US20120177636A1

US20120177636A1 - Carbohydrate-containing pan cancer marker

Info

Publication number: US20120177636A1
Application number: US13/427,109
Authority: US
Inventors: Chi-Yu Gregory Lee
Original assignee: Vancouver Biotech Ltd
Current assignee: Vancouver Biotech Ltd
Priority date: 2007-05-14
Filing date: 2012-03-22
Publication date: 2012-07-12
Also published as: AU2008250953A2; EP2155250A4; WO2008138139A1; AU2008250953B2; CA2687403A1; EP2155250B1; US8143373B2; JP2010529950A; US20110033445A1; EP2155250A1; AU2008250953A1; CN102014955A; JP5683948B2

Abstract

The effective epitope of CA215, a known cancer marker and antigen, has been demonstrated to include a carbohydrate moiety of defined composition and to be non-reactive with anti-human IgG, IgA and IgM, although CA215 is an immunoglobulin heavy chain-like molecule. The defined epitope may be used to prepare immunogenic compositions for treatment and prevention of cancers in humans and may be optimized as to protocol and formulation in animal model systems. Improved protocols for diagnosis and treatment are also described.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a divisional of U.S. Ser. No. 12/599,284 having an international filing date of 14 May 2008, which is the national phase of PCT application PCT/CA2008/000932 having an international filing date of 14 May 2008, which claims priority to U.S. Provisional Application Ser. No. 60/917,906 filed 14 May 2007 and U.S. Provisional Application Ser. No. 61/044,028 filed 10 Apr. 2008. The contents of these documents are incorporated herein by reference.

REFERENCE TO SEQUENCE LISTING SUBMITTED VIA EFS-WEB

The entire content of the following electronic submission of the sequence listing via the USPTO EFS-WEB server, as authorized and set forth in MPEP §1730 II.B.2(a)(C), is incorporated herein by reference in its entirety for all purposes. The sequence listing is identified on the electronically filed text file as follows:


File Name	Date of Creation	Size (bytes)

616342000110Seqlist.txt	Mar. 9, 2012	53,053 bytes

TECHNICAL FIELD

The invention relates to the field of protein markers that can distinguish cancer cells or tissues from normal cells or tissues and are found on many tumors in human subjects. More specifically, the invention relates to the carbohydrate-containing epitope of the known cancer marker CA215 and to methods of using this epitope.

BACKGROUND ART

U.S. Pat. No. 5,650,291 ('291) incorporated herein by reference describes the isolation of a tumor-associated antigen, CA215, which is present on an ovarian tumor cell line, and is also displayed on many tumors in humans. Monoclonal antibodies were prepared to this antigen, including the monoclonal antibody RP215. The hybridoma cell line that produces this antibody was deposited at the American Type Culture Collection under the terms of the Budapest Treaty on 5 Apr. 1989 as ATCC HB10095. The current address of ATCC is P.O. Box 1549, Manassas, Va. 20108. The '291 patent describes CA215 as having a minimum molecular weight of 60 kD on SDS gels when identified with RP215. However, aggregates with molecular weights ranging from 100 kD to 2,000 kD were also shown to be present. CA215 was purified by immunoaffinity chromatographic procedures and could be purified either from an extract of cultured ovarian tumor cells (OC-3-VGH) or from the shed culture medium of these cells. The CA215 antigen is characterized as a “membrane associated” soluble antigen which can be detected by RP215 in sera of patients with ovarian or cervical cancer. The antigen could not be detected in any normal tissue. This antigen and the monoclonal antibody that recognizes it were also described in an article by Lee, C. Y. G., et al., Cancer Immunol. Immunother. (1992) 35:19-26. CA215 was denominated Cox-1 in that article.
In a later paper, authored by the same group, Lee, G., et al., J. Clin. Ligand Assay (2006) 29:47-51, it was reported that treatment with periodate at neutral pH virtually eliminated the immunoreactivity of CA215 in a sandwich assay employing RP215. This led the authors to the conclusion that the epitope of CA215 reactive with RP215 may comprise carbohydrate.
It appears that the epitope of CA215 recognized by RP215 is present on approximately 60% of all cancers. Further information on its distribution is found in Lee, G., et al., J. Clin. Ligand Assay (2006) supra.
The '291 patent further describes a method to determine the location of tumors bearing the antigen CA215 by utilizing the antibodies immunoreactive against it to label cells that produce this antigen. Labeling the monoclonal antibodies with various radioisotopes was described as well as conjugating toxins to these antibodies and administration of the antibodies or immunotoxins for therapeutic use.
The present invention further refines the work described in these publications by demonstrating that the carbohydrate portion of the epitope is located at the variable region of immunoglobulin heavy chain-like molecules, thus making possible compositions which comprise only the relevant portions of CA215 for inclusion in vaccines or for generating and purifying antibodies useful in imaging of targeted cancer cells. This work also demonstrates that there are two forms of CA215—one membrane-bound and another that is secreted.

DISCLOSURE OF THE INVENTION

The invention is directed to compositions that consist essentially of the epitope region of CA215. This epitope region comprises a carbohydrate and optionally at least a portion of an immunoglobulin heavy-chain like variable region amino acid sequence. This epitope is specifically immunoreactive with RP215 monoclonal antibody, but is not significantly immunoreactive with anti-human IgG.
In other aspects, the invention relates to the use of the minimal epitope or antiidiotype antibodies that mimic it as active ingredients in therapeutic and prophylactic methods to treat cancer. The epitope and antiidiotype antibodies can also be used as reagents for affinity purification of and for identification of additional monoclonal antibodies useful as diagnostic or therapeutic reagents for cancers.
In another aspect, the invention relates to improvements in immunoassays for CA215 using an alternative monoclonal antibody directed against this antigen or an antibody that is immunoreactive with IgG as a component in a new sandwich assay.
In other aspects, the invention relates to improved monoclonal antibodies which are modified forms of RP215, including humanized forms. Humanized forms of RP215 are useful in therapeutic methods, and can be conjugated to additional antineoplastic moieties to improve targeting of such moieties.
In additional aspects, the invention relates to protocols that take advantage of the dual secreted/membrane-bound nature of the CA215 antigen. In such protocols, diagnosis in body fluids by detection of the secreted form is effected, optionally using the improved assay system of the present invention followed by localization and treatment of solid tumors using the invention antibodies in humanized form optionally coupled to cytotoxic agents for treatment or radioisotopes for localization and/or treatment.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1-5 are full ESI-MS spectra of released N-linked glycans from human IgG, RP215 mAb, and three samples of CA215 respectively.

FIG. 6 shows the effect of washing on the presence of CA215 in membrane-bound form on OC-3-VGH cells.

FIG. 7 shows the secretion pattern of the OC-3-VGH cell line compared to a hybridoma cell line with respect to the secreted form of CA215.

