CORRECTED
`
`VERSION* PCT
`
`WORLD INTELLECTUAL PROPERTY ORGANIZATION
`International Bumw
`INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PC1)
`
`(51) International
`Patent Qassification:
`C12P 21/00, C12N 5/10, 7/01, 15/00 *
`Al
`
`
`
`(11) International Publication
`
`Number:
`(43) International
`
`
`
`Publication Date:
`
`WO 90/07861
`
`26 July 1990 (26.07.90)
`
`Number: PCT/US89/05857
`
`(21) International Application
`(81) Designated States: AT, AT (European patent), AU, BB, BE
`
`
`BG, BJ (OAPI
`
`(European patent), BF (OAPI patent),
`(22) International
`Filing Date: 28 December 1989 (28.12.89)
`
`
`patent), BR, CF (OAPI patent), CG (OAPI patent), CH,
`CH (European
`
`patent), CM (OAPI patent), DE, DE
`
`FI, FR
`
`
`(European patent), DK, ES (European patent),
`(European
`
`
`patent), GA (OAPI patent), GB, GB (Euro­
`JP, KP, KR,
`pean patent), HU, IT (European patent),
`
`
`
`
`LI<, LU, LU (European patent), MC, MG, ML (OAPI
`patent), MR (OAPI patent), MW, NL, NL (European
`
`
`NO, RO, SD, SE, SE (European
`patent),
`
`patent), SN
`
`
`(OAPI patent), SU, TD (OAPI patent), TG (OAPI pa­
`tent).
`
`data:
`(30) Priority
`290,975
`310,252
`
`28 December 1988 (28.12.88) US
`
`
`
`13 February 1989 (13.02.89) US
`
`PROTEIN DESIGN LABS, INC. [US/US];
`(71)Applicant:
`3181 Porter Drive, Palo Alto, CA 94304 (US).
`(72) Inventors:
`QUEEN, Cary, L. ; 1300 Oak Creek Drive, Palo
`Alto, CA 94304 (US). SELICK, Harold, Edwin ; 1673
`Published
`
`
`Sunnyslope Avenue, Belmont, CA 94002 (US).
`With international search report.
`
`of the time limit for amending the
`Before the expiration
`claims and to be republished in the event of the receipt of
`SMITH, William, M.; Townsend and Townsend,
`(74)Agent:
`One Market Plaza, 2000 Steuart
`Tower, San Francisco,
`amendments.
`CA 94105 (US).
`
`"CHIMERIC IMMUNOGLOBUUNS SPECIFIC FOR pSS TAC PROTEIN OF THE JL-2 RECEPTOR
`(54)Title:
`
`(57) Abstract
`
`Novel methods for designing humanized immunoglobulins having one or more complementary determining regions
`
`
`
`
`
`
`
`
`(CDR's) from a donor irnmuooglobulin and a framework region from a human immunoglobulin comprising first comparing the
`
`
`
`
`
`
`
`framework or variable region amino acid sequence of the donor immunoglobulin to corresponding sequences in a collection of
`
`
`
`
`
`
`human immunoglobulin chains, and selecting as the human immunoglobulin one of the more homologous sequences from the
`
`
`
`collection. Each humanized immunoglobulin chain may comprise about 3 or more amino acids from the donor immunoglobulin
`to a CDR in the donor immunoglobulin.
`
`
`adjacent in addition to the CDR's, usually at least one of which is immediately
`The
`heavy and light chains may each be designed by using any one or all three additional
`
`
`position criteria. When combined into an
`
`
`
`intact antibody, the humanized immunoglobulins of the present invention will be substantially non-immunogenic in humans and
`
`
`
`as the donor immunoglobulin the same affinity to the antigen,
`retain substantially
`such as a protein
`
`
`or other compound contain­
`ing an epitope.
`
`\,
`
`• (Rererroo to in PCT Oau:nc No. 2211990, Section II)
`
`1 of 51
`
`BI Exhibit 1050
`
`

`

`t
`
`'·
`
`_ _.·
`J
`•'° '
`'!
