Europaisches Patentamt
`
`(19)
`
`European Patent Office
`
`Office europeen des brevets
`
`lllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll
`EP 0 504 350 81
`
`(11)
`
`EUROPEAN PATENT SPECIFICATION
`
`(12)
`(45) Date of publication and mention
`of the grant of the patent:
`29.07.1998 Bulletin 1998/31
`
`(21) Application number: 91917169.4
`(22) Date of filing: 04.10.1991
`
`(51) Int c1.s: C12N 15/13, CO?K 16/00,
`C12P 21 /08, A61 K 39/395
`(86) International application number:
`
`PCT/GB91/01726
`
`(87) International publication number:
`
`WO 92/06193 (16.04.1992 Gazette 1992/09)
`
`(54) ANTIBODIES DIRECTED AGAINST CD3
`GEGEN CD3 GERICHTETE ANTIKORPER
`ANTICORPS A EFFICACITE ANTIGONISTE A L'ANTIGENE CD3
`(56) References cited:
`(84) Designated Contracting States:
`AT BE CH DE DK ES FR GB GR IT LI LU NL SE
`
`EP·A· 0 336 379
`
`WO-A-91 /09968
`
`(30) Priority: 05.10.1990 GB9021679
`(43) Date of publication of application:
`23.09.1992 Bulletin 1992/39
`
`(73) Proprietor: BRITISH TECHNOLOGY GROUP
`
`LIMITED
`London EC4M 7SB (GB)
`
`(72) Inventors:
`
`• Gorman, Scott David
`Great Shelford, Cambridge CB2 5JN (GB)
`• Routledge, Edward Graham
`Cambridge CB1 4XQ (GB)
`• Waldmann, Herman
`Cambridge CB4 2ED (GB)
`
`(74) Representative: Percy, Richard Keith et al
`
`Patents Department
`British Technology Group Ltd
`10 Fleet Place
`London EC4M 7SB (GB)
`
`• European Journal of lmmunogology, volume 19,
`1989, VCH Verlagsgesellschaft (Weinheim, DE)
`M. Clark et al.: "The improved lytic function and
`In vivo efficacy of monovalent monoclonal CD3
`antibodies"
`• Clinical Chemistry, volume 35, no. 9, 1989, G.P.
`Moore: "Genetically engineered antibodies",
`pages 1849-1853
`• L. Riechmann et al., Nature, 332, 24 March 1988,
`323-327
`• E. G. Routledge et al., Eur. J. lmmunol. 1991, 21:
`2717-2725
`
`Remarks:
`The file contains technical information submitted
`after the application was filed and not included in this
`specification
`
`,...
`m
`0
`l()
`(")
`111::2'
`0
`l()
`0
`Q.
`w
`
`Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give
`notice to the European Patent Office of opposition lo the European patent granted. Notice of opposition shall be filed in
`a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art.
`99(1) European Patent Convention).
`
`Printed by Jouve. 75001 PARIS (FA)
`
`1 of 38
`
`BI Exhibit 1133
`
`

`

`Description
`
`EP 0 504 350 81
`
`5
`
`10
`
`20
`
`25
`
`lhis invention relates to antibodies, in particular to re-shaped antibodies directed against the CD3 antigen on the
`surface of human T-cells.
`Antibodies, or immunoglobulins, comprise two heavy chains linked together by disulphide bonds and two light
`chains, each light chain being linked to a respective heavy chain by disulphide bonds in a "Y" shaped configuration.
`The two •arms• of the antibody are responsible for antigen binding, having regions where the polypeptide structure
`varies. and are termed Fab' fragments (fragment - antigen - binding) or F(ab')2 which represents two Fab' arms linked
`together by disulphide bonds. The •tail" or central axis of the antibody contains a fixed or constant sequence of peptides
`and is termed the Fe fragment (fragment - crystalline). The production of monoclonal antibodies was first disclosed by
`Kohler and Miistein (Kohler & Milstein, Nature. 256. 495-497 (1975)). Such monoclonal antibodies have round wide­
`spread use as diagnostic agents and also in therapy.
`Each heavy chain has at one end a variable domain followed by a number of constant domains. Each light chain
`has a variable domain at one end and a constant domain at its other end, the light chain variable domain being aligned
`15 with the variable domain of the heavy chain and the light chain constant domain being aligned with the first constant
`domain of the heavy chain (CHI). The constant domains in the light and heavy chains are not involved directly in
`binding the antibody to antigen. The light chain constant region and the CHI region of the heavy chain account for
`50% of each Fab' fragment.
