ZA200500337B - Anty-myelin associated glycoprotein (MAG) antibodies - Google Patents

Description

Antibodies

Field of the Invention . The present: invention relates to altered antibodies that bind to myelirs associated glycoprotein (MAG) and neutralise the function thereof, polynucleotides en coding such antibodies, pharmaceutical formulations containing said antibodies and to the use of such antibodies in the treatmentc and/or prophylaxis of neurological diseases. Other aspects, objects and advantages of the present invention will become apparent from the description below.

Background of the Invention

Stroke is a rmajor cause of death and disability in the Western World.

There is no approv €d therapy for the treatment of stroke other than t-PA which has to be administered within 3h of onset following a CT scan to exclude haemorrhage. To «ate most therapeutic agents directed towards the treatmeant of acute stroke (i.e . neuroprotection), have predominantly involved targeting glutamate receptors and their down stream signalling pathways known to be involved in acute cezll death. However to date these strategies have proved unsuccessful in clin ical trials and are often associated with dose-limiting side effects (Hill & Hachwinski, The Lancet, 352 : (supp! III) 10-14 (1998)). Therefore there is a need for novel approaches directed towards the amelioration of cel death following the cessation of blood flow.

Following the= onset of stroke, some degree of spontaneous functional recovery is observe=d in many patients, suggesting that the brain has the (albeit limited) ability to repair and/or remodel following injury. Agents that have the ’ potential to enhanc-e this recovery may therefore allow intervention to be macie " much later (potentially days) following the onset of cerebral ischaemia. Ageruts which are able to offfer both acute neuroprotection and enhance functional recovery may provicie significant advantages over current potential neuroprotective strategies.

The mechanisms underlying functional recovery are currently unknown.

The sprouting of injured or non-injured axons has been proposed as one possible mechanism. However, although in v#vo studies have shown that treatment of . spinal cord injury or stroke with neurotrophic factors results in enhanced functional recovery and a degree of axonal sprouting, these do not prove a direct link between the degree of axonal sprouting and extent of functional recovery (Jakeman, et al. 1998, Exp. Neurol. 1154 : 170-184, Kawamata et al. 1997, Proc

Natl Acad. Sci. USA., 94:8179-8184, Ribotta, et al. 2000, J Neurosci. 20 : 5144- 5152). Furthermore, axonal sprouting requires a viable neuron. In diseases such as stroke which is associated with exctensive cell death, enhancement of functional recovery offered by a given agent post stroke may therefore be through mechanisms other than axomal sprouting such as differentiation of endogenous stem cells, activation of redundant pathways, changes in receptor distribution or excitability of neurons or glia (Fawcett & Asher, 1999, Brain Res.

Bulletin, 49 : 377-391, Horner & Gage, 2000, Nature 407 963-970).

The limited ability of the central nervous system (CNS) to repair following injury is thought in part to be due to molecules within the CNS environment that have an inhibitory effect on axonal sprouting (neurite outgrowth). CNS myelin is thought to contain inhibitory molecules (Schwab ME and Caroni P (1988) J.

Neurosci. 8, 2381-2193). Two myelin proteins, myelin-associated glycoprotein (MAG) and Nogo have been cloned amd identified as putative inhibitors of neurite outgrowth (Sato S. et al (1989) Bjoc/mem. Biophys. Res. Comm.163,1473-1480;

McKerracher L et al (1994) Neuron 13, 805-811; Mukhopadhyay G et al (1994)

Neuron 13, 757-767; Torigoe K and L_undborg G (1997) Exp. Neurology 150, 254-262; Schafer et al (1996) Neuror# 16, 1107-1113; W0O9522344;

WO09701352; Prinjha R et al (2000) Mature 403, 383-384; Chen MS et al (2000)

Nature 403, 434-439; GrandPre T et al (2000) Nature 403, 439-444; . US005250414A; W0200005364A1; W«O0031235).

Myelin-associated glycoprotein is a cell surface transmembrane molecule expressed on the surface of myelin co nsisting of five extracellular immunoglobulin domains, a single trarmsmembrane domain and an intracellular _2-

domain. MAG expression is restricted to myelinating glia in the CNS and peripheral nervous system (PNS). MAG is thought to interact with neuronal receptor(s) which mediate effects on the neuronal cytoskeleton including . neurofilament phosphorylation and inhibition of neurite outgrowth /n vitro.

Although antagonists of MAG have been postulated as useful for the promotion of axonal sprouting following injury (W09522344, W09701352 and

W09707810), these claims are not supported by /r? vivo data. Furthermore, the role of MAG as an inhibitor of axonal sprouting frorm CNS neurons /n vivo is not proven (Li CM et al (1994) Nature 369, 747-750; Montag, D et al (1994) Neuron 13, 229-246; Lassmann H et al (1997) GLIA 19, 104-110; Li C et al (1998) J.

Neuro. Res. 51, 210-217; Yin X et al (1998) J. Neurosci. 18, 1953-1962; Bartsch

U et al (1995) Neuron 15 1375-1381; Li M et al (1996) 46,404-414).

Antibodies typically comprise two heavy chains linked together by disulphide bonds and two light chains. Each light chain is linked to a respective heavy chain by disulphide bonds. 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 is aligned with the variable domain of the heavy chain.

The light chain constant domain is aligned with the first constant domain of the heavy chain. The constant domains in the light and heavy chains are not involved directly in binding the antibody to antigen.

The variable domains of each pair of light and heavy chains form the antigen binding site. The variable domains on the light and heavy chains have the same general structure and each domain comprises a framework of four regions, whose sequences are relatively conserved, connected by three complementarity determining regions (CDRs) often referred to as hypervariable regions. 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. CDRs and framework regions of antibodies may be determined by reference to Kabat et al ("Sequences of preoteins of immunological interest” US Dept. of Health and Human Services, US Gov ernment Printing Office, 1987). } It has now been found that an anti-MAG monoclonal antibody, described (Pottorak et al (1987) Journal of Cell Biology 105,1893-1899, DeBellard et al (1996) Mol. Cell. Neurosci, 7, 89-101; Tang et al (1997) Mol. Cell. Neurosci. 9, 333-346; Torigoe K and Lundborg G (1997) Exp. Neurofogry 150, 254-262) and commercially available (MAB1567 (Chemicon)) when admanistered either directly into the brain or intravenously following focal cerebral ischaemia in the rat (a mmodel of stroke), provides neuroprotection and enhances Functional recovery. “Therefore anti-MAG antibodies provide potential therapeuti ¢ agents for both acute neuroprotection as well as the promotion of functiormal recovery following stroke. This antibody is a murine antibody. Although mu rine antibodies are

Often used as diagnostic agents their utility as a therapeutic has been proven in

Only a few cases. Their limited application is in part due teo the repeated administration of murine monoclonals to humans usually el icits human immune responses against these molecules. To overcome these intrinsic undesireable properties of murine monoclonals “altered” antibodies desig ned to incorporate regions of human antibodies have been developed and are \well established in the art. For example, a humanised antibody contains complementarity determining regions ("CDR’s") of non human origin and the majority of tie rest of the structure is derived from a human antibody.

The process of neurodegeneration underlies many newtrological disseases/disorders including acute diseases such as stroke, twaumatic brain injury arwd spinal cord injury as well as chronic diseases including A Izheimer's disease, fronto-temporal dementias (tauopathies), peripheral neuropawthy, Parkinson's disease, Huntington's disease and multiple sclerosis. Anti-MMG mabs therefore i} may be useful in the treatment of these diseases/disorders, bby both ameliorating the cell death associated with these diseases/disorders and p.romoting functional recovery.

All publications, both journal and patent, disckosed in this present specification are expressly and entirely incorporated herein by reference. . ‘Brief Summary of the Invention

The invention provides an altered antibody or functional fragment thereof which binds to and neutralises MAG and comprises one or rnore of the following CDR's.

The CDR's are identified as described by Kabat (Kabat et al. (1991) Sequences of protedins of immunological interest; Fifth Edition; US Eepartment of Health and

Human Services; NIH publication No 91-3242. CDRs preferably are as defined by Ka bat but following the principles of protein structure and folding as defined by Ch othia and Lesk, (Chothia et al., (1989) Conforneations of immunoglobulin hypervariable regions; Nature 342, p877-883) it will be appreciated that additional residues may also be considered to be part of the antigen binding regior and are thus encompassed by the present invention.

Light chain CDRs

Heavy chain CDRs

H2 WINTYTGEPTYADDFTG (SEQUENCE ID

SE ii

H3 NPINYYGINYEGYVMDY (SEQUENCE ID co fem

The present invention also relates to an aratibody which binds to the same epitope as an antibody having the CDRs described] above. Competitive inhibition assays are used for mapping of the epitopes on a n antigen. Thus there is also provided an anti-MAG antibody (altered or unaltered) which competitvely inhibits the binding of the altered antibody having the CDRs described supra to MAG, preferably human MAG.

In a further aspect, the present invention provides an altered antibody or functional fragment thereof which comprises a heavy chain variable domain which comprises one or more CDR’s selected from CDRH1, CDRH2 and CDRH3 and/or a light chain variable domain which comprises one or more CDRs selected from CDRL1, CDRL2 and CDRL3.

The invention further provides an altered amti-MAG antibody or functional fragment thereof which comprises: a) a heavy chain variable domain (Vx) which comprises in sequence

CDRH1, CDRH2 and CDRH3, and for b) a light chain variable domain (V, ) which comprises in sequence

CDRL1, CDRL2 and CDRL3

A further aspect of the invention provides a .pharmaceutical composition comprising an altered anti-MAG antibody of the pre sent invention or functional fragment thereof together with a pharmaceutically acceptable diluent or carrier.

In a further aspect, the present invention provides a method of treatment or prophylaxis of stroke and other neurological diseases in a human which comprises administering to said human in need thereof an effective amount of an anti-MAG antibody of the invention or functional fragments thereof.

In another aspect, the invention provides thes use of an anti-MAG antibody } of the invention or a functional fragment thereof in the preparation of a medicament for treatment or prophylaxis of stroke and other neurological diseases.

In a further aspect, the present invention provides a method of inhibiting neurodegeneration and/or promoting functional recavery in a human patient afflicted with, or at risk of developing, a stroke or other neurological disease which comprises administering to said human in need thereof an effective amount of an anti-MAG antibody of the invention or a functional fragment ) thereof.

In a yet further aspect, the inventior provides the use of an anti-MAG antibody of the invention or a functional fraagment thereof in the preparation of a medicament for inhibiting neurodegeneration and/or promoting functional recovery in a human patient afflicted with, or at risk of developing, a stroke and other neurological disease.

Other aspects and advantages of thes present invention are described further in the detailed description and the preferred embodiments thereof.

Description of the Figures

Figure 1: Sequence of a mouse/human chimeric anti-MAG antibody heavy chain (Seq ID No. 27).

Figure 2: Sequence of a mouse/human chimeric anti-MAG antibody light chain (Seq ID No. 28).

Figure 3: Sequence of a mouse/humasn chimeric anti-MAG antibody heavy chain (Seq ID No. 29).

Figure 4: Chimeric anti-MAG antibody binds to rat MAG

Figure 5 Humanised anti-MAG antibody sequences

Figure 6: Humanised anti-MAG antibodies bind to rat MAG

Figure 7: Humanised anti-MAG antibodies bind to rat MAG

Figure 8: Humanised anti-MAG antibodies bind to human MAG

Figure 9: Competition ELISA with mousse and humanised anti-MAG antibodies MAG } Detailed Description of the Invention hy

Anti-MAG Antibody

The altered antibody of the inventior is preferably a monoclonal antibody (mAb) and is preferably chimeric, humanise«d or reshaped, of these humanised is . particularly preferred.

The altered antibody preferably has t he structure of a natural antibody or fragment thereof. The antibody may therefore comprise a complete antibody, a (Fab'), fragment, a Fab fragment, a light ch.ain dimer or a heavy chain dimer.

The antibody may be an 1gG1, 1gG2, IgG3, or IgG4; or IgM; IgA, IgE or IgD or a modified variant thereof. The constant domain of the antibody heavy chain may be selected accordingly. The light chain con-stant domain may be a kappa or lambda constant domain. Furthermore, th=e antibody may comprise modifications of all classes eg IgG dimers, Fc mutants that no longer bind Fc receptors or mediate Clg binding (blocking artibodies) . The antibody may also be a chimeric antibody of the type described in WO86/01533 which comprises an antigen binding region and a non-immunoglosbulin region. The antigen binding region is an antibody light chain variable domain or heavy chain variable domain.

Typically the antigen binding region comprisezs both light and heavy chain variable domains. The non-immunoglobulin region is fused at its C terminus to the antigen binding region. The non-immuncaglobulin region is typically a non- immunoglobullin protein and may be an enzyame, a toxin or protein having known binding specificity. The two regions of this type of chimeric antibody may be connected via a cleavable linker sequence. Irmmunoadhesins having the CDRS as hereinbefore described are also contemplated in the present invention.

The constant region is selected accordi ng to the functionality required.

Normally an IgG1 will demonstrate lytic ability= through binding to complement and/or will mediate ADCC (antibody dependent cell cytotoxicity). An IgG4 will be preferred if an non-cytototoxic blocking antibody is required. However, 1gG4 . antibodies can demonstrate instability in production and therefore is may be more preferable to modify the generally more stable IgGl. Suggested modifications are described in EPO307434 preferred modifications include at positions 235 and 237. The invention therefore provides a lytic or a non-lyEic form of an antibody according to the invention . In a preferred aspect the altered antibody is class IgG, more preferalbly

IgGl.

