CA2777686A1 - Polypeptides and antibodies derived from chronic lymphocytic leukemia cells and uses thereof - Google Patents

Abstract

An isolated antibody, or a CD200-binding fragment thereof, comprising at least one antigen binding site that binds to human CD200 (hCD200), wherein the antibody or CD200-binding fragment thereof is a humanized, fully human, or chimeric antibody or CD200-binding fragment thereof, and wherein a Fab fragment of the antibody inhibits the interaction between hCD200 and hCD200 receptor by at least 90% at a concentration of 6.7 µg/ml. as evaluated in an in vitro fluorescent microsphere assay.

Description

POLYPEPTlIDES AND ANTIBODIES I)LI~hSrIfD FROM
CTIRONIC Li'Aff'EIOCYTIC LEI JKEML4 CELTS AND USES T ITEREOFr This is a division of Canadian 2,574,488 tiled July 19, 2005.
T>= CBNICAL FIELD
Cancer treatments using a therapy that provides a combination of two mechanisms are disclosed. More specifically, this disclosure relates to treating cancer using a therapy that: 1) interferes with the interaction between CD200 and its receptor to block immune Suppression T fetch}, promoting eraL ICallori Ui i le Cancer Cells; and -2) direul.y 111-1111's the cancer cells either by a) complement-mediated or antibody-dependent cellular cytotoxicity or b) by targeting cells using a fi_ision molecule that includes a CD--)00-1? targeting portion.
BACKGROUND
Chronic Lynrphocytic Leukemia (CLIL) is a disease of the white blood cells and is the most common form of leukemia in the Western Hemisphere. CLL, represents a diverse group of diseases relating to the grog nth of malignant lymphocytes that grow slowly bum have an extended life span. CLL is classified in various categories that include, for example, B-cell chronic lymphocytic leukemia (D B-CU-) of classical and mixed types, b-cell and 'T-cell prolymphocytic leukemia, hairy cell leukemia, and large granular lyrnphocytic leukemia.
Of all the different types of CLL, B-CLL accounts for approximately 30 percent of all leukemias. Although it occurs more htequenily in individuals over 50 years of age, S it is increasingly seen in younger people. B-CLL is characterized by accumulation of B-lymphocytes that are morphologically normal but biologically immnature.
leading to a loss of function. Lymphocytes normally function to fight infection. In B-CLL, however, lymphocytes accumulate in the blood and bone marrow and cause swelling of the lymph nodes. The production of normal bone marrow and blood cells is reduced and patients often experience severe anemia as well as low platelet counts. This can pose the risk of life-threatening bleeding and the development of serious infections because of reduced numbers of white blood cells.
To further understand diseases such as leukemia it is important to have snitable cell lines that can be used as tools for research on their etiology, pathogenesis and biology. Examples of malignant human 11-lymphoid cell lines include pre-B
acute lymphoblasticleukemia (Reh), diffuse large cell lymphoma (WSU DLCL2), and WaIdenstroill's macroglobulinen ia (WSU- Wlvl). Unfortunately, many of the existing cell lines do not represent the clinically most common -types of leukemia and lymphoma.
The use of Epstein Barr Virus (E13V) infection in vitro has resulted in some C.L.I.
?0 derived cell lines, in particular B-CLL cells lines, that are representative of the malignant cells. The phenotype of these cell lines is different than that: of the in vivo tumors and instead the features of B-CLL, lines tend to be similar to those of lymphoblastoid cell lines. Attempts to imrnor`ialize B--CLL cells with the aid of EBV intection have had little success. The reasons for this are unclear but it is known that it is not due 1o a lack of Eb\
receptor expression, binding or uptake. Wells et al. found that 1-3-CI _L
cells were arrested in the Gl/S phase of the cell cycle and that transformation associated ED V
DNA was not expressed. This suggests that the interaction of E13V with 13-CLL cells is different from that with normal B cells- EBV-transformed CLI, cell lines moreover appear to differentiate, possessing a morphology more similar t:o lymphoblastoid cell lines (LCL) immortalized by EBV.
An EBV-negative CLL cell line, WSU-C1.1,, has been established previously (Mohammad et al., (1996) Leukemia 10(1):130-7). However, no oilier such cell lines are known.

Various mechanisms play a role in the body's response to a disease state, including cancer and CLL. For example, CD4+ T helper cells play a crucial role in an effective immune response against various malignancies by providing stimulatory factors to effector cells. Cytotoxic T cells are believed to be the most effective cells to eliminate cancer cells, and T helper cells prime cytotoxic T cells by secreting Th1 cyiohines such as IL-2 and IFN-y. In various malignancies, T helper cells have been shown to have an altered phenotype compared to cells found in healthy individuals- One of the prominent altered features is decreased ThI cytokine production and a shift to the production Of T112) cytohines. (See, e.g., Kiani, et at., flaematologica 88:754-761 (2003);
Maggio, et al., Ann Oncol 13 Suppl 1:52-56 (2002); Ito, et al., Cancer 85:2359-2367 (1999);
Podhorecka, et al., Leuk Res 26:657-660 (2002); Tatsumi, et al-, J Exp Ivied 196:619-628 (2002 );
Agarwal, et al., Immunol Invest 32:17-30 (2003); Smyth, ti al., Ann Sung Oncol 10:435--162 (2003); Contasta, et al., Cancer Biother Radiopliarrn 18:549-557 (2003);
Lauerova, et al., Neoplasma 49:159-166(2002).) Reversing that cytokine shift to a Thl profile has been demonstrated to augment anti-tumor effects off cells. (See Winter, et al., Immunology 108:409-419 (2003); .lnagawa, et al., Anticancer Res 18:3957-3964 (1998).) Mechanisms underlying the capacity of tumor cells to drive the cytokine expression of T helper cells from Thl to Th2 include the secretion of cytokines such as 1L-10 or TGF-Ii as well as the expression of surface molecules interacting with cells Of the immune system. OX-2/CD200, a molecule expressed on the surface of dendritic cells which possesses a high degree of homology to molecules of the irnmunoglobulin gene family, has been implicated in immune suppression (Gorczynski et al., Transplantation 61_ .106-1114 (1998)) and evidence that OX-2/CD200--expressing cells can inhibit the stimulation of Thl cytokine production has been provided. Gorczynski et al.
demonstrated in a mouse model that infusion of OX-2/CD200 Fc suppresses the rejection of tumor cells in an animal model using leukaemic tumor cells (Clip Exp Innnunol 126:220-229 (2001)).
Improved methods for treating individuals suffering from cancer or CLL are desirahic, especially to the extent they can enhance the activity of T cells.

SUv1MARY
In one embodiment a CLL cell line of malignant origin is provided that is not established by immortalisation with EBV. The cell line was derived from primary CLL
cells and is deposited under ATCC accession no. PTA-3920. In a preferred embodiment, the cell line is CLL-AAT. CLL-AAT is a B-CLL cell line, derived from a B-CLL
primary cell.
In a fiurther aspect, the CLL-AAT cell line is used to generate monoclonal antibodies useful in the diagnosis and/or treatment of CLL. Antibodies maybe generated by using the cells as disclosed herein as immunogens, thus raising an immune response in animals from which monoclonal antibodies may be isolated. The sequence of such antibodies may be determined and the antibodies or variants thereof produced by recombinant techniques. In this aspect, "variants" includes chimeric, CDR-grafted, humanized and fully human antibodies based on the sequence of the monoclonal antibodies.
Moreover, antibodies derived from recombinant libraries ("phage antibodies") may be selected using the cells described herein, or polypeptides derived therefrom, as bait to isolate the antibodies on the basis of target specificity.
In a still further aspect, antibodies may be generated by panning antibody libraries 13 using primary CLL cells, or antigens derived therefrom, and further screened and/or characterized using a CLL cell line, such as, for example, the CLL cell line described herein. Accordingly, a method for characterizing an antibody specific for CLL
is provided, which includes assessing the binding of the antibody to a CLL, cell line.
In a further aspect, there is provided a method for identif),ing proteins uniquely expressed in CLL cells employing the CLL-AAT cell line, by methods well known to those skilled in the art, such as by immunoprecipitation followed by mass spectroscopy analyses. Such proteins may be uniquely expressed in the CLL-AAT cell line, or in cells t' t~rirtlai )% Cclis derived i IioiiZ CALA. patients, Small molecule libraries (many available commercially) may be screened using 23 the CLL-A.AT cell line in a cell-based assay to identify agents capable of modulating the growth characteristics of the cells. For example, the agents may he identified which modulate apoptosis in the CLL-AAT cell line, or which inhibit growth and/or proliferation thereof. Such agents are candidates for the development of therapeutic compounds.
Nucleic acids isolated from CLL,-AAT cell lines may be used in subtractive hybridization experiments to identify C11-specific genes or in micro array analyses gene chip experiments). Genes whose transcription is modulated in CLL cells may be identified. Polypeptide or nucleic acid gene products identified in this manner are useful as leads for the development of antibody or small molecule therapies for CLL.

In a preferred aspect, the CLL-AAT cell line may be used to identify internalizing antibodies, which bind to cell surface components which are internalized by the cell-Such antibodies are candidates for therapeutic use. In particular, single-chain antibodies, which remain stable in the cytoplasm and which retain intracellular binding activity; may be screened in this manner.

In yet another aspect, a therapeutic treatment is described in which a patient is screened for the presence of a polypeptide that is upregulated by a malignant cancer cell and an antibody that interferes with the metabolic pathway of the upregulated polypeptide is administered to the patient.
The present disclosure further is directed to methods wherein a determination is made as to whether OX-2/CD200 is upregulated in a subject and, if so:
administering to the subject a therapy that enhances immune response. Up-egulat.ion of OX2/CD200 can be determined by measuring OX2/CD200 levels directly, or by monitoring the level of any marker that correlates with OX2/CD200. Suitable imrnunomodulatory therapies include the administration of agents that block negative regulation of T cells or antigen presenting cells, administration of agents that enhance positive co-stimulation of T cells, cancer vaccines, general adjuvants stimulating the immune system or treatment with cytokines such as IL-2, GNI-CSF and IFN-gamma. In particularly useful embodiments, the therapy that enhances immune response includes the administration of a polypeptide that binds to OX-2/CD200, optionally in combination with one or more other immunomochtlatory therapies. In another embodiment, the polypeptide binds to an OX-2/CD200 receptor.

In another aSpeCt, methods in accordance with this disclosure are used to treat a disease state in which OX-2/CD200 is upregulated in a subject by administering a polypeptide that binds to OX-2/CD200 or an OX-2/CD200 receptor to the subject afflicted with the disease state. In one embodiment, the disease state treated by these methods includes cancer, specifically, in other embodiments, CLL.
In a particularly useful embodiment, a cancer therapy in accordance with this disclosure includes i) administering an antibody that interferes with the interaction between CD200 and its receptor to block immune suppression, thereby promoting eradication of the cancer cells; and ii) administering a fusion molecule that includes a CD200-targeting portion to directly kill cancer cells. Alternatively, the antibody directly kills the cancer cells through complement-mediated or antibody-dependent cellular cytotoaicityy.

In another embodiment in accordance with the present disclosure, methods are provided for monitoring the-progress of a therapeutic treatment. The method involves administering an immunomodulat:ory therapy and determining OX-2/CD200 levels in a subject at least twice to determine the effectiveness of the therapy.
In another aspect, the present disclosure provides methods for assessing the immunorriodulatory effect of molecules expressed by cancer cells. In these methods a molecule that is expressed or upregulated by a cancer cell is identified (e.g., experimentally or from a database). Cancer cells, lymphocytes and the molecule that is expressed or upregulated by a cancer cell are administered to a subject and the rate of I growth of the cancer cells is monitored. The number of lymphocytes administered is predetermined to be either a.) sufficient to slow the growth of the cancer cells or b) insufficient to slow the growth of cancer cells. The molecule that is expressed or upregulated by a cancer cell can be administered as the molecule or the active portion of the molecule. itself, or by administering cells that produce or express the molecule or portions thereof natunrally, or by administering cells that have been engineered to produce or express the molecule or portions thereof. If any change in the growth rate Of tilt cancer cells is observed compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the molecule is deemed to have an inimunomodulatory effect. For example, if the number of lymphocytes administered is sufficient to slow the growth of the cancer cells and the rate of growth of the tumor cells observed is higher compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the molecule that is expressed or upregulated by a cancer cell is considered immunosuppr'esstve. As another example, ifthe number of lymphocytes administered is insufficient to slow the growth of the cancer cells and the 23 rate of growth of the tumor cells observed is lowered compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the compound is considered immune enhancing. Once the immunomodulatory effect of the molecule is established, compounds that either enhance or inhibit the activity of the molecule can be identified in accordance with embodiments described herein. The enhancing or inhibiting effect can be the result of direct interaction with tyre molecule expressed or upregulated by the cancer cell or may be the result of an interaction with other molecules in the metabolic pathway of the compound expressed or upregulated by the cancer cell.
In another aspect, the present disclosure provides methods for assessing the imn'runomodulatory effect of a compound. In these methods cancer cells, lymphocytes and the compound to be assessed are administered to a subject and the rate of growth of the cancer cells is monitored. The number of lymphocytes administered is predetermined to be either a) sufficient to slow the growth of the cancer cells or b) insufficient to slow the growth of cancer cells. The compound to be assessed can be administered as the compound itself, or by administering cells that produce the compound naturally, or by administering cells that have been engineered to produce the compound. If any change in the growth rate of the cancer cells is observed compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the compound is deemed to have an iminunomodulatory effect. For exarnpie, if the number of lymphocytes administered is sufficient to slow the growth of the cancer cells and the rate of growth of the tumor cells observed is higher compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the compound is considered immunosuppressive. As another example, if the number of lymphocytes administered is insufficient to slow the growth of the cancer cells and the rate of growth of the tumor cells observed is lowered compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the compound is considered immune enhancing.
BRIEF DESCRIPTION OF TH FIGURES
Fig. 1 schematically illustrates typical steps involved in cell surface panning of antibody libraries by magnetically-activated cell sorting (MACS).
Fig. 2 is a graph showing the results of whole cell ELISA demonstrating binding of selected scFv clones to primary B-CLL cells and absence of binding to normal human PBtv1C. The designation 2 +3 in this and other figures refers to negative control wells stained with Mouse Anti-HA and detecting antimouse antibodies alone- The designation RSC-S Library in this and other figures refers to soluble antibodies prepared from original rabbit scFv unpanned library. The designation R3/RSC-S Pool in this and other figures refers to soluble antibodies prepared from the entire pool of scFv antibodies from round 3 of panning. Anti-CDS antibody was used as a positive control to verify that equal numbers of B-CLL and PBIvIC cells were plated in each well.
Figs. 3a and 3b show the results of whole cell ELISA comparing binding of selected scFv antibodies to primary B-CLL cells and normal primary human B
cells.
Anti-CD19 antibody was used as a positive control to verify that equal numbers of B-CLL and normal B cells were plated in each well. Other controls were as described in the legend to Fig. 2.

Figs. 4a and 4b show the results of whole cell BLISA used to determine if scFv clones bind to patient-specific (i.e. idiotype) or blood type-specific (i.e.
FILA) antigens.
Each clone was tested for binding to PBMC isolated from 3 different B-CLL
patients.
Clones that bound to only one patient sample were considered to be patient or blood type-specific.
Figs. 5a and 5b show the results of whole cell FLI.SA comparing binding of sci v clones to primary B-CLL cells and three human leukemic cell lines. Ramos is a mature B
cell line derived from a Burkitt's lymphoma. RL is a mature B cell line derived from it non-Ilodgkin's lymphoma. TF-l is an erythroblastoid cell line derived from an erythroleukemia.
Figs. 6a, 61) and 6c show the results of whole cell ELISA comparing binding of scFv clones to primary B-CLL cells and CLL-AAT, a cell line derived from a B-CLL
patient. TF-I cells were included as a negative control.
Fig. 7 shows the binding specificity of scFv antibodies in accordance with this disclosure as analyzed by 3-color flow cytometry. In normal peripheral blood mononuclear cells, the antigen recognized by scFv-9 is moderately expressed on B
lymphocytes and weakly expressed on a subpopulation of T lymphocytes. PBMC.
from a normal donor were analyzed by 3-color flow cytometry using anti-CDS-FITC, anti-CD19-PerCP, and scFv-9/Anti-LHA-biotin/streptavidin-PE.
Figs. 8a, 8b and 8c show the expression levels of antigens recognized by scFv antibodies in accordance with this disclosure. The antigens recognized by scFv-3 and scFv-9 are overexpressed on the primary CLL tumor from which the CLL-AAT cell line was derived. Primary PBMC from the CLL patient used to establish the CLL-A A r cell line or PBMC. from a normal donor were stained with scFv antibody and analyzed by flow cytometry. ScFv-3 and scFv-9 stain the CLL cells more brightly than the normal PBMC as measured by the mean fluorescent intensities.
Figs. 9A - 9C provide a summary of CDR sequences and binding specificities of selected scFv antibodies.
Fig. 10 is Table 2 which shows a summary of flow cytometry results comparing expression levels of scFv antigens on primary CLL, cells vs. normal PBMC as described in Figs 8a-8c.
Fig. 11 is a Table showing a summary of Flow cytometry results comparing expression levels of scFv-9 antigen with the percentage ofCD38' cells in peripheral blood mononuclear cells isolated from ten CLL patients.