`
`MG M�r
`ML Mlli
`MR Mauritania
`f;'NI Malawi
`NL NctberLtods
`I'«) Norway
`RO
`SD Sudan
`SE
`Sweden
`� �ncgal
`SlJ Sorict Union
`11) Chad
`TG Togo
`UnUd S1a1CS of America
`l6
`
`Romania
`
`applications
`
`international
`
`FR
`
`FOR THE PURPOSES OF INFORMATION ONLY
`
`Codes used lo iden1ify StalCS party to lhc PCr on lhc front pages of pamphlets publishing
`under lhc Per.
`AT Auma
`AU Auiualia
`88 Bast.dos
`11£ Bclgilm
`BF
`Bwlmf"uo
`BG 8qa:a
`BJ Benin
`BR Bia.zil
`CA Ouwla
`CF Central African Rcpubbc
`CG Co11go
`OI Swiaetl1nd
`CM Cameroon
`DE Germany. fcdc111l Rcpubfoc or
`DIC Ocnmart
`
`ES Spain
`n Fmllnd
`France
`GA Gabon
`GB UaS.cd Kinadom
`i.J Hunpty
`rr
`Italy
`JP Japan
`ICP Dcmocnlii: Pcopk's Rcpubli:
`u Liedltcimio
`SriWh
`lJ(
`w t.uwnboura
`MC MonaQ)
`
`orKon:a
`Rcpubli:orKorca
`
`KR
`
`2 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT /US89/05857
`
`CHIMERIC IMMUNOGLOBULINS SPECIFIC FOR pSS TAC PROTEIN
`OF THE IL-2 RECEPTOR
`
`Field of the Invention
`The present invention relates generally to the
`combination of recombinant DNA and monoclonal antibody
`technologies for developing novel therapeutic agents and,
`more particularly, to the production of non-immunogenic
`antibodies and their uses.
`
`Background of the Invention
`In mammals, the immune response is mediated by two
`types of cells that interact specifically with foreign
`material, i.e., antigens. one of these cell types, B-cells,
`are responsible for the production of antibodies. The second
`cell class, T-cells, include a wide variety of cellular
`subsets controlling the in vivo function of both B-cells and
`a wide variety of other hematopoietic cells, including T­
`cells.
`
`One way in which T-cells exert this control is
`through the production of a lymphokine known as interleukin-2
`(IL-2 ) , originally named T-cell growth factor.
`IL-2 ' s prime
`function appears to be the stimulation and maintenance of T­
`Indeed, some immunologists believe that IL-2 may be
`cells.
`at the center of the entire immune response (� , Farrar, J.,
`et al., Immunol. Rev . .§.1 : 129-166 (1982), which is
`incorporated herein by reference) .
`To exert its biological effects, IL-2 interacts
`with a specific high-affinity membrane receptor (Greene, w.,
`et al., Progress in Hematology XIV, E . Brown , Ed., Grune and
`Statton, New York (1986) , at pgs. 283 ff). The human IL-2
`receptor is a complex multichain glycoprotein, with one
`chain, known as the Tac peptide, being about 55kD in size
`(�, Leonard , w., et al . , J. Biol. Chem. 260:1872 (1985),
`which is incorporated herein by reference) . A gene encoding
`this protein has been isolated, and predicts a 272 amino acid
`peptide, including a 21 amino acid signal peptide (�,
`Leonard, W., et al. , Nature 311: 626 (1984)). The 219 NH2-
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`"
`
`35
`
`3 of 51
`
`BI Exhibit 1050
`
`

`

`W090/07861
`
`PCT/US89/05857
`
`2
`
`terminal amino acids of the p55 Tac protein apparently
`comprise an extracellular domain (see, Leonard, w., et al.,
`Science, 230:633-639 (1985), which is incorporated herein by
`
`reference).
`
`5
`
`Much of the elucidation of the human IL-2
`
`10
`
`15
`
`receptor's structure and function is due to the development
`
`of specifically reactive monoclonal antibodies. In
`
`particular, one mouse monoclonal antibody, known as anti-Tac
`(Uchiyama, et al., J. Immunol. 126:1393 (1981)) has shown
`that IL-2 receptors can be detected on T-cells, but also on
`cells of the monocyte-macrophage family, Kupffer cells of the
`
`liver, Langerhans' cells of the skin and, of course,
`
`activated T-cells. Importantly, resting T-cells, B-cells or
`
`circulating machrophages typically do not display the IL-2
`receptor (Herrmann, et al., J. Exp. Med. �:1111 (1985)).