In preferred formns therefore the antibody of the invention is a full leength non-lytic IgG1 antibody having the CDRs described supra. In most preferred forms we provide a full 1 ength non-lytic IgG1 antibody having the CDRs of

SEQ.I.D.NO:13 and 16 and a ful! length non-lytic IgG1 antibody having the €DRs of SEQ.I.D.NO: 15 and 18.

In a further aspect, the invention provides polynucleotides encoding

CDRH1, CDRH2, CDRH3, CDRL1, CDRL2 and CDRL3. Preferred polynucleotide sequences are

Light chain CDRs

CI

CCAG AAGAACTACCTGGCC

(SEQUIENCE ID NO: 7)

L2 TGGGCCAGCACCCGCGAGAGC (SEQUENCE IDS NO:

Co mem ——

L3 CACCAGTACCTGAGCAGCCTGACC (SEQUENCE

Hr

Heavy chain CDRs : 25

CR

H2 TGGATCAACALCCTACACCGGCGAGCCCACCTAC

GCCGACGACT TCACCGGC

. (SEQUENCE ID NO: 11)

H3 AACCCCATCAACTACTACGGCATCAACTACGAG

GGCTACGTGATGGACTAC (SEQUENCE ID

NO: 12)

In a further aspect of the invention, there is provided a polynucleotide encoding a light chain variable region of an altered anti-MAG antibody including at least one CDR selected from CDRL1, CDRL2 and CDRL3, more preferably including all 3 CDRs in sequence.

In a further aspect of the inwention, there is provided a polynucleotide encoding a heavy chain variable region of an altered anti-MAG antibody including at least one CDR selected from CDRH1, CDRH2 and

CDRH3, more preferably including all 3 CDRs in sequence.

In a particularly preferred aspect, the anti-MAG antibody of the invention is a humanised antibody.

The invention therefore further provides a humanised antibody or functional fragment thereof that birds to and neutralises MAG which comprises a heavy chain variable region comprising one of the following amino acid sequences:-

QVQLVQSGSELKKPGASVKV SCKASGYTFTNYGMNWVRQAPGQGLEWMGWI

NTYTGEPTYADDFTGRFVFSLDTSVSTAYLQISSLKAEDTAVYYCARNPIN

YYGINYEGYVMDYWGQGTLVTVSS (SEQ ID No 13 ) .

QVQLVQSGSELKKPGASVKVSCKASGYTFTNYGMNWVRQAPGQGLEWMGWI

NTYTGEPTYADDFTGRFVFSLDTSVSTAYLQISSLKAEDTAVYFCARNPIN

YYGINYEGYVMDYWGQGTLVTVSS (Sequence ID No 14)

QVQLVQSGSELKKPGASVKVSCKASGYTFTNYGMNWVROAPGQGLEWMGWI

NTYTGEPTYADDFTGRFVFSLDTSVSTAYLQISSLKAEDTATYFCARNPIN

YYGINYEGYVMDYWGQGTLVIVSS (sequence ID No 15)

In a further aspect of the invention there is provided a humanised antibody or functional fragment thereof which binds to MAG which comprises the heavy chain variable re gion of Sequence ID No 13, 14 or 15 together with a light chain variable region comprising amino acid Sequences, Sequence ID No 16, 17, 18, or 19:

DIVMTQSPDSLAVSL GERATINCKSSHSVLYS SNQKNYLAWYQQKPGQPPK

LLIYWASTRESGVPD RFSGSGSGTDFTLTISSLOAEDVAVYYCHQYLSSLT

FGQGTKLEIKRTV (SEQ ID No 1¢)

DIVMTQSPDSLAVSLCGERATINCKSSHSVLYS SNOKNYLAWYQQKPGQPPK

LLIYWASTRESGVPDRFSGSGSGTDFTLTIIN LOAEDVAVYYCHQYLSSLT

FGQGTKLEIKRTV (SEQ ID No 17)

DIVMTQS PDSLAVSL(GERATINCKSSHSVLYSSNQKNYLAWYQQKPGQPPK

LLIYWASTRESGVPDRFSGSGSGTDFTLTISSLHTEDVAVYYCHQYLSSLT

FGQGTKLEIKRTV (SEQ ID No 18)

DIVMTQSPDSLAVSLGERATINCKS SHSVLYSSNQKNYLAWYQQOKPGOPPK

LLTYWASTRESGVPDR FSGSGSGTDFTLTI INLHTEDVAVYYCHQYLSSLT

FGQGTKLEIKRTV (SEQ ID No 19)

In a further aspect of the present invention there is provided a humanised antibody comprising: a heavy chain varia ble fragment comprising SEQ ID No13, 14 or 15 and a constant part: or fragment thereof of a human heavy chain and a light chain variable fragment comprising SEQ ID No 16, 17, 18 or 19 and a constant part or fragment thereof of a human light chain.

Ina preferred aspect the humanised antibody is class 1gG more preferably 19G1. 40

Preferred antibodies of the invention comprise:

Heavy chain variable region com prising Seq ID No 13 and light chain variable region comprising Seq ID No 16;

Heavy chain variable region comprising Seq ID No 13 and light chain ) variable region comprising Seq ID No 17;

Heavy chain variable region comprising Seq ID No 13 and light chain variable region comprising Seq ID No 18;

Heavy chain variable region comprising Seq ID No 13 and light chain variable region comprising Seq ID No 19.

Heavy chain variable region Scomprising eq ID No 14 and light chain variable region comprising Seq ID No 16;

Heavy chain variable region comprising Seq ID No 14 and light chain variable region comprising Seq ID No 17;

Heavy chain variable region comprrising Seq ID No 14 and light chain variable region comprising Seq ID» No 18;

Heavy chain variable region comp rising Seq ID No 14 and light chain variable region comprising Seg ID No 19.

Heavy chain variable region Scom prising eq ID No 15 and light chain variable region comprising Seq ID No 16;

Heavy chain variable region comprising Seq ID No 15 and light chain variable region comprising Seq ID No 17;

Heavy chain variable region comprising Seq ID No 15 and light chain variable region comprising Seq ID No 18;

Heavy chain variable region comprising Seq ID No 15 and light chain variable region comprising Seq ID No 19.

In a further aspect, the invention provides polynucleotides encoding the heavy chain variable region comprising Sequence ID Nos 13, 14 and 15 and light chain variable regions comprising Sequence ID No 16, 17, 18 and 19.

Preferred polynucleotide Sequence encoding the amino acid Sequence SEQ ID

NO13is

CAGGTGCAGCTGETGCAATCTGGCTCTGAGTTGAAGAAGCCTGGGGCCTCA

GTCAAGCTTTCCTGCAAGGCTTCECCATACACCTTCACTAACTACGGCATG

‘ AACTGGGTGCGACAGGCCCCTGCGACAAGCGGCTTGAGTGGATGGGATGGATC

AACACCTACACCGGCGAGCCCACCTACGCCGACGACTTCACCGGCCGGTTT

GTCTTCTCCTTGGACACCTCTGTCAGCACGGCATATCTGCAGATCAGCAGC

CTAAAGGCTGAGGACACTGCCGTETATTACTGTGCGAGRAAACCCCATCAAC

TACTACGGCATCAACTACGAGGGCTACCTGATGGACTACTGGGGCCAGGGC

ACACTAGTCACAGTCTCCTCA (SEQ ID No 20)

Preferred polynucleotide sequence en coding the amino acid Sequence ID No 14 is:

CAGGTGCAGCTGETGCAATCTGGETCTGAGTTGAAGARAGCCTGGGGCCTCA

GTCGAAGGTTTCCTGCAAGGCTTCTI GGATACACCTTCACTAACTACGGCATG

AACTCGGTGCGACAGGCCCCTGGAL.CAAGGGCTTGAGTGGATGGGATGGATC

AACACCTACACCGGCGAGCCCACCTACGCCGACGACTTCACCGGCCGGTTT

GTCTTCTCCTTGGACACCTCTGTCAGCACGGCATATCTGCAGATCAGCAGC

CTAAAGGCTGAGGACACTGCCGTGTATTTCTGTGCGAGAAACCCCATCAAC

TACTACGGCATCAACTACGAGGGEGCTACGTGATGGACTACTGGGGCCAGGGC

ACACTAGTCACAGTCTCCTCA (SEQ ID No 21)

Preferred polynucleotide sequence encoding the amino acid Sequence ID No 15 is:

CAGGTGCAGCTGGCTGCAATCTGGG TCTGAGTTGAAGAAGCCTGGGGCCTCA

CTGAAGETTTCCTGCAAGGCTTCTGGATACACCTTCACTAACTACGGCATG

AACTGGETGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGATGGATC

AACACCTACACCGECGAGCCCACC TACGCCGACGACTTCACCGGCCGGTTT

GTCTTCTCCTTGCACACCTCTGTCAGCACGGCATATCTGCAGATCAGCAGC

CTAAAGGCTCGAGGACACTGCCACCTATTTCTGTGCGAGARACCCCATCAAC

TACTACGGCATCAACTACGAGGGCTACGTGATGGACTACTGGGGCCAGGGC

ACACTAGTCACAGTCTCCTCA (SEQ ID No 22)

Preferred polynucleotide sequence encoding the amino acid Sequence ID No 16 is: 40

GACATCGTGATGACCCAGTCTCCAGACTCCCTGGCTGTGTCTCTGGGCGAG

AGGGCCACCATCAACTGCAAGAGCAGCCACAGCGTGCTGTACAGCAGCAAC

CAGAAGAACTACCTGGCCTGGTACCAGCAGAAACCAGGACAGCCTCCTAAG

CTGCTCATTTACTGGGCATCTACCCGEGGGAATCCGGGGTCCCTGACCGATTC

45 AGTGGCAGCGGGTCTGGGACAGATITCACTCTCACCATCAGCAGCCTGCAG

GCTGAAGATGTCGGCAGTTTATTACTGTCACCAGTACCTGAGCAGCCTGACC

TTTGGCCAGGGGACCAAGCTGCGACGATCAAACGTACGGTG (SEQ ID No : 23)

Preferred polynucleotide sequence encoding SEQ ID No 17 is:

GACATCGTGATGACCCAGTCTCCAGACTCCCTGGCTGTGTCTCTGGGCGAG

AGGGCCACCATCAACTGCAAGAGCAGCCACAGCGTGCTGTACAGCAGCAAC

CAGAAGAACTACCTGGCCTGGTACCAGCAGAAACCAGGACAGCCTCCTAAG

CTGCTCATTTACTGGGCATCTACCCGGGAATCCGGGGTCCCTGACCGATTC

AGTCGCAGCGGGTCTGCGACAGATTTCACTCTCACCATCATCAACCTGCAG

GCTGAAGATGTGGCAGTTTATTACTGTCACCAGTACCTGAGCAGCCTGACC

TTTGGCCAGGGGACCAAGCTGGAGATCAAACGTACGGTG (SEQ ID No 24)

Preferred polynucleotide encoding SEQ ID No 18 is:

GACATCGTGATGACCCAGTCTCCAGACTCCCTGGCTGTGTCTCTGGGCGAG

AGGGCCACCATCAACTGCAAGAGCAGCCACAGCGTGCTGTACAGCAGCAAC

CAGAAGAACTACCTGGCCTGCGTACCAGCAGARACCAGGACAGCCTCCTAAG

CTQCTCATTTACTGGGCATCTACCCGGGAATCCGGGGTCCCTGACCGATTC

ACTGCGCAGCGGETCTECGACAGAT TTCACTCTCACCATCAGCAGCCTGCAC

ACCGAAGATGTGGCAGTTTATTACTGTCACCAGTACCTGAGCAGCCTGACC

TTTCGCCAGGGGACCAAGCTGGAGATCAAACGTACGGTG (SEQ ID No 25)

Preferred polynucleotide encoding SEQ ID No 19 is:

GACATCGTGATGACCCAGTCTCCAGACTCCCTGGCTGTGTCTCTGGGCGAG

AGGGCCACCATCAACTGCAAGAGCAGCCACAGCGTGCTGTACAGCAGCAAC

CAGAAGAACTACCTGGCCTGGTAC CAGCAGARACCAGGACAGCCTCCTAAG

CTGCTCATTTACTGGGCATCTACCCGGGAATCCGGGGTCCCTGACCGATTC

40 AGTGGCAGCGGCTCTGGGACAGAT TTCACTCTCACCATCATCAACCTGCAC

ACCGAAGATGTCGGCAGTTTATTACTGTCACCAGTACCTGAGCAGCCTGACC

TTTGGCCAGGGCGACCAAGCTGGAGATCAAACGTACGGTG (SEQ ID No 26) 45 "Neutralising” refers to inhibition, either total or partial, of MAG function including its binding to neurones and inhibition of neurite outgrowth.

"Altered antibody" refers to a protein encoded by an altered immunoglobulin coding region, which may be obtained by expression in a . selected host cell. Such altered antibodies include engineered antibodies (e.g., chimeric, reshaped, humanized or vectored antibodies) or antibody fragments lacking all or part of an immunoglobulin constant region, e.g., Fv, Fab, or F(ab)? and the like. "Altered immunoglobulin cod ing region” refers to a nucleic acid sequence encoding altered antibody. When the altered antibody is a CDR-grafted or humanized antibody, the sequences that encode the complementarity determining regions (CDRs) from a non-human immunoglobulin are inserted into a first immunoglobulin partner comprising human variable framework sequences.