`The variable domains of each pair of light and heavy chains form the antigen binding site. The domains on the
`light and heavy chains have the same general structure and each domain comprises four framework regions, whose
`sequences are relatively conserved, connected by three complementarity determining regions (CDAs) (Kabat et al,
`Sequences of Proteins of Immunological Interest, U.S. Department of Health and Human Services (1987)). The four
`framework regions largely adopt a beta-sheet conformation and the CDRs form loops connecting, and in some cases
`forming part of, the beta-sheet structure. The CDRs are held in close proximity by the framework regions and, with the
`CDRs from the other domain, contribute to the formation of the antigen binding site.
`In recent years, molecular biology techniques have allowed the production of a wide range of heterologous polypep­
`tides by transformation of host cells with DNA sequences coding for the desired polypeptide. lmmunoglobulin polypep­
`tides have been produced by recombinant DNA techniques, see for example EP-A-0 088 994 (Schering Corporation),
`EP-A-1 I 02 634 (Takeda Chemical Industries Ltd.) and EP-A-0 125 023 (Genentech Inc.). These techniques have also
`allowed the stable introduction of immunoglobulin genes into myeloma cells.
`When murine or rat monoclonal antibodies or even part human chimeric antibodies (antibodies where the antigen
`binding portion of an immunoglobulin is attached to at least part of another protein by a peptide linkage) comprising a
`mouse or rat variable domain is injected into a human in therapy, the human body's immune system could recognise
`that variable domain as foreign and thus produce an immune response. Hence, upon repeated injections of the mouse
`or rat monoclonal or chimeric antibody into humans, the effectiveness would be lost or reduced by the reaction of the
`body's immune system against the foreign antibody.
`EP-A-0 239 400 (Winter) describes a monoclonal antibody in which only the CDAs of the antibody will be foreign
`to the body in order to minimise side effects due to its antigenicity if used for human therapy. Although. for example,
`human, mouse and rat framework regions have characteristic sequences, there seem to be no characteristic features
`40 which distinguish human from mouse and rat CDAs. Thus, an antibody comprised of mouse or rat CDRs in a human
`framework may well be no more foreign to the body than a genuine human antibody.
`It is not clear however that the method of "humanizing• antibodies described in the above application will be suitable
`for application as a general method to all antibodies. Antibodies have either kappa or lambda light chains and one of
`alpha, mu, gamma, epsilon or delta heavy chains, specific combinations of which may make the above method of
`humanising antibodies inapplicable.
`Until now, all of the humanised antibodies have contained a light chain of the kappa type. However, it has now
`been found possible to humanise an antibody directed against the human T-cell CD3 antigen (the monoclonal antibody
`secreted by the rat hybridoma YTHl2.5.14.2 hereinafter referred to as YTHl2.5), even though the antibody has a
`lambda type light chain. The presence of the lambda light chain required a different approach from that used for the
`humanisation of the mouse monoclonal antibody as described in EP-A-0 239 400.
`Accordingly the present invention comprises a ligand, antibody, or antibody fragment with a binding affinity for the
`CD3 antigen, having a human constant region, a human or rat variable framework region, and at least one comple­
`mentarity-determining region (CDR) selected from the amino acid sequences:
`
`30
`
`35
`
`45
`
`so
`
`55
`
`2
`
`2 of 38
`
`BI Exhibit 1133
`
`

`

`EP 0 504 350 81
`
`(a)
`(b)
`
`( c)
`
`(d)
`
`Ser-Phe-Pro-Met-Ala,
`
`Thr-Ile-Ser-Thr-Ser-Gly-Gly-Arg-Thr-Tyr-Tyr-Arg-Asp-Ser-Val­
`
`Lys-Gly,
`
`Phe-Arg-Gln-Tyr-Ser-Gly-Gly-Phe-Asp-Tyr,
`
`Thr-Leu-Ser-Ser-Gly-Asn-Ile-Glu-Asn-Asn-Tyr-Val-His,
`
`5
`
`10
`
`Asp-Asp-Asp-Lys-Arg-Pro-Asp,
`
`His-Ser-Tyr-Val-Ser-Ser-Phe-Asn-Val,
`
`(c)
`(f)
`and conservatively modified variants thereof.
`The term "conservatively modified variants" is one well known in the art and indicates variants containing changes
`which are substantially without effect on antibody-antigen affinity.