Fig. 12 shows the identification of scFv antigens by iiwmunopreeipitation and mass spectrometry_ CLL-AAT cells were labeled with a solution of 0.5mg/m l sulfo-NHS-LC-biotin (Pierce) in PBS, pH8.0 for 30'. After extensive washing with PBS
to remove unreacted biotin, the cells were disrupted by nitrogen cavitation and the microson-ial fraction was isolated by differential centrifugation. The microsomal fraction was resuspended in 1'TP40 Lysis Buffer and extensively precleared with normal rabbit serum and protein A SepharoseT"M. Antigens were immunoprecipitated with I-IA-tagged scFv antibodies coupled to Rat Anti-1L4 agarose beads (Roche). Following immunoprecipitation, antigens were separated by SDS-PAGE and detected by Western blot using streptavidin-alkaline phosphatase (AP) or by Coowassie G-2j0 staining. ScFv-7, an antibody which doesn't bind to CLL-AAT cells, was used as a negative control.
Antigen bands were excised from the Coomassie-stained gel and identified by mass spectrometry (MS). NLALDI-MS was performed at the Proteomies Core Facility of The Scripps Research Institute (La Jolla, CA). tLC/Jv1S/1\4S was performed at the Ilarvard 1 Microchemistry Facility (Cambridge, Iv1A).
Fig. 13 shows that three scFv antibodies bind specifically to 293-EBI'A cells transiently transfected with a human OX-2/CD200 cDNA clone. An OX-2/CD200 cDNA
was cloned from CLL cells by P,.]'-PCR and inserted into the mammalian expression vector pCEP4 (Invitrogen). PCEP4-CD200 plasrnid or the corresponding empty vector pCEP4 was transfected into 293-EBNA cells using PolyfeciT"' reagent (QI: G
EN). Two days after transfection, the cells were analyzed for binding to scFv antibodies by flow cytorn etin'.

Fi + O' 2 / ~i1200 ans fe c r su r tg. r 'i shows that the presence cf _ C1_ t "'s .,ttI ]ls tite:I i:i down-regulation ofTh1 cytokines such as IL-'21 and IFN-'y. Addition of the anti-O
2/CD200 antibody at 30 tghnl hilly restored flit 'I'M response.
Fig_ 1 5 shows that the presence of CLL cells in a mixed lymphocyte reaction resulted in down-regulation of the Thl response for IL-2.
Fig. 16 shows that the presence of CLL, cells in a mixed lymphocyte reaction resulted in down-regulation of the Thl response for FFN-y"
Figs. 17A and B show the mean +/-SD of tumor volumes for all groups of )10D/SCID mice were injected subcutaneously with dlx100 RAJ1 cells either in the presence or absence of human PBL, cells.
Fig. 18 shows the results of statistical analyses performed using 2 parametric tests (Student's t-test and Welch's test) and one non parameh is test (the WVdcox teslj.

L) Fig. 19A shows F1.JSA results of representative IgGI kappa clones after round panning on CD200-Fc captured on goat anti-mouse IgG Fc antibody, Fig. I 9B shows FLISA results of representative IgG2a kappa clones after round panning on CD200-Fc captured on goat anti-mouse IgG Fc antibody.
Fig. 19C shows 1EL[SA. results of representative IgGI kappa clones after round panning on CD200--Fc directly coated on microtiter wet Is.

Fig. 19D shows ELISA results of representative IgG2a kappa clones after round panning on CD200-Fc directly coated on nlicrotiter Wells.
Fig. 20A shows flow cytomeiry results of representative IgGI clones selected on CD200-Fe captured with goat anti-mouse IgG Fe.
Fig. 20B shows flow cytometay results ofrepresentative IgG2a clones selected on CD200-Fc captured with goat anti-mouse IgG Fe.
Fig. 20C shows flow cytometry results of representative IgG I clones selected on directly coated CD200-Fc.
ld Fig. 20D shows flow cytometry results of representative IgG2a clones selected on directly coated CD200-Fe.
Fig. 2 IA shows deduced amino acid sequence of heavy chain complenientarity regions of CD200-specific clones.
Fig. 21B shows deduced amino acid sequence of heavy chain complementarily regions of CD200-specific, clones.
Fig. 22 shows ability of selected clones to block the interaction of CD200 with its receptor (CD200R) in a fluorescent bead assay.
Fig. 23 shows deduced amino acid sequences of selected CD200 Fabs for cllllllerIzalioll.
Fig. 24 shows ELISAresults of chimeric IgG obtained from the culture supernatant of a small-scale transient transfection.
Fig. 25 shows bead inhibition assay results on purified IgG showing dial all antibodies directed against CD200 blocked the receptor ligand interaction very well-Figs. 26A and 26B show that tilt presence of CLL cells completely abrogated 30 1FN--gamma and most of I1_.-2 production observed in the mixed lymphocyte reaction but that the presence of any of the antibodies allowed for production of these ThI
cytokines.
Fig. 26C shows that fL 10 production was downregulated in the presence of the antibodies.

Fig. 27 shows the ability to ]till CD200 expressing tumor cells in an antibody -dependent cell-mediated eytotoxicity assay (ADCC). All of the mouse chimeric antibodies produced similar levels of lysis when cultured with CD200 positive cells.
Fig. 28 shows a representative example of 10 experiments using different PBL
donors compared to a group that received tumor cells only, analyzed by a 2-tailed unpaired Student's t-test. Sigii ficant differences were observed in the groups that received 5 or 10 million PBLs, but not in the group that received 1 million P13Ls from Day 32 on.

Fig. 29 shows a representative example of 10 experiments using different PB1110 donors compared to a group that received tumor cells only, analyzed by 2-tailed unpaired Student's t-test. Significant differences were observed in the groups that received 10 million PBLs for both donors, but not in the group that received 2 million PBLs horn Day $ on.
Fig_ 30(a) shows RIV71 cells infected with CD200 expressing tumor cells l 5 transduced from a lentivirus vector appeared to grow somewhat more slowly than its parental RAJI cells. The growth difterence between the (ransduced and parental cells did not reach statistical significance.
Fig. 30(b) shows the presence of PBLs reduced tumor growth in the RA.I1 cells transduced with the reversed CD-100 (non functional (AD200) by up to f1 1%
when 5 or 20 10x1(}6 PBLs were injected indicating that this particular donor rejects RAJI tumor cells very strongly.
Fig. 30(c) shows results indicating that CD200 expression on tumor cells does indeed prevent the immune system from slowing tumor growth. Also this Out, demonstrates the usefulness of the RAJI/PBL model to assess immunosuppressive 25 compounds or molecules.
lags. 31 (a)-(d) show whether the effects seen in the RAJI/PBL model can also be observed with other tumor cell models.
Fig. 31(a) shows Naulalwa tumor cells resulted in rapid tumor growth with no significant difference between transduced and parental cells.
30 Fig 31(b) shows the presence of PBLs slowed tumor growth by about 3042. IN
2-tailed Stuudent's t-test results showed that the differences between the PBL
treated group versus groups that received only tumor cells were statistically significant.
Figs. 31(c) and (d) show tumor growth in the groups that received CD200 expressing Narnalwa cells and PBLs was similar to the tumor growth in the group ilia[

received Namalwa cells. These data confirm that CIa?00 expression on tumor cells prevents slowing of tumor growth by the human irulrune system.
Fig. 32 shows results demonstrating that tim RATI PBL mode] is an efficient way to assess efficacy of immune-enhancing compounds.
Figure 33 shows that CD200 expression on the tumor cells prevented the immune cells from reducing rumor growth.
Figure 3d shows that CD200 expression on the tumor cells prevented the immune cells from reducing tumor growth.

DETAILED DESCRIPTION OF PREFERRED EIvIBODUvlENTS

In accordance with the present disclosure, methods are provided for determining -whether OX-2/CD200 is upregulated in a subject and, if so, administering to the subject a therapy that enhances immune response. Illustrative examples of suitable immunomodulatory therapies include the administration of agents that block negative regulation of T cells or antigen presenting cells (e.g., anti-CTLA4 antibodies, an(i-PD-LI
antibodies, anti-PDL-2 antibodies, anti-PD-1 antibodies and the like) or the administration of agents that enhance positive co-stimulation of T cells (e.g., anti-CD40 antibodies or anti 4-1BB antibodies) or administration of agents that increase ITK_ cell number or T-cell activity (e.g., anti-CD200 antibodies alone or in combination with inhibitors such as IIvIiDs, thalidomide, or thalidomide analogs).
Furtherrnore, immunomodulatory therapy could be cancer vaccines such as dendrinc cells loaded with tumor cells, tumor RNA or tumor DNA, tumor protein or tumor peptides, patient derived heat-shocked proteins (asp's) or general adjuvants Stimulating the i mmune system at various leVelS such as CpG, LnivacUM, BlostlllhM, RlbominylTII, lmudonlu, Bronchovaxunli I
or art)~ other compound activating receptors ofthe innate immune system (e.g., toll like receptors j. Also, ininlunonlodulatory therapy could include treatment with cytoLines such as IL-2, Gf\,]-CSF and IFN-gamma.

In particularly useful embodiments, the therapy that enhances immune response is the administration of a polypeptide that binds to OX-2/CD200, alone nr in colllbinationi with one ofthe previously mentioned imninnomodulatory therapies. In general, the polypeptides utilized in the present disclosure can be constructed using different techniques which are known to those skilled in the art. In one embodiment, the polypeptides are obtained by chemical synthesis. In other embodiments, the pol.;'peptiLies are antibodies or constructed from a fragment or several fragments of one or more antibodies.
Preferably, the polypeptides utilized in the methods of the present disclosure are obtained from a CLL, cell line. "CLL", as used herein, refers to chronic lymphocytic leukemia involving any lymphocyte including, but not limited to, various developmental stages of B cells and T cells including, but not limited to.. B cell CLL B-CLL, as used herein, refers to leukemia with a manure B cell phenotype which is CD' , ('17)23 %
CD20d"" r, sIgdim and arrested in GOIGI of the cell cycle. In a further aspect, the CLL
cell line is used to generate polypeptides, including antibodies, useful in the diagnosis and/or treatment of a disease state in which OX-2/CD200 is upregulated, including cancer and CLL.
As used herein, the term "antibodies" refers to complete antibodies or antibody fragments capable of binding to a selected target. Included are Fab, by, seFv, Fab' and F(ab')2, monoclonal and polyclonal antibodies, engineered antibodies (including chimeric, 13 CDR-grafted and humanized, hilly human antibodies, and artificially selected antibodies), and synthetic or semi-synthetic antibodies produced using phage display or alternative techniques. Small fragments, such as Fv and scF v, possess advantageous properties for diagnostic and therapeutic applications on account of their small size and consequent superior tissue distribution.
Antibodies may be generated by using the cells as disclosed herein as immunogens, thus raising an immune response in animals from which monoclonal antibodies may be isolated. The sequence of such antibodies may be determined and the n antibodies or variants thereof produced by recombinant techniques. In this aspect, "variants" includes chimeric, CDR-grafted, humanized and fully human antibodies based on the sequence of the monoclonal antibodies, as well as polypeptides capable of binding to OX-2/CD200.
Moreover, antibodies derived from recombinant libraries ("phage antibodies") may be selected using the cells described herein, or polypeptides derived therefrom, as bait to isolate the antibodies or polypeptides on the basis of target specificity.
In a still further aspect, antibodies or polypeptides may be generated by panning antibody libraries using primary CLL cells, or antigens derived therefirom, and further screened and/or characterized using a CLL cell line, such as, for example, The CLL cell line described herein. Accordingly, a method for characterizing an antibody or polypeptide specific for CLL is provided, which includes assessing the binding ofthe antibody or polypeptide to a CLL cell line.

Preparation of Cell Lines Cell lines may be produced according to established methodologies known to those skilled in the art. In general, cell lines are produced by culturing primary cells derived from a patient until immortalized cells are spontaneously generated in culture.
These cells are then isolated and further cultured to produce clonal cell populations Or cells exhibiting resistance to apoptosis.
For example, CLL cells may be isolated from peripheral blood drawn from a patient suffering from CLL. The cells may be washed, and optionally immunotyped in order to determine the type(s) of cells present. Subsequently, the cells may be cultured in a medium, such as a medium containing IL-4. Advantageously, all or part of the medium is replaced one or more times during the culture process. Cell lines may be isolated thereby, and will be identified by increased growth in culture.
In one embodiment a CLL cell line of malignant origin is provided that is not established by immortalization with EBV. "Malignant origin" refers to the derivation of the cell line from malignant CLL primary cells, as opposed to non-proliferating cells which are transformed, for example, with EBV. Cell lines useful according to this disclosure may be themselves malignant in phenotype, or not. A CLL cell having a "malignant" phenotype encompasses cell growth unattached from substrate media characterized by repeated cycles of cell growth and exhibits resistance to apoptosis. .I he cell line, which was derived from primary CLL cells, is deposited under ATCC
accession no. PTA-3920. In a preferred embodiment, the cell line is CLL-AAT. CLL-AAT is a B-CLL cell line, derived tom a B-CLL primary cell.
In one embodiment, proteins uniquely expressed in CLL cells are identified employing the CLL-AAT cell line by methods well known to those skilled in the art, such as by immnunoprecipitation followed by mass spectroscopy analyses. Such proteins may be uniquely expressed in the CLL-AAT cell line, or in primary cells derived from CLL
patients.
Small molecule libraries (many available commercially) may be screened using the CLL-AAT cell line in a cell-based assay to identify agents capable of modulating the growth characteristics of the cells. For example, the agents may be identified which modulate apoptosis in the CLL-AAT cell line, or which inhibit growth and/or proliferation thereof. Such agents are candidates for the development of therapeutic corn pannds.
Nucleic acids isolated from CLL .AAT cell lines may be used in subrTactive hybridization experiments to identify CLL-specific genes or in micro array analyses (e.g-, gene chip experiments). Genes whose transcription is modulated in C.LL cells may be identified. Polypeptide or nucleic acid gene products identified in this manner are useful as leads for the development of antibody or small molecule therapies for CLL.
In one embodiment, the CLL-AAT cell line may be used to identify internalizing antibodies, which bind to cell surface components and are then internalized by the cell.
Such antibodies are candidates for therapeutic use. In :particular, single-chain antibodies., which remain stable in the cytoplasm and which retain intracellular binding activity, may be screened in this manner.

Preparation of Monoclonal Antibodies Recombinant DNA technology may be used to improve the antibodies produced in accordance with this disclosure. Thus, chimeric antibodies may be constructed in order to decrease the immunogenicity thereof in diagnostic or therapeutic applications.
Moreover, immunogenicity may be minimized by humanizing the antibodies by CDR
grafting and, optionally, framework modification. See, U.S. Patent No.
5.225,539-Antibodies may be obtained from animal serum, or, in the case of monoclonal antibodies or fragments thereof produced in cell culture. Recombinant DNA
technology may be used to produce the antibodies according to established procedure, in bacterial or preferably mammalian cell culture. The selected cell culture system preferably secretes the antibody product.
In another embodiment, a process for the production of an antibody disclosed herein includes culturing a host, e.g. L. coli or a mammalian cell, which has been transformed with a hybrid vector. The vector includes one or more expression cassettes containing a promoter operably linked to a first DNA sequence encoding a signal peptide linked in the proper reading frame to a second DNA sequence encoding the antibody protein. The antibody protein is then collected and isolated. Optionally, the expression cassette may include a promoter operably linked to polycistronic_ for example bicistronic, DNA sequences encoding antibody proteins each individually operably linked to a signal peptide in the proper reading frame.

Multiplication of hybridoma cells or mammalian host cells in vitro is carried out in suitable culture media, which include the customary standard culture media (such as, for example Dulbecco's Modified Eagle Medium (DIvIEIvl) or RPIv1I 1640 medium), optionally replenished by a mammalian serum (e.g. fetal calf serum), or trace elements and growth sustaining supplements (e.g. feeder cells such as normal mouse peritoneal exudate cells, spleen cells, bone marrow macrophages, 2-aminoethanol, insulin, transferrin, low density lipoprotein, oleic acid, or the like). Multiplication of host cells which are bacterial cells or yeast cells is likewise carried out in suitable culture media known in the art. For example, for bacteria suitable culture media include medium LE, NZCYNI, NZYM, NZ1vM, Terrific Broth, SOB, SOC, 2 x YT, or M9 Minimal Medium.
For yeast, suitable culture media include medium YPD, YEPD, Minimal Medium, or Complete Minimal Dropout Medium.
In vitro production provides relatively pure antibody preparations and allows scale-up to give large amounts of the desired antibodies. Techniques for bacterial cell, 1 -5 yeast, plant, or mammalian cell cultivation are known in the art and include homogeneous suspension culture (e.g. in an airlift reactor or in a continuous stirrer reactor), and immobilized or entrapped cell culture (e.g. in hollow fibres, microcapsules, on agarose microbeads or ceramic cartridges).
Large quantities of the desired antibodies can also be obtained by multiplying mammalian cells in vivo. For this purpose, hybridoma cells producing the desired antibodies are injected into histocompatible mammals to cause growth of antibody-producing tumors. Optionally, the animals are prirned with a hydrocarbon, especially mineral oils such as pristane (tet;amethyl pentadecane), prior to the injection. after one to three weeks, the antibodies are isolated from the body fluids of those mammals. For example, hybridoma cells obtained by fusion of suitable myeloma cells with antibody-producing spleen cells from Balb/c mice, or transfected cells derived from hybridorna cell line Sp2/0 that produce the desired antibodies are injected intraperitoneally into Balb/c mice optionally pre-treated with pristine. After one to two weeks; ascitic fluid is taken from the animals.
The foregoing, and other, techniques are discussed in, for example, Kohler and Milstein, (1975) Nature 2_56A95-497-1 U.S. Patent 1,]o, 1,376,110; Barlow and Lane, Antibodies: a Laboratory Manual. (1988) Cold Spring Harbor. Techniques for the preparation of recombinant antibody molecules are described in the above references and also in. for example 1c) '9/097/08320; U.S. Patent No. 5,427,908; U.S. Patent No. 5,508,717; Smith, 1985, Science, Vol. 225, pp 1315-1317; Parmley and Smith, 1988, Gene 73; pp 305-318;
De La Cruz at al., 1988, Journal of Biological Chemistry, 263 pp 4318-4322; U.S.
Parent No.
;,403,484; U.S. Patent No. 5223409; W088/06630; W092!15679, U.S. Patent No.
5780279; U.S. PatentNo. 5571698; U.S. PatentNo. 6010136: Davis at at., 1999, Cancer Metastasis Rev., 18(4):421-5; Taylor, et al., Nucleic Acids Research 20 (1992): 6287-0295; Tomizuka at al., Proc. Natl. Academy of Sciences USA 97(2) (2000): 722-727.

