Verification ofTranslation
I have verified the attached document and certify that it is a true
translation of the International Application No. PCT/JP2016/063420
(WO 2016/175307) of 28th April 2016.


DECLARATION
1, Seiichi URUSHIYAMA, of HIRAKI & ASSOCIATES, do solemnly and
sincerely declare as follows:
1. That I am well acquainted with the English and Japanese languages and
am competent to translate from Japanese into English.
2. That I have executed, with the hest of my ability, a true and correct
translation into English of International Application No. PCT/JP2016/063420 filed
on April 28, 2016, a copy of which I attach herewith.


Description
Title of Invention: PHARMACEUTICAL COMPOSITION FOR TREATING AND/OR
PREVENTING CANCER
Technical Field
[0001|
The present invention relates to novel medicinal use of an antibody against CSPG5
protein or a fragment thereof as e.g., a therapeutic and/or prophylactic agent for cancer.
Background Art
[0002]
Cancer is a disease occupying the leading position of cause of death. Therapies
presently employed are primarily based on a surgical therapy in combination with a radiation
therapy and a chemotherapy. Despite of recent development of new surgical techniques and
finding of new anticancer drugs, treatment results of cancers except some cancers have not yet
likely been improved at present. With the advancement of molecular biology and cancer
immunology, antibodies that specifically react with cancers, cancer antigens recognized by
cytotoxic T cells, genes encoding cancer antigens and the like have been identified. Thus,
development of specific cancer treatments targeting cancer antigens has been desired.
[0003]
In the cancer therapy, it is desirable that the peptides (including polypeptides) to be
recognized as antigens are rarely present in normal cells and present specifically in cancer
cells in order to reduce side effects. In 1991. Boon et al. of the Ludwig Laboratory in
Belgium isolated human melanoma antigen MAGE1, recognized by CD8 positive T cells, by
cDNA expression cloning method using an autologous tumor cell line and tumor-reactive T
cells (Non Patent Literature 1). Subsequently. SEREX (serological identification of antigens
by recombinant expression cloning) method was reported, which is a method of identifying a
cancer antigen recognized by an antibody produced in response to autologous cancer in the

body of a patient, by using a gene expression cloning method (Patent Literature 1, Non Patent
Literature 2). Several cancer antigens, which are rarely expressed in normal cells and
specifically expressed in cancer cells, have been isolated by this method (Non Patent
Literature 3). Further, a cell therapy, which uses immune cells targeting a part of the amino
acid sequence of such a cancer antigen and specifically reacting with the cancer antigen, and a
cancer-specific immunotherapy such as a vaccine containing a cancer antigen, are carried out
in clinical trials.
[0004]
In the meantime, various types of antibody medicines for treating cancer, targeting a
specific antigen protein on cancer cells have been known in the world. Most of the target
antigen proteins provide certain levels of medicinal effects as a cancer-specific therapeutic
agent and attract attention; however they express also on a plurality of normal cells. Because
of this, side effects are seriously concerned since not only cancer cells but also normal cells
are damaged as a result of administration of antibodies. Accordingly, it is expected that a
therapy with an antibody medicine having fewer side effects can be realized, if an antigen.
which is specifically expressed only on the cancer cell surfaces and not expressed on normal
cells, can be identified and an antibody targeting the antigen can be used as a medicine.
Citation List
Patent Literature
[0005]
Patent Literature 1: U. S. Patent No. 5698396
Non Patent Literature
[0006]
Non Patent Literature 1: Bruggen P. et al., Science. 254: 1643-1647 (199])
Non Patent Literature 2: Proc. Natl. Acad. Sci. USA, 92: 11810-11813 (1995)
Non Patent Literature 3: Int. J. Cancer, 72: 965-971 (1997)
Summary of Invention

Technical Problem
[0007]
An object of the present invention is to identify a cancer antigen protein specifically
expressed on the surface of cancer cells and provide use of an antibody targeting the cancer
antigen protein as a therapeutic and/or prophylactic agent for cancer.
Solution to Problem
[0008]
The present inventors isolated an antigen specifically expressed in cancer by SEREX
method using cDNA library derived from a canine testicular tissue and the serum of a breast
cancer-bearing dog, and then obtained cDNA encoding CSPG5 protein. CSPG5 protein can
bind to antibodies present in the sera derived from various cancer-bearing living organisms.
The present inventors also found that CSPG5 protein is specifically expressed in breast cancer.
lung cancer, brain tumor, leukemia, malignant lymphoma, adenocarcinoma, mastocytoma.
squamous cell carcinoma, melanoma or neuroblastoma cells: and that a part of CSPG5 protein
is specifically expressed on the surface of these cancer cells. CSPG5 (Chondroitin Sulfate
Proteoglycan 5) protein is type 1 transmembrane protein and one of the neuregulin family
proteins. It is also reported that CSPG5 protein binds to ErbB3 protein to serve as a growth
factor; and that expression of CSPG5 protein increases in ovarian cancer having a mutation of
BRCA1 protein (Kinugasa, Y., et al., 2004, Biochem. Biophys. Res. Commun., 321: 1045;
Press. J. Z., et al, 2010, Neoplasia., 12 (12): 993-1002). It is further known that CSPG5
protein is highly expressed in tissues of the nervous system, such as retinal ganglion cells,
purkinje cells and hippocampus, and serves as a proliferation/differentiation factor of nerve
cells involved in elongation of nerve axon process (Yasuda. Y. et al., 1998, Neurosci. Res., 32:
3)3; Aono, S., et al.. 2006. J. Neurosci. Res., 83: 110; Nakanishi. K„ et al., 2006, J. Biol.
Chem.. 281: 24970). However, there have been no reports that CSPG5 protein has an
immunity inducing activity against cancer cells and thus is useful for treating and preventing
cancer.
[0009]

Also, the inventors prepared CSPG5 protein molecules consisting of amino acid
sequences represented by SEQ ID NOs: 2, 4. 6. 8. 10. 12, 14 and 16 based on the obtained
canine CSPG5 gene and its homologous genes of human, cat and mouse and antibodies against
these CSPG5 protein molecules. Then, they found that an antibody against the portion of
each of these CSPG5 protein molecules expressed on the surfaces of individual cancer cells, in
other words, the extracellular region thereof, damages the cancer cell expressing CSPG5
protein. Based on the finding, the present invention was accomplished.
[0010]
Accordingly, the present invention has the following features.
(1) A pharmaceutical composition for treating and/or preventing cancer, comprising an
antibody or a fragment thereof having immunological reactivity with CSPG5 protein or a
fragment thereof consisting of at least 7 or more consecutive amino acid residues, as an active
ingredient.
(2) The pharmaceutical composition according to {1). wherein the CSPG5 protein consists of
any one of amino acid sequences represented by SEQ ID NOs: 8, 4. 6, 10 and 12. or an amino
acid sequence having an amino acid identity of 80% or more to the amino acid sequence.
(3) The pharmaceutical composition according to (1) or (2), wherein the cancer is leukemia or
malignant lymphoma.
(4) The pharmaceutical composition according to any one of (1) to (3). wherein the antibody is
a monoclonal antibody or a polyclonal antibody.
(5) The pharmaceutical composition according to any one of (1) to (4), wherein the antibody is
a human antibody, a humanized antibody, a chimeric antibody, a single-chain antibody or a
bispecific antibody.
[00111
The specification incorporates the disclosure of JP Patent Application No. 2015-093640
to which the present application claims the priority.
Advantageous Effects of Invention
[0012]

The antibody against CSPG5 protein used in the present invention damages cancer cells.
Accordingly, the antibody against CSPG5 protein is useful for treatment and/or prevention of
cancer.
Brief Description of Drawings
[0013]
[Figure 1] Figure 1 shows expression patterns of identified CSPG5 gene in canine tumor
tissues or cancer cell lines. In the figure, reference number 1 shows expression patterns of
canine CSPG5 gene in individual canine tissues and cell lines: and reference number 2 shows
expression patterns of canine GAPDH gene in individual canine tissues and cell lines.
[Figure 2] Figure 2 shows expression patterns of identified CSPG5 gene in human tumor
tissues or cancer cell lines.
[Figure 3] Figure 3 shows expression patterns of identified CSPG5 gene in mouse tumor
tissues or cancer cell lines. Reference number 3 shows expression patterns of mouse CSPG5
gene in individual mouse tissues and cell lines; reference number 4 shows expression patterns
of mouse GAPDH gene in individual mouse tissues and cell lines.
[Figure 4] Figure 4 shows cytotoxic activity of a polyclonal antibody against CSPG5 protein
(anti-CSPG5 polyclonal antibody) on a leukemia cell line (K562) and malignant lymphoma
cells (L-1236) expressing CSPG5 gene. In the figure, reference number 5 shows the
cytotoxic activity against K.562 cells when a control polyclonal antibody was added; reference
number 6 shows the cytotoxic activity against K.562 cells when an anti-CSPG5 polyclonal
antibody was added; reference number 7 shows the cytotoxic activity against L-1236 cells
when the control polyclonal antibody was added; and reference number 8 shows the cytotoxic
activity against L-1236 cell when the anti-CSPG5 polyclonal antibody was added.
[Figure 5] Figure 5 shows cytotoxic activity of a monoclonal antibody against CSPG5 protein
(anti-CSPG5 monoclonal antibody) on a leukemia cell line (K562) and malignant lymphoma
cells (L-1236) expressing CSPG5 gene. In the figure, reference number 9 shows the
cytotoxic activity against K562 cells when a control monoclonal antibody was added;
reference number 10 shows the cytotoxic activity against K562 cells when an anti-CSPG5

monoclonal antibody was added; reference number 11 shows the cytotoxic activity against L-
1236 cells when the control monoclonal antibody was added; and reference number 12 shows
the cytotoxic activity against L-1236 cell when the anti-CSPG5 monoclonal antibody was
added.
Description of Embodiments
[0014]
The antitumor activity of an antibody against a polypeptide consisting of the amino
acid sequence represented by SEQ ID NO: 2, 4, 6, 8, 10. 12. 14 or 16. used in the present
invention, can be evaluated by checking, in vivo suppression of tumor growth in a cancer-
bearing animal or by checking, whether or not the antibody exerts in vitro cytotoxic activity
against tumor cells expressing the polypeptide via immune cells or a complement, as described
later.
[0015]
The nucleotide sequence of a polynucleotide encoding the protein consisting of the
amino acid sequence represented by SEQ ID NO: 2. 4. 6. 8. 10. 12. 14 or 16 is represented by
SEQ ID NO: 1, 3. 5. 7. 9. 11. 13 or 15. respectively.
[0016]
The amino acid sequence represented by SEQ ID NO: 2 in the sequence listing
disclosed in the present invention is an amino acid sequence of CSPG5 protein isolated as a
polypeptide binding to an antibody specifically present in the serum derived from a cancer-
bearing dog by the SEREX method using a cDNA library derived from a canine testicular
tissue and the serum of a breast cancer-bearing dog; the amino acid sequences represented by
SEQ ID NOs: 4, 6, 8, 10 and 12 are isolated as human homologs of the polypeptide; the amino
acid sequence represented by SEQ ID NO: 14 is isolated as a cat homolog of the polypeptide;
and the amino acid sequence represented by SEQ ID NO: 16 is isolated as a mouse homolog
of the polypeptide (see, Example 1 described later).
[0017]

