`
`US005504190A
`
`United States Patent
`Doughten et al.
`
`[19]
`
`[11] Patent Number:
`
`5,504,190
`
`[ 45] Date of Patent:
`
`Apr. 2, 1996
`
`[54] EQUIMOLAR MULTIPLE OLIGOMER
`MIXTURES, ESPECIALLY OLIGOPEPTIDE
`MIXTURES
`
`[75]
`
`Inventors: R ichard A. Houghten, Solana Beach;
`Julio H. Cuervo, La Jolla; Clemencia
`P inilla; Jon R. Appel, Jr., both of
`Cardiff; Silvie Blondelle, La Jolla, all
`of Calif.
`
`[73] Assignee: Torrey Pines Institute for M olecular
`Studies, San Diego, Calif.
`
`[21] Appl. No.: 253,854
`
`[22] Filed:
`
`Jun.3,1994
`
`Related U.S. Application Data
`
`which is a continuation-in-part
`
`
`which is a continuation-in-part of Ser. No.
`
`[51]
`
`Pinilla et al., Vaccines 92, Cold Spring Harbor Laboratory
`Press, pp. 25-28 (1992).
`Houghten et al., BioTechniques, 13(3):412-421 (1992).
`Appel, et al., Immunomethods 1, Academic Press, Inc.,
`17-23 (1992).
`Pinilla et al., BioTechniques, 13:6, 901-905 (1992).
`Geysen et al., Proc. Natl. Acad. Sci. USA, 81:3998-4002
`(Jul. 1984).
`Geysen et al., Proc. Natl. Acad. Sci. USA, 82:178-182 (Jan.,
`1985).
`Geysen et al., ''The Delineation of Peptides Able to Mimic
`Assembled Epitopes", in 1986 Synthetic Peptides as Anti
`gens, (CIBA Foundation Symposium 119), pp. 130-149
`(1986)-[1986a].
`Geysen et al., Molecular Immunology, 23(7):709-715
`(1986)-[1986b].
`Geysen et al., J. Immunol. Meth., 102:259-274 (1987).
`Mason et al., "Diversity of the Antibody Response", in
`Vaccines 86, pp. 97-103 (1986).
`[60] Division of Ser. No. 797,551, Nov. 19, 1991, abandoned,
`of Ser. No. 701,658, May 16,
`Merrifield,J.Amer. Chern. Soc., 85(14); 2149-2154 (Jul. 20,
`1991, abandoned,
`1963).
`617,023, Nov. 21, 1990, abandoned.
`Rodda et al., Molecular Immunology, 23(6):603-610
`(1986).
`Int. Cl.6 ............................. A61K 38/04; C07K 5/00;
`Schoofs et al., J. Immunol., 140(2):611-616 (Jan. 15, 1988).
`C07K 7 /00; C07K 16/00
`Furka et al., (1988, 14th International Congress of Biochem
`[52] U.S. Cl ........................... 530/329; 530/328; 530/327;
`istry, vol. 5, Abstract FR:013).
`530/326; 530/325; 530/324
`Houghten, Proc. Natl. Acad. Sci., 82:5131-5135 (1985).
`[58] Field of Search . ........ ............................ 530/329, 328,
`Houghten, et al., Biotechniques, 4(6):522-528 (1986).
`530/327, 326, 325, 324; 514/17, 16, 15,
`Devlin et al., Science, 249:404-405 (1990).
`14, 13, 12
`Scott et al., Science, 249:386-390 (1990).
`Fodor et al., Science, 251:767-773 (1991).
`Houghten et al., Vaccines 1986, pp. 21-25 (1987).
`Houghten et al., Nature, 354:84-86 (Nov. 7, 1991).
`Lem et al., Letters to Nature, 354:82 (Nov. 7, 1991).
`Furka et al, Int. J. Peptide Protein Res., 37:487-493 (1991).
`Van derZee et al., Eur. J. Immunol., 19:43-47 (1989).
`Merrifield, Angew. Chern. Int. Ed. Engl. , 24:799 (Oct. 1985).