Optionally, the first immunoglobulin partner is operatively linked to a second immunoglobulin partner. "First immunoglobulin partner" refers to a nucleic acid sequence encoding a human framework or human immunoglobulin variable region in which the native (or naturally-occurring) CDR-encoding regions are replaced by the CDR- encoding regions of a donor antibody. The human variable region can be an immunoglobulin heavy chain, a light chain (or both chains), an analog or functional fragments thereof. Such CDR regions, located within the variable region of antibodies (immunoglobulims) can be determined by known methods in the art. For example Kabat et al. (Sequences of Proteins of Immunological

Interest, 4th Ed., U.S. Department off Health and Human Services, National

Institutes of Health (1987)) disclose rules for locating CDRs. In addition, computer programs are known which are useful for identifying CDR regions/structures. "Second immunoglobulin partmer” refers to another nucleotide sequence encoding a protein or peptide to which the first immunoglobulin partner is fused in frame or by means of an optional conventional linker sequence (i.e., operatively linked). Preferably it is am immunoglobulin gene. The second immunoglobulin partner may include a nucleic acid sequence encoding the entire constant region for the same (i.€., homologous - the first and second altered antibodies are derived from the same source) or an additional (i.e., heterologous) antibody of interest. It may be an immunoglobulin heavy chain OF . light chain (or both chains as part of a single polypeptide). The second immunoglobulin partner is not lirmited to a particular immunoglobulin class or isotype. In addition, the second immunoglobulin partner may comprise part of an immunoglobulin constant region, such as found in a Fab, or F(ab) (i.e., a discrete part of an appropriate human constant region or framework region).

Such second immunoglobulin partner may also comprise a sequence encoding am integral membrane protein exposed on the outer surface of a host cell, e.g., as part of a phage display library, or a sequence encoding a protein for analytical or diagnostic detection, e.g., horseradish peroxidase, p-galactosidase, etc.

The terms Fv, Fc, Fd, Fab, or F(ab), are used with their standard meanings (see, e.g., Harlow et al ., Antibodies A Laboratory Manual, Cold Spring

Harbor Laboratory, (1988)).

As used herein, an "enginesered antibody" describes a type of altered antibody, i.e., a full-length synthestic antibody (e.g., a chimeric, reshaped or humanized antibody as opposed to an antibody fragment) in which a portion of the light and/or heavy chain varia ble domains of a selected acceptor antibody are: replaced by analogous parts from one or more donor antibodies which have specificity for the selected epitopes. For example, such molecules may include antibodies characterized by a humanized heavy chain associated with an unmodified light chain (or chimeric light chain), or vice versa. Engineered antibodies may also be characterized by alteration of the nucleic acid sequences encoding the acceptor antibody ligght and/or heavy variable domain framework regions in order to retain donor antibody binding specificity. These antibodies can comprise replacement of one or more CDRs (preferably all) from the acceptor antibody with CDRs from a donor antibody described herein.

A "chimeric antibody" refers to a type of engineered antibody which contains a naturally-occurring variable region (light chain and heavy chains)

derived from a donor antibody ire association with light and heavy chain constant regions derived from an acceptor antibody.

A "humanized antibody" refers to a type of engineered antibody having its . CDRs derived from a non-human donor immunoglobulin, the remaining immunoglobulin-derived parts of the molecule being derived from one (or more) human immunoglobulin(s). In acidition, framework support residues may be altered to preserve binding affini€y (see, e.g., Queen et al., Proc. Natl Acad Sci

USA, 86:10029-10032 (1989), Hodgson et al., Bio/Technology, 9:421 (1991)). A suitable human acceptor antibody may be one selected from a conventional database, e.g., the KABAT® database, Los Alamos database, and Swiss Protein database, by homology to the nu cleotide and amino acid sequences of the donor antibody. A human antibody cha racterized by a homology to the framework regions of the donor antibody (or an amino acid basis) may be suitable to provide a heavy chain constant region and/or a heavy chain variable framework region for insertion of the donor CDRs. A suitable acceptor antibody capable of donating light chain constant or variable framework regions may be selected in a similar manner. It should be noted that the acceptor antibody heavy and light chains are not required to originate from the same acceptor antibody. The prior art describes several ways of producing such humanised antibodies — see for example EP-A-0239400 and EP-A—054951 "Reshaped human antibody" refers to an altered antibody in which minimally at least one CDR from & first human monoclonal donor antibody is substituted for a CDR in a second human acceptor antibody. Preferrably all six

CDRs are replaced. More preferrably an entire antigen combining region (e.g.,

Fv, Fab or F(ab"), ) from a first human donor monoclonal antibody is substituted for the corresponding region in a second human acceptor monoclonal antibody.

Most preferrably the Fab region from a first human donor is operatively linked to the appropriate constant regions ©f a second human acceptor antibody to form a full length monoclonal antibody.

A "vectored antibody" refers to an antibody to which an agent has been attached to improve transport through the blood brain barrier (BBB). (Review see Pardridge; Advanced Drug Delivery Review 36, 299-321, 1999). The attachment may be chemical or alternatively the moeity can be engineer-ed into the antibody. One example is to make a chimera with an antibody directed towards a brain capilliary endothelial cell receptor eg an anti-insulin rece=ptor antibody or anti-transfersin receptor antibody (Saito et al (1995) Proc. Mati. Acad.

Sci USA 92 10227-31; Pardridge et al (1995) Pharm. Res. 12 807-816;

Broadwell et al (1996) Exp. Neurol. 142 47-65; Bickel et al (1993) Proc ali.

Acad. Sci. USA 90, 2618-2622; Friden et al (1996) J. Pharm. Exp. Ther. 278 1491-1498, US5182107, US5154924, US5833988, US5527527). Once bo und to the receptor, both components of the bispecific antibody pass across the= BBB by the process of transcytosis. Alternatively the agent may be a ligand which binds such cell surface receptors eg insulin, transferrin or low density lipoprote in (Descamps et al (1996) Am. J. Physiol. 270 H1149-H1158; Duffy et al (1.987)

Brain Res. 420 32-38; Dehouck et al (1997) J. Cell Biol. 1997 877-889).

Naturally occuring peptides such as penetratin and SynB1 and Syn B3 winich are known to improve transport across the BBB can also be used (Rouselle eit al (2000) Mol. Pharm.57, 679-686 and Rouselie et al (2001) Journal of

Pharmacology and Exper#mental Therapeutics296,124-131).

The term "donor a ntibody" refers to an antibody (monoclonal, or recombinant) which contributes the amino acid sequences of its variable regions,

CDRs, or other functional fragments or analogs thereof to a first immunoglobulin partner, so as to provide the altered immunoglobulin coding region and reesulting expressed altered antibody with the antigenic specificity and neutralizing activity characteristic of the donor antibody.

The term “acceptor antibody" refers to an antibody (monoclonal, cer recombinant) heterologous to the donor antibody, which contributes all (or any portion, but preferably all) of the amino acid sequences encoding its heawy and/or light chain framew ork regions and/or its heavy and/or light chain constant ~ regions to the first immureoglobulin partner. Preferably a human antibody is the acceptor antibody.

"CDRs" are defined as the complementarity determining region amino acid sequences of an antibody which are the hypervariable regions of immunoglobulin he avy and light chains. See, e.g., Kabat et al., Sequences of Proteins of Immunolog ical

Interest, 4th Ed., U.S. Department of Health and Human Services, National

Institutes of Health (1987). There are three heavy chain and three light chain

CDRs (or CDR regions) in the variable portion of an immunoglobulin. Thurs, "CDRs" as used herein refers to all three heavy chain CDRs, or all three light chain CDRs (or both all Fxeavy and all light chain CDRs, if appropriate). Fhe

Structure and protein fol ding of the antibody may mean that other residue=s are considered part of the antigen binding region and would be understood to be 50 by a skilled person. See for example Chothia et al., (1989) Conformations of immunoglobulin hyperva riable regions; Nature 342, p877-883. For conveniience the CDR's as defined by Kabat in SEQ ID Nos 13-26 are undetlined.

CDRs provide the majority of contact residues for the binding of the antibody to the antigen or epitope. CDRs of interest in this invention are derived from donor antibody variable heavy and light chain sequences, and includes analogs of the naturally occurring CDRs, which analogs also share or retaire the same antigen binding specificity and/or neutralizing ability as the donor antzibody from which they were derived.

A "functional fragment" is a partial heavy or light chain variable sequence (e.g., minor deletions at the amino or carboxy terminus of the immunoglobulin variable region) which retains the same antigen binding specificity and/or neutralizing ability as the antibody from which the fragment was derived.

An "analog" is an amino acid sequence modified by at least one amirao acid, wherein said modification can be chemical or a substitution or a rearrangement of a few armino acids (i.e., no more than 10), which modification permits the amino acid seqquence to retain the biological characteristics, e.g. , antigen specificity and high affinity, of the unmodified sequence. For example, (silent) mutations can be constructed, via substitutions, when certain endonuclesase restriction sites are created within or surrousnding CDR-encoding regions. The present invention contemplates the use of a nalogs of the antibody of the invexntion. It is well known that minor changes in a mino acid or nucleic ) acid sequeances may lead eg to an allelic form of the origirmal protein which retains substantially similar properties. ~~ Thus analogs off the antibody of the invention i ncludes those in which the CDRs in the hyperva riable region of the heavy and light chains are at least 80% homologous, prefearably at least 90 % homologous and more preferably at least 95 % homologowus to the CDRs as defined above as CDRH1, CDRH2, CDRH3, CDRL1, CDRL2 and CDRL3 and retain

MAG neutralising activity. Amino acid sequences are are aat least 80% homologous if they have 80% identical amino acid residue=s in a like position when the sequences are aligned optimally, gaps or insertions being counted as non-identical residues. The invention also contemplates armalogs of the antibodies of the inve=ntion wherein the framework regions are at least 80%, preferably at least 90% and more preferably at least 95% homologous &o the framework regions set forth in Seq ID 1 - 5. Amino acid sequences are at least 80% homologous if they have 80% identical amino acid residue=s in a like position when the sequences are aligned optimally, gaps or insertions being counted as non-identical residues.

Ana logs may also arise as allelic variations. An "alle=lic variation or modification” is an alteration in the nucleic acid sequence. Such variations or modificatio-ns may be due to degeneracy in the genetic cocde or may be deliberately engineered to provide desired characteristics. These variations or modificatio ns may or may not result in alterations in any esncoded amino acid sequence.

The term “effector agents” refers to non-protein carrier molecules to which the altered antibodies, and/or natural or synthetic light or heavy chains of the donor antibody or other fragments of the donor antiboady may be associated by convent ional means. Such non-protein carriers can inclaude conventional carriers used in the diagnostic field, e.g., polystyrene or other plastic beads,

polysaccharides, e.g., as used in the BIAcore [Pharmacia] system, or other non- protein substances useful in the medical field and safe for administration to humans amd animals. Other effector agents may include & macrocycle, for chelating a heavy metal atom, or radioisotopes. Such effector agents may also be useful to increase the half-life of the altered antibodies, e.g., polyethylene glycol.

A neutralising antibody specific for MAG has been described (Poltorak et al (1987) Journal of Cell Biology 105,1893-1899, DeBellard ext al (1996) Mol. Cell.

Neurosci, 7, 89-101; Tang et al (1997) Mol. Cell. Neurosci- 9, 333-346; Torigoe K and Lundborg G (1997) Exp. Neurology 150, 254-262) and is commercially available (MAB1567 (Chemicon)).

Alternatively, one can construct antibodies, altered antibodies and fragments, by immunizing a non-human species (for example, bovine, ovine, monkey, chicken, rodent (e.g., murine and rat), etc.) to ge=nerate a desirable immunoglo bulin upon presentation with native MAG from any species against which antibodies cross reactive with human MAG can be generated, eg human or chicken. . «Conventional hybridoma techniques are employed to provide a hybridoma cell line secreting a non-human mAb to MAG. Such hybridomas are then screened for binding using MAG coated to 384- or 96—well plates, with biotinylated MAG bound to a streptavidin coated plate. or ir a homogenous europium-APC linked immunoassay using biotinylated MAG-

A na-tive human antibody can be produced in a human antibody mouse such as the “Xenomouse” (Abgenix) where the mouse imm unoglobulin genes have been removed and genes encoding the human immuroglobulins have been inserted into the mouse chromosome. The mice are immun ised as normal and develop an antibody reponse that is derived from the human genes. Thus the mouse produces human antibodies obviating the need to humanize the after selection of positive hybridomas. (See Green L.L., J Immurrol Methods 1999 Dec 10;231(1-2):11-23)

The present invention also includes the use of Fab fragments or F(ab')2 fragments dierived from mAbs directed against MAG. Theses fragments are useful as agents protective /n vivo. A Fab fragment contains the entire light ch ain and amino termimal portion of the heavy chain; and an F(ab')y fragment is tie fragment foramed by two Fab fragments bound by disulfide bonds. Fab . fragments amd F(ab’) fragments can be obtained by conventional means, e.g., cleavage of rmAb with the appropriate proteolytic enzymes, papain and/o-r pepsin, or by recombinant methods. The Fab and F(ab"), fragments are useful themselves a s therapeutic or prophylactic, and as donors of sequences irciuding the variable regions and CDR sequences useful in the formation of recormabinant or humanized antibodies as described herein.

The Fab and F(ab")? fragments can also be constructed via a combinatorial phage library (see, e.g., Winter et al., Ann. Rev. Immunol. - 12:433-455 (21994)) or via immunoglobulin chain shuffling (see, e.g., Marks et al., Bio/Techrmology, 10:779-783 (1992), which are both hereby incorporaeted by reference in their entirety.