It has been known from the amino acid sequence that CSPG5 protein is type I
transmembrane protein, and that the extracellular region predicted from its sequence is
expressed on the surface of nerve cells. Owing to the present application, it was continued
that the extracellular region of CSPG5 protein is actually expressed (present) on the surface of
various types of cancer cells. In the present invention, an antibody binding to the
extracellular region of CSPG5 protein on a cancer cell or binding to a polypeptide having an
amino acid identity of 80% or more, preferably 85% or more, more preferably 90% or more.
further preferably 95% or more. 97% or more. 98% or more or 99% or more with the amino
acid sequence of the extracellular region, is preferably used.
[0018]
The antibody against CSPG5 protein used in the present invention may be of any type
of antibody as long as it can exert an antitumor activity. Examples of the antibody include a
monoclonal antibody, a polyclonal antibody, a synthetic antibody, a multi-specific antibody, a
human antibody, a humanized antibody, a chimeric antibody, a single-chain antibody (scFV)
and an antibody fragment (for example, Fab, F(ab')2, Fv). These antibodies and fragments
thereof can be prepared by methods known to those skilled in the art. In the present
invention, an antibody capable of specifically binding to CSPG5 protein is desirable and a
monoclonal antibody is preferable; however, a polyclonal antibody may be used as long as
homogeneous antibodies can be stably produced. When the subject is a human, a human
antibody or a humanized antibody is desirable in order to avoid or suppress a rejection reaction.
[0019]
The phrase "specifically binding to CSPG5 protein" used herein means binding
specifically to CSPG5 protein and substantially not binding to proteins except CSPG5 protein.
[0020]
The subject, which is a target for treatment and or prevention of cancer by the present
invention, is a mammal such as humans, pet animals, domestic animals and animals for
competitive use. preferably a human.
[0021]

Preparation of an antigen, preparation of an antibody and a pharmaceutical composition
according to the present invention is described below.
[0022]

The protein or a fragment thereof used as a sensitizing antigen for obtaining an
antibody against CSPG5 protein (anti-CSPG5 antibody) in the present invention may be
derived from e.g.. humans, dogs, cats, mice, rats, cows, horses and chickens, and the animal
species from which the protein or a fragment thereof is derived is not limited thereto. The
protein or a fragment thereof is preferably selected in consideration of compatibility to a
parent cell to be used in cell fusion. Generally, the protein derived from a mammal is
preferable and particularly the protein derived from a human is preferable. For example, if
the CSPG5 protein is human CSPG5 protein, human CSPG5 protein, a partial polypeptide
thereof, a cell expressing the human CSPG5 protein, and the like can be used.
[0023]
The amino acid sequences and nucleotide sequences of CSPG5 protein and homologs
thereof can be obtained by using, for example. GenBank (NCBI of the United State) and by
using an algorithm such as BLAST and FAST A (Karlin and Altschul. Proc. Natl. Acad. Sci.
USA, 90: 5873-5877,1993; Altschul et al. Nucleic Acids Res. 25: 3389-3402, 1997).
[0024]
For example, if human CSPG5 protein is used as a basis, a nucleotide sequence (SEQ
ID NO: 3. 5, 7. 9 or 11) encoding the human CSPG5 protein and a nucleic acid having a
nucleotide identity of 70% to 100%. preferably 80% to 100%, more preferably 90% to 100%,
further preferably 95% to 100% (for example 97% to 100%, 98% to 100%, 99% to 100% or
99.5% to 100%) to the nucleotide sequence are used as a target. Also, the amino acid
sequence (SEQ ID NO: 4, 6, 8, 10 or 12) of the human CSPG5 protein and a polypeptide
having an amino acid identity of 70% to 100%. preferably 80% to 100%, more preferably 90%
to 100%, or further preferably 95% to 100% (for example, 97% to 100%, 98% to 100%, 99%
to 100% or 99.5% to 100%) to the amino acid sequence are used as a target. The term
"nucleotide identity" used herein refers to the percentage (%) of the number of identical

nucleotides relative to the total number of nucleotides when two nucleotide sequences are
aligned such that they have a maximum degree of similarity by appropriately introducing
gap(s). Similarly, the term "amino acid identity" refers to the percentage (%) of die number
of identical amino acids relative to the total number of amino acids when two amino acid
sequences are aligned such that they have a maximum degree of similarity by appropriately
introducing gap(s).
[0025]
The fragment of CSPG5 protein is specified to have a length equal to or longer than the
amino acid length of an epitope (antigen determinant) and less than the full length of the
protein. The epitope is a polypeptide fragment which is a minimum unit recognized by an
antibody in a mammal, preferably a human and has antigenicity or immunogenicity. and
includes amino acid sequences having a length of about 7 to 12 amino acids, for example. 8 to
11 amino acids.
[0026]
The human CSPG5 protein and a polypeptide containing a partial peptide thereof can
be synthesized, for example, by a chemical synthesis method such as Fmoc method
(fiuorenylmethyloxycarbonyl method) and tBoc method (t-butyloxycarbonyl method)
(Biochemical Experiment Course 1, Chemistry of Protein IV, Chemical Modification and
Synthesis of Peptide, edited by the Japan Biochemical Society. Tokyo Kagaku Dojin (Japan),
1981) or synthesized in accordance with a routine method using a commercially available
peptide synthesizer. Alternatively, a desired polypeptide can be obtained by genetic
engineering technique known in the art (e.g.. Green, M. R. and Sambrook, J., 2012, Molecular
Cloning: A Laboratory Manual 4th Ed., Cold Spring Harbor Laboratory Press, Cold Spring
Harbor. New York, Ausubel et al.. Short Protocols in Molecular Biology, third edition, A
compendium of Methods from Current Protocols in Molecular Biology (1995), John Wiley &
Sons); more specifically by preparing a polynucleotide encoding the polypeptide, integrating
the polynucleotide into an expression vector, introducing the vector into a host cell and
allowing the host cell to produce a polypeptide.
[0027]

The polynucleotide encoding the polypeptide can be easily prepared by a routine
method using a genetic engineering technique known in the art or a commercially available
nucleic acid synthesizer. For example. DNA having the nucleotide sequence of SEQ ID NO:
3 can be prepared by carrying out PCR using a human chromosome DNA library or a human
cDNA library as a template and a pair of primers designed so as to amplify the nucleotide
sequence represented by SEQ ID NO: 3. The reaction conditions of the PCR can be
appropriately defined; for example, the reaction conditions include a condition where using a
heat-resistant DNA polymerase (for example, Taq polymerase) and a Mg '-containing PCR
buffer, a cycle consisting of a denaturation reaction at 94°C for 30 seconds, an annealing
reaction at 55°C for 30 seconds to 1 minute and an elongation reaction at 72°C for 2 minutes.
for example, is repeated 30 times and then carrying out a reaction at 72°C for 7 minutes;
however, the reaction conditions are not particularly limited thereto. The PCR method and
conditions are described, for example in Ausubel et ah. Short Protocols in Molecular Biology.
third edition, A compendium of Methods from Current Protocols in Molecular Biology (1995).
John Wiley & Sons (particularly Chapter 15).
|0028J
A desired DNA can be isolated by preparing an appropriate probe and primers based on
information of the nucleotide sequence represented by SEQ ID NO: 1. 3. 5. 7, 9. 11, 13 or 15
in the sequence listing of the present application and screening a cDNA library of a human etc.
by the probe and primers. The cDNA library is preferably prepared from the cell, organ or
tissue expressing a protein consisting of the amino acid sequence represented by SEQ ID NO:
2, 4. 6, 8. 10, 12, 14 or 16. Examples of such a cell or tissue include, but are not limited to,
cells or tissues derived from cancers or tumors such as testicles or leukemia, breast cancer.
lymphoma, brain tumor, lung cancer, colon cancer, mastocytoma, melanoma and
neuroblastoma. The aforementioned operations, such as preparation of the probe or primers.
construction of acDNA library, screening of acDNA library and cloning of a gene of interest,
are known to those skilled in the art and can be carried out in accordance with the method
described, for example, in Green, M. R. and Sambrook (described above), Ausbel et al.

{described above). From the DNA thus obtained, DNA encoding CSPG5 protein and a
partial peptide thereof can be obtained.
[0029]
As the host cell, any type of cell may be used as long as it can express the above
polypeptide. Examples of prokaryotic cells include Escherichia coli. and examples of
eukaryotic cells include, but not limited to, yeast cells including budding yeast and fission
yeast, insect cells such as silkworm cells. Xenopus egg cells and mammalian cells such as
monkey kidney cells COSl. Chinese hamster ovary cells CHO, human fetal kidney cell line
HEK293 and mouse fetal skin cell line NIH3T3.
[0030]
When a prokaryotic cell is used as a host cell, an expression vector having a replication
origin in a prokaryotic cell, a promoter, a ribosome binding site, a multi cloning site, a
terminator, a drug resistance gene, an auxotrophic complement gene, and the like is used as the
expression vector. Examples of the expression vector for Escherichia coli may include pUC
system. pBIuescriptll. pET expression system and pGEX expression system. The
polypeptide encoded by DNA can be expressed in the prokaryotic host cell, by integrating
DNA encoding the polypeptide into such an expression vector; transforming a prokaryotic
host cell with the vector; and culturing the resultant transformant. At this time, the
polypeptide can be expressed as a part of a fusion protein with another protein.
[0031]
When a eukaryotic cell is used as a host cell, an expression vector for a eukaryotic cell
having a promoter, a splicing region, a poly (A) additional site, and the like is used as the
expression vector. Examples of such an expression vector may include pKAl, pCDM8,
pSVKl pMSG. pSVL. pBK-CMV. pBK-RSV. HBV vector. pRS. pcDNA3.1 and pYES2.
The polypeptide encoded by DNA can be expressed in a eukaryotic host cell, similarly to the
above, by integrating DNA encoding the above polypeptide into such an expression vector.
transforming the eukaryotic host cell with the vector, and culturing the resultant transformant.
The above polypeptide can be expressed as a part of a fusion protein attached with a tag such
as a His tag (for example, (His)^ to (His)io), a FLAG tag, a myc tag, a HA tag and a GFP.

when pIND/V5-His. pFLAG-CMV-2. pEGFP-Nl. pEGFP-CI or the like is used as the
expression vector.
[0032]
Introduction of an expression vector into a host cell can be carried out by using a
method well known in the art, such as an electroporation method, a calcium phosphate method,
a liposome method, a DEAC dextran method, microinjection, viral infection, lipofection and
binding to a cell membrane penetrating peptide.
[0033]
For isolation/purification of a desired polypeptide from a host cell, separation
operations known in the art can be used in combination. Examples of the separation
operations include, but are not limited to, a treatment with a denaturant such as urea or a
surfactant, sonication. enzymatic digestion, salting out and solvent fractionation precipitation,
dialysis, centrifugation. ultrafiltration, gel tilt rat ion. SDS-PAGE. isoelectric point
electrophoresis, ion exchange chromatography, hydrophobic chromatography, affinity
chromatography and reverse phase chromatography.
[0034]

