CA2263744A1 - Process for identification of genes encoding proteins having cell proliferation-promoting activity - Google Patents

Abstract

The present invention is directed to selection systems for the identification of cell proliferation genes based on functional analysis. More specifically, the invention is directed to a process for the identification of a cell proliferation promoting activity, the isolation of genes involved in such cell proliferation promoting activity, and the use of the so identified genes for the diagnosis or treatment of a disease associated with excessive cell proliferation. The invention further is directed to the design and development of antibodies, peptides, nucleic acids, and other compounds which specifically interfere with the function of the identified gene and/or its gene product, and pharmaceutical compositions comprising such compounds, for the treatment of diseases associated with inappropriate or unregulated cell proliferation.

Description

WO ~.'~"~86 PCT/US97/14514 PROCESS FOR ~DENl-~FICATION OF GENES ENCOD~NG PROTElNS HAV~NG CE~L
PROLTFERATION-PROMOTING AC~VITY

I. FIELD OF THE INVENTION
The present invention relates to selection systems for the id~ntific~tion of novel cell proliferation genes. More specifically, the invention relates to a process for the identific~tion of cell proliferation promoting activity, the isolation of genes involved in such cell proliferation promoting activity, and the use of the so identified genes for the 10 rli~gno~ic or tre~tm~nt of a disease related to aberrant or unregulated cell proliferation.
The invention further relates to the design and development of antibodies, peptides, nucleic acids, and other compounds which specifically interfere with the function or regulation of the identified gene and/or its gene product, and pharrn~reutical 15 compositions comprising such compounds, for the targeted tre~tm~nt of diseases related to aberrant or unregulated cell proliferation.

II. BACKGROUND OF THE INVENTION
General Background. In the past decade it has become ~ent that many diseases result from genetic alterations in cign~ling pathways. These include di~e~ces related to unregulated cell proliferation such as cancers, atherosclerosis and psoriasis as well as infl~mm~tory conditions such as sepsis, rhP~m~toid arthritis and tissue rejection.
The finding that these proliferative dise~es are based on genetic defects refocused the medical community to seek new modalities for disease management which essenti~lly consist of ~esignine drugs which modulate cell sign~ling. In order to develop highly specific drugs, i.e., drugs which potently interfere with uncontrolled cell proliferation but have low toxicity or side effects, it is crucial to identify the genes enCotling polypeptides 3 ~ involved in the cellular signal tr~n~duction pathways whose aberrant function may result in the loss of growth control.
Although tremendous progress in underst~ntling relevant signal transduction ~ pathways has been made in recent years, it is quite clear that many of the genes involved 3 5 in the development of proliferative disorders, referred to herein generally as "cell proliferation genes", remain to be discovered.
Cell Proliferotion Genes. Genes whose aberrant ~ ession or function may contribute to cell proliferation disorders fall into two general categories: (I) dominant WO ~)~JU~D6 PCTtUS97/14~14 Llall~ftJl~ lg genes, including oncogenes, and (2) recessive cell proliferation genes, including turnor ~up~lessor genes and genes encoding products involved in progla.~ lcd cell death ("apoptosis").
Oncogenes generally encode proteins that are ~csoci~ted with the promotion of cell growth. Reç~lce cell division is a crucial part of normal tissue development and co..l;..~s to play an illl~ol~ll role in tissue r~g~ .alion~ oncogene activity, plopelly regulated, is essPnti~l for the survival of the organism. However, inal,l,lopl;ate e~ s;on or illlplo~elly controlled activation of ol~cogcl)es may drive u,.col.~rolled cell 10 proliferation and result in the development of severe ~lice~c.os~ such as cancer. Weinberg, 1994, CA Cancer J. Clin. 44:160-170.
Turnor su~r ~sor genes, on the other hand, normally act as "brakes" on cell proliferation, thus opposing the activity of oncogenes. Accordingly, inactivation of 15 tumor suppressor genes, e.g., through mutations or the removal of their growth inhibitory effects may result in the loss of growth control, and cell proliferative ~ice~ces such as cancer may develop. Weinberg, 1994, CA Cancer J. Clin. 44:160-170.
Related to tu~nor slll)pressor genes are genes whose product is involved in the 20 control of apoptosis; rather than regulating proliferation of cells, they influence the survival of cells in the body. In normal cells, surveillance systems are believed to ensure that the growth regulatory m~oc.h~nicm.c are intact; if abnormalities are detected, the surveillance system switches on a suicide prograrn that cl~lmin~tec in apoptosis.
Several genes that are involved in the process of apoptosis have been described.See, for example, Collins and Lopez Rivas, 1993, TIBS 18:307-308; Martin et al., 1994.
TIBS L:26-30. Gene products which have been implicated in the control of or participation in apoptosis include bc1-2 (Korsymeyer, 1992, Immunol. Today 13:285 288), c-myc (Shi et al., 1992, Science 257:212-214; Evan et al., 1992, Cell 69:119-128), p53 3 ~ (Rotter et al., 1993, Trends Cell. Biol. 3:46-49), TRPM-2/SGP (Kryprianou et al., 1991, Cancer Res. 51:162-166), and Fas/~PO-I (Itoh et al., 1991, Cell 66:233-243). Cells that are rei,iSl~i~l to apoptosis have an advantage over normal cells, and tend to outgrow their norrnal cou.lle~ and dominate the tissue. As a consequence, inactivation of genes 3 5 involved in apoptosis may result in the progression of tumors, and, in fact, is an illlpo~ t step in tumorigenesis.

wO 9~ ,7~6 PCT/US97/14514 ~ t~tionC in tumor :,u~ es~ol genes and genes enco iing pl~du,lS involved in the control of apoptosis are typically recessive; i.e., both copies of the gene, the m~tPrn~lly inherited copy and the paternally inherited one, must be inactivated by mutation to 5 remove the effect of the gene product. Usually, a single functional copy of such genes is sufficient to m~int~in tumor ~u~ s~ion. Predisposition to certain h~edi~r cancers involves mutant tumor su~ es~or genes. For example, if an individual inherits a single defective tumor supl).essor gene from her father, initially her health will be ullco~ )lolllised, since each cell still contains a functional copy of the gene inherited 10 from her mother. However, as cells divide, mllt~tio~c ~ccllm~ te. Thus, at one point.
the rem~inine normal copy in a cell may be inactivated by mutation to remove thefunction of the tumor ~u~ ssor, thereby completing one of the steps toward tumorformation. Such a cell may give rise to clescen~nt cells which fe~lescllt the early stages 15 of cancer.
Of course. individuals who inherit a full normal complement of tumor suppressor genes can develop cancer as well. However, because two inactivating mutations are required, the development of the disease is much less frequent in such "normal"
20 individuals, i.e., not predisposed to cancer.
Tumor su~lessor genes and oncogenes participate in growth control pathways in normal cells in such a way that the a~""opliate level of cell division is m~int~in.od Disruption of these pathways by mutation of the colllpolle.lL genes, oncogenes or tumor ~u~ essor genes. is the underlying cause of cancer. Growth control in complex org~ni~smc like humans is a very il~lpGl~lt and complicated process. Thus, multiple genetic pathways for growth control are involved. Some pathways operate in all cell types in the body. Other pathways are much more specific and function only in certain cells.
3 0 D~co,~ Of CeU Proliferation Genes. Oncogenes and tumor suppressor genes have traditionally been identified by different m~thorlc However, each of the approaches currently employed for the identification and isolation of cell proliferation genes hac limitations on the types of genes that can be retrieved.
A first approach involves the detection and identification of transforming retroviruses and chromosomal translocations in tumors, which has provided the means to CA 02263744 l999-02-l8 W O 98/07886 PCT~US97/14514 identify dozens of oncogenes. Bishop, 1983, Annu. Rev. Biochem. 52:350-354; Stehelin e~ al., 1976, Nature 260:170-173; Bishop, 1987, Science 235:305-311. However, this strategy is largely limited to the identification of dolllin&.l~ oncogenes and it rarely leads 5 to the identification of tumor a~ eSSOl genes since inap~)lol.i;a~e tumor SU~JI)1e;>~01 fim~.tionc are recessive. Moreover, viral spread is not f~r.ilit~tP~ by decreased cell growth, thus it serves little ~ )ose for viruses to tr~n~duce tumor su~e~or genes.
Similarly, viral insertion or chromosomal translocations are single events. Thus, d~ t changes are far more likely to be manifested than recessive ch~nges A second traditional method for identifying cell proliferation genes has been the genetic analysis of kindreds, followed by positional cloning. Kindred analysis is, in principle, suited both for the identification of oncogenes as well as recessive cell proliferation genes, including tumor su~ essor genes and/or genes ensorline products 15 involved in the control of apoptosis. Through kindred analysis many recessive cell proliferation genes have been uncovered, including APC (Nishisho et al., 1991, Science 253:665-669), NFl (Xu et al., 1990, Cell 62:599-608), NF2 (Rouleau et al., 1993,Nature 363:515-521), RB (Friend et al., 1986, Na~ure 343:643-646), MLM (Cannon--20 266:66-71), BRCA2 (Wooster e~ al., 1994, Science 265:2088-2090; Wooster e~ al., 1995, Na~ur~ 378:789-792; Tavtigian e~ al., 1996, Na~ure Gene~ics 12:1-6), WTI (Francke et al., 1979. Cytogenet. Cell Genet. 24: 185- 192; Gessler e~ al., 1990, Nature 343 :774-778), and VHL (Latif et al., 1993, Science 260:1317-1320). However, a major disadvantage of the analysis of kindreds is that it is rather slow and limited, because the identification of cell proliferation genes depends on the existence of chance mutations that become established in the cell population, and cause an increased risk that is drarnatic enough to be visible above the level of nonhereditary (sporadic) cancer in the population. Kruglyak et al., 1995, Am. J. Hum. Genet. 57:439-454; Kruglyak et al., 1995, Am. J. Hum. Genet.
30 56:1212-1223-A third al)~)roach traditionally pursued to identify and isolate cell proliferationgenes is the analysis of homozyous or hemizygous genetic lesions in turnor cells. These lesions include regions of loss of heterozygosity (LOH) or homozygous deletions. Horuk 3 5 e~ al., 1993, J. Biol. Chem. 268:541 -546.