The cell culture supernatants are screened for the desired antibodies, preferentially by immunofluorescent staining of CLL cells, by imrnunoblotting, by an enzyme immunoassay, e.g. a sandwich assay or a dot-assay, or a radioimmunoassay.
For isolation ofthe antibodies, the immunoglobulins in the culture supernatants or in the ascitic fluid may be concentrated, e.g. by precipitation with ammonium sulfate, dialysis against hygroscopic material such as polyethylene glycol, filtration through 13 selective membranes, or the like. If necessary and/or desired, the antibodies are purified by the customary chromatography methods, for example gel filtration, ion-exchange chromatography, chromatography over DE_A.E-cellulose and/or (inimuno-) affinity chromatography, e.g. affinity chromatography with one or more surface polypeptides derived from a CLL cell line according to this disclosure, or with Protein _A
or G.
Another embodiment provides a process for the preparation of a bacterial cell line secreting antibodies directed against the cell line characterized in that a suitable mammal, for example a rabbit, is immunized with pooled CLL patient samples. A phage display library produced from the immunized rabbit is constructed and panned for the desired antibodies in accordance with methods well known in the art.

Hybridoma cells secreting the monoclonal antibodies are also contemplated. The preferred hybridoma cells are genetically stable, secrete, monoclonal antibodies described herein of the desired specificity and can be activated from deep-frozen cultures by thawing and recloning.
In another embodiment, a process is provided for the preparation of a hybridorna cell line secreting monoclonal antibodies directed to the CLL cell line is described herein.
In that process, a suitable mammal, for example a Balb/c mouse, is immunized with one or more polypeptides or antigenic fragments thereof derived from a cell described in this disclosure, the cell line itself, or an antigenic carrier containing a purified polypeptide as described. Antibody-producing cells of the immunized mammal are grown briefly in culture or fused with cells of a suitable myeloma cell line. The hybrid cells obtained in the fusion are cloned, and cell clones secreting the desired antibodies are selected. For example, spleen cells of Balb/c mice immunized with the present cell line are tiised with cells of the myeloma cell line PAl or the myeloma cell line Sp)/O-Ag 1-1, the obtained hybrid cells are screened for secretion of the desired antibodies, and positive hyhridoma cells are cloned.
Preferred is a process for the preparation ofa hybridorna cell line, characterized tit that Balb/c mice are immunized by injecting subcutaneously and/or intraperituncally between 10 and 107 cells of a cell line in accordance with this disclosure several times, e.g. four to six times, over several months, e.g. between two and four months.
Spleen cells from the immunized mice are taken two to four days after the last injection and fused with cells of the myeloma cell line PAl in the presence of a fusion promoter, preferably polyethylene glycol, Preferably, the myeloma cells are fused with a three- to twenty-foul excess of spleen cells from the immunized mice in a solution containing about '30% to about 501io polyethylene glycol of a molecular weight around 4000. A licr the fusion, the cells are expanded in suitable culture media as described hereinbefore, supplemented with a selection medium, for example HAT medium, at regular intervals in order to prevent normal myeloma cells from overgrowing the desired hybridoma cells.
In a further embodiment, recombinant DNA comprising an insert coding fur a heavy chain variable domain and/or for a light chain variable domain of antibodies directed to the cell line described hereinbefore are produced. The term DNA
includes coding single stranded DNAs, double stranded DNAs consisting of said coding Lt1-lAs and of complementary DNAs thereto, or these complementary (single stranded) DNAs themselves.
Furthermore, DNA encoding a heavy chain variable domain andlor a light chain variable domain of antibodies directed to the cell line disclosed herein can be enzymatically or chemically synthesized DNA having the authentic DNA, sequence coding for a heavy chain variable domain and/or tot the light chain variable domain, or a mutant thereof. A mutant of the authentic DNA is a DNA encoding a heavy chain variable domain and/or a light chain variable domain of the above-mentioned antibodies in which one or more amino acids are deleted or exchanged with one or more other amino acids. Preferably said modification(s) are outside the MRS of the heavy chain variable domain and/or of the light chain variable domain of the antibody in humanization and expression optimization applications. The term mutant DNA also embraces silent mutants wherein one or more nucleotides are replaced by other nucleotides with the new codons coding for the same amino acid(s). The term mutant sequence also includes a degenerate sequence. Degenerate sequences are degenerate within the meaning of the genetic code in that an unlimited number of nucleotides are replaced by other nucleotides without resulting in a change of the amino acid sequence originally encoded. Such degenerate sequences may be useful due to their different restriction sites and/or frequency of particular codons which are preferred by the specific host, particularly E.
coli, to obtain an optimal expression of the heavy chain marine variable domain and/or a light chain murine variable domain.
The term mutant is intended to include a DNA mutant obtained by in vitro niutagenesis of the authentic DNA according to methods known in the art.
For the assembly of complete tetrameric immunoglobulin molecules and the expression of chimeric antibodies, the recombinant DNA inserts coding for heavy and 13 light chain variable domains are fused with the corresponding DNAs coding for heavy and light chain constant domains, then transferred into appropriate host cells, for example after incorporation into hybrid vectors.
Recombinant DNAs including an insert coding for a heavy chain murine variable domain of an antibody directed to the cell line disclosed herein fused to a human constant domain g, for example yl, y2, 'y3 or 'y4, preferably yl or 'y4l are also provided.
Recombinant DNAs including an insert coding for a tight chain murine variable domain of an antibody directed to the cell line disclosed herein fused to a human constant domain x or 2, preferably xr are also provided Another embodiment pertains to recombinant DNAs coding for a recombinant polypeptide wherein the heavy chain variable domain and the light chain variable domain are linked by way of a spacer group, optionally comprising a signal sequence facilitating the processing of the antibody in the host cell and/or a DNA coding for a peptide facilitating the purification of the antibody and/or a cleavage site and/or a peptide spacer and/or an effector molecule.
The DNA coding for an effector molecule is intended to be a DNA coding for the effector molecules useful in diagnostic or therapeutic applications. Thus, effector molecules which are toxins or enzymes, especially enzymes capable of catalyzing the activation of prodntgs, are particularly indicated. The DNA encoding such an effector molecule has the sequence of a naturally occurring enzyme or toxin encoding DNA, or a mutant thereof, and can be prepared by methods well known in the art.

Uses of the Present Antibodies/Po yl2ep ides The polypeptides and/or antibodies utilized herein are especially indicated for diagnostic and therapeutic applications.
The present antibodies can be administered as a therapeutic to cancer patients, especially, but not limited to CLL patients. In some embodiments, the antibodies are capable of interfering with the interaction of CD200 and its receptors. This interference can block the immune suppressing effect of CD200. By improving the immune response in this manner, such antibodies can promote the eradication of cancer cells.
The anti-CD200 antibody can also be administered in combination with other immunornodulatory compounds, vaccines or chemotherapy. For example, elimination of existing regulatory T cells with reagents such as anti-CD25 or cyclophosphamide is achieved in one particularly useful embodiment before starting anti-CD200 treatment.
Also, therapeutic efficacy of myeloablative therapies followed by bone marrow transplantation or adoptive transfer of T cells reactive with CLL cells is enhanced by anti-CD200 therapy. Furthermore, anti-CD200 treatment can substantially enhance efficacy of cancer vaccines such as dendritie cells loaded with CLL cells or proteins, peptides or RNA derived from such cells, patient-derived heat-shocked proteins (lisp's), tuinor peptides or protein. In other embodiments, an anti-CD200 antibody is used in combination with an immuno-stimulatory compound, such as CpG, toll-like receptor agonists or any other adjuvant, anti-CTLA-l antibodies, and the like. In yet other embodiments, efficacy of anti-CD200 treatment is improved by blocking of immunosuppressive mechanisms such as anti-PDLI and/or 2 antibodies, anti-IL-10 antibodies, anti-IL-6 antibodies, and the like. In yet other embodiments, efficacy of anti-CD200 treatment is improved by administration of agents that increase NK cell number or 'F-cell such as the small molecule inhibitor IMiDs, thalidomide, or thalidomide analogs).
Anti-CD200 antibodies in accordance with the present disclosure can also be used as a diagnostic tool. For example, using blood obtained from patients with hematopoietic cancers, expression of CD200 can be evaluated on cancer cells by F'-'ACS
analysis using anti-CD200 antibodies in combination with the appropriate cancer cell markers such as, e.g., CD38 and CD19 on CLL cells. Patients with CD200 levels at least 1.4I-fold above the levels found on normal B cells can be selected for treatment with anti-0)200 antibodies-In another example of using the present anti-CD200 antibodies as a diagnostic tool, biopsies from patients with malignancies are obtained and expression of CD200 is determined by FACS analysis using anti-CD200 antibodies. If tumor cells express CD200 at levels that are at least 1.4-fold higher compared to corresponding normal tissue, cancer patients are selected for immunomodulatory therapy. Inununomodulatory therapy can be anti-CD200 therapy, but can also be any other therapy affecting the patient's immune system. Examples of suitable irntnunomodulatory therapies include the administration of agents that block negative regulation of'1 cells or antigen presenting cells (e.g., anti-CTL_AA, anti-PD-LI, anti-PDL-2, anti-PD-1) or the administration of agents that enhance positive co-stimulation of T cells (e.g., anti-CD4O or antk1-IB13).
Furthermore, immunomodulatory therapy could be the administration of agents that increase NIL cell number or T-cell activity (e.g., anti-CD200 antibodies alone or in combination with inhibitors such as IMiDs, thalidomide, or thalidomide analogs) or the administration of agents that deplete regulatory T cells (e.g. anti-0)200 antibodies alone or in combination with ONTAK). Furthermore, immunomodulatory therapy could be cancer vaccines such as dendritic cells loaded with minor cells, tumor RNA or tumor DNA, tumor protein or tumor peptides, patient derived heat-shocked proteins (hsp's) or general adjuvants stimulating the immune system at various levels such as CpG, Luivac;
Biostim, Ribominyl, Inmulon, Brorrchovaxonr or any other compound activating receptors of the innate immune system (e.g., toll like receptors). Also, therapy could include treatment wvith cytolaries such as 1L-2, G lvl-CSF and IFN-garnma.
In another embodiment in accordance with the present disclosure, methods are provided for monitoring the progress and/or effectiveness of a therapeutic treahnent. I he method involves administering an inn unomodulatory therapy and determining (-)X-MD200 levels in a subject at least twice to determine the effectiveness of the therapy.
For example, pre-treatment levels of OX-2/0D200 can be ascertained and, after at least one administration of the therapy, levels of OX-2/CD200 can again be determined- A
decrease in OX-2/CD200 levels is indicative of an effective treatment.
Measurement of OX-2/CON levels can be used by the practitioner as a guide for increasing dosage amount or frequency of flit therapy. It should of course he understood that OX-2!CD200 levels can be directly monitored or, alternatively, any marker that correlates with LI -/CD200 can be monitored.

The present antibodies also may be utilized to detect cancerous cells in vivo.
This is achieved by labeling the antibody, administering the labeled antibody to a subject, and then imaging the subject. Examples of labels useful for diagnostic imaging in accordance with the present disclosure are radiolabels such as 1311, 1Illn, 123I, 99mTc;
3'`P, 1251,'H, ~aC
and mnnR.h, fluorescent labels such as fluorescein and rhodamine, nuclear magnetic resonance active labels, positron emitting isotopes detectable by a positron emission tomography ("PET") scanner, chemiluminescers such as luciferin, and enzymatic markers such as peroxidase or phosphatase. Short-range radiation emitters, such as isotopes detectable by short-range detector probes, such as a transrectal probe, can also be employed. The antibody can be labeled with such reagents using techniques known in the art. For example, see Wensel and Meares, Radioin-u-nunolm aging and Radioimmunotherapy,, Elsevier, N.Y. (1983), for techniques relating to the radiolabeling of antibodies. See also, D. Colcher et at., "Use of Monoclonal Antibodies as Radiopharrnaceuticals for the Localization of Human Carcinoma Xenografts in Athymic Mice", Meth.
Enzymol.
1 > 121: 802-816 (1986).

A radiolabeled antibody in accordance with this disclosure can be used for in vitro diagnostic tests. The specific activity of an antibody, binding portion thereof, probe, or ligand, depends upon the half-life, the isotopic purity of the radioactive label, and how the label is incorporated into the biological agent. In immunoassay tests, the higher the specific activity, In general, the better the sensitivity. Procedures for labeling antibodies with the radioactive isotopes are generally known in the art.
The radiolabeled antibody can be administered to a patient where is is localized to cancer cells bearing the antigen with which the antibody reacts, and is detected or "imaged" in vivo using known techniques such as radionuclear scanning using e.g., a gamma camera or emission tomography. See e.g., A. R. Bradwell et al., "Developments in Antibody Imaging", Monoclonal Antibodies for Cancer Detection and Therapy, R. W.
Baldwin et al., (eds.), pp. 65-85 (Academic Press 1985). Alternatively, a positron emission transaxial tomography scanner, such as designated Pet V1 located at Brookhaven 311 National Laboratory, can be used where the radiolahel emits positrons (e.g., 'C_ isp ~'0, and "N).

Fluorophore and chromophore labeled biological agents can be prepared from standard moieties known in the art. Since antibodies and other proteins absorb light having wavelengths up to about 310 nm, the fluorescent moieties should be selected to have substantial absorption at wavelengths above 310 run and preferably above 400 nm.
A variety of suitable fluorescers and chromophores are described by Shyer, Science, 162.526 (1968) and Brand, L. et at., Annual Review of Biochemistry, 41:843-868 (1972).
The antibodies can be labeled with fluorescent chromophore groups by conventional procedures such as those disclosed in U.S. Patent Nos, 3,940,475, 4.289,747, and 4,376,1 10.
In other embodiments, bispecific antibodies are contemplated. Bispecific antibodies are monoclonal, preferably human or humanized, antibodies that have binding specificities for at least two different antigens. In the present case, one of the binding specificities is for the CD200 antigen on a cancer cell, the other one is for any other antigen, and preferably for a cell-surface protein or receptor or receptor subunit.
Methods for mating bispecific antibodies are within the purview of those skilled in the art. Traditionally, the recombinant production of bispecific antibodies is based on the co-expression of two immunoglobulin heavy-chain/light-chain pairs, where the two heavy chains have different specificities (Milstein and Cuello, Nature, 305x37-(1933)). Antibody variable domains with the desired binding specificities (antibody-antigen combining sites) can be fused to immunoglobulin constant domain sequences.
The fusion preferably is with an irrtmunoglobulin heavyy-chain constant domain, including at least part of the hinge, CI-L, and CH3 regions. DNAs encoding the immunoglobulin heavy-chain fusions and, if desired, the immunoglobulin light chain, are inserted into separate expression vectors, and are co-transfected into a suitable host organism. For Further details of illustrative currently known methods for generating bispecific antibodies see, for example. Suresh et al., Methods in Enzymology, 121:210 (1986); WO
96/27011;
Brennan et al., Science 229:81 (1985); Shalaby et al., 1. Exp. Med. 175:217-2'25 (1992);
hostelry et at., J. Irnmunot. 148(5):1547-1553 (1992); Hollinger et al., Proc.
Natl. Acad.
Sci. USA 90:6444-6448 (1993); and Gruber et at., J. Imnunol. 15':5368 (199=1);
and Turi ei al., J. Immunol. 147:60 (1991).

The present antibodies can also be utilized ro directly kill or ablate cancerous cells in vivo- This involves administering the antibodies (which are optionally fused to a cytotoxic drug) to a subject requiring such treatment. Since the antibodies recognize CD200 on cancer cells, any such cells to which the antibodies bind are destroyed.
Where the antibodies are used alone to kill or ablate cancer cells, such killing or ablation can be effected by initiating endogenous host immune functions, such as complement-mediated or antibody-dependent cellular cytotoxicity. Assays for determining whether an antibody kills cells in this maimer are within the purview of those skilled in the art.
The antibodies of the present disclosure may he used to deliver a variety of cytotoxic compounds. Any cytotoxic compound can be fused to the present antibodies.
The fusion can be achieved chemically or genetically (e.g., via expression as a single, fused molecule). The cytotoxic compound can be a biological, such as a polypeptide, or a small molecule. As those skilled in the art will appreciate, for small molecules, chemical fusion is used, while for biological compounds, either chemical or genetic fusion can be employed.
Non-limiting examples of cytotoxic compounds include therapeutic drugs, a compound emitting radiation, molecules of plants, fungal, or bacterial origin, biological proteins, and mixtures thereof. The cytotoxic drugs can be intracellularly acting cytotoxic drugs, such as short-range radiation emitters, including, for example, short-range, high-energy a-emitters. Enzymatically active toxins and fragments thereof are exemplified by diphtheria toxin A frag rent, nonbinding active fragments of diphtheria toxin, exotoxin A
(from Pseudomonas aeruginosa), ricin A chain, abrin A chain, modeccin A chain, .alpha.-sacrin, certain Aleurites fordii proteins, certain Dianthin proteins, Phytolacca americana proteins (PAP, PAPII and PAP-S), Morodica charantia inhibitor, curcin, crotin, Saponaria officinalis inhibitor, gelonin, mitogillin, restrictocin, phenomycin, and enomycin, for example. Procedures for preparing enzymatically active polypeptides of the immunotoxins are described in W084/03508 and W085/03508. Certain cytotoxic moieties are derived from adriamycin, chlorambucil, daunomycin, methotrexate;
neocarzinostatin and platinum, for example.
Procedures for conjugating the antibodies with the cytotoxic agents have been previously described and are within the purview of one skilled in the an.
Alternatively, the antibody can be coupled to high energy radiation emitters;
for example a radioisotope, such as 131I a y emitter, Which, when localized at the tumor site;
2'4 results in a killing of several cell diameters. See; e.g. S. E. Order, "Analysis; Results, and Fun-Lire Prospective of the Therapeutic Use of Radiolabeled Antibody in Cancer Therapy", Monoclonal Antibodies for Cancer Detection and Therapy, R. W. Baldwin et al.
(eds.); pp 303-316 (Academic Press 1985). Other suitable radioisotopes include a--emitters, such as J
Bi, and "'At, and (3-emitters, such as'86Re and "'Y.