An antibody is a hetero-multimer glycoprotein usually containing at least two heavy
chains and two light chains. Four types of immunoglobulins except IgM each are a hetero-
tetramer glycoprotein of about 150 kDa primarily constituted of two identical light (L) chains
and two identical heavy (H) chains. Typically, each of the light chains is linked to a heavy-
chain via a single disulfide covalent bond: whereas the number of disulfide bonds between the
heavy chains varies depending on the isotypes of immunoglobulins. Each of the heavy
chains and light chains has also an intra-strand disulfide bond. Each of the heavy chains has
a variable domain (VH region) at an end. followed by several constant regions. Each of the
light chains has a variable domain (VL region) at an end and a single constant region at the
other end. The light-chain variable domain is aligned with the heavy-chain variable domain.
The constant region of a light chain is aligned with the first constant region following the
heavy-chain variable domain. In the variable domain of the antibody, there are three specific

regions called as complementarity determining regions (CDRs). which are variable parts and
based on which the antibody has binding specificity. In the variable region, a portion
relatively conserved is called as a framework region (FR), Complete heavy chain and light
chain variable domains each contain 4 FRs (FRi. FR2, FR3 and FR4 sequentially from the N
terminal side) connected via three CDRs. The three CDRs in a heavy chain are called
CDRI11. CDRH2 and CDR113 sequentially from the N terminal side and the CDRs in a light
chain are called CDRL1, CDRL2 and CDRL3. CDRH3 is the most important for binding
specificity of an antibody to an antigen. The CDRs of each chain are held together in close
proximity writh each other via FR regions and contribute to formation of an antigen binding
site of the antibody in concert with CDRs of the other chain. The constant region does not
directly contribute to binding of the antibody to an antigen; however, the constant region has
various effector functions, such as involvement in antibody-dependent cell-mediated
cytotoxicity (ADCC), phagocytosis via binding to an Fey receptor, ha If-life/clearance rate via
a neonatal Fc receptor (FcRn). and complement-dependent cytotoxicity (CDC) via Clq
component of a complement cascade.
[0035]
Preparation of antibody>
The anti-CSPG5 antibody of the present invention refers to an antibody having
immunological reactivity with a full-length CSPG5 protein or a fragment thereof.
[0036|
The term "immunological reactivity" used herein refers to a property of an antibody
binding to CSPG5 antigen, in-vivo. Through such a binding, an action to damage (for
example, kill, suppress or induce regression of) a tumor is ex cried. More specifically, the
type of an antibody used in the present invention is not limited as long as it can bind to CSPG5
protein to damage a tumor such as breast cancer, lung cancer, brain tumor, leukemia,
malignant lymphoma, adenocarcinoma, mastocytoma, squamous cell carcinoma, melanoma or
neuroblastoma.
[0037]

Examples of the antibody include a monoclonal antibody, a polyclonal antibody, a
genetic recombinant antibody and an antibody fragment (for example. Fab and F(ab')2). as
mentioned above. Also the antibody may be any class of an immunoglobulin molecule such
as IgG, IgE. IgM, IgA, IgD and IgY or any subclass thereof such as IgGl. lgG2, IgG3. IgG4.
IgAl and IgA2.
[0038]
The antibody may be further modified with acetylation. formylation, amidation,
phosphorylation, pegylation(PEG). or the like as well as glycosylation.
[0039]
Preparation examples of various antibodies are described below.
(I) Monoclonal antibody
Examples of the monoclonal antibody include a human monoclonal antibody and an
animal (non-human) monoclonal antibody (for example, a mouse monoclonal antibody, a rat
monoclonal antibody, a rabbit monoclonal antibody and a chicken monoclonal antibody).
[0040]
The antibody, in the case of a monoclonal antibody, can be prepared by carrying out
immunization in accordance with a general immunization method using a desired antigen
(CSPG5 protein herein) or a cell expressing the desired antigen as a sensitizing antigen, fusing
thus obtained immune cell with a parent cell known in the art in accordance with a general cell
fusion method and screening a monoclonal antibody producing cell (hybridoma) by a general
screening method.
[0041]
First, an animal is immunized with a sensitizing antigen in accordance with a method
known in the art. As a general method, a sensitizing antigen is intraperitoneally or
subcutaneously injected to a mammal, for example, a mouse. More spec i tic ally, a sensitizing
antigen, i.e.. CSPG5 protein, is diluted or suspended to an appropriate amount of PBS
(Phosphate-Buffered Saline), physiological saline, or the like, and if desired, a general
adjuvant, for example. Freund's complete adjuvant, is added thereto in an appropriate amount
and emulsified. Thereafter, the emulsion is administered to a mammal, for example, a mouse,

several times at intervals of 4 to 21 days. An appropriate carrier can be used at the time of
immunization with a sensitization antigen. Alternatively, leukemia cell line K562 expressing
CSPG5 gene or the like may be administered to an animal (to be immunized) in order to
immunize the animal.
[0042]
After a mammal is immunized as described above and an increase of the level of the
desired antibody in the serum is confirmed, immune cells are collected from the mammal and
subjected to cell fusion in order to prepare a hybridoma producing a monoclonal antibody.
As the preferable immune cells for preparing a hybridoma, particularly splenocytes are
mentioned.
[0043]
Mammalian myeloma cells are used as another parent cells to be fused with the
immune cells. As the myeloma cells, various cell lines known in the art such as P3U1 (P3-
X63Ag8Ul), P3 (P3*63Ag8.653) (J. Immunol. (1979) 123. 1548-1550). P3x63Ag8U.l
(Current Topics in Microbiology and Immunology (1978) 81, 1-7), NS-1 (Kohler. G. and
Milstein. C. Eur. J. Immunol. (1976) 6. 511-519). MPC-11 (Margulies. D. H. et al., Cell
(1976) 8. 405-415). SP2/0 (Shulman, M. et al., Nature (1978) 276, 269-270), FO (deSt. Groth,
S.F. et at.. J. Immunol. Methods (1980) 35, 1-21). S194 (Trowbridge, 1. S. J. Exp. Med. (1978)
148, 313-323), R210 (Galfre, G. et al., Nature (1979) 277. 131-133) are suitably used.
[0044]
The cell fusion between the immune cells and myeloma cells can be carried out
basically in accordance with a method known in the art. for example, a method of Kohler and
Milstein et al.. (Kohler, G. and Milstein, C. Methods Enzymol. (1981) 73, 3-46).
[0045]
More specifically, the cell fusion is earned out in the presence of, for example, a cell
fusion accelerator, in a general nutrition culture solution. As the fusion accelerator, for
example, polyethylene glycol (PEG) or Sendai virus (HVJ) is used and, if desired, an auxiliary
agent such as dimethyl sulfoxide can be added in order to enhance fusion efficiency.
[0046]

The ratio of the immune cells and myeloma cells to be used can be arbitrarily
determined. For example, the immune cells can be used in a ratio 1 to 10 times compared
with the myeloma cells. As the culture solution to be used in the cell fusion, for example,
RPM1 1640 culture solution or MEM culture solution suitable for proliferation of the myeloma
cell line, and other culture solutions usually used in culturing these cells can be used. In
addition, a serum-supplement such as fetal calf serum (PCS) can be used in combination with
the culture solution.
[0047]
Cell fusion is carried out by sufficiently mixing predetermined amounts of the immune
cells and myeloma cells in the culture solution and adding a PEG solution (for example,
average molecular weight: about 1000 to 6000) previously heated to about 37°C usually in a
concentration of 30 to 60% (w/v) followed by stirring to form desired hybridomas.
Subsequently, an operation consisting of intermittently adding an appropriate culture solution,
centrifuging the mixture and removing the supernatant, is repeated to remove a cell fusion
agent unfavorable for growth of the hybridoma and the like.
[0048]
The hybridoma thus obtained is selected by culturing in a general selection culture
solution such as HAT culture solution (culture solution containing hypoxanthine. aminopterin
and thymidine). The culture in the HAT culture solution is continued for a time period
(usually, several days to several wreeks) sufficient for cells (non-fused cells) other than a
desired hybridoma to die. Subsequently, a general limiting dilution method is carried out and
screening of a hybridoma producing a desired antibody and single cloning are carried out.
[0049]
As well as obtaining the above hybridoma by immunizing an animal except a human
with an antigen, a hybridoma producing a human antibody having a desired activity (for
example, cytostatic activity) can be also obtained by sensitizing human lymphocytes such as
human lymphocytes infected with EB virus, with a protein, protein-expressing cell or lysate
thereof//; vitro, and fusing the sensitized lymphocytes with human-derived myeloma cells, for
example U266 (registration number TIB 196). having a permanent division potential.

[0050]
Thus prepared hybridoma producing a monoclonal antibody can be sub-cultured in a
general culture solution and stored in liquid nitrogen for a long time.
[0051]
{2) Polyclonal antibody
The antibody, in the case of a polyclonal antibody, can be prepared by immunizing a
small animal such as a mouse, a human antibody-producing mouse or a rabbit, with natural
CSPG5 protein, recombinant CSPG5 protein expressed in the form of a fusion protein with
GST and the like in a microorganism such as Escherichia coli or a partial peptide thereof to
obtain the serum; and purifying the antibody, for example, by ammonium sulfate precipitation.
protein A, protein G column, DEAE ion exchange chromatography, an affinity column
coupled with CSPG5 protein and a synthetic peptide. In Examples described later, a mouse
polyclonal antibody against the extracellular region (outside cancer cell) of the amino acid
sequence of CSPG5 protein is prepared and confirmed to have the antitumor effect.
[0052]
As the human antibody-producing mouse used herein, for example. KM mouse (Kirin
Pharma/Medarex) and Xcno mouse (Amgen) are known (for example, International
Publication Nos. WO02/43478 and 02/092812). When such a mouse is immunized with
CSPG5 protein or a fragment thereof, a complete human polyclonal antibody can be obtained
from the blood.
[0053]
An antigen can be prepared, for example, in accordance with a method using an animal
cell (JP Patent Publication (Kohyo) No. 2007-530068A) or a method using baculovirus (for
example. International Publication No. WO 98/46777). When the immunogenicity of an
antigen is low. the antigen may be bound to a macromolecule such as albumin having
immunogenicity and subjected to immunization.
[0054]
(3) Recombinant antibody

The antibody, in the case of a genetic recombination antibody, can be prepared, in
accordance with a genetic recombination technology, by cloning a gene of the antibody from a
hybridoma. incorporating the gene into an appropriate vector, and introducing the vector into a
host to produce a recombinant antibody (see. for example. Carl, A. K. Borrebaeck. James. W.
Larrick, THERAPEUTIC MONOCLONAL ANTIBODIES, Published in the United Kingdom
by MACMILLAN PUBLISHERS LTD, 1990). More specifically, cDNA of a variable
region (V region) of the antibody is synthesized from mRNA of a hybridoma by using a
reverse transcriptase. When DNA encoding a V region of a desired antibody is obtained, it is
linked to DXA encoding a constant region (C region) of the desired antibody, and the resultant
DNA is integrated into an expression vector. Alternatively, DNA encoding the V region of
the antibody may be integrated into an expression vector containing DNA of the C region of
the antibody. Said DNA may be Integrated to be expressed under the control of an
expression regulatory region, for example, an enhancer and a promoter. Subsequently, a host
cell is transformed with the expression vector and allowed to express the genetic recombinant
antibody.
[0055]
Examples of the genetic recombinant antibody include a chimeric antibody, a
humanized antibody, a single-chain antibody and multi-chain antibody such as a bispecific
antibody.
[0056]
The "chimeric antibody" is an antibody formed of sequences derived from different
animals in combination, for example, an antibody constituted of a variable region of a heavy-
chain and a light chain of a mouse antibody and a constant region of a heavy-chain and a light
chain of a human antibody. A chimeric antibody can be prepared in accordance with a
method known in the art. for example, by ligating DNA encoding an antibody V-region and
DNA encoding a human antibody C-region, incorporating the Hgate into an expression vector.
and introducing the expression vector into a host to allow the host to produce the antibody.
As an example, DNA encoding a human/mouse chimeric antibody can be prepared by ligating
a DNA encoding a variable region of a light chain or a heavy chain of an antibody derived

