CA2026250C - Detection and quantification of neu related proteins in the biological fluids of humans - Google Patents

Abstract

The invention relates to a substantially purified p100 which is a human neu related protein having a molecular weight in the range from about 97,000 daltons to about 115,000 daltons which corresponds substantially to the extracellular domain of the human neu gene product, said protein being detectable in a biological fluid.
In another embodiment this invention relates to assays for detecting this protein.
Finally, this invention also concerns monoclonal antibodies which are capable of binding to p100.

Description

TITLE
Detection and Quantification of neu Related Proteins in the Biological Fluids of Humans FIELDOF THE INVENTION
This invention relates to a substantially purified human neu gene related product, p100, and, mare particularly, to detection and/or quantification of p100 in the biological fluids of humans using monoclonal antibodies which are capable of binding to this protein.
BACKGROUND OF THE INVENTION
Rat neuro/glioblastomas induced by transplacental injection of ethylnitrosourea carry an oncogene that is detectable by transfection into mouse NIH 3T3 cells. (Shih et al., Nature (London) 290: 261-264 (1981), Schubert et al., Nature (London 249: 224-227 (1974)). This gene was designated n~u.
(Schechter et al., Nature (London) 312: 513-516 (1984)). It was found that the neu gene was related to, but distinct from the gene that encoded the epidermal growth factor receptor (EGFR). The transfected NIH 3T3 cells displayed a novel 185,000 dalton tumor antigen (p185) that was not detected when the recipient cells were transformed by other oncogenes. (Padhy et al., Cell 28: 865-871 (1982)).

N~~t>~p~J

The human hom~log of the rat neu oncogene has bet~n isolated and termed c-.~B-2 or HER-2 on the basis of its close relationship to the human EGF receptor gene (also known as the c-g~H-1 gene). (Yamamoto et al., nature 319; 230-234 (1986), Goussens et ai., Science 230: 1132-7139 (1985)). The human neu protein has been rrpo:ted to have a slightly higher apparent molecular weig;i;. of about 190,000 daltons. (Gu111ck et al., Ix~t.
J. Cancer 90: 296-259 (1987)). The DNA sequences of the rat and human clones which have bean isolated predict a 1260 amino acid protein product of the »
gene that is colinear and about 50~ identical with the predicteo a.~.:iryo acid sequence of the EGF receptor. The level of sequence homology of the extracellular domains of the human EGFR and the c-~s-2 is about 93Rs.
HEFt-2/~ differs team EGFR in that it is found on band q21 Of chromosome 17 whereas the EGFR gene is located Of band pll-p13 of chromosome °7. The HER-Z/heu gene generates a messenger RNA (mRNA) of 9.8 kb which differs from the 5.8-10 kb transcripts for the EGC'H gene. Finally, the protein encoded by the HER-i/re~ gene has been found to have a molecules weight in the range from about 185,000 daltons to about 190,000 as ccmpareti tc the 1~/U,UUO dalton protein encoded by the EGFFt gfnw.
By analogy to the EGk'R, the ~ gene product dpr:ears to De a transmembrane protein consisting of a eysteine-rich ertracellular region of about 650 amino acids, a transmem:>rane domain, and an intracellular portion of about 580 amino acids consisting in part of a tyrosine kinasc domain.
Biochemical studies have revealed that the p185 F~r~tein is glycosylated and accessible to antisesa it intact cells, consistent with its lacaliaation at the z ' Z~~~~~~
cell surface. p185 appears to )~e the receptor for an as yet unidentified ligand.
While a single point mutation occurring in the transmembrane domain of the protein converting a valine residue to glutamic acid was s~esponsible for activation of the rat ,~eu oncoyeoe, no such mutation ha~ been found t0 oCCur in the human ~ gene. It is statistically unlikely that such d mutation would occur in the human homolog because two adjacent nucleotide changes would be heeded t0 generate the same mutation in the transmerr~rane domain of the human ~ gene. However, when the double mutation is induced in the human homolog, oncogenic activity is induced. This does not preclude the possibility that other point mutations could activate the human homolog.
The oncogenic potential of the human pgu gene can be achieved by a mechanism other than a point mutation. Unlike the rat L~u gene which is non-transforming at any level of expression unless mutated.
2G the human. neu gene is transforming, in the absence of any mutation, when overexpressed. Alteration of control of er.pression carp be accomplished by increased expression of the pre-existing gene or Dy increasing the number of copies of the gene (gene amplification). Gene 2~ amplification of C-~$-2 haS been identified in primary mammary adenocarcinomas as well as in a salivary gland adenocarcinoma. Researchers have found that the human r,eu gene is amplified relatively frequently in human breast cancer cell lines. n~. Was amplified 2 to 30 greater than 2U times in 30~ of breast tumors. The presence of neu amplification was a significant predictor of both overall survival time and time to relapse. (Slamon et al., Science X35: 177-1.8Z
(January 9, 1987)). Thus, these findings suggest the 3'~ possibility that neu overexpression, whether due to amplification or t4 SUJTIC VLher meChanl9m, contributes to m aplastic growth.
Because the human ~ protein appears to be involved in human malignancy, researchers have tried to study its expression ar,d structure fn human tissues.
aerger et al., Cancer Research 48: 1238-124,3 (1988), tried to correlate c-g~8-2 gene amplification and pr~t.ein expression with lymph node status and nuclear grading as well as with axillary lymph nade involvement. c-~B-2 specific antibodies generated frUm, a synthetic peptide corresponding to residues 1215-1255 of the C-g~.~B-2 open reading frame were used to analyze fifty one primary human breast tumors for amplification of the c-y~B-2 proto-oncogene.
Drebin et al., Nature 312: 545-548 (1984) de~cr=be the generation of monoclonal antibodies that react specifically with cell-surface determinants found on NIH 3T3 cells transformed by transfectian with a group of rat neuroblastoma oncogenes (the rat neu 2b oncogene).
Drebin et al., Cell 41:695-706 (July 1985), describe the rapid and reversible loss of both eeil-surface and total cellular p185 of NzH 3T3 cells transformed with the rat neu oncogerle which were exposed to mcr~oclonal antibodies reactive with the rat ~ gene product.
Drebin et al., Oncogene 2: 2~~3-277 (1988) describe monoclonal antibodies reactive with distinct domains of the rat X11 onCOgene-encoded p185 molecule whic:~ exert synergistic anti-tumor effects ,i.ri .
Drebin et al., Oncogene 2: 38?-394 (1988), describe monoclonal antibodies which bind cell surface domains of the rat ~ gene encoded product.
Er.pression of the C-~2B-l protein in normal 3~ ar~d transformed cells was investigated try Gullick et s al., Int. .1. Cancer 90: 296-254 (1987) using antisera generdt~~3 against two synthetic peptides fxom the predicted sequence of the human c-.~B-2 protein and a monoclonal antibody specific for the rat nEU protein.
Similarly, Ventes et al., The Lancet, ii, pages 69-72 (July 11, 1~8~) describe amplification of the human proto-ancogene c-X8~2 in 12 of 36 human breast tumors which was associated with increased levels of expression of the c-gds-2 protein, measured by i:rununohistological staining and by Western blotting.
Affinity-purified rabbit antibodies, raised to a peptide consisting of residues 1215-1225 of the open reading frame of the c-g~B-2 psvtein, were used in the immunohistological staining.
Tandon et al., Journal of Clinical Oncology, pages 11ZU-1128, Vol. 7, No. 8 (August 1989), describe a method using western blot analysis to quantitate tie HER-2/neu protein levels in 728 human breast tumor specimens for the potential prognostic significance.
Rabbit palyclonal antiserum was generated against the carboxy terminal synthetic peptide of the HER-2/~
protein (residues GTP'L'A~NYEYLGLDVPV from the deduced amino acid sequence) f~i use in the study.
Akiyama et al., Science, pages 1544-1646, vol.
23? (June 27, 1986), also described raising antibodies against a synthetic peptide corresponding to 19 amino acid residues at the Carboxy terminus of the deduced amino acid sequence from the human c-X112$-2 nucleotide sequer:ce. The antibodies were reported to immunoprecipitate a 185,000 dalton glycoprotein from NKN-7 ddenoGarcinoma tells.
Most xecently, regulation of the tyrosine kinase activity of the epidermal growth factor receptor by a truncated receptor of 100 xilodaltons containing a5 the EGF-binding site but not the kinase domain was ity~wlL~l Ly Daau ct al. in Mnloc»1ar and C611~.~1ar Biology, pages 671-677 (February 1989). It was deSCribed that structurally related receptor kinases, such as the platelet-derived growth factor receptor, the ins~.~lin receptor, and Lhe neu receptor, were not ir~hit~ited by the truncated 100 kDa receptor.
~QF_ THF ~N~CE~3.~
This invenr.ic~n relates to a substantially 1G purified p100 which is a human ~ related protein having a molecular weight in the range from about 97,000 da:.t~ns to about 115,000 daltons which corresponds suhstantiaiiy Co trse extrace11u1ar domain of the human f.~ gene product, said protein being detectable in a 1~ biological fluid.
In another embodiment this invention relates to assays for dEtecting this protein.
Finally, this invention also concerns monoclonal antibodies which are capable of binding to l~ p~GO.
F3Ry,F DESCRIPTION OF 'THE EI~tJ
Figure 1 is a schematic representation of plasmid vector pLJdelta neu, created by inserting the ?.5 human ~ gene cDNA indicated onto the pLJdelta expre;;sion vector.
Figure l presents immunoprecipitation seeults indicating that the monoclonal antibodies described herein recognized a human neu related protein.
30 Figure 3 presents imQnunoblot results fox monoclonal antibodies OD3 and P83.
Figure 4 presents immunofluorescent results.
Figure 5 presents slow cytometric results.