MODES OF CARRYING OUT THE INVENTION

The present invention identifies the epitope on CA215 that is immunoreactive with monoclonal antibody RP215 as comprising the carbohydrate portion of this antigen and establishes the identity of the protein portion as a heavy chain immunoglobulin-like molecule, including the immunoglobulins of classes IgG, IgA and IgM. Although some light chain immunoglobulin-like moieties appear to be associated with CA215, they are not present in a 1:1 ratio to heavy chain as in ordinary immunoglobulins and do not bear the carbohydrate-containing epitope associated with the heavy chain-like portion. CA215 exists as undefined aggregates on the cancer cell surface.
The identification of the epitope recognized by RP215 comprising a carbohydrate not present on immunoglobulins in general and associated with the variable region of a heavy chain immunoglobulin-like molecule permits the production of more sophisticated immunogenic compositions which in turn are useful to inhibit the growth of tumor cells that display CA215 at their surfaces and to generate additional antibodies useful as detection reagents or immunotoxins. As demonstrated in the examples below, using the effective epitope of CA215, which, as stated above, is present on approximately 60% of human cancers, more effective immunogenic compositions can be formulated for cancer prevention and treatment.
Thus, rather than employ the entire CA215 antigen as a vaccine to slow the progression of cancer already established or to prevent the appearance of detectable amounts of cancer cells that express this antigen, only the portion of CA215 that bears the epitope relevant to detection and treatment need be employed.
As demonstrated below, the epitope consists essentially of a portion of the antigen which does not immunoreact with anti-human IgG, anti-human IgA, or anti-human IgM. The carbohydrate epitope has a composition distinct from that of these human immunoglobulins and distinct from that of the monoclonal antibody RP215 that immunoreacts with it. The composition of the carbohydrate epitope is approximately 1-3% fucose, 9-15% N-acetyl galactosamine, 27-30% N-acetyl glucosamine, 6-15% glucose, and 47-51% mannose. These are approximate figures, ± at least 1-2% in the latter four cases. The carbohydrate epitope is free of N-acetyl neuraminic acid and N-glycol neuraminic acid.
The epitope may also comprise at least a small portion of the variable region immunoglobulin heavy chain-like protein to which the carbohydrate is bonded.
Thus, the purified antigen CA215 has been identified as an immunoglobulin heavy chain-like molecule wherein the epitope portion immunoreactive with RP215 includes the carbohydrate associated with this amino acid sequence. By “immunoglobulin heavy chain-like molecule” is meant a molecule that includes an amino acid sequence that is able to confer immunoreactivity with anti-immunoglobulin antibodies. Thus, in the present case, CA215 is immunoreactive with anti-IgG, anti-IgA, and anti-IgM. The epitope immunoreactive with RP215, however, is not immunoreactive with anti-IgG, anti-IgA or anti-IgM. The amino acid sequence of the protein portion of CA215, thus, is sufficiently homologous with an immunoglobulin heavy chain that immunoreactivity is exhibited with respect to anti-heavy chain immunoglobulin antibodies. CA215 is also immunoreactive with antibodies that recognize immunoglobulin light chain. Thus, more generally, CA215 might be described simply as an immunoglobulin-like molecule.
The identified epitopes of the invention may be formulated into vaccines for administering to subjects for the treatment and prevention of cancer. Animal model subjects, such as mice, rats, rabbits, guinea pigs, and the like, may be administered such vaccines to optimize the formulation and protocols. Human subjects may be treated with additional therapies such as radiation and chemotherapy along with the immunogenic compositions of the invention.
Immunohistochemical staining studies of normal and cancerous tissues have demonstrated that the epitope identified herein is present on a number of types of human cancers, with varying levels of staining intensity. The epitope shows very intense staining on human cancers of the ovary, cervix, endometrium, colon, stomach, intestine, esophagus, breast, and lung. As noted herein, the tumor tissues from any particular subject can be evaluated using immunostaining for the presence and level of this epitope, thus providing information useful in the design of suitable vaccines, whether composed of the epitope itself or an antiidiotype antibody that mimics it as further described below.
In addition, to the use of the epitopes of the invention to generate antibodies endogenously in cancer-bearing subjects, the epitopes may also be used to generate additional antibodies useful in detection and themselves useful in treatment. As used herein, “antibodies” includes complete immunoglobulins as well as immunospecific fragments thereof, such as Fab, Fab_2′ and F_vfragments. The antibodies may be monoclonal, prepared by standard and well known techniques and under these circumstances may be manipulated recombinantly to obtain humanized forms, chimeric forms in which the variable region associated with one species is coupled to a constant region associated with another or may be single-chain antibodies. Techniques for manipulation of monoclonal antibodies using the tools of recombinant production are well established. The epitopes of the invention may also be used as purification and identification tools for suitable antibodies.
In any of the immunogenic compositions, whether for the purpose of preparing monoclonal or polyclonal antibodies for diagnostic use or as a vaccine formulation, suitable adjuvants may be included in the composition, such as Freund's incomplete adjuvant, alum, and a multiplicity of other adjuvants well known in the art. In addition, the epitope may be coupled to additional moieties such as KLH or tetanus toxoid in order to enhance its immunogenicity. Thus, fusion proteins of the epitopes of the invention with additional heterologous protein are included in the scope of the invention.
The antibodies generated in response to the defined epitopes of the present invention can be labeled with radioisotopes, fluorophores, and in the case of in vitro assays, enzymes, and used to detect the presence of cancer cells. They may also be coupled to toxins for use in therapy.
In addition to the use of the defined epitope of the invention to prepare immunogenic formulations, antiidiotype antibodies which mimic this epitope may be isolated from subjects immunized with RP215 or immunogens that recognize the same epitope as does RP215. Antiidiotype monoclonal antibodies are obtained by immunizing mice or other suitable subjects with purified RP215 mAb or its Fab fragments (or with its humanized form) to elicit an antiidiotypic response against epitopes in the variable region. For example, BALB/C mice may be used. Conventional preparation of monoclonal antibodies by cell fusion and screening using RP215 or its Fab fragments or other moieties that recognize the epitope of the invention will identify monoclonal antibodies that are antiidiotypes. These antiidiotype mAb's then can serve as immunogens to elicit antibodies in subjects to target cancer cells. Thus, the antiidiotype antibodies can be substituted for the CA215 epitope as cancer therapy and in other applications.
Suitable formulations for the defined epitope of the invention are those conventional for immunogenic compositions and are found, for example, in Remington's Pharmaceutical Sciences, latest edition, Mack Publishing Co., Easton, Pa., incorporated herein by reference. Protocols for administration are dependent on the nature of the condition, the judgment of the attending physician, and the severity of the malignancy. Optimization of such protocols on a group or individual basis is well within ordinary skill.
The identification of the CA215 epitope as residing on an immunoglobulin-like moiety has led to improvements in immunoassays for this antigen. Previously, monoclonal RP215 has been used as both members of the “sandwich” employed in standard immunoassays using a variety of labels for detection, including enzymes, radioisotopes and fluorescent molecules. This was possible because CA215 commonly exists in polymeric form and multiple copies of the same epitope are available. An improved form of the assay, however, employs, as one member of the sandwich, antibodies immunoreactive with human immunoglobulin, preferably IgG, so that monomeric forms of CA215 may also be detected.
Improvements are also contemplated in the structure of RP215 through manipulation of the nucleotide sequence encoding the variable region. Thus, RP215 may be humanized or otherwise modified to improve its immunospecificity. The humanized form of this antibody is particularly useful in therapeutic applications. Such humanized forms may be complete immunoglobulins, or may include only variable regions, such as Fab or Fab_2′ portions or may be single chain F_vantibodies produced recombinantly. Any immunospecific portion of RP215 may be modified so as not to raise an immune response in human subjects.
Such antibodies or fragments or modified forms may be coupled to additional biologically active moieties, such as antineoplastic agents including immunoglobulins or fragments thereof immunoreactive with undesirable growth factors. In these conjugates, the RP215 serves as a targeting agent, as well as an anti-tumor factor per se. Thus, these forms of RP215 may be coupled to antineoplastic agents, such as paclitaxel, rapamycin or fumagillin or to moieties that are inhibitors of growth factors or their receptors, such as anti-GNRH receptor, anti-EGF, anti-EGFR, anti-VEGF, anti-VEGFR, and the like.
As used herein, “antineoplastic agent” includes small molecules, such as those set forth above, as well as antibodies or fragments thereof directed against growth factors or the receptors for such growth factors. Thus “anti-growth factor” or “anti-growth factor receptor” refers to immunoglobulins or fragments thereof that are immunoreactive with these moieties.
In the improved protocols of the present invention, an initial step is diagnosis of the presence of malignancies characterized by epithelial cells by virtue of the presence of CA215 antigen in body fluids. Such fluids include sera, plasma, blood, urine, saliva, and the like. The detection can be performed using RP215 or the modified forms thereof described above. The modified assay of the invention may also be used.
Once a diagnosis is made, the location of the tumor may be ascertained by obtaining an image by injection, in the case of humans, of the humanized form of RP215 coupled to an imageable label, such as a radioisotope, fluorescent dye, or luminescent system. Fluorescent proteins may be employed as fusion proteins or otherwise linked to the antibodies. In addition, the antibodies may be used as targeting agents for cytotoxic agents for the treatment of these solid tumors.
The following examples are offered to illustrate but not to limit the invention.

EXAMPLE 1

Characterization of CA215 as an Immunoglobulin-Like Molecule in Both Secreted and Membrane-Bound Forms with an Epitope Comprising a Carbohydrate

In this example, the immunoglobulin-like nature of CA215 was confirmed and it was demonstrated that this cancer-associated antigen is produced simultaneously in both secreted and membrane-bound forms with differing molecular weights. In addition, the epitope of CA215 which is immunoreactive with RP215 comprises a carbohydrate moiety.
CA215 was purified from the shed medium of cultured OC-3-VGH cancer cells using the methods described in U.S. Pat. No. 5,650,291 cited and incorporated by reference above.
NH₂-terminal amino acid sequence analysis of purified CA215 gave a sequence identical to that of normal human IgG (VELVESGA) (SEQ ID NO:1).
The immunoglobulin nature of CA215 was confirmed by Western blot assays. CA215 proteins were separated by, and transferred to, nitrocellulose membrane strips from SDS-PAGE of either OC-3-VGH cancer cell extract, shed medium, or affinity-purified CA215.
After direct incubation with enzyme-labeled RP215 or with alkaline phosphatase (ALP) labeled anti-human IgG, the nitrocellulose strips were incubated with substrates (Bio-Rad Labs) for color detection of protein bands. Indirect binding assays with nitrocellulose membrane strips were also performed using unlabeled RP215 or unlabeled anti-human Ig as primary antibody and enzyme labeled goat anti-mouse IgG or rabbit anti-goat IgG as second antibodies. The results of these extensive studies are summarized in Table 1.