`The anti-Tac monoclonal antibody has also been used
`
`to define lymphocyte functions that require IL-2 interaction,
`
`and has been shown to inhibit various T-cell functions,
`
`including the generation of cytotoxic and suppressor T
`
`20
`
`lymphocytes in cell culture. Also, based on studies with
`
`25
`
`30
`
`anti-Tac and other antibodies, a variety of disorders are now
`associated with improper IL-2 receptor expression by T-cells,
`in particular adult T-cell leukemia.
`More recently, the IL-2 receptor has been shown to
`be an ideal target for novel therapeutic approaches to T-cell
`mediated diseases. It has been proposed that IL-2 receptor
`specific antibodies, such as the anti-Tac monoclonal
`
`antibody, can be used either alone or as an imrnunoconjugate
`(�, with Ricin A, isotopes and the like) to effectively
`remove cells bearing the IL-2 receptor. These agents can,
`
`for example, theoretically eliminate IL-2 receptor-expressing
`
`leukemic cells, certain B-cells, or activated T-cells
`
`involved in a disease state, yet allow the retention of
`
`mature normal T-cells and their precursors to ensure the
`
`35
`
`capability of mounting a normal T-cell immune response as
`
`needed.
`
`In general, most other T-cell specific agents can
`
`destroy essentially all peripheral T-cells, which limits the
`
`agents' therapeutic efficacy. overall, the use of
`
`4 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT /US89/05857
`
`3
`
`appropriate monoclonal antibodies specific for the IL-2
`receptor may have therapeutic utility in autoimmune diseases,
`organ transplantation and any unwanted response by activated
`T-cells.
`Indeed, clinical trials have been initiated using,
`5 �' anti-Tac antibodies (�, generally, Waldman, T., et
`al., Cancer Res. 45:625 ( 1985) and Waldman, T. , Science
`2 3 2 : 727-732 (1986), both of which are incorporated herein by
`reference) .
`Unfortunately, the use of the anti-Tac and other
`non-human monoclonal antibodies have certain drawbacks,
`particularly in repeated therapeutic regimens as explained
`below. Mouse monoclonal antibodies, for example, do not fix
`human complement well, and lack other important
`inununoglobulin functional characteristics when used in
`humans.
`
`10
`
`15
`
`Perhaps more importantly, anti-Tac and other non­
`human monoclonal antibodies contain substantial stretches of
`amino acid sequences that will be immunogenic when injected
`into a human patient. Numerous studies have shown that,
`after injection of a foreign antibody, the immune response
`elicited by a patient against an antibody can be quite
`strong, essentially eliminating the antibody ' s therapeutic
`utility after an initial treatment. Moreover, as increasing
`numbers of different mouse or other antigenic (to humans)
`monoclonal antibodies can be expected to be developed to
`treat various diseases, after the first and second treatments
`with any different non-human antibodies, subsequent
`treatments even for unrelated therapies can be ineffective or
`even dangerous in themselves.
`While the production of so-called "chimeric
`antibodies" (�, mouse variable regions j oined to human
`constant regions) has proven somewhat successful, a
`significant immunogenicity problem remains. In general, the
`production of human immunoglobulins reactive with the human
`IL-2 receptor, as with many human antigens, has been
`extremely difficult using typical human monoclonal antibody
`production techniques. Similarly, utilizing recombinant DNA
`technology to produce so-called "humanized" antib-Odies (see,
`
`20
`
`25
`
`30
`
`35
`
`5 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT /US89/05857
`
`4
`�, EPO Publication No. 0239400), provides uncertain
`results, in part due to unpredictable binding affinities.
`Thus, there is a need for improved forms of human­
`like immunoglobulins, such as those specific for the human
`IL-2 receptor, that are substantially non-immunogenic in
`humans, yet easily and economically produced in a manner
`suitable for therapeutic formulation and other uses. The
`present invention fulfills these and other needs.