`The CDRs of the invention are situated within framework regions of the heavy chain (for (a), (b) and (c)) and light
`chain (for (d), (e) and (f)) variable domains. The variable framework regions may be of human origin or of rat origin. In
`the latter case the variable framework region may, for example, be derived from the YTH 12.5 cell line. Preferably,
`however, the variable framework and constant regions are both of human origin.
`Ligands according to the invention may contain varying numbers of CDRs. Thus, !or example, the entities known
`as molecular recognition units contain a single CDR, but of rather greater interest among ligands which do not contain
`both a heavy and light chain are the single domain ligands described in European Patent Application No. 0 368 684
`25 which contain three CDRs.
`In a preferred embodiment of the invention, therefore, the ligand has three CDRs corresponding to the amino acid
`sequences (a). (b) and (c) above or conservatively modified variants thereof and/or three CD Rs corresponding to amino
`acid sequences (d), (e) and (f) or conservatively modi tied variants thereof, the heavy chain CDRs (a), (b) and (c) being
`of most importance.
`The present invention is however of particular interest in relation to whole antibodies or fragments thereof contain-
`ing both heavy and light chain variable regions. Thus the ligand of the invention preferably has the form of an antibody
`or fragment thereof with a binding affinity for the CD3 antigen having a heavy chain with at least one CDR selected
`from the amino acid sequences:
`
`15
`
`20
`
`30
`
`35
`
`40
`
`45
`
`50
`
`(a)
`(b)
`(c)
`
`Ser-Phc-Pro-Mct-Ala,
`
`Thr-lle-Ser-Thr-Ser-Gly-Gly-Arg-Thr-Tyr-Tyr-Arg-Asp-Ser-Val-Lys-Gly,
`
`Phc-Arg-GI n-Tyr-Ser-Gly-Gly-Phe-Asp-Tyr,
`
`and conservatively modified variants thereof, and/or a light chain with at least one CDR selected from the amino acid
`sequences:
`
`(d) Thr-Leu-Ser-Ser-Gly-Asn-Ile-Glu-Asn-Asn-Tyr-Val-Hi�.
`
`(e) Asp-Asp-Asp-Lys-Arg-Pro-Asp,
`
`
`
`
`
`( f) lfis-Ser-Tyr-Va 1-Ser-Ser-Phe-Asn-Va 1,
`
`and conservatively modified variants thereof.
`Although as indicated hereinbefore, ligands according to the invention do not have to contain both one or more of
`the specified heavy chain CD Rs and one or more of the specified light chain CDRs. the antibodies or fragments thereof
`55 will usually do so. The CDRs (a), (b) and (c) are arranged in the rat hybridoma YTH12.5 heavy chain in the sequence:
`framework region 1/(a)/framework region 2/(b)/framework region 3/(c)/framework region 4 in a leader-? constant region
`direction and the CDRs (d). (e) and (I) are arranged in the hybridoma light chain in the sequence: framework region
`1/(d)/lramework region 2/(e)/framework region 3/(f)/framework region 4 in a leader-? constant region direction. It is
`
`3
`
`3 of 38
`
`BI Exhibit 1133
`
`

`

`15
`
`20
`
`25
`
`30
`
`5
`
`preferred, therefore, that where all three are present the heavy chain CDRs are arranged in the sequence (a), (b), (c)
`in a leader� constant region direction and the light chain CDRs are arranged in the sequence (d), (e), (f) in a leader
`� constant region direction.
`It should be appreciated that it may be possible to have heavy chains and particularly light chains containing only
`one or two of the CDRs (a), (b} and (c) and (d), (e) and (f}, respectively. However, although the presence of all six
`CDRs defined above is therefore not necessarily required in an antibody or fragment thereof according to the present
`invention, all six CORs will most usually be present. A particularly preferred antibody or fragment thereof therefore has
`a heavy chain with three CDRs comprising the amino acid sequences (a), (b) and (c) or conservatively modified variants
`thereof and a light chain with three CDRs comprising the amino acid sequences (d), (e) and (f) or conservatively
`10 modified variants thereof in which the heavy chain CDRs are arranged in the order (a), (b), (c) in the leader constant
`region direction and the light chain CDRs are arranged in the order (d), (e), (1) In the leader constant region direction.
`The invention may be applied to antibodies having a "Y" shaped configuration which have two identical light and
`two identical heavy chains and are thus bivalent with each antigen binding site having an affinity for the C03 antigen.