TABLE 1

Western blot assays using various antibody probes to reveal molecular weight(s) of
detected protein bands on nitrocellulose strips which were derived from those of cancer
cell extract, cultured shed medium or purified CA215.

		Source of	Molecular wt of	Relative
Primary		nitrocellulose	protein band(s)	Staining
Antibody	Secondary Antibody	strips	detected (kDa)	Intensity

RP215-HRP^a	—	OC-3-VGH	55	+
		cancer cell
Anti-human	—	purified CA215	60	++
IgG (Mab)			25	+
RP215 (Mab)	goat anti-mouse IgG-ALP^a	OC-3-VGH	53-70	+++
		cancer cell	(broad)
	goat anti-mouse IgG-ALP	cultured shed	50-54	+
		medium
	goat anti-mouse IgG-ALP	purified CA215	50-52	++
	goat anti-mouse IgG-ALP	purified CA215 after	50-60	++
		pepsin treatment	24	+
Anti-human	goat anti-mouse IgG-ALP	OC-3-VGH	55-70	+++
IgG (Mab)		cancer cell	28	++
	goat anti-mouse IgG-ALP	purified CA215	52-60	++
Anti-human	goat anti-mouse IgG-ALP	OC-3-VGH	57-70	++
IgA (Mab)		cancer cell	(broad)
Anti-human	goat anti-mouse IgG-ALP	OC-3-VGH	53-70	++
IgM (Mab)		cancer cell	(broad)
Anti-human	goat anti-mouse IgG-ALP	OC-3-VGH	46-53	++
Ig light κ		cancer cell	(weak)
chain (Mab)			20-27	+
Anti-human	goat anti-mouse IgG-ALP	OC-3-VGH	56	++
light λ chain		cancer cell	(weak)
(Mab)			23-32	+

^aHRP-horseradish peroxidase
ALP-alkaline phosphatase

Both direct and indirect Western Blot assays using RP215 for detection give the same protein band patterns regardless of whether the protein was cellular extract, shed medium or affinity purified CA215. A strong, broad protein band was observed at molecular weight of 60 kDa and a minor protein band was also detected at 90 kDa.
Anti-human IgG monoclonal antibody (γ-chain-specific) also gave broad protein band(s) of 60 kDa either in a direct assay with ALP-labeled goat anti-human IgG, or in an indirect assay using this antibody as a secondary marker. Anti-human IgG lambda and kappa light chain monoclonal antibodies, also recognize protein band(s) of lower molecular weight (25-30 kDa), although with a much lower staining intensity. Anti-human IgA and IgM monoclonal antibodies recognize the protein bands with similar molecular size of 60 kDa similar to those recognized by RP215 monoclonal antibody.
As shown, the relative concentration of cancer cell-derived IgG is significantly higher than that of human IgA or human IgM (≦5-10% of IgG). Western blot of affinity purified CA215 before and after pepsin digestion showed that after pepsin digestion, the remaining Fab fragment(s) of CA215 can be detected at low molecular range (˜30 kDa) by RP215 monoclonal antibody.
More direct evidence that CA215 in cancer cells cross-reacts with human IgG, human IgA or human IgM, and that unique epitope(s) recognized by RP215 exist in these cancer cell-derived immunoglobulin-like molecules was obtained as follows: Monoclonal anti-human IgG (Cox-100)*, anti-human IgA and anti-human IgM were coated separately on microwells according to standard procedures. Shed medium from OC-3-VGH cells was added to the wells and RP215-HRP was used as the detecting antibody. The sandwich immunoassays were performed in one-step at room temperature overnight with 1/200 RP215-HRP+10 ug/ml normal mouse IgG. * Cox-100 is a monoclonal antibody obtained by immunizing mice with purified CA215, harvesting the spleens, performing cell fusion and screening using standard techniques for preparation of monoclonal antibody. Cox-100 reacts with CA215, and crossreacts with human IgG.
The results of this assay in Table 2 demonstrate the presence of human immunoglobulin-like molecules in cultured shed media of OC3-VGH cancer cells with various antibodies to human immunoglobulin molecules. ODs is the OD value of sample and ODn is the OD value of negative control (culture medium).

TABLE 2

Coating Abs/Ag	Capturing Ab	ODs₆₅₀/ODn₆₅₀

Cox 100	RP215-HRP	13.6
(anti-hIgG Mab)
Anti-hIgA Mab	RP215-HRP	1.7
Anti-hIgM Mab	RP215-HRP	2.2

Additional data were obtained to demonstrate direct binding between various anti-immunoglobulin antibodies to CA215 present in OC-3-VGH cancer cells as shown in Table 3A. In this case the experiment was conducted as a secondary two-step ELISA with primary antibodies incubated with sample overnight at room temperature, followed by goat antibody Anti-Mouse IgG-ALP, for 1 hour at 37° C. ODs is the OD value of sample and ODn is the OD value of corresponding normal mouse IgG concentration.

Table 3

	TABLE 3A

	Capturing Ab	ODs₄₀₅/ODn₄₀₅

	1.25 ug/ml	9.1
	Anti-hIgG₂mAb
	2.50 ug/ml	1.4
	Anti-hIgG₃mAb
	1.125 ug/ml	12.9
	Cox-100
	1.125 ug/ml	11.6
	RP-215
	5.00 ug/ml	1.5
	Anti-hKappa Mab
	5.00 ug/ml	2.8
	Anti-hLamda Mab

Further evidence that CA215 is an immunoglobulin-like molecule mimicking the heavy chain of IgG was obtained using homology analysis. Table 3B shows the results obtained by MALDI-TOF MS system analysis.

TABLE 3B

Amino Acid Sequence Homology Analysis of Tryptic Peptides of CA215 determined by
MALDI-TOF MS

	SEQ
	ID
Peptide Fragments	NO:	Sequence Homology (%)