wo 98/07886 PCT/US97/14514 Finally, a method which has been employed for isolating growth control genes of the tumor :iul~p~:SSOI c}ass involves the selection of variants that have lost certain m~ n~nry traits, namely "revertants". Such revertant lines, however, are typically 5 difficult to identify and separate from the majority of rapidly growing parental cells.
- Still, a number of such revenants have been isolated from populations of cells transforrned by a variety of oncogenes and s~hsequent tre~tmP~t with various cytotoxic agents which are toxic to growing cells or cancer cells. Fischinger et al., 1972, Science 176:1033-1035; Greenberger et al., 1974, Viro~ogy 57:336-346; Ozanne et al., 1974, J.
0 Virol. 14:239 248; Vogel et al., 1974, J. Virol. 14:l404-l410; Cho et al., l976, Science 194:95l 9S3; Steinberg et al., l978, Cell 13:19 32; Maruyama et al., 1981, J. Virol.
37:1028-1043; Varmus et al., 1981, Cell 25:23-26; Varmus et al., 1981, Virology 108:28-46; Mathey-Prevot et al., l984, J. Virol. 50:325-334; Wilson et al., 1986, Cell 15 44:477-487; Stephenson et al., 1973, J. Virol. 1;:218 222; Sacks e~ al., 1979, Virolo~y 97:231-240; Inoue et al., 1983, Virology 125:242 245; Norton et al., 1984, J. Virol.
50:439-444; Ryan et al., l985, Mol. Cell. Biol. 5:3477-3582. Usually, cells are exposed to these agents under such conditions where cells that have reacquired a non-transforrned 2 o phenotype are contact inhibited, and hence, are less susceptible to these cytotoxic agents leading to preferential eliminati~Jn of the transforrned parental cells and. after several cycles, the isolation of morphologic revertants.
In addition to being both inefficient and time conC~ming, the above described selection for tumor s--~plessor genes is based on differential growth parameters of normal versus transformed cells and hence may preclude the isolation of certain classes of revertants. Moreover, the selection procedure itself may induce epigenetic changes or changes in the number of chromosomes. Furthermore, if the cytotoxic agents used are themse}ves mutagenic, then their continuous plesel-ce during the selection period may 3 ~ generate a revertant phenotype resulting from multiple mutational events. While any of these ,l.ech~ m~ may result in the production of a revertant phenotype, the nature of these genetic or epigenetic changes may preclude their analysis by gene transfer eA~ S.
3 5 Obviously, the most constraining factor for the utility of tumor cells in gene discovery is the lack of powerful selection procedures allowing the identification of new Wo 9~ B6 PCTIUS97/14514 genes. It is well recognized that there is a need for a rapid and effi~iPnt selection procedure that would permit the isolation of turnor cell ~ .,.~lt~ resl)lting from a single mutational event. With this objective, Zarbl et al. developed an alternative assay for the 5 selection of revertant tumor cells. Zarbl e~ al., 1991, Environmental Health Perspectives 93:83-89. This selection protocol is based on the prolonged retention of a fluolesc~ t molecule within the mitochondria of a nurnber of t~ rolllled cells relative to non-l,~lsÇullllcd cells. Indeed, in a cignifiG~nt number of cases, retention of fluulesc~
molecules within mitochondria seems coupled to l~lsrolmation. However, because the 10 prolonged dye retention phenotype is neither essenti~l nor sufficient for cell transformation, this approach is limited to some specific types of me~h~nicmc oftransfor nation.
Other methods which have been used to search for cell proliferation genes involve 15 biochemical approaches underlying analysis of cell cycle regulators (Serrano et al., 1993, Nature 366:704-707; Xiong et al., 1993? Nature 366:701-704), random sequencing of ssed sequence tags (ESTs) and homology comparisons (Lermon et al., 1996, Genomics 33:151-152), and methods for identifying differentially c;A~le~sed genes, such 20 as dirr~,relllial display (Liang e~ al., 1995, Methods Enzymol. 254:304-321). None of these approaches, however, offers a way to directly assess the function of the genes.
Tncte~ri candidates are identified based on a presumed (or idrntifi~hle) biochemical function or on an abnor-rnal pattern of eA~ression. These candidates are then tested further for involvement in cancer. Such tests include either mutational alteration in primary cancers or cell lines, experiments using somatic cells (for example? tû determine the effect of ectopic expression), or experiments in transgenic mice or knockout mice co.~ ing inactivated genes.
It is a~ en~ that these selection methods have a number of drawbacks and 3 ~ limit~tionc. Therefore it is desirable, and the objective of the present invention, to develop rapid and efficient selection procedures that would permit the identifir~tiQn and isolation of large numbers of novel genes, particularly cell proliferation genes, based on functional analysis. In accordance with its objective, the present invention provides 35 efficient selection systems which permit the isolation of growth-proficient revertants resulting from a single mutational event in growth arrested cells.

WO 98107886 PCT/US97/14~14 III. SUMMARY O~ rNVENTION
The subject invention is directed to selection systems for the identific~tion of cell proliferation genes based on functional analysis. Generally, the selection procedures of the subject invention involve the use of variants of ~al-~r(~ .cd cells to identify a cell proliferation promoting activity.
The selection systems of the invention may include creation of growth arrested tumor cell lines or cells which may undergo apcjptosis, for example by the c;~y~ ion of a gene encoding a growth SI~ SOI or apoptosis-indl-~ing gene product, under the 10 control of typically, an inducible promoter. When e~y~ ion of the ~uy~-eSSOl or apoptosis-inducing product is in-luce~l growth of the tumor cells is suypr~jsed and/or the cells die. Growth-proficient revertant cells are iclentified by virtue of their continllcd proliferation. Alternatively, if the efficiency of gene transfer is extremely high (as has 15 been reported for certain retroviruses) and selection for cells that have taken up DNA is employed, regulated promoters can be elimin~t~d In this case, the tu~nor suppressor or apoptosis-ind~lcing gene could be carried on the retrovirus along with a selectable marker such as hygromycin re~ict~nre. Rtvc.~ls that express the select~ble marker but do not 2 0 die or undergo cell cycle arrest are then isolated directly.
The invention is further directed to the identification and isolation of genes involved in cell proliferation promoting activity. This may, for example, be accomplished by selecting spontaneous revertant cell lines, analyzing their gene expression pattern, and identifying dir~,e..lially ex~,essed genes.
In other embodiments, revertants are ind~lced with specific molecules or moieties that disrupt a particular biochemical pathway, i. e., "perturbagens". In one embodiment~
the yellwl~agen is a DNA, encoding either a cell proliferation gene, or a protein or protein fragment acting akin to a do-l,;n~ -negative mutant of cell proliferation genes, 3 ~ e.g., by disruption of crucial protein/protein interactions. Revertants are selected, and the cell proliferation gene or protein/protein inl~,.a~;Lion underlying the promotion of cell growth can be determined by means of identification of the nature of the p~ wl agen. If the pelLullJagen is cle~errnin~od to be a cell proliferation gene, the co,lesyonding gene 35 product can be directly analyzed. If the y~llwbagen acts akin to a dol,lh~ negative mutant, e.g., bv disrupting a protein/protein h~tel~clion in a signal tr~n~dllction pathway, WO ~ o86 PCTrUS97/14514 the protein acted on by the dol,lh~ negative mutant is id~ntifie(l employing assays suitable for the identification of protein/protein interactions, e.g., the yeast two-hybrid system.
Analogous to DNA enco~ing protein L~g~ tC, peptides or peptide libraries acting as pcnu-bagens~ typically by protein/protein interaction, may be introduced in the growth ~u~ 2ssed cells in order to select lev~ s. In that case, the protein affected by the p~ g peptide is again identified employing ~says suitable for the identification of protein/protein h~ al;lions.
In still alternative embodim~ntc, revenants are inrl--ced by directing the random insenion of retroviral sequences in the genome as a means of either inaclivali"g cellular genes (e.g., tumor ~u~esso,s) or activating proto-oncogenes. The retroviral insenion is located, and the fl~nking sequences, presumably including genes encoding for cell proliferation associated gene products, are characterized. Perturbagens generated ~s a result of such a retroviral insenion may lepl~,sent aberrantly eA~,es~ed normal cellular proteins or truncated versions of normal proteins. Penurbagens may also derive from RNA that interferes with the stability or translation of specific cellular mRNAs. Most typically, such RNA-based perturbagens would act in an anti-sense manner by binding to complem~nt~ry mRNA sequences in the cell.
The invention is also directed to the use of the cell proliferation genes identified using the methods of the invention for the diagnosis or treatment of a disease. For example, analysis of tumor biopsies to identify the expression of a particular cell proliferation gene may serve as a valuable diagnostic indicator and may assist in guiding the therapeutic choice. Further, the identification of additional cell proliferation genes may help identify individuals who are predisposed for cenain types of cancer.
Predisposed individuals can be surveyed more frequently and thoroughly in order to ensure early diagnosis and 1l~ ,,P ll of the disease.
The invention is further directed to the lred~ t of ~iice~ces relat,~d to inappiol"iate or unregulated cell proliferation. For example, the invention provides methods to design, identify or develop therapeutic compounds, including antibodies, peptides, nucleic acids, etc. which will specifically i~ ,r~,e with the function of the identified cell proliferation gene and/or its gene product.

CA 02263744 l999-02-l8 W O 98/07886 PCTrUS97/14514 Finally, the invention is directed to pharrnaceutical compositions comprising such therapeutic compounds, and the use of such compositions for the treatment of diseases associated with aberrant or unregulated cell proliferation.

IV. BRIEF DESCRIPTION OF THE DRAWINGS
FIGURE l depicts a flow chart of the selection systems of the present invention for the identification of cell proliferation genes.
FIGURES 2A and 2B depict a flow chart exemplifying the use of perturbagens as 10 a tool for the induction of revertants in the selection systems of the present invention for the identification of cell proliferation genes and protein/protein interactions.FIGURES 3A and 3B depict the pOPRSVI.pl6 plasmid~ a means of controlling pl6 tumor suppressor protein expression in cell lines.
FIGURE 4 depicts expression of pl6 in the revertant cell lines derived from HS294T/pl6+ cells.
FIGURE 5 depicts expression of Rb in the revertant cell lines derived from HS294Tlpl6+ cells.
FIGURES 6A and 6B depict a peptide display and genomic fragment library vectors.
FIGURE 6A is a schematic representation of the peptide display library depictingthe insertion point for 45 base oligonucleotides of the composition (NNG/C/T),5 (N =
any base) which encode randomized lS amino acid peptides inserted in frame within GFP.
FIGURE 6B is a schematic representation of the genomic fragment expression library depicting the insertion point between the GFP coding region and the PGKl 3' UTR for small fr~gmen~ of yeast genomic DNA (see Methods and Materials).

V. DEFINITIONS
As used herein, the following term(s), whether used in the singular or plural, will 35 have the me~nin~ indicated:

SIJ~ ITE SHEET (RULE 26) CeU Proliferation Gene. As used herein, the term "cell proliferation gene" refers to a gene which, when aberrantly t;~lessed or regulated, may induce or otherwise be involved in the development of cell proliferative disorders. Such cell proliferative disorders include, but are not limited to cancers, arteriosclerosis and psoriasis viral disease, as well as infl~mm~tory conditions such as arthritis or sepsis. Cell proliferation genes include ~1O~ transforming genes, such as oncogenes and other genes encoding products involved in the induction of cell growth and recessive cell proliferation genes, such as genes encoding tumor suppressors, genes involved in the induction of apoptosis 10 or genes involved in viral growth.
Perturbagens. P~,~LUIIJ&g~lS are molecules or moieties of defined or ~c~ hle nature, e.g., proteins, subdomains of proteins or peptides of defined se~uence which, when introduced into cells or generated internally by forced ex~lession of an endogenous 15 gene or gene fr~gm.ont, complement or disrupt a particular biochemical pathway. For example, they may act in a manner analogous to certain previously described domin~nt mutations. Perturbagen libraries may be generated using techniques that are similar to those employed in construction of conventional gene and cDNA libraries.

VI. DETAILED DESCRIPTION OF THE INVENTION
A. General Overview Of The Invention Utility Of Novel Cell Proliferation Genes. Apart from underst~n~ing the genetic basis for one of the major causes of cell death, discovery of new cell proliferation genes has significant medical and commercial benefits. The potential value of such genes derives from opportunities to diagnose and treat cell proliferation disorders, such as cancer, more succ~s~fully and efficiently.
First, cell proliferation genes can be of medical value in the identification of3 ~ individuals predisposed to cancer. Traditional methods of cancer diagnosis have generally depended on post-~y~ JL~ latic ex~n~in~tion by loc~li7ing the tumor mass, and histological eY~min~tion of tumor biopsies to classify or stage the tumor. Currently, presymptomatic detection is realized more or less routinely for a small number of cancers 35 such as prostate carcinoma. Partin et al, 1995, J. Urol 155:1336 1339; Mettlin et al, 1996, Cancer 77:150-159; Schroder et al, 1995, Cancer 76:129-134; Egawa et al, Wo 98l07886 PCTIUS97/14514 1995, Cancer 76:463-472. I~e~,a-lse early detection and surgical resection play a vital role in survival rates, m-othnds that facilitate early diagnosis are ~ .ely illlpO~
One way to de~.ease the length of time ~ n the ap~ea,a-~ce of tumor tissue and its detection is to survey c~nriid~te patients more rl~luelllly and more thoroughly. However, such m~thoAc of surveillance are e~})e.~sive; thus it is nrce~.y to limit s~;,u~ y to high risk individuals. Conse.lu~.,lly, information about genetic predisposition to cancer is ,el,lely desirable. Because most genes that influence heleli~ y cancer are also involved in tumor plogl~;aaion~ isolation of genes by somatic cell genetics has the 10 potential to uncover such predisposing genes. Germline testing for such genes offers the chance to rate an individual's probability of COllLIa~;lillg cancer, and ~ p~nsive cancer s~ nillg efforts may be limited to those most likely to benefit from them.
Second, cell proliferation genes can be of medical value in the classification of 15 already existing tumors based on genotype. Lowe et al., 1994, Science 266:807-810. In the past, oncologists have relied on histological e~rnin~tion of biopsy specimens.
Though useful, histological analyses are generally h~ pe~d by their subjectivity and imprecision. Methods that classify tumors based on their genetic composition have the 20 potential to improve the reliability of their classification enormously. Detailed knowledge about turnor genotype may serve as a prognostic indicator for the tumor and may assist in guiding the therapeutic choice.
Finally, identification and isolation of cell proliferation genes affords hllyol~
therapeutic opportunities. Numerous approaches may be pursued to use informationabout cell proliferation genes into therapies including, but not limited to the following:
1) transfer of wildtype turnor sup~ ;.sor genes into tumor cells that have lost their activity; 2) inhibition of the activity of oncogenes in tumors, an approach that is being followed by several pharm~eutic~l co.~ niPs in the development of ras farnesylation 3 ~ inhibitors; and 3) selective induction of tumor sul")ressor genes in normal cells to induce a state of t~l,lpOla,r cell cycle arrest. These methodc have the potential to be much more selective and efficacious than conventional chemo- or radiotherapy.
It is desirable to identify as many cell proliferation genes as possible because each 3 5 one will be a candidate for medical utility.