`Alternatively, Fab' or F(ab')2 fragments retaining the CDRs may be prepared. The invention is also applicable to anti·
`bodies and, where appropriate, fragments thereof, in which only one of the arms of the antibody has a binding affinity
`for the CD3 antigen. Such antibodies may take various forms. Thus the other arm of the antibody may have a binding
`affinity for an antigen other than CD3 so that lhe antibody is a bispecific antibody, for example as described in U.S.
`Patent No. 4,47 4,893 and European Patent Applications Nos. 87907123.1 and 87907124.9. Alternatively, the antibody
`may have only one arm which exhibits a binding affinity, such an antibody being termed •monovalent".
`Monovalent antibodies (or antibody fragments) may be prepared in a number of ways. Glennie and Stevenson
`(Nature, 295, 712·71 3, (1982)) describe a method of preparing monovalent antibodies by enzymic digestion. Stevenson
`� fil describe a second approach to monovalent antibody preparation in which enzymatically produced Fab' and Fe
`fragments are chemically cross-linked (Anticancer Drug Design, 3, 219-230 (1989)). In these methods the resulting
`monovalent antibodies have lost one of their Fab' arms. A third method of preparing monovalent antibodies is described
`in European Patent No. 131 424. In this approach the "Y" shape of the antibody is maintained, but only one of the two
`Fab' domains will bind to the antigen. This is achieved by introducing into the hybridoma a gene coding for an irrelevant
`light chain which will combine with the heavy chain of the antibody to produce a mixture of products in which the
`monovalent antibody is the one of interest.
`More preferably, however, the monovalent CD3 antibodies of the invention are prepared by a new method. This
`involves the introduction into a suitable expression system, for example a cell system as described hereinafter, together
`with genes coding for the heavy and light chains, of a gene coding for a truncated heavy chain in which the variable
`region domain and first constant region domain of the heavy chain are absent, the gene lacking the exon for each of
`these domains. This results in the production by the cell system of a mixture of (a) antibodies which are complete
`bivalent antibodies, (b) antibody fragments consisting only of two truncated heavy chains (i.e. an Fe fragment) and (c)
`fragments of antibody which are monovalent for the CD3 antigen, consisting of a truncated heavy chain and a light
`chain in association with the normal heavy chain. Such an antibody fragment (c) is monovalent since it has any only
`one Fab' arm. Production of a monovalent antibody in the form of such a fragment by this method is preferred for a
`number of reasons. Thus, the resulting antibody fragment is easy to purify from a mixture of antibodies produced by
`the cell system since, for example, ii may be separable simply on the basis of its molecular weight. This is not possible
`in the method of European Patent No. 131 424 where the monovalent antibody produced has similar characteristics to
`a bivalent antibody in its size and outward appearance. Additionally, the production of a monovalent antibody fragment
`by the new method uses conditions which can more easily be controlled and is thus not as haphazard as an enzyme
`digestion/chemical coupling procedure which requires the separation of a complex reaction product, with the additional
`advantage that the cell line used will continue to produce monovalent antibody fragments, without the need for con-
`tinuous synthesis procedures as required in the enzyme digestion/chemical coupling procedure.
`The CD Rs of the invention correspond to those present in the rat CD3 antibody YTH 12.5.
`For the humanisation of the rat CDRs, certain human variable domain framework sequences will be preferable for
`combination with the CDR sequences according to the invention, since the 3-dimensional conformation of the CDRs
`will be better maintained in such sequences and the antibody will retain a high level of binding affinity for the antigen.
`Desirable characteristics in such variable domain frameworks are the presence of key amino acids which maintain the
`structure of the CDR loops in order to ensure the affinity and specificity of the antibody for the CD3 antigen, the lambda
`type being preferred for the light chain.
`We have identified human variable region frameworks which are particularly suitable for use in conjunction with
`the CDRs of the present invention. The heavy chain variable (V) region frameworks are those coded for by the human
`VH type Ill gene VH26. D.J. which is from the B cell hybridoma cell line 18/2 (Genbank Code: Huminghat, Dersimonian
`� al., Journal of Immunology, 139, 2496-2501). The light chain variable region frameworks are those of the human
`VL A.type VI gene SUT (Swissprot code; LV6CSHum, Solomon � al. In Glenner � .§.] (Eds), Amyloidosis, Plenum Press
`N.Y., 1986, p.449.