1. K.DVLTITLTPK.V	2	immunoglobulin heavy chain variable region (66%)

2. K.APQVYTIPPK.E	3	immunoglobulin gamma heavy chain 3 (87%)

3. R.VNSAAFPAPIEK.T	4	immunoglobulin gamma heavy chain 3 (88%)

4. K.APQVYTIPPKEQMAK.D	5	Ig gamma-3 chain C region (Heavy chain disease protein)
		(62%)

5. R.SVSELPIMHQDWLNGK.E	6	immunoglobulin heavy chain (72%)

6. K.NTQPIMDTDGSYFVYSK.L	7	immunoglobulin gamma-1 heavy chain constant region
		(61%)

7. K.SSGTSYPDVLK.C	8	immunoglobulin heavy chain variable region (64%)

8. K.VCNYVSWIK.Q	9	immunoglobulin heavy chain (75%)

9. RTLYLQMNSLR	10	immunoglobulin heavy chain variable region (100%)

10. SLVVAAVAPDNRNPAFTTM	11	immunoglobulin heavy chain variable region (70%)
GWLFLK

11. GDRVTITWR	12	immunoglobulin heavy chain variable region (88%)

12. GLSDSVRSCR	13	immunoglobulin heavy chain variable region (75%)

13. TAKGSTGMEILLSTLENTK	14	immunoglobulin heavy chain variable region VH (61%)

14. KVTCVVVDISKD	15	immunoglobulin heavy chain (88%)

15. GPLCGCCPGRSSQK	16	immunoglobulin variable region (43%)

16. AELGGLLSPR	17	immunoglobulin heavy-chain subgroup VIII V-D-J region
		(85%)

17. DGSISILGSDDATTCHIVVLR	18	immunoglobulin heavy chain variable region (100% - 7/7)

18. RTLYLQMNSLR	19	immunoglobulin heavy chain variable region (100%)

19. KCELNCQAMGYR	20	immunoglobulin gamma chain, V region (85%)

20. LSGSCRSTDSLHPCPPTALPR	21	immunoglobulin heavy chain variable region (33%)

21. APTVVLMMTK	22	immunoglobulin heavy chain variable region (85% - 5/6)

22. ATSRGCITIIGGGDTATCCAK	23	immunoglobulin heavy chain variable region (69% - 9/13)

23. MSTRYHQAASDSYLELIK	24	immunoglobulin heavy chain variable region (87% - 7/8)

24. SLPGSPKDSSHLLSPLR	25	Ig heavy chain (VH4) V region (VDJ)

25. GGNSGGSSSICYVLLGFIGTS	26	immunoglobulin heavy chain VJH1 region (77%)
K

26. AEDTAVYYCAKTLTIR	27	immunoglobulin heavy chain variable region (100%)

27. GLECIGYMYSSGSSFYNPSL	28	immunoglobulin heavy chain variable region (100%)
KSR

28. MAYLQQTLAGPSGTR	29	immunoglobulin heavy chain variable region (88% - 8/9)

29. KGHQDSCPFELTACPNEGCT	30	Ig heavy chain variable region (75% - 6/8)
SQVPR

30. GLEWVSAVSGSGTTTYYAD	31	immunoglobulin heavy chain variable region (91%)
SVK

31. LSSVTAADTNVYYCAR	32	immunoglobulin heavy chain VHDJ region (93%)

32. AETLVFSTHAVISMR	33	immunoglobulin heavy chain variable region (70% - 7/10)

Thus, sequences contained in CA215 are homologous to sequences in human immunoglobulin heavy chain.
In Table 3C, partial amino acid sequences of CA215 were mapped with the known sequences of human immunoglobulin heavy chains of several other cancer cells or tissues reported previously. It is now clearly established that CA215 is not a single well-defined molecule, but a mixture of numerous human immunoglobulin heavy chain molecules (e.g., IgG, IgA, IgM and numerous variations in the V regions). A unique character in these immunoglobulin mixtures is the existence of specific carbohydrate-associated epitope that can be commonly recognized by RP215 monoclonal antibody.

TABLE 3C

Comparisons of Partial Amino Acid Sequences of CA215 Deduced from MALDI-TOF
MS with Those of Known Human Ig Heavy Chain from Cancer Cells

		SEQ ID		SEQ ID
	FR1	NO:	CDR1	NO:

T47D (IgG)	EVQLVESGGGLVQPGGSLRLSCAASRFSSR	34	TSGMR	35

ZR75-1 (IgM)	EVQLVQSGAEVKKPGESLKISCKGSGYSFT	36	SYWIG	37

ZR75-1 (IgG)	EVQLLESGGGLVQPGGSLRLSCTASGFNFN	38	TYAMT	39

SKBR3 (IgG)	QVQLQESGPGLVKPSQTLSLTCTVSGGSVS	40	SGYYYWS	41

SKBR3 (IgA)	EVQLVESGGGLVQPGGSLTLSCAVSGLSFS	42	SSGMN	43

MDA-MB-231 (IgM)	EVQLVESGGGLVQPGGSLRLSCAASGFTFS	44	SYWMD	45

LUNG CANCER	EVQLEESGAEVKKPGESLKISCEASGYTFG	46	TYWIG	47

CA215	KSSGTSYPDVLKCKVCN-----	48	YVSW	49
	(SLVVAAVAPDNRNPAFT?)	50	(TMG?)
	(ATSRGCITIIGGGDTATCCAK?)	51
	(MAYLQQTLAGPSGTR?)	52

		SEQ ID		SEQ ID
	FR2	NO:	CDR2	NO:

T47D (IgG)	WVRQAPGKELEVA	53	PFWNGGSQKYCADSVT	54

ZR75-1 (IgM)	WVRQMPGKGLEWMG	55	I IYPGDSDTRYSPSFQG	56

ZR75-1 (IGG)	WVRQAPGKGLEWVS	57	T IAADGTWTSNADFVRG	58

SKBR3 (IgG)	RIRQHPGKGLEWIG	59	YIYYNGSTYENPSLKS	60

SKBR3 (IgA)	WVRQASGKGLEWVG	61	RIGSKAASDTTSYAASVRG	62

MDA-MB-231(IgM)	WVRQVPGKGLVWVS	63	RISPDGRTTTYADSVEG	64

LUNG CANCER	WVRQMPGKGLEWMG	65	IIYPGDSDTTYSPSFRG	66

CA215	IKQ--------GLEWVS	67	AVSGSGTTTYYADSVK	68
	(WLFLK?)	69	YMYSSGSSFYNPSLKSR?)G	71
	(GLECIG	70

		SEQ		SEQ
		ID		ID
	FR3	NO:	CDR3	NO:

T47D (IgG)	GRFTFSETFLRPCSLCKCTVNLRARPS I PAP	72	GITVPHPRLCPRN	73

ZR75-1 (IgM)	QVTISADKSISTAYLQWSSLKASDTAMYYCAR	74	QEIVAFS	75

ZR75-1 (IgG)	RLTISRDNSRNTLYLQMNSLRAEDTAIYFCAK	76	DWYDY	77

SKBR3 (IgG)	RASISVDTSKNQFSLKLSSVTAADTAVYYCAR	78	DIKHTYGPN	79

SKBR3 (IgA)	RFFISRDDSKKTVYLQMNSLKTEDTAVYYCSR	80	QGCGGDCHIPKM	81

MDA-MB-	RFTISRDNAKNTLYLQMNSLRAEDTAVYYCAG	82	GYLSSH	83
231 (IgM)

LUNG	QVTLSVDKFINTAYLQWDSLKASDTAIYYCAR	84	WDVMIGFYTA	85
CANCER

CA215	DRVTITWR RTLYLQMNSLRAEDTAVYYCAK	86	TLTIR	87
	(LSSVTAADTNVYYCAR?)	88
	(MAYLQQTLAGPSGTR?)	89

	JH	SEQ ID	CH	SEQ ID
		NO:		NO:

T47D (IgG)	YFDSGQGTLVTVSS	90	ASTKGPSVFPL	91

ZR75-1 (IgM)	YYYMDVWGKGTTVTVSS	92	GSASPQPFSPS	93

ZR75-1 (IgG)	WGQGTLVTALL	94	TVSTGLHQGPIGLPP	95

SKBR3 (IgG)	YNCYMDVWGKGTTVTVSS	96	GLHQGPIGLPP	97

SKBR3 (IgA)	YYYYGMDVWGQGTTVTVSS	98	ASPTSPKVF	99

MDA-MB-231	DYWGRGTLVTVSS		100	GECIRPNPFPP	101
(IgM)

LUNG	DYWGQGTQVTVS	102	SASTKGPSVFPLAPSSKSTSGGT	103
CANCER			AVLGCLV

			KDYFPEPVTV	104
CA215			TAKGSTGMEILL STLENTK (?)	105

CA215 (constant region)
-----GNSGGSSSICYVLLGFIGTSKLSGSCRSTDSLHPCPPTALPRAELGGLLSPRKDVLTITLTPKVTCVVVDISKDRSVSELPIMHQDWLNGKERVNSAAFPAPIEKTKAPQVYTIPPKEQMAKD--------KGHQDSCPFELTACPNEGCTSQVPRKNIQPIMDTDGSYFVYSKL (SEQ ID NO: 106)
Note:
The following sequences cannot be mapped due to microheterogeneity: 1. GPLCGCCPGRSSQK (SEQ ID NO: 107); 2. APTVVLMMTK (SEQ ID NO:108); 3. 3MSTRYHQAASDSYLELIK (SEQ ID NO:109); 4. SLPGSPKDSSHLLSPLR (SEQ ID NO: 110).

In Table 3D, comparisons of amino acid sequences of CA215 in the constant region deduced from MALDI-TOF MS and RT-PCR are presented together with that of anti-human colon carcinoma heavy chain from the GenBank.

TABLE 3D

Comparisons of amino Acid Sequences in the Constant Region among those of CA215
deduced from RT-PCR and MALDI-TOF MS as well as those from Anti-human Colon
Carcinoma Heavy Chain (AHCCHC gbIAAB28159.1\|) (SEQ ID NOS: 111-113)

AHCCHC	PKDVLTITLTPKVTCVVVDISKDDPEVQFSWFVDDVEVHTAQTQPREEQF

CA215 by MALDI-TOF MS	KDVLTITLTPKVTCVVVDISKD..................................... LSTLE

CA215 by RT-PCR	.......................................... VEVHNAKTKPREEQF

AHCCHC	NSTFRSVSE LP I MHQDWLNGKEFKCRVNSAAFPAPIEKTISK TKG---------

CA215 by MALDI-TOF MS	N -TKRSVSE LP I MHQDWENGKE..... RVNSAAFPAPIEKT...........