`
`[56]
`
`References Cited
`
`U.S. PATENT DOCUMENTS
`12/1970 Johnson et al . ......................... 1561148
`3,544,404
`10/1980 Mornany ................................. 4241177
`4,226,857
`8/1981 Sarantakis .......................... 2601112.55
`4,282,143
`12/1981 Hughes et al . .............................. 260/8
`4,304,692
`11/1984 Urdea et al . ............. ............ 525/54.23
`4,483,964
`3/1985 Miller et al. ......................... 525/54.11
`4,507,433
`12/1986 Houghten .................................. 428/35
`4,631,211
`11/1987 Geysen ...................................... 424/88
`4,708,871
`5/1989 Geysen .................................... 436/501
`4,831,211
`4/1991 Rutter et al . ............................ 530/334
`5,010,175
`1/1993 Huebner et al .
`........................ 530/334
`5,182,366
`3/1993 Geysen .................................... 436/518
`5,194,392
`FOREIGN PATENT DOCUMENTS
`8403564
`9/1984 WIPO .
`9/1984 WIPO .
`8403506
`2/1986 WIPO .
`8600991
`6/1992 WIPO . .... ....................... A61K 37/02
`9209300
`OTHER PUBLICATIONS
`
`Abstract No. 288, Xth International Symposium on Medici
`nal Chemistry, Budapest, Hungary, Aug. 15-18, 1988, p.
`168.
`IN: "Innovation and Perspectives in Solid Phase Synthesis:
`Peptides, Polypeptides and Oligognucleotides", R. Epton,
`ed., Intercept Limited, Andover, pp. 237-239 (1992).
`IN: "Peptides: Chemistry and Biology, Proceedings of the
`12th American Peptide Symposium", J. A. Smith & J. E.
`Rivier, eds., ESCOM, Leiden, pp. 560--561 (1992).
`
`Primary Examiner-Jill Warden
`Assistant Examiner-A. M. Davenport
`Attorney, Agent, or Firm-Welsh & Katz
`ABSTRACT
`[57]
`
`A process for the synthesis of a complex mixture pool of
`solid support-coupled monomeric repeating unit compounds
`such as amino acid derivatives is disclosed in which the
`mixture pool contains an equimolar representation of reacted
`monomeric repeating unit compounds coupled. Also dis
`closed is a process for the stepwise synthesis of a complex
`mixture of coupled or free, unsupported oligomers such as
`oligopeptides. A set of self-solubilizing, unsupported mixed
`oligopeptides having one or more predetermined amino acid
`residues at one or more of the same, predetermined positions
`in the oligopeptide chain in which the set contains equimolar
`amounts of a plurality of different amino acid residues,
`preferably at least six different residues, at one or more of
`the same predetermined positions of the oligopeptide chain
`is also disclosed, as are methods of making and using the
`same.
`
`5 Claims, 11 Drawing Sheets
`
`
`
`Page 1 of 62
`
`ILMN EXHIBIT 1021
`
`
`
`u.s. Patent
`
`Apr. 2, 1996
`
`Sheet 1 of 11
`
`5,504,190
`
`•
`,...
`-
`LL
`
`•
`
`f�
`��
`<( IR2 lll(gJ] �
`�
`
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`. .
`.
`•
`.
`.
`•
`.
`.
`. .
`IIRJ�
`.
`g�
`-� lllg) �
`-� llg)J
`g� II [®IJ r
`-� lltMJ �
`m� lllgj) �
`[g� lltg] �
`
`
`
`Page 2 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 2 of 11
`
`5,504,190
`
`0
`<D
`
`..,.._______...
`
`•
`•
`•
`•
`
`•
`•
`•
`•
`.
`.
`
`r�
`II gJ� 2 •
`.
`. .
`. .
`��,.�1} � r='
`II•J ��
`- 3i8
`11.1] ��
`llrmJJ� �
`II��
`lll-J: �
`
`m
`
`,...
`
`
`
`Page 3 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 3 of 11
`
`5,504,190
`
`a
`
`0
`
`z
`
`C\1
`z � -
`
`•
`
`u.
`
`D..
`
`>- 0..
`
`en
`
`<
`
`8
`
`0
`10
`%CONTROL
`
`0
`
`
`
`Page 4 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 4 of 11
`
`5,504,190
`
`120------......
`100
`80
`60
`40
`20
`o�ua-..a...II..&.&.J ........ L&&.A ........ .....,
`
`p
`
`FIG. 3A
`
`120--------.
`100
`80
`60
`40
`20
`0'-&&.I ........................ L&.&.II ...... ......,
`
`y
`
`120..--------.,
`100
`80
`60
`40
`20
`0 �l..&.&&.a.a..&o�....-...UL&&.-....... ......