from a non-human animal (for example, mouse) to DNA encoding a constant region of a light
chain or a heavy chain of an antibody derived from a human antibody.
[0057]
The "humanized antibody" is a modified antibody also called as a reshaped human
antibody. The humanized antibody is constructed by grafting an antibody CDR derived from
an immunized animal to a complementarity determining region of a human antibody. A
general genetic recombination technique for preparing the humanized antibody is also known
in the art. Specifically, the humanized antibody is obtained by cloning a DNA encoding a
monoclonal antibody; using the resultant as a template to prepare a DNA encoding a light-
chain variable region and a heavy-chain variable region by a RT-PCR method, or the like;
determining the sequences of the variable regions of the light chain and heavy chain or the
sequences of CDR1, CDR2 and CDR3 based on the Kabat EU numbering system (Kabat et ah,
Sequences of proteins of Immunological Interest, 5thEd. Public Health Service. National
Institute of Health. Bethesda. Md. (1991)); subsequently, synthesizing a DNA sequence.
which is designed such that mouse antibody CDRs and framework regions {framework region;
FR) of the human antibody can be ligated, by a PCR method using several oligonucleotides
prepared designed to have an overlapping portion at the ends; ligating the resultant DNA to
DNA encoding a human antibody constant region and then integrating it into an expression
vector; and introducing the expression vector into a host and then allowing the host to produce
a humanized antibody (see, European Patent Publication No. 239400, International Publication
No. WO96/02576). The FRs of the human antibody to be ligated via CDRs are selected such
that the CDRs (complementarity determining regions) form a satisfactory antigen binding site.
If necessary, the amino acids of the framework region in the variable region of the antibody
may be substituted such that the complementarity determining regions of a reshaped human
antibody form an appropriate antigen binding site (Sato K., et al., Cancer Research. 1993, 53:
851-856). Further, the framework regions may be substituted with those derived from
various human antibodies (International Publication No. W099/51743).
[0058]

The framework regions of a human antibody to be ligated via CDRs are selected such
that the CDRs (complementarity determining regions) form a satisfactory antigen binding site.
If necessary, the amino acids of framework regions in the variable region of an antibody may
be substituted such that the complementarity determining regions of a reshaped human
antibody form an appropriate antigen binding site (Sato K. et al., Cancer Research 1993, 53:
851-856).
[0059]
An amino acid in the variable region (for example, FR) and the constant region may be,
for example, substituted with another amino acid, after a chimeric antibody and a humanized
antibody are formed.
[0060]
The substitution of amino acids includes substitution of amino acids of, for example.
less than 15. less than 10, 8 or less, 7 or less. 6 or less, 5 or less. 4 or less. 3 or less or 2 or less.
preferably 1 to 5 amino acids, and more preferably 1 or 2 amino acids. The antibody having
a substitution should be functionally equivalent to the antibody having no substitution.
Substitution is desirably conservative amino acid substitution, which is a substitution between
amino acids having analogous properties in view of charge, side chain, polarity and
aromaticity, and the like. The amino acids having analogous property can be classified into,
for example, basic amino acids (arginine, lysine, histidine). acidic amino acids (aspartic acid.
glutamic acid), uncharged polar amino acids (glycine, asparagine, glutamine, serine, threonine,
cysteine, tyrosine), nonpolar amino acids (leucine, isoleucine, alanine, valine, proline,
phenylalanine, tryptophan, methionine), branched amino acids (threonine, valine, isoleucine)
and aromatic amino acids (phenylalanine, tyrosine, tryptophan, histidine).
[0061]
Modified antibodies include, for example, antibodies bound to various types of
molecules such as polyethylene glycol (PEG). In the modified antibody of the present
invention, the material to be bound to the antibody is not limited. Such a modified antibody
can be obtained by chemically modifying the obtained antibody. The chemical modification
methods have been already established in this field.

[0062]
The term "functionally equivalent" used herein means that the antibody ofinterest has
the same biological or biochemical activity as the antibody of the present invention, more
specifically means that the antibody of interest has a tumor damaging action, and a rejection
reaction does not basically occur when the antibody is applied to a human. Such activities
include, for example, cytostatic activity or binding activity.
[0063]
A method of introducing a mutation into the polypeptide is known as a method well
known to those skilled in the art for preparing a polypeptide which is functionally equivalent
to a predetermined polypeptide. Those skilled in the art can employ a site-specific
mutagenesis (1lashimoto-Gotoh, T.et al., (1995) Gene, 152,271-275: Zoller, MJ., and Smith.
M. (1983) Methods Enzymol.100, 468-500; Kramer, W.et al„ (1984) Nucleic Acids Res. 12,
9441-9456; Kramer. W. and Fritz. H.I.. (1987) Methods Enzymol.154. 350-367, Kunkel. TA.,
(1985) Proc. Natl. Acad. Sci. USA.. 82, 488-492; Kunkel (1988) Methods Enzymol.. 85. 2763-
2766) to appropriately introduce a mutation into the antibody of the present invention, thereby
preparing an antibody functionally equivalent to the antibody.
[00641
An antibody recognizing an epitope of CSPG5 protein that anti-CSPG5 antibody
recognizes can be obtained byr a method known to those skilled in the art. for example, the
antibody is prepared by determining the epitope of CSPG5 protein recognized by anti-CSPG5
antibody by a general method (for example, epitope mapping) and preparing the antibody
using a polypeptide having an amino acid sequence contained in the epitope as immunogen.
In addition to this method, the antibody can be obtained by a method, for example,
determining the epitopes of the antibodies prepared by a general method and selecting an
antibody having the same epitope as in the anli-CSPG5 antibody.
[0065]
The affinity constant (binding constant) Ka (kon/koff) of the anti-CSPG5 antibody of
the present invention for the CSPG5 protein on a cancer cell surface is at least 10' M" . at least
108 M"1, at least 5 x 108 M"1, at least \Q9 M'1, at least 5 x \09 M"!, at least I010 M"1, at least 5 x

10I(IM'\ at least 10n M"1. at least 5 x 1011 M'1. at least 101:M~' or at least 1013 M"1. As the
binding affinity increases, a stronger antitumor activity can be obtained. Accordingly, if an
anti-CSPG5 antibody having a high binding affinity for CSPG5 protein can be obtained, a
stronger antitumor effect can be expected, the antibody can be applied to a pharmaceutical
composition for treating and/or preventing cancer.
[0066]
The "single-chain antibody" is an antibody obtained by linearly Hgating a heavy-chain
variable region and a light-chain variable region via a linker. DNA encoding a single-chain
antibody can be prepared by ligating DNA encoding a heavy-chain variable region. DNA
encoding a linker and DNA encoding light-chain variable region. The heavy-chain variable
region and light-chain variable region used herein are those preferably derived from a human
antibody or those derived from a human antibody, in which CDRs alone are replaced by those
of an antibody derived from a non-human animal (for example, mouse, rat, chicken). A
linker consists of 12 to 19 amino acids, and includes for example, (G4S)3 consisting of 15
amino acids (G. B. Kim et aL Protein Engineering Design and Selection, 2007. 20 (9): 425-
432).
[0067]
In the case of the "bispecific antibody (diabody)". which is an antibody capable of
specifically binding to two different epitopes. DNA encoding the bispecific antibody can be
prepared by binding, for example. DNA encoding heavy-chain variable region A. DNA
encoding light-chain variable region B, DNA encoding heavy-chain variable region B and
DNA encoding light-chain variable region A sequentially in this order (provided that DNA
encoding light-chain variable region B and DNA encoding heavy-chain variable region B arc
connected via DNA encoding a linker as mentioned above). The heavy-chain variable
regions and light-chain variable regions each are preferably derived from a human antibody or
derived from a human antibody, in which CDRs alone are replaced by those of an antibody
derived from a non-human animal (for example, mouse, rat, chicken).
[0068]

A recombinant antibody can be prepared by integrating the recombinant DNA prepared
as described above into a single or a plurality of appropriate vectors, introducing the vector(s)
into a host cell (for example, mammalian cells, yeast cells, insect cells) and allowing the host
cell to (co-)express the recombinant DNA (P. J. Delves., ANTIBODY PRODUCTION
ESSENTIAL TECHNIQUES.. 1997 WILEY. P. Shepherd and C. Dean.. Monoclonal
Antibodies., 2000 OXFORD UNIVERSITY PRESS; J. W. Goding., Monoclonal Antibodies:
principles and practice., 1993 ACADEMIC PRESS).
[0069]
The antibody as mentioned above preferably has a cytotoxic activity and can produce
an antitumor effect due to the cytotoxic activity.
[00701
The antibody of the present invention can be conjugated with another antitumor agent.
The antibody and the antitumor agent can be conjugated via a spacer having a group reactive
to an amino group, a carboxyl group, a hydroxy group, a thiol group, and the like (examples of
the reactive group include a succinimidyl group, a formyl group, a 2-pyridyldithio group, a
maleimidyl group, an alkoxycarbonyl group and a hydroxy group).
[0071]
Examples of the antitumor agent include the following antitumor agents known to
public by literatures etc.. such as paclitaxel. doxorubicin, daunorubidn. cyclophosphamide.
methotrexate. 5-lluorouracil. thiotepa. busulfan. improsulfan. piposulfan. benzodopa.
carboquone, meturedopa, uredopa, altretamine. triethylenemelamine,
triethylenephosphoramide. triethilenethiophosphoramidc, trimethylolomelamine, bullataein,
bullatacinone, camptothecin. bryostatin, callystatin, cryptophycin 1. cryplophycin 8, dolastatin.
duoeannycin. eleutherobin. pancrati statin, sarcodictyin. spongistatin. chlorambucil.
chioRNAphazine. cholophosphamide. estramustine. ifosfamide. mechlorethamine.
mechlorethamine oxide hydrochloride, melphalan, novembichin, phenestcrine, prednimustine,
trofosfamide, uracilmustard, carmustine, chlorozotocin, fotemustine. lomustine, nimustine,
ranimustine. calicheamicin. dyncmycin, clodronatc. csperamicin. aclacinomycin, actinomycin,
authramycin. azaserinc. bleomycin, cactinomycin. carabicin. carminomycin. carzinophilin.

