CA2054725A1 - Ptpase vectors, diagnostics, therapeutics and uses thereof - Google Patents

Abstract

Protein phosphotyrosyl phosphatase (PTPase) nucleotide segments and nucleotide probes are disclosed. Preferably, the PTPase gene is PTP1B. Their use as diagnostics is also shown.

Description

W O 91/131~3 2 0 ~. 4 7 2 ~ PCT/US9l/01~32 PTPasie Diagnostics and Uses Thereof The prcsen~ invention i3 direct~d co the use of vectors, transformanes and cell lines containing a supressor gene and the use of such gene for expression, diagnostic and eherapeucic means. More particularly, th~ inven~ion is direcced to the use of PTPase gones, prc~er~bly PTPlB, for dingnoselc and tharapoue~c uscs.
Ch~ngcs in thc phosphorylaeion scnco of protaln tyro~yl ro3~duos s~c ~n import~ne nspact Oe erow~h r0~ulacion.
Ligand-induc~d ty~o~lne pho~phorylnelon by r~cup~o~cyro~ln~
kin~cs loads to chAnga~ in ~on~ ~xprc~3l0n and ~-lt~r$ tho ~row~h o cells ~Yardon, Y., et ~l ~n~ R~v. Bioch~m. 57:443-478 (1988)]. In addltion, neoplnstic trans~ormatlon by oncoproteins, ~uch as ~X~ nnd fp~, app~rs to bc depondent upon inerinslc tyrosine ~inase acti~ity iSnyder, M.A., et al, Mol. C~ll Biol 5:1772-1779 (1985); Weinmaster, G., et al., EMB0 J. 5:69-76 ~1986)]. The role of dephosphorylation of phosphotyrosyl residues in regulating ~rowth remains unclear. For example, a net increase in the phosphotyrosine content of cells tue to the action of protein tyrosine kinases is reported [Sefton, B.M., et al, Cell 20:807~-816 (1980); Cooper, J.A., et al, J. Virol 46:15-28 (1983)], but the same net increase could easily arise due to loss or reduction in the acti~ity of a proeein phosphotyrosyl phosphatase (PTPase:protein-tyrosine-phosphate phospho-hydrolase, EC 3.1.3.48~. In either case, the resulting W O 9l/13173 2 ~ ~ ~ 7 2 ~ PCT/US91/01432~

net increase in tyrosine phosphorylation could lead to oncogenic transformation. To the extent that PTPases may re~ulate the overall phosphotyrosine content of cells, it has been suggeseed that these enzymes could function as growth suppressors tCharbonneau, H., et al, Proc. Natl. Acad. $ci USA 85:7182-7186 (1988): Huntor, T., Ç~ll 58:1013-1016 (1989)]. Vanadat~, a pot~nt inhlbitor of PTPasc~, cause~ a transient incr~ase In tho cellular phosphotyrosino cone~nt a3 woll as a phenotypic transformation oi somo fibroblast coll lines tKlarlund, J.K., Cell ~:707-717 (1985) ] .
Protein phosphotyrosyl phosphatases havo been identified in many differenc ouk~ryotLc coll ey,o~s in a broad rang~ of slz~s ro~ both particulnte and solubl~ c~Ll fracelons (Joncs, S.W., ct kl._Çh~ ?6l~:7747~7753; Lau, K.~l.W., ot al, ~ h~n.~L
23-36 ~1989); Tank~, N.K., ~ nl, ~ Ql__5 6722~6730 (1988) which nro Lncorpor3to~ by ro~oncQI. Th~
bost charActocL~cd Oe th~o i9 tho 37000 mol~cul~r wolght PTPa~Q
lB ~rom human pl~c~nta whlch hns b~n pur~Ei~d ~nd has h3d an amino acid soquenc~ rcport~d tCharbonneau, H., et al, Proc. ~atl.
e5l~L_~i~5i 86:5252~5226 (1989) Which is incorporneod by r~Perencel. In addltion, a cDNA encoding A putativo T
cell-speciiic PTPase has been isolated and sequenced [Cool, D.E., et al, Proc. Natl. Acad. Sc. USA 86:5257-5261 (1989) which is incorporated by reference]. It has also been reported that the transmembrane glycoproteins CD45 (also raferred to as L-CA or T200) and LAR have some amino acid similarity with PTPase lB, and that CD45 exhibits PTPase activity [Charbonneau, H., et al, Proc.
Natl. Acad. Sci. USA 85:7182-7186); Charbonneau, H., et al, Bioch. 86:5252-5256 ~1989); Tonks, N.K., et al, Biochem.
27:8695-8701 ~1988); Riener, P.A., et al, J. Immunol ~g~:23-28 (1989); Streuli, M., et al., Proc. Natl. Acad. Sci, USA
86:8698.8702 ~1989); Ostergaard, H.L, et al, Proc. Natl. Acad._ Sci. USAI ~:89S9-8963 (1989)]. Thus, PTPas~s co~pri~ a dlv~r.~Q

20~72~, famlly of proteins consisting of both transmembrane glycoprcteins and cytosolic procelns. While the enzymology of intracellular PTPases is being studied tLau, K.H., et al., Biochem J
257:23-36 (1989); Tonks, N.K., et al, Adv. Prot. PhosDh~tases ~:149-180 ~1989)], physiologically relevant subscrates ~or these cnzymcs havu not yet been identified, nor has the mode of rogulation of onz~atic ~ctivity bccn elucidatad, However, PTPase 18 from placonta and PTPase 5 ero~ bovlne 'orain have been reportod to dephosphorylate the phosphotyrosyl forms of Lnsulin receptor and v-src, respectively, ~ vitro ~Jones, S.W., et al, J. Biol. Chem. 264:7747-7753 (1989); Tonks, N.K., et al, J. Biol.
Cheml ~Ç~:6731-6737 (1988)~ and purlEied PTPa~o lB in~octed into ~gQQ~ oocytos delays insulLn-depandcnt m~turation tl'onk~, N.K., at al, Aclv. P~ot. Pho~q~n~.se~ 5:149-180 ~}989)1.
Ma}ignant trnn~ormacion i~ bclLovod to ~o~ult fram th~
cumul~iv~ o~Ecc~ o~ multlplo g~natic la~lonD. Two cLnfl~ of genotic lo~ion~ h~vo bo~n d~Lnu~. rho bost undar~eood cl~D~ l~
the do~inan~, actlvating mutntion~ th~e g~nornte oncoeenes ~Bishop, J.M., ~lQnS~ 305~311 ~1987~. A second class is the 1099 oi~ unction mutations that lead to neoplAstic growth.
Such mutations are associatod with ~rowth ~uppressor genas ~Hansen, M.F., et al., Ç~ll 53:172-173 ~1988); Weinberg, R., ~iochemistry 28:8263-8269 ~1989) which are incorporated by re~erence], and are sometimes referred to as tumor suppressor genes or anti-oncogenes.
It would be desirable to have a method of diagnosis for the presence of neoplastic growth (cancer) based upon identifying the absence of a functional portion of a suppressor gene or an alteration in the sequence of such gene and/or by the reduced activity of the protein product of the suppressor gene.
.It would also be useful to have a means oP counteraating this neoplastic growth by being able to supply a therapeutlc amount oP

~ ..

...... . ,. . .. , ~ .. ,; . ,. , . ,. ,. , ~ ;, . . .

f 7 2 ~
W O 91tl3173 PCT/US91/0143 the suppressor gene product to counceract the neoplastic growth caused by its loss or al~eration.
In addition, it would be desirable to have a vector that can be used in a wide range of expression systems to increase the expression of the gene products. It would be useul to have a means whereby l~rge amounts of the suppressor ~ne produce can be grown to mora fully dafine lts functlon, mode oE oparaeion, and to ba availabla for thar~p~utic usc, Summary of Invention We have now discovered PTPase genes are associated with neoplast1c transformation. Preferably, the gene ls a member o the PTPase gRna famLly, such as PTPlB, LAR, CD~5, PTPn~a 5, or T~cell PTPa~c. Moro prafer~bly, the gono i~ a PTPlB gano.
Thu hum~n PTPlB g~no i~ locac~d on ~ho long nrm Oe chromo~o~a 20, And loc~llza~ n~nr th~ lntarenca oE b~nd~ ~13.1 ~n~l qL3.2;
reP~rred to ~ loc~lizlng botwoo~ 20 ql3.1 13.2.
~ he pra~onc~ o~ m~lignnncy ln hum~lns c~n bo d~t~r~lna~ by assaying ior a del~tion Oe a functional portion or nlteration o a supprussor gana such as a PTPasa gen0, preiarably, the PTPlB
gena. This c~n be accomplished by tha usa o a nuclaic acLd probe based on the nucleotide sequence o tha suppressor gene or by the use of an antibody probe to the gene's expression product. In one embodiment an alteration in the suppressor gene can be determined by comparing a preselected cell from an lndiv~dual with a normal cell from that individual. In an alternative embodiment, the preselected cell is compared against a predetermined base line. Preferably, one compares the level of the suppressor gene product to a Nnormal" base level, with a low level being indicative of a neoplastic condition.
The suppressor gene is used to diagnose the presence o~ a cancer or precancerous condition, preferably for the presence of , a cancer such ns leukemia. ~ore preferably, acuee nonlymphocy~ic ~ W O 91/13173 2 ~ ~ ~ 7 2 ~ PCT/US91/01432 leukemia (ANLL). Preferred precancerous eonditions diagnosed are ~yelodysplastic syndrome ~MDS) such as idiopathic acquLred sideroblastic anemia or refractory anemia with excess blasts, as well as myeloproliferative disorders (~PDs) sueh as polycythemia vera.
The adtition of a thcrapeutic amoune of ehe suppressor ~ne produet to a diseased eoll will eountaraet th~ neoplaseie eondltlon.

Brlef Description Of the Fi~ures Figure l is a sche~atic of the nucleotide and predicted amino acid sequence for PTPlB eDNA.
Figuro 2A is n sehematic of the prim~ry transl~tlon prod~let of tho eDNA oE PrP~so lB ~nd T~enll PTPa~o.
Figur~ 2B i~ thQ amlno ~el~ nmont of th~ ~dduc~ c~rboxy tormlnl o~ PTP~ lB ~n~ T~eall PTP~
Figuro 3 L~ n pie~u~o o~ ~n a~1to~Adio~r~ph of th~ ln ~15EQ
t~an~l~tion p~oduet o~ th~ PTYlB cDNA.
Figure 4A is ~ ~eh~tie rapr~sontatlon of the PTPlB eDNA and the genomic PTPlB DNA.
~ lgure 4B is a r0presentation o~ the DNA sequ~ne~ oE the intron/exon and exon/intron ~une~ions ~or the 5 exons ~A-E) identified in Figure 4A.
Figures 5A-C are fluoreseenee ~ si~u hybriderization loealization of the PTPlB gene.
Figure 5D is a sehematic representation of human ehromosome 20 with a summary of the PTPlB gene plaeement.
Figures 6A-F are pictures of NIH 3T3 cells and 3T3 3-18 eells.
Figure 7 is a sehematic of a plasmid eontaining the PTPlB
cDNA.
Figure 8 is a sehematie of a elone eontaining the PTPlB
genome.

:., ~ , :.
,, ~

W O 91/13173 20~472S PCT/US9l/~143 ~

Figures 9 is ~els of PTPase lB. ¦
Figure 10 is a Western analysis of PTPlB monoclonal ` .
antibodies CE5-lK and AE4-2J.
Figure 11 is a picture of an autoradiograph of lmmunoprecipitation of 3T3 and 3-18 cells wlth PTPlB monoclonal antibodias .
Figura 12 is a Westo~n analysls Oe PTPaso expressed Erom vaccln~a inf~cted cells and PT~ onoclonal antlbodLes.
Figur~ 13 is a Western analysis of tumor cells from nude mice that were injected with neu TM infected cells.
Figure 14 is a Western analysis of 3T3, 3-18 and 43-5, a 3T3 cell llne transfectcd with tho PTPlB 8en~ undcr thh conerol o an inducibl~ promotcr.
Figure 15 i9 a picturc of an ~utorndiogrnm Oe an 1~ Yl~Q
tyrosino kin~o A~say.
Fi6urc 16 1J ~ ~C ~ tIC Oe p~bT~587.
F~guro 17A lo ~ ~ohomntlc Oe pAbT9110 and F~ura 17 sche~tLc oE pAbT9112.
Fi~ure 18 is a schematic oE an inducible PTPase expression plas~id.