Thus h uman antibody fragments (Fv, scFv, Fab) specific for MAG can be isolated using human antibody fragment phage display libraries. A library of bacteriophages particles, which display the human antibody fragment proteins, are panned against the MAG protein. Those phage displaying antibody framgments that bind the EMAG are retained from the library and clonally amplified. Th-e human antibody genes are then exicised from the specific bacteriophage and inserted into leuman IgG expression constructs containing the human IgG constant regio ns to form the intact human IgG molecule with the variable regions from the isolated bacteriophage specific for MAG.

The doror antibodies may contribute sequences, such as variable fmeavy and/or light chmain peptide sequences, framework sequences, CDR sequences, : 30 functional frag ments, and analogs thereof, and the nucleic acid sequences encoding thera, useful in designing and obtaining various altered antibodies which are characterized by the antigen binding specificity of the donor antibody.

Taking i nto account the degeneracy of the genetic code, various coding sequences may’ be constructed which encode the variable heavy and light chain amino acid seq uences, and CDR sequences as well as functional fragments and analogs thereof which share the antigen specificity of the donor ant ibody.

Isolated nucleic aci d sequences, or fragments thereof, encoding the variable chain peptide sequences or CDRs can be used to produce altered aratibodies, e.g., chimeric or husmanized antibodies, or other engineered antibod ies when operatively combined with a second immunoglobulin partner.

Altered immunoglobulin molecules can encode altered antibodies which include engineered antibodies such as chimeric antibodies and huma nized antibodies. A desired altered immunoglobulin coding region contains CDR- encoding regions that encode peptides having the antigen specificity of an anti-

MAG antibody, preferably a high affinity antibody, inserted into a first immunoglobulin partner (a human framework or human immunoglobulin variable region).

Preferably, the first immunoglobulin partner is operatively linke=d to a second immunoglobulin partner. The second immunoglobulin partner is defined above, and may include a sequence encoding a second antibody region of interest, for example an Fc region. Second immunoglobulin partners rmay also include sequences erscoding another immunoglobulin to which the lighet or heavy chain constant regiore is fused in frame or by means of a linker sequerace.

Engineered antibodies directed against functional fragments or analogs of MAG may be designed to elicit enhanced binding.

The second immunoglobulin partner may also be associated witl effector agents as defined above, including non-protein carrier molecules, to which the second immunoglobulin partner may be operatively linked by conventional means.

Fusion or linkagie between the second immunoglobulin partners, e.q., antibody sequences, and the effector agent may be by any suitable means, e.g., by conventional covalent or ionic bonds, protein fusions, or hetero-bifurctional cross-linkers, e.g., carbodiimide, glutaraldehyde, and the like. Such tecahniques are known in the art amd readily described in conventional chemistry and biochemistry texts.

Additionally, conventional linker sequences wvhich simply provide for a desired amount of space between the second immunoglobulin partner and the effector agent may also be constructed into the alteered immunoglobulin coding . region. The design of such linkers is well known tos those of skill in the art. In further aspects of the invention we provide diabodiees (bivalent or bispecific), 1 0 triabodies, tetrabodies and other multivalent scFV protein species having one or more CDRs as described supra that bind to and neutralise MAG function.

In still a further embodiment, the antibody of the invention may have attached to it an additional agent. For example, thes procedure of recombinant

DNA technology may be used to produce an engine ered antibody of the invention 1% in which the Fc fragment or CH2-CH3 domain of a complete antibody molecule has been replaced by an enzyme or other detectable molecule (i.e., a polypeptide effector or reporter molecule).

The second immunoglobulin partner may also be operatively linked to a non-immunoglobulin peptide, protein or fragment thereof heterologous to the

CDR-containing sequence having the antigen specifi city of anti-MAG antibody.

The resulting protein may exhibit both anti-MAG antigen specificity and characteristics of the non-immunoglobulin upon expression. That fusion partner characteristic may be, e.g., a functional characteristic such as another binding or receptor domain, or a therapeutic characteristic if th e fusion partner is itself a therapeutic protein, or additional antigenic characteristics.

Another desirable protein of this invention maay comprise a complete antibody molecule, having full length heavy and light chains, or any discrete fragment thereof, such as the Fab or F(ab) fragme.nts, a heavy chain dimer, or any minimal recombinant fragments thereof such as an Fy or a single-chain antibody (SCA) or any other molecule with the same specificity as the selected donor mAb. Such protein may be used in the form o=f an altered antibody, or may be used in its unfused form.

Whenever the second immunoglobulin partner is derived from an antibody different from the donor antibody, e.g., any isotype or class of immunoglobulin framework or constant regions, an engineered antiboedy results. Engineered antibodies can comprise immunoglobulin (Ice) constant regions and variable framework regions from one source, e.g., thee acceptor antibody, and one or more (preferably all) CDRs from the donor antibody. In addition, alterations, e.g., deletions, substitutions, or additions, off the acceptor mAb light and/or heavy variable domain framework region at the nucleic acid or amino acid levels, or the donor CDR regions may be made in o rder to retain donor antibody antigen binding specificity.

Such engineered antibodies are desig ned to employ one (or both) of the variable heavy and/or light chains of the ant1-MAG mAb or one or more of the heavy or light chain CDRs. The engineered antibodies may be neutralising, as above defined.

Such engineered antibodies may inclusde a humanized antibody containing the framework regions of a selected human #mmunoglobulin or subtype, or a chimeric antibody containing the human heavy and light chain constant regions fused to the anti-MAG antibody functional fragments. A suitable human (or other animal) acceptor antibody may be one selected from a conventional database, e.g., the KABAT® database, Los Alamos database, and Swiss Protein database, by homology to the nucleotide and amino ac id sequences of the donor antibody.

A human antibody characterized by a homology to the framework regions of the donor antibody (on an amino acid basis) mays be suitable to provide a heavy chain constant region and/or a heavy chain wariable framework region for insertion of the donor CDRs. A suitable acce ptor antibody capable of donating light chain constant or variable framework re-gions may be selected in a similar manner. It should be noted that the acceptor antibody heavy and light chains are not required to originate from the same acceptor antibody.

Desirably the heterologous framework. and constant regions are selected from human immunoglobulin classes and isotypes, such as IgG (subtypes 1 through 4), IgM, IgA, and IgE. However, the acceptor antibody need not comprise only human immunoglobulin proteir sequences. For instance a gene may be constructed in which a DNA sequence encoding part of a human 3% immunoglobulin chain is fused to a DNA sequeence encoding a non-

immunoglobulin amino acid sequence such as a polypeptide effector or reporter molecule.

Preferably, in a humanized antibody, the variable domains in both human heavy and light chains have been engineere=d by one or more CDR replacements.

Itis possible to use all six CDRs, or various «combinations of less than the six

CDRs. Preferably all six CDRs are replaced. lt is possible to replace the CDRs only in the human heavy chain, using as lighrmt chain the unmodified light chain from the human acceptor antibody. Alternatively, a compatible light chain may be selected from another human antibody by’ recourse to the conventional antibody databases. The remainder of the engineered antibody may be derived from any suitable acceptor human immunoglobulin.

The engineered humanized antibody thus preferably has the structure of a natural human antibody or a fragment thereof, and possesses the combination of properties required for effective therapeutic use.

It will be understood by those skilled in- the art that an engineered antibody may be further modified by changes in variable domain amino acids without necessarily affecting the specificity and high affinity of the donor antibody (i.e., an analog). It is anticipated that heavy and light chain amino acids may be substituted by other amino acids either in the variable domain frameworks or CDRs or both.

In addition, the constant region may be altered to enhance or decrease selective properties of the molecules of the instant invention. For example, dimerization, binding to Fc receptors, or the abidity to bind and activate complement (seg, e.g., Angal et al., Mol. Immunol, 30:105-108 (1993), Xu et al,

J. Biol. Chem, 269:3469-3474 (1994), Winter et al., EP 307,434-B).

An altered antibody which is a chimeric antibody differs from the humanized antibodies described above by providing the entire non-human donor antibody heavy chain and light chain variable regions, including framework regions, in association with immunoglobulin constant regions from other species, preferably human for both chains.

Preferably, the variable light a nd/or heavy chain sequences and the CDRs of suitable donor mAbs, and their encoding nucleic acid sequences, are utilized in the construction of altered antibod ies, preferably humanized antibodies, of this invention, by the following process. The same or similar techniques may also be employed to generate other embodiments of this invention.

A hybridoma producing a selected donor mAb is conventionally cloned, and the DNA of its heavy and light chain variable regions obtained by techniques known to one of skill in the art, e.g., the techniques described in Sambrook ef al, (Molecular Cloning (A Laboratory Manual), 2nd edition, Cold Spring Harbor

Laboratory (1989)). The variable heavy and light regions containing at least the CDR-encoding regions and those portions of the acceptor mAb light and/or heavy variable domain framework regions required in order to retain donor mAb binding specificity, as well as the remaining immunoglobulin-derived parts of the antibody chain derived from a human immunoglobulin are obtained using polynucleotide primers and reverse tra nscriptase. The CDR-encoding regions are identified using a known database and by comparison to other antibodies.

A mouse/human chimeric antibody may then be prepared and assayed for binding ability. Such a chimeric antibody contains the entire non-human donor antibody VH and V|_ regions, in association with human Ig constant regions for both chains.

Homologous framework regions of a heavy chain variable region from a human antibody may be identified using computerized databases, e.g., KABAT®, and a human antibody having homology to the donor antibody will be selected as the acceptor antibody. A suitable light chain variable framework region can be designed in a similar manner.

A humanized antibody may be derived from the chimeric antibody, or preferably, made synthetically by inserting the donor mAb CDR-encoding regions from the heavy and light chains appropriately within the selected heavy and light chain framework. Alternatively, a humanized antibody can be made using standard mutagenesis techniques. Thus, the resulting humanized antibody contains human framework regions and donor mAb CDR-encoding regions.

There may bre subsequent manipulation of framework residues. The resulting humanized a ntibody can be expressed in recombinant host cells, e.g., COS, CHO or myeloma ells.

A conwentional expression vector or recombinant plasmid is produced by placing these coding sequences for the antibody in operative association with conventional regulatory control sequences capable of controlling the replication and expression in, and/or secretion from, a host cell. Regulatory sequences include promoter sequences, e.g., CMV promoter, and signal sequences, which can be derived from other known antibodies. Similarly, a second expression vector can be produced having a DNA sequence which encode s a complementary antibody light or heavy chain. Preferably this second expression vector is identical to thee first except insofar as the coding sequences and selectable markers are concerned, so to ensure as far as possible that each polypeptide chain is functionally expressed. Alternatively, the heavy and light chain coding sequences for the altered antibody may reside on a single vector.

A selected host cell is co-transfected by conventional techniques with both the first and second vectors (or simply transfected by a single wector) to create the transfected host cell of the invention comprising both the recombinant or synthetic light and heavy chains. The transfected cell is then cultured by conventional techniques to produce the engineered antibody of the invention.

The humanized antibody which includes the association of both the recombinant heavy chain an dor light chain is screened from culture by appropriate assay, such as ELISA or RIA. Similar conventional techniques may be employed to construct other altered antibodies and molecules.

Suitable wectors for the cloning and subcloning steps employed in the methods and construction of the compositions of this invention rmay be selected by one of skill ir the art. For example, the conventional pUC series of cloning vectors, may bes used. One vector, pUC19, is commercially available from supply houses, such as Amersham (Buckinghamshire, United Kingdom) or Pharmacia (Uppsala, Sweden). Additionally, any vector which is capable of replicating readily, has an abundance of cloning sites and selectable genes (e.g., antibiotic resistance), and is easily manipulated may be used for cloning. Thus, tine selection of the clo ning vector is not a limiting factor in this invention.

Similarly, thes vectors employed for expression of the antibodies ranay be . selected by one of skill in the art from any conventional vector. The vectors also contain selected re gulatory sequences (such as CMV promoters) which clirect the replication and exp ression of heterologous DNA sequences in selected hsost cells.

These vectors contain the above described DNA sequences which code for the antibody or altered immunoglobulin coding region. In addition, the vect-ors may incorporate the selected immunoglobulin sequences modified by the insexrtion of desirable restriction sites for ready manipulation.

The express@ion vectors may also be characterized by genes suitable for amplifying expressi-on of the heterologous DNA sequences, e.g., the marmmalian dihydrofolate reductase gene (DHFR). Other preferable vector sequences include a poly A signal sequence, such as from bovine growth hormone (BGH) a nd the betaglobin promote=r sequence (betaglopro). The expression vectors useful herein may be synt hesized by techniques well known to those skilled in this art.

The components of such vectors, e.g. replicons, selection genes, enhancers, promoters, signal sequences and the like, may be obtained from commercial or natuTal sources or synthesized by known procedures for Lmse in directing the expression and/or secretion of the product of the recombinant DNA in a selected host. Other appropriate expression vectors of which numer-ous types are known in the art for mammalian, bacterial, insect, yeast, and faungal expression may also be selected for this purpose.

The present invention also encompasses a cell line transfected witch a recombinant plasmi«d containing the coding sequences of the antibodies or altered immunoglobeulin molecules thereof. Host cells useful for the clon@ing and other manipulations: of these cloning vectors are also conventional. However, most desirably, cellss from various strains of £. coli are used for replicatiomn of the cloning vectors and other steps in the construction of altered antibodies of this invention.