Figure 6 s)ivws Lhe binding curves when TA-1 and N13-3 were used to Compete with the binding of biotinylated TA-1.
figure 7 shows the k~inding curves when TA-1 and NB-3 were used to compete with the binding of biotinylated N8-3. ' Figure 8 shows results of a capture inununoassay in which lysates fx.om a variety of human breast carcinoma cell lines were tested for the presence of the 11~d. related protein using the capture ELaSA
system. An anti-human neu monoclonal antibody tTA-1) was used to capture neu related protein and anti-human nPu monoClondl antibody, NA-3, was biotinylated and Cietected using Streptavidin-horseradish peroxidase (SA-HRF' ) .
Figure 9 shows results (microgram of tumor lysate vs. optical density) of a comparison between tumor lysate of a nude mouse tumor expressing ~,t1 (X-3-5) and tumor lysate of a zlg3i negative tumor (3T3/ras). The assay was performed with an anti-human L~, monoclonal antibody designated NB-3 as the capture reagent and the anti-human t1e.11 monoclonal antibody TA-1 was biotinylated and detected using SA-HRp.
Figure 1D shows the results of a capture immunoassay of cEll lysates prepared Lrom a normal piece of human breast tissue iz~47-O1-050) or a breast carcinoma (2?4~-O1-D5U) and tested for the presence of related protein using the Capture foLmat. The assay was performed with an anti-human ~d monoclonal antibody designated TA-1 as the capture,reagent and the monoclonal antibody BD-5 was coupled to biotin and detected using SA-HRP.
Figure 11 shows results of a capture arr~r~ur~oassay of supernatant fluids from 18-3-7 cells 3r (N1H 3T3 cells transformed with the human I~ genQ $nd 2i~~~~~~

c>;pr~reFing the p185 trot e; n nn the cel l surfdCe) , 3T3 ras Ce115 (NIH 3T3 cells transformed with the ras gene and not expressing the human p185 protein an the cell SurtaCe), and Stc-8F-3 human breast carcinoma cells arid Culture media-DMEM supplemented with 10% fetal Galf serum. In the capture immunoassay, an anti-n~
man~c:lonal antibody, Ns-3, was used as the capture reagfnt and biotlnylated anti-human neu monoclonal antibody, TA-1, was used as part of the detection , system.
Figure 12 s)rows results of a capture irrununoassay using sera from mice bearing tumors derived from T194 wrrich 15 a ras-~ransformed NIH 3T3 cell line derived from an activated ras gene obtained ~rom breast 75 carcinoma Dell line HS0578t, sera from mice bearing tumors expressing the p185 protein (18-3-7 mouse), and sera from mice bearing tumors not expressing the p185 protein (3Tj (ras)). These seza assayed using an anti-human reu monoclonal antibody, TA-1, as the capture 2(; antibody and a biotinylated anti-luuman ~,yt monoclonal draibody, BD-5, was used as part of tire detection system.
Figure 13 shows the results of a capture imtr~unoassay in which the antihuman neu monoclonal 2'_~ antibody TA-1, was used as the capture reagent and biotinylated BD-5 was used as part of the detection system. Samples for analysis included normal human plasma and plasma from two breast carcinoma patients.
Figure 14 presents immunoprecipitation results 30 showing the relationship of Y100 to p185.
Figures 15°18 present immunoblot results showing detection of p100 in human plasma samples obtained from breast, gastric, and ovarian carcinoma patients.

~Q~~~ ~~

~TATk;MF,N~'_ QF n~
The fVllowing hybridoma cell lines were deposited with the ArueLiGan Type Culture Collection (ATCC), 12305 Parklawn Drive, Rockville, Maryland, 20852 under the Buciapesc Treaty and were accorded the following accession numbers:
Hybridoma Cell lire f3D5-2d was accorded ATCC
Accession number HB 9689 and deposited on April 18, 1988.
IU Hybridoma cell line OD3 was accorded ATCC
Acce~~ior~ number Hs 10204 and deposited on August 11, 1989.
Hybridoma cell line Na-3 was accorded ATCC
Accession number HB 10205 and deposited on August 11, 1989.
tiybridomd Cell line TA-1 was accorded ATCC
Accession number HB 10206 and deposited on August il, 1989.
ZO J?ETAIL~L), DESCR1PTIOh OF THE It~TVENTI~N
The terms human ~, e-~B-2, HER-2/n~li, and Y.x:n-2 are used iuLerchangeably herein.
The term oncogene, as used herein, refers to a ger';e a'terEdr in some fashion or by some mechanism, such 2;; t:~a~ 1t contributes to conver.Ling a normal cell to a cancer cell. For example, the rat tl~la oncogene appears to contribute to malignancy in rats through a point mutation Occurring in the transmembrane domain. On the other hand, the term human neu oncogerre has been used to :iU neudescribe the human n proto-oncogene which is believed to be altered in Some way to contribute to malignancy in humans through ever-expression of the human ,per gene product. It has been shown that overexpression of the normal human c-~B-2 protein loads to transformation of ~~W w~3~~
N1H/3T3 Cells. (Di Fiore et al., Science, 237:178-182 (1987) ) .
The term "human neu gene product'° refers to a growth factor receptor-like giycoprotein having an 5 intracellular tyrosine kinase damain, a transmembrane domain arid an extracellulax domain which is produced by the human neu gene. This protein has been reported to nave a molecular weight of about 185 or 190 kilodaltons.
The abbreviation "p185" will bE used interchangeably 10 herein with the term human neu gene product.
The term "substantially purified" means SyriLhesized or, if naturally occurring, isolated free of other cellular components with which it is normally associated.
While the function of human c-~H-2 protein in normal growth and differentiation of cells remains unknown, it appears that the increased expression of a grokth factor-receptor like protein encoded by the human c-~Yr B-2 gene might play an important role in the ~'v initiation or progression of neoplasia.
One of the important aspects of this invention concerns a human ~ related protein which is detectable in a t:uTan biological fluid such as blood, serum, plasma, urine, cerebrospinal fluid, supernatant from norrr,al cell lysate, supernatant from preneoplastic cell lysste, supernatant from neoplastlc cell lysate, supernatants from carcinoma cell lines maintained in tissue culture, and y east aspirates.
More specifically, this invention Concerns a 3G substan;iz;ly purified p100 which is a humar5 ~eu related protein having a molecular weight in the range from ' about 97,000 daltons to at~out 115,000 daltons which corresponds substantially to the extracellular domain of the human nEU gene product, the protein being detectable :i', ire a biological fluid such as those described above.

ThC torm "corn-espondz substantially" provides for conservdtive additions, deletions and/or substitutions.
It i.s believed that p100 is a cleavage/degradaLion product of p185. However, it is also possible L)rat p100 may be independently Synthesiz~:d. 1t is also possible that p100 may be furtl'~eL modified and/or cleaved.
zhe molecular weight range of p100 was determined using an immunoblot format or immunoprecipitation format as described below.
This invention also concerns a method of detecting preneoplastic or neopiastic cells in a human which comprises testing a biological fluid from a human for the presence of a p100 by:
(a) Contacting the fluid with at least one mor.oc'_onal antibody which is caYable of binding the protein, and (b) determining whether antibody binding has occurred.
In another embodiment Ltuis invention concerns a;. immunoassay for detecting or quantifying the presence cf p:00 whicri comprises (a) reacting the fluid with at least one first mo:~oclonal antibody which is capable of binding to F~100:
(b) reacting the product of step (a) with at least one delectably-labeled second monoclonal antibody whic:~ is capable of binding to p100 at an epitope different from the epitope bound by the first antibody;
and (c) detecting or quantifying the product of step (b) .
The monoclonal antibodies which can be used to detect plDO constitute another aspect of this invention.

Immunore.~orivc fragments of these antibodies Can also be Used to practice the invention.
These monoclonal antibadies or immunoreaCtive fragments thereof are specific for the extrace11u1ar domain of p185 as is shown below. Briefly, these antibodies were generated by immunizing mice using a protocol consisting of administering an NIH 3T3 cell line, cyclophosphami9e, and a transfected NIH 3T3 cell line which expressed the full length human neu gene product. This pra~edu~.e is set forth in greater detail below.
In accc~rdarce with this invention, an antibody cr cocktail of antibodies can tae used fox detection.
These antibodies can be labeled directly with a reporter or indirectly with a member of a specific binding pai=~
using conventional techniques.
Specific binding pairs can be of the immune or non-immune type. Immune specific binding pairs are e~:er;~plifiea by antigen-antibody systems of hapten/anti-?G hapten systems. There can be mentioned fluoresceinlanti-fluorescein, dinitrophenyl/anti-dinifrophenyl, biotin/anti-biotin, peptide/anti-peptide and the like. The antibody member of the specific bin~:iin7 pair can be produced by customary methods , 2~ farr~Sliar to those skilled in the art. Such methods involve immunizing an animal with the aaatigen member of the specific binding pair. If the antigen member of the specific binding pair i5 not immunogcnic, e.g., a hapte.~,, it can be covalently coupled to a carrier 3U protein: :~ render it immunogenic.
Non-immune binding pairs include systems wherein the two components share a natural affinity for each other but are riot antibodies. Exemplary non-immune -pairs az'e biotin-streptavidin, intrinsic fmctos-vitamin 812 folic acid-folate binding protein and the like.

2~~~'~ ~~

A variety of methods ire available to covalently label anLik~odies with members of specific binding pairs. Metal ds are selected based upon the nature of the member of the specific binding pair, the type of linkage desired, and the tolerance of the antibody to various conjugation chemistries. Biotin can be Covalently coupled to antibodies by utilizing commercially availa)alE active derivatives. Some of these are biotin-N-t~yd~oxy-succinimide Which binds to 1C amine groups on proteins; biotin hydrazide which binds to carbohydrate moieties, aldehydes and carboxyl groups via a carbodiimide coupling; and biotin maleimide and iododCetyl biotin k)~ich bind to sulfhydryl groups.
Fluorescein can be coupled to protein amine groups using 1°_: fluores~ein isothiocyanate. Dir~itrophenyl groups can be coup:ed to protein amine groups using 2,9-dinitrobenzene su7tate or Z,4-dinitrofluorobenzene. Other standard methods of conjugation can be employed to couple monoclonal antibodies to a member of a specific binding 2~~~ pair including dialdehyde, carbodiimide coupling, homofu:~ctior,al crosslinking, and heterobifunctional riosslinking. Cdrbodilmide Coupling is an effective method of coupling Carboxyl groups on one substance to arf,_r~e atvups on another. Carbodiimide coupling is ?.5 facilita:ed by using the commercially available reagent 1-ethyl-3-(dimethyl-aminopropyl)-carbodiimide (EDAC).
Nomobifunctional crosslinkers, including the bif~~nctional imidoesters and bifunctionai N-hydroxy-surcinimidc estErs, are oommercjally available and are 30 er~ployed ;or coup:ing amine gr,pups on one substance to amine groups on another. Heterobifunctional crosslinkers are reagents which possess different fvnr~t Tonal groups. The most Common commercially available heterobifunctional crasslinkers have an amine :,~: reactiva N-hydroxysuccinimide ester as one functional 2'~,'~2:~~