Detailed Descri~tion o~ the InventLon We have now discovered that, in humans, PTPase genes are associaced with neoplastic growth. One mechanism for ~enerating loss of function ~utations is deleeion of a putative growth .
suppressor gene which may be accompanied by observable ;-chromosomal abnormalities. Two growth suppressor ~enes have been isolated based on this criteria lFriend, S.H., et al, ~at~re ~28:643-647 (1986); Fung, Y.K.T., et al, Science 236:1657-1661 (1987); Lee, W.H., et al, Science 235:1394~1399 (1987); Fearon, E.R., et al, Science 247:49-56 11990) which ara incorporated by reference], We ha~e ~ound that a daletion in a chro~o~o~e o~ a ~unctional portion o~ auch a PTPa~c gen~ i~ aasociat&d ~lth a .

~ W O ~I/I3l73 2 0 5 4 7 2 ~ PCT/US9l/~1432 neoplastio transformation. Thus, by using a probe for the ~ene, lt is possible to deter~ine whecher or not there is cancer. The probe can be a nucleotide probe or an antibody probe.
The pres~nce of malignancy can be d~termined by assaying for the del~tion of a functional por~ion or alteration o~ a ~uppr~sor e~ne. Prnf~rably, th~ suppresso~ gene is a member of th0 PTPas~ g~ne ~a~lly, e.g., PTPlB ~sometlmQs r~ferred to 3S
PTPase lB), T~eell PTP~se, LAR, CD45 a~d PTPase 5.
Figure 1 shows the nucleotido sequence and predicted amino aeid sequence for the PTPlB cDNA. The cDNA sequence of the PTPlB
gene has a 1305 nucleotide open reading frame, which predicts a proteLn oE l35 alnino ncids h~vlng ~ mol~eular weL~hc of abou~
49,966 ~in Fi~ura 1 the termin~tLon codon L~ m3rkod by nn X).
Nueleotido numberlng Ln Flgur~ 1 ~s on tha ri~ht wlth th~
~equonco numbared Ero~n th~ Eir~t ATG codon ob~or~d ln tho ~in~l~
opan rcading Pr~n~. ~Th~ und~rlln~d a~lno ~eid ao~u~ned~
eorre~pond to tho olL~op~p~ldo~ eho~ or ~on~truetlon o~ the degenerate oligonuelootide~ usQd to 3cre~n the eDNA library).
Tho Asn321 rosidue markLng the carboxy terminus of the p~rified 37000 dalton PTPase lB is underlined. A putati~e polytA)additLon site is obser~ed at nueleotides 3123-3128.
There is a similarity between the amino terminal 276 amino acids of PTPase lB and the expression product of the T e~ll PTPase cDNA. Figure 2 is a sehematic representaeion of the amino acid alignments of PTPase lB and T cell PTPase. Figure 2A shows the primary translation product of each cDNA represented by the individual lines. The filled boxes connecting the two lines indieates oligopeptides of seven or more amino aeids that are identieal between the two proteins. The numbers above the lines refer to PTPase lB a~ino aeid numbering. The PTPase lB
Asn321Gly322 peptide pair, whieh is also present in T-eell PTPase, ls indieated by NG. Tho eross-hateh~d box repregen~s the hydrophobie re~ion a~ the carb~xy tQr~inu~ oE the two prot~ina.

................ .

W O 91/13173 2 0 5 ~ 7 2 5 Pcr/us9~ 43~

Figure 2B shows the alignment of ehe deduced carboxy-termini Of PTPase lB (labelled PTP lB) and T-Cell PTPase (labelled PTP TC~.
Amino acids are shown in one-lecter code, and the a~ino acid idencities between the two regions are indicated. The hydrophobic region containing 8 out of l9 identical amino acids is boxed.
Th~ cDNA sequenca can b~ used ln preparing nucleotldo probes for datection of d~latlons ~nd alc~r~tions in eh~ suppressor gene. In one preferred embodiment one would use the cDNA
sequence to obtain a genomic clone by screening a human genomic library by standard techniques. The genomic sequence is preferably used to determine the chro~oso~al location. For exampl~, Fl~uro 4 i9 ~ sche~atic rQpro3enc~cLon oE che ~no1nlc structure and cDNA structuro or ~hc PTPlB gonc wiCh oxon/~ntron structure oE this gona shown. Soueharn an~ly~L~ oE t~ nomlc DNA id~nel~icd 3in~1u ~coRI, Bnm~ nd ~1ln~XII ~rA6rnan~ wh~ th~
coding r~gion oE th~ ~DNA w~ UDOd aa ~ prob~, Hownv~r, ~lnco a s1ngle HindXII ~lto i3 pra~ont ln tho cDNA ~aqu~nco, ch~ro ~y b~
an additional smaller gonomic HlndI~I fra8ment noc seen on the Southern blots.
Figure 4A is a schcmatic reprcs~ntation oE the cDNA clone with the open box representlng the 1305 nucleotide open readLng frame, upper figure, and the 13 kb genomlc fragment, lower figure shown. The genomic figure is not drawn to scale. Exons, identlfied by determining exon/intron ~unctions through comparison of cDNA and genomic DNA sequences, are shown as open boxes labelled A through E and are mapped onto the cDNA. Various restriction endonuclease sites are shown for both clones (Bm -BamHI; R ~ EcoRI; H - HindIII; Q - SacI; S - Sphl). Figure 4B is the DNA sequences of the intron/exon and exon/intron ~unctions for the ive exons (A-E) identified in Figure 4A. Ineron sequences are in lower case latters; exon sequences ar~ in upper case letters. Numbers above the exon se~uences corr~spand to the ~ ~VO ~1/13173 2 0 5 4 7 2 ~ PCI/US91/~1432 cDNA numbering in Figure 1. Consensus nucleotides for intron sequences [Shapiro, M.B.,et al, Nucl. Acids Res. 15, 7155-7174 tl987)] are underlined. The reading frame and translation products are shown beneath the exon sequences. Exon E has only an ineron-exon ~unction because it conta~ns the stop codon and the 3' untranslatud region.
A ~unc~ion~l portion of the ~ana i~ a scquencu the d~lation oE which nagatlvely aEects eha growth suppressing activity of the resultant gene product. Alterations of the ~ene can be distinguished from allelic variation by comparing the PTPase gene or gene product in the cell being teseed with a PTPase ~ene or gene product ~ro~ a "normal" cell (i.c. one not belleved co be malignnnt) fro~ the s~mu indivLd~
A delueion of ~ functlonal unit u~n bo any nw~bcr o nual~ocidu~ thnt rosul~ ln ~n alcer~d or l~ceiv~ ~on~ produc~.
Whun using ~ nuclootld~ p-sobo, tho dol~tlon L~ pr~Eo~bly a~
loa~ ~bout 100 nuolootldu~, moro pro~orably ~t ~OA9~ abouc 250 nuclaotidcs, ~t~ll mcrc pro~or~bly n~ la~t about 400 nuclcotidcs. In one pre~err~d umbodLmont the deletion is at least about one exon of the gena. For example, with the PTPlB
gene lt prufcrnb}y is a dele~ion of at least about 250 nucleotides, more pre~erably at least about 400 nucleotLdes, still more preferably, at least about 600 nucleotides. A
deletion of at least most of the nucleotides, preferably all the nucleotides, of one of the five exons, for example, exon A, ls also preferred for detecting the presence of cancer.
Preferably, the nucleotide probe corresponds to at least about 20 nucleotides of a human PTPase gene. More preferably, ~t corre~ponds to a~ least about 25~ of the cDNA of a PTPase gene, still more preferably, it corresponds to at least about 50~ of the cDNA of PTPase ~ene, even more preferably it corresponds to at least about 70~ o~ the cDNA of a PTPa~e gene, In another embodimene, the nuc}qotide probe corra~pondin~ ~o .

., ~ .. . . . . . ................. .

: ,: : . ,., , , : ,. . . : . : . ;: . . , : . :

~: .. . . .. : : . . :: : ; . : : . . . . .

W O 91/13173 2 0 ~ ~ 7 2 ~ PCT/US91/0143 ~

at least about 25~ of the coding sequence of a PTPase gene, more preferably, at least about 50% of the coding sequence of a PTPase gene, still more preferably, ae least about 70~ of the codLng sequence of a PTPase gene. Preferably, the PTPase gene is the PTPlB gene. The coding sequence refers to eLther the cDNA or the genomlc ~equences (i.e. the exons and che interv~ning sequences), bu~ the prob~ prefor3bly 19 ba~ed upon tho cDNA sequences. Ona can use a probe or RNA or DNA, but a DNA probe is pr~ferred.
When using a nuclcotide probe, one can determine whether there is a deletion in a functional portion of the gene by a variety of techniques well known in the art such as Southern blo~
analysis [Southern, E.~ 91~ L~ ~:503 (l975)1 or ln ~ieu hybridixation tLawrone~, J,B., ÇQ11 ~:Sl (1988)].
For ex~mplo, u~Lng ln ~ hybridization, le i~ po~slblo to dctermine whcthar or noe th~ prob~ wllL hybrldlx~ to ~ nuel~otld~
~oquonca occurrlng ln A call. I~ th~ probe ~1Oo~ hybridlzo wlth A
nuclootldu 3~quonca, 1~ lndlc~e~ th~t C~ c~ll eoncALn~ ~
nucleotido saquenco correspondlng to the proba, and chereEore does noe have a deletion o that sequence. This can be doterminad by ordinary techniq~es well known in the art. The probe can be labelled, pre-Ecrably with a radioactLve isotope, enzyme, fluorogenic, chemiluminescent or electro-chemical material. Preferably, one uses a probe corresponding to a genomic sequence to determine the presence or absence of a functional portion of the suppressor gene in a chromosome of the human tissue or plasma cell of interest. ' In another embodiment, one uses a nucleotide probe to compare the nucleotide sequence of the test cell (i.e. predetermined cell sa~ple) with a "normal~ cell from the individual to be tested.
Thus, one compares the hybridization in the test cell with the "normal" cell from that individual to determine if ehere is a difierence. Detecting a diEference in binding in calls irom the sa~e individual is indicativ~ o~ a change in tha "~ast" call, ~,;. ~ . . .. ` , '; "' ', ' ;. ' `; ` , ' `" ' . , ~

~; W ~ ~1/13173 2 0 ~ 4 7 2 ~- P~T/US91/01432 which is diagnostic of cancer for that cell. The predetermined cell sample Ls the cell sample you wish to test for maLignaney.
This can be from tissue, plasma, etc. The alteration or deletion of a functional portion of the suppressor gene, preferably PTPase suppressor gene, can indirectly be determined by the use of antibody probes using standard techniques.
The pre~qne~ of a malignnncy in humana ean be de~ermined as indieated h~r~ln. Whon th~ g~ne ls a PTpAs~ ~ene it i9 preferably used ~or the diagnosis of eaneer of precancerous condition More preferably, the diagnosis of a cancer, preferably leukemia such as acute nonlymphocytic leukemia (ANLL). Preferred precancerous condltions dia~nosed are myelodysplas~ic ~yndromo ~M~S) auch ~s ldlopnchie ~equired siderobl~stie an~ or roEraetory ~no~in wich ~XCQ3S bl~t~ nnd myolop~olleraclve disordars ~MPD9) sueh as polyey~h~ml~ vor~.
In sueh ~n a~s~y lookln~ eor deloelon~ or ~lt~rnntlon~ ln th~
PTPlB gono i~ pre~rrod.
In on~ p~oorrod ombodlmone, i~ wo(lld b~ us~ul ~o uso ~
eoektail eontainin~ probes to ~arious funetional portions of the suppressor (eDNA) gene as well as to the suppressor (genomic) gene.
The PTPlB suppressor gene maps as a single gene eopy on the long arm of chromosome 20 in the region ql3.1-ql3.2.
For example, loealization of the gene to a ehromosome ean be determined by non-isotopie, flouresenee deteetion of in situ hybridiza~ion with a probe based on the genomie suppressor gene.
Chromosome banding analysis ean be periormed by a ~ariety of means well known in the art. For example, one ean use DAPI-staining of BrdUrd-ineorporated ehromosomes or Giemsa-t~ypsin banding.
See, for example, Figure 5 whieh illustrates the loealizacion of the PTPlB gene to a single ehromosome loeus. As illustraced : :~ :: : :: .: :, :, :: .::.. : : . - : .: . ~ :: ::, : . ::;. :: ::

W O 91/13173 2 0 ~ ~ ~ 2 ~ ` PCT/~S91/0143 ~

in Figure 5 A and B, hybridization can be unambigously visualized in one cell by the identical labelling of both sister chromat~ds on each of the two ho~ologous chromosomes. Figure 5A shows DAPI-staining of metaphase chromosomes from normal human lymphocytes. The bands are sufficient to allow identification of individual chromosomes. The large arrow ac the top of the figure points to a pair of chromoso~e 20s. Tho small arrows indicaee chromoso~e 19, which ls dL3tinguish~d by a more prominent centromere and paler arms.
Figure 5B shows fluorescein avidin detection o hybridi~a~ion with the biotin-labelled PTPl~ genomic DNA probe. Both homologs .
of chromosome 20 label on both sister chromatids and no other chromo~omal sites arc laboled. Tho bandin8 nnalysis d~monstrated that this w~ls chromosome 20 nnd r~sulced in a ragional localization oE the 3ignal to tho raglon ql3.l-~13.2. Flgur~ 5C
illustrato~ th~ po3ition o~ tho c~ro~osom~ nal on ~ovGr~l ~nlarg~d chromo~omo~ ~t ~ hLgh~ ~a~nLElGa~lon ~howlng nll~nm~nt o~ the tot~l chro~o~o~ D~PX~fluDro~conc~ on th~ loE~ oE ~aah pair wlth the Eluoroscain avidin hybrLdization signnls on the right. Figure 5D is a schematic representation of chromosome 20 with ~ 3uMmary of th~ PTPlB gen~ placemont baaed on tho analysia from l0 DAPI~banded chromosomes scored independently. The PTPlB
gene localizes near the interface of bands ql3.l and ql3.2.
Similarly, localization of the PTPlB gene by measurement along the length of the chromosome based on the average of 27 determinations maps the gene in the middle of ql3.l. ,-Thus, one can use a probe to directly determine the deletion of a segment of the gene on the chromosome.
One can also prepare a nucleotide se&ment corresponding to the eDNA suppressor gene or corresponding to a functional portion thereof by standard techniques, which can be used to produce the ~ene product ~protein). For example, one can use a degenerative oligonucleotide probe to obtain a cDNA probe irom a human ,: .~ . .. .