`
`p
`
`FIG. 38
`
`FIG. 3C
`
`
`
`Page 5 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 5 of 11
`
`5,504,190
`
`120 r-----------,
`100
`80
`60
`40
`N
`20
`o -"LILlO_•-·----- ·--- LA-lLa
`
`FIG. 30
`
`120r---------,
`100
`80
`60
`40
`L
`20
`0 ._._._.a •• I • • • •
`
`, • _.__
`
`FIG. 3E
`
`120t-------__,
`100
`80
`60
`40
`I
`II
`s
`20
`0 \-&.A..I .................... L&.LII ........ .-.........JL..J
`
`FIG. 3F
`
`
`
`Page 6 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 6 of 11
`
`5,504,190
`
`Ac-AOXXXX-NH2
`
`2.0
`
`1.0
`
`2.0
`
`1.0
`
`0
`
`2.0
`
`1.0
`
`A DE F GH I K LM NPO RS TV Y FIG. 4A
`
`r....
`
`Ac-DOXXXX-NH2
`
`A DE F GH I K LM NPQ RS TV Y
`
`FIG. 48
`
`Ac-EOXXXX-NH2
`
`1--
`
`A De F G H 1 K L M N P a R s T v v
`
`I-
`
`FIG. 4C
`
`
`
`Page 7 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 7 of 11
`
`5,504,190
`
`Ac-FOXXXX-NH 2
`
`2.0
`
`1.0
`
`0
`
`2.0
`
`1.0
`
`0
`
`2.0
`
`1.0
`
`0
`
`A DE F GH I K LM NPa RS TV Y
`
`FIG. 40
`
`Ac-GOXXXX-NH 2
`
`A o E F G H 1 K L M N P a As T v v FIG. 4E
`
`Ac-HOXXXX-NH2
`
`A o E F G H 1 K L M N P a R s T v v FIG. 4F
`
`
`
`Page 8 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 8 of 11
`
`5,504,190
`
`Ac-IOXXXX-NH 2
`
`-
`
`A o E F G H 1 K L M N P a R s T v Y FIG. 4G
`
`Ac-KOXXXX-NH2
`
`2.0
`
`1.0
`
`2.0
`
`1.0
`
`0
`
`A DE F GH I K LM NPO R S TV Y FIG. 4H
`
`Ac-LOXXXX-NH2
`
`2.0
`
`1.0
`
`0
`
`,II
`A DE F GH I K L M NPO RS Tv y FIG. 41
`
`�
`
`
`
`Page 9 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 9 of 11
`
`5,504,190
`
`Ac-MOXXXX-NH2
`
`2.0 ---
`
`1.0
`
`0
`
`2.0
`
`1.0
`
`0
`
`2.0
`
`A DE F GH I K L M NPQ R S TV Y
`
`FIG. 4J
`
`Ac-NOXXXX-NH2
`
`A DE F GH I K LM NPQ RS TV Y FIG. 4K
`
`Ac-POXXXX-NH2
`
`A 0 E F G H I K L M N p Q R s T v y FIG. 4L
`
`
`
`Page 10 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 10 of 11
`
`5,504,190
`
`Ac-QOXXXX-NH2
`
`2.0 -=
`
`1.0
`
`A DE F GH I K LM NPQ R S Tv y FIG. 4M
`
`Ac-ROXXXX-NH2
`
`2.0
`
`1.0
`
`o�
`
`A o E F G H 1 K L M N P a R s T v v FIG. 4N
`
`�
`
`Ac-SOXXXX-NH2
`
`2.0 -----
`
`1.0
`
`A o e F G H 1 K L M N P a R s T v v FIG. 40
`
`
`
`Page 11 of 62
`
`
`
`U.S. Patent
`
`Apr. 2, 1996
`
`Sheet 11 of 11
`
`5,504,190
`
`Ac-TOXXXX-NH2
`
`2.0
`
`1.0
`
`0
`
`A o e F G H 1 K L M N P a R s T v v FIG. 4P
`
`Ac-VOXXXX-NH2
`
`2.0 -----l-all�tt.-1----
`
`1.0
`
`2.0
`
`1.0
`
`0
`
`--.-I
`A o e F G H 1 K L M N P a R s T v v FIG. 4Q
`
`Ac-YOXXXX-NH2
`
`f-
`
`� ll
`A 0 E F G H I K L M N p 0 R s T v y FIG. 4R
`
`
`
`Page 12 of 62
`
`
`
`5,504,1 90
`
`1
`EQUIMOLAR MULTIPLE OLIGOMER
`MIXTURES, ESPECIALLY OLIGOPEPTIDE
`MIXTURES
`
`CROSS-REFERENCE TO RELATED
`APPLICATION
`
`This is a division, of application Ser. No. 07/797,551, filed
`Nov. 19, 1991 and now abandoned, which is a continuation
`in-part of application Ser. No. 07/701,658 filed May 16,
`1991 and now abandoned, that was a continuation-in-part of
`application Ser. No. 07/617,023, filed Nov. 21, 1990 and
`now abandoned, whose disclosures are incorporated by
`reference.