gr:pup, and a su7fliydryl reactive group as the second functional group. The most r:o~runon sulfhydryl reactive groups are maleimides, pyzidyl disulfides and active halvg~ris, Pne of the functional groups can be a plootoactive aryl nitrene, which upon irradiation reacts with a variety of groups.
The detectably-labeled antibody, detectably-labeled antibodSes, or detectably~-labeled member of the specific binding pair is r:oupled to a reporter which can 1U be a radioactive isotope, enzyme, fluorogenic, Ghe;niluminescent or electrochemical materials. Two Commonly used radioactive isotopes are 1251 and 3H.
Standard radioactive isotopic labeling procedures include the chlaramit~e T, lactoperoxidase and Bolton-Hun:.er metriods for 1251 and reduction methylation for 3H.
Enzymes suitable for use in this invention include, but are not limited to, horseradish peroxidase, , dikaline phosphatase, ~i-galactosldase, glucose oxidase, 2~: luciterase, (~-lactamase, mease and lysoxyme. Enzyme laY>eling is facilitated by using dialdchyde, carbodiimide Coupling, homobifunCtional crosslinkers and heterobifunctional crosslinkers as desczibed above for coupling an antibody with a member of a syecific binding pail.
The labeling method clr~sen depends on the functional groups available on the enzyme and the material to be labeled, and the tolerance of both to the conjugation conditions. The labeling method used in the 3G I>resent invention can be one of, but not limited to, any conventi~~nal methods currently employed including those described by Engvall and Pearlmann, Immunochemistry g, $71 (1971), Avrameas and Ternynck, Immunochemistry ~.
1175 (1975), zshikawa et al" a. Immunoassay 4(3):209-'x:. 327 (1983) and Jablonski, Anal. 8iochem. 198:199 (1985).

2~2~~~~~
l.abelinr~ ran be accomplished by indirect methods such as using spacers or other members of Specific binding pairs. An example of this is the detection of a biotinylat,ed antibody with unlabeled 5 streptavidin and biotinylated enayme, with streptavidin and biotinylated enzyme being added either sequentially or simultaneously. Thus, according to the present invention, the ar~~.~body used to detect can be dF~tectdbly-labeled directly with a reporter or lU indirHCtly with a first member of a specific binding pair. When the antibody is coupled to a first member of a specific binCiing pair, then detection is effected by xeacting the antik~udy-first member of a specific binding compeer, with the second member of the binding pair which 7~ is labeled or unlabeled as mentioned above.
Moreover, the unlabeled detector antibody can be detected by reacting the unlabeled antibody with a labeled antibody specific for trre unlabeled antibody.
Such an anti-antibody can be labeled directly or :U indirectly using any of the approaches discussed above.
:cr example, the anti-antibody can be coupled to biotic which is detected by reactir~rg with the streptavidin-horseradisr: peroxidase system discussed above.
Une of the preferred embodiments of this invention utilizes biotin. The biotinylated antibody is in turn reacted with streptavidin-horseradish peroxidase comple;:. Urthophenylenediamine, 4-chloro-naphthol, or tetramethylbenzidine ('rMB) can be used to effeet chromogenic detection.
The preferred immunoassay format far practicing this invention is a farward sandwich assay in which thF capture reagent has been immobilized, using conventional techniques, on the surface of the Support.
Suitable supports used in assays include 3~ synthetic polymer supports, such as polypropylene, ~0~~~~~

polystyrene, suLstituted polystyrene, e.g., aminated or carboxylaLed polystyrene; polyacrylamides; polyamides;
Yulyvinylchloride, etc.; glass beads: agarase;
nitrocellulose, etc.
T2'» examples discussed below are intended to illustrate the invention and should not be construed 8s limitations.
lU ~'x'~s~~J7..1'f neu-Syecific."~~i~oclonal p~ntibodias R. F'rodvILQ~ Hyhrido The hybridomas described below were generated by immunization of mite with viable cells tthe 18-3-7 Celi iine described below) which express the full length protein encoded key the human I~u gene. i.e., p185.
Using the full length protein presented by viable cells a5 the immunogen, it is Yussible to generate a election of monoclonal antibodies with specificities that. span the entire length of tree extracellular domain ' of the protein. This is as Opposed to the use of peptide immunogens, or short polypeptides generated by l:roY,aryotic systems, which present only a limited number of epitopes from the original protein, and hence rare o:: ir,~mune resYurnse Gf limited specificities.
2.'_~ F'artt~ermore, by presenting the protein antigen in what is believed to be its native state, the immune ~ystem will bE responding to an antigen which most closely resembles that which will be seen when the antibodies are later used 2or diagnostic or therapeutic 30 ap~~licatmns.
B. ~.t3Q ~~f ~ 8-3-7 ~~:11&
18-3-7 cells are a transfected NIH 3T3 tell line that express lull length normal human ~t protein.
The human nEU gene is expressed by a Murine leukemia 3~~ virus LTR tpromoter and enhancer). This cell line ~~2~'~~~

ehhibits all the characteristics of transform~~d NIH 3T3 cells. They grow in soft agar, form tumors in nude tniGC, and display altered morphological characteristics.
This Cell line was used as the immunogen fos the isolation of anti-human ~y speCi~ic monoclonal antibodies.
T!1P pi.,7 rPtrovitdl vector was modified to remove the polyoma early region, thereby eliminating the endogenous transforming activity of the pLJ vector.
GOnStruCtion of the modified vector fs shown in Fig;lre 1. The modification was accomplished by restricting pLJ with Apa I and isolating the 6300 base pa:.= fragmeirL, and reoircularixing it with Tq ligase.
The resulting plasmid tpdelta LJ or AbT 5009, shown in :figure 1) was digested at the unique Bam HI sate, filled with Xlenow, and ligated to a Xlenow treated r:c:ol-tiind111 fragment containing the entire human gay protein coding region. The resulting plasmid (pdelta LJ
n~u or pAbT 5'», shown in figure 1) was transfected into 2C NIH 3T.5 cells by the calcium phosphate precipitation procedure. Transfected cells were selected in 6918 (pdelta LJ has a SV40 promoted neon gene). The colonies were Screened for .oeu expression by RNA dot blots. 18-3-7 was one o: the highest expreJcOrV out of appror.imately SO screened.
C , lrtu_yLi Zat ion of Mice Two adult female BalD/c mice were immunized intraperitoneally (I.P.) with 1.9 X 106 viable NIH 3T3 Cells per animal. This was followed immediately by an 34 I.F~, injection of cyciophosphamide in H20, 30 mg/kg.
The cyclopt~osphamide treatment was repeated 29 and 96 hours after the primary injection. On day 14 following immunization, the mice were injected I.F. with 1.5 X 106 viable 18-3-7 cells. The animals were allowed to rest far another 14 days, at which tame the entire 2~2~~~~