W O 98/07886 PCTnUSg7/14514 .~ele~ SystemsForTheD~scoi_,J, or CeUProliferationGenes. Thepresent invention is directed to selecti( n systems for the i~entific~tiorl of cell proliferation genes based on functional analysis. More specifically, the invention is directed to a process for 5 selecting revertant cell lines which can be used to identify and isolate the isolation of genes involved in such cell proliferation ,orolllotiilg activity, and the use of the so-identified genes for the fli~nocic or Ir~ 1 of a disease ~ccoci~ted with a~ t orunregulated cell proliferation. The invention is further directed to the design and development of antibodies, peptides, nucleic acids, and other co~ ,oullds which 10 specifically interfere with the function of the idPntified gene and/or its gene product, and pharrn~eutical compositions comprising such colllyoullds~ for the targeted ~ l of fli~e~ces related to illa~ ol,l;ate or unregulated cell proliferation.
More particularly, the selection systems of the invention involve construction of 15 growth arrested tumor cell lines or cells which may undergo apoptosis. for example by the e~lession of a gene encoding a growth ~u~ or or apoptosis-inr~ ing product under the control of an inducible promoter followed by selection of revertant cells.
Alternatively, revertant cells can be selected that no longer require specific growth 20 factors. When e~lession of the SU~)leS501 gene is inrlllce~ or specific growth factor(s) are withheld, the growth of the tumor cells is arrested. From these arrested cells, growth-proficient revertant cells can be identified by virtue of their continuedproliferation. The selection systems of the invention are scl~ tically depicted in FIGURE 1.
The selection strategy provided by the present invention has several advantages.
};irst, contrary to previously suggested methods which involved the isolation and molecular characterization of non-~ransformed revertants from populations of tumor cells (Fischinger et al., 1972, Science 176:1033-1035; C~ elgel et al., 1974, Virology 30 57:336-346; Ozanne et al., 1974, J. Vfrol. 14:239 248; Vogel e~ al., 1974, ~ Virol.
14:1404-1410; Cho e~ al., 1976, Sc~ence 194:951 953; Steinberg e~ al., 1978. Cell 13:19 32; Maruyarna e~ al., 1981, J. Virol. 37:1028-1043; Varrnus e~ al., 1981, Cell 25:23-26;
Varmus et al., 1981, Virology 108:28-46; Mathey-Prevot et al., 1984, J. Virol. 50:325-35 334; Wilson et al., 1986, Cell 44:477-487; Stephenson et al., 1973, J. Virol. 11:218-222;
Sacks et al., 1979, Virology 97:231-240; Inoue e~ al., 1983 Virology 1~5:242-245:

WO ~Xl~oa6 PCT/US97/14514 Norton et al., 1984, J. Yirol. 50:439-444; Ryan et al., 1985, Mol. Cell. Biol. 5:3477-3582; Zarbl et al., 1991, Environmental Health P~ C~;IJCS 93:83-89), the assays closed herein involve positive selection; i.e., selection for growth, rather than the 5 ce~s~tion of growth. It is easier to identify and separate growing cells from growth-arrested cells than to isolate non-transformed rcve.t~ilts.
Second, cultured tumor cell lines generally grow vigorously in culture. Thus, the assays of the invention can be perforrned in a time-efficient manner, as growing colonies can be identifie~l isolated, and analyzed very quickly.
Third, red-m~l~nry in growth control palhwt.~s is not a problclll in the growth au~ cssed turnor cell lines provided and used for the selection s~atc.lls of the invention, as is the case in assays based on selection for non-l,u,~o,~,.cd cells. For exarnple, in the case where a cell line is enEinePred to contain a gene encoding a wildt,vpe turnor 15 suppressor, one single le~lldilll to growth remains. This growth rc~L~dint can be overcome by a variety of second~ry changes, for example alterations in genes dow,lslle~ull of the particular tumor Su~plcSaOr gene in the genetic pathway of growth control. Because of the fact that a single change can be sufficient to overcome the 20 growth fe~ lt of tumor su~ple;,sol-me~i~t~cl arrest, m~th<-rlc that induce mutation (or pc.lull~dtion) in a manner that allows recovery of the targeted gene in the cell permit isolation of additional cell proliferation genes. Accordingly, such cell proliferation genes are selected based on their inherent function as growth regulators in cells B. Selection Systems Based On Tumor Suppressor Genes In one embodiment of the invention, selection systems are generated based on the growth suppression of tumor cell lines by the cxyl~;saion of a tumor sup~ie;~a gene, and proliferating revertants are selec~e~
1. Tumor Su~ cr Genes Many twnor a~pressol genes cause growth arrest when ovelcA~l~,ssed in norrnal cells, as well as in certain turnor cell lines. Examples for tumor au~ essor genes include pS3 (Lin et al., 1992, Proc. Natl. Acad. Sci. U.S.A. 89:9210-4), Rb (Francke et al., 1976, CyJogene~. Cell (:~enet. 16:131-134; Cavanee et al., 1983, wO 98/07886 PCT/US97/1~14 Nature 305:779-784; Friend et al., 1987, Proc. Natl. Acad. Sci. U.S.,4. 84:9059; Lee et al., 1987, Nature 329:642-645; Huang et al., 1988, Science 242:1563-1566; Harbour et al., 1988, Science 241:353-357; Yokota et al., 1988, Oncogene 3:471-475) andpl6 5 (Karnb et al., 1994, Science 264:436-440; Nobori et al., 1994, Nature 368:753-756).
Generally, tumor ~u~,essor genes trigger growth arrest in cells at one of several positions in the cell cycle. Most frequently, however, tumor SU~ CS501S are found to cause growth arrest at the G,/S stage.
Though the details of growth controi pathways are known in only a few cases, it 10 is generally believed that ovel~A~,~,ssion of tumor s.lppressor genes in cell lines that contain inactivating mutations dow"~ c~" in the ~e~e~ e growth control l)alh~r~. will not have a growth inhibitory effect. In order to result in a growth arresting effect in the target cell, any particular turnor ~.u~ es~.or must be eA~"~s~.ed in an a~l.,op"ate cell line 15 which CO~ S intact do~"sl,~al" components of its respective growth control pathway.
For example, ove.cA~,ies~ion of pl6 in cells that are retinoblastoma-negative (Rb-) has little or no effect on growth, while ovelcA~ .ion of pl6 in a wide variety of Rb~ lines, for exarnple the Rb+ mel~nom~ cell line HS294T (Horuk et al., 1993, J. Biol. Chem.
20 268:541-546), causes G, arrest. Stone et al., 1996, Cancer Res., in press. The reason is that Rb participates in a growth control pathway along with pl 6~ acting do~h"sl,e~ull of pl6; conse.luently, ovel~lcssion of pl 6 in the absence of Rb protein has no growth arresting effect.
In one embodiment of the invention, a selection system has been designed based on the tumor ~.up~ s~or pl 6, which is described in more detail, infra.
In another embodiment of the invention, selection systems are designed based on the gene encoding rb. ovelek~le~7~7ion of rb is known to cause arrest in many cell lines.
As rb acts do~ll~ ,~ll of pl6, revertants of rb-arrested cells are expected to contain 3 ~ alterations in a set of genes that overlaps with the pl 6-arrested revertants. DirJ;lellces between these sets of genes identified by analysis of rb and pl6 revertants"e~c~ ely, may, however, give inlcle~ g insight into so far unknown cellular events. Co,l,l)~dlive e.i,l,~.,lS involving rb and pl6 arrest could, for example, elucidate alternative 35 pathways used by p16, if, for example, the growth control pathways branch u~slle~ull of rb so that pl6 acts in parallel through other dowl,sl,~a.n mediators besides rb.

Wo ~I'~ B6 PCT/US97/14514 In still another embo~im~ont, selection systems are generated based on the breast cancer ~usc~,ptibility tumor ~u~-ei,~o~ gene BRCAI. BRCAl has been shown to arrest growth of breast epithelial cell lines (Holt et al., 1996, Nat. Genet. 12:298 302).
however, little is known about BRCAl 's palhway of growth control. Thus, selection systems based on BRCAl su~ essed tumor cells are of compelling interest and potential utility. Analysis of revertants of BRCAI-arrested cells, e.g, in a BRCAI-ov~le~ lesaing breast cancer cell line, e.g, MCF7, can be used to identify do~llsL~n merli~tors of BRCAI tumor ~u~ ssor function.
In another embodiment of the invention, selection systems are ~ecign~d based on the pS3 pal~ y. Regulated e~ ion of p53 and its downstream targets, such as the CDK inhibitor p21 induces either apoptosis or Gl arrest in a variety of cell lines. Given the prominent role of pS3 in human cancer, i.e., roughly 50% of human cancers contain 15 pS3 slutations, information about other cGlllponc.l~s of the p53 pathway will be extremely valuable.
In still other embodiments of the invention, other tumor ~u~pl'essor genes are used in order to design selection systems for the identification of novel cell proliferation 20 genes. In principal, any gene whose ~ ~es~ion can be manipulated to cause cell growth arrest, can be used. Examples include, but are not limited to, ~YTI, YHL, B~CA2, NFI.
NF2, Pl5, P21, P18, Pl9, P27, P57.

2. Reversion Of Growth Arrested Phenotypes Once arrested by exp.ession of the turnor suppressor gene. revertant cells which continue to grow can be isolated.
In one embodiment of the invention, growth proficient random revertants are isolated. In other embo~im~ontc, reversion is induced using specific agents, i.e., 30 perturbagens, which are introduced into the growth ~u~le3sed target cell. See. infra ~ trnf~ R~._,tu~.b. Generally, growth-proficient random revertants may proliferate for one of several reasons. First, they may have gained c~ a~ion of an oncogene located downstream (or possibly u~sll~alll) of the tumor Supl)lc~SOI in the same 35 genetic pathway. If this is the case, tumor genes can be directly identified. Second, the rel~lt cells may have undergone an alteration of a ~ign~ling ~alhw~y that is parallel to W O 98t07886 PCTrUS97/14514 the pathway within which the ectopically cA~ ;.sed tumor ~.l~lessor gene acts.
A~;vely, the lev~ cells may have lost eA~lc~aloll of the tumor ~u~ple~or gene used to arrest them in the first place. Finally, the cells may have lost tAI,ies~ion of tumor :iU~ ,SSor genes that act dow"sL.cam of the ectopically tAIlles~ed tumor suy~,lessor gene.
In a specific embodiment of the invention, pl 6 is ectopically cA~lessed in an Rbt cell line, such as the melanoma HS294T line, under the control of an IPTG inducible promoter. As a consequence of the pl 6 eAyicssion~ the cells arrest in the G, phase of the 10 cell cycle, rçsl~ltin~ ultimately in death of the vast majority of these cells. The l~re.~nt cells are identified and isolated, for example, by placing the arrested cells in 96-well plate wells. After about three (3) weeks, revertant clones are transferred into new culture dishes, exr~n~leo and characterized.
In a specific working example described herein, in~a, six revertant cell lines derived from the HS294T/pl6 cell line, which eA~"~,;,sed pl6 when inducecl with IPTG, were isolated. Interestingly, the revc.~-~ cell lines typically exhibited growth properties that are similar to their parental line. For exarnple, det.~ ....;.. ~I;on of the proportion of 20 cells in G, under conditions of asynchronous growth by flow cytometry revealed that our of the six lines had a p,opol~ion of cells in G, similar to the parental HS294T line used to engineer the arrestable line, i.e., roughly 3-4 times as many G, cells as G2 cells.
One line ~peal~d to have a more equal plop(,-~ion of G, and G2 cells. The sixth line turned out to have some residual sensitivity to pl 6. since the percentage of G, cells varied depending on whether ~A~,lcs~ion of pl6 was indllced by addition of IPTG to the mediurn. Unlike the other five lines, this line had more cells in the G, phase when pl6 was eA~.~ssed than in its ~hs.on~e, suggesting that the line had not become fully insensitive to pl 6 ~AI ,~ssion, but only partially incPncitive.
3 ~ When the six lines were characterized for the presel~ce of various proteins thought to be involved in the pl6 growth control pathway, ill~clc~li,lg results were obtained.
Four of the six lines had lost eAples~ion of pl6. Presumably these lines escaped from in~ ced pl6 arrest by elimin~ting the tumor s.l~ple~sol gene, or alternatively, by 35 preventing its ex~,es~ion. The fifth cell line had lost eApicssion of Rb. This is concict~nt with the notion that Rb acts dO~ s~ of pl6 in a co..~..oll pathway for CA 02263744 l999-02-l8 W03~ &6 PCTAUS97/14514 growth control. Finally, the sixth cell line ~c~d to contain the expected levels of pl 6 and Rb genes. The levels of the potential oncogenes CDK4 and cyclin Dl, also thought to act in the pl 6 growth control pathway, appe~ed normal as well. Thus, the sixth 5 revertant cell line cont~in~d alterations in the t;A~Jleaaion or function of a gene of unknown identity. Based on its function, this gene is involved in the induction of the ~-conLlolled cell proliferation and thus possibly in the development of cancer. This cell line permits the identification of a novel cell proliferation gene.
Rever~ants. In another embodiment of the invention, the identification 10 of cell proliferation genes does not rely on the selection of random revertants. Growth-proficient revertants are in~ce~ using specific types of "mutagenic" agents, r~;re~,~d to as "~ ulbagens". Revertant cells are selected, and the gene or genes that allow escape from arrest are identified.
In one embodiment, the perturbagen is DNA encoding a cell proliferation gene, or, alternatively, dominantly active protein subdom~in~ or peptide sequences, used to disrupt the action of endogenous tumor allpplea5Cjla or oncogenes, e.g, by illt~lr~ g with crucial protein/protein interactions. Revertants are selecte~l and the cell2 o proliferation gene or protein/protein interaction underlying the promotion of cell growth is ~let~rmin,d by means of identification of the nature of the perturbagen.
If the perturbagen is determined to be a cell proliferation gene, it can be directly analyzed. For example, the perturbagen sequence is recovered using the Polymerase Chain Reaction (PCR) and sequenced using standard methods. If the perturbagen sequence is identical or similar to sequences in a public d~t~h~e such as GenBank or dbEST, then it can be directly identified Alternatively, if a portion of the sequence is known, or even in the ~bsence of any identification, the entire sequence of the perturbagen can be identified by isolating cDNA clones and standard recombinant DNA
3 ~ methodology.
The target of the perturbagen can be identified using a variety of methods. For example, if the p~.lull,agen is acting akin to a dominant-negative mutant, e.g., by dislLIl~tillg a protein/protein interaction in a signal tr~n~d~lction pathway, the protein 35 affected by the dolllh~lt-negative mutant is identified using assays suitable for the identification of protein/protein interactions, e.g., the yeast two-hybrid system.