`
`EP 0 504 350 81
`
`4
`
`35
`
`40
`
`45
`
`so
`
`55
`
`4 of 38
`
`BI Exhibit 1133
`
`

`

`EP 0 504 350 81
`
`The one or more CDRs of th heay chain of the rat anti-C03 antibody are therefore preferably present in a human
`variable domain framework which has the following amino acid sequence reading in the leader � constant region
`direction, CDR indicating a CDR (a), (b) or (c) as defined hereinbefore, a conservatively modified variant thereof or an
`alternative CDR:-
`
`
`
`Glu-Val-Gln-Leu-Leu-Glu-Ser-Gly-Gly-Gly-Leu-Val-Gln-Pro-Gly-Gly­
`
`Ser-Leu-Arg-Leu-Ser-Cys-Ala-Ala-Ser-Gly-Phe-Thr-Phe-Ser-/CDR/-
`
`1rp-Val-Arg-Gln-Ala-Pro-Gly-Lys-Gly-Leu-Glu-Trp-Val-Ser-/COR/­
`
`
`
`Arg-Phe-Thr-l le-Ser-Arg-Asp-Asn-Ser-Lys-Asn-Thr-Leu-Tyr-Leu-Gln­
`
`
`
`
`Met-Asn-Ser-Leu-Arg-A la-Glu-Asp-Thr-Ala-Val-Tyr-Tyr-Cys-A la-Lys­
`�Val-Ser-Ser.
`/COR/-lrp-Gly-Gln-Gly-Thr-Leu-Val-Thr
`
`In a preferred antibody containing all three CDRs, the heavy chain variable region comprises the following se­
`quence:-
`
`
`
`Glu-Val-Gln-Leu-Leu-Glu-Ser-Gly-Gly-Gly-Leu-Val-Gln-Pro-Gly-Gly­
`
`Ser-Leu-Arg-Leu-Ser-Cys-Ala-Ala-Ser-Gly-Phe-Thr-Phe-Ser-Ser-Phe­
`
`Pro-Met-Ala-Trp-Val-Arg-Gln-Ala-Pro-Gly-Lys-Gly-Leu-Glu-Trp-Val­
`
`Ser-Thr-lle-Ser-Thr-Ser-Gly-Gly-Arg-Thr-Tyr-Tyr-Arg-Asp-Ser-Val­
`
`Lys-Gly-Arg-Phe-Thr-lle-Ser-Arg-Asp-Asn-Ser-Lys-Asn-Thr-Leu-Tyr­
`
`
`
`Leu-Gln-Met-Asn-Ser-Leu-Arg-Ala-Glu-Asp-Thr-Ala-Val-Tyr-Tyr-Cys­
`
`Ala-Lys-Phe-Arg-Gln-Tyr-Ser-Gly-Gly-Phe-Asp-Tyr-Trp-Gly-Gln-Gly­
`
`Thr-Leu-Va l-Thr-Val-Ser-Ser.
`
`Similarly, the one or more CDRs of the light chain or the rat CD3 antibody are therefore preferably present in a
`human variable domain framework which has the following amino acid sequence reading in the leade r � constant
`region direction, CDR indicating a CDR (d), (e) and (I) as de tined hereinbelore, a conservatively modified variant thereof
`or an alternative CDR:-
`
`Asp-Phe-Met-Leu-Thr-Gln-Pro-His-Ser-Val-Ser-Glu-Ser-Pro-Gly-Lys­
`
`Thr-Val-l le-l le-Ser-Cys-/CDR/-Trp-Tyr-Gln-Gln-Arg-Pro-Gly-Arg-A la­
`
`
`
`Pro-Thr-Thr-Val-lle-Phe-/COR/-Gly-Val-Pro-Asp-Arg-Phe-Ser-Gly-Ser­
`
`lle-Asp-Arg-Ser-Ser-Asn-Ser-Ala-Ser-Leu-Thr-Ile-Ser-Gly-Leu-Gln­
`
`Thr-Glu-Asp-Glu-Ala-Asp-Tyr-Tyr-Cys-/COR/-Phe-Gly-Gly-Gly-Thr-Lys­
`
`leu-Thr-Val-Leu.
`
`In a preferred antibody containing all three CD Rs the light chain variable region comprises the following sequence:-
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`5
`
`5 of 38
`
`BI Exhibit 1133
`
`

`

`EP 0 504 350 81
`
`Asp-Phe-Met-Leu-Thr-Gln-Pro-His-Ser-Val-Ser-Glu-Ser-Pro-Gly-Lys­
`Thr-Val-I le-l le-Ser-Cys-Thr-Leu-Ser-Ser-Gly-Asn-lle-Glu-Asn-Asn­
`Tyr-Val-His-Trp-Tyr-Gln-Gln-Arg-Pro-Gly-Arg-Ala-Pro-Thr-Thr-Val­
`I le-Phe-Asp-Asp-Asp-Lys-Arg-Pro-Asp-Gly-Val-Pro-Asp-Arg-Phe-Ser­
`Gly-Ser-Ile-Asp-Arg-Ser-Ser-Asn-Ser-Ala-Ser-Leu-Thr-Ile-Ser-Gly­
`Leu-Gln-Thr-Glu-Asp-Glu-Ala-Asp-Tyr-Tyr-Cys-His-Ser-Tyr-Val-Ser­
`Ser-Phe-Asn-Val-Phe-Gly-Gly-Gly-Thr-Lys-Leu-Thr-Val-Leu.