Seq;~c~ncB Of injecting NIH 3T3 calls, cyclophosphamide, and 1B-3-7 cells was repeated. Four days following the s-r_cnnci injection of 1B-3-7 Cells, the animals were Sacrificed and their spzeens obtained for the first fusion. A second, identical experiment was performed, in four female Balb/C mice and four female CB6 , (nalb/c X C5~BLl6) nice, using 1.8 x 106 N1H 3T3 cells, and 9.B X 106 18-3- cells pez mouse in first round, and 8.5 X 106 NIH 3'1'3 Ge5ls and 2.7 X 106 18-3-7 cells in 30 the second round of immunisations.
~1y1_nridoma Methodoloav Hybridomas wore produced by fusing cells from immuni2ec~ mice with SP2/0 myeloma cells (ATCC CRL 1518) key a polyethylene glycol (PEG) method. Spleens were 75 removed asceptically from inu"unized mice, and a single cell suspension of the spleen cells was obtained by pexfusing the Spleen with serum-free media (DME).
Spleen cells and SP2/0 cells (harvested from a log phase growth Culture) were mixed togetheL at a ratio of 5:1, 20 spleen Cell:myeloma cell. The cells Were Centrifuged at 2U0 X g for 70 minutes at 9°C, an d the supernatant removed by aspiration. After loosening the cell pellet by gently tapping the bottom vi the tube, 1 ml of sterile, 37°C, l0a P~G in UME was added dropwise. The 2''~ tube was gently swirled while adding the PEG ever a 1.5 minute period. An additional 10 ml of 3"~°C
serum-free DME was then added dropwise, followed by another 20 ml of media. The suspension was tner~
cenC.rifuged at 2U0 X g for 10 minutes at room 30 temperature. Media was aspirated from t):e cell pellet, and media containing peritoneal macrophages (2 X 104 cells per ml) in the presence of 20% fetal calf serum, 0.2 mM hypoxanthine, 0.4 uM aminopterin, and 0.032 mM .
thymidine (HAT media) was used to resuspend th6 Cell 3~ pellet. (Peritoneal macz'ophages were obtained from unirtununlzed mice, eiLtm salb/c or CH6, depending on which Spleen cells were used for fusion. These cells :ere obtained by injecting and immediately removing Serum-f nee rnCtiia into the peritoneum of euthanized animals.) The post-fusion cells were resuspended In a final Cell concentration (not including the peritoneal macrUy?'~dgeS) of 5 X 105 cells/ml. One milliliter of this cell mixture Was distributed to each well of 24 well plates.
E. ELISA Pz~,Fdnre and Preliminary SC,~Ctlinc Hyk~ridc>mas which grew often the fusion procedure were initially screened for the secretion of anCi-human r,e~.: antibodies by an ELISA assay on a cell lysa;.e of 18-3-~ cells. Lysates were prepared by ir~cuY~ating freshly harvested 18-3-7 cells in the presence or a hypotonic lysis buffer (10 mM Tris, 10 mM
KC1, 5 cnM ~J'1'A, pH 8.0) followed by the addition of 'lvitor. X 100 to a final concentration of one percent. A
lysa=a of Nor 3T3 cells wa-s prepared similarly for use "; as a negative control. Microti.ter plates (Nunc, mrr,vnoplate#II) were i:vated overnight at room ter-,perature witt-~ 50 )tl of lysate, at a total protein concentration o_' S00 ~1g/ml. After aspirating to remove unr~ound antSoen, EL:SAS were performed by first '2t i:iCUbaLir~a 50 ~tl of culture supernatnnt obtained from the viable hybridoma colonies in the antigen-coated microtiter wells. A 3 h~uz incubation at 37°C was fo_lowed by 3 washes with a washing buffer (0.05'1 Tween 2C, 20 mM Tris, pH 7.6) and then a one hour 30 incubation at s7°c: witri 50 ~tl horseradish peroxidase labeled goat anti-mouse IgG +'igA ~ ZgM (HRP-GAM-GAM).
The wells were again washed t~~ree times with washing buffer, and the assay was developed by the addition of 50 ~tl of a tetramethylbenzidina (TM8) solution. This 3~ s~.lutior~ was prepared by dissolving 10 mg of TM8 in 1 ml # Trade Mark 19 Zo of dimethylsulfoxide (DMSO), arid adding 100 ~tl of this solution to 5 ml of TMB buffer (0.1 M sodium acetate, to pH 6.0 with 0.1 M Citric acid) along with the addition Of 10 X11 Of 3b )rydrogen p~;roxide. Color was allowed to develop for 5 minutes, at which time the enzymatic raaction was stopyed by adding 50 ~1 of 2 N HZSO4. The optical density (OD) of the resulting yellow color was rCad at 950 nm on a micrvtitex plate reader. A po;.itive reaction, as indicated by a greater yellow color developed on 18-3-7 gall-canted wells than on NIH 3T3 cell-coated wells, signaled that there was antibody present in the culture supernatant which recognized the hurr,an ,~eu gene product .
F. ~uhc7Qnin$~ybrid.S?m~.
Hybridomas which yielded positive results upon initial screening were expanded and cloned by limiting dilution to assure that the cells and resulting antibodies were indeed monoclonal. A ~eeder cell population was first prepared by r~btaining thymocytes 24 front; S week old unimmuni2ed mica, and making a single cell suspension at a concentratiUrl of 2 X lO4 cells/ml in HAT media. Hybridoma colonies which tested positive for the presence of antibody to the human neu gene product were diluted in the media containing thymocytes 2~ to a concentration of 5 hybridoma ceils/ml. Two hundred microliters of this solution was then delivered to each well of 96 well microtiter plates, Once colonies had grown, the supernatants were again tested far the presence of antibody to the human neu gene product. If 34 the results were positive when tested by the ELISA assay as described above, the colonies were cloned by limiting dilution a second timQ.
Hybridomas which were obtained in the manner described above following the fizst fusion secrete 35 monoclonal antibodies which have been designated Hb5-2d, TA-1-lC, RC1-9c, NA3-6a, and OD3-lOj. Following the socond fusion, l~ybW domas wire obtained which secrete antibodies named PH3, RC6-2, NH-3, IDS, and IE3-4.
G. tiTlL.~3?~S~Y._Is.QxY~'..~-~~_~t~.'~~s.~rmlrnation ELISA assays were performed to determine the isotype and light chain class of the antibody produced by the hybridomas, and to determine the IgG subcla~s.
For this purpose, a kit was purchased from 8oehringer Mannheim (Indianapolis, IN) which contained all of the necessary immunoreagents. Tissue culture supernatants obtained from t?re cloned hybridoma colonie~ were incubated on lysates of 18-3-7 cells as described above.
This was fo110wed by ari incubation with goat antisera specific for mouse immunogiobulin isoLypes, light chain classes, and 1gG subclasses, and then with horseradish peror.idase labeled with swine anti-gnat IgG as the second antibody. The assay was developed using ABTS
(2,2'-azino-bis-f3-ethylbenzthiazoline-6-sulfonic acid)) as per the manufacturer's instructions, and the OD of 2C the resulting green color was read at 405 nm.
Using this method, it was determined that 3 of the monoclonal antibodies from the first fusion, DD5-2d, RC1-4c, and TA-1-1C. are IgGl/kappa antibodies, and NA3-ba, and OD3-lOj are IgM/kappa antibodies. The monoclonal antibodies RC6-2, NH-3, IDS, and IB3-4 obtained from the second fusion are IgGl/kappa and the antibody PB3 iw IgG2a/lambda.
H. Radioit)1_munp~reci~itat~on ImmunopreCipitation of radioactively labeled 18-3-7 cells was done usiry each of the monoclonal antibodies to determine Whether the antibodies recognized a protein of 185 kd molecular weight, the expected molecular weight of the human neu gene product.
A near confluent monolayer of 18-3-7 cells (or NI~i 3T3) cells in a 10 cm petri dish was incubated overnight in media containing 500 ~Ci of 355-ldveled cysteine. The CC11S were harvested the following morning, and were Iyscd in rr dcter9ent buffer (IP buffer: 1~ ~'riton X-100; 1~ sodium d~oxycholate, 0.1~ SoS, 10 mM Tris, 0.65 M NaCl, pH '1.2) containing the protease inhibitors YN,SF and soybean trypson inhibitor. Approximately 1 )tCi or the labeled cell preperaticrn was then incubated nvPrnight dt 9°C with 500 y11 of culture supernatant from each of the hybriC~Urnas. During this incubation period, 50 ~tg of purified rabbit anti-mouse ZgG (Kirkegaard t Ferry Labs) was mixed with 50 ~tl of a 1:1 slurry of Protein A-Sepharose#(Pharmacia) in IP buffer overnight at 9°C. The excess Labbit antibody was removed by washing the Protein A-Sepharose once with IP buffer, and the slurry was then added to the incubation mixture containing the labeled cells and the monoclonal antibody. This mixture was allowed to react overnight at 4°C;. The Protein A-Sepharose was--pclleted by centrifugation and. was-washed-font-t.iiries with IP rbvffer, followe3 by one wash with-:TBS.~10=mM--Tris~=-150 mM NaCT, pH 8.2), and the pellet Was allowed to dry. Each pellet was resuspended in 50 N1 of sample Luffer for SDS gels ( 10 r~s: T: is, 45. SDS, 20t glycerol, lOt 2-mercaptoethanol, C.04~: bromphenol k~lue) . One-half of Cast. of the samples was run On SDS p~lyacrylarnide gels, using a 4.5~ acrylamide stacking gel, and a separating gel. The gels were dried end then autor3diographed.
Results of the immunoprecipitations indicated that all of the monoclonal antibodies recognized a protein of approximately 185 kd molecular weight (p185) ire the 18-3-7 cells which was not present in the NIH 323 cells. This was determined by the presence of a dark band on the autoradiograph which Corresponded to the distance travelled in the gel key a 185 kd molecular * trade mark z3 weight protein as indicated by standard protein markers.
R Similar experiment was done using SKeR-3 cells(a human breast carcinoma) and A431 cells (a human epidermoi~i carcinoma). The SKBR-3 cells have been shown by other investigators tv expjess high levels of the human ngy gene pr~duc;L, and immunoprecipitations with the monoC10na1 antibodies described above yielded confirming results. The band observed migrated the same distance as the r~~rnd which was precipitated from the labeled 18-3-7 ceW S. Fused upon these experiments and the immunoblot analysis described below it was concluded that the monoclonal antibodies raised to the 183-7 cells were specific for the huinan n,~ gene product, and did not cross react With human EGFR.
kiqure 2 presents immunopreeipitation rc~ults.
Panel A: 7:mmunopiecipitation of 18-3-7 and NIN/3T3 cell lysates with the IgG monoclonal antibodies.
La;~e 1 contains molecular weight standards. Lanes 2, 4, 6, 8 and 10 contain 1$-3-7 lysates, and Lanes 3, 5, ~, 9 2Ca and 77 contain NIH/3T3 lysates. Lanes 2 and 3:
precipitation with TA1. Lanes 9 and 5: precipitation with BD5. Lanes 6 and l: precipitation with N83. Lanes 8 arid 9: precipitation with P83. Lanes 10 and 11:
precipitation with MaPC-21.
Panei ti: Immunoprecipitation of 18-3-7 and NI:;/:iTs cell lysates with the IgM monoclonalS. Lane 1 contains molecular weight Standards. Lanes 2, 9 and 6 contain 18-3-7 lysates~ Lanes 3, 5 and 7 contain NIIif3T3 lysates. Lanes 2 and 3: prec~.i.vn with Ob3. Lanes 3U 4 and 5: precipitation with NA3. Lanes 6 and precipitation with TEPC 183.
Panel C: Immunoprecipitatiur~ of SISBR-3 and A-431 cell lysates with 1gG monoClonals. Lane 1 contains molecular weight standards. Lanes 2, 4, 6, 8, 10 and 12 3' Contain SKBR-3 lysates. Lanes 3, S, 7, 9, 11 and 13 2~~