chromomycin. dactinomycin. detorbicin. 6-diazo-5-oxo-L-norleucine. ADRIAMYCIN.
epirubicin. esolubicin. idarubicin. marcellomycin. mitomycin C, mycophenolic acid.
nogalamycin. olivomycin, pepromycin. potfiromycin. puromycin. quelamycin. rodorubicin.
streptonigrin. strcptozocin. tubercidin, ubenimex, zinostatin. zorubicin. denoplerin, pteropterin,
trimetrexate, fludarabine, 6-mercaptopurine, thiamipurine, thioguaninc. ancitabine, azacitidine,
6-azauridine. carmofur, cCarabine, dideoxyuridine, doxiiluridine, enocitabine. floxuridine;
androgens such as calusterone, dromo stand one propionate, epitbiostanol. mepitiostane.
testolactone. aminoglutethimide. mitotane, trilostane. frolinic acid. aceglatone.
aldophosphamide glycoside. aminolevulinic acid, eniluracil. amsacrine. bestrabucil. bisantrene.
edatraxate. defofamine, dcmecolcine. diaziquone. elfornithine. elliptinium acetate, epothilone,
etoglucid, lentinan. lonidamine, maytansine, ansamitocine, mitoguazone. mitoxantrone,
mopidanmol. nitraerine, pentostatin, phenamet. pirarubiein. losoxantronc. podophyllinic acid.
2-ethylhydrazide. procarbazine, razoxane. rhizoxin. schizophyllan, spirogermanium.
tenuazonic acid, triaziquone. roridine A. anguidine. urethanc. vindesine. dacarbazine.
mannomustine. mitobronitoL mitolactol. pipobroman. gacytosine. doxetaxcl. chlorambucil,
gemcitabine, 6-thioguanine. mercaptopurine, cisplatin. oxaliplatin. carboplatin. vinblastine.
ctoposide, ifosfamide, mitoxantrone, vincristine, vinorelbine. novantrone, teniposide,
edatrexate, daunomycin. aminoptcrin, xeloda, ibandronate. irinotecan, a topoisomerase
inhibitor, difluoromethylomithine (DMFO), retinoic acid, capecitabinc and pharmaceutically
acceptable salts or derivatives thereof.
[0072]
A radioactive isotope known to public by literatures etc.. such as " At, J I, ""I, Y,
186Re; 1S8Re, 153Sm, :i2Bi, 32P, 175Lu and l76Lu, may be bound to the antibody of the present
invention. As the radioactive isotope, preferably, a radioactive isotope is effective for
treating and diagnosing a tumor.
[0073]
The antibody of the present invention is an antibody having immunological reactivity
with CSPG5 protein or an antibody specifically recognizing CSPG5 protein. The antibody
should be an antibody having a structure such that no or little rejection reaction is produced in

the target animal to which the antibody is administered. Examples of such an antibody in the
case where a target animal is. a human etc. include a human antibody, a humanized antibody
and a chimeric antibody (for example, a human-mouse chimeric antibody).
[0074]
A hybridoma capable of producing a human antibody or non-human animal antibody
(for example, mouse antibody) against human CSPG5 protein, is prepared. A monoclonal
antibody produced by the hybridoma is recovered and determined as to whether it is a desired
antibody or not based on immunological binding property to the human CSPG5 protein and
cytotoxic activity as an indicator. In this manner, a desired monoclonal antibody-producing
hybridoma is identified and selected. Thereafter, as described above, DNA encoding the
variable regions of the heavy chain and light chain of a desired antibody is prepared from the
hybridoma and the nucleotide sequence thereof is determined. The information of the
nucleotide sequence of the DNA is used for preparing another antibody.
[0075]
The present invention further provides DNA encoding the antibody of the present
invention described above, DNA encoding the heavy chain or light chain of the antibody
described above, or DNA encoding a variable region of the heavy chain or light chain of the
antibody described above.
[0076]
CDRs encoded by these DNA molecules are regions determining specificity of the
antibody. The nucleotide sequence encoding the region (more specifically, constant region
and framework region) other than CDRs of the antibody may be a nucleotide sequence derived
from another antibody. The "another antibody" used herein may include an antibody derived
from an organism other than a human; however, an antibody derived from a human is
preferable in order to reduce side effects. More specifically, in the above DNA. the regions
encoding individual framework regions of the heavy chain and the light chain and the regions
encoding individual constant regions thereof preferably contain nucleotide sequences encoding
the corresponding amino acid sequences derived from a human antibody.
[0077]

DNA of an antibody serving as an active ingredient of the present invention, can be
obtained, for example, by the above method or the following method. First, total RNA is
prepared from a hybridoma relating to the antibody of the present invention by a commercially
available RNA extraction kit, and then, cDNA is synthesized with a reverse transcriptase by
using random primers etc. Subsequently, cDNA encoding the antibody is amplified by a
PCR method using oligonucleotides of the conserved sequences in the variable regions of the
heavy chain gene and light chain gene of a known mouse antibody, as primers. The constant
region-encoding sequence can be obtained by amplifying a known sequence by a PCR method.
The nucleotide sequence of DNA can be determined by a routine mediod, for example, by
integrating DNA into a sequencing plasmid or a phage etc.
[0078]
It is considered that the antitumor effect of the anti-CSPG5 antibody to be used in the
present invention on CSPG5 protein expressing cancer cells is produced by mechanism of
antibody-dependent cell-mediated cytotoxicity (ADCC) via effector cells and complement
dependent cytotoxicity (CDC).
[0079]
Accordingly, the activity of the anti-CSPG5 antibody used in the present invention can
be evaluated by measuring the ADCC activity or CDC activity against the cancer cells
expressing CSPG5 protein in vitro, as specifically described in Iixamples,
[0080]
The anti-CSPG5 antibody is considered to be useful for treating or preventing cancer.
since the antibody used in the present invention binds to the extracellular region of CSPG5
protein present on the cancer cell surface and exerts an an I i-tumor action based on the
activity(s) mentioned above. More specifically, the present invention provides a
pharmaceutical composition containing an anti-CSPG5 antibody as an active ingredient for
treating and/or preventing cancer. When the anti-CSPG5 antibody is used for administration
to a human body (antibody treatment), the anti-CSPG5 antibody is preferably prepared as a
human antibody or a humanized antibody in order to reduce immunogenicity.
[0081]

The ability of an antibody to bind to CSPG5 protein can be specified, for example, by a
binding assay such as HI.ISA. Western blotting, immunofluorescence and flow cytometric
analysis, as described in Examples.
[0082]

With respect to the antibody recognizing CSPG5 protein, the reactivity thereof with
CSPG5 protein can be checked by using tissues and slices thereof in accordance with
immunohistochemical staining method well known to those skilled in the art. For example, a
tissue obtained from a patient during a surgical operation; or a tissue obtained from an animal
having a heterologous tissue grafted by administering a cells expressing CSPG5 protein
naturally or after transfection; frozen tissue slice fixed with paraformaldehyde or acetone; or
paraffin-embedded tissue slice fixed with paraformaldehyde.
[0083]
An antibody having a reactivity with CSPG5 protein can be stained with various
methods for immunohistochemical staining. For example, visualization can be made by
reacting a horseradish peroxidase-conjugaled goat anti-mouse antibody and a horseradish
peroxidase-conjugated goat anti-rabbit antibody.
[0084]
pharmaceutical composition>
A target of the pharmaceutical composition for treating and/or preventing cancer
according to the present invention is not particularly limited as long as it is cancer (cells)
expressing CSPG5 protein on the cell surface.
[0085]
The terms "tumor" and "cancer" used herein refer to malignant neoplasms and arc
interchangeably used.
[0086]
In the present in\ cniion. a target cancer is a cancer expressing CSPG5 gene, preferably,
in particular, cancers expressing genes encoding amino acid sequences represented by SEQ ID

NOs: 2, 4, 6, 8, 10. 12. 14 and 16 or polypeptides eontaining partial sequences of the amino
acid sequences consisting of at least 7 consecutive amino acids, more preferably, cancers
except ovarian cancer, further preferably, breast cancer, lung cancer, brain tumor, leukemia,
malignant lymphoma, mastocytoma, melanoma or neuroblastoma, more preferably, leukemia
or malignant lymphoma.
|0087]
Examples of these specific cancers include, but are not limited to, breast
adenocarcinoma, complex breast adenocarcinoma, mammary gland malignant mixed tumor,
ductal papillary adenocarcinoma, lung adenocarcinoma, squamous cell carcinoma, small cell
cancer, large cell cancer, glioma which is a neuroepithelial tissue tumor, ependymoma.
neurocytoma, fetus neuroectodermal tumor, neurinoma, neurofibroma, meningioma, chronic
lymphocytic leukemia, llodgkin's lymphoma, gastrointestinal-tract lymphoma, gastrointestinal
lymphoma, small to medium cell lymphoma, cecal cancer, ascending colon cancer, descending
colon cancer, transverse colon cancer, sigmoid colon cancer and rectal cancer.
[0088]
An animal of the interest for the pharmaceutical composition of the present invention is
a mammal; for example, primates, pet animals, domestic animals and animals for competitive
use. particulary preferably, humans, dogs and cats.
[0089]
When the antibody according to the present invention is used as a pharmaceutical
composition, the antibody can be formulated by a method known to those skilled in the art.
For example, it can be used parenterally in the form of a sterile solution or suspension for
injection when the antibody is mixed with water or a pharmaceutically acceptable liquid. It is
also contemplated that the antibody is appropriately mixed with a pharmaceutically acceptable
carrier or medium: for example, sterile water, physiological saline, a vegetable oil. an
emulsifying agent, a suspending agent, a surfactant, a stabilizer, a flavoring agent, an excipient.
a vehicle, an antiseptic agent and/or a binder, in a unit dose required for generally accepted
pharmaceuticals to prepare medicinal agents. The amount of active ingredient in these

medicinal agents is specified such that the dose within a predetermined range can be
appropriately obtained.
[0090]
The sterile composition for injection can be formulated by using a vehicle such as
distilled water for injection in accordance with a routine manner for preparation of medicinal
agents.
[0091]
Examples of an aqueous solution for injection include physiological saline and isotonic
solutions containing glucose and other adjuvant(s); for example, D-sorbitol, D-mannose, D-
mannitol and sodium chloride, and it can be used in combination with an appropriate
solubilizing agent such as an alcohol, for example, elhanol. a polyalcohol such as propylene
glycol and polyethylene glycol, and a nonionic surfactant such as polysorbateSO'™1 and HCO-
60.
[0092]
Examples of an oily solution include sesame oil and soybean oil. A solubilizing agent
such as benzyl benzoate and benzyl alcohol may be used in combination. Furthermore, a
buffer such as a phosphate buffer and a sodium acetate buffer, a soothing agent such as
procaine hydrochloride, a stabilizer such as benzyl alcohol and phenol, and/or an antioxidant
may be blended in combination. The injection solution prepared is usually stored in
appropriate ampoules.
[0093]
Examples of administration method include oral administration or parenteral
administration, preferably parenteral administration, in particular, injection, nasal
administration, transpulmonary administration and transdermal administration. Examples of
the injection include intravenous injection, intramuscular injection, intraperitoneal injection
and subcutaneous injection for systemic administration or local administration.
[0094]
The administration method can be appropriately selected depending upon the age,
weight, sex and symptom of the patient. As a dose of a pharmaceutical composition

containing an antibody or a polynucleotide encoding an antibody may be selected from the
range of e.g.. 0.0001 mg to 1000 mg per time per body-weight of" 1 kg, or from the range of
e.g., 0.001 to 100000 mg/body per patient. However, the dose is not limited by these
numerical values. The dose and administration method may vary depending on the body
weight, age, sex. symptom of the patient, and the like and can be appropriately selected by
those skilled in the art.
Examples
[0095]
The present invention is more specifically described below by way of Examples;
however, the scope of the present invention is not limited by these examples.