`
`summary of the Invention
`The present invention provides novel compositions
`useful, for example, in the treatment of T-cell mediated
`human disorders, the compositions containing human-like
`immunoglobulins specifically capable of blocking the binding
`of human IL-2 to its receptor and/or capable of binding to
`the p55 Tac protein on human IL-2 receptors. The
`immunoglobulins can have two pairs of light chain/heavy chain
`complexes, typically at least one pair having chains
`comprising mouse complementarity determining regions
`functionally joined to human framework region segments. For
`example, mouse complementarity determining regions, with or
`without additional naturally-associated mouse amino acid
`residues, can be used to produce human-like antibodies
`capable of binding to the human IL-2 receptor at affinity
`levels stronger than about 108 M-1•
`The immunoglobulins, including binding fragments
`and other derivatives thereof, of the present invention may
`be produced readily by a variety of recombinant DNA
`techniques, with ultimate expression in transfected cells,
`preferably immortalized eukaryotic cells, such as myeloma or
`hybridoma cells. Polynucleotides comprising a first sequence
`coding for human-like immunoglobulin framework regions and a
`second sequence set coding for the desired immunoglobulin
`complementarity determining regions can be produced
`synthetically or by combining appropriate cDNA and genomic
`DNA segments.
`The human-like immunoglobulins may be utilized
`alone in substantially pure form, or complexed with a
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`6 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT /US89/05857
`
`5
`
`cytotoxic agent, such as a radionuclide, a ribosomal
`inhibiting protein or a cytotoxic agent active at cell
`surfaces. All of these compounds will be particularly useful
`in treating T-cell mediated disorders. The human-like
`immunoglobulins or their complexes can be prepared in a
`pharmaceutically accepted dosage form, which will vary
`depending on the mode of administration.
`The present invention also provides novel methods
`for designing human-like immunoglobulin chains having one or
`more complementarity determining region� (CDR's) from a donor
`immunoglobulin and a framework region from a human
`immunoglobulin, the preferred methods comprising first
`comparing the framework or variable region amino acid
`sequence of the donor immunoglobulin to corresponding
`sequences in a collection of human immunoglobulin chains, and
`selecting as the human immunoglobulin one of the more
`homologous sequences from the collection. The human
`immunoglobulin, or acceptor immunoglobulin, sequence is
`typically selected from a collection of at least 10 to 2 0
`immunoglobulin chain sequences, and usually will have the
`highest homology to the donor irnmunoglobulin sequence of any
`sequence in the collection . The human immunoglobulin
`framework sequence will typically have about 65 to 70%
`homology or more to the donor immunoglobulin framework
`sequences. The donor immunoglobulin may be either a heavy
`chain or light chain (or both) , and the human collection will
`contain the same kind of chain. A huma.nized light and heavy
`chain can be used to form a complete humanized immunoglobulin
`or antibody, having two light/heavy chain pairs, with or
`without partial or full-length human constant regions and
`other proteins.
`In another embodiment of the present invention,
`either in conjunction with the above comparison step or
`separately, additional amino acids in an acceptor
`irnmunoglobulin chain may be replaced with amino acids form
`the CDR-donor irnmunoglobulin chain. More specifically,
`further optional substitutions of a human framework amino
`acid of the acceptor immunoglobulin with a corresponding
`
`5
`
`10
`
`15
`
`2 0
`
`25
`
`30
`
`35
`
`7 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT/US89/05857
`
`6
`
`amino acid from a donor immunoqlobulin will be made at
`
`positions in the immunoglobulins where:
`
`(a}
`
`the amino acid in the human framework region
`
`of an acceptor immunoglobulin is rare for that position and
`
`5
`
`the corresponding amino acid in the donor immunoglobulin is
`
`common for that position in human immunoglobulin sequences;
`
`..
`
`10
`
`15
`
`or
`
`(b} the amino acid is immediately adjacent to one
`
`of the CDR' s; or
`
`(c} the amino acid is predicted to be within about
`3A of the CDR' s in a three-dimensional immunoglobulin model
`and capable of interacting with the antigen or with the CDR's
`
`of the humanized immunoglobulin.