e;Urit.ain A-431 lysates. l,ar~es 2 and 3: precipitation with TA1. Ldnes 4 and 5: precipitation with HDS. Lanes 6 and 7: precipitation with PB3. Lanes B dnd 9:
precipitation with NB3. Lanes 10 and 11: precipitation with MOPC 21. Lanes 12 and 13: precipitation with rabbit anti-EGFR.
Panel D; Inunur~oprecipitation of SKBR-3 and A-931 cell lysates with IgM monoclonals. Lane 1 contains molecular weiyt~t standards. Lades 2, 9, 6 and 8 contain i0 SKBFC-3 lysates. Lanes 3, 5, 7 and 9 contain A-931 lysdtes. Lanes 2 and 3: precipitation with OD3. Lanes 9 dnd 5: precipitation with NA3. Lanes 6 and 7:
preCipitdtion with TEPC 183. Lanes 8 and 9:
precipitation with rabbit anti-EGFR.
I. Immunoblot Lysates of SKBR-3 cells (ATCC HTB 30) and A-431 cells (ATCC GRL 1555) were electrophoresed on 1.5 mm thick 7~ SUS-polyacrylamide gels, using a 4.5% stacking gel. The separated piuteins were transferred onto 2G nitrocellulose iSchleicher s 5chue1l) using the BioRad Transblot*apparatus. The nitr~c:ellulose filter was then k~locked for i hour in Blotto*(3% dry milk, 2~ normal goat serum., 0.1~ Tween-20*in PBS) and incubated for 3 hours at room temperature with either 0.5 )tg/ml OD3, or 2 )tg/ml Pb3 (both diluted with Filotto), or with 20 pg/ml 291-3A
(in culture supernatanLl. 291-3A is an anti-ECFR
monpclonal antibody generdled using a peptide derived from the tyrosine kinase domain of EGFR. (291-3A was a gift from Randall Schatzmann, Syntex ResearcYr, Palo Alto, GA.) Filters were rinsed 3 times in a High Salt Wash buffer (?.0 mM Tris-HC1, 1 M NaCi, 0.05% Tween--20, pH 7.6) and were then incubated with alkaline phosphatase labeled goat anti-mouse IgG+1gA+IgM (Kirkegaard & Perry Labs) for 1 hour at room temperature. They were washed again three times with the high salt wash buflex, and the bands were * trade mark ~~Z~~~~
vlSUdiized using a HCIP NBT substrate kit (kirkegaard b Perry Labs).
In all cases, lane 1 contains 80 ~g of SKBR-3 cell lysate and Lane 2 contains 80 ~g of A-931 lysate, 5 separated by SDS-PAGE on a 7.5% SDS-polyacrylamide gel, and electrophoretically transferred to nitrocellulose.
Tmmunoblot results ate presented in Figure 3. Panel (a) detection with 0.5 ug/ml purified OD3J panel (b) deteCtiOn with 2 ug/ml purified P83: and panel (c) 10 detection with 20 ug/ml purified 291-3A.
J. mmmnf~~orc~o Cell lines were grown to confluence an F1-c:hnmbered LabTek*tissue culture slides (Miles Scientific) Overnight. T2ley were briefly washed in 15 Lulbecco's PB5 (containing Ca++ and Mg+*) and were fixed with 3~ formalin for 30 minutes at room temperature. A
1;50 dilution of TA-1 ascites fluid (diluted in 50%
normal goat serum) was incubated with the cells for 1 hour at room temperature. The slides were washed again 20 with PBS, and were then incubated for 1 hour at room temperature with fluorescein labeled goat anti-mouse IgG
(c:appel) .
Immunofluorescent results are presented in Figure 9: panel (a) 18-3-7 cells; panel (b> NIH3T:3 25 cells; panel (c) SKBR-3 cells and panel (d) A-931 sells.
Positive fluorescent staining wds observed on 18-3°7 and SKBFt-j cells. No staining was observed on N~H3T3 arid A-431 r.Plls.
K . F 1 ow C~/~~,.~,Y.
3p Cells were harvestad.from culture, washed once, and resuspended to a Goncentrati~i~ of 2x106 viable cells per Sample in Leibovitz L-15, They were l~han incubated with 1 ~tg of purified TA-1 or with the isotype-matched control MUPC°2a fir 1 hour at 4°C. The cells were washed three times with PBS and incubated * trade mark with 1 Ng of goat anti-mouse Ig-FITC for 1 hour at 9°C.
This inCUbdtivn was fulluwed by three additional washes in YRS. The Cells wW a analyzed using an EPICS V flow c:ytumetEr with an argon laser tuned to 4~B nm.
UiscrimindLors were set such that <SRs of the oells were positive with t?~e isotype-matched control antibody. The percei~Lage of cells positive aid the mean fluorescence inT.ensity fUa each histogram was determined using the Easy 85*software (Coulter). In all panels: MOPC21(M21) is re.pre$ented by ... and TA1 is represented by Flow ~ytometric results are presented in Figure 5: Upper left: NIH/3T3/ras, an NIH/3T3 cell line tran5fected with tloe ras oncogene. Upper right:
17-7-B, an NIH/3T3 cell line :o-transfected with the ra~
and human neu oncogenes. LoweL left: X-3-S. sn NIH/3T3 cell line co-Lransfected with the ras and human t~
oncagenes. Lower right: 1B-3-7, an NiH/3T3 tell line transfected only wit)u the human ~ onCOgene. The p185-positive cell lines demonstrated a mean-fluorescence intensity which was approximately 10-fold greater than background.
Table 1 summarizes the results obtained from ~)1e above-described evaluations.
* trade mark -0 t s. :.~ .. .
I~RJ~ Ir ~ 1 . ' d* r~i~ ~'t~ ,. . ... . . ~ .
1 ~ M' '17 W 4.s Gd ry m~ble 1 ~~'?dr~~lQr~i~S_of Mabs IsotyPe ~..... ._ . ReactiY.ii.Y____ and WesteYn Flow MaD -..xv4class_HloL ._ _._. ,F'~-uQrs~csnc~
_~ Ip _ Cytoms.tZY

BD5 IgGa/K - + + +

RCl IgG~/x - + + +

TA-1 IgG~/K + + +

NA3 IgM/K - + ND*~~ ND

OD3 1gM/K + + HD +

PB3 IgGld/~l + t ND +

RC6 IgG;/K - + ND ND

NB-3 IgGa/K - + ND +

ID5 IgG;/K - + ND +

1 ~ IB3 IgG;/K - + ND ND

~ All e~sbys we=e performed as described.
IP-itnmunopseCipitation f ~Iuo>:escenCe-llnmun0~luOte9CenCE
"~~ Na-not done ~,XAMPLE 2 J2~~on~~rdiion that monoclorrel antibodis,~ TA-1 and N83 ~g~ogn~z.~ d_stinr~t ep~~pPS an th_e X185 molccu~e ~5 Competitive enzyme immunoassays were perfarmed Lc; demo:~:~tz'a:e that anti-hum~sn p,~u monoclonal antibodies TR-7 and ?~B-3 bind to different epitopes on the evtxacellular domain of p185. This ws~s shown by co-inCUbating the two antibodies with plBS and demonstrating 3U tha_ neither car. inhibit the specific binding of the othez.
N
Microtiter plates were coated with 50 X11 of a cell lysate of 1'i-;i-1-3 cells wtaich were used ss a source 35 of pl$5, at a total protein concentration of 10 ~tg/ml.

~~2~2~~

(NcKenzie et al., 1989 On4ogene 9:543-548). The 17-3-1-3 cEll line is an NIH 3T3 cell line stobly transfected with the full lengtt: human ~~u gene. The plates wore Coated overnight at room temperature, and then washed three times with ELISA wa5l~ (0.05% Tween 20 in phosphate buffered saline, pH 7.4). Serial two-fold dilutions of the competing antibody, fr.um 4 ~tg/ml to 0.03 )ag/ml, was then added to the wells and incubated for 1 hour at 37°C.
The plate was washed three times and the biotinylated test antibody was added to each well. Either a 1:800 dilutiau <rf 7 mg/ml biotinylated-TA-1, or a 1x504 dilution of 1 my/rol biotinylated-NB-3, was used.
Following a 3 hour incubation at 37°C, the plate was washed again three times with ELISA wash, and a 1:9000 dilution of avldin-labeled horseradish peroxidase tavidin-HFcP; Sigma) was Lheri added and incubated for 1 hour et 37°C. Tl~e plate was Washed a final three times, was developed using tetramethylbenxidine (TMH:
5ig;~,a; 10 mg TM8 in 1 ml dirn-r.thylsulfoxide, added to ?.:: 50 ml of a 0.1 t~: acetate buffer, pH 6.0, plus 100 ~1 of 3~E 1:200 , and the reaction stopped after 5 minutes using 2.5 h H25Uq. The resulting yellow Golor was read on an >;l.ISh plate reader at q50 nm.
F,esults 2'~ Figure 6 shows the binding curves with TA-1 and NB-.~ were used to cum~ete with the binding of biUtm ylated-z'A-1. It can be seen that increasing amounts of unlabeled TA-1 completely inhibits the binding of biotinylated-'fA-1 (open squares), as indicated by the 30 reduction of the OD 950 signal. Increasing amounts of NB-3 (closed triangles) had no effect. Figure 7 shows the curves when TA-1 and NB-3 were used to compete with biotinylated-1JB-3. In this carer increasing amounts of NF's-3 (closed triangles) completely inhibited ttae binding 2d~~~i~~~