CA215 by RT-PCR	NSTYRVVSVLTVVHQDWLNGKEYKCKVSNKGLPAPIEKTISKTKGGTRGC

AHCCHC	-------------------------------------------------------RPKAPQVYT IPPPKEQMAKDKV

CA215 by MALDI-TOF MS	........................................................KAPQVYT IPP -KEQMAKDKV

CA215 by RT-PCR	EGHMDRGQLGPPSALGVTAVPTSVPTGQPREPQVYTLPPSREEMTKNQV

AHCCHC	SLTCMI TDEFPEDITVEWQWNGQPAENYKNIQPIMDTDGSYFVYSKL

CA215 by MALDI-TOF MS	--- TCVVVD- ISKD... ........................KNTQPIMDTDGSYFVYSKL

CA215 by RT-PCR	SLTCLVKGFYPSDIAVEWE SNGQPENNYK TT

The carbohydrate portion of CA215 was analyzed following verification of results demonstrating the ability of periodate to destroy the immunoreactivity of this antigen in a sandwich assay, thus establishing the presence of carbohydrate in the epitope. Verification was performed as follows:
PBS-washed OC-3-VGH cancer cells (conc. 1×10⁶cells/ml) were incubated with 100 mM NaIO₄for 30 minutes, the cells were washed with PBS containing 0.5% BSA, and then dried on microwells at 1×10⁴cells/well. The cell-coated microwells were then blocked with 0.5% BSA in PBS and direct binding enzyme immunoassays using RP215 labeled with horseradish peroxidase (RP215-HRP) were performed at 37° C. for one hour followed by extensive washes and color development with TMB substrate. For comparison wells coated with cancer cells without NaIO₄treatments served as control. The binding between RP215-HRP and cancer cell-coated wells treated with periodate was drastically reduced. In addition, the presence of 10 to 100 μg/ml goat anti-human IgG reduced the binding to RP215-HRP to the wells coated with cancer cells in a dose-dependent manner, indicating that goat anti-human IgG competes with RP215-HRP to bind the complete CA215 antigen. The results of this study are summarized in Table 4.

TABLE 4

Direct binding assays to reveal the effect of NaIO₄treatment
to OC-3-VGH cancer-cells on the binding of RP215-HRP to
the cancer cell-coated wells as well as its binding
inhibition in the presence of goat anti-human IgG

Optical Density at 450 nm

	Without	With
Assay Conditions	NaIO₄treatment	NaIO₄treatment

RP215-HRP (10 ug/ml) +	2.153	(100%)^a	0.797 (37%)
normal mouse IgG (10 ug/ml)
RP215-HRP (10 ug/ml) +
normal mouse IgG (10 ug/ml) +	0.961	(45%)	0.333 (15%)
goat anti-human IgG (20 ug/ml)
RP215-HRP (10 ug/ml) +
normal mouse IgG (10 ug/ml) +	0.471	(22%)	0.252 (12%)
goat anti-human IgG (50 ug/ml)
RP215-HRP (10 ug/ml) +
normal mouse IgG (10 ug/ml) +	0.163	(8%)	0.158 (7%)
goat anti-human IgG (100 ug/ml)

^apercent maximum binding

The presence of both secreted and membrane-bound forms of CA215 in OC-3-VGH cells was further confirmed as shown in Table 5 below, and in FIGS. 6 and 7. Table 5 shows the results of Western blot when the strips treated with detection reagents either contained immobilized OC-3-VGH whole cells, OC-3-VGH culture medium or purified antigen. As shown, regardless of the detection method, the purified antigen produced a result at only 54-55 kD molecular weight, as did the culture medium. However, the whole cells showed results at molecular weights of both 50-56 and 68-73 kD. In Table 5, N/A represents “not applicable” and ND represents “not done.”

TABLE 5

Western Blot of OC-3-VGH cells, Culture Medium and
Purified CA215 with RP215 and
Anti-Human IgG Probes Under Reducing Conditions

	Molecular weight of detected
	bands from various
Western	sources of nitrocellulose strips (KDa)

Blot Conditions

OC-3-VGH

Purified

Primary	Secondary	whole	culture medium	CA215
Antibody	Antibody	cell strips	strips	strips

RP215	GAMIgG-	56/68	55	55
	ALP
MAHIgG	GAMIgG-	55-73	55	54
	ALP	(broad)
MAHIgA-	N/A	50	ND	ND
ALP
MAHIgM-	N/A	70	ND	ND
ALP

FIG. 6 shows the result of consecutive PBS washes on the fraction of binding of RP215 to OC-3-VGH cells. As shown, even after five consecutive washes, no appreciable change in the fraction binding the cells occurs. These results were obtained on isolated OC-3-VGH cells not in culture.
This is in contrast with the results in FIG. 7 which show that when culture medium is assessed from either OC-3-VGH cells or a comparable standard hybridoma secretion system, similar patterns of secretion are obtained based on absorbance at 450 nm using the HRP detection system.
Thus, the secreted and membrane-bound forms appear to be produced simultaneously and have differing molecular weights. The secreted form has a molecular weight of approximately 55 kD and the membrane-bound form has a molecular weight of approximately 73 kD.

EXAMPLE 2

Carbohydrate Composition of Affinity-Purified CA215

The carbohydrate composition of CA215 was analyzed through a contract service by Complex Carbohydrate Research Center (Athens, Ga., USA). For comparison, composition analyses of normal human IgG, and RP215 monoclonal antibody were also performed.
The results of this comparative carbohydrate composition analysis are summarized in Table 6. The values of glucose were not determined because glucose was identified as a major contaminant and thus not susceptible to accurate measurement.

TABLE 6

Neutral and Amino Sugar Composition (excluding glucose)

Sample ID	Types of Amino-/Sugar	nmoles/μg	Molar %

Human IgG-	Fucose	0.0077	7.93
Salt free	N-acetyl-galactosamine	0.0068	6.99
	N-acetyl-glucosamine	0.0394	40.72
	Galactose	0.0053	5.46
	Glucose	ND	ND
	Mannose	0.0376	38.90
	N-acetyl-neuraminic acid	0.0000	0.00
	N-glycol-neuraminic acid	0.0000	0.00
	Total:	0.0967	100.0
RP215 Mab-	Fucose	0.0085	5.03
Salt free	N-acetyl-galactosamine	0.0000	0.00
	N-acetyl-glucosamine	0.0762	45.16
	Galactose	0.0076	4.48
	Glucose	ND	ND
	Mannose	0.0765	45.33
	N-acetyl-neuraminic acid	0.0000	0.00
	N-glycol-neuraminic acid	0.0000	0.00
	Total:	0.1688	100.0
CA215	Fucose	0.0018	1.10
Lot No. 070305 A	N-acetyl-galactosamine	0.0154	9.45
	N-acetyl-glucosamine	0.0442	27.05
	Galactose	0.0229	14.05
	Glucose	ND	ND
	Mannose	0.0789	48.35
	N-acetyl-neuraminic acid	0.0000	0.00
	N-glycol-neuraminic acid	0.0000	0.00
	Total:	0.1633	100.0
CA215	Fucose	0.0018	2.46
Lot No. 070305 B	N-acetyl-galactosamine	0.0095	13.04
	N-acetyl-glucosamine	0.0208	28.40
	Galactose	0.0042	5.68
	Glucose	ND	ND
	Mannose	0.0369	50.41
	N-acetyl-neuraminic acid	0.0000	0.00
	N-glycol-neuraminic acid	0.0000	0.00
	Total:	0.0731	100.0

Normal human IgG and RP215 (mouse IgG) have similar carbohydrate compositions, though human Ig contains N-acetyl galactosamine, which is absent from RP215. CA215 exhibits different sugar content from either normal human or mouse IgG. CA215 contains lower percentage of N-acetylglucosamine (27-28% vs. 40-45%) but a significantly higher amount of mannose (48-50% vs. 38-45%).