`
`The humanized immunoglobulin chain will typically
`comprise at least about 3 amino acids from the donor
`immunoglobulin in addition to the CDR's, usually at least one
`
`of which is immediately adjacent to a CDR in the donor
`
`immunoglobulin. The heavy and light chains may each be
`
`designed by using any one or all three of the position
`
`2 0
`
`criteria.
`
`When combined into an intact antibody, the
`
`humanized light and heavy chains of the present invention
`
`will be substantially non-immunogenic in humans and retain
`
`substantially the same affinity as the donor immunoglobulin
`
`25
`
`to the antigen (such as a protein or other compound
`
`containing a.n epitope}. These affinity levels can vary from
`about 108 M-1 or higher, and may be within about 4 fold of the
`donor immunoglobulin' s original affinity to the antigen.
`
`8 of 51
`
`BI Exhibit 1050
`
`

`

`WO 90/07861
`
`PCT /US89/05857
`
`7
`
`BRIEF DESCRIPTION OF THE FIGURES
`Figure 1. comparison of sequences of anti-Tac
`heavy chain (upper lines) and Eu heavy chain (lower lines).
`The 1-letter code for amino acids is used. The first amino
`Identical amino
`acid on each line is numbered at the left.
`acids in the two sequences are connected by lines. The 3
`CDRs are underlined. Other amino acid positions for which
`the anti-Tac amino acid rather than the Eu amino acid was
`used in the humanized anti-Tac heavy chain are denoted by
`an *.
`
`Figure 2 . Comparison of sequences of anti-Tac
`light chain (upper lines) and Eu light chain (lower lines).
`The single-letter code for amino acids is used. The first
`amino acid on each line is numbered at the left.
`Identical
`amino acids in the two sequences are connected by lines. The
`3 CDRs are underlined . Other amino acid positions for which
`the anti-Tac amino acid rather than the Eu amino acid was
`used in the humanized anti-Tac heavy chain are denoted by
`an *·
`
`Figure 3. Nucleotide sequence of the gene for the
`humanized anti-Tac heavy chain variable region gene. The
`translated amino acid sequence for the part of the gene
`encoding protein is shown underneath the nucleotide sequence.
`The nucleotides TCTAGA at the beginning and end of the gene
`are Xba I sites. The mature heavy chain sequence begins with
`amino acid #20 Q.
`Figure 4. Nucleotide sequence of the gene for the
`humanized anti-Tac light chain variable region gene. The
`translated amino acid sequence for the part of the gene
`encoding protein is shown underneath the nucleotide sequence.
`The nucleotides TCTAGA at the beginning and end of the gene
`are Xba I sites. The mature light chain sequence begins with
`amino acid #21 D.
`Figure 5. A. Sequences of the four
`oligonucleotides used to synthesize the humanized anti-Tac
`heavy chain gene, printed 51 to 31• B. Relative positions
`of the oligonucleotides. The arrows point in the 3 '
`direction for each oligonucleotide.
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`9 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT/US89/05857
`
`8
`
`Figure 6.
`oligonucleotides used to synthesize the humanized anti-Tac
`
`(A) Sequences of the four
`
`light chain gene, printed 5' to 31•
`
`(B) Relative positions
`
`of the oligonucleotides.
`
`The arrows point in the 3'
`
`5
`
`direction for each oligonucleotide.
`
`The position of a Hind
`
`III site in the overlap of JFD2 and JFD3 is shown.
`Figure 7. Schematic diagram of the plasmid
`pHuGTACl used to express the humanized anti-Tac heavy chain.
`
`Relevant restriction sites are shown, and coding regions of
`
`10
`
`the heavy chain are displayed as boxes. The direction of
`
`15
`
`20
`
`transcription from the im:munoglobulin (Ig) promoter is shown
`by an arrow. Eu = heavy chain enhancer, Hyg = hygromycin
`resistance gene.
`Figure a. Schematic diagram of the plasmid pHuLTAC
`used to express the humanized anti-Tac light chain. Relevant
`
`restriction sites are shown, and coding regions of the light
`
`chain are displayed as boxes. The direction of transcription
`
`from the Ig promoter is shown by an arrow.