0f tile biotif7ylated-NB-3, whereas TA-1 (open squares) had no effect.
These results indicated that the antibodies recognized twv distinct epitopes on the human n.~.~i gene product.
~X_AMPLE
n~~'~.~>S~r~.~.~i9n that~?a~..~s~~z~t.a.~s~~~._~.Q~~.a.l.
f..~ a i d s ~.s.~'.~ 1 a t e.d t o p1&~.
The SKBR-3 cell line is a continuous cell line which originated from a human breast tumor, and is known tc enp:eSS high levels uI the human ps.Lt gene product, p185 (Kraus, et al. t1997l, Embo J. 6:605-610).
Monoclonal antibodies speciTic for p185 as described l; above were shown to detect a protein of approximately :OO,UGO daltvns in the culture media removed from these cel:s. A competitive binding a55ay was performed in order to confirm that p100 was related to the human ri8ll gene product, p185.
G Tr're p185 p~esEnt in lysatcs of cells transfected with the ~ gene was used to compete far antibody binding with a radiolabeled p100 in the SltBR-3 supernatdra. The amount of antl-~y antibody used in t?'~e competition assay was titrated in order to 2~~ demonstrate two points:
.. the amount of antibody was a limiting reagent in the assay, such that the unlabeled p185 and the radiolabelled p100 would actually compete for Linding; and 3G 2. the amount of a>Stlbody used was sufficient to visualize the radiolabelled p100 band following a~.~t hradiography.
One 10 cm petri dish containing subconfluent (approximately 5 X IG61 SnFSR-~ cells (ATCC HTB 30! was 35 incubated with 500 ~tCi of 3'S-cysteine tDu Pont) in CyStCiue-free media overnight at 37°C, 5~ COp. The culture supernatant was used as the source of radiolabelled p100.
An N1H 3T~ cell line stably tamnsfected with the Iull length human n~ gene, end designated 17-3-1-3, was used as a source of p185. Call lysates were prepared by scraping cells Pram 10 (I5 cm) petri dishes into 10 ml of p?~c~sphate buffered saline (PBS) . The tells were pclleted by centrifugation at 200 X g for 10 lp minutes, the pellet resuspended in 5 ml of hypotonic lysi5 buffer (10 mM Tris, 10 mM !(Cl, 5 mM EDTA, pH 8.0), and then homogenized with a bounce homogenizes. The homogenate was centrifuged at 200 X g far 10 minutes, and the resulting supernatant was sonicated for 15 seconds. All remaining cellular debris was then removed by centrifugation in a mic:.r~-centrifuge, and the final supernatant was used as the cell lysate. Similarly, a lysate of nontransfected NIN 9T3 cells was prepared.
Total protein concentration of each lysate was determined using a kit based on the Bradford method (BioRad'~, using HSA as a standnLd. The concentration of the 17-3-1-3 lysate was determined to be 2.69 mg/ml total protein, and the N1H 3T3 lysate was 1.01 mg/tnl.
The 1~-3-1-3 lysate was used at a 1:3 dilution for these 2> assays.
The amount of human ~ related protein present in the SKBR-3 supernatant and the 17-3-1-3 lysate was determined by ELISA. The monoclonal antibody TA-1 was used to coat tire wells of Nunc*Immunoplates.
This was followed by incubation with either the supernatant or the Iysate, and then by incubation with biotinylated PB3 antibody (IgG2x). The assay was developed by a final incubation with avidin-labeled horseradish peroxidase tHRP), and tetramethylberrzidine tT:~:fi) as the colorimetric substrate far HRP. This * trade mark 31 ~~~~~ 3~
Capture FLISA detected 3.72 0D units/ml of ~neu related pfotein irr t)ue culture supernatant from the SKBR-9 Cells, and 1295 OD units/rnl in the 17-3-1-3 lysate. The activity in the NIH 3T3 cell lysate was zero.
Protein A-Sepharose (Pharmacia) was swollen and Washed w9th immunoprec:ipitation buffer (IP buffer:
lib Triton X-100, 1% sod9um deoxy-cholate, 0.1$ sodium dodecylsulfate, 10 mM Tris, 650 mM NaCl, pH ~.2), and resuspended 7:1 (vol/vol) in IP buffer.
Samples were prepared by mixing 103 )Sl of 17-3-1-3 lysate (diluted 1:3; approximately 95 Ofd units of ~ activity) with 1 fig, 0.3 fig. 0.01 fig, or 0.003 ~g of purified PB3 ar7tibody. These were incubated pvernight. at 9°C with mixing. Control samples were prepared by incubating 277 ~tl of NIH 3T3 lysate5 with Pt33 in the same fashion. Following the overnight incubation, 1.028 ml of radiolabeled SKBR-3 supernatant (approximately 9 Uv units of ~y, and 4 ~tCi of total labeled material) was added to each sample. All samples 2C. were then incubated again overnight at 4°C with gentle mixing. Following this second incubation, 50 pl of the Protein-A 5epharose slurry was added to each sample, and each was incubated with mixing for 1 hour at 9°C. The 5ephardse was pelleted by centrifugation for 1 minute in a micro-centrifuge, and was washed 4 times by resusgending the SeYhatose in 1 ml of IP buffer, vortexing briefly, and centrifuging for 30 seconds in the micro-Centrifuge. The samples were washed a final time in TE5 (iris buffered saline, pH 7.5). 5amp7.es wErE air dried, resuspended in 30 ~1 of SbS-PAGE Sample buster, and then incubated at 100°C for S minutes. The entire sample was loaded and run on a 7$ SbS
polyaCrylamide gel. After the gel was fixed, it was rinsed with k:n3HanCe*(bu Pont)r and dried. An aut.oradioc~raph was produced by exposing X-OMAT~'AR film * trade mark 2~'~'~u'~~~
3z (Kodak) t0 the dried gcl in the presence of an it~Lwsifying screen at -70°C for f days.
Results are pmsented 1n Figure 19.
Lanc 1: 1 ~g of PB3 + l~-3-1-3 lysate + SxBR-3 supernatant Lanc 7.: 0.3 dig of PI33 + l7-3-1-3 lysate + SKBR-3 supernatant Lane 3: 0.1 erg of PH3 t 17-3-1-3 lysate + SxBR-3 supernatant Lane 9: 0.03 Ng of PB3 + 1'~-3-1-3 lysate + SKBR-3 supernatant Lane 5: 0.01 fag of P83 + 1~-3-1-3 lysate + SKBR-3 supernatant Lane 6; U.003 ~1g of PB3 + 17-3-i-3 lysate + SKBR-3 supernatant Lane 7: 1 yag of P83 + SKBR-3 supernatant Lane 8: 0.003 ~tg of PB3 + Sx$R-3 supernatant Lane 9: 1 qty of Pa3 * NIH 3T3 lysate + SKBR-3 supernatant 2G Lane 10; U.3 ~g of PB3 + NiH 3T3 lysate + SKBR-3 supernatant Lane 11; 0.1 ~tg of PB3 + NIH 3T3 lysate ~ SKBR-3 supernatant Lane 12: 0.03 ~tg of P83 + N1H 3T3 lysate + SKaR-3 supernatant Lane 13: 0.01 dig of PB3 + NIH 3T3 lysate + SKBR-3 supernatant Lane 19: 0.003 Ng of PB3 + N1H 3T3 lysate + SKBR-3 supeznatant 3p The right half of the gel showed titration of the PBa antibody in the presence of the NIH 3T3 lysate a5 the competitive agent. As there should be 170 competition between molecules pzesent in the 3T3 lysate and the p100 protein, the disappearance of the band at 100,000 daltans as the antibody is titrated indicates that the antibody is becoming the limiting zQagent.
T)~e left side of the gel showed titration of the YB3 antibody ire the presence of the 17-3-i-3 lysate.
In all lanes, except the 1 dig 863 sample, the p100 band wdS absent. This was particularly significant in the lanes xhich contained 0.3 ~g and 0.1 fag of PB3, as there was St111 discernable band in the comparable samples when the Nlri 3T3 lysate was used as the competitor. The presence of some p100 in the 1 erg PB3 sample indicated that the antibody was in such excess that it could bind the radi0labeled material as well as the unlabeled p185 which was added.
The disappearance of the p100 band when the 17-3-1-3 lysate was added indicated that the plDD
molecule was indeed aerated to p185. The ~ixe correlated with the predicted molecular weight of the extracellular domain of the human ~ gene product.
0 EX~L~.~.
~ etect.ion p~ ,pl 0 logiGa sam~lg~.
These examples illustrate detection of the human ~ related protein in cell lysates, tumor lysate&, and in human blood plasma and Sera.
A. ~8p~'.lr~n~s~~'.i~.~'~fral F~cttzcol Polystyrene plates (Nutrc) were coated with either 20 microgram per milliliter (~glml) of an anti-p.~3 monoclonal antibody IMab), a combination of anti-~y 3L Mobs, or a polyclonal dr~tibody for the purpose of capturing hu>s~an rev protein from various biologioal specimens. Mobs were diluted in 0.1 M carbonate buffer (pH 9.6) and 100 microliters (ul) added to each arell of the microtiter plate. The plates were then incubated overnight at_ 9°C.

34 ~~~~9~~~
After incubation, the coating material was decanted from the plates and 250 ~1 of blocking buffer (F'p5 with Z% bovine serum albumin (BSA), 10~ Beta-lactose and O.OI$ l.lrimerosal) was rdded to each well. The S blOCking buffer was decanted and 250 )xl of fresh blocking t7uffer added to each well in ojder to block sites on the miCrOtiter wells not coated with the anti-human n~u antibody. Plates were incubated for 2 hours at room temperature. The blocking buffer was decanted and plates blotted with payez towels. Plates were dried overnight in a hood at room temperature and then stored covered at 9°C until use.
Specimcirs to be evaluaC.ed for the human neu protein consisted of lysates prepared from normal, :' prFneop=antic or neoplastic cells or human body fluids such as serum, plasma or urine, Tire specimen was then added to the antibody coated wells in order to capture the human neu protein from Line specimen. The plates wore incubated overnight at room temperature. After 2~:: incubati.on, the plates were was?red six times with Du Pont*
Flat.e h-asrr F~uttEr (Pl3S, 0.05, Tween 20) and a Dynatech*
i~°late a skier in order to remove unbound biological spec:rr.er~ .
Another anti-human neu Mab coupled to biotin 25 was added to each well and incubated for 30 minutes at roo~. Lerr,FeraLUr~. Flates were Lhen washed six times with Du Pont Plate wash Buffer. To detect the biotinylated anti-neu Mab, Streptavidirr-horseradish peroxidase was added at a 1:2500 dilution crrd allowed to incubate for 3C 15 r,.inutes aL room temperatuze. Plates were then washed sir. times with Du Yont Plate Wash Suffer. To complete the: react ion, the substrate orthophenylenediamine (OPD) was added for 1 hour aL room temperature. xhe reaction was stopped with sulfuric acid and the optical density * trade mark 35 ~D~~w~~
was deCermined using a Molc:c;ular Devices Elate Reader at a wavelength of 990 nm.
Pxtscz~.Qi?__s~f P?~5..fr.Q.m~el7, ~~-~
lyual"g~~sing ~~:~.r~~ssa~s Several caYtuze immunoassays have been perlprmed t0 determine usefulness of this assay on binlogiCdl materials. Figure 9 shows the results of a Capture immunoassay in which the first antibody is TA-1 t~nc9 the 5ec;ond antibody is biotinylated NA-3. Cell lySdCes were prepared from several human tumor cell lines. neu RNA levels have been published for several of these Cell lines (5K-BR-3, ZR-'75-1, MCF-7). The relative levels of human n~u detected by this assay area in agrQpment with the published RNA data. The results of these assays and several others (not shown) using cell lines with known levels of human y~ indicated this assay can be used to determine the relative level of human ~t in cell lysates. Results also indicated that differences in expression of human neu related protein can be used to classify the carcinoma cell lines shown in Figuse B.
In order to determine if this assay could detect human neu in tumor lysates, tumors that either expressed human neu (X-3-5) or did not express human ~d (3T~ ras) were grown in nude mice. The two NIH 3T3 2~ derived cell lines are isogenic except that X-3-5 er,pressed of the human ~ gene. Figuia 9 shows the results of a capture immunoassay using A18-3 as the capture antibody and biotinylated TA-1 as the detector antibody. A human neu related gene product was detected in t);e lysate of the X~3-5 tumor but not in the lysate of the 3T3-ras tumor, indicating the assay can specifically detected humermeu in tumor lysates.
Several investigacvls have shown that many human breast tumors express human neu at high levels. In 3S order to determine if human ~ Can be detected in human ~U~~~~O