(1) Preparation of cDNA library
Total RNA was extracted from the testicular tissue of a healthy dog in accordance with
the Acid guanidium-Phenol-Chloroform method, and then, poly A RNA was purified by using
Oligotex-dT30 mRNA purification Kit (manufactured by Takara Shuzo Co.. Ltd.) in
accordance with the protocol attached to the kit.
[0096]
Dog testis cDNA phage library was synthesized using the mRNA (5 ug) obtained
above. The cDNA phage library was prepared by using cDNA Synthesis Kit. ZAP-cDNA
Synthesis Kit or ZAP-cDNA Gigapacklll Gold Clonig Kit (manufactured by Agilent
Technologies) in accordance with the protocol attached to the kit. The size of the cDNA
phage library prepared was 7.73 x 10" pfii/mL.
[0097]
(2) Screening of cDNA library with serum
Immuno-screening was carried out using the dog testis cDNA phage library prepared
above. More specifically, host Escherichia coll (XL 1-Blue MRF') was infected with phages
so as to obtain 2210 clones in a NZY agarose plate of § 90 x 15 mm and culture was carried
out at 42°C for 3 to 4 hours to fonn plaques. The plate was covered with nitrocellulose

membrane (Hybond C Extra: manufactured by GE Healthcare Bio-Scinece) impregnated with
IPTG (isopropyl-fi-D-thiogalactoside) at 37°C for 4 hours to induce protein expression and the
protein was transferred to the membrane. Thereafter, the membrane was taken, soaked in
TBS (10 mM Tris-HCl, 150 mM NaCl pH7.5) containing 0.5% skim milk powder and shaken
at 4°C overnight lo suppress a nonspecific reaction. This filter was allowed lo react with the
500-fold diluted serum of a disease dog at room temperature for 2 to 3 hours,
[0098]
As the disease-dog serum mentioned above, the serum taken from a dog with breast
cancer was used. The serum was stored at -80°C and pretreated right before use. The
pretreatment is carried out in accordance with the following method. First, host Escherichia
coli (XLl-BIure MRF) was infected with X ZAP Express phages having no inserted foreign
gene and cultured on a NZY plate medium at 37°C overnight. Then, a buffer (0.2 M
NaHCOs pi 18.3) containing 0.5 M NaCl was added to the plate, and allowed to stand still at
4°C for 15 hours. Thereafter, the supernatant was recovered as an Escherichia coft/phage
extraction liquid. Subsequently, the Escherichia co///phage extraction liquid recovered was
passed through a NHS-column (manufactured by GE Healthcare Bio-Science) to allow
proteins derived from Escherichia coli/phage lo immobilize to the column. The disease dog
serum was passed through the protein-immobilized column to react to remove the antibody
adsorbed to Escherichia coli and phage (protein) from the serum. The serum fraction passed
though the column was diluted 500 fold with TBS containing 0.5% skim milk powder and
used as a sample for immuno-screening.
[0099]
The serum thus treated and the above fusion protein were blotted on a membrane and
the membrane was washed four times with TBS-T (0.05% Tween (registered trade mark)
20/TBS), then allowed to react with goat anti-dog IgG (Goat anti Dog IgG-h + I HRP
conjugated; manufactured by BETHYL Laboratories) as a secondary antibody, which was
diluted 5000-fold with TBS containing 0.5% skim milk powder, at room temperature for one
hour. Detection was made by an enzymatic chromogenic reaction using NBT/BC1P reaction
solution (manufactured by Roche). Colonies corresponding to chromogenic reaction

positive-sites were picked up from the NZY agarose plate of <|)90 x 15 mm and dissolved in
500 uL of SM buffer solution (100 mM NaCl. 10 mM MgClS04. 50 mM Tris-HCl. 0.01%
gelatin. pH7.5). The secondary and tertian' screening were carried out by repeating the
above method until the chromogenic reaction positive colonies were unified. Through
screening of 9110 phage clones reacting with IgG in the serum, a single positive clone was
isolated.
[0100]
(3) Homology search of isolated antigen gene
In order to subject the single positive clone isolated by the above method to nucleotide
sequence analysis, an operation for transferring from a phage vector to a plasmid vector was
carried out. More specifically, a solution (200 pL) containing host Escherichia coli (XL1-
Blue MRF) prepared so as to show an absorbance at ODWHI of 1.0. a purified phage solution
(100 pL) and further 1 ^L of ExAssist helper phage (manufactured by Agilent Technologies)
were mixed and allowed to react at 37CC for 15 minutes. 3 mL of LB medium was added
and culture was carried out at 37°C for 2.5 to 3 hours. Immediately after cultivation, the
medium was kept warm in a water bath of 70°C for 20 minutes, and centrifuged at 4°C and
1000 x g. for 15 minutes. The supernatant was recovered as a phargemid solution.
Subsequently, a solution (200 pL) containing a phargemid host Escherichia coli (S01..R) was
prepared so as to have an absorbance at OD60o of 1.0 and a purified phage solution (10 pL)
were mixed and reacted at 37°C for 15 minutes. The resultant solution (50 pL) was seeded
on an ampicillin (final concentration: 50 pg/mL)-containing LB agar medium and cultured at
37°C overnight. A single transformed SOI.R colony was picked up. cultured in ampicillin
(final concentration: 50 pg/mL)-containing LB medium at 37°C and thereafter, purified by
QIAGEN plasmid Miniprep Kit (manufactured by QIAGEN) to obtain a plasmid DNA having
a desired insert.
[0101]
The purified plasmid was subjected to primer walking using T3 primer represented by
SEQ ID NO: 17 and 17 primer represented by SEQ ID NO: 18 to analyze the full-length
sequence of the insert. The gene sequence represented by SHQ ID NO: 1 was obtained by

the sequencing. Using the nucleotide sequence and amino acid sequence of the gene,
sequence identity search (search for identical sequence with known genes) was carried out by
a sequence identity search program. BLAST search (http://www.ncbi.nlm,nih.iZov/BLAST/),
As a result, it was found that the gene obtained above is CSPG5 gene. In the human CSPG5
gene, which is a human homologous factor with a canine CSPG5 gene, a nucleotide-sequence
identity was 87%. and in human CSPG5 protein, an amino acid sequence identity was 87%.
In cat CSPG5 gene, a nucleotide sequence identity was 92%. In cat CSPG5 protein, an
amino acid sequence identity was 91%. In mouse homologous factor, i.e., mouse CSPG5
gene, a nucleotide sequence identity was 84%. In mouse CSPG5 protein, an amino acid
sequence identity was 85%. The nucleotide sequences of the human CSPG5 gene are
represented by SEQ ID NOs: 3, 5. 7, 9 and 11. The amino acid sequences of the human
CSPG5 protein are represented by SEQ ID NO: 4, 6, 8, 10 and 12. The nucleotide sequence
of the cat CSPG5 gene is represented by SEQ ID NO: 13. The amino acid sequence of the
cat CSPG5 protein is represented by SEQ ID NO: 14. The nucleotide sequence of the mouse
CSPG5 gene is represented by SEQ ID NO: 15. The amino acid sequence of the mouse
CSPG5 protein is represented by SEQ ID NO: 16.
[0102]
(4) Gene expression analysis in tissues
Expression of the gene obtained by the above method in normal tissues and cancer
tissues of a dog. human and mouse and a cancer cell lines was checked by a RT-PCR (Reverse
Transcription-PCR) method. The reverse transcription reaction was carried out as follows.
First, total RNA was extracted from individual tissues (50 to 100 mg) and individual cell lines
(5 to 10 x 10(1 cells) by use of TRIZOL reagent (manufactured by Thermo Fisher Scientific) in
accordance with the attached protocol. Using the total RNA, cDNA was synthesized by
using Superscript First-Strand Synthesis System for RT-PCR (manufactured by Thermo Fislier
Scientific) in accordance with the protocol attached. As the cDNA of the human normal
tissues (brain, hippocampus, testicles, colon, placenta), gene pool cDNA (manufactured by
Thermo Fisher Scientific). QUICK-Clone cDNA (manufactured by Clontech Laboratories,
Inc.) and Large-Insert cDNA Library (manufactured by Clontech Laboratories, Inc.) were used.

The PCR reaction was carried out by using the obtained gene specific primers (canine primers
are represented by SEQ ID NOs: 19 and 20, human primers are represented by SEQ ID NOs:
21 and 22, mouse primers are represented by SEQ ID NOs: 23 and 24) as follows. First,
reagents and the attached buffer were added to 0.25 pL of the sample prepared by the reverse
transcription reaction to obtain a mixture having a total amount of 25 uL containing the above
primers (2 pM for each), dNTPs (0.2 mM for each) and a 0.65 U ExTaq polymerase
(manufactured by Takara Shuzo Co., Ltd.). The reaction mixture was subjected to a Thermal
Cycler (manufactured by BIO RAD) in which a cycle consisting of a reaction at 94°C for 30
seconds, a reaction at 55°C for 30 seconds and a reaction at 72°C for one minute, was repeated
30 times. For comparison, a GAI'DH-specific primer (canine and human GAPDH primers
are represented by SEQ ID NOs: 25 and 26. mouse GAPDH primers are represented by SEQ
ID NOs: 27 and 28) were simultaneously used. As a result, as shown in Figure 1, the canine
CSPG5 gene was not expressed in almost all normal canine tissues, but strongly expressed in
the canine tumor tissue. Similarly to the canine CSPG5 gene, expression of human and
mouse CSPG5 genes in normal human and mouse tissues was rarely confirmed: however.
expression thereof was detected in cancer cells, for example, breast cancer, lung cancer, brain
tumor, ovarian cancer, leukemia, malignant lymphoma cells (Figures 2 and 3).
[0103]

(1) Cloning of full-length cDNA encoding human CSPG5 and cDNA encoding the
extracellular region of human CSPG5
A full-length cDNA encoding human CSPG5 gene was obtained by cloning in
accordance with the following method based on the gene represented by SEQ ID NO: 7
obtained Example 1. PCR was carried out as follows: Reagents and the attached buffer were
mixed to obtain a total amount of 50 pL of mixture containing the cDNA molecule (1 pL),
which was one of those taken from various tissues and cells (prepared in Example 1) and
whose expression was confirmed by the RT-PCR method, two types of primers (0.4 pM for
each) having Kpnl and EcoRI restriction enzyme cleavage sequences (represented by SEQ ID
NOs: 29 and 30), 0.2 mM dNTPs, and 1.25 U PrimeSTAR HS polymerase (manufactured by

Takara Shu/o Co.. I Ad.).: and the resultant was subjected In ;i Thermal C\clcr (manufactured
by BIO RAD), in which a cycle (PCR) consisting of a reaction at 98°C for 10 seconds and a
reaction nt 68'C lor 2.5 minutes was repeated 30 times. Incidentally, the above two types of
primers were used for amplifying a region encoding a full length amino acid sequence
represented h> SEQ ID NO: 7. After the PCR. the amplified DNA was eleetrophoresed on a
1% agarose gel and a DNA fragment of about 1.7 khp was purified by use of QIAquiek Ciel
Extraction Kit {manufactured by QIAGEN). The amplified product obtained by the above
PCR reaction was inserted in pcDNA3.1 (Thermo Fisher Scientific) (hereinafter referred to as
CSPG5/pcDNA3.1) and confirmed to be a cDNA sequence encoding human CSPG5 gene by
sequencing using a DNA sequencer. The sequence represented by SEQ ID NO: 7 represents
the nucleotide sequence of human CSPG5 gene and the sequence represented by SEQ ID NO:
8 represents the amino acid sequence of human CSPG5 protein.
[0104]
A PCR reaction was carried out based on SEQ ID NO: 7 as follows. Reagents and the
attached buffer were mixed to obtain a total amount of 50 p.L of mixture containing two types
of primers (0.4 pM for each)(represented by SEQ ID NOs: 29 and 30) containing Kpnl and
EcoRI restriction enzyme cleavage sequences. 0.2 mM dNTPs and 1.25 U PrimeSTAR HS
polymerase (manufactured by 1 akara Shu/o Co.. I.id.), and the resultant was subjected to a
Thermal Cycler (manufactured by BIO RAD) in which the cycle consisting of a reaction at
98°C for 10 seconds and a reaction at 68°C for 2.5 minutes was repeated 30 times.
Incidentally, the above two types of primers were used for amplifying a region encoding the
amino acid sequence of the extracellular region of CSPG5 protein represented by SEQ ID NO:
7. After the PCR. the amplified DNA was eleetrophoresed on a 1% agarose gel and a DNA
fragment of about 1.3 kbp was purified by use of QIAquiek Gel Extraction Kit (manufactured
by QIAGEN). The amplified product obtained by the above PCR reaction was ligated to
pcDNA3.1 to which cDNA encoding a mouse IgG2a Ec protein is inserted to obtain an
expression vector (hereinafter referred to as pcDNA-hCSPG5 ECD-IgG2al;e) encoding a
human CSPG5 extracellular region/mouse IgG2a Fc fusion protein (hereinafter referred to as
hCSPGS ECD-mIgG2aFc) and confirmed to be a cDNA sequence encoding hCSPG5 ECD-