`Figure 9 . Fluorocytometry of HUT-102 and Jurkat
`cells stained with anti-Tac antibody or humanized anti-Tac
`
`antibody followed respectively by fluorescein-conjugated goat
`
`anti-mous� Ig antibody or goat anti-human Ig antibody, as
`
`labeled.
`
`In each panel, the dotted curve shows the results
`
`when the first antibody was omitted, and the solid curve the
`
`25
`
`results when first and second (conjugated) antibodies were
`
`included as described.
`
`Figure 10.
`
`(A) Fluorocytometry of HUT-102 cells
`
`stained with 0-40 ng of anti-Tac as indicated, then with
`
`biotinylated anti-Tac, and then with phycoerythrin-conjugated
`
`30
`
`avidin.
`
`(B) Fluorocytometry of HUT-102 cells stained with
`
`the indicated antibody, then with biotinylated anti-Tac, and
`
`then with phycoerythrin-conjugated avidin.
`
`10 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT /US89/05857
`
`9
`
`DETAILED DESCRIPTION OF THE INVENTION
`In accordance with one embodiment of the present
`invention, human-like immunoglobulins specifically reactive
`with desired epitopes, such as those on the IL-2 receptor on
`human T-cells, are provided. These immunoglobulins, which
`have binding affinities of at least about 108 M-1, and
`preferably 109 M-1 to 1010 M-1 or stronger, are capable of,
`�. blocking the binding of IL-2 to human IL-2 receptors.
`The human-like immunoglobulins will have a human-like
`framework and can have complementarity determining regions
`(CDR's) from an immunoglobulin, typically a mouse
`immunoglobulin, specifically reactive with an epitope on p55
`Tac protein. The immunoglobulins of the present invention,
`which can be produced economically in large quantities, find
`use, for example, in the treatment of T-cell mediated
`disorders in human patients by a variety of techniques.
`The basic antibody structural unit is known to
`comprise a tetramer. Each tetramer is composed of two iden­
`tical pairs of polypeptide chains, each pair having one
`"light" (about 25kD) and one "heavy" chain (about 50-70kD).
`The NH2-terminus of each chain begins a variable region of
`about 100 to 110 or more amino acids primarily responsible
`for antigen recognition. The COOH terminus of each chain
`defines a constant region primarily responsible for effector
`function.
`
`Light chains are classified as either kappa or
`lambda. Heavy chains are classified (and subclassified) as
`gamma, mu, alpha, delta, or epsilon, and define the
`antibody's isotype as IgG, IgM, IgA, IgD and IgE,
`respectively. Within light and heavy chains, the variable
`and constant regions are joined by a "J" region of about 12
`or more amino acids, with the heavy chain also including a
`"D" region of about 12 more amino acids.
`(See, generally,
`Fundamental Immunology, Paul, W. , Ed., Chapter 7, pgs. 131-
`166, Raven Press, N.Y. (1984), which is incorporated herein
`by reference.)
`The variable regions of each light/heavy chain pair
`form the antibody binding site.
`The chains all exhibit the
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`11 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT/US89/05857
`
`10
`
`same general structure of relatively conserved framework
`regions joined by three hypervariable regions, also called
`CDR's (see, "Sequences of Proteins of Immunological
`Interest," Kabat, E., et al., U.S. Department of Health and
`Human Services, (1983}; and Cholthia and Lesk, J. Mol. Biol.,
`196:901-917 (1987}, which are incorporated herein by
`reference}. The CDR's from the two chains of each pair are
`aligned by the framework regions, enabling binding to a
`specific epitope.
`As used herein, the term 11immunoglobulin11 refers to
`a protein consisting of one or more polypeptides
`substantially encoded by immunoglobulin genes. The
`recognized immunoglobulin genes include the kappa, lambda,
`alpha, gamma, delta, epsilon and mu constant region genes, as
`well as the myriad immunoglobulin variable region genes. The
`immunoglobulins may exist in a variety of forms besides
`antibodies; including, for example, Fv, Fab, and F(ab)2, as
`well as in single chains (�, Huston, et al., Proc. Nat.