bred5t tumors, Lwu samples from the same individual were preYeaed. Lysates were prepared from a human breast tumor (2797-01-050) and from normal breast tissue (2747-01-050) from the same patient. In this assay TA-1 was used as the first antibody and biotinylated BD-5 was used aS the detector antibody. Figure 10 shows that human neu ran bP detected in the tumor.
These assays showed that the neu capture immunoassay Specifically detected a human ~ related gene product from either cell or tumor lysates. The data also indicated the assay tan determine relative levels of ~ between samples.
LySdtes of human breast tumor and normal breast tissue (2747-pl-050) from the same patient were evaluated '°~ by immunoblot. The immunoblot results showed that the A J
n » related protein detected tiom the breast carcinoma lysate was p185.
C. Detection of p100 i« L~~plasma and a era TJ~ese examples 111ustiate detection of human ~ in sera and plasma from mice and humans bearing neoplastic tumors.
Tn order to determine ii human nru can be specifically detected in human sera or plasma, several control experiments were performed. These included detection of human neu in the culture supernatant of cell lines that express high levels of human ~ and in the sera of nude mice bearing tumors.
Figure 11 shows the results of a capture immunoassay of human neu from Culture supernatants of cell lines using NB-3 as the capture antibody and biotinylated TA-1 as the detector antibody. The results show that a human p.Sy related gene product can be detected in the supernatant of murine (1~-3-7) or human (SK-BH-3) cell lines that express high levels of human neu but not in the supernatant of a cell line that does ~~~~2~~~
npt express human reu, 3T3-ras, or in media alone. ~'wo of these cell li»e:s aye able to grow as tumors in nude mice (18-3'7 dnd 323-ras). Mi<:e bearing tumors derived from injecting these cell lines subcutaneously into nude mice were bled and thEir sera was analyzed for the presence of human n~ by a capture immunoassay using TA-1 as the capture arm ibody and biotinylated s~-5 as the d4tector dntibody. Tire zesults of this assay are shown in Fi.guxe 12. As with the cell or tumor lysates and the cell Culture supernatants, only the sera of the nude mouse bearing a tumor that exy essed human ~ reacted in tire assay. Both normal nudE mouse sera and sera from a nude mouse bearing a tumor that did not express human neu did not react in the assay.
These experiments irrdiCtited that a human neu related protein was found in the sera of nude mice bearing tumors expressing human ~y as well as the cell line Causing the nude mouse tumor. Human neu was found in the supernatant of human cell lines expressing n 2~ (SY,-BR-3) .
ASSays were performed that were designed to test the hypothesis that patients with tumors expressing high levels of human ~ will Jrave sera containing a human ~. related protein. One series of assays used :~5 anti-r,eu Mab TA-1 as the capture reagent and biotinylated BD-5 as the detection reagent. Samples for analysis included normal human plasma and plasma from two breast carcinoma patients. Results show that normal piasrna and plasma from patient A,7AC were virtually unreactivc in 30 this particular assay whereas plasma from patient PSU1.
showed significant reactivity irr this assay suggesting that a human ~ related protein was present in the plasma of the breast carcinoma patient PSUI. (Figuxe 13).
This experiment was repeated on a larger number 3~ of patients with Mab NB-3 as the cayture reagent (affixed 3~

~~~b'~'~~
t0 the solid support) and biotinylated MAg TA-1 as the detection reds3ent. Using this assay, about 725 separate F>lasrna samples were evalu~sted for reactivity with P~u-~peCiIiC monoclonal antibodies. The samples were oht.ained frUm Dr. Daniel Hayes and Dr. Don Kufe of the Dana-FdrbPr Cancer Institute. The specimens consisted of plasma samples Pram normal individuals, plasma samples from individuals with Lenign breast disease, plasma SamplPS from individuals with breast cancer, plasma Samples from individuals with gastric carcinoma and individuals with ovarian cancer. It appears that the BvFrage ?'mrr~an nay valve for this particular study was c~b0ut 600 far tt~e riarmal plasma samples, i.e., plasma samples from nuimal individuals and individuals with normal (benign) breast disease. Table 2 presents the individual data obtained for samples. Table 3 presents additional data for about 66 plasma samples from ovarian cancer patients which were evaluated for reactivity with rev-specific monoclonal antipodies. It may be possible 2ti to convert the human neu values presented herein to fmol/ml.

2U~~~~i~

~'~blQ,.2 n~V ~ SIJLTS
..~APT~
~I~7~Al~E~
PAN_E~..
~.~L~5hL9_~~~_ Human Human Human neu n~u neu y~n~.'=~Y~11?~ ~53mP1~. Values $am~le yalues 102 897.8 251 892.9 386 733.1 1DS 460.1 253 1091.0 390 1099.0 109 1877.8 264 815.4 392 674.0 110 <500 266 919.3 395 5675.7 10112 1087.9 27a 776.8 397 669.5 115 892.1 280 979.7 398 1474.3 116 11D6.5 281 391.4 909 10278.7 118 928.6 274 61730.0 412 1068.0 121 1:x'19.2 284 744.7 418 1142.7 15122 826.5 290 583.0 920 1363.7 129 887.9 291 2939.3 425 895.2 134 1133.9 292 715.9 428 1106.0 13G 1390.3 300 763.6 930 928.6 140 81915.0 305 175573.3 935 562.6 2D19~J 962.2 316 660.6 436 5568.0 149 1261.9 317 841.8 437 560.4 358 1363.1 320 576.7 ail 773.3 165 'JO'~1.5 322 857.0 474 1099.0 :6~ 3226.5 324 649.2 510 558.9 251'lb Bzi.E 330 927.6 517 865.8 15i 1126.0 339 878.0 521 2156.7 192 699.3 355 1128.0 522 3042.0 153 10'6.2 356 662.6 526 599.5 209 7359.0 360 859.0 529 1542.0 3U2U'! 1524.8 370 677.8 530 591.3 225 19b9.S 376 3486.7 533 932.6 292 853.9 378 834.2 537 1174.8 385 922.3 542 582.9 .-. .....;,.;
,.,:: .,. . ..
y",...

l~N, :.t 9 0 ~~~~3~~

Table _~
~C9.nt~dL

~L1SA_ Fte~~l~.

Human Human Human n.~u n~.v ~s?m~le ~zunpl~. valt~~. SY.alnes Values 556 645.8 '~15 1053.0 511 443.5 560 845.4 721 997.1 520 1084.0 583 8534.7 728 815.9 539 1143.0 587 669.8 732 669.3 567 502.4 599 1054.0 796 725.3 596 b67.9 6U4 533.0 749 666,1 606 675.4 6(~9 78266.7 753 1108.5 618 646.8 611 988, 764 10905.0 710 991.8 615 1065.0 130 598.8 718 556,6 6'19 839.9 191 1069.0 759 106x.5 621 1238.5 169 822.7 1303 9277.0 622 1211.5 187 812.7 1350 2068.0 623 1061.0 198 328.2 1552 804.6 625 866.4 265 664.1 1562 951.3 2U 63'l 956.1 2'73 752,1 1810 9397.5 638 400.1 283 1515.5 1944 1339.0 f59 1493.0 299 589.2 1978 828.3 662 1079.0 299 586.3 1983 2699.0 67's 2130.3 391 674.2 2004 1190.5 2_~ 679 644.3 359 851.6 2105 760.

68a 71C.5 364 729.7 2236 279.1 68E. 757.5 383 969.0 2553 1298.5 687 7209.3 910 760.7 2661 21668.5 691 583.9 424 563.1 2669 2831.5 3U 692 647,2 427 458.0 2823 714.8 fi5~ilJlv.S 929 6'77.1 2892 1'158.0 702 1927.0 431 1267.3 2904 31408.0 708 860.2 489 423,4 2934 7129.3 Td~7l~
~lCOnli~~5~1 F l~ I
SA ~~
Sl? ~.L.S.

Human Human Human n~ n.~u n~

~am.F?.~~v~~ ~~m.F?lg Ya~~ .~a~~l~ y la ues 3282 882.4 C24 422.1 0108 651.8 3527 621.0 C25 692.0 C109 468.1 BU1 713.5 C27 880.b C110 634.6 802 586.9 C28 327.8 C111 617.4 X03 581.0 C29 636.5 C112 373.7 804 560.9 C94 790.0 C114 529.9 F~05 731.1 C59 669.8 G1 378.0 806 586.3 C55 570.0 G2 542.8 k30~ 1038.0 CBO 605.1 G3 653.1 BOB 1027.0 C82 664.3 G5 523.6 T~09 900.6 C83 782.2 G6 750.2 810 455.1 C84 574.5 G7 473.6 811 332.5 C86 593,6 G8 600.1 812 627.3 C89 565.9 LWOl 476.7 2U 813 515.9 C90 598.5 LW02 445.$

I~14 505.5 C91 567.7 LW03 586.5 815 421.1 C92 679.2 LW04 693.0 A16 574.4 C9:i 624.3 LW05 565.9 817 679.2 C96 717.0 LW06 404.6 t5 H7B 665.6 C97 505.9 LW07 '59.7 819 564.7 C98 437.2 LWO8 886.9 820 475.3 C99 571.1 LW09 555.6 R?1 1297.0 C100 783.9 LW10 567.7 822 344.1 C101 73'1.6 LW11 454.4 30 823 824.2 0102 572.8 LW12 411.5 C10 99f~.0 C103 528.2 LW13 677.5 C17 597.8 C109 823.4 Lwl4 540.2 C21 362.9 C105 608.8 LwlS 617.4 i.wl6 1263.0 4z .~ab~ ~~
p~l~ Capture FLISA

Qy~ r i ~~~La sn~.a.s.

~amPl~ ~m~n n~12 ~~~2~.5. ;~7m,~ Wr1 n~u yalues E01 938.8 E036 566.?

E02 495.1 E03? 512.2 E03 794.9 E038 901.4 E09 1057.9 E039 709.9 E05 829.2 E090 488-8 10E06 1769.8 E041 442.3 E07 757.1 E092 483.8 E08 1138.0 E093 613.1 E09 900.7 E099 634.6 E011 917.7 E045 >10000 15E012 b9"l,5 E04b 422.3 E016 748.8 E047 525.7 E017 807.0 E048 590.0 EU19 632.6 E049 55?.8 E02U 729.1 E050 -w- --- 803.9 2cEozl 491. Eo51 so9.1 E022 549.9 E052 ?82.6 E~,23 946.2 E053 386.6 E024 516.5 E054 890.8 EG25 550.8 E055 558.9 l5EU<< 1099.2 E056 >10000 EO?.7 691.9 E057 595.6 E02b 633.3 E058 923.5 E029 394.9 E059 983.9 E030 4'3.8 E060 X60.3 30EU3~ 852.1 E061 489.0 E032 829.7 ' E062 485.6 E033 698.8 , E063 574.4 E0a4 760.4 E069 739.1 E035 1~U3.0 E065 522.3 Tak~~3_ _Ic~i.ns~

neu Cap~~e._ .FIT ~A_ Qv~.x~n.~~sm~

~dr'~plg ~yimzn nBl~_ Values E066 667.5 E067 445.0 E06B 582.5 E069 609.7 E070 792.6 E0~1 1817.5 D. ~mmlrnQ~lot E~al~ation.~f~li~smas Some of the samples which were evaluated by Lhe ELISA protocol a5 described above were evaluated by immunoblot as described below using the anti-human pgy related monoclonal antibody designated oa-3. Immunoblot results presented in Figures 15-17 are set forth in Tdble 4 along with corresponding human ~y values as determined by the human neu ELISA.
xhe immunoblot procedure entailed the following protocol:
Appraximately 10 ~1 of the plasma samples were prerleared with rabbit anti-mouse protein A agarose for Z5 30 minutes at 9°C using a rotating wheel or rocker. The sample was centrifuged in an eppendorf centrifuge. The supernatant was decanted and saved. About 50 ~1 of sodium dodecyl sulfate reducing bufrer with methyl green (sample loading buffer, SLB) was added to each sample.
a0 50 ~1 of SIrB was added to the high molecular weight standard. The samples were then heated in a hot oil bdtt~ for f ive minutes at 100°C.
The samples were electrop)~oresed on a i.5 mm or 3.0 mm thick 5~ SDS-polyacrylamide gels, using a 3.0%
35 stacking gel. The separated proteins were transferred o«t_o nitrorE°alulose usi=ng the DioRad Transblot apparatus. The nitroc:e11u1ose filter was then blockad fOr one hour in Blotto (3% dry milk, 2% normal goat serum, 0,1~ Tween-20 in paS) and incubated for 3 hours at room temperature with about 10 ug/ml of On3 or a class matched cont.ml diluted to 10 ~g/ml in 50 ml Blotto. Filters were rinsed 3 times in a high salt wash buffer t20 mM Tzis-HC1, 1 M NaCI, 0.05% Tween-20, pH
7.5) and were then incubated with alkaline phosphatase labeled goat anti-mouse IgM (Kirkegaard 6 Parry LabB) for at least. one haur at room temperature. They were washed again 3 times with the high salt wash buffer, and the bands were visualized using a BCIP NBT substrate kit (Kirkegaard 6 Perry Labs).
Figure 15 tNlO) shows that OD3 detected e~
human Day related protein in human plasma, ovtained from a lactating woman, an individual having benign breast disease and a breast carcinoma patient, having an approximate molecular weight of about 100,000 daltons in 2G lanes 2, 4, b, and 8. No such protein tp100> was detected in lanes 3. 5. S. and 9. Lane 1 contained mol~:cular weight markers. Lanes 2 and 3 contained plasma froth a lactating woman. Lanes 4 and S contained controls. Lanes 5 and ? contained plasma from breast carcinoma patient 1109. tPiasma taken from patient X109 had a human neu value of 1,877.8 as determined by ELISA.) Lanes 6 and 9 contained plasma from breast carcinoma patient ~2B3. (Plasma taken from patient X283 had a human neu value of 1,515.5 as determined by ELISA.) Lanes 2, 9, 6 and 8 were blotted with monoclonal antibody OD3. Lanes 3, 5, 7 and 9 were blotted with the class matched negative control monoclonal antibody, TEPC 1S3 (a myeloma %gM purchased fmm Litton Bionetic5) .

