=C=0
`\oyen=n 7 Ne
`
`R—NH--CO—NH—R’
`WA _
`NHo—R
`\we
`
`R—NH—COOH ~—» R—NH,
`Scheme 8—Curtius rearrangementof azide.
`
`1, Protected Hydrazide Methods
`Peptide esters containing Arg(NO,), Asp(O-t-Bu), Asp(OBzl), or Glu(OBzl)
`(where t-Bu and Bzl represent tert-butyl and benzyl, respectively) cannot be con-
`verted to the corresponding hydrazide by hydrazinolysis. In such cases, a protected
`hydrazide is introduced at the C terminus of a peptide or carboxyl group of an
`amino acid. After elongation ofthe peptide chain, the substituent on the hydrazide
`is removed, and the resulting hydrazide is converted to an azide.
`Various protected hydrazides have beenreported. Naturally, the substituent
`chosen has to have a selectivity of removal different from that of the a-
`amino protecting group. The usual protecting groups for hydrazine are Z (Hof-
`mann ef al., 1950, 1952), tert-butoxycarbonyl (Boc; Boissonnas et al., 1960;
`Schwyzer et al., 1960), 2,2,2-trichloroethyloxycarbony] (Troc; Yajima and Kiso,
`1971; Watanabe et a/., 1974), and 2,2,2-trichloro-tert-butoxycarbonyl (Tcboc;
`Shimokura er al., 1985).
`
`2, Azide Method Using Diphenylphosphorylazide
`Azides have long been obtained exclusively via hydrazides since their
`first preparation by Curtius. Shioiri et al. (1972), however, devised an inter-
`esting alternative to the hydrazide-mediated azide method, and reported that
`
`of
`CHs—O,§
`DoNg
`TN¢\-0 C2Hs—-O
`
`
`0
`
`(9)
`
`(10)
`
`Figure 4
`
`Structures of DPPA (9) and DEPC (10).
`
`MPI EXHIBIT 1059 PAGE 19
`
`

`

`Yoshiaki Kiso and Haruakiyt aiiby
`Cigi?
`PN
`N’ O-N’
`Veo©
`
`N
`
`PF,
`(12)
`
`)
`
`—_
`
`(en
`N=C
`y
`7
`O-N"'N
`Prem
`
`(15)
`
`50
`
`Ny
`
`“N
`
`14
`
`s NeN
`oN
`:
`“N ‘O-N’
`
`PFO
`(11)
`
`\
`
`ar C)
`—
`‘ed * @N
`YO-N'
`IN cor =
`[
`‘O-N
`no O
`J
`
`(13)
`
`(14)
`
`b
`
`HO-N*
`
`°N
`
`Ny
`\vy
`
`No
`\8 2
`UN,
`N-C
`PFO vs
`
`‘O-N7
`
`°N
`
`[
`
`(16)
`
`(17)
`
`Figure 5 (a) Structures of BOP (11), PyBop (12), HBTU (13), TOPPipU (14), and BOI(15),
`(b) Structures of HOAt (16) and HATU (17).
`
`diphenylphosphorylazide (DPPA, 9 in Fig. 4) directly converts a free carboxy!
`group to an azide (Shioiri and Yamada, 1974). The coupling reaction is consid-
`ered to proceed via an azide rather than an acyloxyphosphoric mixed anhydrie,
`whichisalsoa likely intermediate. A similar reagent, diethylphosphorocyanidate
`(DEPC, 10 in Fig. 4) was also reported (Yamadaet al., 1973; Shioiri et al.
`1976a,b). The advantagesof the reaction using DPPA or DEPCare noformation
`of insoluble by-products and less racemization in fragment condensation.
`
`F. Phosphonium and Uronium Salt Type Coupling Reagents
`Very high performance of coupling reagents has been foundto be neces
`sary, especially in the field of SPPS. Benzotriazole-1-yloyxtris(dimethylamin0)
`phosphonium hexafluorophosphate (BOP,11 in Fig. 5a), reported by Castro ¢
`al. (1975, 1977), is an excellent coupling reagent offering high reactivity #
`easy handling. The reaction between BOP and an N-protected amino acid
`proceeds via an acyloxyphosphonium as an intermediate to producethe corres
`ponding benzotriazole ester and easily removable by-products such as her
`amethylphosphoric triamide (HMPA)and salts (Scheme 9a). Underbasic condi:
`tions, no side reaction accompanies the main reaction, in contrastto the ust °
`DCC,andthe reaction rate is generally very fast. For example, in the cas¢ °
`
`MPI EXHIBIT 1059 PAGE 20
`
`

`

`sone
`\ ow
`
`
`ee tee
`TR eee7
`
`\1
`
`
`
`2, Amide Formation, Deprotection, and Disulfide Formation 512,AmeeaecisultideFormationOT
`
`
`
`coupling of Boc-Ie with Val-OBzl, the reaction is completed within a few min-
`utes. BOPis used in both solution and solid-phase syntheses.
`However, HMPA,the starting material in the synthesis and a by-product in
`the reaction of BOP, has been reported to have respiratory toxicity. It is therefore
`recommended to replace BOP by benzotriazole- | -yloxytrispyrrolidinophospho-
`nium hexafluorophosphate (PyBOP, 12 in Fig. 5a), which has similar reactivity to
`BOP and forms no carcinogenic by-products (Martinez er al., 1988: Coste et al.,
`1990a).
`2-(1H-Benzotriazole-!-yl)-oxy-1,1 ,3,3-tetramethyluronium hexafluorophos-
`phate (HBTU, 13 in Fig. 5a; Knorr et al., 1989), 2-[2-oxo-1(2H)-pyridyl)-1,1,3,3-
`bispentamethyleneuronium tetrafluoroborate (TOPPipU,14 in Fig. 5a; Henklein et
`al., 1991), and 2-(benzotriazole-1-yl)-oxy-1 ,3-dimethylimidazolidinium hexafluo-
`rophosphate (BOI, 15 in Fig. 5a; Kiso et al., 1992b) have similar reactivities to
`BOP. Phosphoniumis substituted by uronium in these reag