The route of antibody administration of the present antibodies (whether the pure antibody, a labeled antibody, an antibody fused to a toxin, etc.) is in accord with known methods, e.g., injection or iniiision by intravenous, intraperitoneal, intracerebral, intramuscular, subcutaneous, intraocular, intraarterial, intrathecal, inhalation or intralesional routes, or by sustained release systems. The antibody is preferably administered continuously by infusion or by bolus injection. One may administer the antibodies in a local or systemic manner.
The present antibodies may be prepared in a mixture with a pharmaceutically acceptable carrier. Techniques for formulation and administration of the compounds of the instant application may be found in "Remington's Pharmaceutical Sciences,"
Mack Publishing Co., Easton, PA, latest edition. This therapeutic composition can he administered intravenously or through the nose or lung, preferably as a liquid or powder aerosol (l),ophilized). The composition may also be administered parenterally or subcutaneously as desired. When administered systemically, the therapeutic composition should be sterile, pyrogen-free and in a parenterally acceptable solution having due regard for pH, isotonicity, and stability. These conditions are known to those skilled in the att.
Pharmaceutical compositions suitable for use include compositions wherein one or more of the present antibodies are contained in an amount effective to achieve their intended purpose. More specifically, a therapeutically effective amount means an amount of antibody effective to prevent, alleviate or ameliorate symptoms of disease or prolong the survival of the subject being treated. Determination of a therapeutically effective amount is well within the capability of those skilled in the art, especially in light of the detailed disclosure provided herein. Therapeutically effective dosages may he determined by using in vitro and in vivo methods.
In some embodiments, present CD200 binding antibodies provide the benefit of blocking immune suppression in CLL by targeting the leukemic cells directly through CD200. Specifically, stimulating the immune system can allow the eradication of GILL
cells from the spleen and lymph nodes. Applicants are unaware of any successful eradication of 'CU cells from these microenvironnments having been achieved with agents that simply target B cells (such as alemtnzumab). In contrast, CLL reactive 'f-' cells can have better access to these organs than antibodies. In oilier embodiments, direct cell killing is achieved by tagging the CLL cells with anti-CD200 Abs-In particularly useful embodiments, the combination oC direct cell killing and driving the immune response towards a Thl profile provides a particularly powerful approach to cancer treatment. Thus, in one embodiinent, a cancer treatment is provided wherein an antibody or antibody fi-agment, which binds to CD200 and both a) blocks the interaction between CD200 and its receptor and h) directly kills the cancer cells expressing CD200, is administered to a cancer patient- The mechanism by which the cancer cells are killed can include, but are not limited to antibody-dependent cellular cytotoxicity (ADCC) or complement dependent cytotoxicity (CDC); fusion with a toxin;
fusion with a ratliolabel; Fission with a biological agent involved in cell killing, such as granzyme B or perthrin; fusion with a cytotoxic virus; fusion with a cytokine such as TNF-cL or TFN-a. In an alternative embodiment, it cancer treatment involves adminisiering an antibody that both a) blocks the interaction between CD200 and its receptor and b) enhances cytotoxic T cell or NTH. cell activity against the tumor. Such enhancement of the cytotoxic T cell orNh cell activity may, for example, be combined by fusing the antibody with cytokines such as e-g. TL-2; IL- 12, 1L-18, II,-13, and IL--5. In addition, such enhancement may be achieved by administration an anti-CD200 antibody in combination with inhibitors such as IIViiDs, thalidomide, or thalidomide analogs.
In yet another embodiment, the cancer treatment involves administering an antibody that both a) blocks the interaction between CD-200 and its receptor and b) attracts T cells to the tumor cells. "T cell attraction can be achieved by fusing the Ab with chemokines such as MiG, II'-10, 1-TAC, CCL21, CCLJ or LIGHT. The combined action of blocking immune suppression and killing directly through antibody targeting of the tumor cells is a unique approach that provides increased efficacy.
While the above disclosure has been directed to antibodies, in some embodiments polypeptides derived front such antibodies can be utilized in accordance with the present disclosure.

Uses of the CLL Cell Line There are many advantages to the development of a CLL cell line, as it provides an important tool for the development of diagnostics and treatments for CLL, cancer, and other disease states characterized by upregulated levels of 'OX-2/CD-200, e.g., melanoma-A cell line according to this disclosure may be used for in vitro studies on the etiology, pathogenesis and biology of CLL and other disease states characterized by upregulated levels of OX-2/CD200. This assists in the identification of suitable agents that are useful in the therapy of these diseases.
The cell line may also be used to produce polypeptides and/or monoclonal antibodies for in vitro and in vivo diagnosis of CLL, cancer, and other disease states characterized by upregulated levels of OX-2/CD200 (e.g., melanoma), as referred to above, and for the screening and/or characterization of antibodies produced by other methods, such as by panning antibody libraries with primary cells and/or antigens derived from CLL patients.
The cell line may be used as such, or antigens may be derived therefrom.
Advantageously, such antigens are cell-surface antigens specific for CLL. They may be isolated directly from cell lines according to this disclosure. Alternatively, a cDNA
expression library made from a cell line described herein may be used to express CLL-specific antigens, useful for the selection and characterization of anti-CLL
antibodies and the identification of novel CLL-specific antigens.
Treatment of CLL using monoclonal antibody therapy has been proposed in the art. Recently, Hainsworth (Oncologist 5 (5) (2000) 376-384) has described the current therapies derived from monoclonal antibodies. Lymphocytic leukemia in particular is considered to be a good candidate for this therapeutic approach due to the presence of multiple lymphocyte-specific antigens on lymphocyte tumors.
Existing antibody therapies (such as Rituximab u"t, directed against the CD20-antigen, which is expressed on the surface of B-lymphocytes) have been used successfully against certain lymphocytic disease. However, a lower density CD20 antigen is expressed on the surface of B-lymphocytes in CLL (Almasri et al.. Am. J.
Henmatol., 40 (4) (1992) 259-263).
The CLL cell line described herein thus permits the development of novel anti-CI-1, antibodies and polypeptides having specificity for one or more antigenic determinants of the present CLL cell line, and their use in the therapy and diagnosis of CLI,, cancer, and other disease states characterized by upregulated levels of OX-2/CD200.
The antibody or polypeptide may bind to a receptor with which OX-2/CD200 normally interacts, thereby preventing or inhibiting OX.-2/CD200 from binding to the receptor. As yet another alternative, the antibody can bind to an antigen that modulates expression of OX-2/CD200, thereby preventing or inhibiting normal or increased expression of OX-2/CD200. Because the presence of OX-2/CD200 has been associated with reduced immune response, it would be desirable to interfere with the metabolic pathway of OX-2/CD200 so that the patient's immune system can defend against the disease state, such as cancer or CLL, more effectively.
In a particularly useful embodiment, the polypeptide binds to OX-2/CD200. In one embodiment, the polypeptide can be an antibody which binds to OX-2/CD200 and prevents or inhibits OX-2/CD200 from interacting with other molecules or receptors. As CLL cells and other cells overexpressing OX-2/CD200 greatly diminish the production of 1-5 ThI cytokines, the administration of anti-CD200 antibody or a polypeptide which binds to OX-2/CD200 to a subject having upregulated levels of OX-2/CD200 restores the Th 1 cytokine profile. Thus, these polypeptides and/or antibodies can be useful therapeutic agents in the treatment of CLL and other cancers or diseases overexpressing OX-2/CD200.
Thus, in another embodiment, the method of the present disclosure includes the steps of screening a subject for the presence OX-2/CD200 and administering a polypeptide that binds to OX-2/CD200. It should of course be understood that the presence of OX-2/CD200 can be directly monitored or, alternatively, any marker that correlates with OX-2/CD200 can be detected. In a particularly useful embodiment, a CLL patient is screened for overexpression of OX-2/CD200 and an antibody that hinds to OX-2/CD200 is administered to the patient. One such antibody is the commercially available anti-CD200 antibody from Serotec Inc. (3200 Atlantic Ave, Suite 105, Raleigh, NC 2760=1). As described in detail below, another such antibody is scFv-9 (see Fig. 9B) which binds to OX-2/CD200.
In another aspect, the present disclosure provides methods for assessing the immunomodulatory effect of molecules expressed by cancer cells. In these methods a molecule that is expressed or upregulated by a cancer cell is first Dent lied.
The molecule can be identified from a database or experimentally. Databases that identify molecules that are expressed or upregulated by cancer cells are known and include, for example. the NC160 cancer microarray project (Ross et al.. Nature Genetics 24.
227-3d_ 2000}_ the Carcinoma classification (Andrew I. Su et al.; "Molecular Classification of Human Carcinomas by Use of Gene Expression Signatures. " Cancer Research 61 :7388-7393;
2001), and the Lymphoma/Leukemia molecular profiling project (Alizadeh et al..
Nature 403: 503-11, 2000).

Experimental methods useful for identifying molecules that are expressed or upregulated by cancer cells are also known and include, for example microarray experiments, quantitative PCR, FAGS, and Northern analysis, Cancer cells, lymphocytes and the previously identified molecule that is expressed by a cancer cell are administered to a subject and the rate of growth of the cancer cells is monitored. any type of cancer cells can be employed in the present methods. In some embodiments, the cancer cells express an immunosuppressive compound. In particularly useful embodiments, the cancer cells express or even overexpress CD200.
Suitable 1 cancer cells include, but are not limited to lymphoma cell lines such as the PAR or Namalwa cell lines. The amount of cancer cells administered may range from about I x 10 to about 20 x 106.
Any type of lymphocyte may be employed in the present process. Suitable lymphocytes include, for example, PBLs, T cells, cytotoxic T cells, dendritic cells or NK
cells. In particularly useful embodiments; the lymphocytes are human lymphocytes, specifically human PBLs. The number of lymphocytes administered is predetermined to be either a) sufficient to slow the growth of the cancer cells or b) insufficient to slow the growth of cancer cells- The amount of lymphocytes administered may be greater than or equal to the number of cancer cells administered when the number of lymphocytes administered is intended to be sufficient to slow the growth of the cancer cells. In embodiments, the amount of lymphocytes administered may be from abou15x 10 to about 10 x 10 when the number of lymphocytes administered is intended to be sufficient to slow the growth of the cancer cells. The amount of lymphocytes administered may be less than the number of cancer cells administered when the number of lymphocytes administered is intended to be insufficient to slow the growth of the cancer cells. In embodiments, the amount of lymphocytes administered may range from about I x 10 to about 4 x 10 when the number of lymphocytes administered is intended to be insufficient to slow the growth of the cancer cells. The foregoing amounts are illustrative and other suitable amounts may be experimentally determined and used in the present methods.

The molecule that is expressed or upregulated by a cancer cell can be administered as the molecule or the active portion thereof itself (either isolated or recombinantly generated), or by administering cells that produce the molecule naturally, or by administering cells that have been engineered to produce the molecule or portions thereof. IViethods for engineering cells to express a desired molecule are known to those skilled in the art. The amount of molecule administered can be any amount above the amount found in a healthy individual. For example, the amount of the molecule administered can be from about 1.4-fold above what is found in a healthy individual in the same type of cell to about 10,000-fold above what is found in a healthy individual in the same type of celL It should be understood that the molecule being assessed may or may not be known to have some degree of immunomodulatory activity. Thus, the present methods may he used to confirm the immunomodulatoiy effect of a molecule as well as to determine such activity ab initio.
Any small animal may be chosen as the subject to which the cancer cells, lymphocytes and the molecule are administered. The subject may advantageously be immunocornpromised. Suitable small animals include, for example, immunodeficient Mice, irradiated rats, irradiated guinea pigs and the lihe.

The rate of cancer cell growth can be monitored using conventional techniques.
For example, tumor growth can be monitored by measuring length and width with a caliper. Tumor volume can be calculated, for example, based on multiplying the length of the tumor by the width of the tumor and then multiplying by one-half the width of It tumor. The rate of cancer cell growth can be measured periodically, such as, fir example, three Was a week. The We of growth of cancer cells when cancer cells and lymphocytes alone have been administered may be determined by administering cancer cells and lymphocytes alone to a control subject and periodically measuring trunor size.
Alternatively, cancer cells and lymphocytes alone can be administered initially, and once a baseline growth rate is established, the molecule that is expressed or upregulated by a cancer cell can be subsequently administered to he subject and the rate of growth of the cancer cells abet the second administration can be measured. Alternatively, with systemic models the rate of cancer cell growth can be monitored using F ACS, survival or other conventional techniques.
If any change in the growth rate of the cancer cells is observed compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the molecule is deemed to have an immunomodulatory of ect. If the number of lymphocytes administered is sufficient to slow the growth of the cancer cells and the rate of growth of the tumor cells observed is higher compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the molecule that is expressed or upregulated by a cancer cell is considered immunosuppressive. If the number of lymphocytes administered is insufficient to slow the growth of the cancer cells and the rate of growth of the tumor cells observed is lowered compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the molecule is considered immune enhancing. Typically, the rate of cancer cell growth observed may be about 20 to about 1,000% higher than the rate of growth when the tumor cells and donor lymphocytes are administered alone for the immunosuppressing or immune enhancing effect to be statistically significant.
Once the immunomodulatory effect of the molecule is established, compounds that either enhance or inhibit the activity of the molecule can be identified in accordance with embodiments described herein. The compound that either enhances or inhibits the activity of the molecule previously found to have immunomodulatory effect can be any compound that alters the protein/protein interaction that provides the immunomodulatory effect. The enhancing or inhibiting effect can be the result of direct interaction with the compound expressed or upregulated by the cancer cell or may be the result of an interaction with other compounds in the metabolic pathway of the compound expressed or upregulated by the cancer cell.
For example, antibodies or functional antibody fragments can be identified that interact with the molecule previously found to have an innnunomodulatory effect, the receptor with which the molecule interacts, or some other molecule in the n-metabolic pathway of the molecule responsible for the immunomodulatory effect.
Techniques for making antibodies (including antibody libraries) and screening them for an inhibitory or enhancing effect will be apparent to those skilled in the art. As another example, small molecules may be screened for an inhibitory or enhancing effect. Techniques for screening small molecule libraries for an inhibitory or enhancing effect will be apparent to those skilled in the art.
In another aspect, methods for assessing the immunomodulatory effect of a compound are contemplated by the present disclosure. Demonstrating fill munomodulatory properties of compounds or molecules acting on the human inunune system is very challenging in small animal models. Often, the compounds do not act on the immune system of small animals, requiring reconstitution of the human immune system in mice. Reconstitution can be accomplished by grafting of various fetal immune organs into mice or by injection of human lymphocytes, but none of the models described to date has proven to be useful for demonstrating inrrmanornodulatory properties of compounds or molecules. The immune system is believed to play an important role in eradicating cancer cells. Cancer cells have found ways to evade the immune system by upregulation of immunosuppressive receptors.
The present methods of assessing the immunomodulatory effect of a compound is accomplished by a model that mimics the graft versus leukemia effect observed in patients with leukemia that are infused with donor lymphocytes (e.g., PBLs) resulting in remission in up to 80% of patients. The method involves administering cancer cells, lymphocytes and the compound to be assessed to a subject and the rate of growth of the cancer cells is monitored-Any type of cancer cells can be employed in the present methods. In some embodiments, the cancer cells express an immunosuppressive compound. In particularly 1.5 useful embodiments, the cancer cells express or even overexpress CD200.
Suitable cancer cells include, but are not limited to lymphoma cell lines such as the RAJI or Namalwa cell lines. The amount of cancer cells administered may range from about 2 x10 to about 20 x 10 .
Any type of lymphocyte may be employed in the present process. Suitable lymphocytes include, for example, PBLs, dendritic cells, T cells, cytotoxic T
cells, NK
cells. In particularly useful embodiments, the lymphocytes are human lymphocytes, specifically human PBLs. The number of lymphocytes administered is predetermined to be either a) sufficient to slow the growth of the cancer cells orb) insufficient to slow the growth of cancer cells. The amount of lymphocytes administered may be greater than or equal to the number of cancer cells administered when the number of lymphocytes administered is intended to be sufficient to slow flit growth of the cancer cells. In embodiments, the amount of lymphocytes administered may be from about 5 x 106 to about 10 x 106 when the number of lymphocytes administered is intended to be sufficient to slow the growth of the cancer cells. The amount of lymphocytes administered may be less than the number of cancer cells administered when the number of lymphocytes administered is intended to be insufficient to slow the growth of the cancer cells. In embodiments, the amount of lymphocytes administered may range h-on-r about I x 10 to about 4 x 10 when the number of lymphocytes administered is intended to be insufficient to slow the growth of the cancer cells. The foregoing amounts are illustrative and other suitable amounts may be experimentally determined and used in the present methods.
The compound being assessed can be any compound whose immunomodulatory effect is sought to be determined. Illustrative examples of compounds include the antibodies and peptides described herein above. The amount of the compound being assessed administered may range from about I nig/kg to about 200 mg/kg. The compound to be assessed can be administered as the compound itself, or by administering cells that produce the compound naturally, or by administering cells Thal have been engineered to produce the compound. It should be understood that the compound being assessed may or may not be knoyvn to have some degree of immunomodulatory activity.
Thus, the present methods may be used to confirm the immunomodulatory effect Of &
compound as well as to determine such activity ab initio.
Any small animal may be chosen as the subject to which the cancer cells, lymphocytes and the compound are administered. The subject may advantageously be unmunocompromised. Suitable small animals include, for example, immunodeficient mice, irradiated rats, irradiated guinea pigs and the like.
The rate of cancer cell growth can be monitored using conventional techniques.
For example, tumor growth can be monitored by measuring length and width with a caliper. `humor volume can be calculated, for example, based on multiplying the length of the tumor by the width of the tumor and then multiplying by one-half the width of the tumor. The rate of cancer cell growth can be measured periodically, such as, for example, We times a week. The We of growth of cancer cells when cancer culls and lymphocytes alone have been administered may be determined by administering cancer cells and lymphocytes alone to a control subject and periodically measuring tumor size.
Alternatively, cancer cells and lymphocytes alone can be administered initially, and once a baseline growth rate is established, the compound to be assessed can be subsequently administered to the subject and the rate of growth of the cancer cells afier the second administration can be measured.
When injecting cancer cells such as the lymphoma cell lines RAiI or Hainaiwa into imnnine-deficient mice, administration of five to 10 million PBLs results in significantly slower tumor growth. In contrast, low PBL numbers (1-2 million depending on donor) do not slow tumor growth.
If any change in the growth rate of the cancer cells is observed compared to the rate of growth when the tumor cells and donor lymphocytes are administered alonc, the molecule is deemed to have an immunomodulatory effect. If the number of']
ymphocytes administered is sufficient to slow the growth of the cancer cells and the rate of arowth of the tumor cells observed is higher compared to the rate of growth when the rumor cells and donor lymphocytes are administered alone, the compound is considered immunosuppressive. If the number of lymphocytes administered is insufficient to slow the growth of the cancer cells and the rate of growth of the tumor cells observed is lowered compared to the rate of growth when the tumor cells and donor lymphocytes are administered alone, the compound is considered immune enhancing.
In order that those skilled in the art may be better able to practice the compositions and methods described herein, the following examples are given for illustration purposes.