`
`TECHNICAL FIELD
`
`The present invention relates to the organic synthesis of
`oligomeric sequences of compounds. More particularly it
`relates to stepwise synthesis of multiple
`independent
`sequences, especially oligomeric peptide chains.
`
`BACKGROUND AND RELATED ART
`
`20
`
`2
`Synthesized peptides are released from the resin by acid
`catalysis (typically with hydrofluoric acid or trifiuoroacetic
`acid), which cleaves the peptide from the resin leaving an
`amide or carboxyl group on its C-terminal amino acid.
`5 Acidolytic cleavage also serves to remove the protecting
`groups from the side chains of the amino acids in the
`synthesized peptide. Finished peptides can then be purified
`by any one of a variety of chromatography methods.
`Though most peptides are synthesized with the above
`10 described procedure using automated instruments, a recent
`advance in the solid phase method by R. A. Houghten allows
`for synthesis of multiple independent peptides simulta
`neously through manually performed means. The "Simulta
`neous Multiple Peptide Synthesis" ("SMPS") process is
`15 described in U.S. Pat. No. 4,631,211 (1986); Houghten,
`Proc. Natl. Acad. Sci., 82:5131-5135 (1985); Houghten et
`al., Int. J. Peptide Protein Res., 27:673-678 (1986);
`Houghten et al., Biotechniques, 4, 6, 522-528 (1986), and
`Houghten, U.S. Pat. No. 4,631,211, whose disclosures are
`incorporated by reference.
`Illustratively, the SMPS process employs porous contain
`ers such as plastic bags to hold the solid support synthesis
`resin. A Merrifield-type solid-phase procedure is carried out
`with the resin-containing bags grouped together appropri-
`25 ately at any given step for addition of the same, desired
`amino acid residue. The bags are then washed, separated and
`regrouped for addition of subsequent same or different
`amino acid residues until peptides of the intended length and
`sequence have been synthesized on the separate resins
`30 within each respective bag.
`That method allows multiple, but separate, peptides to be
`synthesized at one time, since the peptide-linked resins are
`maintained in their separate bags throughout the process.
`The SMPS method has been used to synthesize as many as
`35 200 separate peptides by a single technician in as little as
`two weeks, a rate vastly exceeding the output of most
`automated peptide synthesizers.
`A robotic device for automated multiple peptide synthesis
`has been recently commercialized. The device performs the
`sequential steps of multiple, separate solid phase peptide
`synthesis through
`iterative mechanical-intensive means.
`This instrument can synthesize up to 96 separate peptides at
`one time, but is limited at present by the quantity of its
`peptide yield.
`Several research groups have reported the synthesis of
`synthetic combinatorial libraries of peptides. Those reports
`are discussed below.
`Of interest is work by Geysen et al., which deals with
`methods for synthesizing peptides with specific sequences of
`amino acids and then using those peptides to identify
`reactions with various receptors. Geysen et al.' s work pre
`supposes that one has a prior knowledge of the general
`nature of the sequences required for the particular receptors,
`55 so that the appropriate group of peptides can be synthesized.
`See U.S. Pat. Nos. 4,708,871 and 4,833,092; P.C.T. Publi
`cations Nos. WO 84/03506 and WO 84/03564; Geysen et
`al., Proc. Natl. Acad. Sci. U.S.A., 81:3998-4002 (1984);
`Geysen et al., Proc. Natl. Acad. Sci. U.S.A., 82:178-182
`60 (1985); Geysen et al., in Synthetic Peptides as Antigens,
`130---149 (1986); Geysen et al., J. Immunol. Meth.,
`102:259-274 (1987); and Schoofs et al., J. Immunol.,
`140:611-616 (1988).