Suitable host cells or cell lines for the expression of the antibody of the invention are preferably mammalian cells such as NSO, Sp2/0, CHO, COS, a fibroblast cell (e.g., 3T3), and meyeloma cells, and more preferably a CHO or a myeloma cell. Human cells may" be used, thus enabling the molecule to be modified with human glycosylation patterns. Alternatively, other eukaryotic ce ll lines may be employed. The sel ection of suitable mammalian host cells and methods for transformation, culture, amplification, screening and product production and purification are known in the art. See, e.g., Sambrook et al, cited above.

Bacterial cells may prove useful as host cells suitable for the expression of the recombinant Fabs of the present invention (see, e.g., Plickthun, A,,

Immunol. Rev., 130:151-188 (1992)). However, due to the tendency of proteins expressed in bacterial cells to be in an unfolded or improperly folded form or ire a non-glycosylated form, any recormbinant Fab produced in a bacterial cell wouid have to be screened for retentior of antigen binding ability. If the molecule expressed by the bacterial cell was produced in a properly folded form, that bacterial cell would be a desirable host. For example, various strains of £. coli used for expression are well-known as host cells in the field of biotechnology.

Various strains of B. subtilis, Streptomyces, other bacilli and the like may also be employed in this method.

Where desired, strains of yeast cells known to those skilled in the art ares also available as host cells, as well as insect cells, e.g. Drosophila and

Lepidoptera and viral expression systems. See, e.g. Miller ef al, Genetic

Engineering, 8:277-298, Plenum Press (1986) and references cited therein.

The general methods by wrhich the vectors may be constructed, the transfection methods required to produce the host cells of the invention, and culture methods necessary to produce the altered antibody of the invention frorm such host cell are all conventional techniques. Likewise, once produced, the antibodies of the invention may b e purified from the cell culture contents according to standard procedures of the art, including ammonium sulfate precipitation, affinity columns, cot umn chromatography, gel electrophoresis and the like. Such techniques are within the skill of the art and do not limit this invention. For example, preparation of altered antibodies are described in WO 99/58679 and WO 96/16990.

Yet another method of expression of the antibodies may utilize expression in a transgenic animal, such as described in U. S. Patent No. 4,873,316. This relates to an expression system rising the animal's casein promoter which when transgenically incorporated into a mammal permits the female to produce the desired recombinant protein in its milk.

Once expressed by the dessired method, the antibody is then examined for in vitro activity by use of an appropriate assay. Presently conventional ELISA assay formats are employed to assess qualitative and quantitative binding of the antibody to MAG. Additionally, other /n vitro assays may also be used to verify neutralizing efficacy prior to subs-equent human clinical studies performed to evaluate the persistence of the antibody in the body despite the usual clearance mechanisms.

The therapeutic agents of this invention may be administered as a prophylactic or post injury, or as «otherwise needed. The dose and duration of treatment relates to the relative cluration of the molecules of the present invention in the human circulatior, and can be adjusted by one of skill in the art depending upon the condition beiing treated and the general health of the patient.

The mode of administratio n of the therapeutic agent of the invention may be any suitable route which delivers the agent to the host. The antagonists and antibodies, and pharmaceutical compositions of the invention are particularly useful for parenteral administration, i.e., subcutaneously, intramuscularly, intravenously, or intranasally.

Therapeutic agents of the invention may be prepared as pharmaceutical compositions containing an effectzive amount of the antagonist or antibody of the invention as an active ingredient in a pharmaceutically acceptable carrier. In the prophylactic agent of the invention, an aqueous suspension or solution containing the engineered antibody, preferably buffered at physiological pH, in a

> form ready for injection is preferred. The compositions for parenteral admimnistration will commonly comprise a solution of the anstagonist or antibody of ’ the invention or a cocktail thereof dissolved in an pharmaceutically acceptable

Carriesr, preferably an aqueous carrier. A variety of aqueous carriers may be employed, e.g., 0.9% saline, 0.3% glycine, and the like. These solutions are sterile and generally free of particulate matter. These solutions may be sterilized by conventional, well known sterilization techniques (e.qg., filtration). The compasitions may contain pharmaceutically acceptable auxiliary substances as required to approximate physiological conditions such as pH adjusting and buffering agents, etc. The concentration of the antagonist or antibody of the invention in such pharmaceutical formulation can vary widezly, i.e., from less than about 0.5%, usually at or at least about 1% to as much as 15 or 20% by weight and will be selected primarily based on fluid volumes, viscosities, etc., according to the particular mode of administration selected.

Thus, a pharmaceutical composition of the invention for intramuscular injection could be prepared to contain 1 mL sterile buffered water, and between about 1 ng to about 100 mg, e.g. about 50 ng to about 30 rng or more preferably, about 5 mg to about 25 mg, of an antagonist or antibody of the invention. Similarly, a pharmaceutical composition of the invention for intravexnous infusion could be made up to contain about 250 ml of sterile Ringer's solution, and about 1 to about 30 and preferably 5 mg to about 25 mg of an engineered antibody of the invention. Actual methods for preparing parenterally administrable compositions are well known or will be apparent to those skilled in the art and are described in more detail in, for example, Rernington's

Pharmaceutical Science, 15th ed., Mack Publishing Company, Easton,

Pennsylvania.

Itis preferred that the therapeutic agent of the invention, when in a pharmaceutical preparation, be present in unit dose forms. The appropriate therapeutically effective dose can be determined readily by those of skill in the art. To effectively treat stroke and other neurological diseases in a human, one dose of up to 700 mg per 70 kg body weight of an antagonist or antibody of this -32._

invention should be administered parenterally, preferably iv. or i.m. (intramuscularly). Such dose may, if necessary, be repeated at appropriate time intervals selected as appropriate by a physician.

The antibodies described herein can be lyophilized for storage and reconstituted in a suitable carrier prior to use. This technique has been shown to be effective with conventional immunoglobulins and art-known lyophilization and reconstitution techniques can be employed.

In another aspect, the invention provides a pharmaceutical composition comprising anti-MAG antibody of the present invention or a functional fragment thereof and a pharmaceutically acceptable carrier for treatment or prophylaxis of stroke and other neurological diseases.

In a yet further aspect, the invention provides a pharmaceutical composition comprising the anti-MAG antibody of the present invention or a functional fragment thereof and a pharmaceutically acceptable carrier for inhibiting neurodegeneration and/or promoting functional recovery in a human patient suffering, or at risk of developing, a stroke or other neurological disease.

The following exam ples illustrate the invention.

Example 1 — Anti-MAG antibody in stroke model

Materials and Methods Anti-MAG monoclonal antibody

Anti-MAG monoclonal antibody was mouse anti-chick MAG antibody MAB 1567 obtained from Chemicon. The antibody has the following characteristics:

Antigen: myelin-associated glycoprotein (human, mouse, rat, bovine, chick, frog)

Isotype: 1gG1

Neutralising ability: see DeBellard et al (1996) Mol. Cell. Neurosci. 7, 89- 101; Tang et al (1997) Mo/. Cell. Neurosci. 9, 333-346; Torigoe K and Lundborg

G (1997) Exp. Neurology 1 50, 254-262.

Control IgG1 mab was purchased from R+D Systems.

S Intra-cerebral ventricular cannulation (for stuc?y 1 only)

Under halothane anaesthesia intra-cerebral ventricular (i.c.v.) cannulae were positioned in the left lateral cerebral ventricle (coordinates :1.6mm from the midline, 0.8mm caudal from bregma, 4.1m m from skull surface, incisor bar — 3.2mm below zero according to Paxinos and Watson, 1986) All rats were singly housed to avoid damage to the guide or dummy cannula. 7 days following surgery, correct cannula placement was verified by an intense drinking response to Angiotensin II (100ng, Simpson, et al. 1978). Nine days later, animals underwent cerebral ischaemia.

Transient Focal Cerebral Ischaemia

Transient (90 min) focal cerebral ischaernia was induced in male Sprague

Dawley rats, each weighing between 300-350g . The animals were initially anaesthetised with a mixture of 5% halothane, 60% nitrous oxide and 30% oxygen, placed on a facemask and anaesthesia subsequently maintained at 1.5% halothane. Middle cerebral artery occlusion (M<CAQ) was carried out using the intraluminal thread technique as described previously (Zea Longa, et. al., 1989).

Animals were maintained normothermic throug hout the surgical procedure, allowed to recover for 1h in an incubator, before being singly housed. Only those animals with a neurological score of 3 1h post-occlusion were included in the study (as assessed using a 5-point scoring system: 0, no deficit; 1, contralateral reflex; 2, weakened grip; 3, circling; 4, immobile; 5, dead). Animals were maintained for up to 1 week at which time animals were killed by transcardial perfusion of 0.9% saline followed by 4% paraformaldehyde in 100mM phosphate buffer. The brains were post-fixed in 4% paraformaldehyde at 4°C for 48h at which time they were removed from the skulls and cut into 2mm blocks using a rat brain matrix. The 2mm sections were then paraffin embedded using a Shandon Citadel 1000 tissue processor, cut into 6um sections using a microtome and mounted on poly-L-lysiree coated slides. Sections were then processed for Cresyl Fast Violet (CFV) staining.

Dosing regime

Anti-MAG monoclonal antibody and mouse IgG1 isotype control antibody were dialysed against sterile 0.9 % sodium chloride overnight and concentrated appropriately.

Study 1 : Animals received 2.5pg of anti-MAGmab or 2.5ug mouse IgGi.c.v. 1, 24 and 72h following MCAO (Sul per dose).

Study 2 : Animals received 200ug of anti-MAG mab or 200ug mouse IgG iv. 1 and 24 following MCAO.

Investigator was blinded to the identity of each dosing solution.

Neurological assessment

Prior to induction of cerebral ischaemia, rats for Study 1 received training in beam walking and sticky label test. Animals not reaching criteria in both tests were excluded from further study. Following training, the remainder of the animals were stratified according to performance into two balanced groups.

Throughout the neurological assessment, the investigators were blinded to the treatment group of the animal.

Bilateral Sticky label test

The bilateral sticky label test (Schallert et al., Pharmacology Biochemistry and Behaviour 16 : 455-462, (1983)) was used to assess contralateral neglect/ipsilateral bias. This models tactile extinction observed in human stroke patients (Rose, et al. 1994). This test has been described in detail previously (Hunter, et al., Neuropharmacology 39 : 806-816 (2000); Virley et al Journal of

Cerebral Blood Flow & Metabolism, 20 : 563-582 (2000). Briefly, a round, sticky paper label was placed firmly around the hairless area of the forepaws with equal pressure with order of placement randomised (left, right). Training sessions were conducted for 6 days prior to MCAO, day 6 data was utilised as the pre-operative baseline (Day 0). Animals were given two trials 24 and 7d following MCAO, the data represents a mean of the two trials. The latency to contact and remove the labels were recorded and analysed using the logrank test (Cox, J. Royal Statistical Society B 34 : 187-220 (1972)).

Beam walking

Beam walking was used as a measure of hind-limb and fore-limb co- ordination by means of distance travelled across an elevated 100cm beam (2.3cm diameter, 48 cm off the floor) as previously described in detail (Virley et al Journal of Cerebral Blood Flow & Metabolism, 20 : 563-582 (2000)). Rats were trained to cross the bean from start to finish. For testing, each rat was given 2 trials 24 h and 7d following MCAO, the data represents a mean of the two trials. Statistical analysis wvas ANOVA followed by Student's t-test.

The 27-point neurological scores (Study-1)

This study consists of a battery of tests to assess neurological status including, paw placement, visual forepaw reaching, horizontal bar, contralateral rotation, inclined plane, righting reflex, contralateral reflex, motility & general condition, as described previou sly (Hunter, et al. Neuropharmacology 39 : 806- 816 (2000)) with the addition -of grip strength measurements (scores 2 for good right fore-limb grip, 1 for weak grip). Total score = 27 for normal animal.

For study 2 this test wass modified further : Grip strength - normal scores 3, good — 2, weak — 1, very werak — 0; Motility - normal scores 4, excellent — 3, very good ~ 2, good — 1, fair — 0; General Condition - normal scores 4, excellent - 3, very good — 2, good — 1, fair — 0; Circling — none scores 5, favours one-side scores 4, large circle — 3, medium circle — 2, small circle — 1, spinning — 0). Total score = 32 for a normal animal.

In both studies animals wvere tested 1, 24, 48h and 7d following MCAOQ, a healthy normal animal scores 277 or 32 respectively. Data are presented as median values, Statistical analysis was Kruskil Wallis ANOVA.

Lesion assessment

Study 1 - For each animal, lesion areas were assessed in sections from three pre-determined levels in the brain (0, -2.0 and —6.0 mm from Bregma respectively). Neuronal damage was assessed using cresyl fast violet staining and the area of damage measured using an Optimas 6.1 imaging package. Data is expressed as mean area (mm?) + sem.

Study 2 - For each animal, lesion areas were assessed in sections from seven pre-determined levels in the brain (43mm to -8mm w.r.t. Bregma). Neuronal damage was assessed using cresyl fast violet staining and the area of damage measured using an Optimas 6.1 imaging package. Data is expressed as mean area (mm?) + sem.

Results

Study 1 — Intra-cerebral ventrical (i.c.v.) administration of anti-MAG mab

Neurological score

One hour following MCAO animals in both treatment groups showed marked impairment in neurological score (median score 12 in each group).

There was no significant difference between groups at this time. However, 24 (p=0.02), 48 (p=0.005) h and 7d (p="0.006) following MCAQO animals treated with anti-MAG mab (2.5 pg, 1, 24 an d 72h post-MCAO) showed significantly improved Total Neurological score compared to those treated with control IgG.