E1 A1v1PLE 1 Isolation of Cell Line CLL-AAT
Establishment of the cell line Peripheral blood from a patient diagnosed with CLL was obtained. The VJBC
count was ] .6xl0S/ml. Mononuclear cells were isolated by HistopaqueTM-1077 density gradient centrifugation (Sigma Diagnostics, St. Louis, Iv[O). Cells were washed -PV"
Ice with Iscove's Modified Dulbecco's Medium (IlvIDlvI) supplemented with 10% heat-inactivated fetal bovine serum (FBS), and resuspended in 5 ml of ice-cold 1MDM./I0%
FBS. Viable cells were counted by staining with trypan blue. Cells were mixed with an equal volume of 85 %o FBS/I5% DMSO and frozen in 1 ml aliquots for storage in liquid nitrogen.
lmmunophenotyping showed that >90% of the CD45T lymphocyte population expressed IgD, kappa light chain, CDS, CD19, and CD23. This population also expressed low levels of ],-M and CD20. Approximately 50% of the cells expressed high levels of CD38. The cells were negative for lambda light chain, CD10 and CD138 An aliquot of the cells was thawed, washed, and resuspended at a density of 107/mL in 1lv )M supplemented with 20% heat-inactivated FBS, 2m1vt L-glutalnine, 100 units/ml penicillin, 100 gg/ml streptomycin, 50 1v1 2-mercaptoethanol, and 5 ng/ml recombinant human IL-4 (R Sr D Systems, Minneapolis, MN). The cells were cultured at 37 C in a humidified s% CO_ atmosphere. The medium was partially replaced every 4 days until steady growth was observed. After 5 weeks, the number of cells in the culture began to double approximately every 4 days. This cell line was designated CLL-_f-'~_AT.

3=1 Characterization of the cell line lmmunophenotyping of the cell line by flow cytomeny showed high expression of IgM, kappa light chain.. CD23, CD38, and CD138, moderate expression of CDlp and CD20, and weak expression of IgD and CD5. The cell line w,vas negative for lambda light chain, CD4, CD8, and CD10.
i Immunophenotyping of the cell line was also done by whole cell ELISA using a panel of rabbit scFv antibodies that had been selected for specific binding to primary B-CLI, cells. AlI of these CLL-specific scFv antibodies also recognized the CLL -AAT
cell line. In contrast, the majority of the scFvs did not bind to two cell lines derived from B cell lymphomas: Ramos, a Burkitt's lymphoma cell line, and RL, a non-Hodgkin's lymphoma cell line.

Selection of scFv Antibodies for B-CLL-specific Cell Surface Antigens using Antibody Phase Display and Cell Surface Panning Immunizations and scFv antibody library construction Peripheral blood mononuclear cells (PBMC) were isolated from blood drawn from CLL patients at the Scripps Clinic (La Jolla, CA). Two rabbits were immunized with 2x107 PBh4C pooled from 10 different donors with CLL. Three immunizations, two subcutaneous injections followed by one intravenous injection, were done at three week intervals. Serum titers were checked by measuring binding of serum IgG to primary CLL
cells using flow cytometry. Five days after the final immunization, spleen, bone marrow, and PBMC were harvested from the animals. Total RNA was isolated from these tissues using Tri-ReagentT"' (Molecular Research Center, hnc). Single-chain Fv (scEv) antibody phase display libraries were constructed as previously described (Barbas at aL, (2001) Phage Display: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring harbor, New York). For cell surface panning, phagemid particles from the reamplitied library were precipitated with polyethylene glycol (PEG'), resuspended in phosphate-2 buffered saline (PBS) containing 1M bovine serum albumin (BSA), and dialysed overnight against PBS.

Antibody selection by cell surface panning The libraries were enriched for CLL cell surface-specific antibodies by positive-negative selection with a magnetically-activated cell sorter (MACS) as described by Siegel at al. (1997, J. Immunol. Methods 2206:73-85). Briefly, phagerriid particles from the scFv antibody library were preincubated in 1v1PBS (2% nonfat dry milk, 0.02%
sodium azide in PBS, p1-1 7.4) for 1 hour at 25 C to block nonspecific binding sues.
Approximately 10 primary CLL cells were labeled with mouse anti-CD5 IgG and mouse anti-CD-19 lgG conjugated to paramagnetic microbeads (Miltenyi Biotec, Sunnyvale, CA). Unbound microbeads were removed by washing. The labeled CLL cells ("target cells") were mixed with an excess of "antigen-negative absorber cells", pelleted, and resuspended in 50 l (1010-1011 cfu) of phage particles. The absorber cells serve to soak tip phase that stick non-specifically to cell surfaces as well as phage specific for "common" antigens present on both the target and absorber cells. The absorber cells used were either TF-1 cells (a human erythroleukemia cell line) or normal human B
cells isolated from peripheral blood by imrnLill ornagnetic negative selection (SteinSepTM1 system, StemCell Technologies, Vancouver, Canada), The ratio of absorber cells to target cells was approximately 10 fold by volume. After a 30 minute incubation at 25 C, the cell%phage mixture was transferred to a MiniM.ACST"l MS+ separation column.
The column was washed twice with 0.5 tnl of MPBS, and once with 0.5 ml of PBS to remove the unbound phage and absorber cells. The target cells were eluted from the column in I rnl of PBS and pelleted in a microcentrifuge at maximum speed for 15 seconds. The captured phage particles were eluted by resuspending the target cells in 201) l of acid elution buffer (0.] N HCI, pH adjusted to 2.2 with glycine, plus I mg/ml BSA).
After a 10 minute incubation at 25 C, the buffer was neutralized with 12 L of 2M Tris base, pH10.5, and the eluted phage were amplified in E. colt for the next round of panning. For each round of panning, the input and output phage titers were determined. The input titer is the number of reamplified phage particles added to the target cell/absorber cell mixture and the output titer- is the number of captured phage eluted from the target cells. An enrichment factor (F) is calculated using the formula E=(R, outputfRr, input)/(Itj output/Ri input), where R1 = round 1 and R,= round 2, 3, or 4. In most cases, an enrichment factor of 10'-] 03 fold should be attained by the third or fourth round.

Analysis of enriched antibody pools following panning After 3-5 rounds of panning, the pools of captured phage were assayed for binding to CLL cells by flow cytometry and/or whole cell ELISA:
1. To produce an entire pool in the form of 1--LA -tagged soluble antibodies, 2ml of '& non-suppressor strain of E. cols (e.g. TOPI OF') was infected with 1 1 (109-101D
chi) of phagemid particles. The original, unparined library was used as a negative control.
Carbenicillin was added to a final concentration of 10 M and the culture was incubated at 37 C with shaking at 250rpm for 1 hour. Eight ml of SB n-iedium containing 50 g/ml carbenicillin was added and the culture was grown to an OD

of -0.8. IPTG was added to a final concentration of 1mM to induce scFv expression from the Lac promoter and shaking at 37 C was continued for 4 hours. The culture was centrifuged at 3000Kg for 15'. The supernatant containing the soluble antibodies was filtered and stored in 1 ml aliquots at-20 C.
2. Binding of the scFv antibody pools to target cells vs. absorber cells was determined by flow cytometry using high-affinity Rat Anti-HA (clone 3F10, Roche iviolecular Biochemicals) as secondary antibody and PE-conjugated Donkey Anti-Rat as tertiary antibody.
3. Binding of the antibody pools to target cells vs. absorber cells was also determined by whole-cell .ELISA as described below.

Screening individual scFv clones following panning To screen individual scFv clones following panning, TOPIDF' cells were infected with phage pools as described above, spread onto LB plates containing carbenicillin and tetracycline, and incubated overnight at 37 C. Individual colorises were inoculated into deep 96-well plates containing 0.6-1.0 ml of SB-carbenicillin medium per well.
The cultures were grown for 6-8 hours in a HiGroTM shaking incubator (GeneMachines. San Carlos, CA) at 520 rpm and 37 C. At this point, a 90 l aliquot flow each well was transferred to a deep 96-well plate containing 10 pL of Dh1SO. This replica plate was stored at -80 C. II'TG was added to the original plate to a final concentration of I mlv1 and shaking was continued for 3 hours. The plates were centrifuged at 3000xg for 15 minutes. The supernatants containing soluble schv antibodies were transferred to another deep 96-well plate and stored at -20 C.

A sensitive tiwhole cell ELISA method for screening HA-tagged scFv antibodies was developed:
I An ELISA plate is coated with concanavalin A (1 Ong/nil in 0.1 ivt I~lalICO3, PIf9.6, 01 mlvi CaCI).
3 >. After washing the plate With PISS, 0.5-1110' target cells or absorber cells in 501x1 of PBS are added to each well, and the plate is centrifuged at 250xg for 10 minutes.
3. 50 l of 0.02% glutaraldehyde in PBS are added and the cells are fixed overnight at =1 C-4- After washing with PBS, non-specific binding sites are blocked with PBS
containing '1~ o non-fat dry milk for 3 hours at room temperature.

5. The cells are incubated with 501d of soluble, HA-tagged scbv or flab antibody (TOPI OF' supernatant) for 2 hours at room temperature, then washed six tunes with PBS.

6- Bound antibodies are detected using a Mouse Anti-11A secondary antibody (clone 12CA5) and an alkaline phosphatase (AI')-conjugated Anti-Mouse IgG tertiary antibody. An about 10-fold amplification of the signal is obtained by using AMDEX
AP--conjugated Sheep Anti-N.touse IgG as the tertiary antibody (Anrersham Pharmacia Biotech). The AIVIDEX antibody is conjugated to multiple _P molecules via a de~tran backbone. Color is developed with the alkaline phosphatase substrate PHPP
and measured at ,405nm using a microplate reader.
Primary screening of the sclkv clones was done by ELISA on primary CLL culls versus normal human PBIC. Clones which Were positive on CLL cells and negative on normal PBMC were rescreened by ELISA on normal human B cells, human B cell lines, 'VF- 4 cells, and the CLL-AAT cell line. The clones were also rescreened by ELISA on CLL cells isolated from three different patients to eliminate clones that recognized patient-specific or blood type-specific antigens. Results from representative ELISAs are shown in Figures 2-6 and summarized in Figs. 9A- 9C.
The number of unique scFv antibody clones obtained was determined by DNA
fingerprinting and sequencing. The scFv DNA inserts Were amplified h-om the plasmids by PCR and digested with the restriction enzyme BstNl. The resulting t3agrnents were separated on a d% agarose gel and stained with ethidium bromide. Clones with different restriction fi-agnment patterns must have different amino acid sequences.
Clones with identical patterns probably have similar or identical sequences. Clones with unique BstN I
Fingerprints were further analyzed by DNA sequencing. Twenty-five different sequences were found, which could be clustered into 16 groups of antibodies with closely related complementarity determining regions (Figs. 9A- 9C).

Characterization of scFv antibodies by flow cytometry The binding specificities of several scFv antibodies were analyzed by 3-color flow cytometry (Fig. 7). PBMC isolated from normal donors Were stained with FITC-conjugated anti-CDC and PerCP-conjugated anti-CD 19. Staining with scFv antibody was done using biotin-conjugated anti-HA as secondary antibody and PF-conjugated streptavidin. Three antibodies, scFv-2, scFv-3, and scFv-6, were found to specifically recognize the CD19-' .B lymphocyte population (data not shown). 7 lie fourth antibody, scFv-9, recognized two distinct cell populations: the CD19' B lymphocytes and a subset of CDS" T lymphocytes (Fig. 7). Further characterization of the "1 cell subset showed that it was a subpopulation of the CD4' CDW III cells (data not shown).
To determine if the antigens recognized by the scFv antibodies were overexpresseia on primary CLL cells, PBIVIC from five CLL patients and five normal donors were stained with scFv and compared by flow cytonletry (Fig. 8 and Table 2). By comparing the mean fluorescent intensities of the positive cell populations, the relative expression level of an antigen on CLL cells vs. normal cells could be deterrnineil. One antibody, scFv-2, consistently stained CLL cells less intensely than normal PBAv1C, whereas scFv-3 and scFv-6 both consistently stained ('.LL cells more brightly than normal P13h-IC. The fourth antibody, scFv-9, stained two of the five CLL samples much more intensely than normal PUNIC, but gave only moderately brighter staining tier the ether three CLL samples 04g. 8 and Table 2). This indicates that the antigens for scl, V-3 anti scbv-6 are overexpressed approximately Ld fold on most if not all CLL tumors, whereas scFv-9 is overex_pressed 3 to 6-fold on a subset of CLL tumors.
CLL patients can be divided into two roughly equal groups: those with a poor prognosis (median survival lime of 8 years) and those with a favorable prognosis (median survival time of 26 years). Several unfavorable prognostic indicators have been idenlitled for CLL, most notably the presence W N genes lacking somatic mutations and the 3t) presence of a high percentage of CD38' B cells. Since scFv-9 recognizes an antigen overexpressed in only a subset of CLL patients, it was sought to determine if scFv-9 antigen overexpression correlated with the percentage of CD38" cells in blood samples from ten CLL patients (Fig. U). The results indicate that scFv-9 antigen overexpressiou and percent CD38+ cells are completely independent of one another.

identification of antigens recognized by scFv antibodies by immunoprecipitation (IP) and mass spectrometry (MS) To identify the antigens for these antibodies, scFvs were used to immunoprecipitate the antigens from lysates prepared from the microsomal fraction of cell-surface biotinylated CLL-A-AT cells (Fig. 12). The immunoprecipitated antigens were purified by SDS-PACE and identified by matrix assisted laser desorption ionization mass spectrometry (IvLa1DI-MS) or microcapillary reverse.-phase H PLC nano-electrospray tandem mass spectrometry (gLC/MS/MS) (data not shown). ScFv-2 immunoprecipitated a 110 kd antigen from both Rh and CLL-AAT cells (Fig. 12).
This antigen was identified by MAI_.DI-MS as the B cell-specific marker CD 19. ScFv-3 and scFv-6 both immunoprecipitated a 45 kd antigen from CLL-AAT cells (not shown).
This antigen was identified by NLALDI-MS as CD23, which is a known marker for CLL
and activated B cells- ScFv-9 immunoprecipitated a 50 kd antigen from CLL-AAT
cells (Fig.
12). This antigen was identified by gLC/NIS/MS as OX-2/CD200, a known marker for B
cells, activated CD4+ T cells, and thymocytes. OX-2/CD200 is also expressed on some non-lymphoid cells such as neurons and endothelial cells.

The capability of cells overexpressing OX-=2/CD200 to shift the cytokine response from a Thu response (IL-2, IFN-y) to a Th2 response (IL-4, IL-10) was assessed in a mixed lymphocyte reaction using monocyte-derived tnacrophages/dendritic cells from one donor and blood-derived T cells from a different donor. As a source of OX-2/CD200-expressing cells, either OX-2/CD200 transfected EBNA cells as described below or CLL patient samples were used.

Transfection of 293-EBNA cells 293-EBNA cells (Invitrogen) were seeded at 2.5x10 per 100n1m dish. 24 hours later the cells were transiently transfected using Polyfect reagent (QLSGEN) according to the manufacturer's instructions. Cells were cotransfected with 7? g of OX-cDNA in vector pCEP4 (Invitrogen) and 0.8 g of pAdVAntageTM vector (Promega).
As a negative control, cells were cotransfected with empty pCEP4 vector plus pAdVAntage.
48 hours after transfection, approximately 901//0 of the; cells expressed OX-2/CD200 on their surface as determined by flow cytometiy with the scFv-9 antibody.

Ivlawration of dendritic cells/macrophages from blood monocytes Buffy coats were obtained from the San Diego Blood Bank and primary blood lymphocytes (PBL) were isolated using FicollT,,1. Cells were adhered for 1 hour in Eagles Minimal Essential Medium (EMEM) containing 2% human serum followed by vigorous washing with PBS. Cells were cultured for 5 days either in the presence of GIvI-CSF, IL-4 and IFN-)/ or M-CSF with or without the addition of lipopolysaccharide (LPS) after 3 days- Matured cells were harvested and irradiated at 2000 R AD using a 'y-11radiator (Shepherd Mark I Model 30 irradiator- (Cs137)).