mIgG2aFc by sequencing using a DNA sequencer. The sequence represented by SEQ ID
NO: 32 represents the nucleotide sequence encoding hCSPGS ECD-mIgG2aFc and the
sequence represented by SFQ ID NO: 33 represents the amino acid sequence of hCSPGS
ECD-mIgG2aFc.
[0105J
(2) Preparation of hCSPGS ECD-mIgG2aFc
As an immunizing antigen for preparing an antibody against CSPG5 protein, hCSPGS
ECD-mIgG2aFc was prepared.
[0106]
An expression vector. pcDNA-hCSPG5 ECD-mIgG2aFc was introduced into a human
fetal kidney cell line. HEK293 cell, by a lipofection method. hCSPGS ECD-mIgG2aFc was
purified from the culture supernatant 7 days after introduction. The culture supernatant was
applied to a Hi Trap protcinG HP (GE Healthcare Bioscienee) column, washed with a binding
buffer (20 mM sodium phosphate (pH 7.0)). and eluted with an elulion buffer (0.1 M glycine-
HC1 (pH 2.7)). The eluted liquid was placed in a tube containing a neutralization buffer (1 M
Tris-MCl {pH 9.0)) and immediately neutralized. Then, the eluted liquid obtained by the
above method wras subjected to ultrafiltration using NANOSEP 1 OK OMEGA (manufactured
by PALL) and replacement with a physiological phosphate buffer solution (manufactured by
NISSUI PHARMACEUTICAL CO.. LTD.). and then aseptically filtered by HI' Tuffryn
Acrodisc of 0.22 u.m (manufactured by PALL) and used in the following experiments.
[0107]

(1) Preparation of polyclonal antibody against CSPG5
To obtain an antibody binding to the extracellular region of CSPG5. 0.1 mg of hCSPGS
ECD-mlgG2aFc prepared as described above as an antigen and an equivalent amount of
complete Freund's adjuvant (CFA) solution were mixed and the resultant mixture was
subcutaiieously administered to a mouse 4 times every two week. Thereafter, blood was
taken to obtain an anti-scrum containing a polyclonal antibody. The anti-serum was purified
by Protein G carrier (manufactured by GE Healthcare Bioscienee) to obtain a polyclonal

antibody against hCSPG5 ECD-m]gG2aFc. The serum of a mouse to which the antigen was
not administered was purified by use of Protein G carrier in the same manner as above and
used as a control antibody.
[0108]
(2) Establishment of cells constantly expressing full-length human CSPG5
CSPG5/pcDNA3.1 prepared as described above was introduced into CHO-K1 cells
(ATCC) by a lipofection method. Screening was carried out by a 500 ug/mL G418 (Nacalai)
to establish a CHO cell line constantly expressing full-length human CSPG5 (CHO-CSPG5),
An expression vector having no cDNA encoding CSPG5 inserted therein (hereinafter referred
to as emp/pcDNA3.t) was introduced and screened in the same manner as above to obtain
cells to be used as control cells (hereinafter referred to as CHO-emp).
[0109]
Similarly. CSPG5/pcDNA3.1 was introduced to murine leukemia cell line EL4
(ATCC) by a lipofection method. Screening was carried out by a 500 pg/mL G41 8 (Nacalai)
to establish EL4 cell line constantly expressing full-length human CSPG5 gene (EL4-CSPG5).
An expression vector having no cDNA encoding CSPG5 gene inserted therein (hereinafter
referred to as emp/pcDNA3.1) was introduced and screened in the same manner as above to
obtain cells to be used as control cells (hereinafter referred to as EL4-emp).
[0110]
(3) Analysis of expression of antigen protein on cell surface
It was examined whether the polyclonal antibody prepared in step (1) specifically
reacts with CSPG5 protein expressed on the surface of the cell established in step (2). CIIO-
CSPG5 cells and CHO-emp cells (106 cells for each) were separately placed in 1.5 mL-volume
micro-centrifuge tubes and centrifuged. To each of the tubes, the polyclonal antibody (2 pg
(5 pL)) against CSPG5 protein prepared in the above step (1) was added. The mixture was
further suspended with PBS (95 pL) containing a 0.1% fetal bovine serum and allowed to
stand still on ice for one hour. After washing with PBS. the mixture was suspended with 5
pL of an FITC-labeled goat anti-mouse IgG antibody (manufactured by Santacruz) and 95 pL
of PBS containing a 0.1% fetal bovine serum (FBSJ and allowed to stand still on ice for one

hour. After washing with PBS, fluorescence intensity was measured by FACS Calibur
(manufactured by BD). On the other hand, the control antibody prepared in the above step
(1) was subjected to the same operation as above in place of the polyclonal antibody against
CSPG5 protein and used as a control. As a result, the CHO-CSPG5 cells to which the anti-
human CSPG5 antibody was added exhibited increase of fluorescence intensity of about 221%
compared with the control. The same operation was applied to CHO-emp cells. As a result,
the CHO-emp cell to which anti-human CSPG5 antibody was added exhibited the same
fluorescence intensity as the control. It was demonstrated from these results that the anti-
human CSPG5 antibody specifically binds to CSPG5 protein expressed on the surface of a cell
membrane.
[0111]
The increase rate of the fluorescence intensity was expressed by an increase rate of
mean fluorescence intensity (MFI value) of each cell and calculated in accordance with the
following formula:
Increase rate of mean fluorescence intensity {fluorescence intensity increase rate) (%) =
((MFI value of cells to which an anti-human CSPG5 antibody was reacted)-(control MFI
value)) ■*■ (control MFI value) x 100
[0112]
Next, it was examined whether or not CSPG5 protein is expressed on the cell surface
with respect to two types of leukemia cell lines (K562, THP-1) and two types of malignant
lymphoma cell lines (L-1236. P3HR-1) on which CSPG5 gene was confirmed to be highly
expressed. Individual human cell lines (106 cells) on which gene expression were confirmed
in the above were separately placed in 1.5 mL-volume micro-centrifuge tubes and centrifuged.
To each of the tubes, the polyclonal antibody (2 ug (5 uL)) against CSPG5 protein prepared in
the above step (1) was added. The mixture was further suspended with PBS (95 uL)
containing a 0.1% fetal bovine serum and allowed to stand still on ice for one hour. After
washing with PBS. the mixture was suspended with 5 f.iL of an FITC-labeled goat anti-mouse
IgG antibody (manufactured by Santacruz) and 95 uL of PBS containing a 0.1% fetal bovine
serum (FBS) and allowed to stand still on ice for one hour. After washing with PBS,

fluorescence intensity was measured by FACS Calibur (manufactured by BD). On the other
hand, the control antibody prepared in the above step (1) was subjected to the same operation
as above in place of the polyclonal anlibodx against CSPG5 protein ;nui used LIS a control.
As a result, the cell to which the anti-human CSPG5 antibody was added exhibited increase of
fluorescence intensity of 30% or more compared with the control. More specifically, K562
exhibited 184% increase of fluorescence intensity, THP-1 51% increase, L-1236 115%
increase, and P3HR-1 82% increase. It was confirmed from these results that CSPG5 protein
are expressed on the surface of cell membrane of the human cancer cell lines.
[0113]

Next, it was examined whether a polyclonal antibody against CSPG5 protein can
damage tumor cells expressing CSPG5 protein. Evaluation was made by using the
polyclonal antibody against human CSPG5 prepared in Example 3. Human leukemia cell
line K562 and malignant lymphoma ceil line L-1236 (106 cells for each) on which expression
of CSPG5 protein was confirmed, wrere separately collected in a centrifuge tube of 50 mL in
volume. To the tube, 100 [aCi chromium 51 was added and the tube was incubated at 37°C
for two hours. Thereafter, the cells were washed three times with RPMI 1640 medium
containing a 10% fetal bovine serum and added to wells of a 96-wrell (with a V-shape bottom)
plate in a ratio of 10' cells per well. To this, the above polyclonal antibody against human
CSPG5 protein was added in a ratio of 1 u.g per well and further lymphocytes separated from
the peripheral blood of a rabbit was added in a ratio of 2 x 10~ cells per well. The plate was
cultured at 37°C in a 5% C02 condition for 4 hours. After culturing, the amount of
chromium (Cr) 51 released from damaged tumor cells in the culture supernatant was measured
and the ADCC activity of a polyclonal antibody against human CSPG5 protein on each of the
cancer cells was calculated. As a result, it was confirmed that the ADCC activities on K562
and L-1236 cells are 23.2% and 18.7%. respectively (see. Figure 4). On the other hand, the
activity was not virtually confirmed (see. Figure 4) when the same operation was conducted by
using the control antibody (Example 3) prepared from the peripheral blood of a mouse not

immunized with the antigen and using a sample to which no antibody was added.
Accordingly, it was clearly demonstrated that the tumor cells expressing CSPG5 protein can
be damaged by an antibody against CSPG5 protein based on the ADCC activity.
[0114]
The cytotoxic activity was obtained by mixing the antibody against CSPG5 protein
used in the present invention, rabbit lymphocytes and the cell lines (10J cells) into which
chromium 51 was incorporated, culturing the mixture for 4 hours, measuring the amount of
chromium 51 released in the medium after culture, and estimating the cytotoxic activity on
each of leukemia cell lines using the following formula*.
♦Formula: Cytotoxic activity (%) = (the amount of chromium 51 released from K562
and L-1236 when an antibody against CSPG5 protein and rabbit lymphocytes were added) +
(the amount of chromium 51 released from target cells to which IN hydrochloric acid was
added)x 100.
[0115]