`In published PC T application PC T/AU85/00165 (WO
`86/00991), Geysen describes a method for determining
`so-called "mimotopes". A mimotope is defined as a catamer
`(a polymer of precisely defined sequence formed by the
`
`40
`
`Over the last several years, developments in peptide
`synthesis technology have resulted in automated synthesis of
`peptides accomplished through the use of solid phase syn
`thesis methods. The solid phase synthesis chemistry that
`made this technology possible was first described in Mer
`rifield et al. J. Amer. Chern. Soc., 85:2149-2154 (1963). The
`"Merrifield method" has for the most part remained
`unchanged and is used in nearly all automated peptide
`synthesizers available today.
`In brief, the Merrifield method involves synthesis of a
`peptide chain on solid support resin particles. These particles
`typically consist of polystyrene cross-linked with divinyl
`benzene to form porous beads which are insoluble in both
`water and various organic solvents used in the synthesis
`protocol. The resin particles contain a fixed amount of
`amino- or hydroxylmethyl aromatic moiety which serves as
`the linkage point for the first amino acid in the peptide.
`Attachment of the first amino acid entails chemically
`reacting its carboxyl-terminal (C-terminal) end with deriva
`tized resin to form the carboxyl-terminal end of the oli
`gopeptide. The alpha-amino end of the amino acid is typi
`cally blocked with a t-butoxy-carbonyl group (t-Boc) or with
`a 9-fiuorenylmethyloxycarbonyl (F-Moc) group to prevent
`the amino group which could otherwise react from partici
`pating in the coupling reaction. The side chain groups of the 50
`amino acids, if reactive, are also blocked (or protected) by
`various benzyl-derived protecting groups in the form of
`ethers, thioethers, esters, and carbamates.
`The next step and subsequent repetitive cycles involve
`deblocking the amino-terminal (N-terminal) resin-bound
`amino acid (or terminal residue of the peptide chain) to
`remove the alpha-amino blocking group, followed by
`chemical addition (coupling) of the next blocked amino
`acid. This process is repeated for however many cycles are
`necessary to synthesize the entire peptide chain of interest.
`After each of the coupling and deblocking steps, the resin
`bound peptide is thoroughly washed to remove any residual
`reactants before proceeding to the next. The solid support
`particles facilitate removal of reagents at any given step as
`the resin and resin-bound peptide can be readily filtered and 65
`washed while being held in a column or device with porous
`openings.
`
`45
`
`
`
`Page 13 of 62
`
`
`
`5,504,190
`
`3
`condensation of a precise number of small molecules),
`which in at least one of its conformations has a surface
`region with the equivalent molecule topology to the epitope
`of which it is a mimic. An epitope is defined as the surface
`of an antigenic molecule which is delineated by the area of
`interaction with an antibody molecule.
`The mimotopes are synthesized on a series of solid
`polymer (e.g. polyethylene with a coating of grafted acrylic
`
`acid) rods having a diameter of about 4 mm and a length of
`about 50 mm. A spacer formed by reaction of the e-amino
`
`group of t-Boc-lysine methyl ester and then t-Boc-alanine
`was added to the resins, followed by removal of the t-Boc
`group to provide an amino group to be used to begin the
`syntheses.
`A mixture of blocked amino acids containing different
`amounts of each of the blocked twenty amino acids to be
`used was dissolved in dimethyl formamide and then coupled
`to the rods. That first coupling was repeated three times
`using conventional solid phase synthesis techniques. Twenty
`amino acid residues were individually next added so that
`twenty 5-mer sequences were prepared, each having a
`single, known amino acid residue at the amino-terminus and
`a mixture of amino acid residues at each of the four other
`positions of the chain. Each of those twenty rod-linked
`peptides was then individually reacted with each of the
`twenty amino acid residues to form 400 (20x20) 6-mer
`peptides having the two amino-terminal positions defined
`and the four remaining positions as mixtures. Two more
`positions of mixtures of amino acids were then added, and
`the terminal amine acetylated to form N-acetyl 8-mers 30
`linked to the rods whose first two amino acid positions were
`undefined (mixtures), followed by two defined positions,
`followed by four undefined positions (mixtures), followed
`by the spacer and then the supporting rods.