WO 9~ PCT/US97/14514 P~ agc,~s can also act at the RNA level, in which case yeast two-hybrid analysiswould be incufficient to identify the perturbagen target. In most cases such p.,.lu~bagens are expected to act through an anti-sense ~ecl.A~.;cm and the targets would are id~ntifiçd 5 based on the co~ e,.l of the perturbagen sequence.
~ or introduction of the pcllulbagens, if genes enroAin~ for entire proteins or protein fr~grn~ntc are employed as pc.lull,agens, pe~ bagell libraries are consL~.n,led from mRNA of any cell line or tissue and inserted into cA~icssion vectors such as retroviruses which serve as highly efficient delivery systems. Such libraries, when 10 introduced into cells, may act as mutagens. If the cells are placed under stringent selection for a particular trait such as growth, l~c~ agen-in~uceA. variants can be isolated. Wildtype cells die, but cells that receive specific p~ bagen sequences that ~ e.r~le with growth regulation pathw~ys grow. Contrary to somatic mutations of 15 growth a~plcsaor genes, which are recessive, p.,~ bagens that impair the activity of a gene product by, e.g., interfering with protein/protein interactions, are domin~nt affecting the products of both alleles of a genetic locus. Alternatively, perturbagens are introduced into cells and monitored in a transient fashion. Transient gene cxl~ie,aion is 20 efficient and readily achieved. ElecL op-,~dlion and various other methods of gene delivery are suitable fc r transient c~ ssion monitoring of the introduced pe~ l)agens.
In cases where do~llinat~l negatively acting perturbagens are employed, libraries may be constructed from randomly primed mRNA and inserted into e~ ion vectors, such as retroviruses. Alternatively, the libraries are fused to degradation promoting domains.
In altemative embodiments, DNA libraries that encode random peptide are employed. Altematively, combinatorial chemical libraries, most typically peptidelibraries, may be employed as pclLu~bagens.
In still alternative embo~imrntc revertants are inrlured by directing the random3 ~ insertion of retroviral sequences in the genome as a means of either inactivating cellular genes (e.g., tumor aUpJJlCaaUla) or activating proto-oncogenes. The retroviral insertion is located, and the flAnking seq~rnres, presumably including genes encoding for cell proliferation associated gene products, are cha~acl~,izcd. Pc.lull,agens generated as a 35 result of such a retroviral insertion may lc~ scl.l ab~ lly cx~lcssed normal cellular proteins or truncated versions of normal proteins. Perturbagens may also derive from CA 02263744 1999-02-18.

W 098/07886 PCTnUS97/14~14 RNA that interferes with the stabi}ity or translation of specific cellular mRNAs. Most typically, such RNA-based p~,.lu,l,agens would act in an anti-sense manner by binding to colllyl~lllc~lt~y mRNA sequences in the cell.
Recovery and identification of the pc.lull)agcn se~uences and their targets is accomplished with standard procedures, including the polymerase chain reaction (PCR) and the yeast two hybrid system. See, SecJion VI.E., infia.
The use of p~.lu.l,agens for the induction of reve~ s in the selection systems of the present invention is depicted sch~rn~tic~lly in FIGURES 2A and 2B.
Once isolated, the pc.lulbagen can be reintroduced into the same cell it was isolated from, or into different cell types to further chalacleli~c the ~)rop~.lies of the molecule.

C. Other Selection Systems CDI~lnh~bit~.~. In one embodiment of the invention, selection systems are generated based on cA~es~ion of CDK inhibitors in suitable host cells.
All CDK inhibitors defined to date, including pl5, pl6, pl8, pl9, p21, p27, p57 20 cause cell cycle arrest when they are overcAI,lessed in certain cell lines. In some cases, such as pl 6, some details are alre ldy known with respect to downstream pathwaycomponents. In other cases, most details of the pathway of growth control within which the genes function are still to be elucidated. Apart from their preferred in vitro targets, i.e., CDK4 and CDK6 in the cases of pl5, pl 6~ and pl8, and CDK4, CDK6. and CDK2(and CDK4, CDK6) in the case of p21, p27, and p57, the identification of con,ponc~ of the pathways that act downstrearn by reversion selection systems will greatly facilitate the ability to manipulate these growth control pathways to achieve a therapeutic advantage.
Many cell lines respond to ectopic ~Ap,~ssion of CDK inhibitors by entering a state of arrest, and may be used for CDK inhibitor based selection systems accordingly.
Exceptions are lines that have lost the activity of dowllsk~ l mediators of the CDK
inhibitor pathways. For example, Rb--cell lines cannot be forced into arrest by 35 oveleA~"~ssion of pl6. In addition, certain cell lines may have incurred mutations in downstream genes other than Rb. For in.ct~nre, specific mutations in CDK4 render the wo 98/07886 PCT/US97/14514 mutant protein resistant to inhibition by pl 6. This defect has been shown to result from single amino acid substitutions in CDK4 protein that prevent binding of pl 6 to the enzyme without hllpai~ g catalytic activity. Wolfel eJ al., 1995, Science 269:1281-1284.
5 Similar mutations could illl~.Ç~l~ with the ability of other CDK inhibitors to carry out their tumor ~u~ eisor activity. Thus, it is critical to select cell lines that have intact growth control pathways dOwll~ ll of the particular CDK inhibitor such that theyrespond to ectopic CDK inhibitor eA~le~ion by entering cell cycle arrest.
Oncogene rulh~ . In another emborlim~nt, selection systems are g~n.,.aled 10 based on dissection of oncogene pathways. For example, a dolllil~t-negative oncogene or a dominant-negative fragment of an oncogene of interest may be ectopically ~ lessed such that growth is inhibited or apoptosis is ind~lced Selection and analysis of revertants results in the identification of genes encoding products which play a role do~ll~ in 15 the oncogene's pathway.
Many forms of dolllillalll-negative oncogene m~t~ntc have been engin~ered. For example, in the case of receptor tyrosine kin~ces, receptor ,~ n~ lacking an intact enzymatic domain have been shown to clo.~ -negatively inhibit the function, and thus 20 signal tr~n~d~lction, of the wild-type receptor. R~d~m~nn et al., 1992, Mol. Cell. Biol.
12:491 498; Kashles et al., 1991, Mol. Cell. Biol. 11:1454-1463; Millauer et al., 1994, Nature 367: 576-579. Further, naturally occurring do~ negative oncogenes havebeen identified, which have variable effects that depend heavily on the specific cell line in which they are expressed. Below (TABLE 1) are listed several exarnples from the lil.,lalule of the effects of do.,.i~ t negative proto-oncogenes on the growth and/or transformation prope.~ies of specific cells.

W O ~ /o86 PCTrUS97/14514 TABLE I

GENE RECIPIENT CELL EFFECT REFERENCE
c-JlnV MCF7 inhibition of colony Chen et al., 1996, formation Mol. Carcinog.
15:215-226 EGF-R Rat-l inhibition of DNA Daub et al., 1996, synthesis Nature 379:557-560 GRB2 NIH3T3 inhibition of Xie et al., 1995,J.
transformation Biol. Chem.
270:30717-30724 RAF NIH3T3 inhibition of growth Den~o et al.. 1995, in soft agar ~'omat. Cell. Mol.
Genet. 21:241-253 RAF GH4 ras-in~lced promotor Pickett et al., 1995, activation Mol. Cell. Biol.
S:6777-6784 ~X NIH3T3 natural growth Arsura et al., 1995, regulation Mol. Cell. Biol.
15:6702-6709 RAS SK-N-M~ inhibition of ERK2 van Weering et al., activation 1995, Oncogene 11:2207-2214 SRC endothelial inhibition of c-FOS Simonson et al. .
activation 1996,J. Biol. Chem.
271:77-82 In principle, dominant negative proto-oncogenes can serve in the sarne way as 30 tumor suppressor genes to arrest cells or prevent cell growth under certain conditions, thus providing a basis for selection of revertants.
Tumor Fo, ~tion And M~ In Vivo. In another embodiment, selection systems are generated based on the observation that some tumor cell lines do not forrn 35 tumors when injected into immunocompromised mice, while others do. For example~
pre~lign~nt melanoma cell lines typically are nontumorigenic when placed in immunoco~l.plo,..ised mice. In one embodiment~ such premalignant melanoma cells are W 098/07886 PCT~US97114t,14 injected sl~hcllt~neously in nude mice, and tumors are selected following injection of such prPm~lign~nt cells. These tumors arise from variants of the prem~lign~nt parental cells that have acquired mutations that permit growth in the mouse, llltim~t~ly forming iclçntifi~hle tumors. Thus, such a mouse nlmor formation system provides a meçh~nicm for selectinp cell revertants that have activated proto-oncogenes or inactivated tumor su~ essor genes that are involved directly in the ~,~ulsÇo,l.lation from a nontumorigenic phenotype to a tumorigenic one. These revertants can ~ubse~lu~ ly be studied to identify the proto-oncogenes or tumor ~up~ ,or genes involved in tumor formation.
In addition, overeA~ression of particular genes in tumor cell lines can render a tumorigenic line non-tumorigenic. Again, if such cells are injected in immnn~ con~ ised ~nim~lc, for example subcl~t~nt-ously, revertant cells may be selected that contain alterations in illlpOl ~ cell proliferation genes. Genes that 15 contribute to tumor formation in vivo may be directly analyzed and recovered.Apoptosis. In another embodiment, selection systems are gel.."ated based on the phenomenon of apoptosis, i.e., the ability of cells to undergo programmed cell death.
Apoptosis, is a mPçh~nicm h..po.~ll for the proper development of tissues. It is2 0 also implem~ntPd by the body during Iymphocyte maturation in order to remove self-reactive Iymphocytes. Finally, it serves as an illl~Ol~lt lllech ~.ic~ for m~ir~t~ g the integrity of fullv developed tissues in the context of various types of damage. For in~t~n~e, skin cells exposed to significant levels of ultraviolet light undergo apoptosis, presumably to elimin~te cells that have a high likelihood of being damaged in a way that is har nful to the long term health of the organism. Such "sunburned" cells. if they were not removed, might give rise to cancerous growth at an increased frequency. Ziegler et al., 1994, Nature 372:773-776.
In fact, apoptosis appears to be a general mech~ni~m used in many tissues for 30 elimin~ting prem~ n~nt, partially transformed cells. When these meçh~ni~mc are inactivated by mutation of genes such as pS3, cancer cells are at a selectivc advantage coll.l)~cd to normal cells and colllpaled to tumor cells in which apoptotic pathways are still intact. Such apoptosis-deficient cells are able to grow (or avoid self-inflicted death) 35 where others are not. Ziegler et al., 1994, Nature 372:773-776.