Mixed lymphocyte reaction Mixed lymphocyte reactions were set up in 24 well plates using 500,000 dendritic cells/macrophages and 1x 10 responder cells. Responder cells were T cell enriched lymphocytes purified from peripheral blood using Ficoll. T cells were enriched by incubating the cells for 1 hour in tissue culture flasks and taking the non-adherent cell fraction. 500,000 OX-2/CD200 transfected EBNA cells or CLL cells were added to the macrophages/dendritic cells in the presence or absence of 30 gghnl anti-CD200 antibody (scFv-9 converted to full IgG') 2-4 hours before the lymphocyte addition.
Supernatants were collected after 48 and 68 hours and analyzed for the presence of cytokines.

Conversion of scFv-9 to full Igo Light chain and heavy chain V genes of scPv-9 were amplified by overlap PCR
with primers that connect the variable region of each gene with human lambda light chain constant region gene, and human IgGI heavy chain constant region CHI gene, respectively. Variable regions of light chain gene and heavy chain gene of scFv-9 were amplified with specific primers and the human lambda light chain constant region gene and the IgGl heavy chain constant region CH1 gene were separately amplified with specific primers as follows:

R9VL-Fl QP: 5' GGC CTC TAG ACA GCC TGT GCT GAC TCA GTC
GCC CTC 3' (SEQ ID 110: 103);
R9VL/hCL2-rev: 5' CGA GGG GGC .4GC CTT GGG CPG ACC TOT
GAC GGT CAG CTG GGT C 3' (SEQ ID NO: 104);

R9VL/3 CL F: 5' GAC CCA OCT GAC CGfh CAC AGG TCA GCC
C:k\ GGC TOC CCC CTC G 3' (SEQ IP NO: 105):
R9VH-F1: 5' TCT AAT CTC GAG CAG CAG CAG CTG A TG GAG
TCC G 3' (SEQ M NO: 106);
R9VFUhCG-rev: 5' GAC CGA TGG GCC C17 GGT GGA GGC TG A
GGA GAC GGT GAC CAG GGT GC 3' (SEQ ID 1v10: 107);

R9VTI/hCG-F: 5' GCA CCC TGG TCAA1 CC(] TCT CCT CAG OCT CCA
CCA AGG GCC CAT C:GG TC 3' (SEQ ID N0: 108);
hCL2-rev : 5' CCA CTG TCA GAG CTC CCG GGT AGA AGT C 3' (SEQ M- N0: 109);
hCG-rev : 5' GTC ACC GGT TCG GGG AAG TAG TC 3' (SEQ 1D NO: 110).
Amplified products were purified and overlap PCR was performed.
Final products were digested with Xba I/Sac I (light chain) ando I1Pin As (heavy chain) and cloned into a human Fab expression vector, PAX243hGL (see published International Application WO 2004/078937). DNA clones were analyzed for PCR errors by DNA sequencing. The hCMV IE promoter gene was inserted at Not 1/Xho I
site (in front of the heavy chain). The vector was digested with Xha I/Pin AI/EcoR I/Nhe 1 and a 3472 bp fragment containing the light chain plus the hCMV IE promoter and the heavy chain gene was transferred to an lgGl expression vector at the Xba 1/Pin Al site.
Cytokiue analysis The effect of the scFv-9 converted to full IgG on the cytokine profile in the mixed lymphocyte reaction was determined.
Cy-tokines such as IL-2, IFN-y, IL-4, IL-10 and IL-6 found in the tissue culture supernatant were quantified using ELISA. Ivlatched capture and detection antibody pairs for each cytokine were obtained from R+D Systems (IVlinneapolis, DEN/f and a standard q"_' curve for each cytohine was produced using recombinant human eytolcine. Anti c1 tokine capture antibody was coated on the plate in PBS at the optunurn concentration.
After overnight incubation, the plates were washed and blocked for 1 hour with PBS
containing 1 % BSA and 5% sucrose. After 3 washes with PBS containing 0.03% TweenT"^
supernatants were added at dilutions of two-fold or ten-fold in PBS containing 1 ,t BSA..
Captured c}tokines were detected with the appropriate biotinylated anti-cytokirne antibody followed by the addition of alkaline phosphatase conjugated streptavidin and SigrnaS substrate. Color development was assessed with an ELISA plate reader (Molecular Devices).
As shown in Figure 14, the presence of OX-2/CD200 transfected but not untransfected cells resulted in down-regulation ofTh] cytokines such as IL-2 and IFII-1'.
Addition of the anti-CD200 antibody at 30 g/ml fully restored the Th 1 response, indicating that the antibody blocked interaction of OX -2/CD200 with its receptor.
As set forth in Figures 15 and 16, the presence of CLL cells in a mixed lymphocyte reaction resulted in down-regulation of the Thl response. (Figure 15 shows the results for IL-2; Figure 16 shows the results for IFN-y).. This was not only the case for cells over-expressing 0X-2/CD200 (IB, EM, HS, BH), but also for CLL cells that did not overexpress OX-2/CD200 (JR, JG and GB) (the expression levels for these cells are set forth in Figure 11). However, the anti-CM00 antibody only restored the Till response in cells over-expressing OX-2/CD200, indicating that for patients overexpressing OX-2/CD200, abrogating OX-2/CD200 interaction with its receptor on macrophages was sufficient to restore a Thl response. In patients that did not overexpress OX-2/CD200, other mechanisms appeared to he involved in down-regulating the Thl response.

Animal Models To Test An Effect Of Anti-CD200 On Tumor Rejection A model was established in which RAIL lymphoma tumor growth is prevented by the simultaneous injection of PBLs. NOD/SCID mice were injected subcutaneously with 4x] 0 R4JI cells either in the presence or absence of human PBLs from different donors vas at lx10 , 5x10 or 10x10 cells. Tumor length and width as well as body weight \-determined 3 times a week. Mean +/-SD of armor volumes for all groups is shown in Figures 17 A and B. Statistical analysis was performed using ? parametric tests (Student's t-test and Welch's test) and one non-parametric test (the Wilcox test). Results of the statistical analysis are found in Figure 15. RAJ] cells form subcutaneous tumors 'I?

with acceptable variation. Rejection is dependent on the specific donor and the PBL cell number. 1x10 PBLs were insufficient to prevent rumor growth. Donor 2 at 510 PBLs from day _2? 43 and donor 3 at 5x10 or 1x10 PBLs starting at clay 36 significantly reduced r ur,or growth. Donor 4 is very close to being significant after day 48.
To test for an effect of anti-CD200, RA.IT cells are stably transfect.ed with CD200.
Animals are injected as described in the previous paragraph. In the presence o_f'CD 200-transfected cells, tumors grow even in the presence of human PBLs. Anti-CD200 antibody is administered to evaluate tumor rejection in this model.
Also, a liquid tumor model is established. RAJI cells are injected intraperitoneally into NOD/SCID mice. Cells disseminate to bone marrow, spleen, lymph node and other organs resulting in paralysis. Concurrent injection of human PBLs prevents or slows tumor growth. Tumor growth is monitored by assessing the mice for signs of movement impairment and paralysis. Once these signs are observed, mice are sacrificed and the number of tumor cells is assessed in various organs including bone marrow, spleen, lymph nodes and blood by FACS analysis and PCR.
Similar to the subcutaneous model, CD200 transfected cells are injected intraperitoneally. They grow even in the presence of luiman PBLs. Treatment with anti-CD200 results in tumor rejection or slower tumor growth.

E`1AIvB'LE 4 Library construction A mouse was immunized alternately with baculovirus expressed recombinant CD200 extracellular domain fused to mouse IoG Fe (CD200-Fc) (Orbigen Inc., San Diego, CA) and 293-EBNA cells transiently trarrsfected with a vector containing full length CD200. Total RNA was prepared from mouse spleen using TRI reagent (Molecular Research Center, Inc., Cincinnati, 01) according to the manufacturer's protocol. Messenger RNA (mRNA) was purified using OligotexTu (QIAGEN Inc..
Valencia, CA) according to the manufacturer's manual. First strand cDISIA was synthesized using SuperScript 11 RTaseTM (Invitrogen Life Technologies, Carlsbad, CA) according to the manufacturer's protocol. First strand cDNA was digested with restriction endonuclease and second strand cDNA was synthesized according to the method fully described in published PCT application W003/025202A2, published March 2?, 2003 Second strand cDNA was cleaned up with PCR purification kit (QIAGEN) and single primer amplification was performed according to the method described in published PCT

application W003/025202A2, published March 27, 2003. Amplified products were pooled and purified with PCR purification kit. Kappa light chain was digested with 5ha I
and BspE 1, and IgGI and IgG2a heavy chains were digested with Abe I and Bin I.
Digested fragments were purified from the agarose gel using Gel extraction kit (QIAGEN) and cloned into PAX313m/h(3 vector as described in published PCT
application WO/04078937A2 published September 16, 2004.

Library . afninQ
The libraries (IgGI kappa and IgG2a kappa) were panned on CD200.-Fc either directly coated on the microliter wells (Costar Group, Bethesda, PAD) or captured with goat anti-mouse IgG Fc specific antibody (Sigma Aldrich Corp., St. Louis, MO).
For the preparation of library phage, electrocompetent l-Blue cells (Strataeene, La Jolla, CA) were eleetroporated with library DNA and grown in SOC medium for 1 hour and in SB
medium for 2 hours with carbenicillin. Phage production was induced with the addition of VCS M13 helper phage (Ainersham Biosciences Corp., Piscataway, NJ) and I mM
fPTG
at 30 C overnight. The culture was spun down and phage were precipitated with 4%
polyethylene glycol and 3% NaCl. The phage were spun down and resuspended in 1 i%
BSA/PBS containing unrelated antigen, FLJ32028 that is also baculovirus expressed extracellular domain fused to mouse IgG Fc (FLJ32028-Fe) (Orbigen, San Diego), as a soluble competitor. For the panning on directly coated CD200-Fc, four wells were coated with 100 l of CD200-Fc (5 l.tg/ml in 0.1 lvi NaHCO3 pH8.6) at 4'C overnight.
The wells were washed 5 times with phosphate buffered saline (PBS) pII7.0 and blocked with 1%
bovine serum albumin (BSA)IPBS at 37 C for I hr. For the panning on CD200-Fc captured on goat anti-mouse IgG Fe, four microtiter Wells were coated with 100 41 goat anti-mouse IgG Fc (20 ttg/ml in PBS) at 4 C overnight. The wells were washed 5 times with PBS and incubated with 100 pd CD200-Fc (20 g/ml in PBS) for 1 hour at 37 C.
The wells were washed 5 times with PBS and blocked with 1% B SA/PBS at 37 C
for I
hour. For both directly-coated and captured methods of panning, the blocker was replaced with the mixture of soluble Fabs obtained from the panning of another library (the library described in Example 3 of PCT Publication WO 2004/1 10369, on FL132028 to mast:
epitopes on Mouse IgG Fe and the wells were incubated for 30 ruin at 37 C.
These masking Fabs were shown to also bind to CD200-Fc_ Library phage were added on top of the masking Nabs and the wells were incubated for approximately 1.5 hours at 37'C. The unbound phage were washed with PBS with increasing stringency (3 times in the first round, 5 times in the 2"i round and 10 times in the 3'L1 and the 4`i' rounds) with 5 minute incubation and pipetting up and down 5 times for 3 each wash. The bound phage were eluted twice with 100 pJ 0.1 Ivl BC1 with I
mg/1-111 BSA, p1-12.2 and neutralized with 2 M Tris Base p1-l 11.5. The freshly grown cells were infected with eluted phage and tiitratcd onto LB agarose plates containing carbenicillin and glucose. The remaining phage were propagated overnight at 30-with the addition of VCS M 13 Helper phage and 1 mM IPTG for the next round of panning.

hih a y screenii Ninety five colonies from round 3 and 4 titration plates were grown in I nil containing 1.2.5 lig/ml tetracycline and 50 ftg/m1 carhenicillin for approximately 6 hours at 37 C. VCS M13 helper phage were added and the culture was incubated for 2 hours at 37'C. I mM IPTO and 70 pg/ml kananiycin were added and Fab-phage production was induced at 30 C overnight. Microtiter wells were coated with 50 ltl of rabbit anti-mouse IgG F(ab')c (4 fig/ml in PBS), CD200-Fc (4 fcg/ml in 0.1 M NalIC03 p1-I8.6), or FLJ32028-Fc (I fig/ml in 0.1 hvl NafI('O3 1*18.6) at 4'C overnight. The wells were washed 3 times with PBS and blocked with 100 p1 11,'o NSA/PBS for 1 hour at 37'C. The culture was spun down. The blocker was replaced with the culture supernatant containing Pab-phage and the wells were incubated for 1.5-2 hours at 37'C. The remaining Fab-phage was stored at -80 C for flow cytometry. The plates were washed 3 times with PBS
and the binding Was detected with 50 ltl alkaline phosphalase (AP)-conjugated goat anti-mouse 1gG F(ab')2 antibody (Pierce)(l:500 in 1 'b BSA/PBS) for 1 hr at 37 C.
The plates were washed 3 times with PBS and developed with AP Substrate (Sigma-Aldrich) in pNPP buffer. Almost all of the clones from round 3 were already specifically positive to CD200 (Figs. I 9A-D). Clones were also screened by high throughput flow cylomeiry analysis. One hundred tuicrolilers of 293 cells transiently transfected with CD200 (I x 10' cells) were ali"quoted into a 96 well plate (Costar). Fifty microliters Fah-phage was added to the cells and mixed by pipettiug and incubated on ice for 30 minutes. The cells vverc washed twice with 1 `;4, BSA/PBS containing 0.01 ,,o NaN3. The cells were resuspended in 100 l PE-conjugated groat anti-mouse 1gG antibody (Sigma-Aldrich) in I%
BSA/PBS

=16 containing 0.01 % NaN3 and incubated on ice for 30 minutes. The cells were washed twice with 1% BSA/PBS containing 0.01% NaN3 and resuspended in 200 l I/'(, paraformaldehyde in PBS. Representative clones showing positive binding To CD-expressing cells are shown in Figs. 20A-D.
DNA sequences were analyzed and deduced amino acid sequences of the hea 't chain were grouped according to the complementarity determining region 3 (CDR3) (rigs. 21A, B). They were divided into 17 groups.

Fluorescent bead assay 23 clones were selected for further analysis. They were cG2aR3B5, dGIR3A5, cG2aR3A2, dG2aR3B2, dG1R3AI, cG2aR3A1, cG2aR3B, dG1R3B, cGIR3A, cG1R3A, cG1R3A1,dG1R3B,dG1R3B,cGIR3C,dG2aR3C,dG2aR3AI,cO2aR3B,cG2aR313, dG1R3B, cG2aR3B, cG2aR3C, dG1R3H, and dG2aR3.A6. DNA of selected Fabs was digested with Spe IfNhe I for gene 11I removal and soluble Fab expression and purification. The purified Fabs were evaluated for their ability to block the interaction of CD200 with its receptor (CD200R) in a fluorescent bead assay.
TransFluoSpheresT"' carboxylate-modified microspheres (488/645) (Molecular Probes invitrogen Detection Technologies, Eugene, OR) were coated with streptavidin followed by a biotin-labeled anti-human Fc antibody and baculovirus-produced CD200-Fc protein. 293 cells were transiently transfected with CD200R. Cell surface expression was confirmed by FACS
analysis- 1 million CD200-coated beads were pre-incubated with various amounts of anti-CD200 Fabs or chimeric 1gO for 10 minutes before the addition of 50,000 CD200R
transfected cells. After a 30 minute incubation at 37'C, the cells were washed in Tris buffer containing 1% BSA and analyzed using a FACS CaliburTM. Fabs c1A10, c2aB7, and d 1 A5 showed the best blocking of CD200 and CD200R interaction at 6.7 p.g/ml of Fab (Fig. 22). These clones are referred to as cGIR3A 10, cG2aR3B7 and dGlR3A5, respectively in Figs. 2) IA and/or B.

Chimerization and 1QG conversion Six antibodies were selected for chimerization and IgG conversion. (See Fig.
23.) They were ciA10 (cGIR3A10), c2aA10 (cG2aR3A10), c2aB7 (cG2aR3B7), dlA5 (dG1R3A5), d1B5 (dGIR3135), and d1B10 (d(731R3B10). For the chimerization overlap PCR was performed to connect mouse kappa chain variable region and human kappa chain constant region. Mouse heavy chain variable region was amplified with a 3' primer that contains a partial human IgGI constant region and Apa I site for cloning.
Amplified kappa chain fragments and heavy chain fragments were cloned into PAX243h(3I
vector (see published International Application WO 2004/078937) that contains human 1gGI
constant region at Kba IINot I for kappa light chain and Mio I/Apa I for heavy chain fragment. Binding of chimeric Fab to CD200 was confirmed by ELISA and flow cytometry. These chimeric Fabs were converted into IgG by insertion of human cytornegalovirus immediate early promoter 0hCIv1V IE Pro) sequence for the heavy chain expression at Not I/:k ,o I, then the transfer of the light chain and heavy chain into a human IgGI expression vector at Xba I/Pin Al sites. This vector has an additional hCIvIV
LE Pro sequence upstream Xba I site for the light chain expression in mammalian cells.
The DNA sequences were confirmed and maxi prep DNA was prepared using HiSpeed"rM
Maxi prepTM columns (QIAGEN) for mammalian cell transfection. Transient transfection was performed in 293-EBNA cells using EffecteneTM (QIAGEN) according to the manufacturer's protocol with the addition of pAdVAntage vector (Promega US, Madison, WI). Stable cell line transfection was performed hn NSO cells using Effectene according to the manufacturer's protocol. After a small scale transient transfection, culture supernatant for each antibody was tested by ELISA (Fig. 24). After a large scale transient transfection, each IgG was purified from the culture supernatant by anti-human IgG
F(ab' )r afrtnity column using FPLC (Amersham Biosciences).
The Purified IgG were tested in a bead inhibition assay as described for the Fabs.
All antibodies directed against CD200 blocked the receptor ligand interaction very well as shown below in Fig. 25.