2~ll7~5 ~ W O 91/13173 PCT/US91/01432 -~f~

placental cDNA library. Alternatively, one can chemically synthesize the predicted nucleotide sequence. By either means one can prepare a nucleotide sequence corresponding to a sufficient portion of the suppressor gene cDNA to produce a protein havLng growth suppressor actLvity. For example, a nucleotide i~equenca for the PTPlB cDNA is set forth in Figure 1.
Thls nucleotlde ssgment ls 3143 nucluotldes ~nd has an observed 1305 nuclootida open readlng frame which results in a predicted protein of 435 amino acids. This nucleotide segment or a functional portion thereof can be used to generate a clone capable of expressing the PTPlB gene product by standard techniques.
For exa~ple, a vector concaLning ~ PTPaso ~ono cDNA a~qucnc~, preferably, the PTPlB gcne cDNA, polyndonyl~tlon ~q~enco~ ' downstraam 3' ~rom th~ saquonce nnd preEer~bly ~1aQ cont~inin~ fl replicntLon origLn c~n bo u~u~ ~o tr~ns~oct A CO11~ Th~ vootor wlll ~l~o con~ln 1~ promot-~r to parln~e ~xpra~Lorl o~ th~
suppr~ssor gene. Pro~otar~ u~ed Ln t~le vector c~n be ~ny o the known promoters and the choico is governed by the host cell one wishes to use to thereby permit expression of the desired product in the host cell of choice. Ratroviral promoters aro preferred.
Such promoters include retroviral promoters such as Akv, SL3-3 and Friend viruses. The vector can preferably also contain an enhancer, such enhancers are well known in the art, for example, ~iral enhancer. ~ore preferably, it contains an enhancer which is tissue specific. Preferably, the vector also contains a warker gene to aid in detection of transformed cells. Such markers are well kno~n in the art and can include an antibiotic resistance gene such as the bacterial neomycin resistance gene, which confers a dominant selectable resistance to the antibiotic G418 in eukaryotic cells. A vector such as one corresponding to a defective virus or benign virus, e.g., reco~binant vaccinia virus, can be used.

W O ~l/l3l73 ~ O ~ 4 7 2 5 YCT/US91/0143 ~

Although different cell lines can differ in cheir ability to take up and express the transfected suppressor gene DNA by use of appropriate promoters, a wide range of host oells can be used.
For example, NIH3T3, CHO, COS, SF-9(ATCC No. CRL 1711) and SF-21 are preferred. Typically, mammalian cells would be preferred, but the cell lines of the present invencion are not limited ther~to. For oxamplo, one can use a b~oterial cell or insect eell to expres~ the protein.
The veceor can b~ used to transform cell lines, for example, by being in~roduced into psi/2 ~ecotropic) and psiAM
(amphotropic) cell lines by a variety of methods well known to the art. The preferred method is the calcium phosphate co-preclpitation method [Sce Wigler, ~t nl, C~ll lh:777~785 ~1979), which is incorporated by ro~oronce~. Th~so cell llnos can conseitutively produco in~octlous, roplLeation d~Eeetlve murine loukomi~ vlru~s eon~ainlng n ~ono~n~ dorL~tl fro~ eho ~octor. ~Cone, ~t ~ Q~ 63~ 6353 ~1984); Mann, 153~159 ~1983)~. Two ~ys Eollowln~ tr~n~etion, e~
~an bo sslected by lookLn~ ~or thQ markor, i.e. ~ntlbiotic resistanee gene, which the voctor would also preferably contain.
Antlbiotic r~sistant elones, ~.g. G418, would b~ evidenc in a short period of tl~e, typically 7-10 days.
The antibiotic resistant psi/2 and psitAM clones 3re isolated and clones produc~ng large amounes of the suppressor gene product are selected. These cells are then cultured in a standard medium and harvested as needed. For example, the protein may be expressed as insoluble inclusion bodies. Such inclusion bodies ean be isolated by standard means, e.g., centrifugation of cell lysates. Thereafter, the protein can be readily isolated by standard isolation techniques.
The protein produced from these eells ean then be us&d to generate an antibody to the suppressor gene produet (protein).
The antibody generated ean be polyelonal or monoelonal depending ~3~ ~h3 W O 91/13173 PCT/US91/0~43 upon the particular application for which ic is designed, Such antibodies can be prepared by techniques well know to the skilled artisan. For example, the protein or an antigenic portion thereof can be con~ugated to keyhole limpet hemocyanin (KLH) and used to raise an antibody in an animal such as a rabbit.
Typically, the poptide-KLH conJugate is in~ected sevnral times over a perlod of ~bout two months to generat~ nncibodies~ The antibody is then collected from serum by seandard techniques.
Alternatively, monoclonal antibodies can be produced in cells which produce antibodies to the proeeins by using standard fusion techniques for forming hybridoma cells. EKohler, G., et al., Natur~ 256:495 (1975) which is lncorporated by reference].
Typically, this involves usiny. an antibody producing c~ll with an immertal cell lln~ ~uch ~g ~I myelomn cell Co prod-lco tho hybrid c~ll. Alt~rnativoly, monoclonnl ~ntibo~le~ can b~
produccd from aoll~ by th~ matho~ ef 11u~a, ~t nl, ~Ql~M~
1275 (1989) whlch 13 lncorpor.qtod h~roln by r~E0r~nc~.
~ or axampl~, hybrido~9 can b~ gen~r~tad by lmmuni~ntlon of mice with viable calls transfor~ed by the suppressor gcno~cont~inLng vactor, which ~xpross tho suppressor geno product. Preorably, thosc colls express the i'ull langth protein such as the iull length PTPlB gene product shown in ~igure l.
Using the full length protein as the immunogen, it is possible to generate a collection of monoclonal antibodies with specificities that span the entire length of the protein. This is as opposed to the usé of peptide immunogens, or short polypeptides generated by prokaryotic systems, which present a more limited number of epitopes from the original protein and hence, raise an immune response of more limited specificity, such as would be the case with the Charbonneau protein (Proc. Natl. Acad. Sci. USA 86, ~y~). The mice can be immuni2ed intraperitoneally ~I.P.) with a sufficient number of ~iable cells oF the host cell, which is to be transfected by th~ suppre~sor ~ene veccor. This can th~n b~

W O 91/13173 2 0 ~ ~ ~ 2 5 PCT/US91/0143~

followed i~mediately by an I.P. injection of, for example, cyclophosphamide in H20. The cyclophosphamide treatment is repeated one and two days following the primary injection. About two weeks following immunization, mice are then injected with a sufficient amount of the transEormed host cells and then allowed to rest for another cwo weeks, ae which time thc eneire procedure ls ropeat~d. Four days following eho ~ccond in~eceion of ehe transormad Cell3, tho al1im~ls nre ~acrLficod 3nd their spleens ob~ainad for the fLrse fuslon.
Hybridomas are produced by fusing cells by typical techniques, such as from immunii~ed mice with SP2/O myelo~a cells by a polyeehylene glycol ~PEG) method. Cells are asceptically removed from immunized mice and a single cell suspension of thc spleen cells obeaLned by parfuslng thc splaen wleh ~rwn-frflc media ~e.g. DME). Sploon cclls And my~lo~a cclls ar~ ~1ix~d togecher ~t a ratio, ~or ex~plo, o~ 5 to l, 3pl~0n ~ co myoloma colls. Tho c~lls nr~ ch~n c~n~rifugod ~nd th~
supe~natant r~mo~ed by asplr~tlon, Tho c0115 aro than ~rown in metium by standard tcchni~ues. Hybridomas, which grow aEeer the usion procedure, are then screened for secretion of anei-human suppr0ssor gono product antibodies by an ELISA assay o~ a cell lysate. Hybridomas thae produce posleive results, are expanded and cloned by limiting dilution to assure that the cells and resulting antibodies are indecd monoclonal. Hybridoma colonies, that test positive for the presence oi antibody to the human suppressor gene product are diluted in media to a concentration of, for example, 5 hybridoma cells per milliliter. Once colonies grow, the supernatants are again tested for the presence of antibody to the human suppressor 8ene product. If the results are positive when tested by an SLISA assay, the colonies are cloned again by limiting dilution.
The resultant antibodies can then be used as probes in the assay method discussed above. Antibody probes are particularly . " " .. . . .... . ... . .. . . . . . . . . . . .. . . .

W O 91/13173 PCTtUS91/01432 sensiti~e to alterations or deletions in the suppressor gene, such as a PTPase gene, preferably PTPlB, which affect the resultant protein products. Thus, the an~ibody probes provide a simple and efficient means of determinin~ whether a deletion or alceration of the gene has affected a functional unic, By using ~ntibody probes it is possible to det~rmin~ both the level of cxpr~ssed prot~ln and whether thero has baen a change in expression. One c~n co~para ros~lts against base linc levels obtained for the materlal belng sampled, e.g., level of suppressor gene product in blood, or by comparing a test cell (preselected cell) from an individual with another cell from that individual, which i9 noe believed eo show malignancy. IE ehcre i5 a change (Q.g" absenco of r~ctivo proteln o~ altor~d ~lectrophoretic mobility) lt l~ LndLcatlvo that tha t~ c~Ll 1 ~lignant, Furthor, on~ c~n tak~ coll ~mplo~ ~rom th~ ~amu lndlvldual ~t ~arlo~ timo~ to provl~Q lovol8 Oe co~p~ioon.
Thl3 c~n ~130 ~0 don~ ~or tha nuclao~ probo~.
In accordanco wlth thls in~ention, an anti~ody or c~ck~ail o probes, e.g. nntibody probcs, can be used ~or detection. The probes, e.g. ~ntibodics, can be labelled directly with ~ reporter or indirectly with a member of a specific binding palr using conventional techniques.
Specific binding pairs can be of the immune or non-immune type. Immune specific binding pairs are exemplified by anti~en-antibody systems of hapten/anti-hapten systems. These - include fluorescein~anti-fluorescein, dinitrophenyl/anti-dinitrophenyl, biotin/anti-biotin, peptide/anti-peptide and the like. The antibody member of the spec1fic binding pair can be produced by customary methods familiar to those skilled in the art, Such methods involve immunizing an animal with the antigen member o~ the specific binding pair. If the antigen member o~ the sp~cific bindin~ pair is not immunogenic, e.g., a hapten, it can be covalontly coupl~d W O 91/13173 2 0 ~ ~ 7 2 ~ PCT/US91/0143 to a carrier protein eo r~nder it immunogenic.
Non-lmmune binding pairs include systems wherein the two components share a natural affinity for each other but are not antibodies. Exemplary non-immune pairs are biotin-streptavidin~
intrinsic factor-vitamLn B12, folic acid-folate binding protein and the like.
A variety of methods ar~ ~vailablc co covalently labol antibodios with membors o~ spociflc blnding pairs. Methods are selected based upon tho naturo of tho member of ~he 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 available actlve derivatives. Somc of these arc biotin-N-hydroxy-sucinimide which binds to aMin~ group~ on proteins; bieoin hydrazido whLeh binds to c~rbohydrn~ ~oio~io~, ~ldohydo~ nndi carboxyl ~roup~ n c~rbotllLmld~ couplin~; rlnd biotin maloi~ido ~nd iodo~ic~tyl biotin which blnd to ~ hydiryl group~. Fluor~sco~n ~fl bu COUplfld to procaLn ~ln~ groups u~l-ng ~luoroscein isothiocyanate. Dinitrophenyl groups can b~ coupled to prot~in a~ine groups usng 2,4-dinitrobenzcne sulat~ or 2,4-dinitrofluorobcnzene. Other standard methods of con~ugaCion can bo employ~d to couple monoclonal antibodies to a me~bcr of a specific binding pair including dialdehyde, carbodiimide coupling, homofunctional crosslinking, and heterobifunctional crosslinking. Carbodiimide coupling is an effective method of coupling carboxyl groups on one substance to amine groups on another. Carbodiimide coupling is facilitated by using the commercially available reagent l-ethyl-3-(dimethyl-aminopropyl)-carbodiimide (EDAC). ~, Homobifunceional crosslinkers, including the bifunctional imidoesters and bifunctional N-hydroxy-succinimide esters, are commercially available and are employed for coupling amine groups on one substance to amine groups on another.