EXAMPLE 3

N-Linked and O-Linked Oligosaccharide Profiling

In addition to overall carbohydrate content, the structures of N-linked glycans associated with human IgG, RP215, and CA215 were determined by electrospray ionization mass spectrometry (ESI-MS).
In carrying out this determination, the samples were dissolved in 1 mL of nanopure water. Eight hundred microliters of each of human IgG, RP215 mAb, and CA215 sample B, 900 μL of CA215 sample A and all of CA215 sample C were pipetted into screw-cap tubes and lyophilized The dried samples were dissolved with 100 μL ammonium bicarbonate buffer (50 mM, pH 8.4) and followed immediately by reduction with 25 mM dithiothreitol (45 min at 50° C.) and carboxyamidomethylation with 90 mM iodoacetamide (45 mM at room temperature in the dark) prior to trypsin digestion (37° C., overnight). A second enzyme, peptide N-glycosidase F (New England BioLabs) was added to each of the tryptic digests and incubated at 37° C. for 18 hours to release the N-linked glycans. After enzymatic digestions, the samples were passed through a C18 reversed phase cartridge. The N-linked glycans from each sample were eluted with 5% acetic acid and then lyophilized
The lyophilized N-linked fraction of each sample was dissolved in dimethylsulfoxide and then methylated with NaOH and methyl iodide. The reaction was quenched by addition of water, and per-O-methylated carbohydrates were extracted with dichloromethane. The organic phase was concentrated to dryness and then dissolved with methanol for glycan structural analysis.
The profiles of N-linked glycans from all five samples were analyzed by electrospray ionization mass spectrometry (ESI-MS) using an LCQ-MS (Thermo Finnigan) quadrupole ion trap. Each sample (˜5 pmol/μL) was infused directly into the instrument at a constant flow rate of 1 μL/min via a syringe pump (Harvard Apparatus) and sprayed at 3.5 kV. A normalized collision energy of 35 and an isolation mass window of 2 Da was applied to obtain MSn.
The results are shown in Tables 7-10 and FIGS. 1-5. No table is provided for CA215 sample C, as apparently the sample was too small and defined peaks could not be obtained. In Tables 9 and 10, the highlighted areas represent structures that are found in CA215, but not in IgG or RP215.
In addition, it appears that sialic acids were not detected, again, possibly due to low sample size.

TABLE 7

Profile of N-linked glycans of human IgG by ESI-MS

Observed Mass{M + Na}	Charge state	Proposed Structure	Structure

1032	double	GlcNAc₄Man₃Hex₁Fuc₁

1134	double	GlcNAc₄Man₃Hex₂Fuc₁

1228	double	GlcNAc₄Man₃Hex₂NeuAc₁

1257	double	GlcNAc₅Man₃Hex₂Fuc₁

1315	double	GlcNAc₄Man₃Hex₂Fuc₁NeuAc₁

1350	double	GlcNAc₅Man₃Hex₂NeuAc₁

1408	double	GlcNAc₄Man₃Hex₂NeuAc₂

1437	double	GlcNAc₅Man₃Hex₂Fuc₁NeuAc₁

1495	double	GlcNAc₄Man₃Hex₂Fuc₁NeuAc₂

1618	double	GlcNAc₅Man₃Hex₂Fuc₁NeuAc₂

1836	single	GlcNAc₄Man₃Fuc₁

1866	single	GlcNAc₄Man₃Hex₁

1907	single	GlcNAc₅Man₃

Legend:

TABLE 8

Profile of N-linked glycans of RP 215 Mab by ESI-MS

Observed Mass {M + Na}	Charge state	Proposed Structure	Structure

1032	double	GlcNAc₄Man₃Hex₁Fuc₁

1134	double	GlcNAc₄Man₃Hex₂Fuc₁

1242	double	GlcNAc₄Man₃Hex₂NeuGc₁

1314	double	GlcNAc₄Man₃Hex₂Fuc₁NeuAc₁

1453	double	GlcNAc₅Man₃Hex₂Fuc₁NeuGc₁

1417	single	GlcNAc₃Man₃

1663	single	GlcNAc₄Man₃

1836	single	GlcNAc₄Man₃Fuc₁

1866	single	GlcNAc₄Man₃Hex₁

Legend:

TABLE 9

Profile of N-linked glycans of CA215 Sample A by ESI-MS

Observed Mass {M + Na}	Charge state	Proposed Structure	Structure

1047	double	GlcNAc₄Man₃Hex₂

1172	single	GlcNAc₂Man₃

1228	double	GlcNAc₄Man₃Hex₂NeuAc₁

1243	double	GlcNAc₄Man₃Hex₂NeuGc₁

1330	double	GlcNAc₄Man₃Hex₂Fuc₁NeuGc₁

1452	double	GlcNAc₅Man₃Hex₃NeuAc₁

1467	double	GlcNAc₅Man₃Hex₃NeuGc₁

1418	single	GlcNAc₃Man₃

1621	single	GlcNAc₃Man₃Hex₁

1663	single	GlcNAc₄Man₃

1836	single	GlcNAc₄Man₃Fuc₁

1866	single	GlcNAc₄Man₃Hex₁

1907	single	GlcNAc₅Man₃

Legend:

TABLE 10

Profile of N-linked glycans of CA215 Sample B by ESI-MS

Observed Mass {M + Na}	Charge state	Proposed Structure	Structure

1047	double	GlcNAc₄Man₃Hex₂

1134	double	GlcNAc₄Man₃Hex₂Fuc₁

1169	double	GlcNAc₅Man₃Hex₂

1228	double	GlcNAc₄Man₃Hex₂NeuAc₁

1242	double	GlcNAc₄Man₃Hex₂NeuGc₁

1315	double	GlcNAc₄Man₃Hex₂Fuc₁NeuAc₁

1330	double	GlcNAc₄Man₃Hex₂Fuc₁NeuGc₁

1366	double	GlcNAc₅Man₃Hex₂NeuGc₁

1438	double	GlcNAc₄Man₃Hex₂NeuGc₂

1525	double	GlcNAc₄Man₃Hex₂Fuc₁NeuGc₂

1580	single	GlcNAc₂Man₅

1663	single	GlcNAc₄Man₃

1785	single	GlcNAc₂Man₆

1837	single	GlcNAc₄Man₃Fuc₁

1907	single	GlcNAc₅Man₃

Legend:

Generally, in both human IgG and RP215, the major ion detected has a glycosylated structure (N/Z 1836). Although this was also detected in the CA215 samples, the signal was not as dominating as in human IgG and RP215. Other fucosylated and sialylated glycans were detected in all samples.
In addition, the O-linked sugar content of these materials was also determined with the results shown in Table 11.

TABLE 11

Monosaccharide Composition of O-glycans Analyzed by HPAEC.

			Molar
Sample name	Analyte	nmoles	percentage

Human IgG-Sigma	Fucose	0.0702	4.1
	N-acetyl-galactosamine	0.3360	19.7
	N-acetyl-glucosamine	0.7958	46.8
	Galactose	0.3717	21.8
	Glucose	nd	nd
	Mannose	nd	nd
	N-acetyl-neuraminic acid	0.1276	7.6
	N-glycol-neuraminic acid	nd	nd
	Total	1.7013	100.0
RP215 Mab #070801-A	Fucose	0.0692	6.2
	N-acetyl-galactosamine	0.2004	18.1
	N-acetyl-glucosamine	0.3569	32.2
	Galactose	0.3475	31.4
	Glucose	nd	nd
	Mannose	nd	nd
	N-acetyl-neuraminic acid	0.0838	7.6
	N-glycol-neuraminic acid	0.0501	4.5
	Total	1.1079	100.0
CA215 1 #070801-1	Fucose	0.0981	14.8
	N-acetyl-galactosamine	0.0961	14.5
	N-acetyl-glucosamine	0.1371	20.7
	Galactose	0.1473	22.3
	Glucose	nd	nd
	Mannose	nd	nd
	N-acetyl-neuraminic acid	0.1320	20.0
	N-glycol-neuraminic acid	0.0508	7.7
	Total	0.6614	100.0
CA215 4 #070801-4	Fucose	nd	nd
	N-acetyl-galactosamine	nd	nd
	N-acetyl-glucosamine	nd	nd
	Galactose	nd	nd
	Glucose	nd	nd
	Mannose	nd	nd
	N-acetyl-neuraminic acid	0.3512	100.0
	N-glycol-neuraminic acid	nd	nd
	Total	0.3512	100.0
CA215 5 #070801-5	Fucose	nd	nd
	N-acetyl-galactosamine	0.4961	34.2
	N-acetyl-glucosamine	0.3506	24.2
	Galactose	0.3539	24.4
	Glucose	nd	nd
	Mannose	nd	nd
	N-acetyl-neuraminic acid	0.2307	15.9
	N-glycol-neuraminic acid	0.0198	1.3
	Total	1.4511	100.0

nd = not detected.