`The 400 rod-linked N-acetyl 8-mer peptide mixture 35
`preparations were then screened in an ELISA assay using a
`monoclonal antibody to a desired antigenic protein. The
`8-mers having the best binding to the antibody were iden
`tified. Two sets of further 8-mers that contained the identi
`fied best-binding 2-mer sequences within those 8-mers were 40
`prepared.
`A first set contained mixed amino acids at the three
`C-terminal positions, followed toward the N-terminus, by a
`position containing each of the twenty amino acids made by
`twenty separate couplings, the identified 2-mer sequences,
`two further mixtures at the next two positions, and an
`N-terminal acetyl group. The second group contained mixed
`amino acids at the four C-terminal positions, the identified
`2-mer sequences, a position made by separate couplings of
`each of the twenty amino acids, mixed amino acids as the
`terminal residues and an N-terminal acetyl group.
`Each of those rod-linked N-acetyl 8-mers was again
`screened in an ELISA with the monoclonal antibody. The
`best binding sequences for each group were identified, and
`thus 4-mer, best-binding sequences were identified.
`The above process of separately adding each of the amino
`acids on either side of identified best-binding sequences was
`repeated until an optimum binding sequence was identified.
`The above method, while elegant, suffers from several
`disadvantages. First, owing to the small size of each rod
`used, relatively small amounts of each peptide is produced.
`Second, each assay is carried out using the rod-linked
`peptides, rather than the free peptides in solution. Third,
`even though specific amounts of each blocked amino acid 65
`are used to prepare the mixed amino acid residues at the
`desired positions, there is no way of ascertaining that an
`
`4
`equimolar amount of each residue is truly present at those
`positions.
`In addition, Furka et al., (1988, 14th International Con
`gress of Biochemistry, Volume 5, Abstract FR:013)
`5 described the synthesis of nine tetrapeptides each of which
`contained a single residue at each of the amino- and car
`boxy -termini and mixtures of three residues at each position
`therebetween. The abstract futher asserts that those authors'
`experiments indicated that a mixture containing up to 180
`10 pentapeptides could be easily synthesized in a single run. No
`biological assays were reported.
`Recent reports (Devlin et al., Science, 249:404-405
`[1990] and Scott et al., Science, 249:386-390 [1990]) have
`described the use of recombinant DNA and bacterial expres-
`15 sian to create highly complex mixtures of peptides. For
`example, a 45-nucleotide base pair stretch of DNA was
`synthesized in which the individual nucleotide bases were
`varied to contain all four possible nucleotide bases (guanine,
`adenine, cytosine and thymidine) at every position in the
`20 synthesized DNA chain, except at each third position (3, 6,
`9, etc.) which contained only guanine and cytosine. The
`omission of adenine and thymidine at every third position in
`the synthesized DNA removed the possibility of chain
`terminator triplet codons ending in A or T, such as TAA or
`25 TGA.
`The resulting DNA sequence would then code for a
`mixture of 15-mer peptides with all combinations of the 20
`naturally occurring amino acids at each position.
`Those investigators fused the 45 synthetic nucleotide
`sequence to a gene coding for the coat protein of a simple
`bacteriophage and created a large library of these bacte
`riophages. Each member of the library contained a different
`45-mer DNA fusion sequence and therefore each member of
`the library resulted in a different 15-mer peptide fused to the
`outer coat protein of its corresponding fully assembled
`bacteriophage particle. Screening of the recombinant bacte-
`riophage particles in a biochemical assay allowed the inves
`tigators to find individual peptide-coat protein fusions (bac
`teriophages) that were active in that assay by enrichment,
`selection and clonal isolation of the enriched bacteriophages
`that contained active peptide fusions. By determining the
`DNA sequence of the cloned bacteriophages, the investiga
`tors could deduce which peptide sequences were active in
`their assay.
`That method yielded several peptide sequences from a
`mixture of 107 or more recombinant bacteriophages. Each of
`the 15-mer peptides found contained the same four-amino
`acid sequence somewhere within its overall sequence,
`thereby allegedly validating the assay accuracy and meth
`odological approach.
`The recombinant DNA method is extremely powerful for
`screening large numbers of peptides. However, it is limited
`in that the peptides must be fused to a larger protein as a
`55 result of and integral to the design of the method. The
`peptide-protein fusions (and corresponding bacteriophage
`particles) are likely to be unreactive in many biochemical,
`biological and in vivo assays where the peptides must be
`present in solution without steric hindrance or conforma-
`tional distortion. In addition, the method results in an
`over-representation of some sequences of peptides due to the
`inherent redundancy of the genetic code which has several
`codons per amino acid in some cases and only one codon per
`amino acid in others.