`
`5
`
`10
`
`15
`
`20
`
`25
`
`35
`
`40
`
`The variable domains comprising one or more CD Rs as described above, preferably in the humanised form having
`human antibody-derived framework regions, may conveniently be al1ached to another protein or carrier, or to constant
`domains of light and heavy chains of antibodies.
`The nature of the heavy and light chain constant regions has less effect on binding affinity than that of the variable
`domain framework and these can be based on antibodies of different types as desired, but are preferably of or are
`derived from those of human origin and may be of various different classes although for the light chain the constant
`region will most usually be of the lambda type and for the heavy chain it may conveniently be of an lgG class, particularly
`lgGI. Thus the constant domains may conveniently be selected to have desired effector functions appropriate to the
`intended therapeutic use of the antibody.
`It will also be appreciated that an antibody according to the invention may be used in a form which retains lhe
`CDRs but lacks other parts of the molecule not essential lo its binding function. In particular as indicated hereinbefore,
`Fab' and F(ab')2 fragments may be used, or the variable regions incorporating the CDRs of the invention may be
`attached to a suitable protein or carrier molecule.
`It is well recognised in the art that the replacement of one amino acid with another amino acid having similar
`properties, for example the replacement of a glutamic acid residue with an aspartic acid residue, may not substantially
`alter the properties or structure of the peptide or protein in which the substitution or substitutions were made. Thus,
`the invention includes those CDR amino acid sequences in which such a substitution or substitutions have occurred
`30 without substantially altering the binding affinity and specificity of the CDRs. Alternatively, deletions may be made in
`the amino acid residue sequence of the CORs or the sequences may be eX1ended at one or both of the N- and C­
`termini whilst still retaining activity.
`As indicated, therefore, the invention eX1ends to ligands in which the CDRs may be conservatively modified to
`provide a variant thereof which retains a binding affinity for the CD3 antigen. Preferred ligands are such that the affinity
`constant for the antigen is 1 OS mole·t or more, for example up to 1012 mole·t. Ligands of different affinities may be
`suitable for different uses so that, for example, an affinity or 1 os, 107 or 1 os mole·1 or more may be appropriate in some
`cases. However ligands with an affinity in the range of 10s to 1 os mole·1 will often be suitable. Conveniently the ligands
`also do not exhibit any substantial binding affinity for other antigens. Binding affinities of the ligand and ligand specificity
`may be tested by assay procedures such as those described in the Examples section hereinafter, (Effector Cell Re-
`targetting Assay), or by techniques such as ELISA and other immunoassays.
`The ligands of the invention may be prepared in a number of ways. Most conveniently, however, appropriate gene
`constructs for the constant and variable regions of the heavy and light chains which are present in the ligand are
`separately obtained and then inserted in a suitable expression system. Antibody fragment may be prepared from whole
`antibody molecules in the usual manner or, as described for monovalent antibody fragments hereinbefore, may be
`produced directly by the expression system.
`Genes encoding the variable domains of a ligand of the desired structure may be produced and conveniently
`attached to genes encoding the constant domains or an antibody of the desired isotype and therapeutic applicability.
`These constant genes may be obtained from hybridoma cDNA or from the chromosomal DNA or by mutagenesis (site
`directed) of such genes to produce constant regions with novel properties. Genes encoding the variable regions may
`also be derived by gene synthesis techniques used in the identification of the CDRs contained herein. Suitable cloning
`vehicles for the DNA may be of various types.