EXAMPLE 4

Location of the Carbohydrate Epitope

Using Western blot, it has been demonstrated that the RP215-specific carbohydrate-associated epitope is localized in the Fab region of cancer cell-derived Ig heavy chain of CA215. Amino acid analysis of the CDR1, CDR2 and CDR3 regions of a number of immunoglobulin heavy chains were analyzed to locate the position of the glycosylation site. These comparisons are shown in Table 12.

TABLE 12

	Last 6	SEQ		SEQ		SEQ		SEQ
Cell lines	amino acid	ID		ID		ID		ID
or tissues	in FR1	NO:	CDR1	NO:	CDR2	NO:	CDR3	NO:

T47D	SRFSSR	114	TSGMR	115	PFWNGGSQKYCA	116	GITVPBPRLCPRN	117	IgG
					DSVT

ZR75-1	SGYSFT	118	SYWIG	119	IIYPGDSDTRYSPS	120	QBIVAFS	121	IgM
					FQG

ZR75-1	SGFNFN	122	TYAMT	123	TIAADGTWTSNA	124	DWYDY	125	IgG
					DFVRG

SKBR3	SGGSVS	126	SGYYY	127	YIYYNGSTYENPS	128	DIKHTYGPN	129	IgG
			WS		LKS

SKBR3	SGLSFS	130	S SGMN	131	RIGSKAASDTTSY	132	QGCGGDCHIPKM	133	IgA
					AASVRG

MDA-	SGFTFS	134	SYWMD	135	RISPDGRTTTYAD	136	GYLSSH	137	IgM
MB-231					SVEG

Lung	SGYTFG	138	TYWIG	139	IIYPGDSDTTYSPS	140	WDVMIGFYTA	141
Cancer					FRG

Dakiki	SGFTFS	142	DYGMT	143	GITSSVLTTYYAD	144	AQGFAPPAS	145
					SVKG

IM-9	SGFRFD	146	DYAMH	147	GISWNSDTIDYAD	148	TKEGGVTDIDPFDI	149
					SVKG

MC116	SGYRFT	150	GYYMH	151	RINPNSGGINYAQ	152	TREDSGSYEY	153
					RFQG

Daudi	SGYSIT	154	SYYIH	155	KTDNDGRDADYA	156	VRENGQKCFDY	157
					QRFQG

A consistent O-link glycosylation site with serine or threonine was always located proximal to the junction between FR1 and CDR1, thus indicating that the RP215-specific epitope is associated with the presence of a serine or threonine residue in this location. Absence of this O-glycosylation site results in the failure of RP215 to recognize CA215.

EXAMPLE 5

In Vivo Efficacy of RP215

Although addition of RP215 monoclonal antibody to cell cultures of OC-3-VGH ovarian cancer cells had no effect on growth, these antibodies were successful in inhibiting tumor growth in vivo. Cell culture growth was also not inhibited either by human IgG or goat anti-human IgG in vitro up to a concentration of 200 μg/ml in the cell culture.
Groups of four nude mice were implanted subcutaneously with 2×10⁶cells in 0.2 ml per mouse at sites near the breast for a growth period of 2-3 weeks. Treatments were performed after visible apparent growth of tumors. The experimental design is shown in Table 13. The radioactive labeling of the mAb was at a specific radioactivity of 12.5 μCi/mg.

TABLE 13

		Ani-
Exp.		mal
No.	Group No.	(n)	Dosage

1.	Negative Control	4	Medium only
2.	Positive Control	4	Cyclophosphonamide
			(60 mg/Kg)
3.	Antibody (High Dose)	4	RP215 Mab (10 mg/Kg)
	(Naked)
4.	Antibody (Low Dose)	4	RP215 Mab (2 mg/Kg)
	(Naked)
5.	I¹³¹-labeled Antibody	4	RP215 Mab (10 mg/Kg +
	(High dose)		125 μCi)
6.	I¹³¹-labeled Antibody	4	RP215 Mab (6 mg/Kg + 75 μCi)
7.	I¹³¹-labeled Antibody	4	RP215 Mab (2 mg/Kg + 25 μCi)

The mice were sacrificed on day 16 after treating with antibody, and the size of tumors in each mouse was determined by weight together with the body weight. These results are shown in Table 14.

TABLE 14

				Average
	Mouse	Body	Tumor	of Tumor
Group ID	#	Weight	Weight	Weight	Percent

Negative Control	1	22.23	0.148	0.13075	100
	2	22.51	0.133
	3	23.16	0.104
	4	21.14	0.138
Positive Control	1	21.18	0.098	0.09575	73.2
(Cyclophosphonamide	2	21.42	0.096
60 mg/Kg)	3	23.37	0.098
	4	21.18	0.091
Antibody	1	21.43	0.088	0.0865	66.2
(high dose	2	23.05	0.077
10 mg/Kg)	3	22.15	0.095
	4	23.56	0.086
Antibody	1	21.76	0.103	0.10225	78.2
(low dose	2	20.57	0.122
2 mg/Kg)	3	21.98	0.080
	4	22.64	0.104
I¹³¹-labeled	1	20.26	0.034	0.4575	35
Antibody	2	25.71	0.075
(High dose	3	20.25	0.048
10 mg/Kg)	4	22.60	0.026
I¹³¹-labeled	1	20.47	0.068	0.0705	53.9
Antibody	2	22.91	0.076
(Mid dose	3	23.00	0.049
6 mg/Kg)	4	23.05	0.089
I¹³¹-labeled	1	20.03	0.154	0.11675	89.2
Antibody	2	21.46	0.083
(Low dose	3	22.12	0.108
2 mg/Kg)	4	20.38	0.122

As shown, antibodies dosed at 10 mg/kg without radioactive label reduced the tumor size more significantly than the positive control which employed 60 mg/kg of cyclophosphonamide. The I¹³¹labeled antibody at the same dose reduced the tumor size even more.

EXAMPLE 6

Nucleotide Sequences of RP215 Variable Regions

For purposes of humanizing RP215, and as a target for mutagenesis, to obtain additional antibodies with favorable properties immunoreactive with CA215, the nucleotide sequences encoding the heavy and light chain variable regions of RP215 have been determined and are shown in Table 15, along with the deduced amino acid sequences.

TABLE 15

Nucleotide and the Deduced Amino Acid Sequences of the variable
regions of RP215 Monoclonal Antibody

	Length
Region	(bp)	Nucleotide Sequence

H Chain Variable Region- 19 amino acids

Signal Peptide (SEQ ID NO: 158)

1 atgagatggagctgtatcatcctcttcttggtagcaacagctacaggtgtcagctcc 57

M R W S C I I L F L V A T A T G V S S

H Chain Variable	1	caggtccaactgcagcagcctggggctgagcttgtgatgcctggg
Region (SEQ ID NO:159)		Q V Q L Q Q P G A E L V M P G
	46	gcttcagtgaagatgtcctgcaaggcttctggctacacattcact
		A S V K M S C K A S G Y T F T
112 amino acids	91	gactactggatgcactgggtgaagcagaggcctggacaaggcctt
		D Y W M H W V K Q R P G Q G L
	136	gagtggatcggagcgattgatacttctgatagttatactaggtac
		E W I G A I D T S D S Y T R Y
	181	aatcaaaagttcaaggacaaggccacattgactgtagacgaatcc
		N Q K F K D K A T L T V D E S
	226	tccagcacagccttcatgcagctcagcagcctgacatctgaggac
		S S T A F M Q L S S L T S E D
	271	tctgcggtctattactgtgcaagatccatctatgactggggccaa
		S A V Y Y C A R S I Y D W G Q
	316	gggactctggtcactgtctctgca 339
		G T L V T V S A

L Chain Variable	1	atggaatcacagacccaggtcctcatgtttcttctgctctgggta
Region-Signal Peptide		M E S Q T Q V L M F L L L W V
(SEQ ID NO: 160)	46	tctggtggtgcctgtgca 63
21 amino acids		S G G A C A

L Chain Variable	1	gacattgtgatgacacagtctccatcctccctggctatgtcagta
Region (SEQ ID NO: 161)		D I V M T Q S P S S L A M S V
	46	ggacagaaggtcactatgagctgcaagtccagtcagagcctttta
		G Q K V T M S C K S S Q S L L
112 amino acids	91	aatagtagcaatcaaaagagctatttggcctggtaccagcagaaa
		N S S N Q K S Y L A W Y Q Q K
	136	ccaggacagtctcctaaacttctggtatactttgcatccactagg
		P G Q S P K L L V F A S T T R
	181	gaatctggggtccctgatcgcttcataggcagtggatctgggaca
		E S G V P D R F I G S G S G T
	226	gatttcactcttaccatcagcagtgtgcaggctgaagacctggca
		F T T L T I S S V Q A E D L A
	271	gattacttctgtcagcaacattatagcactccgtccacgttcgga
		D Y F C Q Q H Y S T C S T F G
	316	ggggggaccaagctggaaataaaa 339
		G G T K L E I K

Using this information, alternative forms of monoclonal antibodies immunoreactive with CA215 were designed and produced, including humanized forms thereof.