;'`~.

, . , ! , ' . ~. . ' : ~ ' : ` : ' ~ W O 91/13]73 2 0 ~ ~ 7 2 ~ PCT/~S91/01432 HeterobLfunctional crosslinkers are reagents which possess different functional groups. The most common co~mercially available heterobifunctional crosslinkers have an amine reac~ive N-hydroxysuccinimide ester as one functional group, and a sulfhydryl reactive group as the second functional group The most common sulfhydryl reactive groups are maleimides, pyridyl dLsulfides and nctlvo halogen~. Ono of the functional groups cAn bo a photo~ctlvo arJl nitrcna, which upon irradia~ion reacts with a variety of groups.
The detectably~labelled probe, e.g., antibody, detectably-labelled antibodies, or detectably-labelled member of the specific binding pair is coupled to a rcporter which can be a radioactive i~otopo, onzyme, fluorogenic, ch~miLum~nesccnc or electroch~mLcal matorials. Two commonly used radlonctlv~
isotope~ ~r~ 125~ ancl 3~l, Scand~rd rndio~c~lva i~ocoplc labolin~ procodura~ includ~ tho chlorn~in~ T, lnctop~roxLdfl~ ~nd Bolton-~lunt~r mothod~; ~or 12gI nn~l roducti~n m~thyl~lon ~or 3~
Enzymes suitable ~or use in this invention include, bue are not limitcd ~o, horsoradish peroxidase, alkaline phosphaease, ~-g~l~ctosLdase, glucose oxidasc, luciferase, ~-lactamase, urease and lysozyme. Enzymo labeling is facilitated by using dialdehyde, carbodiimide coupling, homobifunctional crosslinkers and heterobifunctional crosslinkers ~s described above for coupling an antibody with a member of a specific binding pair.
The labeling method chosen depends on the functional groups available on the enzyme and the material eO be labeled, and the tolerance of both to the conjugation conditions. The labeling method used in the present invention can be one of, bue not limited to, any conventional methods currently employed including those described by En~vall and Pearlmann, Immunochemistry 8:871 (1971), Avrameas and Ternynck, l~mgneghgmiLs~Y 8:1175 ~1975)l - ~ . . . .

W O 91/13173 ~ ~ ~ 4~ ?J ~ PCT/US91~0143 Ishikawa et al., J. Immunoassav 4 (3):209-327 (1983) and Jablonski, ~ 8iochem. 148:199 (1985), which are incorporated by reference.
Labeling can be accomplished by indirect methods such as using spacers or other me~bers of specific bindin8 pa~rs. An exnmple o~ this is th~ do~ec~ion o a biotinylatcd antLbody with unlabellod str~pcnvidln and blotinylnccd anzyme, wlch ~troptnv~dLn and biotinylae~d ~nzym~ belng added either sequentially or simultaneously. Thus, according to the present invention, the antibody used to detect can be detectably-labelled directly with a reporter or indLrectly with a first member of a speciiic binding p~ir. When the ~ntLbody i~ couplod to ~ first momber o~ a speciPic bindin~ pnir, then d~toction i9 e~fccc~d by r~acting tho Dintibody-flrst mu~bor oE n ~pocific bindlng complax with tha soaond m~mbar oE thQ binding p~ilr which 1~ boll~i or unl~ibcllod ui~ m~nti~n~ bova. ;~
Morcov~r, tho unlnbolLed do~ioctor nnelbody cnn bo d~e~ct~d by reacting the unlnb~llcd ~n~ibody with ~ lnbellad an~ibody speciic for the unlabelled antibody. Such an anti-antibody can be labelled directly or indLrcctly using any of tho approaches discussed above. For example, the anti~antibody can b~ coupLod to biotin which is detected by reacting with the streptavidin-horseradish peroxidase system discussed above.
One preferred embodiment utilizes biotin. The biotinylated antibody is ln turn reacted with streptavidin-horseradish peroxidase complex. Orthophenylenediamine, 4-chloro-naphehol~ or tetramethylbenzidine (TMB) can be used to effect chromogenic detection.
Thie preferred immunoassay format for practicing thLs ; inve~tion is a forward sandwich assay in which the capture reagent has been immobilLzed, using conventional techniques, on the surface of tha support.

~ W O 91/13173 2 V ~ ~ 7 2 5 PCr/US91/01432 Suitable supporcs used in ass ys include synehetic polymer supports, such as polypropylene, polystyrene, substituted polystyrene, e.g., aminated or carboxylated polystyrene;
polyacrylamides; polyamides; polyvinylchloride, etc.; glass beads; agarose; nitrocellulose, etc.
It ls also possible to treat an individual suffering from a cancerous or precancerous conditlon, preferably tha ones b~ing diagnosed by the probos by supplying a therapeutic amount of the suppressor gene product. Thls can be accomplished by a number of methods known in the art. [Williams, D.A., et al, Nature 310:476-480 (1984); Cepko, C.L., et al, Cell 37(3~:lO53-lO62 (1984) which are incorporated by reference]. For example, one can use geno transE0r tcchniqucs to propare a vector containing n nucleotid~ ~aquencc corraspondLn~ to ~he suppros~or gen~ or ~unction~l ~ragm~nt ther~oE, pro~ably A PTPa3~ ~en~, ~or~
pro~cr~ly PTPlB ~nd u~e ~uch ~ voctor to ~ran~or~ tho mAllgnnne ccll~, Tha vuctor 1~ prceor~bly ~ rctrovi~l voceor ~uch dc~cribed by Brown ~nd Sco~t, in 2~ 5L~D1D~, vol ~II, A
Practical Approach, ch. 9, ~Rotroviral Vectors~, CRL Press ~l987), which is incorporated herein by reer~nce, Preferably, thc suppressor gene used is the PTPlB ~ene or a ~unct50nal fragment thereof.
Alternatively, one can use, for example, a PTPase transformed cell line to produce large amounts of the PTPase gene product, e.g., PTPlB gene product, which is then isolated and purified so that ~t is substantially free of pyrogens and endotoxins. The gene product is preferably purified so that is is at least about 90~ pure, more preferably at least about 95~ pure, still more preferably at least about 98~ pure. The purified protein Ls then packaged by standard pharamaceutical procedures so that it can be delivered to the malignant cells, such as by in~ection, carrier-linked preparations, etc. A therapeutically effective ,",, ~, ~" "~ ; ,.,, ,~ ' ' W O 9J/13173 2 0 ~ 4 ~ 2 5 PCT/US9l/0143 amount of the purified protein is used. The therapeueically effective amount can readily be determined empirically based upon the present disclosure.
For example, a PTPase lB overexpressing cell line, which is subsequently infected by a vlrus expressing the a ~u oncogene, e.g., MLV TM D~-neo virus, has greater resistance to neoplasclc tran3form~tLon than an oquivnlent c~ll not having PTPase overexpras310n. The focL ar~ sm~ller and the cells appear less morphologically transEormed.
See, for example, Figure 6 wh$ch $11ustrates the reduced susept$bility of PTPase lB express$ng cells to transformation by the human ~_ oncogene. As illustrated in FLgure 6 thc foci inducad in NI~13T3 cells ~Fi~uro 6B) nro largcr and donsor than th~ foci ind~ced in 3T3 3-18 cells ~6E) and the NLtl3T3 calls show ~ charnctarl~tlc tr~nsEormo~ morphology ~6C), whlle 3T3 3~1~
cells do not ~6F). F~Kur~ 6A ~howa A monol~r o~ mock ln~cted NI~13T3 coll~. FLguru 68 ~how~ ~ Ino~ol~yor oE MLV ~M ~ noo virus inEcctod NIH3T3 c~ . FLgur~ 6C show~ ~ C~8 rcsistant colony of MLV TM nQ_-noo virus infected 3T3 cells. Figure 6D-F
are the sa~a as 6A~C but use 3T3 3-18 cells instcad of NIH3T3.
The prcsant invention is furth~r illustr~ed by the ollowLng examples. These examples are provided to aid in the understanding of the invention and are not to be construed as a li~itation thereof.
Isolation and Characterization of a PTPlB cDNA. -~
On the basis of partial amino acid sequence for a truncated PTPase lB [Charbonneau, H., et al, Proc. Nael Acad. Sci. USA
85:7182-7186 (1988)], two mini~ally degenerate oligonucleotides were used to screen a human placental cDNA library (Clontech).
The oligopeptide sequences (amino acid numbering based on Charbonneau, et al. ~çh~ 86:5252-5256 (1989)~ and the corresponding degenerate oligonucleotides used were (degenerate positions are shown within parenthesis) 1) Trp96-Glnl02 and the , 3 ~ J ~ ~

anti-sense oligonucleotide 5'TG(T/C)TCCCA(~/A/T/C)ACCAT(T/C)TCCCA; Z) Glul32-Thrl38 and the sense oligo~ucleotide GA(C/A)ATGAT(A/C/T)TT(T/C)GA(G/A)GA(T/C)AC. The antisense oligonucleocide designed from oli~opeptide l was chos~n for fur~hcr usos aa n ~equencLng prim~r to idcntify th~ S' end of any oycn re~ding ~rame in putativ~ cDNA clon~s, 500,000 plaque forming unLts of a ~gtll recombinant cDNA
bacteriophage llbrary, prcpared from human placental poly(A)+mRNA (Clontech), were plated at a density of 25000 bacteriophagl per lS0 mm petri dish. Nitrocellulose filter lifts taken from each plate were chall~nged for hybridization wLth the two degencrate oligonucleotides. Radiolabelling of oligonucleotLdas wlth T4 polynucleoeide kinnse, ollgonuclaocLd~
puriEicatlon, ~nd fLlt~r hybrLdiz~elon conclit;Lon~ wor~ a~a describad by Au3ubal, F.M, ct nl, Ç~b~ 3~h LoL~LL__~
Grone PublL~hln~ A~ocL~ nd Wil~y~Int~rsci~n~, N,Y" N,Y, (1987), The ~inal hybridlza~lon w~3h conditions wer~ 2X SSC, O,l~ SDS at 42C ~or 30 minutes, Bacterlophage DNA was prepared and subJected to restriction digestion and Southern analysis by standard methods tAusubel, F,M,, et al, Curr~nt Protocols I~ Molecular Biologv (supra)] using the radiolabelled degenerate oligonucleotides separately as probes.
DNA fragments that hybridized to the degenerate ol~gonucleotides were introduced into the sequencin~ vector pGEM3Z (Promega) for DNA sequence determination by the chain-terminating method [Sanger, F., et al, Proc. Natl ~Acad.
Sci. USA 74:5463-5467 tl977)~ using the Sequenase reaction (United States Biochemical).
The ~ itro transcription reactions were carried out using the Riboprobe GeminL System II ~Promega). RNA transcripts ware translated 1~ YlS~ using a rabbit reticulocyte lysate ~Promega), The [35S]m~thionine labellod ~ranal~tion reaa~ion W O 91/13173 2 0 5 4 7 2 ~ PCT/US91/0143 ~

products were displayed by SDS/PAGE [Laemmli, U.K., Nature 227:680-685 (1970)] and visualized by autoradiograph [Ausubel, F.M., et al, su~ra]
A human placental cDNA library was screened with de8enerate oligonucleotldes deduced from che partial amlno acid sequence of the PTPaso 1~ as described above. One recombinant bacceriophage wa~ shown by Southern analy31s to contain a DNA fragm~nt of 3.2 kb that hybridlzod to both of tho dagenerat& oligonucleotides.
The fragment was subcloned into pGEM3Z by standard techniques, described above and design~ted pAbT9611. Figure 7 is a schemaeic of pAbT 9611, whLch is a plasmid containing a cDNA clone encoding the entire protain coding region of PTPlB and a bacterial replicon ~pCEM3Z, Promega) which codes Eor an ~mplcillin resistnnce gene or growth and s~l~ctlon ln ~ çQl~ A snmple oE
thi~ plas~id hn~ bcun dapo3it~d wi~h tho Am~ric~n Typ~ Culturo Colloction ~ATCC) undor tha Bud~po~t T~nty Imd ~ivon J\TCC
Acc~s3ion No. 40768. Th~ DNi\ ~qu~n~ WMg tlctar~lnfl~ 'hd cDN~
s~quencQ o th~ PTPlB ~nu, tho do~uc~d eransl~tlon product, nnd sequences corresponding to the degoncrate ol~gonucleotidcs used in the cDNA ~loning are shown in FLgure 1. Ihe observed 1305 ~.
nucleotide open readLng ~rame pradicts a protein of 435 amino acids of molecular weight 49,966. The amino terminal 321 amino acids encoded by the cDNA are identical to the empirically determined a~ino acid sequence of the 37000 dalton PTPase lB of Charbonneau, H., et al., Proc. Natl Acad. Sci USA 86, su~ra.
There is similarity between the amino terminal 276 amino acids of the PTPase lB and the translation product of the T
cell-specific PTPase cDNA [Cool, D.E., et a}, Proc, Natl. Acad.
Sci. USA 86:5257-5261 (1989)]. There is ~ery little similarity in the remainder of the amino acid sequence deduced rom the cDNAs except for a region ~lear the carboxy-terminus in which there are 8 out o~ 19 identical amino acid residues. This re~ion in both proteLn se~uences also exhibi~ a general hydrophobic .~ ,.
, :;'.. :''. ' ': ' .: ,, .,, :;.. ,, .. , . , ,'~ . ,.: ' ;'' ` . , . :