wo 98/07886 PC rluS97/14514 Cells in culture can be inrl~lced to undergo apol~olic death by a variety of stimuli, d~ g on the particular cells. For example, certain cells enter apoptosis after CA~O:IUlC to glucocorticoids, tumor necrosis factors, or other natural agents. In addition, 5 many cell types undergo apoptosis when exposed to radiation or chemothc. ~p~ cFurther, cells may be engi..~,.cd to contain genes which have been implicated in the control of or participation in apoptosis under the control of an inducible promoter. Such genes include, but are not limited to bc1-2 (Korsymeyer, 1992, Immunol. Today 13:285 288), c-myc (Shi et al., 1992, Science 257:212-214; Evan et al., 1992, Cell 69:119-128), 10 p53 (Rotter et al., 1993, Trends Cell. Biol. 3:46-49), TRPM-2/SGP (Kryprianou et al., 1991, Cancer Res. 51:162-166), and Fas/APO-I (Itoh et al., 1991, Cell 66:233-243).
Cell types which can be int~ ed to undergo apoptosis include, for example, Iymphocytes and tumor cells derived from Iymphocytes. Activation of the FAS antigen receptor in 15 maturing Iymphocytes activates an apoptosis program. If the FAS antigen is activated either by exogenous application of a FAS antibody (Velcich et al., 1995, Cell Growth Di~er. 6:749-757) or by ectopic eApie~ion of an activated form of the receptor, revertants that survive can be selected. Some of these rever~nts contain mutations in 20 genes do~llsl.ca.ll of the FAS antigen that operate in the sarne apoptotic pathway as FAS. Tre~tm~nt with certain steroid hormones or cross-linking of the T cell r~ceptors on the cell surface using, for example, an antibody, can also induce apoptosis in Iymphocytes and related cell or tumor lines. The 3DO line, for instance, responds to receptor cross-linking by undergoing apoptosis (Vito et al., 1996, Science 271:521-525)~
while murine thymoma W7 cells undergo apoptosis in response to dexamethasone (Bourgeois et al., 1993, Mol. ~ndocrinol. 7:840-851). Other cell lines undergo apoptosis when cultured at low density or in the ~bsen~e of specific serum factors (T~hi7~ki et al., 1995, Mol. Endocrinol. 7:840-851). In Friend erythrole~k~ cells, ove~cApr~ssion of 30 p53 results in apoptosis (Abrahamson et al., 1995, Mol. Cell. Biol. 15:6953 6960).
Ovc.cA~ s~ion of certain oncogenes in some turnor lines can, paradoxically, also induce apoptosis (Harrington et al., 1994, Curr. Opin. Genet. Dev. : 120-129). The morphogen retinoic acid induces programmed cell death in the P19 embryonic stem cell (Oka_awa er 35 al., 1996, .~. Cell Biol. 132:955 968). It is also possible to use various forms of traurna to induce apoptosis in a variety of cell types. For in~ es, ~ .,t of many cell types W O ~ a6 PCT~US97/14514 by DNA~ m~ing agents (e.g., certain chemot~,el~eulics, radiation) causes an apoptotic es~onse. Each of these methods provides the basis for selecting r~ S that fail to undergo apoptosis. These revertants can be used in turn to recover genes involved in 5 pathways of apoptosis.
Accordingly, the ability of cells to initiate apoptosis is used for the development of a genetic selection system; revertants that fail to die are i~ol~t~l Contac~ Ir hi~;t;D In still another embo~lim~nt selection systems are generated based on the fact that loss of growth regulation of cells is frequently reflected in the loss 10 of contact inhibition of cell proliferation. Accor.lingly, pools of cells which have lost contact inhibition are used to isolate contact-inhibited r~,r~.l~lls.
Most normal cells and many cell lines do not grow indefinitely in the body or inculture, rather they are inhibited by contact with their neighbors; this state of arrest is 15 known as contact inhibition. For exarnple, melanoma cell lines can be cultured under conditions where they become inhibited by contact (Valyi-Nagy et al., 1993, Int. ~
Cancer 54:159-165), as can neural precursor lines transfonned by polyoma large tumor (T) gene (Galiana e~ al., 1995, Proc. Natl. Acad. Sci. U.S.A. 92:1560-1564), derivatives 2 o of colon HT29 cells (Velcich e~ aL, 1995, Velcich et al., 1995, Cell Growth Diffrer.
6:749-757), human umbilical vein endothelial cells (Gaits e~ al., 1995, Bio~hem. J.
311:97-103), nonparenchymal epithelial cells (Johnson et al., 1995, Cancer Lett. 96:37-48), and many others.
In the past, the phenomenon of contact inhibition of cells has been used to select variants that continue to grow when saturation of the culture dish bottom has been reached. Foci have been isolated, comprised of cells that no longer respond to contact inhibitory signals and are often more likely to form tumors in animals than their parental counterparts. The initial identification of cellular oncogenes involved such an ~A~clhllental approach. Land et al., 1983, Na~ure 304:596-602; Copeland e~ al., 1979, Cell 17:993-10~2.
Growth Factors. In still another embol1im-ont, selection systems are generated based on the growth factor requilcln~ of m~mm~ n cells.
Many m~mm~ n cells in culture require the plesence of factors in the media which perrnit growth. In the absence of such factors, many cell types do not grow in WO ~ Oa6 PCTAUS97/1i~14 tissue culture. In several cases the relevant factors have been defin~(l For exarnple, in the absence of exogenous interleukin-2, certain T cells do not proliferate in culture.
MelQnomQ formation proceeds via a series of steps through which normal melanocytes evolve into fully m~tQctQtic melanomas. During this process the progressing tumor cells gradually lose their re~ -e.ll~ for specific exogenous factors (TABLE II). Norrnal melanocytes require factors such as phorbol ester, fibroblast grow~ factor (FGF), mf l~nocyte stimlllQtiTlg hormone-alpha (MSH-a), insulin, or insulin-like growth factor-l (IGF-l). In contrast, metQ~t~tic melanoma cells often require none of these factors. Cell 10 lines with interrnediate phenotypes require progressively fewer factors. This transition can be studied in culture such that factor-independent variants are isolated from earlier stage lines. These variants contain mutations that allow the cell to bypass the lc~luhe~llc~ll for one of the factors. Thus, they can be used as the starting point for 15 identification of genes that participate in the pathway of tumorigenesis involving escape from growth factor requirements.

.

CA 02263744 l999-02-l8 WO ~ /Do6 PCT~US97/14514 TABLE II
GRADUAL LOSS OF THE GROWTH FACTOR REQUIREMENT OF
MELANOCY'rES DURING MELANOI~ FORMATION.

MELANOMA PROGRESSION

Cell Type Requi.ell.cn~ Phenotype Melanocyte TPA Normal FGF
1 0 a-MSH
IGF-I
Nevus TPA Similar to melanocyte FGF
a-MSH
IGF-l Early melanoma FGFI Immortalized ~-MSH
IGF-l Primary melanoma IGF-I Tumorigenic Metastatic melanoma Migratory Accordingly, the growth factor requirement of cells can be exploited to provide a powerful selection system. More specifically, a particular growth factor is removed from the media, resulting in death of the vast majority of cells. Subsequently, variants that 25 continue to grow in the absence of the factor are selected; the mutations that have elimin~te~l the function of the regulatory pathway that prevents growth in the absence of the factor are identified and the co..~onding genes recovered.
Growth In rrO/nti'!P~ In still another embodiment, selection systems are 3 0 generated based on the observation that many cells in culture do not grow in isolation or at low density in culture. They requ re neighboring cells, presumably because these cells produce and secrete into the media growth factors that are nececs~-y for growth. If these factors do not reach a certain threshold concentration, the cells cannot grow.
Many cell lines can be grown in isolation from other neighbors only with great difficulty. For example, many T cell lines can only be cloned when the individual cells are placed on a "feeder layer" of other cells, i.e., cells that have been treated such that CA 02263744 l999-02-l8 W O 98/07886 PCTrUS97/14514 they do not divide, but cGlllin~e to produce growth factors that allow the T cell to proliferate into clone of descP~ nl~ This process of clonal growth can be used to select variants that are able to grow in the absen~e of a feeder layer.
Accordingly, revertant cells are selected that do grow at low density in colonies.
These cells presumably contain alterations in genes involved in a pathway of growth depen~çnre on neighbors, and hence, on secreted factors.
Immortnliz~tion. In still another embodiment, selection systems are g~l.e.aLed based on the observation that normal cells, e.g., primary ~ .. AIi~n cells, have a finite 10 life span in culture; they undergo a certain number of cell doublings and then die. The length of their life in culture depends on a variety of factors including the tissue of origin, the age of the animal from which the cells were derived. and the nature of the growth media. The period during which massive cell death occurs as the cells reach 15 their age limit is known as the crisis phase.
Accordingly, variants are selected that survive the crisis phase; these cells have undergone changes that lead to immortalization. In principle, this serves as a selection for imrnortalized cells with mutations in genes that normally limit life span.

D. Generation Of Growth Arrested Tumor Cell Lines As Selection Systems Where the generation of the selection systems of the invention involves the tA~,e~sion of a growth suppressing or apoptotic gene in cultured cells. the nucleotide 25 sequence encoding for the apoptosis regulator or inducer. or the growth suppressor. e.g., a tumor suppressor gene or a dominant-negative oncogene or oncogene mutant, or afunctional equivalent thereof, is inserted into an a~ lu~llate ~A~,ie;,~ion vector, i.e., a vector which contains the necesc~ry elements for the l-ails~ ion and translation of the 30 inserted coding sequence, and is introduced into the host cell system.
Expression Systems For The E~r~ Of The GrowJh S~ ng Genes.
Typically, where a gene encoding the growth S~lppl~,sSor or the apoptosis-in~lcing product is introduced in a transformed cell, an inducible promoter system is used for the control of its expression. An inducible promotor permits growth of the cells that contain the eApl~saion construct under conditions where the promotor is turned off. Whendesired, the promotor can be inrhlced and the cells become growth arrested due to the ~ . , CA 02263744 l999-02-l8 W O 9~n~6 PCTrUS97/14514 e~ c~sion of the tumor su~ or gene. ~ltern~tively, if the efficiency of gene transfer is ~AL~ cly high (as has been reported for certain letruvhuses) and a selection for cells that have taken up DNA is employed, regulated promoters may be ~licp~n~ed with. In this case, the tumor su~pl~,i,sor gene or apoptosis-in~ inp gene is carried on the retrovirus along with a select~hle marker such as h~ llly.;in re~i~t~n~e. Revertants that express the selectable marker but do not die or undergo cell cycle arrest are then isolated directly.
Several suitable inducible promoters are established for use in m~mm~ n cells~
10 and many more can be envisioned. Ex~ll~lcs include, but are not limited to, interferon inducible promoter systems, such as the promoter for 3'-5' poly (A) synthPt~se or the Mx protein, which are in~luced by, e.g., a poly inosine and cytosine duplex S~hun~ch~r et al., 1994, Virology 203:144-148. Other examples include the HLV-LTR, which can 15 be inrluced with, e.g, dimethylsulfoxide (DMSO), or the metallothionein promoter system, which is inducible by heavy metal ions Haslinger ef al., 1985, Proc. NatL Acad Sci. U.S.A. 82:8572-8576. Other inducible promoters include the tetracycline and lac lc~)icssor systems, where a rel)icssor, i.e., tetracycline or IPTG, respe~ ely, m~int~in~
20 the promotor in an inactive state. Thus in the ~bsence of exogenous tetracycline or IPTG
the promoter is s~,iesse-d (tet system: Gossau and Bujard, 1992, Proc. Natl. Acad. Sci.
U.S.A. 89:5547-5551; lac system: Fieck et al., 1992, Nucleic Acid Res. 20:1785-1791).
In a specific embodiment of the invention, a IPTG lac switch system has been used as inducible promoter system. Specifically, the promotor of the inducible construct contained sequences from the Rous Sarcoma Virus (RSV) long terminal repeat (LTR)that act as a potent transcriptional initiator located upsllcaln of the coding sequence of the gene to be t~ cssed (in the specific example disclosed herein, pl6). Between the translational start site and the RSV LTR were operator sequences derived from the E.
3 ~ coli lac operon. These sequences are sufficient for binding of the lac repressor. In the presence of functional lac rcl)ressol, transcription from the RSV LTR is drarnatically reduced by the lac operator sequences. However, when IPTG is added to the culture media, lac lepl~ ,SOl molecules are prevented from blocking the ~ sc,.~tion of the gene 35 to be e~ ssed; the desired mRNA is synth~si7~1 and protein is produced. Further.
proper ~ ession of genes encoding the growth slll)plessor or the apoptosis-in~lucing product may require specific initiation signals for efficient translation of inserted cell proliferation gene encoding sequences. These signals include the ATG initiation codon and ~ ent sequences. In cases where the entire gene, inrll~ing its own initiation codon and ~dj~c~nt sequences, is inserted into the ayyloyl;ate c;A~resaion vector, no additional tr~n~l~tion~l control signals may be needed. However, in cases where only a portion of the gene encoding the growth auyplessor or the apoptosis-inrlueing product is inserted, exogenous translational control signals, including the ATG initiation codon, must be provided. Furthermore, the initiation codon must be in phase with the reading 10 frame of the cell proliferation gene encoding sequence to ensure translation of the entire insert. These exogenous translational control signals and initiation codons can be of a variety of origins, both natural and synthetic. The efficiency of e.~lei,aion may be enh~nc~(l by the inclusion of ~propl,ate transcription ~nh~n~er elements, transcription 15 terminators, etc. See, Bittner et al., 1987, Methods in Enzymol. 153:516-544.Though transient cl~yression of the growth su~ ssor or the apoptosis-inducing product might be sufficient in some cases, most typically the gene encoding the growth au~ylessor or the apoptosis-in~lncine product will be stably eAyrcssed in the host cells.
2 0 Host cells are transforrned with DNA encoding the desired product controlled by apployl;ate t;~ esaion control elements, including a promoter, which typically is inducible, see, supra, enh~ncer sequences, llallsclil~tion terminators, polyadenylation sites, etc., and a selectable marker.
Following the introduction of foreign DNA, engin~-ered cells are allowed to grov.
for 1-2 days in an enriched media, and then switched to a selective media. The select~hle marker in the recombinant plasmid confers resistance to the selection and allows cells to stably integrate the plasmid into their chromosomes and grow to form colonies. The colonies are cloned and exr~nded into cell lines.
A variety of transfection techniques are cllllclllly available to transfer DNA in viJro into cells; including calcium phosphate-DNA p~cipit~lion, DEAE-Dextran transfection, ele~llopoldlion, liposome me~ ted DNA transfer or transduction with recombinant viral or retroviral vectors, and may be used in the metho-is of the present 3 5 mvenhon.