EXAMPLE 7

Humanization of RP215

Using the information set forth in Example 6, humanized forms of RP215 (hRP215) were prepared.
In addition, a chimeric RP215 mAb (chRP215) was prepared using the same human constant regions and the murine variable regions set forth above in Example 6.
RP215, HRP215, and chRP215 were obtained in purified form and compared as follows:
Each was coated separately at 5 μg/ml overnight in microwells. The microwells were treated with concentrated cell culture shed medium of OC-3-VGH cells (to supply CA215) and then with HRP-labeled RP215 as a detection antibody. The detection antibody was incubated with each well for 60 min at 37° C. After 30 min, TMB substrate was added for 20 min of color development and, after stopping the reaction, the intensity determined at 450 nm in an ELISA reader. In some cases, 10 μg/ml human IgG was added to the wells, but this had no evident effect on the signal intensity. The results are shown in Table 16.

TABLE 16

	Relative Signal Intensity	Relative Signal Intensity
Coated	in Sandwich EIA	in Sandwich EIA
Antibodies	No Human IgG	10 μg/ml Human IgG

RP215
	100%	100%
hRP215	20.6%	21.4%
chRP215	14.5%	15.4%

As shown, hRP215 and chRP215 showed lower binding affinity, but nevertheless, were able to bind CA215 specifically and not affected by human IgG.
In additional experiments, RP215-coated microwells showed no binding to alkaline phosphatase-labeled goat anti-human IgG or to the Fab or Fc regions of this antibody. However, RP215 showed strong binding to goat anti-mouse IgG.
HRP215 showed little or no binding to goat anti-mouse IgG but strong binding to goat anti-human IgG. The results were similar for chRP215.
These results are shown in Table 17.

TABLE 17

Antibodies coated in Wells

Detecting		hRP215	chRP215	hIgG
Antibodies		Relative	Relative	Relative
Used	RP215	Intensity	Intensity	Intensity

Goat Anti
100%	~5%	~7%	~3%
mouse IgG^a
Goat Anti-	<0.5%	100%	~10%	70%
human IgG^a
Goat Anti-	<0.5%	150%	45%	100%
human IgFc
Goat Anti-	<1%	70%	10%	40%
human IgFab

^aSignal intensity for goat anti-mouse IgG and goat anti-human IgG were adjusted to 100% for comparative purposes

As shown in Table 17, wells were coated with either RP215, human RP215 (hRP215), chimeric human RP215 (chRP215) or human IgG. In row 1, the results were normalized to 100% for the interaction between RP215 and goat anti-mouse IgG. As shown in the first row, goat anti-mouse IgG bound comparatively poorly to the humanized or chimeric human forms.
In row 2, when goat anti-human IgG is used as a detector, and humanized RP215 was set at 100%, decreased binding was shown for human IgG, and very reduced binding for the chimeric or murine RP215.
In row 3, when the detecting antibody was goat anti-human IgFc, strong binding was shown to the humanized RP215 and to human IgG, but reduced binding to chimeric human RP215 and virtually no binding to RP215 itself.
In row 4, when goat anti-human IgFab was used, strong binding was detected in the humanized RP215 but relatively poor binding of human IgG and very weak binding, as expected, to RP215 and chimeric RP215.
From this, it was concluded that
(1) The humanized antibody, hRP215 and chimeric antibody, chRP215 showed comparable low crossreactivity to goat anti-mouse IgG to that of human IgG.
(2) Both hRP215 and chRP215 showed similar high binding activity to goat anti-human IgFc antibodies to that of human IgG as compared to the very weak binding of RP215, which is of mouse origin. chRP215, in contrast, has very low binding activity to goat anti-human IgFab.
These results demonstrate that the humanized antibody retains antigen binding specificity to CA215 and has human characteristics to the exclusion of murine characteristics.

EXAMPLE 8

Analysis of Cancer Cell Lines for the Presence of CA215

The presence of RP215-specific epitope(s) in more than 30 cancer cell lines from ATCC and others has been tested using Western blot assay and sandwich EIA. The following cancer cell lines were shown to be positive for the presence of RP215-related epitope (>90%) in cancer cell extracts and in cell cultured shed media.
Breast Cancer Cell Lines:
MCF7 (HTB-22), MDA-MB-231 (HTB-26), MDA-MB-468 (HTB-132), MDA-435, SW-48 (CCL-231), T-47D (HTB-133)
Cervical Cancer Cell Lines: C-33A (HTB-31), ME-180 (HTB-33)
Colon Cancer Cell Lines:
HCT 115 (ABM), HCT 116 (CCL-247), HT29 (HTB-38)
Liver Cancer Cell Lines:
Hep3B (HB-8064), HepG2 (HB-8065), Hep-2 (CCL-23)
Kidney Cancer Cell Lines:
293 (UBC)
Lung Cancer Cell Lines:
A549 (CCL-185), Calu-6 (HTB-56), H441(HTB-174), MRC-5 (CCL-171), WI-38 (CCL-75)
Lymphoma
HEL (ABM)
Melanoma
MMAN, MMRU, SK-Mel-3 (HTB-69)
Neuroblastoma
SH-SY5Y (CRL-2266), Neuro2A (CCL-131),
Bone Cancer Cell Line
U-20S (HTB-96)
Ovarian Cancer Cell Line
Skov-3 (HTB-77), OC-3-VGH (Taiwan)
Prostate Cancer Cell Line
DU 145 (HTB-81), PC-3 (CRL-1435)
However the existence of RP215-specific epitope cannot be easily demonstrated in several cancer cell lines. They are: SiHa (HTB-35, cervical), JEG-3 (HTB-36, placenta) and Jurkat (TIB-152, T-cell leukemia).

Claims

1. A method to elicit an immune response to cancer in a subject, wherein said cancer expresses CA215, which method comprises administering to said subject a formulation comprising a molecule that is immunoreactive with RP215 monoclonal antibody, but not significantly immunoreactive with antihuman IgG, said molecule consisting of an FR1 and CDR1 sequence of CA215 and a carbohydrate, wherein said carbohydrate is coupled to a threonine or serine glycosylation site proximal to the junction between FR1 and CDR1.

2. The method of claim 1 wherein said molecule is coupled to a heterologous moiety.

3. The method of claim 1, wherein the subject is human and which method further comprises treating said subject with chemotherapeutic agents or with radiation.

4. The method of claim 1, wherein the subject is an animal tumor model.

5. A method to elicit an immune response to cancer in a subject, wherein said cancer expresses CA215, which method comprises administering to said subject a formulation comprising antiidiotype antibodies to RP215 monoclonal antibody.

6. The method of claim 5, wherein the subject is human and which method further comprises treating said subject with chemotherapeutic agents or with radiation.

7. The method of claim 5, wherein the subject is an animal tumor model.

8. A method to detect CA215 in a sample, which method comprises

(a) contacting said sample with a solid support to which is bound an anti-human IgG, followed by contacting said support with the monoclonal antibody RP215 or an antibody cross-reactive therewith or with immunospecific fragments of either, that is labeled or modified subsequently to contain a label; or

(b) conducting a sandwich immunoassay in which one member of the sandwich is RP215 antibody and the other member of the sandwich is immunoreactive with antihuman IgG.

9. A protocol for diagnosis and treatment of solid tumors in a human patient which method comprises

a) diagnosing said patient by assaying a sample of body fluid from said patient for the presence or absence of CA215, wherein the presence of CA215 indicates the presence of a solid tumor in said patient; and

b) treating said patient by administering to said patient a humanized antibody that binds an epitope on membrane-bound CA215.

10. The method of claim 9, wherein the antibody is a humanized form of RP215.

11. The method of claim 9, wherein the antibody is coupled to a cytotoxic agent.