~~~~i~~~
Figure 16 (N12) Shows that OD3 detected a Dil7.
related protein in human plasma, obtained from an individual having benign breast disease and from a breast carcinoma patient, having an approximate 5 molecular weight of about 100,000 daltons in lanes 1, 5, and 7. No such band was detected in lanes 2, 3, 4, 6, and 8. Lanes 1 and 2 contained plasma from an individual with benign breast disCase. Lanes 3 and 4 Contained plasma fxom an individual with gastric cancer.
10 Ldnes 5 and 6 Gantained plasma from breast carcinoma patient X2661. (Plasma taken from patient #2661 had a human meu value of 21,668.5 as determined by ELISa.) Lanes 7 and B contained plasma Pram breast carcinoma patient 112909. (Plasma taken from patient X2904 had a human nay value of 31,008.0 as determined by hLISA.) Lane 10 contained molecular weight markers. Lanes 1, 3, 5, and ~ were blotted with monociorral antibody OD3.
Lanes 2, 4, 6, and 8 were blotted with the negative control antibody TEPC 183.
2D Figure 17 (N13) shows that OD3 detected a human ~e~ related protein in human plasma, obtained from do individual with benign breast disease and a breast carcinoma patient, having an approximate molecular weight of about 100,000 daltons in lanes 4, 6 and 8.
25 Na suc', bane was detected in lanes 2, 3, 5. 7, and 9.
Lane 1 contained molecular weight markers. Lanes 2 and 3 contained plasma from an individual with ga3trie carcinoma. Lanes 9 and 5 contained plasma from an individual with benign breast disease. Lanes 6 and 7 30 contained p:asr'~a from an breast carcinoma patient ~190. , (Piasn,a taken from patient 1140 had a human neu valve of 81,915.0 as determined by EL1SA.) Z,anes 8 and g contained plasma from breast carcinoma patient X305.
(Plasma taken from patient 8305 had a human ~~u value of 35 175573.3 as determined by ELISA.) Lanes 2, 4, 6. and 8 46 ~~r~~~'~!~
wCre blOttEd with anti-human n~, monoclonal antibody OD3. Lanes 3, 5, ~, arid 9 were blotted with the n2gative Control ~nonoolonal antibody TEPC 183.
Figure 18 (NB8) slyc~ws that OD3 detected a human 11~~1 related protein in T'~uman plasma obtained from an individual with ovarian carcinoma having an approximate molecular weight of about 100,000 daltons in lanes 2. 3, 6, and 7, No such band was detected in lanes 4, 5, 8. 9. and 10. l.ancs 2, 4. 6, and B
contained ~riasma Pram ovarian cancer patient X95.
(Plasma taken from patient ~A5 had a human n~ti value Of >10,00U as determined by EL~SA.) Lanes 3, 5, 7, and 9 contained plasma from ovarian cancer patient X35.
(Plasma taken from patient X35 had a human , » value of 1,703 as determined by ELISA.) Lane 10 did not contain a plasma sample. Rather, it conta'ned the Control which was tumor er.tract electrophoresed from a cell line, 17-7-8 known to contain human neu p185. Lanes 2 rind 3 were blotted with anti-human rieu monoclonal antibody OD3 for one hour. Lanes 9 and 5 were blotted with the negative Control monoclonal antibody TEPC 183 for one hour.
Lanes 6 and 7 were blotted with anti-human p~li monoclonal antibody OD3 for three hours. Lanes 8 and 9 were blotted with the negative control antibody TEPC 183 2=~ for three hours. Lane 10 was blotted with OD3 which detecr.ed the human ~ related gene product p185 in the tumor extract.

~'~~?1a~

Human ~ley.

Plasma Sample t;LZSA Results lmm~!?~.~.l.Qt...Results .IFi~I,ient~.l ,i1'~Vman _n~v.~lu~l.p18~ n100 .~

109 1,877.8 - +

283 1,515.5 - +

2661 21, b68.5 - +

2909 31,008.0 - +

190 81,915.0 - +

305 175,573.3 - +

Claims (13)

1. A method of detecting preneoplastic or neoplastic cells in a human which comprises testing a biological fluid from the human for the presence of the extracellular domain of the human neu gene product by:
(a) contacting the fluid with at least one monoclonal antibody which is capable of binding the extracellular domain of the human neu gene product;
and (b) determining whether antibody binding has occurred.

2. A method according to claim 1, wherein the cancer is ovarian, breast, stomach, pancreatic, colon or lung cancer.

3. A method according to claim 1, wherein the biological fluid is selected from the group consisting of blood, serum, plasma, urine, cerebrospinal fluid, supernatant from normal cell lysate, supernatant from preneoplastic cell lysate, supernatant from neoplastic cell lysate and breast aspirates.

4. An immunoassay for detecting or quantifying the presence of the extracellular domain of the human neu gene product in a biological fluid obtained from a human which comprises (a) reacting the fluid with at least one first monoclonal antibody which is capable of binding to the extracellular domain of the human neu gene product;
(b) reacting the product of step (a) with at least one detectably-labeled second monoclonal antibody which is capable of binding to the extracellular domain of the human neu gene product at an epitope different from the epitope bound by the first antibody; and (c) detecting or quantifying the product of step (b), and thereby detecting or quantifying the presence of the extracellular domain of the human neu gene product in the fluid.

5. An assay according to claim 4, wherein an immunoreactive fragment of the first monoclonal antibody and/or the second monoclonal antibody is used.

6. An assay according to claim 4, wherein the detectable label is selected from the group consisting of radioisotopes, enzymes, fluorogens, chemiluminescent compounds and electrochemicals.

7. An assay according to claim 4, wherein the second antibody is conjugated to biotin.

8. An assay according to claim 7, wherein the biotin-conjugated antibody is detected by reacting the biotinylated complex first with streptavidin-horseradish peroxidase followed by reaction with orthophenylenediamine.

9. An assay according to claim 4, wherein the first antibody is selected from the group consisting of antibodies produced by the hybridoma cell lines having ATCC Accession Numbers HB 10204, HB 10205 and HB 10206.

10. An assay according to claim 4, wherein the second antibody is selected from the group consisting of antibodies produced by the hybridoma cell lines having ATCC Accession Numbers HB 10204, HB 10205 and HB 10206.

11. Hybridoma cell line, designated OD-3, having ATCC accession number HB 10204.

12. Hybridoma cell line, designated NB-3, having ATCC accession number HB 10205.

13. Hybridoma cell line, designated TA-1, having ATCC accession number HB 10206.