~ u ~
~ W O 91/13l73 PCT/VS91/01432 ~,;J'5'~

character as ~alculate~ by the mechod of Hopp and Uoods, Proc.
Natl. Acad. Sci. USA 78:3824-3828 (1981). Figure 2 shows a schematic of the ove~all similarity of PTPase lB to the T cell PTPase and an alignm~nt of the ted-~ced carboxy-terminl of the two proteins.
Th~ eDNA clone, pAbT9611, was used as a t~mpl~ce for ~n vitro tran~cripcion and translation. U~ing eh~ full~length cDNA, the ma~or translation product w~s ~ prot~in wlth an apparent moleeular weight o~ 50,000 as measured by SDS/PAC~.
When the cDNA was cleaved prior eo transcription with HindIII
at nucleotide 106 of the coding region, no high molec~lar weighc translation products were obtain~d. Although several lower mol~cular walght sp~cies wor~ obtain~d u~lng th~ full l~ngth transcript, no 37000 dalton translatLon producc was ob~orv~d.
The lowar mol~eular walght sp~ei~ y eorr~pond ~o t~n~lA~ion produet~ Inltiatod at lnt~frlAl m~thlonlno~, or, ~lt~fl~tlv~ly, ~
pr~nturQ tr~nsl~tlon ~r~lna~Lon produe~. T~o r~ult~ erom tho in vitro translation are shown ln Flgure 3.
t35S}m~t~ionine labell~d transl~tion products derived from th~ DNA clone were displayed on a 120 polyacrylamid~/SDS gel.
N~mbers to the left of Figur~ 3 refer to protein moleeular weights in kilodaltons. Lane A: Radiolabelled molecular weighc markers (Amersham); Lane B: Translation products from transcriptiorl of the eDNA digested with BamHI which eleaves the plasmid at a site 3' to the eDNA lnsert; Lane C: Translation.~.
produets from transeription of the cDNA digested with HindIII
whleh eleaves within the eDNA coding sequence, Lane D: No added RNA. ;
The analysis o~ the eDNA indicates that the primary ~;
translation product of the PTPlB gena is a 435 amino aeid protein with a predieted molecular weight o~ 49,966, which is in contrast to the 321 amino aeid PTPase lB protein previously puriEied and .' sequanc~d [Charbonneau, H., et al, ~iQgh~ ~:5252~5256 ~1989~].

W 0 91/13173 2 0 ~ ~ 7 2 ~ PCT/US91/0143~

Using a PTPase lB antibody fraceion (provided by N. Tonks, University ~f Washington), we observed thac PTPase lB expressed from cDNA clones in NIH 3T3 cells appears eo be a 50,000 dalton protein.
Thus, we expect that the 435 amino acid PTPase lB is the functional species, and that the carboxy-terminus may perfo~m a necessary role, perhsps in regulating enzymatic actlvity, controlllng access to substratns, or possibly Cargetine the protein eo the cytoplasmic surface Oe th~ membrane. Th~ overall hydrophobic character and conserved amino acids at the carboxy-ter~inus of the PTPase lB and the T cell PTPase (deduced from cDNA sequences) suggests that this region of the protein is important. The fact that 50~ oP the PTPasc lB ac~Lvity is obtalned ~rom tho particulatc ~ractlon [Tonks, N.K., ~t ~l, I.
~LL~ 5h~ , 6722~6730 ~l988)] au~go~es that PTPa30 lB mny bo mcmbrnn~ as30cl~tod.

A hum~n gonomic lLbrAry, p~pnro~ ln thu lambd~ v~ctor EM~L-3 Sp6/T7 (Clontoch), was screened Eor recombinant bacterioph~g~ cont~ining soquences that hybrldized to the PTPase lB cDNA clon~. Approximately 2xlO6 rocombinant bacterlophage were screened as described by Ausubel, F.M., et al, supra with the probe radiolabelled using [c~32p]CTp in a random hexanucleotLde priming reaction [Feinberg, A.P. et al, Anal.
Biochem. l~:6-13 (1983)]. Genomic clones were subjected to Southern analysis using either the cDNA clone or various oligonucleotides derived from the cDNA sequence as probes.
A genomic clone was isolated by screening a human genomic library with PTPlB cDNA, and designated LAbTlO-l. See Figure 8. LAbTlO-l is a lambdaphage clone containing 13 kilobases of the human genome that encode a portion of the PTPlB gene, The insert was cloned as a Sau 3A ragment into the Ba~Hl si~e of . .. :

. . . .

~ W O 91/]3173 2 0 5 ~ 7 2 5 PCT/USg1/01432 the lambda cloning vector EMBL-3 Sp6/T7. In the cloning process the human DNA replaces the stuffer fragment and the restriction sites for Eco Rl, BamHl and Xba 1 were removed. A sample has been deposited with the ATCC under the Budapest Treaty and glven ATCC Accession No. 40767. Two genomic fragments of approximately 13 kb ~ach were isolated. Restriction analysis indicates that the two clones ~xhiblt noarly 90~ overlap.
Southern analysls and DNA 3~quenclng shows thae the genomic clones e~compass the 3' 2700 nucleotldes of the cDNA. Neither clone contains the 5' portion of the PTPlB gene, howev~r. The carboxy-terminal 272 amino acids of PTPase lB plus the stop codon are encoded by ive exons. A single exon encodes amino acid~ Asp289-Clu326 which includcs Asn321, the c~rboxy-termin~l ~ino acid oE tho 37000 dalton Eorm o thc PTP~o lB. Tho ~xon/intron ~tructu~ o~ PTPlB i~ ~hown in Fly,uru 4. Sou~h~rn ~n~lysl~ oP g~nomlc DNA idon~iPlfl~ 4inelo EcoR~, Bm~ nd ~lndIII ~rA~m~nt~ wh~n th~ co~lng rcglon Oe tho cDU~ wa3 usad ~3 a proba. A slngla HindIII 9i~ i9 present in th~ cDNA sequence suggesting that there ~ay be an additional, smaller genomic HlndIII ira~m~nt not seen on the Southern blots. Also, the geno~ic clone contains an EcoRI site in the intron sep~rating exons C ant D (Fig. 4). A second genomic EcoRI fragment encompassing exons D and E may not have been observed since these exons contain only 218 nucleotides that are homologous to the open reading frame portion of the cDNA.
The recombinant lambta phage genomic clone (LAbT lO-l) was used for gene mapping by fluorescence in situ hybridization.
The basic procedure for localizing single copy genes with fluoresence detection has been described by Lawrence, J.B., e~
al, Ce; 52:51-61 (1988); Staunton, D.E,, et al, J. Ex~. Med.
lO87-lO99 (1989), which are incorporated by reference, and a detailed analysis of banding procedures described by Lawrence, .B., e~ al, ~l~n~ ~1990) which is lncorporated by raP~renca.

.' " ~

W 0 91/13173 2 0 ~ ~ 7 2 ~ PCTtUS91/0143 ~

The denaeured probe, previously labelled by nick-translation with biotin d-UTP, was present at a final concentration of 5 ~g/ml in the hybridization solution. An excess oE
non-specific competitors including total human placental DNA was also lncluded to compete hybridization to repeeitlve sequence olemenca, Speciflc hybridization was detcctod by 9taining with ~luorosc~in avidln.
Metaphase chromosom~s w~re id~ntified by diaminophenylindole (D~PI) banding, enhanced by prior incorporation of S-bromodeoxyuridine into chromosomal DNA. In addition, ~iemsa-trypsin banding prior to hybridization was done to confirm theso results. Tho locali7ation o~ tho ~cne and interprecation Oe banding pnttorns in 10-12 motaphn~as wn~
performed by two lndep~ndont obsorvcr~. The locall6atlon WAa al~o ~nalyz~i by ~lmpl~ moa~ur~m~ne of si~nal po3icion r~lA~Lvo to tho tot~l lon~th Oe th~ chroM~olno ln 20 ml~t~ pha0H ~iKu~o~.
Th~ d~po~it~d ~onomlc clon~ w~ls nick ~ n~l~tqd ln tho presonce o biotLnylnt~d d~TP ~nd then used to probe human metaphas~ chromosomo~. Loaalization to a sing}o chromosomal locus was datermined by non~isotopic, ilouresconc~ d~oction of ~ u hybridization. Over 90~ of metaphase figures showed hybridizat$on which was consistently on a chromosomal pair having the small metacentric morphology of F group chromosomes.
Chromosome banding analysis was performed primarily by DAPI-staining of BrdUrd-incorporated chromosomes and was confirmed with Giemsa-trypsin banding. As illustrated in Figure 5A and B, hybridization could be unambiguously visualized in one cell by the identical labelling of both sister chromatids on each of the two homologous chromosomes. Banding analysis demonstrated this to be chromosome 20 and allowed regional localization oE ~he signal to ql3,1-ql3.2. Figure SC
illustrates the position of the chromosome signal on ~e~eral enlar~ed chro~oaomes; 5D su~marlzos the data fro~ ten DAPI

, .

~ W O ~ 3173 2 0 ~ 4 ~ 2 ~ PCT/US91/0143~

banded chromosomes, scored independenely by two observers with essentially identical results. The PTPlB ~ene localizes near the interface of bands ql3.1 and ql3.2. Placement o the signal was also analyzed by measure~ent along the length of the chromosome; b~sed on the average of 27 determinations, thc PTPlB
gane would m~p In tha mlddle o ql3.1. Howover, becnus~ bnnd~n8 analysis i~ mor~ rel~ablo, wa conclud~ that PTPlB is localized to 20ql3.1~13.2.
Interestingly, chromosome 20q deletions spannin~ the qll-ql3.3 region have been observed in some patients wlth myeloid dis~rders tLe Beau, M.M., et al. Proc. NaCI. A~ld. Sci , 66q2-6696 tl985); Davls, M.P., at nl, ~Inçl_QQoQ~_ cYto~eno~ . 63~71 ~1984); Ml~olman, F-, 2~ DoL ~ L~LL~LL~
516 ~1984); Tosta, J.R., ot al, ~L~Q~ ~2- 868-877 ~1978)~. Whll~ ~ono~ o~hor t~lan PTPl~ could bo nltarnd or o~h4rwLs~ ~ut~t~ on ~0~ ql3.3 ln tha~o p~tlnt~, wo b~llovq that los~ o~ PTP~a ~c~lvLty could contrLbutc to th~ progression of cer~ain myeloid disorders, For examplo, the myeloid growth factor CSFl stimulatcs mltogenesis through a racep~or tyrosine kinase ~Sherr, C.J., et al, ~ , 665-676 ~1985)]. Loss of PTPase lB activity in CSFl-stLmulated myeloid cells could result in increased levels of phosphotyrosine. Phenotypically, this would appear similar to neoplastic transformation induced by an oncogenic tyrosine kinase, but would instead result from a loss of a growth suppressing function.