A nurnber of selection systems may be used in the invention, il~clulhlg but not limited to the herpes simplex virus thyrnidine kinase (Wigler et al., 1977, Cell 11:223), hy~ hil~c-guanine phosphoribosyltransferase (Szybalska and Szybalski, 1962, Proc.
5 Natl. Acad. Sci. U.S.A. 48:2026), and a~lenine phosphoribos~ ,r~,ase (Lowy et al., 1980, Cell 22:817) genes can be employed in tk-, hgprt~ or aprt- cells, le*,e~ ely. Also, h~olite re~;c~ re can be used as the basis of selecti-)n for dhfr, which confersrecict~n~e to methu~ ate (Wigler et al., 1980, Proc. Natl. Acad. Sci. U.S.A. 77:3567;
O'Hare et al., 1981, Proc. NatL Acad. Sci. U.S.A. 78:1527); gpt, which confers l~s;~ n~e 10 to mycophenolic acid (Mulligan and Berg, 1981, Proc. Natl. Acad. Sci. U.S.A. 78:2072);
neo, which confers le~;c~ -ce to the arninoglycoside G-418 (Colberre-Garapin et al., 1981, J. Mol. Biol. 150:1); and hygro, which confers recict~nl~e to hygromycin (Santerre et al., 1984, Gene 30:147) genes. Additional select~hle genes have been described, 15 namely trpB, which allows cells to utilize indole in place of tryptophan; hisD, which allows cells to utilize histinol in place of histidine (Hartrnan and Mulligan, 1988, Proc.
Natl. Acad. Sci. U.S.A. 85:8047); and ODC (o~ hine dec~l,o~ylase) which confers recict~nce to the o,.,illline decarboxylase inhibitor, 2-(difluoromethyl)-DL-ornithinP, 20 DFMO (McConlogue, 1987, in: Current Co~ "~lications in Molecular Biologv, Cold Spring Harbor Laboratory ed.).
In the case where high-efficiency retroviral delivery systems are used for the generation of cell lines, selection systems are not nPce~c~rily le~ ed due to the high efficiency of retroviral gene transfer. Retroviral ~ ession systems which are enpinPPred to encode and express the desired recombinant gene may involve the use of infectious or non-infectious particles that undergo only a single initial round of infection. In the forrner case, the genome of the virus ...~ inc regulatory seqL~.,ces, structural genes, and p~Clr~ging signals nPces~ry for the generation of new virus particles, while genes 30 conferring oncogenic potential are removed. After the retroviral proteins are synthPsi the host cell packages the RNA in new particles, which can undergo further rounas of infection.
Preferably, however, non-infectious retroviral vectors are used in the present 3 5 methods, which require a helper virus to provide the structural genes necessary to encode viral structural proteins. The helper virus' pac~ging signal which is required to wo gs/07886 PCT/USg7/14514 c .~ ..s~ te the helper viral RNA into particles is destroyed, and as a result only the recombinant retroviral vector C~ tlg a functional paeL aging signal and the gene of interest, but lacking the retrovirus' structural colllponents can be incorporated in an 5 particle. Conse~uclllly, the resl-lting retrovirus can infect a target cell, and its genetic inforrnation may be inserted into the host's genome; however, the so transferred genetic h~,lllation is biologically c~ ;..ed becallse genes ecernti~l for viral growth are not provided. Methods for consllucling and using retroviral eA~,-e~ion systems are well known in the art and reviewed, for eA~,l~lc, in Miller and Rosman, 1992, Bio~echni~ues 1 ~ 7:980-990.
~ dentif cation Of Transfectants Or Transformants Tha~ Express The Growth 5~, r c~ 8 Or Aropt~ Cene. The host cells which contain the coding sequence and which express the biologically active gene product may be identified by at least four 15 general approaches; (1) DNA DNA or DNA-RNA hybridization; (2) the ples~.lce or ~bs~n-~e of "marker" gene functions; (3) ~esescing the level of transcription as llleaauled by the expression of cell proliferation gene mRNA transcripts in the host cell; and (4) detection of the gene product as Ill.,~ulcd by imm~.o~c.~ y or by its biological activit~v.
2 0 In the first approach, the presence of the sequence en~o~ling the desired product inserted in t~e ~A~lession vector is detected by DNA-DNA or DNA-RNA hybridization using probes comprising nucleotide sequences that are homologous to the sequenceencoding the desired product, lei,l,e~,lively, or portions or derivatives thereof.
In the second approach, the recombinant e~ es~ion vector/host system is identified and selected based upon the presence or absence of certain "marker" gene functions (e.g, thymidine kinase activity, recict~nce to antibiotics, resict~n~e to methotrexate, trans~llllalion phenotype. For example, if the gene encoding the growth ~-lpplei.~or or the apoptosis-inrlucing product is inserted within a marker gene sequence 30 of the vector, recombinants CQI.I~ g the gene encoding the gro~,vth ~u~lessor or the apoptosis-inducing Froduct are identified by the absence of the marker gene function.
Alternatively, a marker gene is placed in tandem with gene encoding the growth s.~ ssol or the apoptosis-in~ cing product under the control of the same or different 35 promoter used to control the eA~,~ssion of the sequence encoding the growth suppressor or the apoptosis-in~ Cing product. Expression of the marker in response to induction or W O 98107886 PCT~US97/14514 selection indicates t:A~r~ ion of the sequence encoding the growth S~)l'e.~OI or the apoptosis-in-l-lcing product.
In the third approach, transcriptional activity for the gene encoding the growth~iessor or the apoptosis-in~iucing product is ~ceessed by hybri~li7~tion assays. For plc~ RNA is isolated and analyzed by Northern blot using a probe homologous to the gene encoding the desired product or particular portions thereof. Alternatively, total nucleic acids of the host cell are extracted and assayed for hybridization to such probes.
In the fourth a~pl~ ach, the ~A~leision of the gene protein product is ~cseseed 10 immlmologically, for ~A~llple by Western blots, immllnn~ee~ys such as radioi~ r..~.o-~,.eci~,;~lion, enzyme-linked imm~mo~ceays and the like. The ultimate test of the success of the e~c~iession system, however, involves the detection of the biologically active gene product. A number of assays can be used to detect activity of the gene encoding the 15 growth ~up~ie~,or or the apoptosis-inducing product including, but not limited to, transformation assays, growth assays, etc.

E. Identifi~ t ~n And Isolation Of Novel Cell Proliferation Genes 2 0 The identified revertant cells obtained as described above are used in the methods of the invention to reveal and characterize novel cell proliferation genes.
In one embollim~nt~ subtractive hybridization of cDNA is used to identify sequences that are responsible for the reverted phenotype. In this process, cDNA probes are ~ cd from the parental cell line and one of the selected revertants. These probes are then hybridized to filters generated from cDNA libraries or genomic DNA libraries.
Most preferably, the source of the libraries may be the parental cell line, the revertant line or both. The filters are probed with the two probes sepaldlely, most conveniently in duplicate, and diff~lences in signal intensity are noted. Clones of interest col~t~il.;..g 3 ~ sequences that show different signal strengths in the hybridi_ations to the parental probe and the revertant probe are idertified and isolated. A general protocol for subtractive hybridization of cDNA can be found, arnong other places, in Sambrook et al., 1994, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Press, New York.
In another embollim~nt~ a probe is g~u~.~t~d that is enriched for diff.,.~ ially~iA~I~ssed sequences. More specifically, cDNA is synth~si7~od from the parental or CA 02263744 l999-02-l8 W O ~ /oo6 PCT~US97/14514 r~ .,kLnt cell lines and hybridized in solution against rnRNA from the other line. Shared sequences are removed from the cDNA probe by, for example, avidin-biotin capture, by binding to hydroxyapatite, or by any other suitable procedure. The r~m~ining single str~nll.od and thus unique cDNA sequences are then used to probe cDNA library filters to identify and isolate the differentially t;AI~less~d se~ ,ces, ~lt~ tively~ they may be cloned and eY~Inin~d directly. Hedrick et al., 1984, Nature 308:149-153.
In still another embo~imPnt, differentially ~A~,lessed genes are detected by cloning and sequencing of high numbers of cDNA sequence rlc~g~ from the parent and 10 l~ sources. Comparison of the sequences then leads to information about relative eAI"ession levels. This, for example, can be accomplished by sequence analysis of 3' eAI,iessed sequence tags (ESTs) a method pioneered by The Institute for Human Genome Research (TIGR) and by Hurnan Genome ~ciences, Inc. (HGS). Lennon et al.,15 1996, Genomics 33:1Sl-152. An alternative is to analyze small sequence tags cloned in multiple copies into plasmids or phage, a method known as Serial Analysis of Gene Expression (SAGE). Velculescu eJ al., 1995, Science 270:484-487.
In still another embodiment polymerase chain reaction (PCR) is employed for 20 identification of differentially ~A~l~ssed sequences, in an approach known as "differential display." The method takes advantage of the pseudo-random amplification that ensues when multiple primers of ~bi~ y sequences are placed in a reaction tube with random-primed cDNA. Cenain fragment~ amplify and these are analyzed by denaturing gel electrophoresis. If two different cDNA sarnples are used separately, i.e., one from a parental line, one from a revenant. the two PCR-amplified product sets can be run side-by-side on a gel. The intensities of different sized bands are compared, bands of different illl.,.lsily are excised from the gel, reamplified and cloned for further analysis.
Zhao e- al., 1995, Biotechniques 18:842-850.
In still other embo~iimPnt~ gene ex~iession is monitored and col,lpared using protein levels as the output parameter. One method of differential protein analysis, for example, involves colnp~ison of two-dimPn~ional protein gels, whereby one dimension is non-den~ g, the second ~lim~n~ion is denaturing, to identify protein spots ehat are 35 non-identically ~x~ ssed in the two samples. Differences in samples of total protein isolates from the parental line are identified, the coll~s~,onding proteins are then purified WO ~ 70D6 PCTAUS97tl4514 and sequenced, in order to 1-1tim~tely gain enough information for cloning of the co~ onding gene or cDNA.
In still another embodiment, an array of oligonucleotides or cDNA fr~emt~nt~
gridded out on a solid support is used as the probe against labeled cDNA prel,~ed from the re~,~.l cell line. The hybridization signals of the cDNAs are used to del~ ...i..c which sequences are e~e;,~ed at different levels. Schena et al., 1995, Science 270:467-470.
~ ol~* Of The Cell Proliferation Gene Or Its cDNA. Once a DNA or peptide 10 fragment of the cell proliferation gene has been id~rltifie~ and sequenced, the co.,~ onding gene or cDNA clone is isolated by standard methods described in, for exarnple, Sambrook et al., supra.
Gene Recovery In Reir.,tu, b Induced By P~th~_gL~.s. If the revertants are 15 in-luced by a specific agent, i.e., a p~ bagen, the relevant ge-.e or genes may be recovered even more easily. As outlined above, see, Section VI.B.2., supra, a p~ ulbagen may be nucleic acid encoding a cell proliferation associated protein, or a protein fragment acting akin to a doll,i~ negative mutant that disrupts crucial 20 protein/protein interactions involving cell proliferation genes.
In cases where revertants are based on he eA~"es~ion of a cell proliferation gene, or on a protein or protein fragment acting akin to a dolllinhl,l-negative mutant of signal transduction pathways, the cell proliferation gene or protein/protein interaction underlying the promotion of cell growth can be determined by means of identification of the nature of the p~.lulbagen. In most cases, one of two scenarios will apply. First, the ~,.lul~agen may be deterrnined to be a cell proliferation gene itself. In such cases.
detel"lillation of the nature of the cell proliferation gene is simply accomplished by analysis of the p~ u~bagen product. Second, the perturbagen may act by disrupting a 3 ~ protein/protein interaction in a growth control pathway. ln this case, two steps are required: first, the cDNA encoding the p~ bagen needs to be identified and isolated.
In a second step, the protein/protein interaction affected by the dominant-negative mutant is identified employing assays known in the art, suitable for the identification of 35 protein/protein interactions such as the yeast two-hybrid system. Bartel et al., 1995, Methods Enzymol. 254:241-263.