Median neurological scores 24, 48h and 7d following MCAO in the IgG; treated group were 15, 14 and 18 respective ly compared to 19.5, 21.5 and 22 in the anti-MAG mab treated animals. On further analysis of the individual behaviours comprising the total score, this significant improvement was mainly attributed to improved performance in the followirg tests : paw placement (24h, p=0.045; 48h, p=0.016; 7d, p=0.008), grip strength (24h, p=0.049 48h, p=0.0495; 7d, p=0.243), motility (24h, p=0.199; 48h, p=0.012; 7d, p=0.067), horizontal bar (24h, p=0.065; 48h, p=0.005; 7d, p=0.016), inclined plane (24h, p=0.006; 48h,

p=0.006; 7d, p=0.169), visual forepaw reaching (48h, p=0.049, 7d, p=0.049) and the degree of circling (24h, p=0.417; 48h, p=0.034; 7d, p=0.183).

Beam Walking

Prior to surgery all animals were trained to cross the beam (100cm).

Twenty four hours following surgery there was a significant impairment on the distance travelled on the beam in both arti-MAG (50+18cm) and IgG, (22+14cm) treated animals compared to pre-operative values. Although not significant, anti-MAG treated animals showed marked improvement over IgG; treated animals in that they travelled twice the distance of IgG; treated animals 24h following tMCAO. Seven days following surgery however, while both groups showed marked improvement over time, the performance of animals treated with

IgG remained significantly impaired compared to baseline (55+15cm; p=0.005).

In contrast however 7d following MCAO, animals treated with anti-MAG mab (2.5 ng 1, 24 and 72h, i.c.v post MCAO) performance was not significantly different from baseline (75+15cm; p=0.07). This data shows that anti-MAG mab treatment accelerated recovery of this bea m walking task compared to mouse

IgG; treated controls.

Sticky label

Prior to surgery, animals in each of the treatment groups rapidly contacted and removed the labels from each forepaw, there was no significant difference in the groups prior to treatment (Table 1). Twenty-four hours and 7d following

MCAQ the latency to contact the left paw in each of the treatment groups remained relatively unaltered, while that of the right was markedly increased.

However there was no significant differences between removal times in anti-MAG and IgG, treated animals. In addition 24h following MCAQ, the latency to removal from both the left and right forepaw was significantly increased in both treatment groups compared to baseline , however in anti-MAG treated animals the latency to removal from the left paw was significantly shorter than that of

IgG; treated animals (p=0.03). There was also a trend for reduced latency to removal from the right paw in anti-MAG treated animals compared to those treated with IgG; (p=0.08) (Table 1). At 7d there was some degree of recovery in IgG; treated animals in that the latency to removal times for each forepaw were reduced compared to those at 24h (Table 1). This data suggests that treatment of rats with anti-MAG mab accelera te the recovery in this sticky label test following tMCAOQ.

Table 1 ~ Sticky label data

Day | Treatment Contact Time (s) Removal Time (s) i =e

Left Right Left Forepaw Right er |e 0 | Ae | Zawo2 | sew0s | mea | wer 0 | de [seed | saiar | ETTORE (*p=0.03 Anti-MAG V's IgG, using the logrank test)

Lesion area measurements

Administration of anti-MAG mab i.c.v, significantly reduced lesion area in two of the three brain levels examined compared to those animals treated with equal amounts of mouse IgG; when examined 7 days following tMCAO (Table 2).

Table 2

Venskre d sen a Eo B00

[eww TT eww

Anti-MAG mab *9Q + 2 4+ 3 3+1

Fl I I (*p=0.02, ¥p=0.03, *p=0.06, anti-MAG V's IgG;, One—way, unpaired Students T-

Test)

Study 2 — Intra-venous (i.v.) administration

Neurological score

One and 24 hours following MCAO animals in both groups showed marked impairment in neurological score. There was no significant difference between groups at this time, median scores 24h following anti—MAG mab and IgG; treatment were 20 and 18 respectively (p=0.5). Forty—eight hours following

MCAO, animals treated with anti-MAG mab (200ug, i.w. 1 and 24h post-MCAO) showed significant improvement in paw placement (p==0.048) and grip strength (p=0.033). Seven days following the onset of cerebral ischaemia animals treated with anti-MAG mab continued to improve (paw placenent p=0.041; grip strength, p=0.048; motility, p=0.05) and showed sigruificant improvement in total neurological score (median score 25) compared to thoese treated with mouse IgG; (Median score 23, p=0.047).

Lesion area measurements

The anti-MAG antibody when administered i.v. EMCAO significantly reduced lesion area at 5 out of 7 pre-determined brain levels (+3 to -8 mm w.r.t.

Bregma) compared to isotype controls, when examined 7d following MCAO.

Brain level | Mean lesion area + Mean lesion area + wrt SEM (mm?) SEM (mm?)

Bregma — Anti- MAG treated — Mouse IgG ; treated

: * p< 0.05 — Unpaired, one-way Students T-test

Conclusions

An anti-MAG monoclonal antibody ad ministered either directly into the CSF or intravenously following transient middie-cerebral artery occlusion in the rat, both reduced the area of cell death and imp-roved functional recovery compared to control treated animals. The degree of neuroprotection seen in these studies suggests that this effect can not be attributed to axonal sprouting as this would not result in neuronal sparing. The improvernent in functional recovery seen 24 and 48h following MCAO probably reflects the degree of neuroprotection offered by this antibody compared to control treated animals. However, over time the animals appear to improve further, suggesting that blocking MAG activity can also enhance functional recovery over time.

The studies presented here provide evidence that blocking the actions of

MAG provide both neuroprotection and enhanced functional recovery in a rat model of stroke, and therefore anti-MAG anttibodies provide potential therapeutic agents for acute neuroprotection and/or the promotion of functional recovery following stroke. The low amounts of antibody administered via the i.v route and the resulting low serum levels of the antiboaly would in turn suggest extremely low antibody concentrations in the brain dues to the constraints of the blood brain barrier for antibody penetration. Surprisingly”, however, this still resulted in both, neuroprotection and enhanced functional recovery being observed. Anti-MAG antibodies also have potential use in the treatment of other neurological : disorders where the degeneration of cells amd or nerve fibres is apparent such as spinal cord injury, traumatic brain injury, peripheral neuropathy, Alzheimer's disease, fronto-temporal dementias (tauopathies), Parkinson's disease,

Huntington's disease and Multiple Sclerosis. In the examples that follow the

CDRs of the chimeric and humanised antiboclies disclosed therein are the CDRs of the antibody of example 1.

Example 2~ Chimeric antibody

Altered antibodies include chimeric antib<dies which comprise variable regions deriving from one species linked to consstant regions from another species. Examples of mouse-human chimeric arati-MAG immunoglobulin chains of the invention are provided in Figures 1, 2, and 3. Mouse-human chimeras using human IgG1, IgG2, IgG3, IgG4, 1gA, IgE, IgM, XgD constant regions may be produced, as may chimeras associating the mousse variable regions with heavy or light chain constant regions from non-human sp ecies.

Figure 1 (Seq ID No. 27) provides the amino acid sequence of a chimeric immunoglobulin heavy chain in which the murine anti-MAG heavy chain variable region is associated with a functional immunoglobulin secretion signal sequence, and with an altered form of the human IgG1 corustant region, in which Kabat residues 248 and 250 have been mutated to alamine in order to disable the effector functions of binding to FcyRI and complement protein Clq (Duncan, A.R. and Winter, G. Localization of the C1q binding sit-e on antibodies by surface scanning. Nature 332, 738-740, 1988. Duncan, A.R., Woolf, J.M., Partridge,

L.J., Burton, D.R. and Winter, G. Localisation of t-he binding site for human FcR1 on IgG. Nature 332, 563-564, 1988). Such mutations are optionally made in order to customise the properties of an altered amtibody to achieve a particular therapeutic effect — for example binding to and blocking the function of an antigen without activating lytic effector mechanisms.

Figure 2 (Seq ID No. 28) provides the amiro acid sequence of a chimeric immunoglobulin light chain in which the murine aeti-MAG light chain variable region is associated with a functional immunoglobeulin secretion signal sequence, and with the human kappa constant region.

Similarly, the anti-MAG variable regions ma-y be associated with immunoglobulin constant regions which lack mutations disabling effector functions. Figure 3 (Seq ID No. 29) provides the amino acid sequence of a chimeric immunoglobulin heavy chain in which the- murine anti-MAG heavy chain variable region is associateed with a functional immunoglobulin secretion signal sequence, and with a wild-type form of the human IgG1 constant region.

From the information provided in Figures 1 to 3, cDNA inserts encoding . these chimeric chains may be prepared by standard molecular biology techniques well known to those skilled in the art. Briefly, the genetic code is used to identify nucleotide codons encoding the desired amino acids, creating a virtual cDNA sequence encoding the chirneric protein. If the cDNA insert is desired to be expressed in a particular organism, then particularly favoured codons may be selected according to know n codon usage biases. The desired nucleotide sequence is then synthesised by means of PCR amplification of a template comprising overlapping synthetic oligonucleotides which, as a contig, represent the desired sequence. The resulting product may also be modified by PCR or mutagenesis to attach restriction sites to facilitate cloning into a suitable plasmid for expression or further manipulations.

Example 3-Chimeric antibody binds to rat MAG in ELISA

Chimeric anti-MAG antibody containing the light and heavy chain CDRs of the invention was produced by transient transfection of CHO cells. For this,

Transfast transfection reagent (Promega; E2431) was used and transfections carried out according to manufactures instructions. In brief, ~ 108 CHO cells were plated out per well of é-well culture plates. The following day mammalian expression vector DNA encoding the appropriate heavy or light chain were mixed at 1:1 ratio (5ug total DNA) in medium (Optimem1 with Glutamax; Gibco #51985-026). Transfast tramnsfection reagent was added and the solution transferred to wells with corfluent cell layers. After 1h at 37°C in the cell incubator, the DNA/Transfast mixture was overlaid with 2mi Optimem medium and left for 48-72h in the incubator. Supernatants were harvested, cleared by : centrifugation and passed through 0.2 um filters. Antibody concentration in CHO cell culture supernatant was determined by ELISA and estimated to be around 0.5 pg/ml. For MAG binding, commercially available ratMAG-Fc was used. Due to the fusion with human Fc bound chimeric antibodies could not be detected using anti-human IgG secondary antibodies. Instead, anti-human kappa light chain-specific antibody was used. Figure 4 shows that this chimeric antibody binds to MAG even at 1/64 di lution. An unrelated humanised antibody and culture supernatant from mock transfected cells did not generate any signal in this assay.

Procedure:

ELISA microtiter plates (Nunc Maxisorp) were coated with 1 pg/ml rat MAG-Fc fusion protein (R&D systems; 538-MG) in PBS at 4°C overnight. Plates were washed twice with PBS and then blocked with PBS/BSA (1% w/v) for 1h at room temperature (RT). Culture suprernatants from transiently transfected CHO cells were passed through 0.2um fidters and serial diluted in PBS/BSA starting at neat supernatant to 1/64 dilution. Sample dilutions were left at RT for 1h. Plates were then washed three times with PBS/Tween 20 (0.1% v/v). Detection antibody was goat anti-human kappa light chain specific-peroxidase conjugate (Sigma A-7164) diluted at 1/2000 in PBS/BSA. ~The detection antibody was incubated for 1h at RT and the plates washed as above. Substrate solution (Sigma Fast OPD P-9187) was added and incubated until appropriate colour development was detected and then stopped using 3M H,SO4. «Colour development was read at 490nm.

Example 4 — Humanised antibodies

Altered antibodies include humanised antibodies which comprise

J humanised variable regions linkexd to human constant regions. Examples of humanised anti-MAG immunoglobulin chains of the invention are provided in

Figure 5. Humanised antibodies using human IgGl, IgG2, IgG3, IgG4, IgA, IgE, : IgM, IgD constant regions may be produced.

Figure 5 (Seq ID No: 30 ) provides an example of the amino acid sequence of a humanised immunoglobulin heavy chain in which the humanised anti-MAG heavy chain variable region is associated with a functional immunoglobulin secretion signal sequence, and with an altered form of the human Ig&1 constant region, in which Kabat residues 248 and 250 have been mutated to- alanine in order to disable the effector functions of bindi ng to FcyRI ) and complement protein C1q (Duncan, A.R. and Winter, G. Localizatzion of the

Clq binding site on antibodies by surface scanning. Nature 332, 738-740, 1988.

Duncan, A.R., Woolf, J.M., Partridge, L.J., Burton, D.R. and Winter, G. Localisatior of the binding site for human FcR1 on IgG. Nature 332, 563-564, 1988). Such mutations are optionally made in order to customise the properties of an altere=d antibody to achieve a particular therapeutic effect ~ for example binding to and blocking the function of an antigen without activating lytic effector mechanisms.

Figure 5 (Seq ID No. 31) also provides an example of the amimo acid sequence of a humanised immunoglobulin light chain in which the humanised anti-MAG light chain variable region is associated with a functional immunoglobulin secretion signal sequence, and with the human kappa constant region.

Similarly, the anti-MAG variable regions may be associated wit h immunoglobeulin constant regions which lack mutations disabling effector functions. Fi gure 5 (Seq ID No. 32) provides the amino acid sequence of a humanised ismmunoglobulin heavy chain in which the humanised anti~MAG heavy chain variable region is associated with a functional immunoglobulin Secretion signal seque nce, and with a wild-type form of the human IgG1 constant region.