`Still further, neither group reported data as being defini
`tive for the determination of optional peptide ligands for
`strepavidin (Devlin et al.), or for the two monoclonal
`
`45
`
`50
`
`60
`
`
`
`Page 14 of 62
`
`
`
`5,504,190
`
`5
`antibodies raised against myohemorythinin (Smith et al.).
`Neither group provided a single specific answer comparable
`to the expected sequence.
`More recently, Fodor et al., Science, 251:767-773 (1991),
`described the solid phase synthesis of mixtures of peptides 5
`or nucleotides on glass microscope slides treated with ami
`nopropyltriethoxysilane to provide amine functional groups.
`Predetermined amino acids were then coupled to predefined
`areas of the slides by the use of photomasks. The photolabile
`protecting group NVOC (nitroveratryloxycarbonyl) was 10
`used as the amino-terminal protecting group.
`By using irradiation, a photolabile protecting group and
`masking, an array of 1024 different pep tides coupled to the
`slide was prepared in ten steps. Immunoreaction with a
`fluorescent-labeled monoclonal antibody was assayed with 15
`
`epifiuorescence microscopy.
`This elegant method is also limited by the small amount
`of peptide or oligonucleotide produced, by use of the syn
`thesized peptide or nucleotide affixed to the slide, and also
`by the resolution of the photomasks. This method is also less 20
`useful where the epitope bound by the antibody is unknown
`because all of the possible sequences are not prepared.
`The primary limitation of the above new approaches for
`the circumvention of individual screening of millions of 25
`individual peptides by the use of a combinatorial library is
`the inability of the peptides generated in those systems to
`interact in a "normal" manner with acceptor sites, analogous
`to natural interaction processes (i.e., in solution at a con
`
`6
`lular receptor, etc.). If one could devise a means to prepare
`and screen a synthetic combinatorial library of peptides,
`then the normal exquisite selectivity of biological affector/
`acceptor systems could be used to screen through vast
`numbers of synthetic oligopeptides.
`The availability of a wide variety of clearly identified
`peptides in relatively limited mixtures would greatly facili
`tate the search for the optimum peptide for any particular
`therapeutic end use application. At
`the present
`time,
`researchers are hampered by the inability to rapidly create,
`identify and screen large numbers of peptides with specific
`receptors. Work such as reported by Geysen has been
`valuable where the general nature of the required amino acid
`residue sequence could be previously determined, so that the
`specific peptides of interest could be individually formu
`lated. However, such techniques cannot insure that the
`optimum peptides are identified for testing.
`It would therefore be of considerable interest to have a
`method for the precise synthesis of mixtures of peptides in
`which individual peptide sequences can be specifically
`defined, such that a comprehensive array of peptides is
`available to researchers for the identification of one or more
`of the optimum peptides for reaction with receptors of
`interest, from which one can derive optimum therapeutic
`materials for treatment of various organism dysfunctions. It
`would also be of value for such a process to have the
`capability to produce equivalent sequences of other types of
`oligomeric compounds.
`
`centration relevant to the receptors, antibody binding sites, 30
`
`enzyme binding pockets, or the like being studied without
`the exclusion of a large percentage of the possible combi
`natorial library). Secondarily, the expression vector systems
`do not readily permit the incorporation of the D-forms of the
`natural amino acids or the wide variety of unnatural amine 35
`acids which would be of interest in the study or development
`of such interactions.
`The interest in obtaining biologically active peptides for
`pharmaceutical, diagnostic and other uses would make
`or a single peptide within a mixture with optimal activity for
`desirable a procedure designed to find a mixture of peptides 40
`a target application. Screening mixtures of peptides enables
`the researcher to greatly simplify the search for useful
`therapeutic or diagnostic peptide compounds. Mixtures con
`taining hundreds of thousands or more peptides should be 45
`readily screened since many biochemical, biological and
`small animal assays are sensitive enough to detect activity of
`compounds that have been diluted down to the nanogram or
`even picogram per milliliter range, the concentration range
`at which naturally occurring biological signals such as 50
`peptides and proteins operate.