W O 98/07886 PCT~US97/14~14 P~olo~.cal Fu r~;~l And Relevance Of The IdentiSed Cell Proliferation Gene And Its rrl ~ The metho~c of the invention deseribed above pennit the identifir~tion of highly yrcselc~,led c~nrli~tes for crueial co,ll~,onellla of eellular grow~h proliferation 5 pathways. In order to eonfirm their speeific biological funetion and relevanee, these ç~n~ t~s are tested in suitable in vitro and in vivo assays. The design of the assays will vary ~iepen~ling on the growth eontrol pathway whieh was targeted by a partieular seleetion system. For example, genes identified with seleetion systems based on, e.g, the o~rel~AI,.e ,sion of a twnor a.lpplesaor may be cA~lessed in eultured eells, e.g., NIH3T3 10 eells, and their effeet on eell growth, DNA synthesis, foeus form~tion, growth in soft agar, modifieation, e.g, phosphorylation of co~ ol,cl,ls or substrates in signaltr~n~ etion pathways, complex formation of signal tr~n~duetion COIlll)Oll-.n~S, including adapter moleeules, changes in the pattern of gene eA~uic;,~ion, e.g., induetion of 15 l.anscli~tion faetors, ineluding cjun, c-fos, c-myc, ete. is determined. ln vitro assays are d~sign~(l to determine substrate or ligand binding, phosphorylation signal tr~n~d~ ion moleeules, ete. Further, loss of funetion mutations may be generated in mice (knockout miee) or l~lsgenic miee may be produced in which the gene is ectopically ~A~ ;.aed.
20 Dominant-negative lllu~lta may be enginPered in mouse or in human eells. Anti-sense eonstructs or oligonueleotides may be employed to downJegulate e~Jle~aion of thespeeific gene. In certain eases, the gene or its homologs may be studied in yeast cells.

F. Expression Of The Cell Proliferation Gene In Cultured Cells In order to express a biologically active cell proliferation gene in eultured eells, the nucleotide sequence encoding the cell proliferation gene, as identified and isolated as deseribed in Section VI.E., supra, or a funetional equivalent, is inserted into an ~p.o~l;dte ~,le~;on vector, i.e., a veetor which coll~ains the nPce~ elements for 30 the l,a.~s.;l;l,lion and tramslation of the inserted coding sequence. The cell proliferation gene produets as well as host eells or eell lines transfeeted or transformed with reeombinant cell proliferation gene eA~iesaion vectors cam be used for a variety of ~,u".oses. These include ~i~gnnstic uses, and gen~.aling antibodies (i.e., monoclonal or 35 polyclonal) that bind to the cell proliferation gene, as well as the identifieation of analogues or drugs that act on the cell proliferation gene, and for diagnostic purposes.

CA 02263744 l999-02-l8 W 0~ 6 PCT~US97/14514 1. Expression Systems Methods which are well known to those skilled in the art are used to consku~ ,ssion vectors cont~ininE the cancer gene coding sequence and 5 a~ op.iate ~ s~;li,utionaUtranslational controi signals. These methods include in vitro ,.,co...bi..ant DNA techniques, synthetic techniques and in vivo recombination/genetic recombination. See, for example, the techniques described in S;il--block et al., supra;
and Ausubel et al., 1989, Current Protocols in Molecular Biology~ Greene Publishin~
Associates and Wilev In~ ,iellce. N.Y
A variety of host~ es~ion vector systems are utilized to express the cell proliferation gene encoding se~uc;nce. These include, but are not limited to, microorg~nicm~ such as bacteria transformed ~,vith recombinant bacteriophage DNA, plasmid DNA or cosmid DNA ~ is~ion vectors co.~ .;..g the cell proliferation gene 15 encoding sequence; yeast transformed with recombinant yeast ~ es~ion vectors co ~ E the cell proliferation gene PnCorling sequence; insect cell systems inf~cted with recombinant virus ~ re~ion vectors (e.g., baculovirus) co.-~;..;ng the cell proliferation gene encoding sequence; plant cell systems infected with recombinant virus expression 20 vectors (e.g., cauliflower mosaic virus, CaMV; tobacco mosaic virus, TMV) or transforrned with recombinant plasmid e~ ion veClors (e.g., Ti pl~smid) cu..~ i..g the cell proliferation gene encoding sequence; or animal cell systems infected with recombinant virus ~.ei~ion vectors (e.g., adenovirus, vaccinia virus) including cell lines engint~ered to contain multiple copies of the cell proliferation gene DNA either stably amplified (C~O/dhfr) or unstably amplified in double-minute chromosomes (e.g., murin~
cell lines) The t;~lu-ession elements of these systems vary in their sl.e~ and specificitiesDepending on the host/vector system Utili7~ any of a number of suitable transcription 30 and translation elements, including con~ uli~e and inducible promoters, may be used in the exp~es~ion vector For exarnple, when cloning in bacterial systems, induciblepromoters such ~ pL of bacteriophage ~, plac, ptrp, ptac (ptrp-lac hybrid promoter) and the like are used; when cloning in insect cell systems, promoters such as the baculovirus 3 5 polyhedrin promoter are used; when cloning in plant cell systems, promoters derived from the genome of plant cells (e.g., heat shock promoters; the promoter for the small wo 98/07886 PCrlUS97/1~514 subunit of RUBISCO; the promoter for the chlorophyll a/b binding protein) or from plant viruses (e.g., the 35S RNA promoter of CaMV; the coat protein promoter of T~V) are used; when cloning in ~..,...,..~li~n cell systems, promoters derived from the genome of ,..~...n~ n cells (e.g., metallothionein promoter) or from ".~ Ali~n viruses (e.g, the adenovirus late promoter; the vaccinia virus 7.5K promoter) are used; when g~ ,ldt.illg cell lines that contain multiple copies of the cell proliferation gene DNA SV40-, BPV-and EBV-based vectors are used with an ~,fol,liate select~hle marker.
In bacterial systems a number of ~ ;on vectors are ad~ tdgeo~sly sel-~t.~l 10 depending upon the use jntend~cl for the cell proliferation gene el~ple;,~zd. For example, when large quantities of cell proliferation gene product are to be produced for the generation of antibodies or to screen peptide libraries, vectors which direct the ~ aion of high levels of fusion protein products that are readily purified are desirable. Such 15 vectors include, but are not limited to, the E. coli e~,les~ion vector pUR278 (Ruther et al., 1983, EMBO J. 2:1791), in which the cell proliferation gene coding se~lence may be ligated into the vector in frame with the lacZ coding region so that a hybrid AS-lacZ
protein is produced; pIN vectors (Inouye and Inouye, 1985, Nucleic Acids ~es. 13:3101-20 3109; Van Heeke and Schuster, 1989, J. Biol. Chem. 264:5503-5509); and the like.
pGEX vectors may also be used to express foreign polypeptides as fusion proteins with glutathione S-transferase (GST). In general, such fusion ~)luleins are soluble and are easily be purified from Iysed cells by adsorption to glutathione-agarose beads followed by elution in the presence of free glutathione. The pGEX vectors are designed to include thrombin or factor Xa protease cleavage sites so that the cloned polypeptide of interest can be released from the GST moiety.
In yeast, a number of vectors col.l~init.g constitutive or inducible promoters may be used. For a review, see, Current Protocols in Molecular Biolo~y, Vol. 2, 1988, Eds.
30 Ausubel et al., Greene Publish. Assoc. & Wiley Interscience, Ch. 13; Grant et al., 1987, Expression and Secretion Vectors for Yeast, in: Methods in Enz~molo~Y, Eds. Wu &Groccm~n, 1987, Acad. Press, N.Y., Vol. 153, pp. 516-544; Glover, 1986, DNA Clonin~
Vol. II, IRL Press, Wash., D.C., Ch. 3; and Bitter, 1987, Heterologous Gene Expression 35 in Yeast, Methods in Enzvmolo~Y. Eds. Berger & Kimrnel, Acad. Press, N.Y., Vol. 152, .