Production of PTPase Antibodies Expression Of PTPase Protein PTPase lB was cloned into the bacterial expression vector pET 3A (Studier and Moffatt, 1986) expressed from the T7 010 pro~oter and used to transform pLYS S cells an E. coli strain ~BL21DE3) containing a T7 po~ym~ras~ ganQ oxpr~s~ed Prom a lac inducible pro~otcr and constitutivQ ~xpression oP ly~ozyma.
' W O 91/131~3 r PCr/US91/0143~ ' 2~4723 ,~

Cells were grown overnighe ln LB ~edia containing 200 ~g/ml ampicillin and 25 ~g/ml chloramphenicol. The cells were diluted lOO fold in LB, 200 ~g amp and grown to an OD600 of l. At ~his point IPTG was added to a final concentration of 0.4~M and the cells were incubated for an ndditlonal 3 hours.
Bacterial exprasscd PTPasR lB resulted in the protoln being seque~ter~d ~ insoluble inclu3ion bodles, The inclusion bodies were then isolated from the bactoria by using the following protocol.
Bacterial cells were collected by centrlfugation and sub~ected to three cycles of freeze-thaw. The thaw~d cells wore resuspended in 1 ml of ST ~lO~M Trl~ p~l8, lOOm~ NaCl~ contnining 0.05~ NP40, 2.5mM MgC12, lmg/rnl DNAso pcr lO ~l Oe cultura, Thc resultlng su3p~n3ion wn~ lncubat~d on lcn Eor 15 mSnu~o~ nnd th~n sonlc~ecd Eor 1 mlnuto. ~ha ~u3ponslon wn~ th~n c~ntrl~ugod at lO,OOOxg ~or lO ~Lnueo~. Tho pollot w~
rosuspondod in l ml oP Herman~ bu~~r ~lOmM Tris pH 7.2, lOOmM
NaCl, lmM EDTA, l~ NP40, O.S4 Nndooxycholate) vortexed and centri~ugod at lO,OOOxg for lO minut~s. The pellot wa3 resusponded in lml of Hermans containlng 2mM
beta-mercaptoethanol~BME) and centri~uged at lO,OOOxg or lO
minutes. The resulting pellet was resuspended in lml of 25mM
Tris pH8, 2~ NaCl, 2mM BME and centrifuged in lO,OOOxg for lO
minutes. The resulting pellet was resuspended in lOO~l of TE.
Figure 9A shows the protein content of the inclusion bodies and is a coomasie blue stained gel of the PTPase lB inclusion body ~aterial. The first lane shows the material before solubilization by the addition of of 6M guanidinium-HCL, 5mM
Dithiothreitol ~DIT) (starting material) and the remaining lan~s, after the guanidine was dialyzed out of the sample ~post-dialysis). The slowest migrating band represent i~tact ,r~ . r- .

~ W 0 91/13173 2 0 ~ 4 7 2 ~ PCT/US91/~1432 PTPasP lB protein. The ocher lower bands are a mixture of contaminaeing bacterial proteins and proteolytic degrad~tLon products of PTPase lB. Afeer dialysis to ~emove the guanidine approximately 50~ of the inc~ct PTPase lB remained soluble.

Producti~ of ~n~bod~es The PTPa~o lB ~nclu3~0n bodl~s described above, wero Chen u~ed a3 an lmmunogen. Four ~ic~ wor~ immunized wieh 50~g of incl~sion body prot0in and then boost~d at day 14 and day 36 with 50~g of the sa~e inclusion body material, and fused on day 40. These mice generated a response that recognized the inclusion body materials in an EL~SA. 96 well plates w~r~
coaeed with O.l~g/ml of eieher PTPase lB incluslon body material or unLnduc~d lysntas in 0.05M Cnrbona~e buf~er, p~l 9.6 overnLght at 4C. The protcin ~olution was asplr~d o~E And 50~ o~ hybrido~a 3upornatant, im~uni~od ~ou~o ~orn d~lut~d l to lOOO in blotto or norln~l ~ou~o ~ora dllutod l ~o lOOO in blotto was addcd ~nd incub3tod at 37C Eor 90 mlnueo3. The platos werc wnshad 3 ti~es wich PBS cont~lning Tween 20 and then incubated with 50 ~l w~th HRP CAM ~Pierce Chemical Comp~ny) diluted l to 5000 in blotto for l hour at 37C. The pl~tos were washed 3 times with PBS containing Tween 20 and de~eloped with TMB (Transgenic Sciences Inc.) for five minutes and stopped with 2.5N sulphuric acid. The OD was read at 450nm. Spleens from two of these mice were fused by standard techniques. Two hundred-five hybridomas were generated and screened by ELISA
against the inclusion bodies. Twenty-seven of the hybridomas ;
screened by ELISA that were l.2 - 6 times over background were rescreened by Western blot analysis against PTP lB protein, which was expressed from vAbT 477-2-2 ~described below). Seven of the hybridomas were positive by the Western bloe analysis and single cells wero cloned. The seven were labeled AE4, AC5, AD2, AB3, AF5, 5E5 and BA6. These primary clones were screen~d by ., ~

Western analysis with vAbT477-2-2 expressed PTP 1B protein the following results:
There was then a second of single cell cloning of two AE4 and two CE5 positives. These secondary clones were again screened by Western analysis against the vAbT 477-2-2 expressed PTP 1B protein the following results:
These hybridomas resulted in two monoclonal antibodies (MAbs) AE4-2J (ATCC Accession No. Hb10666) and CE5-1K (ATCC Accession No.
HB10667). The 2 hybridomas grew as ascites.
A second fusion was performed using an additional mouse boosted as described above with the following exceptions, one additional boost on day 70, the final boost was on day 174 with fusion on day 177. 197 hybridomas screened by western blot reacted with the 49K PTP1B
protein. These hybridomas were single cell cloned and rescreened by western blot analysis. Twenty-two of the primary clones were positive ~ W O 91/13173 2 0 ~ ~ 7 2 ~ PCr/US91/01432 and were subjeceed to a second round of single cell cloning and screening by western blot. This resulted in l9 new hybrido~a cell lines. Thireeen of the monocl~nal antibodies produced by these hybridomas, as well as the original two hybridomas, were fureher characterized to determine the isotype of the antibodies, whether they react with thc 37kd core PTPase domain and whether th~y cross react wlth the rolat~d T c~ll PTPas~.
The lso~yp~s oE the antibodi~s waro dotermined using the Amersham Mouse Monoclonal Antibody Isotyping Kit ~cat. #RPN.29). The reactivity .
was de~ermined by western blot analysis using cell lysates from BSC-40 cells infected with vAbT 477-2-2 (PTPase lB) or vAbT 479 3 l tT cell PTPase, d~scribed below) and a 37kD PTPase l3 fragm~nt expressed in as a 64kd fusion protein with glutathi~no 3 tran~farnso. The bacterially ~xpra3sot 37kd PTPa~elB Eusion protoin wa3 ~lppll~c1 by Dr.
Ray Erick~on, Hnrvard UnLvorsLt~ and pla~Ld pCEX~PTP~21 w~s con~truat~ to oxpro3s tho n~Lno ~or~Sn~l 3~1 A~Lno ~cLd~1 ~WQCI eo th~
glut~thiono S-t~n~for~o gano ln th~ v0ctor paEX~2T ~Ph~rr~Ac~a, product nu~bor 27-480l). Tho ro~ul~ ~rom thLs nnnlysis are shown below:
Antlbody,~ ot~p0 3~K_P~se~ coll PTPas~
FC6-lg IgG2a +
FD9~2A IgG2a + +
FD2-2b IgG2a + +
FC9-2c IgG2a + +
DFlO-le IgG2a + +
BGl2-le IgG2a + +
BC2-2d IgG2a + +
CA6-ld IgG2b +
DAll-lf IgGl CD9-2a IgG2b +
DH8-lb IgG2b EAll-la IgG2a EF3-lg IgG2a - -CE5~lK IgGl - -EG3-lh IgG2a AE4~2J I~Cl , .

W O 9l/l3l73 2 0 ~ 4~ 2 '3 PCT/US91/~143 This analysis permits classification of these antibodies into five classes that react with unique epitopes. The class represented by the antibody FG6-lg (ATCC Accession No. HB 10690) reacts with an epitope in the 37kd core domain of PTPaselB thac is also present in T cell PTPase. The class represented by the antibody CA6-ld (ATCC Accession No. HB 10691) reacts wLth a dif~cr~nt epitope in thc 37kd core domain of PTPnselb nnd does not cross react wlth T cell PTPnse. The remainlng three classes represented by the an~ibodies DH8~1b (ATCC Accession No. HB
10689), AE4-2J and CE5-lK recognize unique epitopes in the carboxy ter~inal domain of PTPase lB and do not cross react with T cell PTPase. These three antibodies are distinguished from each other by their cross reactivi~y to oeher unidontiiod proteLns from cithcr vaccinin or ehe ~SC-40 c0119 u~d ~or th~
infcctions.
3T3, 3 18 ~-nd 3-18 neU C~ W~r~ 1Y~9Qd in hypotonlc Iy9i~
buEer contalnln~, protonD~ lnhlbLcors. ~pproxlm~coly ~0 ~
of ~ach of th~ cell ~yDnto wn~ run on SDS~PAG~ els an~ bloeced onto nitrocellulose. The blots wéxo probed with undiluted supernntant fro~ thc hybridoma cell lines. See Fi~ure 10.
Th~s~ antLbodies detect a 49kD protein in 3-18 cells, but not in 3T3 cells. MAb AE4-2J also detected a protein o~ approxi~ately 90kD. This protein was detected in both mouse cells snd hu~an cells.
lOcm plates of 3T3 and 318 cell lines were labeled overnight with 500~Ci of the 35S-methionine, then lysed in 500 ~1, per pla~e, RIPA buffer containing protease inhibitors. 250 ~1 of each lysate were immunoprecipitated with 500~1 of hybridoma supernatant from each of the 5 MAbs. A rabbit anti-mouse IgG heavy and light chain was used as the capture antibody for the antibodies AE4~2J and CES-lK.
Figure 11 indicaees that AE4 2J, FG6-lg and DH8~1b recognize a 49kD protein in 3-}8 cells but not 3T3 cells. CE5-lk appears to weakly ilmnunopr~cipitat~ this ~l9kD prot~in and C~6 1d doos not detect this protein at all. l'he Mab AE4-2J did not appear ~ W O 9Itl3173 2 0 ~ ~ 7 2 ~ PCT/US91/01432 ~o immunoprecipitate the 90kD protein seen in the weseern blot analysis.

~xpression of P~Pase P~oteins Using A~Vacclnia Vector Conseruction of plasmid pAbT 9110 for insertion of the PTPAse lB cDNA into the vaccinia virus HlndIII M fragmont.
The entire PrPa~o lB codlng region w~s isolated as a 1.4 Kb Ba~HI/SphI restriction fragment. The Sph I slte was made blunt ~sing T4 polymerase. This frag~enc was subcloned into Bam HI/Sma I dLge~ted pAbT4587 (Figure 16). pAbT4587 is 4069 base pairs and derived from pAG3 and vaccinia ~lind M frag plus the 40K promoter. The resulting plasmid, pAbT 9110 tFigure 17~) oxpres3es PTPa3e lB ~rorn thc v~ccinia 40 K prornotar.
Con3truction of pln3mld p~bT 9112 or in~ortlon o~ th~ T
c~ll PTPaso cDNA Lnto eha v~cclnia vLr~ ILntltIX M ~ra~m~nt.
~ h~ ontira r c~ll PTP~c co~in~ r~lon w~ oln~l n~ ~ ~,3 Kb Eco R1 r~trictlon i'r~m~nt ~Cool, ue nl,, ~
~5lL.~ Q~:S257 ~1989)]. Thig fragment was 5ubcloned into Eco Rl dige3eed pAbT4587 (Figure 16). The resultin~ plasrnid, pAbT
9112 ~Figurc 17B) expreases T celL PTPase rom tho vacclnla 40 promoter.
Vaccinia virus has been developed as an infectious eukaryotic cloning vector. PTPase lB and T cell PTPase have been expressed in vaccinia by the method of Mackett, M., et al., J, Gen Virol, 67:2067 (1986) except the selection scheme for the recombinant virus was a host range selection, Smith, K., PCT
89102486.

Insertion Of The PTP lB_And T Cell PTP Genes Into Vaccinia Strain vAb~33 To in~ert ehe PTPase lB gene inco che vaccinia virus genome at the HindlII M region of vaccinia virus ~train vAb~33 ~ATCC
VR2240), a selaceion ~ch~mo ba~od upon eho 29K hose~ran~ na, ::,: . .. , ., : : , .... . . .

~........ : . ~.... . : . . ..