`Acad. Sci. U.S.A., 85:5879-5883 (1988) and Bird, et al.,
`Science, 242:423-426 (1988), which are incorporated herein by
`reference).
`(See, generally, Hood, et al., "Immunology",
`Benjamin, N.Y., 2nd ed. (1984), and Hunkapiller and Hood,
`Nature, .1£2:15-16 (1986), which are incorporated herein by
`reference}.
`Chimeric antibodies are antibodies whose light and
`heavy chain genes have been constructed, typically by genetic
`engineering, from immunoglobulin gene segments belonging to
`different species. For example, the variable (V) segments of
`the genes from a mouse monoclonal antibody may be joined to
`human constant (C) segments, such as 11 and 73•
`A typical
`therapeutic chimeric antibody is thus a hybrid protein
`consisting of the V or antigen-binding domain from a mouse
`antibody and the c or effector domain from a human antibody
`(�, A.T.C.C. Accession No. CRL 9688 secretes an anti-Tac
`chimeric antibody), although other mammalian species may be
`used.
`
`As used herein, the term "framework region" refers
`to those portions of immunoglobulin light and heavy chain
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`12 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT /US89/0S8S7
`
`11
`variable regions that are relatively conserved (�, other
`than the CDR's) among different immunoglobulins in a single
`species, as defined by Kabat, et al., .211• cit. As used
`herein, a "human-like framework region" is a framework region
`that in each existing chain comprises at least about 70 or
`more amino acid residues, typically 75 to 85 or more
`residues, identical to those in a human inununoglobulin.
`As used herein, the term "human-like
`inununoglobulin" refers to an immunoglobulin comprising a
`human-like framework and in which any constant region present
`is substantially homologous to a human inununoglobulin
`constant region, i.e., at least about 85-90%, preferably
`about 95% identical. Hence, all parts of a human-like
`inununoglobulin, except possibly the CDR's, are substantially
`homologous to corresponding parts of one or more native human
`inununoglobulin sequences. For example, a human-like
`inununoglobulin would not encompass a chimeric mouse variable
`region/human constant region antibody.
`In accordance with another general aspect of the
`present invention, also included are criteria by which a
`limited number of amino acids in the framework of a human­
`like or h�manized inununo9lobulin chain are chosen to be the
`same as the amino acids at those positions in the donor Ig
`rather than in the acceptqr Ig, in order to increase the
`affinity of an antibody comprising the humanized
`inununoglobulin chain.
`This aspect of the present invention is based in
`part on the model that two contributing causes of the loss of
`affinity in prior means of producing humanized antibodies
`(using as examples mouse antibodies as the source of CDR's)
`are:
`
`(1) When the mouse CDR's are combined with the
`human framework, the amino acids in the framework close to
`the CDR's become human instead of mouse. Without intending
`to be bound by theory, we believe that these changed amino
`acids may slightly distort the CDR's, because they create
`different electrostatic or hydrophobic forces than in the
`donor mouse antibody, and the distorted CDR's may not make as
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`13 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT /US89/05857
`
`12
`
`effective contacts with the antigen as the CDR's did in the
`donor antibody;
`(2) Also, amino acids in the original mouse
`antibody that are close to, but not part of, the CDR's (i.e.,
`still part of the framework), may make contacts with the
`antigen that contribute to affinity. These amino acids are
`lost when the antibody is humanized, because all framework
`amino acids are made human.
`To avoid these problems, and to produce humanized
`antibodies that have a very strong affinity for a desired
`antigen, the present invention uses the following four
`criteria for designing humanized immunoglobulins. These
`criteria may be used singly, or when necessary in
`combination, to achieve the desired affinity or other
`characteristics.