Mixed lymphocyte reaction Whether blocking of CD200 interaction with its receptor also prevents the cytokine shift from Thl to Th2 observed in mixed lymphocytes reactions in the presence of CD200 was evaluated. Buffy coats were obtained from the San Diego Blood Bank and primary blood lymphocytes (PBL) were isolated using Flistopaque (Sigma-Aldrich). Cells were adhered for 1 h in E1\4EM containing 20ro human serum followed by vigorous washing with PBS. Cells were cultured for 5 days in the presence of GIMI-CSF, 1L-4 and IFN-^i. Matured cells were harvested and irradiated at 2000 RAD using a y irradiator (University of California San Diego). Mixed lymphocyte reactions were set up in 24 well plates using 500,000 dendritic cells and I x 10 responder cells. Responder cells were T

cell enriched lymphocytes purified from peripheral blood using Histopaque. T
cells were enriched by incubating the cells for 1 hour in tissue culture flasks and taking the non-adherent cell fraction. Five hundred thousand CD200 expressing primary irradiated CLL
cells were added to the dendritic cells in the presence or absence of various amounts of anti-CM00 antibodies 2-4 hours before the lymphocyte addition. Supernatants were collected after 48 and 68 hours and cytokines such as IL-2, IFN-y, TL-4, IL-10 and IL-6 were quantified using ELISA. Matched capture and detection antibody pairs for each cytokine were obtained from R+D Systems (Minneapolis), and a standard curve for each cytokine was produced using recombinant human cytokine. Anti-cytokine capture antibody was coated on the plate in PBS at the optimum concentration. After overnight incubation, the plates were washed and blocked for l h with PBS containing 1 %
BSA and 5% sucrose. After 3 washes with PBS containing 0.05% Tween, supernatants were added at the indicated dilutions in PBS containing 1% BSA. Captured cytokines were detected with the appropriate biotinylated anti-cytokine antibody followed by the addition of alkaline phosphatase conjugated streptavidin and SigmaS substrate. Color development was assessed with an ELISA plate reader (Molecular Devices Corp., Sunnyvale, CA). As shown in Figs. 26A and B, the presence of CLL cells completely abrogated IFN-gamma and most of IL -21 production observed in the mixed lymphocyte reaction.
Presence of any of the antibodies allowed for production of these ThI cytokines (Figs. 26A and B). In contrast, IL-10 production was downregulated in the presence of the antibodies. (See Fig.
26C.) Antibod -dependent cell-mediated cfnotoxicity.assay Furthermore, the six chimeric mouse anti-CD200 antibodies were evaluated for their ability to kill CD200 expressing tumor cells in an antibody-dependent cell-mediated cytotoxicity assay (ADCC). 293-EBNA cells transfected with CD200 were labeled with 100 ttCi/million cells in 0.Sml medium for lhr at 37 C. After 3 washes, cells were counted, resuspended in medium (RPIvfl supplemented with 10% human AB serum) at 0.2 million/ml and 50 td (10,000 cells/well) was dispensed in triplicate into a 96 well round bottom plate. 20 l of anti-CD200 antibodies were dispensed into each well so as to achieve a final concentration of 20 tglml. Peripheral blood mononuclear cells (effector cells) were isolated on a Ficoll gradient, red blood cells were lysed with ammonium chloride, washed and resuspended in culture medium and 50 l of cells were dispensed =19 into each well. The assay plates were spun (1,500 rpm/5 minutes/low brake) and transferred to the cell culture incubator. After 4 hours, assay plates were spun as before-36 l of the supernatants were transferred to pico plates and mixed with 250 l microscint-20 cocktail, and placed on the orbital shaker for 2 minutes and read on a Top count. As illustrated in the Fig. 27, all of the mouse chimeric CD200 antibodies produced similar levels of lysis when cultured with CD200 positive cells. No lysis was observed with CD200 negative cells. In addition, the extent of lysis was statistically significant (p <
0.05) when compared to isotype control antibody, d2A6 (anti-M32028 antibody).

EXAMPLE, 5 Raji/PBL Model NOD.CB17-Prkdc<scid> mice (Jackson Laboratory) were injected with 200 l 1''I containing 4x106 RA.M cells (ATCC) s.c. along with 0, 1, 5 or 10 million PBLs.
Nine or ten mice were included per group. PBLs were isolated from 250 m1 whole ammonium chloride. Tumor growth was monitored three times a week by measuring length and width with a caliper. Tumor volume was calculated based on length x width x width/2.

Differences between the groups that were injected with PBLs compared to the group that received tumor cells only were analyzed by 2-tailed unpaired Student's t-test.
Significant differences were observed in the groups that received 5 or 10 million PBLs, but not in the group that received 1 million PBLs horn Day 32 on. The data shown in Figure 28 are a representative example of 10 experiments using different PBL
donors.
Namalwa PBL model NOD.CB17-Prkdc<scid> mice (Jackson Laboratory, Bar Harbor, Maine) were injected with 200 1 RPhv11 containing 4x106Namalwa cells (ATCC) s.c. along with 0, 2 or 10 million PBLs. 9-10 mice were included per group. PBLs were isolated from 250 ml whole blood on a histopaque gradient followed by red blood cell lysis using 0.9uo ammonium chloride. Tumor growth was monitored three times a week by measuring length and width with a caliper. Tumor volume was calculated based on length x width x width/2-Figure 29 shows differences between the groups that were injected with PBLs compared to the group that received tumor cells only analyzed by 2-tailed unpaired Student's t-test. Significant differences were observed in the groups that received 10 million PBLs for both donors, but not in the group that received 2 million PBLs from day bon.

Creation of stable CD200-expressing cell lines Stable CD200-expressing Raji and Namahwa cell lines were generated using the VirapowerTM Lentiviral Expression System (invitrogen, Carlsbad, CA). A CD200 cDNA
was isolated from primary CLL cells by RT-PCR using forward primer 5'-GACA-^>_GCTTGCAAGGATGGAGAGGCTGGTGA-3' (SEQ ID NO: 212) and reverse primer 5'-GACG(--jATCCGCCCC=ITCCTCCTOCTTTTCTC-3' (SEQ ID h10:
213). The PCR product was cloned into the Gateway entry vector pCR8/GW/1'OPO-TA
and individual clones were sequenced. Clones with the correct sequence were recombined in both the sense and antisense orientations into the lentiviral vectors pLenti6/V5/DEST and pLenti6/UbC/V5/DEST using Gateway technology (Invitrogen, Carlsbad, CA)_ The primary difference between these two vectors is the promoter used to drive CD200 expression: pLenti6/V5/DEST contains the human CNN immediate early promoter, whereas pLenti6/ TbC/V5/DEST contains the human ubiquitin C
promoter.

High-titer, VSV-G pseudotyped lentiviral stocks were produced by transient coiransfection of 293-FT cells as recommended by the manufacrurer. Raji or Narnalwa cells were transduced by resuspending 106 cells in lml of growth medium containing and adding 1 ml of lentiviral stock. After incubating the cells 12 g/ml PolybreneTM
overnight at 37 C, the medium containing virus was removed and replaced with 4111 of fresh medium. Two days later, the infected cells were analyzed for CD200 expression by flow cytometry. In all experiments, k70% of the cells were CD200+, whereas CD-200 was undetectable in the parental cell lines and in cells transduced with the negative control (antisense CD200) viruses.

To isolate clonal cell lines that overexpress CD200, the infected cells were selected with blasticidin for 13 days. The concentrations of blasticidin used were 6 I.ig/ml for Raji cells or 2 gghnl for Namalwa cells. Stable clones were then isolated by limiting dilution of the blasticidin-resistant cells into 96-well plates. Clones were screened in 96-well format by flow cytometry using PE-conjugated Mouse Anti-Human CD200 (clone mRC Ox104, Serotec) and a BD FACSCalibur equipped with a High Throughput Sampler. After screening a total of 2000 Raji and 2000 Namalwa clones, those clones titi~ith the highest CD200 expression were expanded for further characterization using conventional techniques.

lnimunosugpressive effect of CD200 in flit RAJI/PBL model In accordance with the methods disclosed, it has been demonstrated that CD200 is upregulated on CLL cells. Upregulation of this molecule might potentially be inununosuppressive. To test whether cancer cells expressing CD200 prevent the immune system horn eradicating the cancer cells, RAE cells, that normally do not express CD200, were infected with a Ientivirus vector system encoding [or CD200 as described above.
RA.11 clones stably expressing CD200 were selected. As a control to ensure that there was no effect of vector infection, clones expressing a reversed, nonhtnctional form of CD200 (CD200rev) were also selected. RAJI cells expressing CD200, CD200REV or the parental RAJI cells were injected subcutaneously into NOD.CBI7-Prkdc<scid>
mice.
The following groups were included in the study:

group 1: =1 x 10 RAJI s.c.; 9 mice This group was needed to ensure that the lentivirus transduced cells show similar growth as the parent cells.
group 2: -l x 10 RAJ1CD200 s.c.; 9 mice This group was needed to ensure that the CD200 transduced cells showed similar growth as the parent cells. Also, this group will give the maximum ?5 tumor growth. Group 3 and group 4 were compared to this group.
group 3: 4 x 10 RAJICD200 + 5x.10 PBL s.c.; 9 mice This PBL number has been shown to reduce tumor growth in some mice in previous experiments. Rejection is not as strong as with 10 million cells, but in order to determine whether CD200 can affect only a certain number of cells, and 5x 10 is the minin-min amount of PilLs which can be used to get rejection:.
rejection should be prevented by the presence of CD200.
group 1: 1 x 10 RAI1CD200 H- IOxl06 P13L s.c.; 8 mice This is the optimum number ofPBLs to see rejection in the RA_11/1'I3L model.

The design was that CD200 expression would prevent this rejection.
group 5: 4x106 RAJICD200rev s.c.; 9 mice This group was needed to ensure that the lentivirus transduced cells show similar growth as the parent cells.
group 6: 4x10 RAJ1CD200rev + 2x10 PBL s.c.; 9 mice This number of PBLs should not result in strong rejection or reduction of tumor growth. This is the positive control for group 3 and group 4 (maximum expected tumor growth). If there is no rejection in this group, then the donor PBLs were hyperactivated to start with which could explain lack of an effect by CD200.
group 7: 4x 10 RAJICD200rev + 5x10 PBL s.c.; 9 mice Controls that any observed effects in the CD200 group 3 are really related to CD200 and not to lentivirus transduction.
group 8: 4 x 10" RAJICD200rev -F 10x10" PBL s.c.; 8 mice iS Controls that any observed effects in the CD200 group 4 are really related to CD200 and not to lentivirus transduction.

Animals were sacrificed at clay 38 based on tumors reaching a size above acceptable limits. Tumors from 4 animals/group were removed. Two tumors/group were frozen in OCT, the other 2 were used to isolate cells and analyze by FAGS for expression. Figures 30(a-c) demonstrate the results for this study.
Although RAIICD200 cells appeared to grow somewhat more slowly, the growth difference between transduced and parental cells did not reach statistical significance as shown in Figure 30(a).
The presence of PBLs slowed tumor growth by up to 8/11 % when 5 or 10110 PBLs were injected, although generally IOxI 0 PBL resulted in a stronger reduction over time compared to 5x106 PBL. The reduction in growth compared to the parent tumor cells was significant from day 20 on. 2x1O" PBL resulted in a significant tumor growth reduction from d22-d29, but that reduction was overcome at later timepoints (Fig. 30(b)).
This study indicated that this particular donor rejects RAJI tumor cells very strongly.
Tumor growth in the groups that received CD200 expressing RAJI cells and PBLs was not significantly different from the tumor growth in the group that only received RAJ1 cells although mice that received I Ox10 P13L showed a trend Of reduced tumor growth, but the difference reached no statistical significance at any time point after tumors reach 100 rum3. Every mouse in the group that received RAJI cells and 5x10 PBL developed a second tumor, some mice as early as d7, while this was not observed in any other group. For analysis, the second tumor was added to the first tumor and the combined size is shown in Fig. 30(c).

These results indicate that CD200 expression on tumor cells does indeed prevent the immune system fi-on-i slowing tumor growth. Also, this study demonstrates the usefulness of the RAJI/PBL model to assess immunosuppressive compounds or molecules.

hmnunos ressive effect of CD200 in the Namalwa/PBL model To evaluate whether the effects seen in the RAJI/PBL model can also be observed in other tumor models, Namalwa tumor cells were also infected with the lentivirusCD200 system and stable clones selected. As a control to ensure that there is no effect of vector infection, clones expressing a reversed, nonfunctional form of CD200 (CD200rev) were also selected. NOD.CB17-Prkdc<scid> mice were injected according to the following scheme as shown in Figures 31 (a)-(d):

group 1: 4 x 106 Namalwa s.c.; 9 mice This group was needed to ensure that the lentivirus transduced cells show similar growth as the parent cells.
group 2: 4 x 10 Namalwa CD200 (ID12Ub) s.c.; 9 mice This group was needed to ensure that flit CD200 transduced cells showed similar growth as the parent cells. Also, this group will give the maximum tumor growth. Group 3 and group 4 were compared to this group.
group 3: 4 x 106 Namalwa CD200 (1D12Ub)+ 5x10 PBL s.c.; 9 mice This PBL number has been shown to reduce tumor growth in some mice in previous experiments. Rejection is not as strong as with 10 million cells, but if CD200 can affect only a certain number of cells, and this is the minimum PBI-we can use to get rejection; rejection will be prevented by the presence of CD200.
group 4: 4 x 10 Namalwa CD200 (ID12Uh) + 10x10 PBL s.c.; 8 mice This is the optimum number of PBLs to see rejection in the Namalwa /.PBL
model. This was done to show that CD200 expression prevents this rejection.

group 5: 4 x 106 Namalwa CD200rev (C5Ubrev) s.c.; 9 mice This group was needed to ensure that the lentivirus transduced cells showed similar growth as the parent cells.
group 6: 4 x 106 Namalwa CD200rev + 2x10 PBL s.e.; 9 mice This number of PBLs should not result in strong rejection of reduction of tumor growth. This is the positive control for group 3 and group 4 (maximum expected tumor growth). if there is no rejection in this group, then the donor PBLs were hyperactivated to start with which would explain lack of an effect by CD200.
group 7: 4 x 106 Namalwa CD200rev + Sxl06 PBL s.c.; 9 mice This group was needed as a control to detect any effects in the CD200 group 3 that were related to CD200 and not to lentivirus transduction.
group 8: 4 x 106 Namalwa CD200rev + 10x10 PBL s.c.; 8 mice This group was needed as a control to detect any effects in the CD200 group 4 that were related to CD200 and not to lentivirus transduction.

Tumor length and width were assessed three times/week.
All tumor cells resulted in rapid tumor growth. There was no significant growth difference between transdueed and parental cells. The tumor grows more aggressively than previously observed as shown in Figure 31(a).
Figure 31(b) shows the presence of PBLs slows tumor growth by about 50%.
This trend was observed from d12 on. The differences of the PBL treated group versus groups that received only tumor cells are statistically significant (as determined by 2-tailed Student's t-test) at d17 and d19 when 2 million or 10 million PBLs were injected.
Injection of 5 million PBLs resulted in tumor growth reduction, but did not reach significance.
Tumor growth in the groups that received CD200 expressing Namalwa cells and PBLs was similar to the tumor growth in the group that only received Namalwa cells (Figure 31(c)).
These data confirm that CD200 expression on tumor cells prevents slowing of tumor growth by the human immune system.

Blockage of the immunosup ressive effect of CD200 in the RAJI/PBI, model by anti-CD200 antibodies To evaluate whether anti-CD200 antibodies can block the immunosuppressive effect of CD200 expressed on tumor cells, RAIl cells transduced with CD200 were injected s.c.
into NOD.CB17-Prkdc<scid> mice, and the ability of PBLs to reduce tumor growth in the presence or absence of chimeric anti-CD200 antibodies d1135 and c2aB7 or a control antibody that does not bind tuinor cells (alxn4100) Was assessed. Antibodies Were administered initially at 10 mg/kg or 2.5 mg/kg with the tumor cells, and then at concentrations indicated below twice/week i.v.. The following groups were set up:
1. 4x 10 RAJICD200; 10 mice 2. 4x 10 RAJ] CD200 + 5 x 106 PBL; 10 mice 3. 4x 10 RAJICD200 + 5 x 106 PBL + 100 mg/kg d 1B5; 12 slice 4. 4 x 10 RAJICD200 + 5 x 106 PBL + 20 mg/kg d I B 5; 9 mice 5. 4 x 106 RAJICD200 + 5 x 106 PBL + 100 mg/kg c2aB7; I1 mice 6. 4 x 106 RAJICD200 + 5 x 10 PBL + 20 mg/kg c2aB7; 9 mice 7. 4 x 10 RAJICD200 + 5 x 10 PBL + 100 mg/kg alxn4100; 9 mice Tumor length and width was measured 3 times a week, and the tumor volume was calculated by tumor length*width*width/2. Figure 33 shows that as expected, expression on the tumor cells prevented the immune cells from reducing tumor growth.
However, addition of anti-CD200 antibodies reduced the tumor- volume by 50-75 %. The reduction in growth by the antibodies was statistically significant as determined by Student's t-test and Mann Whitney test from day 18 on through the end of the study. In contrast, treatment with the control antibody did not reduce the tumor growth.
These data demonstrate the usefulness of anti-Cf)200 in anti-cancer treatment. Also, this study demonstrates the usefulness of the RAJI/PBL model to assess immunomodulatory therapeutics.