From the information provided in Figure 5, cDNA inserts encod ing these humanised c hains may be prepared by standard molecular biology techniques well known to those skilled in the art. Briefly, the genetic code is used to identify nucleotide codons encoding the desired amino acids, creating a virtua 1 cDNA sequence encoding the protein. If the cDNA insert is desired to be expressed in a : particular organism, then particularly favoured codons may be selected according to known codon usage biases. The desired nucleotide sequence is them synthesised bby means of PCR amplification of a template comprising o-verlapping synthetic oligonucleotides which, as a contig, represent the desired sequence.

The resulting product may also be modified by PCR or mutagenesis to attach restriction sites to facilitate cloning into a suitable plasmid fog" expression or further manipulations.

Example 5 : Humanised anti-MAG antibodies bind to rat and human

MAG.in Elisa 1) Direct bimding ELISA to rat MAG-Fc fusion protein Of normalised amounts of culture supernatant for 9 humanised heavy and light chain cormbinations

Humanised anti-MAG antibodies containing the light and heawy chain CDRs of the invention were produced by transient transfection of CHO cells. For this,

Transfast trarssfection reagent (Promega; E2431) was used amd transfections carried out ac cording to manufactures instructions. In brief, ~= 108 CHO cells were plated o ut per well of 6-well culture plates. The following day mammalian expression vector DNA encoding the appropriate heavy or lighat chain were mixed at 1:1 ratio (5 pg total DNA) in medium (Optimem1 with Glutammax; Gibco #51985-026). Transfast transfection reagent was added and the solution transferred to wells with confluent cell layers. After 1h at 37°C in the cell incubator, the DNA/Transfast mixture was overlaid with 2ml O=ptimem medium and left for 48 -72h in the incubator. Supernatants were harvested, cleared by centrifugation and passed through 0.2 pm filters. 9 heavy and light variable chain combinations wvere produced from the sequences shown in the table below and the IgG1 heavy chain constant regions were functional accordi ng to Seq.ID .

oo

X 5

Antibody concentration was determined by ELISA and &he amounts of supernatant used in the assay normalised to a starting concentration of 250 or 500 ng/ml (depending on concentration of culture supematant). As antigen, commercially available ratMAG-Fc was used (R&D Systems; 538-MG). Due to the fusion of this antigen with human Fc, bound chimeric antibodies could not be detected using general anti-human IgG secondary antibodies. Instead, anti- human kappa light chain-specific antibody was used. Figure 6 shows that all 9 humanised antibodies examined here bound to rat MAG with very similar binding curves down to ~ 4ng/ml. The chimeric antibody used .as a reference showed binding characteristics that fell within the group of hurmanised antibodies examined here. Although not absolute, this may suggest that the affinities of the humanised antibodies examined here lie very closely within the affinity range of fhe non-humanised chimeric antibody used as a refererce here.

Procedure

O6-well Nunc Maxisorp plates were coated overnight at 4°C with rat MAG-Fc f=usion protein (1 pg/ml; R&D Systems; Cat.No. 538-MG ) in PBS. Plates were washed twice with PBS containing Tween20 (0.1% v/v; PBST) and blocked with

PBS containing BSA (1%w/v) for 1h at room temperature (RT). Variable amounts of culture supernatants were serial diluted in blocking bauffer and added to the blocked wells starting at approximately 500 or 250 ng/ml. Antibody concentrations of supernatants were based on independent assays measuring tthe amount of humanised antibody present in each culture supernatant. Chimeric mnouse-human (non-humanised) antibody was also inclu ded as reference.

Antibody samples were incubated 1h at RT and plates then washed 3x with

PBST. Secondary antibody (Goat anti-human light chain specific-peroxidase conjugate; Sigma A-7164) was added diluted 1/5000 ir blocking buffer and incubated for 1h at RT. Wells were washed three times as above and binding dete cted by adding substrate (OPD tablets dissolved according to instructions;

Sigma P-9187). Colour development was monitored an d the reaction stopped using 3M H,S0,, Colour development was read at 490n m. 2) Direct binding ELISA to rat MAG-Fc fusion protein of two purified lmumanised anti-MAG antibody heavy-light chain combinations

Two humanised antibodies consisting of heavy and light chain variable region combinations BVh1/CVi1 and BVh3/CVI3 (table figure 5) and a mutated IgG1 constant region as exemplified by SEQ.L.D.NO:30 (which is BVh1/CVil mutated

IgG1 , those skilled in the art can readily derive the sequence for the BVh3/CVI3 equivalent) were produced by a scaled-up version of the transient transfection described in example 3 and purified using protein A affgnity chromatography.

Purified antibody material was dialysed against PBS and the concentration determined by OD280 reading. Antibody concentrations were adjusted to 5000 ng/ml and used as serial dilutions in a rat MAG-Fc bindi ng ELISA. Figure 7 shows that purified antibody material binds rat MAG-Fc and that both heavy and light chain variable region combinations tested here are extremely similar.

Method: 96-well Nunc Maxisorp plates were coated overnight at 4°C with rat MAG-Fc fusion protein (2.5 pg/ml; R&D Systems; Cat.No. 538-MIG) in PBS. Plates were washed twice with PBS containing Tween20 (0.1% v/v; PBST) and blocked with

PBS containing BSA (1%wi/v) for 1h at room temperatures (RT). Purified humanised antibody was adjusted to a starting concentration of 5 pg/ml in blocking buffer and then serial diluted. Antibody sample s were incubated 1h at

RT and plates then washed 3x with PBST. Secondary aretibody (Goat anti-human light chain specific-peroxidase conjugate; Sigma A-7164) was adde d diluted 1/5000 in blocking buffer and incubated for 1h at RT. Wells were wrashed three times as above and binding detected by adding substrate (OPD tab lets dissolved according tos instructions; Sigma P-9187). Colour development was monitored and the reaction stopped using 3M H,S0, Colour development was read at 490nm.

Results:

Both purifiedE humanised antibodies carrying none or several framevaork mutations sheow extremely similar binding to rat MAG. The results are seen in

Figure 7. 3) Binding to human MAG expressed on CHO cells of norma#ised amounts of culture supernatant for two humanised heavy and light chain cosnbinations

The same hurmanised variable heavy and light chain combinations de=scribed in example 5 2) were tested as cleared culture supernatants against hu-man MAG expressed on the surface of CHO cells. The amount of culture supermatant used for each antib-ody was normalised based on antibody concentrations Cletermined by ELISA. For this, 96-well plates (Nunc Maxisorp) were coated overmight at 4°C with goat anti~human IgG (gamma) chain (Sigma I-3382; in bicarborate buffer pHO.6; 2ug/mB). Following day, plates were washed twice with wash buffer (PBST) and blocked by adding at least 75ul blocking buffer (PBS containing BSA } 1% w/v) for 1 at RT. Antibody sample solution were serial diluted in blocking buffer (starting dilution neat or 42) in duplicate. Ab standard was purified humanised IgG 1 antibody of an unrelated specificity and known conce=ntration.

The standard solution was also serial diluted across plate startineg at 500ng/m.

All antibody soluations were incubated for 1h at RT. Plates were vvashed 3x as above and then incubated with goat anti-human light (kappa) ctmain specific (free and bound) peroxidase conjugate (Sigma; A-7164) at 1/5000 in blocking buffer for 1h @ RT. Plates were again washed 3x as above and incuba ted with substrate solution (OPD tablets; Sigma P-9187 until strong colou r development.

Colour development was stopped by adding 25ul 3M H2504 and the plate read at 490nm.

Figure 8 shows that both antibodies tested here are recognising human MAG and are very similar i n their binding characteristics. CHO/- are negative controls of

CHO cells with no MAG expressed.

Method for Fu cell-based ELISA 96-well plates (Costar 3595) were filled with 100p! cell suspensio=n/well (see table below for recommended cell number for performing assay on days 1, 2, 3 or 4).

Culture medium wvas removed and plates blocked with DMEM/F12 (Sigma D6421) containing FCS ( 10%), BSA (1%), NaN3 (1%; blocking buffer) for 1 hour at RT.

Blocking solution was then removed and sample added (in blockimg buffer

50ul/well). Incubated samples at 4°C for 1 h. Plates were then washed 3x with

PBS using a Skatron plate wash er. After wash, cells were fixed with 0.5% paraformaldehyde (diluted in PES) for 20 minutes at 4°C and again washed as above. 50ul/well Europium-conj ugated secondary antibody diluted in Europium buffer (50mM Tris base, 150mM1 NaCl, 0.5% BSA, 0.1g/l, Tween 20, 7.86mg/!

DTPA at pH 7.3) was added and incubated for 1 h at 4°C.

Washed plates as above and ad ded 200ul Delphia enhancement solution/well.

Incubated solution at RT for 5 — 10 minutes. Wells were read within 24 hours on a Victor . 4) Competition ELISA for bianding to rat MAG-Fc fusion protein of two purified humanised antibeodies and the non-humanised mouse monoclonal antibody

Method: 96-well Nunc Maxisorp plates were coated overnight at 4°C with rat MAG-Fc fusion protein (2.5 ng/ml; R&D Systems; Cat.No. 538-MG) in PBS. Plates were washed twice with PBS containirg Tween20 (0.1% v/v; PBST) and blocked with

PBS containing BSA (1%w/v) for 1h at room temperature (RT). Purified humanised antibody was adjusted to a concentration of 200ng/ml and mixed at equal volume with competitor molecules made up in blocking buffer ranging from 6000 to 0 ng/ml. Competitors were either parental mouse monoclonal antibody (anti-MAG) or an unrelated mou se monoclonal antibody (INN1) at the same concentrations (BVh1/CVI1 only). Antibody/competitor solutions were incubated 1h at RT and plates then washed 3x with PBST. Secondary antibody (Goat anti- human light chain specific-peroxcidase conjugate; Sigma A-7164) was added diluted 1/5000 in blocking buffer and incubated for 1h at RT. Wells were washed three times as above and binding detected by adding substrate (OPD tablets dissolved according to instructions; Sigma P-92187). Colour development was measured at 490 nm.

Results:

Both purified antibody preparations are equally” competed by the original mouse monoclonal antibody but not by a mouse monoclonal antibody that has an unrelated specificity - see Figure 9. This shows that the original mouse monoclonal antibody and the humanised antibodies tested here are probably recognising the same epitope and possibly have very similar affinities to rat MAG.

Claims (1)

1. An altered antibody or functional fragment thereof which binds to and neutralises MAG and which comprises one or more of the following CDRs. Light chain CDRs According to Kabat KSSHS\/LYSSNQKNYLA WASTRES cia Heavy chain CDRs Accorddng to Kabat WINTYTGEPTYADDFTG NPINYYGINYEGYVMDY

2. An altered antibody or functional fragment thereof which comprises a heavy chain variable domain which comprises one or more CDRs selected from CDRH1, CDRH2 and CIDRH3 and/or a light chain variable domain which comprises one or more CDRs selected from CDRL1, CDRL2 and CDRL3.

3. An altered anti-Mag antibody or functional fragment thereof which ) comprises: a heavy chain variablex domain (Vu) which comprises in sequence hypervariable regions <CDRH1, CDRH2and CDRH3 and Jor a Bight chain variable domain (Vi. ) which comprises in sequence hy-pervariable regions CDRL1, CDRL2 and CDRL3. : 4, An antibody of claim 3 which is monoclonal.

5. An antibody of claim 4 which is humanised.

6. An antibosdy or functional fragment thereof that according to claism 5 which comprises a heavy chain variable region comprising one of the following amino aci d sequences OVOLVOSGSEXEKKPGASVKVSCKASGYTFTNYGMNWVRQAPGQGL EWMGWI NTYTGEPTYADDFTGRFVFSLDTSVSTAYLQISSLKAEDTAVYYCARNPIN YYGINYEGYVIMMDYWGQGTLVTVSS (SEQ ID No 13). QVOQLVQSGSEILLKKPGASVKVSCKASGYTFTNYGMNWVRQAPGQGLIEWMGWI NTYTGEPTYADDFTGRFVFSLDTSVSTAYLQISSLKAEDTAVYFCARNPIN YYGINYEGYVMDYWGQGTLVTVSS (Sequence ID No 14) QVQLVQSGSELKKPGASVKVSCKASGYTFTNYGMNWVROAPGQGLEWMGWI NTYTGEPTYADDFTGRFVFSLDTSVSTAYLQISSLKAEDTATYFCARNPIN YYGINYEGYVMDYWGQGTLVTVSS (sequence ID No 15)

7. An antibocly or functional fragment thereof according to claim 6 farther comprising a light chain variable region comprising amino acid Sequence ID No 16, 17,18 or 19: DIVMTQSPDSILAVSLGERATINCKSSHSVLYSSNQKNYLAWYQOKEGOPPK LLIYWASTRESCGVPDRFSGSGSGTDFTLTISSLOAEDVAVYYCHQYLSSLT FGOGTKLEIKRTV (SEQ ID No 16) : 35 DIVMTQSPDSIAVSLGERATINCKSSHSVLYSSNOKNYLAWYQOKEFGOPPK LLIYWASTRES GVPDRFSGSGSGTDFTLTIINLQAEDVAVYYCHQYLSSLT FGQGTKLEIKRTV (SEQ ID No 12)

DIVMTOSPDSLAVSL GERATINCKS SHSVLYS SNQKNYLAWYQQK PGOPPK LLIYWASTRESGVPDRFSGSGSGTDFTLTISSLHTEDVAVYYCHQ YLSSLT . FGQGTKLEIKRTV (SEQ ID No 18) DIVMTQSPDSLAVSLGERATINCKSSHSVLYSSNQKNYLAWYQQK PGQPPK LLIYWASTRESGVPDRFSGSGSGTDFTLTIINLHTEDVAVYYCHQ YLSSLT FGQGTKLEIKRTV (SEQ ID No 19)

8. An antibody acco rding to claims 6 — 7 comprising: a heavy chain variable fragment comprising SEQ ID No 13,14 or 15 and a constant part or fragment thereof of a human heavy chain and a light chain variable fragment comprising SEQ ID No 16, 17, 18 or 19 and a constant part ows fragment thereof of a human light chain.