`
`Criterion I: As acceptor, use a framework from a p�rticular
`human immunoglobulin that is unusually homologous to the
`donor immunoglobulin to be humanized, or use a consensus
`framework from many human antibodies. For example,
`comparison of the sequence of a mouse heavy (or light) chain
`variable region against human hea-yy (or light) variable
`regions in a data bank (for example, the National Biomedical
`Research Foundation Protein Identification Resource) shows
`that the extent of homology to different human regions varies
`greatly, typically from about 40% to about 60-70%. By
`choosing as the acceptor immunoglobulin one of the human
`heavy (respectively light) chain variable regions that is
`most homologous to the heavy (respectively light) chain
`variable region of the donor immunoglobulin, fewer amino
`acids will be changed in going from the donor immunoglobulin
`to the humanized immunoglobulin. Hence, and again without
`intending to be bound by theory, it is believed that there is
`a smaller chance of changing an amino acid near the CDR's
`that distorts their conformation. Moreover, the precise
`overall shape of a humanized antibody comprising the
`humanized immunoglobulin chain may more closely resemble the
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`14 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT /US89/0S857
`
`13
`
`shape of the donor antibody, also reducing the chance of
`distorting the CDR's.
`Typically, one of the 3-5 most homologous heavy
`chain variable region sequences in a representative
`collection of at least about 10 to 20 distinct human heavy
`chains will be chosen as acceptor to provide the heavy chain
`framework, and similarly for the light chain. Preferably,
`one of the 1-3 most homologous variable regions will be used.
`The selected acceptor immunoglobulin chain will most
`preferably have at least about 65% homology in the framework
`region to the donor immunoglobulin.
`Regardless of how the acceptor immunoglobulin is
`chosen, higher affinity may be achieved by selecting a small
`number of amino acids in the framework of the humanized
`immunoglobulin chain to be the same as the amino acids at
`those positions in the donor rather than in the acceptor.
`The following criteria define what amino acids may be so
`selected. Preferably, at most or all amino acid positions
`satisfying one of these criteria, the donor amino acid will
`in fact be selected.
`
`Criterion II: If an amino acid in the framework of the human
`acceptor immunoglobulin is unusual (i.e., "rare", which as
`used herein indicates an amino acid occurring at that
`position in no more than about 10% of human heavy
`(respectively light) chain V region sequences in a
`representative data bank) , and if the donor amino acid at
`11C0Jnin0n11 I
`that position is typical for human Sequences (j,.,z.,g.:..t
`which as used herein indicates an amino acid occurring in at
`least about 25% of sequences in a representative data bank),
`then the donor amino acid rather than the acceptor may be
`selected. This criterion helps ensure that an atypical amino
`acid in the human framework does not disrupt the antibody
`structure. Moreover, by replacing an unusual amino acid with
`an amino acid from the donor antibody that happens to be
`typical for human antibodies, the humanized antibody may be
`made less immunogenic.
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`15 of 51
`
`BI Exhibit 1050
`
`

`

`W0 90/07861
`
`PCT/US89/05857
`
`14
`
`Criterion III: In the positions immediately adj acent to the
`3 CDR's in the humanized immunoglobulin chain, the donor
`amino acid rather than acceptor amino acid may be selected.
`These amino acids are particularly likely to interact with
`the amino acids in the CDR's and, if chosen from the
`acceptor, distort the donor CDR's and reduce affinity.
`Moreover, the adjacent amino acids may interact directly with
`the antigen (Amit et al. , Science, �' 747-753 (1986), which
`is incorporated herein by reference) and selecting these
`amino acids from the donor may be desirable to keep all the
`antigen contacts that provide affinity in the original
`antibody.
`
`Criterion IV: A 3-dimensional model, typically of the
`original donor antibody, shows that certain amino acids
`outside of the CDR ' s are close to the CDR ' s and have a good
`probability of interacting with amino acids in the CDR's by
`hydrogen bonding, Van der Waals forces, hydrophobic
`interactions, etc. At those amino acid positions, the donor
`amino acid rather than the acceptor immunoglobulin amino acid
`may be selected. Amino acids according to this criterion will
`generally have a side chain atom within about 3 angstrom
`units of some site in the CDR ' s and must contain atoms that
`could interact with the CDR atoms according to established
`chemical forces, such as those listed above. Computer
`programs to create models of proteins such as antibodies are
`generally available and well known to those skilled in the
`art (see, Loew et al. , Int . J. Quant . Chem . , Quant. Biol.
`Symp., 15:55-66 (1988) ; Bruccoleri et al. , Nature, 335,
`564-568 (1988) ; Chothia et al . , Scien