W O 91/13173 20~4 ~'~ J PCT/VS91/01432 which i5 located in this region, was used. Gillard, et al., Proc. Natl. Acad Sci. USA 83:5573 (1986). Recombi~ant vaccinia virus vAbT33 contains the lacZ gene in place of a portion of the 29K g4ne. Thls lacZ insertion destr~ys the funceion of the 29K
gene; therefore, vAbT33 grows poorly on RK-13 cells, which roquire the 29K gonc product. Furtharmorc, vAbT33 forms bluc plaque3 on par~l~sivu cells in th~ pr~senc~ oE the chromogenic substra~e for ~-galactosidase, Bluogal, due to the presence of the lacZ gene. See, U.S. application Serial No. 205,189, filed June lO, 1988.
IVR vector pAbT9llO was transfected into BSC-40 cells which had been infected with vaccinla virus vAbT33. Vlral infaction and plas~Ld transfection worQ purformed eaiqontlally ns dcscrlbnd ln Spyropoulos, ct ~ . Vlr~ 7:lO46 ~l988). Racambinane virs~s w~ro soloctod a~ whlt~ plaquos in the pra30nc~ oE
Bluogal on RK 13 coll,~J. Pla~u~i~ w~r~ plck~d and purlel~d, ~nd thc final roco~bLnant, d~slgni~t~d vAbT477, was amplifiad on RK-13 cells and purifiod ov~r a 360 sucrose cushion.
The BSC-40 cells inected with either PTPlB (vAbT 477-2-2) or NYCB~, a wild-type vaccinla virus were lysed in hypotonic lysis buffer containing protease inhibitors. Approximately 20~g of each lysate was run on SDS-PAGE gels and blotted onto nitrocellulose. The blots were probed with undiluted hybridoma supernatant from either of the two Mabs (See Figure 10) .
Lysatçs of BSC-40 cells infected with either PTPlB (477-2-2) or NYCBH were run on polyacrylamide gels and transferred to nitrocellulose. The blots were cut into strips and probed with sera from immunized mice, PTP lB, normal mouse sera (NMS) or hybridoma supernatsnt from the PTP lB MAbs AE4~2J or CE5-1K.
See Figure 12. The sera was dlluted l:lO00 in Blotto and the supcrnatant was used strai~ht, ~he lino ln all the a~rip5 at 200 kD ls an o~ientation ~ark u~od to allgn the strips, ~S~

:. . . ..

~ W O 91/13173 2 0 ~ ~ 7 2 ~ PCT/US9~/01432 Figure 12). The PTP lB hybridoma supernatant detected a ca. 50 kD protein that was specific to cells infected with PTPlB. In all cases, the sera did not react with lysate from cells infected with NYCBH virus and normal mouse sera did not react with the vaccinia expressed PTPase proteins (Sea Figure 12).
The T cell PTP~a cDNA was lnsortetl into the ~indIII M
r~gion of vacclnin virus u5ing tho ~mo mathodology described above for PTPase lB. The IVR vector pAbT9112 was used resulting in the recombinant vaccinia virus vAbT 479. Expression of the T
cell PTPase was confirmed by western blot analysis o~ vABT479 infected BSC-40 cells using antLsara raised by lmmunlzing mic~
with T cell PTPase protein purifiad from reco~nblnant bnc~llo vlrus in~octod insoct colls.

ÇY ~ ~Q~ Q
~ rt nn~q ~m~
Ovor~xpr~s~ion o~ n PTPn~ cnpablo of InhLblting cQlLular transfor~ation. A coll linc overoxpressing PTPase was prepared 4s indicated bolow. The PTPase ovorexpressing cell line was then infected with an onco~enic re~rovirus and tho lnfectecl cells were analysed for parameters of transformation.
NIH 3T3 cells were cotransfected with a plasmid that expresses PTPase lB (pAbT9703) from the MLV LTR promoter and a plasmid that expresses the gene encoding hygromycin resistance pTKhy~ro from the Herpes virus thymidine kinase promoter.
Transfected cells were then plated in the presence of lOOug/ml hy~romycin and 22 colonies were isolated. The selected colonies were then expanded and RNA was isolated from cells derived from each colony. RNA dot blot analysis was performed using a fraBment containing the PTPase lB codlng region ~rom pAbT9611 as the probe and one colony showed hiBh levels of PTPase lB mRNA.
This colony was designa~ed 3r3 3~18 and PTPase lB expresaiQn was .

''`'''''''.'''''',"''` '''"''.',"`",' ;`'`', '' ;`. ''' '' ';
, .. . : . . ~ . . ~, W O 91~13173 2 0 ~ ~ 7 2 ~ PCT/US91/0143 ~ ~

confirmed by Western analysis using an antib~dy fraction supplied by Dr. Edmund Fiseher o~ the University of ~ashington, MLV TM~ neo virus expresses an allele of the hu~an neu oncogene which Includes a point mutat~on ln the transmembrane domain that enhances transforming activity. The transformin~
aetivity o nQ~ i9 thought to ba dua to its tyrosine kinas~
activlty and oneoganlc activation by tha transm~mbrana point mutation results in an iner~ase in tyrosine kinase activity. In addition, MLV TM-neu neo expresses the gene for G418 resistance. Infectious, replication defective viral stocks are derived from Psi 2 cells transected with tha plasmid pAbT9003.
Infection o NIH 3T3 cclls with MLV TM~Q~ nao rasule~ in detactable morphologlcnl trnnsfor~nation of in~aetad eoll~.
Morphological trnnsformncion ls dat~rminud boch by vlsu~l Inspection Oe G418R colonlu~ anel abLlity Oe in~oet~d e~ to ~orm Eoel in monoloyor~, 100~ oE th~ G~18R colonL~ ~ro morphologically tr~n~ormed and In axp~rimanes whore ono plata o~ ineeted c0119 ware split into both C418 salection media snd n focus ~s9ay equal numbers o G418R eolonies and foci appear indicating thst 100~ o tho inaetad eells are ~orphologically transformed.
PTPase activity was assayed by measuring the release of orthophosphate from tyrosine-phosphorylated RCM lysozyme, Tonks, N.K., et al, J. of Biol. ~h~ 263:14, 6722-6730 ~1988).
Control cells or retrovirally infected cells expressing neu TM
were harvested in a hypotonic buffer without deter~ents and i lysed by dounee homogenization. After centrifugation, the supernatant fraction, referred to as the "soluble" fraction, was reserved and the insoluble sediment was resuspended in buffer eontaining 1~ Triton. The resuspended material was subJeeted to eentrifugation and the resulting supernatan~ waa re~erred to as "Triton soluble" and reserved. The sediment, re~err~d to aa "Triton insolubl~ a r~suspen~ed. All threo e~llulnr .

~ W O 91/13173 2 0 ~ ~ 7 2 5 PCT/US91/01432 fractions were then assayed for PTPase activity PTPase activity was expressed as nanomoles of Pi release per minute.
Table 4 provides the results of the enzymatic assays for uninfected and nQ~-TM infected 3T3 and 3-18 cell lines.
~L~
PT~aso lB AC~IVITY IN 3T3 CELLS
CELL LINE PTPASE ACTIVITY
SOLUBLE PELLET
TRITON SOL TRITON INSOL.
NIH3T3 194 ~81 261 3-18 333 g79 320 NI~13T3 NEU2BS ~26 23 3~18 N~V 370 l2~ 437 3~18 ~nd ~g~ T~lnEectod 3~18 cull~ expro~ P~ s~ Aa~lvley tho.~ i3 two to th~o~ tim~s avor l~ndogunou~ ~ctLvlty. 5'ho mn~ority of tho oxc~aD PTP~a ~c~l-vley is eound Ln tha Trlton solublc fraction of aoll ly3ate3 which is simil~r to that ob~erved with endogcnous activity.
Both PTPaso-expressing 3-}8 cells and tho pnrontal NI~13T3 cells were pLat~d at a density oi' l.5Xl05 calls per 60 mm dlsh. The followlng day the cells were either mock infected or lnfected with MLV TM-neu neo virus. Eighteen hours later the cells ~ere split into nor~al ~rowth ~edia and growth media containin~ 800ug/ml G418. After l0 days the plates were analysed for the presence of foci or G418 resistant colonies, respectively. Both infected NIH 3T3 and 3T3 3-18 cells give rise to ~418 colon~es, but only tha NIH 3T3 colonies were ~orphologically altered (Figure 6B, E). In the focus assay both cell lines gave rlse to foci but the 3T3 3-18 foci were smaller and the cells in the foci appearad to be less morphologically transformed (Figure 6C, F).
Neu protein levals were detarminad by captu~a ELISA u~ing two monoclonal antibodias ~desi~natad NB3 and TAl) dirocted W 0 9l/13173 2 0 ~ 4~ 2 5 PCT/US91/0143 ~

against the neu gene product, for two individual G418~
colonies from both NIH 3T3 and 3T3 3-18. Although the 3T3 3-18 colonies are not morphologically ~ransformed, they con~ain actlve neu protein. This indicates that the res~stance to transformation is due to PTPase qxpression and not dacraased axprassion, The NIH 3T3 cqll linas, 3T3 nau TM cell lines (3T3 c811 lines transformed by the MLV TM-neu expressing virus), 3-18 hygro uninfected cell lines, 3-18 hygro/neu TM cell lines, (3-18 hygromycin sensitive cell lines transformed by the neu expressing TM virus) and 3-18 TM ne4 cell linas wera suspendcd In 0.354 ag~r and grown for 28 days, ~acroscopicnlly observablo colonies werc ¢ountqd on ~n lll~inatod colony countor. The '~
3-18 hygro ~ninfactod nnd 3~18 hy~ro/~Q~ TM cQll lLno~ woro pas3n~ad in modln conta~ning hy~romycin prIo~ to pla~lng Ln ~o~e ~g~r.
The 3-18 ~Q~ ~M c011~ were pn~sn~ed in ~ho absance oE
hygromycin. The results o this soc agar assay ara set forth below.
Tabla 1 Golony formation in Soft Agar Cell Line Colonies 3T3 uninfected 2 3T3neu TM 32 3-18hygro uninfected 0 3-18hygro/neuTM1 & 4 3-18neuTM 5 & 9 These results show that the PTPase lB expressing 3-18 cells inected with neu TM virus produce very ew, if any, colonies in soft agar whereas the neu expressing control cell lines all form signiicantly more colonies. Multiple antries in the tabla indicatq colonies ~ro~ duplica~e pla~as.
Nude ~ico were ~ubcu~aneou~ly in~oct~d wlth aither con~rol ... . .

~ W O 91/l3173 2 0 ~ 4 7 2 ~ PCT/US91/01432 unlnfected cells or cPlls previously infected with the retrovirus expressing neu TM. Cells infected wi~h MLV TM neuneo were expanded under selection, G418 (800 ~g/ml) for 3T3 neu, 21-1 n~, 3~12 neu, 3-24 neu, 3-18 ~_ or C418 t800 ~g~ml) snd hygromycln (lOO~g/ml) for approximatsly 3 wcoks. Cells were harv0stad by tryp3inization ~nd 1 x 106 cell~ were in~oct~d. Atar 3 weaks, tu~or volume w~s determinod by measurement of the tumor. Mica in~ected with neu TM-infect~d cells from either NIH 3T3 or 3-18 cell lines developed large tumors, while those in~ected with neu TM-infected 3-24 cell lines developed only small tumors. See, Tablo 2 balow, Prlor to in~ection tho 3-18 cells had shown a higher levcl of PTPaaQ
lB exprasslon than 3 24 colls.

Av~ Tu~or Size ~m ~
~E~ LI~ BI~E~ L ~EE~ X_~ EXP~RIM~T 3 3T3neu ND 7778 ~/- 4172 (6) 7843 ~/- 4058 (5) 21-lneu 370~ ~/- 2162 (2) ND ND
3-12neu 4649 ~/- 3219 (4) ND ND
3-24neu 365 ~/- 150 (4) 197 ~/- 190 (5) ND
3-18neu 3246 +/- 1052 (4) 4387 +/- 2346 (10) 4861 ~/- 2751 (5) 3-18hygneu ND ND 4841 +/- 1673 (4) 1) No uninfected cell lines caused tumors when injected into nude ~ice.
2) Numbers in parenthesis indicate the number of injections.
ND - Not done.
The following experiment was performed using early passage cultures of 3-18 cell lines. The mice were in~ected, as described above, with 3-18 cells in~ected with the neu TM
retrovirus, and again, developed larga t~ors. The mice inJected with inf~ctcd 3~2~ aalla only exhiblted ~mall tumor~.
Seo, Tablo 2 abovo. Tumors wera oxcise~ ~rom tha mico, which W O 91/13t73 2 ~ ~ 4 7 2 ~ PCT/US91/0143 ~

had recP1ved the 3-18 neu TM cells, and examined by Western blot analysis, for PTPase lB express$on. (See, Table 3, experimene 2 below). Although initially 3-18 cells show hi~her levels of PTPase lB then 3-24 cells, the tumor derived 3-18 cells showed no observable expression of PTPase lB except for one small tumor, which showed a slight PTPase expression upon examination of twico ag much tumor s~mpl~.