`Expression of these genes through culture of a cell system to produce a functional CD3 ligand is most conveniently
`effected by transforming a suitable prokaryotic or particularly eukaryotic cell system, particularly an immortalised mam­
`malian cell line such as a myeloma cell line, for example the YB2/3.01/Ag20 (hereinafter referred to as YO) rat myeloma
`cell, or Chinese hamster ovary cells (although the use of plant cells is also of interest), with expression vectors which
`include DNA coding for the various antibody regions, and then culturing the transformed cell system to produce the
`desired antibody. Such general techniques of use for the manufacture of ligands according to the present invention
`are well known in the very considerable art of genetic engineering and are described in publications such as "Molecular
`
`45
`
`so
`
`55
`
`6
`
`6 of 38
`
`BI Exhibit 1133
`
`

`

`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`EP 0 504 350 81
`
`Cloning" by Sambrook, Fritsch and Maniatis, Cold Spring Harbour Laboratory Press, 1989 (2nd edition). The techniques
`are further illustrated by the Examples contained herein.
`Accordingly, the invention further comprises DNA sequences encoding the CDRs of the ligand/antibody of the
`invention. A group of nucleotide sequences coding for the CDRs (a) to (f) described hereinbefore is as indicated under
`(a) to (f) below, respectively, but it will be appreciated that the degeneracy of the genetic code permits variations to be
`made in these sequences whilst still encoding for the CDRs' amino acid sequences.
`
`(a) AGCllTCCAA TGGCC
`(b) ACCATTAGTA CTAGTGGTGG TAGAACTTAC TATCGAGACl CCGTGAAGGG C
`(c) TTTCGGCAGT ACAGTGGTGG CTTTGATTAC
`(d) ACACTCAGCT CTGGTAACAT AGAAAACAAC TATGTGCAC
`(e) GATGATGATA AGAGACCGGA T
`(f) CATTCTTATG TTAGlAGTTT TAATGTT
`
`The invention also particularly includes larger DNA sequences which comprise (1) DNA expressing human heavy
`chain framework regions and one or more of (a), (b) and (c), and (2) DNA expressing human light chain framework
`regions and one or more of (d), (e) and (f). A specific example of such DNA is that sequence (1) indicated below which
`codes for the CDRs (a), (b) and (c) arranged in the heavy chain framework coded for by the human VH type Ill gene
`VH26. D.J. as discussed hereinbefore and that sequence (2) indicated below which codes for the CDRs (d), (e) and (f)
`arranged in the light chain framework coded for by the human VLA. type VI gene SUT The CDR sequences (a), (b),
`(c), (d), (e) and (f) have been underlined.
`
`( l)
`
`GAGGTCCAAC lGClGGAGTC TGGGGGCGGT TlAGTGCAGC ClGGAGGGTC
`
`CCTGAGAClC TCCTGTGCAG CCTCAGGATT CACTTTCAGT AGCTTTCCAA
`
`TGGCCTGGGT CCGCCAGGCT CCAGGGAAGG GTCTGGAGTG GGTCTCAACC
`
`ATTAGTACTA GTGGlGGTAG AACTTACTAT CGAGACTCCG TGAAGGGCCG
`
`ATTCACTATC 1CCAGAGA1A ATAGCAAAAA TACCCTATAC CTGCAAATGA
`
`ATAGTCTGAG GGCTGAGGAC ACGGCCGTCl ATTACTGTGC AAAATTTCGG
`
`CAGTACAGTG GTGGCTTTGA TTACTGGGGC CAAGGGACCC TGGTCACCGT
`
`CTCCTCA
`
`(2)
`
`GACHCATGC TGACTCAGCC CCACTCTGTG TCTGAGTCTC CCGGAAAGAC
`
`AGTCATTAH TCTTGCACAC TCAGCTCTGG TAACATAGAA AACAACTATG
`
`TGCACTGGTA CCAGCAAAGG CCGGGAAGAG CTCCCACCAC
`
`GATGATGATA AGAGACCGGA IGGTGTCCCT GACAGGTTCT
`
`TGTGATTTTC
`
`CTGGCTCCAT
`
`TGACAGGTCT TCCAACTCAG CCTCCCTGAC AATCAGTGGT CTGCAAACTG
`
`AAGATGAAGC TGACTACTAC TGTCATTCTT ATGTTAGTAG TTTTAATGTT
`
`TTCGGCGGTG GAACAAAGCT CACTGTCCTT
`
`The humanised ligands in accordance with the invention have therapeutic value. In particular, a reshaped antibody,
`especially a humanised antibody, with a specificity for the antigen CD3 has valuable applications in immunosuppres-
`sion, particularly in the control of graft rejection, and potentially also in other areas such as the treatment of cancer,
`especially of lymphoid malignancies and indeed lymphomas in general.