The antigen protein (hCSPG5 ECD-mlgG2aFc) (100 ug) represented by SEQ ID NO:
33 and prepared in Example 2 was mixed with the equivalent amount of MPL + TDM
adjuvant (manufactured by Sigma). The mixture was used as an antigen solution per mouse.
The antigen solution was intraperitoneally administered to 6-week old Balb'c mice
(manufactured by Japan SEC. Inc.) and further administered 4 times every week to complete
immunization. The spleens were excised out three days after the last immunization and
ground by sandwiching each of the spleens between two sterilized slide glasses, washed with
PBS {-) (manufactured by Nissui) and centriiuged at 1500 rpm for 10 minutes and then the
supernatant was removed. This operation was repeated three times to obtain spleen cells.
The obtained spleen cells and mouse myeloma cells SP2/0 (purchased from ATCC) were
mixed in a ratio of 5: 1. To the mixture, a PEG solution prepared by mixing RPM1 1640
medium (200 uL) containing 10% FBS and heated to 37°C and 800 uL of PEG1500
(manufactured by Boehringer) heated to 37°C was added. The mixture was allowed to stand
still for 5 minutes to perform cell fusion. The mixture was centriiuged at 1700 rpm for 5

minutes. After the supernatant was removed, the cells were suspended with 150 mL of RPMI
1640 medium (HAT selection medium) containing 15% FBS and a HAT solution
manufactured by Gibco in an equivalent of 2%, and seeded in fifteen 96-well plates
(manufactured by NUNC) in an amount of 100 uL per well. The cells were cultured for 7
days at 37°C in a 5% C02 condition to obtain hybridomas. i.e.. fusion cells of spleen cells and
myeloma cells.
[0116]
A hybridoma was screened based on the binding affinity of the antibody produced by
the hybridoma prepared for hCSPG5 ECD-mIgG2aFc. A 1 ug/mL solution of hCSPG5
FXD-mIgG2aFc protein prepared in Example 2 was added to a 96-well plate in an amount of
100 uL per well and allowed to stand still at 4°C for 18 hours. After individual wells were
washed three times with PBS-T. a 0.5% Bovine Serum Albumin (BSA) solution
(manufactured by Sigma) was added in an amount of 400 uL per well and allowed to stand
still at room temperature for 3 hours. The solution was removed and wells were washed
three times with 400 \±L of PBS-T per well. Then each culture supernatant of the hybridomas
obtained above was added in an amount of 100 uL per well and allowed to stand still at room
temperature for 2 hours. After individual wells were washed three times with PBS-T. HRP-
labeled anti-mouse IgG (H - L) antibody (manufactured by Invitrogen) diluted 5000 fold with
PBS was added in an amount of 100 uL per well and allowed to stand still at room
temperature for one hour. After the wells were washed three times with PBS-T. a TMB
substrate solution (manufactured by Thermo) was added in an amount of 100 uL per well and
allowed to stand still for 15 to 30 minutes to perform a chromogenic reaction. After the color
was generated. 1 N sulfuric acid was added in an amount of 100 pL per well to terminate the
reaction. The absorbance values at 450 nm and 595 nm were measured by an absorption
spectrometer. As a result, hybridomas producing antibodies exhibiting high absorbance
values were screened.
[0117|
The screened hybridomas were added to a 96-well plate in a ratio of 0.5 cell per well
and cultured. After one week, a hybridoma forming a single colony in wells were observed.

The cells in these wells were further cultured. A hybridoma was screened based on the
binding affinity of the antibody produced by the hybridoma cloned for CSPG5 protein. A 1
jig/mL solution of hCSPG5 ECD-mIgG2aFc protein prepared in Example 2 was added to a
96-well plate in an amount of 100 pL per well and allowed to stand still at 4°C for 18 hours.
After individual wells were washed three times with PBS-T. a 0.5% BSA solution was added
in an amount of 400 pT per well and allowed to stand still at room temperature for 3 hours.
The solution was removed and the wells were washed three times with PBS-T in an amount of
400 pL per well. Each of the culture supernatants of hybridomas obtained above was added
in an amount of 100 uL per well and allowed to stand still at room temperature for 2 hours.
After individual wells were washed three times with PBS-T, HRP-labeled anti-mouse IgG (H
+ L) antibody (manufactured by Thermo Fisher Scientific) diluted 5000 fold with PBS was
added in an amount of 100 pL per well and allowed to stand still at room temperature for one
hour. After the wells were w:ashed three limes with PBS-T. a TMB substrate solution
(manufactured by Thermo) was added in an amount of 100 pL per well and allowed to stand
still for 15 to 30 minutes to perform a chromogenic reaction. After color was generated. 1 N
sulfuric acid was added in an amount of 100 pi. per well to terminate the reaction. The
absorbance values at 450 nm and 595 ran were measured by an absorption spectrometer. As
a result. 312 hybridoma cell lines producing monoclonal antibodies reactive to CSPG5 protein
were obtained.
[0118]
Subsequently, monoclonal antibodies reactive to the surface of a cell expressing
CSPG5 protein were screened from the monoclonal antibodies. Specifically. 10^ cells (CHO-
CSPG5) expressing CSPG5 protein and established in Example 2 were placed in a 1.5 mL-
volume micro-centrifuge tube and centrifuged. To this, each of the hybridoma culture
supernatants obtained above (100 pL) was added and allowed to stand still on ice for one hour.
After washing with PBS, FITC-labeled goat anti-mouse IgG antibody (manufactured by
Thermo Fisher Scientific) diluted 500 fold with I'BS containing 0.1% FBS was added and the
mixture was allowed to stand still on ice for one hour. After washing with PBS. fluorescence
intensity was measured by a FACS Calibur (manufactured by BD). CHO cells (CHO-emp)

expressing no CSPG5 protein were subjected to the same operation as above and used as a
control. As a result, monoclonal antibodies whose fluorescence intensities are higher than
the control, in other words, 18 monoclonal antibodies (#1 to #18)) reacting with the surface of
the cell expressing CSPG5 protein, were screened.
[0119]

(1) Antitumor effect (ADCC activity) of monoclonal antibody against CSPG5 protein on
cancer cells
The cytotoxic activities (ADCC activity) of monoclonal antibody #1 against CSPG5
protein screened above to cancer cells were evaluated. The hybridomas producing a
monoclonal antibody were cultured by using hybridoma SFM (manufactured by Thermo
Fisher Scientific) medium. The resultant supernatant was purified by use of H Strap proteinA
SepharoseFF (manufactured by GE Healthcare), replaced with PBS (-), and filtered by a 0.22
u.m filter (manufactured by Millipore). and the resultant product was used as an antibody for
measuring activity. Human leukemia cell line K.562 and malignant lymphoma cell line L-
1236 (10ft cells for each) were collected separately in 50 mL-volume centrifuge tubes and 100
pCi of chromium 51 was added and incubated at 37°C for 2 hours. Thereafter, the cells were
washed three times with RPMI 1640 medium containing a 10%FBS, and added to a 96 well
(with a V-shape bottom) plate in a ratio of 10 cells per well and used as target cells. To this,
the above purified antibody was added in an amount of 1 ug per cell, and mouse lymphocytes
(2 x 10^ cells) separated from a mouse spleen were added and cultured at 37°C in a SVoCO?
condition for 4 hours. After culture, the amount of chromium 51 released from the damaged
tumor cells in the culture supernatant was measured and the ADCC activity of anti-CSPG5
monoclonal antibody to cancer cells was calculated.
[0120]
(2) Antitumor effect (CDC activity) of monoclonal antibody against CSPG5 protein on cancer
cells
The cytotoxic activity7 (CDC activity) of monoclonal antibody #1 against CSPG5
protein screened as described above on cancer cells was evaluated. Blood was taken from a

rabbit, placed in an Eppendorf tube, allowed to stand still at room temperature for 60 minutes.
and centrifuged at 3000 rpm for 5 minutes to prepare a serum for CDC activity measurement.
Human leukemia cell line K562 and malignant lymphoma cell line L-1236 (lO'^ cells for each)
were collected separately in 50 mL-volume centrifuge tubes and 100 uCi of chromium 51 wras
added, incubated at 37°C for 2 hours, washed three times with RPMI medium containing a
10%FBS. suspended with RPMI medium containing 50% of rabbit serum prepared as
described above and added to a 96 well (with a V-shapc bottom) plate in a ratio of 10 cells
per well. To this, the monoclonal antibody #1 used in the above step (1) was added
individually in an amount of 1 ^g and cultured at 37°C. in a 5%CC>2 condition for 4 hours.
After culture, the amount of chromium 51 released from damaged tumor cells in the culture
supernatant was measured and the CDC activity of anti-CSPG5 monoclonal antibody in
hybridoma supernatant to K562 and L-1236 was calculated. As a result, monoclonal
antibody #1 has a CDC activity of 26%. The monoclonal antibody prepared in Example 5
and reacting with CSPG5 protein itself but does not react with the surface of cancer cells, was
subjected to the same operation. As a result, no cytotoxic activity was observed.
Accordingly, it was demonstrated that the monoclonal antibody (#1) against CSPG5 protein
damages tumor cells expressing CSPG5 protein also based on the CDC activity.
[0121]

The in-vivo antitumor effect of monoclonal antibody #1 (obtained above) against
CSPG5 protein in a cancer-bearing mouse was evaluated. The antibody used herein was
obtained by purifying each hybridoma culture supernatant by a column, in the same manner as
above.
[0122]
The antitumor effect of the monoclonal antibody #1 against CSPG5 protein was
examined using a cancer-bearing mouse obtained by grafting mouse-derived leukemia cell line
EL4-CSPG5. which expresses CSPG5 protein and was established in Example 3-(2). To the
subcutaneous portion of the back of each of thirty C57BL/6 mice (manufactured by Japan SEC,
Inc.), EL4-CSPG5 cells (10h cells/mouse) were grafted and the mice were allowed to grow

until a tumor reached a size of about 7 mm in diameter. To ten cancer-bearing mice out of
these mice, monoclonal antibody #1 against CSPG5 protein was intraperitoneal!)' administered
in a dose of 100 ug (100 uL) per mouse. To another ten mice, a monoclonal antibody-, which
was prepared in Example 5 and reacts with CSPG5 protein itself but does not react with the
surface of cancer cells, was intraperitoneally administered in a dose of 100 jig (100 pL) per
mouse. Thereafter, each antibody in the same dose was intraperitoneally administered to
individual cancer-bearing mice once every three days for three times in total. Every day, the
size of tumors was measured to observe the antitumor effect. PBS (-) was administered in
place of the antibody to the remaining ten cancer-bearing mice, and they were used as a
control group. As a result of observation of the antitumor effect, in the group to which the
monoclonal antibody (#1) against CSPG5 protein was administered, the tumor volume was
reduced to about 90% on Day 10 and about 70% on Day 20 and 60-somc % on Day 30 based
on the tumor volume at the i nitial day of administration of the antibody as 100%. In contrast,
in the control group, the tumor volume was increased up to about 260%, 350% and 550% on
Day 10. Day 20 and Day 30, respectively. In the group to which a monoclonal antibody,
which reacts with CSPG5 protein itself and does not react with the surface of cancer cells, was
administered, the antitumor effect was not obtained and the tumor volume was increased in the
same manner as in the control group. It was demonstrated from the results that the
monoclonal antibody (#1) against CSPG5 protein exerts a strong in-vivo antitumor effect on
the leukemia cancer cells expressing CSPG5 protein. The size (volume) of the tumor was
calculated in accordance with the formula:
Major axis x Minor axis x Minor axis x 0.5.
Industrial Applicability
[0123]
The antibody of the present invention is useful for treating and/or preventing cancer.
[0124J
All publications. Patents and Patent Applications cited in the specification arc
incorporated in the specification in their entirety by reference.

We claim
[Claim 1]
A pharmaceutical composition for treating and/or preventing cancer, comprising an
antibody or a fragment thereof having immunological reactivity with CSPG5 protein or a
fragment thereof consisting of at least 7 or more consecutive amino acid residues, as an active
ingredient.
[Claim 2]
The pharmaceutical composition according to claim 1, wherein the CSPG5 protein
consists of
any one of amino acid sequences represented by SEQ ID NOs: 8. 4, 6, 10 and 12. or
an amino acid sequence having an amino acid identity of 80% or more to the amino
acid sequence.
[Claim 3]
The pharmaceutical composition according to claim 1 or 2, wherein the cancer is
leukemia or malignant lymphoma.
[Claim 4]
The pharmaceutical composition according to any one of claims 1 to 3, wherein the
antibody is a monoclonal antibody or a polyclonal antibody.
[Claim 5]
The pharmaceutical composition according to any one of claims 1 to 4. wherein the
antibody is a human antibody, a humanized antibody, a chimeric antibody, a single-chain
antibody or a bispecific antibody.