TABI,E 3 TUMORICENICITY STUDY ASSAYS

nou ELISA

EKPE~IMENT 2 3T3nou inpuC coll~ 382 3 24nou inpu~ coll~ 90 3~18neu lnpu~ cqll~ 105 Largol 264 Large2 144 Smalll 215 S~all2 253 ~/,1 3T3neu input cells 640 Tumorl 1848 Tumor2 1170 3-18neu input cells 385 +
Tumor3 612 Tumor4 856 3-18hygroneu input cells 252 +
Tumor5 898 Tumor6 9g7 1) 200~g of protein was necessary to see a ~aint PTPase lB
band. This was twice the amount of control 3-18 cell line lysate needed to observe a strong band.

,- .. . -, . , . ..... . . . . : ~ . .............. ;

.,. ;: : ,, ~ ~ ... :~.

ln Table 3, one can s0e that when the pre-injection cells were analyzed by neu ELISA, the 3-18 neu cells showed lower levels of neu TM than 3T3 neu cells. ~owever, the 3-18 n~_ tumors all showed levels of neu TM expression closer to that of the 3T3 cell lines, indicating that the tumors arose from thc inJected cu113 and suggestlng th~t the tumors arose irom cells whore thoro ~i~9 a 10s9i oE PTPas~ lB ~xprossion and an lncrease in n~ TM expression.
Another experiment was performed using essentially the same procedures described above in which 3-18 neu TM cells were selected and expanded in the presence of hygromycin to select Eor retention of PTPaso lB expres~ion. Six MiCe WQre ill~QCted wi~h nQ~ TM infocted 3~18 colLs ~1 x 10~) grown in d~al solection mcdiia tboth C4 18 and hy~romycin) whilo ~Lx ~lcu rcc~ivod simllarly in~ctadi 3~18 calls grown withouti hygromycin. Six con~rol mlc~ racolv~d ln~octLons Oe 3T3 nQ~ TM
cell3. Mico lnJoct~d wlCh uninEocta~ control c~ dld nor dovelop tumors, whil~ all other ~ice ~xhLbit~d largo tumors.
(See Table 2, experimene 3). Tumors in both sets af 3-18 n~Y TM
mice woro slightly delayed in appeariance comparcd to 3T3 noui TM
in~ected mice, but developed to approximately the same large size.
Two tumors were excised from each set of mice and analyzed for both PTPase lB and neu TM expression levels along with the pre-in~ection cells.
MAb AE4 2J was used to probe cell lysates from neu TM
lnfected cells or tumor deri~ed cells, which had been electrophoresed and transfered to nitrocellulose. See Figure 13. Lanes 1 and 2 are uninfected control cell lysates (3T3 and 3-18 respectively~. Lanes 3-5 are lysates of infected cells bePore ~njections into mice. Lanes 6-11 are lysates of t~mor derived cells from each set of in~ctions. The ar~ows on the right-hand side of tho page indic~ta MW wni~ht, w~aroas the W O 91/13173 PCT~US91/01432~
~0~472~ ~

arrow on the left-hand side of the pa~e indicates PTPase lB.
The 3-18 n~_ TM cumor samples showed no detectable PTPase lB
expression and this result was indep~ndent of ehe application of selectLve pressure to 3-18 ~ TM cells. See Figur~ 13 lanes T3 to T6. Western analysls also showed high lcvels of nq-l TM
~xpru3sion In all tumor samplo~ n~ did n neu ELISA. ~See, Table 3, experlment 3).
These experiments suggest that large tumors are formed only after there is almost a complete loss of PTPase lB expression.

Xnducible Expre~sion o Y~IB.
NIH 3T3 cells woro transEect~ with the M'r prolnotar PTPlB
plasmLd ~AbT 9710 tFLguro 18). Thls plasmLd w~s cons~ruce~cl by i~olating ~n SacII/Cla I fragmont thnc con~ain~ ~ha anCiro PTPa30 lB codin~ ra~lan Erom th~ pln~ld pAbT9703. Tha SaaXI
sitc w~ blunt~d u~ln~ th~ Kl~now i'~lgmont o~ Q~ll DNA
poly~er~se I. Ait~r tha addltLon of Cla I linkers th~ PTPase lB
fragment was subcloned into the unique Cla I siee of the pNUTC
vec~or, Tho resulting plasmid, pAbT9710, contained a P'rPase lB
cDNA expres~ed from the heavy ~etal induciblo Mtl~l proMoter and a DHFR gene for selection in methotrexate. 10 ~8 of plasmid DNA was used eo transfect 3T3 cells using standard calcium phosphate transfections. Individual colonies were isolated by selection in lOOnM methotrexate (MTX) and tested for inducible PTPase lB expression. After overnight growth in various concentrations of either zinc or copper, the clones were examined by RNA dot blot analysis for levels of PSPlB
expression. Specific and sharp induction of PTPlB occurs ae 80~M zinc. Expression is still high as zinc levels are increased, but cell number begins to rapidly decrease due to cell death or loss of adherence. Although induction occurs with copper, concentrations o~ 320~400 ~M were r~quired and tho expresaion lovol9 woro lowor than wlth zinc, and induction be~an W O 91~l3173 2 0 5 ~ 7 2 ~ PCT/US91/01432 ;

at about the same time as loss of cell number.
The transfected cell lines were examined by Western analysis for zinc inducible PTPase lB expression. Clones were obtained which exhibit no basal (uninduced) level and no induced expresslon, or higher basal expresslon than desired, as well as one clone which had the desired low bas~l expression and induction of PTPase lB expression to at loasc the same level as constitutively expressed PTPase lB ln 3-18 cells. This clone was called 43-5. Cell lysates of these cells were prepared by sonication in a Trlton X-100 buffer and 150 ~g of protein loaded onto a 12~ polyacrylamide ~el. After electrophoresis, proteins wero tr2~sfcrrecl to nitrocoLluloso and probad with nn anti-T ccll PTP rabbit ~n~isera. Lysncas WerQ prep~rad fro~ ~Itl 3T3 and 3-18 cell~ ~ nogntive nn~l po3Ltivo con~rol~, rcspec~ivoly. S~e, Fieure 14, Tho nrrow in F1guro 1~ ln~lLc~
PTPa3e 1.~.

NIH 3T3 colls or PTPase lB expressing cell lines (3-18 and 3-24) wer~ infacted with neu TM a~ dcscribed ~bove, ly~ed in tho presence o~ the PTPase inhibitor, sodium orthovanadate, and immunoprecipitated with the h-neu specific antibody, PB3, McKenzie, S.J., et al., Oncogene 4:543-548 (1989). After incubation of the immunoprecipitate with ~32P-ATP, the ~inased products were examined by ele~trophoresis and autoradiography. Figure 15 shows that the neu TM-infected 3T3 cells (lanes 2 and 3) had high levels of autophosphorylation on the ~ TM molecule pl85neU, while neu-infected 3-18 or 3-24 cell lines, lanes 5-6 and 7-8, respecti~ely, exhibit 1/2 to 2/3 less pl85~- autophosphorylation. The arrow in the figure ~ndicates pl85.
~ _ ELISA results on the cell lysates, which are lndicated below the ~el lanes in Fi~ure lS show that 3T3 nQ~ TM a~lls alsQ

YVO 91/13173 2 0 5 ~ 12 3 PCT/US91/0143 ~' ' conealn cwo to three times more neu IM protein than the 3-18 neu TM or 3-24 neu TM cell lines. Accordingly, it appears that, although PTPase lB-expressing cell lines show less neu TM
expression, the specLfic activity of ~_ TM is the same.

It is evldenc that thosa skillad in tha art, given tha ban~fit o tha oragoing disclo9ura m~y m~ke nu~arous modificatlons thereof, and departuras Erom the specific embodiments described herein, without departing from the inventive concepts, and the presenc invention is to be limlted solely by the scope and spirlt of the appended claims.

Claims (43)

In the claims:

1. A vector containing:
(a) a nucleotide segment comprising a sufficient number of nucleotides corresponding to a functional human PTPase gene or a functional fragment thereof to express a protein having a growth suppressing ability, wherein the vector does not contain an entire human chromosome; and (b) a promoter upstream of the nucleotide segment.

2. The vector of claim 1, wherein the PTPase gene is PTP1B.

3. The vector of claim 2, wherein the prometer is n viral prometer and the vector also contains an enhancer and polyadenylation sequences.

4. A DNA segment containing a functional PTPase gene or a functional fragment thereof that will express a PTPase gene product (protein) having growth suppressing activity, but wherein the DNA segment does not contain an entire human chromosome.

5. A substantially pure recombinant protein having growth suppressing activity produced by the functional PTPase gene of claim 4.

6. A protein produced by the vector of claim 2.

7. The protein of claim 6, wherein the protein contains about 435 amino acids.

8. An antibody that react specifically with the protein of claim 7.

9. An antibody to the subsequential pure recombinant protein of claim 5.

10. A method of assay for the presence of cancer or precancerous condition in a human cell which comprises;
(a) taking a predetermined cell sample from a human;
(b) adding a probe for a human PTPase gene or its gene product to the predetermined cell sample; and (c) determining from the probe whether there has been an alteration or deletion of a functional portion of the PTPase gene in the predetermined cell sample, wherein an alternation or deletion of the PTPase gene or its gene product is indicative of the cancer.

11. The method of claim 10, wherein the assay is for the presence of cancer.

12. The method of claim 11, wherein the cancer is acute nonlymphoctyic leukemia.

13. The method of claim 10, wherein the PTPase gene is PTP1B, LAR, CD45, PTPase 5 or T-cell PTPase.

14. The method of claim 13, wherein the PTPase gene is PTP1B.

15. The method of claim 10, wherein the probe is a nucleotide probe.

16. The method of claim 10, wherein the probe is used to determine whether there has been a deletion.

17. The method of claim 15, wherein the probe is a probe for a deletion in a chromosome containing the PTPase gene.

18. The method of claim 17, wherein the PTPase suppressor gene is PTP1B and the probe is for a deletion in chromosome 20 in the region q13.1-13.2.

19. The method of claim 18, wherein the probe corresponds to at least about 25% of the coding sequence of the PTP1B gene.

20. The method of claim 18, wherein the probe corresponds to at least about 50% of the coding sequence of the PTP1B gene.

21. The method of claim 19, wherein the probe corresponds to at least 70% of the coding sequence of the PTP1B gene.

22. The method of claim 10, wherein the determination as to whether or not a deletion or alteration has occurred in the PTPase gene of the predetermined cell sample is made by comprising the results from the probe of the predetermined cell sample with a predetermined standard base line.

23. The method of claim 10, wherein the determination as to whether or not a deletion or alteration of a functional portion of the PTPase gene is mate by comparing the predetermined cell sample with a nonmalignant cell from the human, to see if there are any differences.

24. The method of claim 10, wherein the probe is an antibody probe.

25. The method of claim 12, wherein the PTPase gene is PTP1B.

26. The method of claim 22, wherein the probe is an antibody probe and the comparison is made by ELISA.

27. The method of claim 23, wherein the probe is an antibody probe.

28. A method of treating acute nonlymphocytic leukemia, myelodysplastic syndrome and myeloproliferative disorders which comprises:
adding a therapeutically effective amount of a PTPase gene product to diseased cells exhibiting nonlymphocytic leukemia, myelodysplastic syndrome or myeloproliferative disorders.

29. The method of claim 28, wherein the PTPase gene product is the PTP1B gene product.

30. The method of claim 28, wherein the therapeutically effective amount of the PTPase gene product is added to the diseased cells by gene transfer technology using a retroviral vector containing a functional PTPase suppressor gene.

31. A PTPase nucleotide probe which comprises a nucleotide segment corresponding to at least about 20 nucleotides of a PTPase cDNA sequence.

32. A PTPase nucleotide probe which comprises a nucleotide segment corresponding to at least about 25% of the coding sequence of the PTPase gene.

33. The PTPase nucleotide probe of claim 32, wherein the nucleotide segment corresponds to at least about 50% of the coding sequence of the PTPase gene.

34. The PTPase nucleotide probe of claim 33, wherein the nucleotide segment corresponds to at least 70% of the coding sequence of the PTPase gene.

35. The PTPase nucleotide probe of claim 32, wherein the PTPase gene is the PTP1B gene.

36. The PTPase nucleotide probe of claim 32, wherein the probe is a RNA probe.

37. The PTPase nucleotide probe of claim 32, wherein the probe is a DNA probe.

38. The PTPase nucleotide probe of claim 31, wherein the PTPase cDNA sequence is the PTP1B cDNA sequence.

39. The method of claim 10, wherein the assay is for the presence of a precancerous condition.

40. The method of claim 39, wherein the precancerous condition is myelodysplastic syndrome and myeloproliferative disorders

41. A purified antibody to the PTPase protein.

42. A monoclonal antibody which specifically binds to PTP 1B
protein.

43. A monoclonal antibody having the binding properties to PTP
1B protein of a monoclonal antibody selected from the group of PTP 1B antibodies consisting of AE4-2J, CE5-1K, DH8-1b, CA6-1d and FG6-1g.