`In a further aspect. the invention thus includes a method of treating patients with lymphomas or for immunosup­
`pression purposes, for instance in a case where graft rejection may occur, comprising administering a therapeutically
`
`7
`
`7 of 38
`
`BI Exhibit 1133
`
`

`

`EP 0 504 350 81
`
`effective amount of a ligand in accordance with the invention.
`Ligands in accordance with the invention may be formulated for administration to patients by administering the
`said ligand together with a physiologically acceptable diluent or carrier. The ligands are preferably administered in an
`injectable form together with such a diluent or carrier which is sterile and pyrogen free. By way of guidance it may be
`stated that a suitable dose of ligand is about 1-10 mg injected daily over a time period of, for example 10 days. In order
`to avoid a severe first dose response, suitable anti�ytokines may be administered with the first injection. Such a
`procedure facilitates the use of a dosage towards the upper end of the 1-10 mg range or even somewhat higher.
`The invention is illustrated by the following Examples which are illustrated by the drawings listed below:
`
`5
`
`10
`
`DESCRIPTION OF THE DRAWINGS
`
`Figure 1 :
`
`Figure 2:
`Figure 3 :
`
`15
`
`Figure 4:
`Figure 5 :
`Figure 6 :
`
`Figure 7 :
`
`Figure 8 :
`
`Figure 9 �
`
`Figure 10:
`
`EXAMPLES
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`so
`
`55
`
`Position and sequence of the oligonucleotide forward and backward primers used in the cDNA syn­
`thesis and PCR amplification of the rat YTH12.5 VL lambda gene.
`Cloning and reshaping of the YTH 12.5 VH gene.
`Reshaping of the YTH12.5 VL gene and construction of the YTH12.5 immunoglobulin light chain ex­
`pression vector.
`Construction of the reshaped YTH12.5 immunoglobulin heavy chain expression vector.
`Construction of the truncated human lgGI heavy chain (tH) gene expression vector.
`Native PAGE of protein-A purified total immunoglobulin secreted by cells co-transfected with the hu­
`manised CD3 heavy, light and truncated heavy chain gene expression vectors.
`Reduced (7a) and non-reduced (7b) SOS-PAGE of the antibody molecules corresponding to native
`PAGE bands 1 , 2 and 3 (lanes 1, 2 and 3 respectively).
`Humanised bivalent and monovalent CD3 antibodies were tested for their ability to direct T-cell killing
`of Fe receptor-bearing U937 cells. Rat bivalent YTH1 2.5 CD3 monoclonal antibody and the humanised
`CDw52 antibody were tested as controls.
`Comparison or antibody binding ol humanised monovalent and bivalent CD3 antibodies with rat biva­
`lent YTH1 2.5 CD3 monoclonal antibody. The humanised CDw52 antibody was included as a negative
`control.
`Humanised bivalent and monovalent CD3 monoclonal antibodies were tested for their ability to direct
`complement mediated lysis of human T-cell blasts. Rat bivalent YTH12.5 CD3 monoclonal antibody
`was tested for comparison.
`
`The invention is illustrated by the following Examples, which employ techniques such as those described in Mo-
`lecular Cloning by Sambrook et al.
`
`EXAMPLE 1: CULTURE OF RAT HYBRIDOMA AND CHINESE HAMSTER OVARY CELLS
`
`YTH12.5 rat hybridoma cells secreting rat gamma-2b antibody specific for the Epsilon chain of the human CD3
`antigen complex (Clark��. Eur. J. lmmunol., � 381-388 (1989)) were grown or maintained in lscove's modification
`of Dulbecco's medium with antibiotics and 5% bovine foetal calf serum respectively. YO cells, a non-antibody secreting
`rat myeloma cell line were similarly cultured (Clark and Milstein, Somatic Cell Genetics. z (6) 657-666, (1981) and
`European Patent 0043718).
`Chinese hamster ovary (CHO) cells with a dihydrololate reductase negative (dhfr) phenotype were cultured in
`medium supplemented with hypoxanthine and thymidine.
`
`EXAMPLE 2: CLONING THE YTH12.5 HYBRIDOMA IMMUNOGLOBULIN VARIABLE
`
`HEAVY (VH) AND VARIABLE LI GHT (Vl) REGION GENES
`
`YTH12.5 cells were lysed with a solution of guanidine thiocyanate and total RNA was isolated by centrifugation
`through a CsCI cushion (Chirgwin et al, Biochemistry, 18, 5294 (19879)). Messenger RNA was prepared from this by
`affinity chromatography on oligo-dT cellulose (Maniatis et al, Molecular Cloning. A laboratory manual. Published by
`Cold Spri