09. An humanised antibody according to any preceding claim selecte d from antibodies comprising: Heavy chain variable region comprising Seq ID No 13 and light cain variable region comprising Seq ID No 16; Heavy chain variable region comprising Seq ID No 13 and light chain variable region comprising Seq ID No 17; Heavy chain varia ble region comprising Seq ID No 13 and light chain variable region comprising Seq ID No 18; Heavy chain varia ble region comprising Seq ID No 13 and light clhain variable region co mprising Seq ID No 18. Heavy chain varia ble region Scomprising eq ID No 14 and light chain variable region co mprising Seq ID No 16; Heavy chain varia ble region comprising Seq ID No 14 and light chain variable region co mprising Seq ID No 17; Heavy chain variable region comprising Seq ID No 14 and light chain variable region comprising Seq ID No 18; Heavy chain variable region comprising Seq ID No 14 and light chain variable region coamprising Seq ID No 19.

Heavy chain variable rexgion comprising Seq ID No 15 and light chain variable region comprising Seq ID No 16; Heavy chain variable region comprising Seq ID No 15 and light chain } variable region comprising Seq ID No 17; Heavy chain variable region comprising Seq ID No 15 and light chain variable region comprising Seq ID No 18; Heavy chain variable region comprising Seq ID No 15 and light chain variable region comprising Seq ID No 19.

10. A polynucleotide encod ing the heavy chain variable region comprising Sequence ID No 15. CAGGTGCAGCTGGTGCAAT CTGGGTCTGAGTTGAAGAAGCCTGGGGCCTCA GTGAAGGTTTCCTGCAAGGCTTCTGGATACACCTTCACTAACTACGGCATG AACTGGGTGCCACAGGCCCECTGGACAAGGGCTTGAGTGGATGGGATGGATC AACACCTACACCGGCGAGC'CCACCTACGCCGACGACTTCACCGGCCGGTTT GTCTTCTCCTTGGACACCT CTGTCAGCACGGCATATCTGCAGATCAGCAGC CTAAAGGCTGAGGACACTG-CCGTGTATTACTGTGCGAGAAACCCCATCAAC TACTACGGCATCAACTACG-AGGGCTACGTGATGGACTACTGGGGCCAGGGTC ACACTAGTCACAGTCTCCT CA

11. A polynucleotide sequemce encoding the amino acid Sequence ID No 14 is: CAGGTGCAGCTGGTCCAAT CTGGGTCTGAGTTGAAGAAGCCTGGGGCCTCA GTGAAGGTTTCCTGCAAGG CTTCTGGATACACCTTCACTAACTACGGCATG AACTGGGTGCGACAGGCCC CTGGACAAGGGCTTGAGTGGATGGGATGGATC AACACCTACACCGGCGAGC-CCACCTACGCCGACGACTTCACCGGCCGGTTT GTCTTCTCCTTGGACACCT CTGTCAGCACGGCATATCTGCAGATCAGCAGC CTAAAGGCTGAGGACACTG CCGTGTATTTCTGTGCGAGAAACCCCATCAAC TACTACGGCATCAACTACGAGGGCTACGTGATGGACTACTGGGGCCAGGGC ACACTAGTCACAGTCTCCTCA

12. A polynucleotide sequerce encoding the amino acid Sequence ID No 15 is: CAGGTGCAGCTGGTGCAATCTGGCTCTGAGTTGAAGAAGCCTGGGGCCT CA GTGAAGGTTTCCTGCAAGGCTTCTGGATACACCTTCACTAACTACGGCATG AACTCGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGATGGATC 40 AACACCTACACCGGCGAGCCCACCTACGCCGACGACTTCACCGGCCGGTTT GTCTTCTCCTTGGCACACCTCTGTCAGCACGGCATATCTGCAGATCAGCAGC

CTAAAGGCTGAGGACACTGCCACCTATTTCTGTGCGAGAAACCCCATCAAC TA CTACGGCATCAACTACGAGGGCTACGTGATGGACTACTCRGGCCAGGGC . ACACTAGTCACAGTCTCCTCA 13 A polynucleotide encoding amino acid Sequence ID No 16 is: : 10 GACATCGTGATGACCCAGTCTCCAGACTCCCTGECCTCTEGTCTCTGCRCGAG AGGGCCACCATCAACTGCAAGAGCAGCCACAGCGTGCTGTACAGCAGCAAC CAGAAGAACTACCTGGCCTGGTACCAGCAGAAACCAGGACAGCCTCCTAAG CTGCTCATTTACTGGGCATCTACCCGGGAATCCGGEGTCCCTCACCCGATTC AGTGGCAGCGGGTCTGGGACAGATTTCACTCTCACCATCAGCACCCTCGCAG GC TGAAGATGTGGCAGTTTATTACTGTCACCAGTACCTGAGCAGCCTCACC TT TGGCCAGGGGACCAAGCTGCGAGATCAAACGTACEATG

14. A polynucleotide sequence encoding amin acid SEQ ID No 17 is: GACATCGTGATGACCCAGTCTCCAGACTCCCTGGCT CTGTCTCTGGGCCAG AGGGCCACCATCAACTGCAAGAGCAGCCACAGCCTGCTGTACAGCAGCAAC CAGAAGAACTACCTGGCCTGGTACCAGCAGAAACCA GGACAGCCTCCTAAG CTGCTCATTTACTGGGCATCTACCCGGGAATCCGGECTCCCTGACCGATTC AGT GGCAGCGGGTCTGGGACAGATTTCACTCTCACCATCATCAACCTCCAG GCT'GAAGATGTGGCAGTTTATTACTGTCACCAGTAC CTGAGCAGCCTGACC TTT GGCCAGGGGACCAAGCTGGAGATCARACGTACGECTG

15. A polynucleotide encoding amino acid SEQ ID No 18 is: GACATCGTGATGACCCAGTCTCCAGACTCCCTGGCTCTGTCTCTCGGGCGAC AGG GCCACCATCAACTGCAAGAGCAGCCACAGCGTGCTGTACAGCAGCAAC CAGAAGAACTACCTGGCCTGGTACCAGCAGAAACCACGGEACAGCCTCCTAAG CTGCTCATTTACTGGGCATCTACCCGGGAATCCGCCGGTCCCTGACCGATTC AGT GGCAGCGGGTCTGGGACAGATTTCACTCTCACCATCAGCAGCCTCCAC ACCGAAGATGTGGCAGTTTATTACTCTCACCAGTACCTGAGCAGCCTGACC TTTGGCCAGGGGACCAAGCTGGAGATCAAACGTACGCGTG

16. A polynucleotide encoding amino acid SEQ ID No 1 5 is: 40 GACATCGTGATGACCCAGTCTCCAGACTCCCTGGCTGETGETCTCTCGECGAG AGGGCCACCATCAACTGCAAGAGCAGCCACAGCGTGCTGTACAGCAGCAAC Te ear Ss AVABV ALL AAC } CAGAAGAACTACCTGGCCTGGTACCAGCAGAAACCAGGACAGCCTCCTAAG CTGCTCATTTACTGGGCATCTACCCGGGAATCCCECEETCCCTGACCGATTC 45 AGTGGCAGCGGGTCTCGCACAGATTTCACTCTCACCATCATCAACCTGCAC ACCGAAGATGTGGCAGTTTATTACTGTCACCAGTACC TGAGCAGCCTGACC TTTGGCCAGGGGACCAAGCTGGAGATCAAACGTACGETG

SX h EE PCT/EXP2003/008749

17. A pharmaceutical composition comprising an altered anti-M1AG antibody or functional fragment thereof according to claims 1-8 togeather with a pharmaceutically acceptable diluent or carrier.

18. A method of prophylaxis of stroke and other neurological dise ases/disorders in a human which comprises administeriang to said human an effective amount of an anti-MAG antibody, acco rding to claims 1-8 including altered antibodies or a functional fragment thesreof.

19. The use of an anti-Mag antibody according to claims 1-8, ircluding altered antibodies or a functional fragment thereof in the m anufacture of a preparation for treatment or prophylaxis of stroke and other neurological dise ases/disorders.

20. A m ethod of inhibiting neurodegeneration and/or promoting functional reco very in a human subject at risk of developing, a stroke or other neurological disease/disorder which comprises administering to said hum an an effective amount of an anti-MAG antibody according to claims 1-6, including altered antibodies or a functional fragment th ereof.

21. The use of an antibody according to claims 1-8, including a Itered antibodies or a functional fragment thereof in the manufacture of a preparation for inhibiting neurodegeneration and/or promotirg functional recowery in a human subject afflicted with, or at risk of developing, a strolkce and other neurological disease/disorder. -58 - AMENDED SHEET

NY A PCT/EP2C03/008749

22. A method of prophylaxis of stroke or other neurological diseasse/disorder in a human comprising the step of parenteral administration off an effective amount of an anti-MAG antibody to said human.

23. The method of claim 22 wherein the anti-MAG antibody is adrmninistered intravenously.

24. The method of claim 18, 20 or 22 wherein the other neurological disease/disorde r is selected from the group consisting of; traurmatic brain injury, spinal cord, Alzheimer’s disease, fronto-temporal demertias (tauopathies), peripheral neuropathy, Parkinson's disease, Huntington's disease and multiple sclerosis.

25. A method of promoting axonal sprouting comprising the step of contacting a hurnan axon with an anti-MAG antibody of claims 1 to 8.

26. The method of claim 25 wherein the method is in vitro.

27. Use of claim 19 or 21 wherein the other neurological disease/dissorder is selected from the group consisting of; traumatic brain injury, sp inal cord, Alzheimer’s disease, fronto-temporal dementias (tauopathies), peripheral neuropathy, Parkinson’s disease, Huntington's disease and multiple sclerosis. :

28. Use of an anti-M.AG antibody of claims 1 to 8 in the manufacture of a preparation for promoting axonal sprouting. - B69 - AMENDED SHEET

PCT/EP2003/008749

29. A substance or composition for use in a nrnethod of treatment or prophylaxis of stroke and other neurological diseases/disorders in a human, said substance or composition co mprising an anti-MAG antibody. according to claims 1-8 including altered antibodies or a functional fragment thereof, and said method comprising administering to said human an effective amount of said substance or composition.

30. A substance or composition for use in a rmethod of inhibiting neurodegeneration and/or promoting func tional recovery in a human subject suffering, or at risk of developing _, a stroke or other neurological disease/disorder, said substance or comp osition comprising an anti-MAG antibody according to claims 1-6, includimg altered antibodies or a functional fragment thereof, and said method comprising administering to said human an effective amount of said substance or composition.

31. A substance or composition for use in a method of treating or prophylaxis of stroke or other neurological disease/dissorder in a human, said substance or composition comprising an anti-MAG antibody, and said method comprising the step of parentera | administration of an effective amount of said substance or compositiom to said human.

32. A substance or composition for use in a method of treatment or prevention of claim 31 wherein the anti-BIMAG antibody is administered intravenously.

33. A substance or composition for use in a method of treatment or prevention of claim 29, 30 or 31 wherei n the other neurological - 60 - AMENDED SHEET

SE TIPE NE FPCT/EP2003/008749 disease/dissorder is selected from the group consisting: of; traumatic brain injury, spimal cord, Alzheimer’s disease, fronto-tempor-al dementias (tauopathies), peripheral neuropathy, Parkinson’s dise=ase, Huntington's disease armd multiple sclerosis.

34. A substan<ce or composition for use in a method of promoting axonal sprouting, said substance or composition comprising an anti-MAG : antibody of claims 1 to 8, and said method comprisirmg the step of contacting a human axon with said substance or conmposition.

35. A substan ce or composition for us in a method of promoting axonal sprouting of claim 34 wherein the method is in vitro.

36. An antibody or functional fragment thereof of any ornue of claims 1 to 9, substantially as herein described and illustrated.

37. A polynucleotide of any one of claims 1 to 16, subst antially as herein described and illustrated.

38. A composition of claim 17, substantially as herein deescribed and illustrated .

39. A method of any one of claims 18 to 24, substantial ly as herein described and illustrated.

40. A method of claim 25 or claim 26, substantially as h erein described and illustrated . -61 - AMENDED SHEET

[I RPE PCT/EP2003/008749

41. Use of any one of claims 19, 21, 27 or 28, substantially as herein described and illustrated.

42. A substance or composition for use in a method of treatm ent or prevention of any one of claims 29 to 35, substantially as herein described and illustrated.

43. A new antibody or functional fragment thereof, a new polynucleotide, a new coemposition, a new non-therapeutic method of treatrment, a new method of promoting axonal sprouting, a new use of an a nti-MAG antibodly of any one of claims 1 to 8, or a substance or composition for a new usse in a method of treatment or prevention, substantially as herein describ ed. -B62 - AMENDED SHEET