Blockage of the immunosuppressive effect of CD200 in the Namaliia/PBL, model by anti-CD200 antibodies To evaluate whether the effect seen with the anti-CD200 antibodies in the RAJ1 PBL
model can also be observed in other humor models, RAJl cells transduced with were injected s.c. into NOD.CB 17-Prkdc<scid> mice, and the ability of PBLs to reduce tumor growth in the presence or absence of chimeric anti-CD200 antibodies d1B5 and c2aB7 or a control antibody that does not bind tumor cells (alxn4100) was assessed.
Antibodies at concentrations indicated below were administered initially with the tumor cells, and then twice/week s.c. within 0.5 cm of the tumor. The following groups were set up (10 mice/group unless indicated otherwise):

1. 4 x 106 NamalwaCD200 2. 4 x 10 NamalwaCD200 + 5 x 10 PBL; 12 mice 3. 4 x 10 NamalwaCD200 + 5 x 106 PBL + 10 mg/kg d1B5; 12 mice 4. 4 x 106 NamalwaCD200+ 5 x 106 PBL+2.5 mg/kg d1B5 5. 4 x 10 NamalwaCD200 + 5 x 106 PBL + 10 mg/kg c2aB7; 12 mice 6. 4 x 10 NamalwaCD200 + 5 x 106 PBL + 2.5 mg/kg c2aB7 7. 4 x 10 NamalwaCD200 + 5 x 106 PBL + 10 mg/kg alxn4 100 Tumor length and width was measured 3 times a week, and the tumor volume was calculated by tumor length:"width*width/2. Figure 34 shows that as expected, expression on the tumor cells prevented the immune cells from reducing tumor growth.
However, addition of anti-CD200 antibodies reduced the tumor volume by up to 97' o.
The reduction in growth by the antibodies was statistically significant as determined by Student's t-test and lvlann Whitney test from day 12 on through the end of the study. In contrast, treatment with the control antibody did not reduce the tumor growth.
These data confirm the usefulness of anti-CD200 in anti-cancer treatment, and the use of the tumor/PBL models in assessing immunomodulatoiy therapeutics.

Detection of a potential immune-enhancing effect of compounds in the RAI PBL
model T cells isolated from PBLs using CD3 columns (Miltenyi) were incubated in vitro with an ascites preparation of AZND 1 (anti-DC-SIGN antibody) or an ascites preparation of a control antibody (BB5). NOD.CB17-Prkdc<scid> mice were injected according to the following scheme:

Group 6: 8 mice, 4x10 RAJI -F0.8xl0 AZNDI-potentiated T cells -F 2x10 PBL
s.c.
Group 5: 10 mice, 4x106 RAJI s.c.
Group 4: 8 mice, 4x106 RAJI + 8x106 PBL s.c.
Group 3: 8 mice, 1x106 RAJ, +0.8x10 freshT cells s.c.
Group 2: 8 mice, 1x106 RAJI +0.8x106AZNDI-potentiated T cells s.c.
Group 1: 8 mice, 4x106 RAJI +0.8x106 BB5.1-potentiated T cells s.c.

Tumor length and width were measured 3 times/week.
While T cells incubated with the negative control 13135 did not reduce tumor growth, AZNDI treated T cells did reduce tumor growth significantly (See, Figure 32).
These results demonstrate that the RAJI PBL model can be used to assess efficacy of immune-enhancing compounds.

EXAIvfPLE 6 Determination of CD200 U pre elation in CLL Patients Lymphocytes from 15 CLL patients were stained with FITC-conjugated anti-CD5 (e-bioscience), APC-conjugated anti-CD19 (e-bioscience) and PE-conjugated anti-(Serolec). Lymphocytes from healthy donors were stained accordingly. CD200 expression on CD5-+-C.D19-+- cells was determined. As shown in the table below, although the level of CD200 expression varied among CLL patient samples, all CLL
samples showed elevated levels (1.6-4-fold range) higher CD200 expression compared to expression on normal B cells. The CLL patients showing elevated levels of expression are selected for anti-CD200 treatment in accordance wvith the methods described herein.

FACS analysis of CD200 expression on B-CLL cells in comparison to normal B cells CLL sample Healthy donor Donor ID B-CLL CD200 Normal B Ratio(CLL/normal B) (GMMMMFI) CD200 (GNMMFI) RC0 11731 93 58 1.6 RF020934 659 185 3.6 JA073031 334 64 5.2 GR011846 1.56 64 2.4 BB101735 420 95 4.4 DM6988172 290 97 2.9 CB8267677 300 97 3.1 GB 1325248 178 77(7) 2.3 VN7029373 154 77(7) 2.0 DG8942820 146 77(7) 1.9 MA48451869 237 77(7) 3.1 J R4539931 215 77(7) 2.8 IIS6787771 305 77(7) 4.0 VB040439 123 41 3.0 MEAN= 3.1 STDEV= 1.0 REFERENCES

The following references more fully describe the state of the art to which the present invention pertains. Any inconsistency between these publications below or those noted above and the present disclosure shall be resolved in favor of the present disclosure.

1) Agarwal, at al., (2003;). Disregulated expression of the Th2 cytokine gene in patients with intraoral squamotts cell carcinoma. Immunol Invest 32:17-30.
2) Almasri, lvIlvl at al. (1992). Am .1 Ilemato140 259-263.
3) Contasta, et al., (2003). Passage from normal mucosa to adenoma and colon cancer: alteration of normal sCD30 mechanisms regulating THl/TH2 cell functions.
Cancer Biother Radiophar n 18:549-557.
4) Gorezynski, et al., (1998). Increased expression of the novel molecule O _-2 is involved in prolongation of murine renal allograft survival. Transplantation 65:1106-1114.
S) Gorczynski, at al., (2001). Evidence of a role for CD200 in regulation of immune rejection of leukaernic tumour cells in C57BL/6 mice. Clin Exp Imnnunol 1?6:220 2 9.
6) Hainsworth, JD (2000). Oncologist 2000;5(7):376-04.
7) Inagawa, et al., (1998). Mechanisms by which chemotherapeutic agents augment the antitumor effects of tumor necrosis factor: involvement of the pattern shift of cytokines from Th2 to Thl in tumor lesions. Anticancer Res 18:3957-3964.

8) Ito, at al., (1999). Lung carcinoma: analysis of T helper type I and 2 cells and T
cytotoxic type 1 and 2 cells by intracellular cytokine detection with flow cytometry.
Cancer 85:2359-2367.

9) Mani, at al., (2003). Normal intrinsic 7:h1iTh2 balance in patients with chronic phase chronic myeloid leukemia not treated with interferon-alpha or iniatinib.
Haematologica 88:754-761.

;9 10) L,auerova, et al., (2002). Malignant melanoma associates with Thl/Th2 imbalance that coincides with disease progression and immunotherapy response. Neoplasma 49:159-11 ) Maggio, at al., (2002). Chernokines, cytokines and their receptors in I-lodgkin's lymphoma cell lines and tissues. Ann Oncol 13 Suppl 1:52-56.

12) Nilsson; K (1992). Burn Cell. 5(1):25-41.

13) Podhoreclca, et al., (2002). T type 1/type 2 subsets balance in B-cell chronic lymphocytic leukemia--the three-color flow cytometry analysis. Leuk Res 26:657-660.
1-1) Pit, QQ and Bezwoda, W (2000). Anticancer Res. 20(4):2569- 78.
15) Smyth, et al., (2003). Renal cell carcinoma induces prostaglandin E2 and T-helper type 2 cytokine production in peripheral blood mononuclear cells. Ann Surg Oncol 10:455-462.
16) Tatsumi, et al., (2002). Disease-associated bias in T helper type 1 (Thl)/Th2 CD4(+) T cell responses against IvLAGE-6 in HLA-DRB 10401(+) patients with renal cell carcinoma or melanoma. I Ex p Med 196:619-628.
17) Walls A Vet al. (1989). Int. J Cancer 44846-853.
IS) Winter, at al., (2003). Tumour-induced polarization of tumour vaccine-draining lymph node T cells to a type 1 cy,tokine profile predicts inherent strong immunogenicity of the tumour and correlates with therapeutic efficacy in adoptive transfer studies.
Immunology 108:409-419.

It will be understood that various modifications may be made to the preferred embodiments disclosed herein. For example, as those skilled in the art will appreciate, the specific sequences described herein can be altered slightly without necessarily adversely 23 affecting the functionality of the polypeptide, antibody or antibody fragment used in binding OX-2/CD200. For instance, substitutions of single or multiple amino acids in the antibody sequence can frequently be made without destroying the functionality of the antibody or fragment. Thus, it should be understood that polypeptides or antibodies having a degree of identity greater than 70 %o to the specific antibodies described herein are contemplated. In particularly useful embodiments; antibodies having an identity greater than about 80% to the specific antibodies described herein are contemplated.
In other useful embodiments, antibodies having an identity greater than about 90 i%, to the specific antibodies described herein are contemplated. Therefore. the above description should not be construed as limiting, but merely as exemplifications of preferred embodiments. The scope of the claims should not be limited by the preferred embodiments set forth herein; but should be given the broadest interpretation consistent with the disclosure as a whole.

Claims (17)

WHAT IS CLAIMED IS:

1. An isolated antibody, or a CD200-binding fragment thereof, comprising at least one antigen binding site that binds to human CD200 (hCD200), wherein the antibody or CD200-binding fragment thereof is a humanized, fully human, or chimeric antibody or CD200-binding fragment thereof, and wherein a Fab fragment of the antibody inhibits the interaction between hCD200 and hCD200 receptor by at least 90% at a concentration of 6.7 µg/mL as evaluated in an in vitro fluorescent microsphere assay comprising the following steps:

(i) forming an Fab-microsphere mixture by contacting, in an aqueous solution, I
million hCD200-coated carboxylate-modified fluorescent microspheres with the Fab fragment at a concentration of 6.7 µg/mL for 10 minutes to thereby allow for binding of the Fab fragment to the hCD200 present on the microspheres;

(ii) forming an Fab-microsphere-cell mixture by further contacting the mixture of (i) with 50,000 mammalian cells expressing hCD200 receptor protein for 30 minutes at 37°C to thereby allow for binding of the hCD200 protein present on the microspheres to the hCD200 receptor present on the cells; and (iii) subjecting the Fab-microsphere-cell mixture of step (ii) to flow cytometry to thereby determine the amount of fluorescent signal associated with the cells, as a function of the amount of interaction between the hCD200 protein present on the microspheres and the hCD200 receptor protein present on the cells, wherein at least a 90%
reduction in the fluorescent signal associated with the cells, as compared to a control fluorescent signal, indicates that the Fab fragment inhibits the interaction between hCD200 and hCD200R by at least 90% at a concentration of 6.7 µg/mL, and wherein the control signal is the amount of fluorescent signal associated with a corresponding set of expressing cells incubated under the same conditions as (ii) with 1 million hCD200-coated carboxylate-modified fluorescent microspheres that were not contacted with the antibody fragment.

2. The isolated antibody, or CD200-binding fragment thereof, according to claim 1, wherein the CD200-binding fragment is selected from the group consisting of an Fv, scFv, Fab' or F(ab')2 fragment.

3. The isolated antibody, or CD200-binding fragment thereof, according to claim 1 or 2, wherein the antibody or CD200-binding fragment comprises: a light chain CDR1 having the sequence set forth in residues 26-36 of SEQ ID NO:209; a light chain CDR2 having the sequence set forth in residues 52-58 of SEQ ID NO:209; a light chain CDR3 having the sequence set forth in residues 91-99 of SEQ ID NO:209; a heavy chain CDR1 having the sequence set forth in SEQ ID

NO: 149; a heavy chain CDR2 having the sequence set forth in SEQ ID NO: 174;
and a heavy chain CDR3 having the sequence set forth in SEQ ID NO: 195.

4. The isolated antibody, or CD200-binding fragment thereof, according to claim 1 or 2, according to any one of claims 1-3, wherein the antibody or CD200-binding fragment comprises:
a light chain CDR1 having the sequence set forth in residues 26-36 of SEQ ID
NO:208, a light chain CDR2 having the sequence set forth in residues 52-58 of SEQ ID NO:208, a light chain CDR3 having the sequence set forth in residues 91-99 of SEQ ID NO:208, a heavy chain CDR1 having the sequence set forth in SEQ ID NO: 114, a heavy chain CDR2 having the sequence set forth in SEQ ID NO: 165, and a heavy chain CDR3 having the sequence set forth in SEQ ID
NO:189.

5. The isolated antibody, or CD200-binding fragment thereof, according to claim 1 or 2, wherein the antibody or CD200-binding fragment comprises: a light chain CDR1 having the sequence set forth in residues 26-35 of SEQ ID NO:207; a light chain CDR2 having the sequence set forth in residues 51-57 of SEQ ID NO:207; a light chain CDR3 having the sequence set forth in residues 90-98 of SEQ ID NO:207; a heavy chain CDR1 having the sequence set forth in SEQ
ID NO:111; a heavy chain CDR2 having the sequence set forth in SEQ ID NO: 118;
and a heavy chain CDR3 having the sequence set forth in residues 100-112 of SEQ ID NO:201.

6. The isolated antibody, or CD200-binding fragment thereof, according to claim 1 or 2, wherein the antibody or CD200-binding fragment comprises: a light chain polypeptide comprising: a light chain CDR1 having the sequence set forth in residues 26-41 of SEQ ID NO:210; a light chain CDR2 having the sequence set forth in residues 57-63 of SEQ ID NO:2 10; a light chain CDR3 having the sequence set forth in residues 96-104 of SEQ ID NO:210; a heavy chain CDR1 having the sequence set forth in SEQ ID NO:140; a heavy chain CDR2 having the sequence set forth in SEQ ID NO: 162; and a heavy chain CDR3 having the sequence set forth in SEQ ID
NO:187.

7. A pharmaceutical composition comprising a pharmaceutically acceptable carrier and the isolated antibody, or CD200-binding fragment, according to any one of claims 1-6.

8. The isolated antibody, or CD200-binding fragment, according to any one of claims 1-6 for use in treating a cancer in a subject.

9. The isolated antibody fragment, or CD200-binding fragment thereof, for use according to claim 8, wherein the cancer comprises cancer cells overexpressing CD200.

10. The isolated antibody fragment, or CD200-binding fragment thereof, for use according to claim 8 or 9, wherein the cancer is chronic lymphocytic leukemia.

11. The isolated antibody fragment, or CD200-binding fragment thereof for use according to claim 10, wherein the chronic lymphocytic leukemia is B cell chronic lymphocytic leukemia.

12. The pharmaceutical composition according to claim 7 for use in treating cancer.

13. A method for determining whether a human subject is afflicted with a cancer, the method comprising:
(a) determining the level of expression of human CD200 by cancer cells obtained from a human subject, and (b) comparing the level of expression in step (a) to the level of expression of human CD200 in normal cells of the same histological type as the cells from which the cancer cells are derived, wherein a level of human OX-2/CD200 expression in the human subject's cancer cells that is elevated by at least 1.4 fold, as compared to the level of expression in the normal cells, indicates that the human subject has a cancer.

14. The method of claim 13, wherein the cancer cells are chronic lymphocytic leukemia cells.

15. The method of claim 11, wherein the chronic lymphocytic leukemia cells are B cell chronic lymphocytic leukemia cells.

16. The method of any one of claims 13 15, wherein a level of human OX-2/CD200 expression in the human subject's cancer cells that is elevated by at least 1.6 fold, as compared to the level of expression in the normal cells, indicates that the human subject has a cancer.

17. The method of any one of claims 13-16, wherein the level of expression of OX-2/CD200 is determined using an antibody, or a CD200-binding fragment thereof, selected from the group consisting of:

(i) an antibody, or a CD200-binding fragment thereof, comprising: a light chain CDR1 having the sequence set forth in residues 26-36 of SEQ ID NO:209; a light chain CDR2 having the sequence set forth in residues 52-58 of SEQ ID NO:209; a light chain CDR3 having the sequence set forth in residues 91-99 of SEQ ID NO:209; a heavy chain CDR1 having the sequence set forth in SEQ ID NO: 149-1 a heavy chain CDR2 having the sequence set forth in SEQ ID NO: 174; and a heavy chain CDR3 having the sequence set forth In SEQ ID
NO: 195;

(ii) an antibody, or a CD200-binding fragment thereof, comprising: a light chain CDR1 having the sequence set forth in residues 26-36 of SEQ ID NO:208, a light chain CDR2 having the sequence set forth in residues 52-58 of SEQ ID NO:208, a light chain CDR3 having the sequence set forth in residues 91-99 of SEQ ID NO:208, a heavy chain CDR1 having the sequence set forth in SEQ ID NO:114, a heavy chain CDR2 having the sequence set forth in SEQ ID NO: 165, and a heavy chain CDR3 having the sequence set forth in SEQ ID
NO: 189;

(iii) an antibody, or a CD200-binding fragment thereof, comprising: a light chain CDR1 having the sequence set forth in residues 26-35 of SEQ ID NO:207; a light chain CDR2 having the sequence set forth in residues 51-57 of SEQ ID NO:207; a light chain CDR3 having the sequence set forth in residues 90-98 of SEQ ID NO:207; a heavy chain CDR1 having the sequence set forth in SEQ ID NO:111; a heavy chain CDR2 having the sequence set forth in SEQ ID No: 118, and a heavy chain CDR3 having the sequence set forth in residues 100-112 of SEQ ID NO:201; and (iv) an antibody, or a CD200-binding fragment thereof, comprising: a light chain polypeptide comprising: a light chain CDR1 having the sequence set forth in residues 26-41 of SEQ ID NO:210; a light chain CDR2 having the sequence set forth in residues 57-63 of SEQ ID
NO:210; a light chain CDR3 having the sequence set forth in residues 96-104 of SEQ ID
NO:210; a heavy chain CDR1 having the sequence set forth in SEQ ID N0:140; a heavy chain CDR2 having the sequence set forth in SEQ ID NO:162; and a heavy chain CDR3 having the sequence set forth in SEQ ID NO: 187.