`Almost all of the broad diversity of biologically relevant
`ligand-receptor (or affector-acceptor) interactions occur in
`the presence of a complex milieu of other substances (i.e.,
`proteins make up approximately 5-10 percent of plasma,
`e.g. albumin 1-3 percent, antibodies 2-5 percent-salts, lip
`ids/fats, etc.). This is true for virtually all biologically active
`compounds since most are commonly present, and active, at
`nanomolar and lower concentrations. These compounds are
`also, in most instances, produced distant from their affection
`sites. That a small peptide (or other molecule) can readily
`"find" an acceptor system, bind to it, and affect a necessary
`biological function prior to being cleared from the circula
`tion or degraded suggested that a single specific peptide
`sequence can be present in a very wide diversity, and 65
`concentration, of other individual peptides and still be
`recognized by its particular acceptor system (antibody, eel-
`
`BRIEF SUMMARY OF THE INVENTION
`In one aspect, the invention herein contemplates a process
`that provides for the synthesis of complex mixtures of
`step-growth oligomers, especially peptides, wherein each
`position in the oligomeric sequence chain contains an
`equimolar representation of reacted bifunctional monomeric
`repeating unit compound, such as an amino acid residue,
`added at that step. In peptide synthesis, the method circum-
`vents the problem of unequal reaction yields during addition
`of blocked amino acids reacted as a mixture in a coupling
`step in the Merrifield solid phase synthesis procedure. Use
`of the present method also provides a relatively much larger
`amount of coupled oligomer than previously contemplated.
`In its preferred embodiment, the invention contemplates
`the organic synthesis of complex equimolar mixtures of
`oligopeptide sequences on a solid support material. The
`equimolar oligopeptide sequences consist essentially of
`chains of amino acid residues linked end-to-end by peptide
`bonds wherein the amino acid residue incorporated at any
`one position in the chain can be varied, such as to contain all
`or a combination of the twenty naturally occurring amino
`acids and/or their derivatives. The invention enables syn
`thesis of these peptide mixtures with equal and precise
`55 representation of any amino acid residues at any position in
`the chain at which a mixture of amino acid residues is
`intended to be represented. The process can use any type of
`peptide addition chemistry and protocols, but preferably
`uses the Merrifield solid phase synthesis procedure in pro-
`tocols similar to that of the Houghten SMPS process.
`In yet another aspect, the invention comprises a method
`for the identification of one or more optimum peptides for
`reaction with a designated acceptor, such that design of
`therapeutic materials for treatment of organism dysfunctions
`involving such receptor can be facilitated.
`In its broadest form, a process of this invention is defined
`as a process for the synthesis of a complex mixture pool of
`
`60
`
`
`
`Page 15 of 62
`
`
`
`5,504,190
`
`7
`solid support-coupled monomeric repeating unit com
`pounds, wherein the mixture pool contains a substantially
`equimolar representation of the reacted monomeric repeat
`ing unit compound, such as amino acid residues, coupled at
`that step. In accordance with this method,
`(a) a plurality of solid supports is provided, each solid
`support comprised of a particle linked to reactive functional
`groups. The functional groups of the solid support react with
`a functional group of each of the monomeric repeating unit
`compounds to be reacted. In a preferred embodiment, each
`of the solid supports is within a porous container, the solid
`support is of a size that is larger than the pores of the
`container, and both the container and solid support are
`substantially insoluble in a liquid medium used during the
`stepwise synthesis.
`(b) A plurality of liquid media is provided, each medium
`containing a different monomeric repeating unit compound
`from a plurality of monomeric repeating unit compounds
`from which the oligomers are to be formed. Each of the
`monomeric repeating unit compounds has a first reactive
`functional group that reacts with the reactive functional
`group of the solid support and a second reactive functional
`group that is capable of reacting during the reaction of the
`solid support functional group and the first reactive func
`tional group, but is protected from so reacting by a selec
`tively removable, covalently linked protecting group.
`(c) Each of the solid supports is placed in a different one
`of the liquid media and the reactive functional group of each
`solid support is therein reacted with a first reactive func
`tional group of a monomeric repeating unit compound in
`that respective medium to couple that monomeric repeating
`unit compound to the solid support.
`(d) Each of the reactions is maintained for a time period
`and under conditions sufficient for all of the reactive func
`tional groups of the solid support to couple to the monomeric 35
`repeating unit compound to form a plurality of monomeric
`repeating unit-coupled s