W O ~t~ooC PCT~US97114514 pp. 673-684; and The Molecular Biolo~l of the Yeast Saccl~ y~es. 1982, Eds.
Strathem et al., Cold Spring Harbor Press, Vols. I and II.
In cases where plant c:xylca~ion vectors are used, the e~ ion of the cell proliferation gene enro~ling sequence may be driven by any of a number of promoters.
For example, viral promoters such as the 35S RNA and 19S RNA promoters of CaMV
(Brisson et al., 1984, Nature 310:511-514), or the coat protein promoter of TMV
(T~k~m~t~u et al., 1987, EMBO J. 6:307-311) may be used; altematively, plant promoters such as the small subunit of RUBISCO (Coruzzi et al., 1984, EMBO J.
10 3:1671-1680; Broglie et al., 1984, Science 224:838-843); or heat shock promoters, e.g, soybean hspl7.5-E or hspl7.3-B (Gurley et al., 1986, Mol. Cell. Biol. 6:559-565) may be used. These constructs are introduced into plant cells using Ti pl~mi~l$, Ri plasmids, plant virus vectors, direct DNA transfomnation. microinjection, cle~ opo.alion, etc. For 15 reviews of such techniques, see, for exarnple, Weissbach and Weiseb~cll~ 1988, Methods for Plant Molecular Biolo,ey~ Aç~dennic Press, NY, Section VIII, pp. 421-463; and Grierson and Corey, 1988, Plant Molecular Biolo~v, 2d Ed., Blackie, London, Ch. 7-9.
An altemative eAp-~;,sion system which can be used to express the cell 20 proliferation gene is an insect system. In one such system, Autographa californica nuclear polyhidrosis virus (AcNPV) is used as a vector to express foreign genes. The virus grows in Spodoptera frugiperda cells. The cell proliferation gene encodingsequence may be cloned into non ess~ l regions (for exa"l~,le the polyhedrin gene) of the virus and placed under control of an AcNPV promoter (for example the polyhedrin promoter). Successful insertion of the cell proliferation gene encoding sequence results in inactivation of the polyhedrin gene and production of non-occluded recombinant virus (i.e., virus lacking the proteinaceous coat coded for by the polyhedrin gene). These recombinant viruses are then used to infect Spodoptera frugiperda cells in which the 30 inserted gene is eApressed. See, for example, Smith e~ al., 1983, J Viol. 46:584; U.S.
Patent No. 4,21 ~,051.
In m~mm~ n host cells, a number of viral based t:~le~ion systems may be utili7~i In cases where an adenovirus is used as an e~l~le~aion vector, the cell35 proliferation gene encoding sequence is ligated to an adenovirus L,~s~ tion/translation control complex, e.g, the late promoter and tripartite leader sequence. This chimeric W 098/07886 PCT~US97/14514 gene may then be inserted in the adenovirus genome by in vitro or in vivo recombination.
Insertion in a non-es~nti~l region ofthe viral genome (e.g, region El or E3) will result in a recombinant virus that is viable and capable of c~les:,;ng the cell proliferation gene 5 in infected hosts. See, for example, Logan and Shenk, 1984, Proc. Natl. Acad. Sci.
U.S.A. 81 :3655-3659. Alternatively, the vaccinia 7.5K promoter may be used. See, for example, Mackett et al., 1982, Proc. Natl. Acad. Sci. U:S.A. 79:7415-7419; Mackett et al., 1984, J. Virol. 49:857-864; Panicali et al., 1982, Proc. Natl. Acad Sci. U.S.A.
79:4927-4931.
Specific initiation signals may also be required for efficient ~ lalion of inserted cell proliferation gene encoding se~luences. These signals include the ATG initi~tion codon and ~dj~cent sequences. In cases where the entire cell proliferation gene~including its own initiation codon and adjacent sequences, is inserted into the ~ ;ale 15 cx~ ion vector, no additional translational control signals may be needed. However, in cases where only a portion of the cell proliferation gene en~o~ing sequence is inserted, exogenous translational control signals, including the ATG initiation codon, must be provided. Furthermore, the initiation codon must be in phase with the reading frame of 2 o the cell proliferation gene encoding sequence to ensure translation of the entire insert.
These exogenous translational control signals and initiation codons can be of a variety of origins, both natural and synthetic. The efficiency of cAI,le;,~ion may be enh~n~ed by the inclusion of a~p,o~,;ate ~ sc,il)tion enh~n~er elements, lld"sc,ilJ~ion terminators, etc.
See, Bittner et al., 1987, Methods in ~nzymol. 153:516-544.
In addition, a host cell strain may be chosen which modulates the t:~p,ession of the inserted sequences, or modifies and processes the gene product in the specific fashion desired. Such modifications (e.g., glycosylation) and pl'OCC;~ lg (e.g., cleavage) of protein products may be hlll)GI~lt for the function of the protein. Dirr~.cl~ host cells 30 have characteristic and specific mech~ni~m~ for the post-translational processing and modification of proteins. Appropriate cells lines or host systems can be chosen to ensure the correct modification and processing of the foreign protein cx,~lcssed. To this end, eukaryotic host cells which possess the cellular m~chin.ory for proper processing of the 35 primary transcript, glycosylation, and phosphorylation of the gene product may be used.

WO 98~'~.70D6 PCT/US97/14~14 Such ~ n host cells include, but are not }imited to, CHO, VERO, BHK, HeLa, COS, MDCK, 293, W138, etc.
For long-term, high-yield production of recombinant proteins, stable t"E,lei,sion is pl~f~ c;d. For example, cell lines which stably express the cell proliferation gene may be ~ngin~ered~ Rather than using eA~ ion vectors which contain viral origins of replication, host cells are transformed with the cell proliferation gene encoding DNA
controlled by a~propl;ate e~l~ ;on control elc-n-,llL~ ~e.g, ~IOllIVI_-, enh~nr,çr, sequences, ~ sc.;~ion te.--~ alors, polyadenylation sites, etc.), and a select~hle marker.
10 Following the introduction of foreign DNA, e~ cd cells are allowed to grow for 1-2 days in an enriched media, and then are switched to a selective media. The select~hle marker in the recombinant plasmid confers recict~nre to the selection and allows cells to stably integrate the plasmid into their chromosomes and grow to form colonies which in 15 turn can be cloned and exp~n~ed into cell lines.
A nurnber of selection systems may be used, including but not limited to, the herpes simplex virus thymidine kinase (Wigler et al., 1977, Cell 11:223), h~ox~llh;lle-guanine phosphoribosyl~ Çe.dse (Szybalska and Szybalski, 1962, Proc. Na~l. Acad Sci.
20 U.S.A. 48:2026), and adenine phosphoribosyltransferase (Lowy et al., 1980, Cell 22:817) genes can be employed in tk, hgprt~ or aprt~ cells"~ .e~ ely. Also, ~ntimPt~bolite recict~nce can be used as the basis of selection for dhfr, which confers recict~nce to methotrexate (Wigler et al., 1980, Proc. Natl. Acad. Sci U.S.A. 77:3567; O'Hare et al., 1981, Proc. Natl. Acad Sci. U.S.A. 78:1527); gpt, which confers recict~nGe to mycophenolic acid (Mulligan and Berg, 1981, Proc. Natl. Acad. Sci. U.S.A. 78:2072);
neo, which confers .~ e to the aminoglycoside G-418 (Colberre-Garapin et al.~
1981, J. Mol. Biol. 150:1); and hygro, which confers lec;s~ e to hy~,olllycill (Santerre et al., 1984, Gene 30:147) genes. Additional selectable genes have been described, 3 ~ namely trpB, which allows cells to utilize indole in place of ~ tophan; hisD, which allows cells to utilize histinol in place of histidinP (Hartman and Mulligan, 1988, Proc.
Natl. Acad Sci. U.S.A. 85:8047); and ODC (ornithine decarboxylase) which confersrecict~nre to the o,llilhi,-e decarboxylase inhibitor, 2-(difluoromethyl)-DL-ornithine, 35 DFMO (McConlogue, 1987, in: Current Co"ll"u,lications in Molecular Biolo~v~ Cold Spring Harbor Laboratory ed.).

wo 3~7Oo~ ~CT/US97/14514 Host cells cf..~ the coding sequence and which express the biologically active cell proliferation gene product may be idPntified by several general approaches, including DNA-DNA or DNA-RNA hybridization, the p~csence or ~bsPr~e of "marker"
gene functions, ~c~ec~ p~ ofthe level of ~lansc~;~tion as measured by the ~ ion of cell proliferation gene mRNA ~ S~,liyL~ in the host cell, and the detection of the gene product as measured by immlm~cs~y or by its biological activity. These approaches are described in more detail in Section VI.D., supra.
The cell proliferation gene products can also be e,-~.essed in ~ g, ~ic ~nim~lc 10 Animals of any species, including, but not limited to, mice, rats, rabbits, guinea pigs, pigs, micro-pigs, goats, and non-human primates, e.g., baboons, monkeys, and rh;~ 7~es may be used to generate llansg.nic animals using methods known in the art to introduce the cell proliferation associated transgene into animals to produce the 15 founder lines of ~ sgellic animals. Such techniques include, but are not limited to, pr~n~lcle~r microinjection (Hoppe, P.C. and Wagner, 1989, U.S. Pat. No. 4,873,191);
retrovirus mP~ tPcl gene transfer into germ lines (Van der Putten et al., 1985, Proc.
Natl. Acad. Sci. U.S.A. 82:6148-6152); gene targeting in embryonic stem cells 20 (Thompson et al., 1989, Cell 56:313-321); electroporation of embryos (Lo, 1983, Mol.
Cell. Biol. 3:1803-1814); and sperm-me~ tP~ gene transfer (Lavitrano et al., 1989, Cell 57:717-723); etc. For a review of such techniques, see Gordon, 1989, Transgenic ,~nim~ls, Intl. Rev. Cytol. 115:171-229, which is incorporated by lel~.~,nce herein in its entirety.

The present invention provides for Iransgenic animals that carry the cell proliferation associated t.dnsgene in all their cells, as well as animals which carry the transgene in some, but not all of their cells, i.e., mosaic ~nim~lc The ~ lsgel-e may be integrated as a single transgene or in con~t~mPrs, e.g., head-to-head t~nfl~rnc or head-to-30 tail t~nrlentc The l~1sgel1e may also be selectively introduced into and activated in aparticular cell type by following, for example, the teaclling of Lasko et al. (Lasko, M. e~
al., 1992, Proc. Natl. Acad. Sci. U.S.A. 89: 6232-6236). The regulatory sequences required ~or such a cell-type specific activation will depend upon the particular selected 35 cell type, and will be appal~el-t to those of skill in the art.

W O 98107886 PCT~US97114514 When it is desired that the cell proliferation ~ccoci~ted l,a,lsgene be illlegldt-,d into the chromosomal site of the endogenous cell proliferation gene homologue, gene targeting is preferred. Briefly, when such a technique is to be utili7Pd vectorscor.~ .g some nucleotide sequences homologous to the endogenous cell proliferation gene homologue are ~lecignPd for the purpose of illt. ~ ing, via homologous recomhin~tion with chromosom~l sequences, into and disrupting the function of the nucleotide sequence of the endogenous cell proliferation gene. The ~a~lsgene may als be selectively introduced into a particular cell type, thus inactivating the endogenous cell 10 proliferation gene in only that cell type, by following, for exarnple, the te~hing of Gu et al. (Gu e~ al., 1994, Science 265:103-106). The regulatory sequences ~equired for such a cell-type specific inactivation will depend upon the particular cell type of interest, and will be ~palcn~ to those of skill in the art.
Once ~ sgc.-ic animals have been generated, the t;~l"es ,ion of the recombinant cell proliferation gene is assayed utili7ing standard techniques. Initial scree~ g may be ~ecornplichP(l by Southern blot analysis or PCR techniques to analyze animal tissues to assay whether integration of the ~ sgel1e has talcen place. The level of mRNA
2 0 ~ ession of the transgene in the tissues of the transgenic animals may also be ~csesced using techniques which include, but are not limited to, Northern blot analysis of cell type samples obtained from the animal, in situ hybridization analysis, and RT-PCR. Sarnples of the cell proliferation gene-e~plessil1g tissue, are evaluated irnrnunocytoch~ mic~lly using antibodies specific for the cell proliferation associated L,ansg~ne product.

G. Use Of The Identified Cell Proliferation Gene Sequences For Diagnosis Of Aberrant Or Uncontrolled Expression Of Cell Proliferation Gene Products Related To Cell Proliferation Disorders Or Cell Proliferation D.sc~de~ Pr~di~l)r~ ~c-3 ~ The cell proliferation gene DNAs identified with the selection systems of the present invention have a number of uses for the diagnosis of ~liC~ces resulting from their aberrant ~ ei,;,ion. For exarnple, probes generated accoldh~g to the cell proliferation gene DNA are used in hybridization assays of autopsies or biopsies to 35 diagnose abnorrnalities in their ~ lession, thereby providing a basis for a defined and targeted treatment of the disease.

W 098/07886 PCTrUS97/14514 A variety of mPthn-lc can be employed for the di~gnostic and pr~lG~Iic evaluation of f~i~ç~5 related to aberrant c;Apr~ssion of cell proliferation associated genes, including cancer, and for the identification of subjects having a predisposition to such disorders. Such m~thnAc may, for example, utilize reagents such as the cell proliferation gene's nucleotide sequences described in Section Vl.E., supra, and antibodies directed to the cell proliferation gene product, as described, in Section V~.I., infra. Specifically, such reagents may be used, for exarnple, for: (l) the detection of the l,les~.lce of cell proliferation gene mutations, or the detection of either over- or under-~l,les~ion of the 10 cell proliferation gene's mRNA relative to the state found in normal cell activation;
(2) the detection of either an over- or an under-abl-nd~rlre of cell proliferation gene product relative to the normal state; and (3) the detection of p~llulba~ions or abnormalities in the signal transduction pathway mç~ tPd by the cell proliferation gene 15 product.
The methods described herein may be performed, for example, by ~Itili7ing pre-packaged diagnostic kits comprising at least one specific cell proliferation gene nucleotide sequence or antibody reagent directed to its gene product described herein, 20 which may be conveniently used, e.g, in clinical settin~.c, to diagnose patients exhibiting cell proliferation disorder abnormalities.
For the detection of cell proliferation gene mutations, any nucleated cell can be used as a starting source for genomic or messenger nucleic acid. For the detection of the cell proliferation gene's ~le;,sion or its gene products, any cell type or tissue in which the cell proliferation gene is e~r~ssed, most typically the afflicted tissue exhibiting a disease related to ul,conl.olled cell proliferation~ may be utili7P~
Nucleic acid-based detection techniques are described in Section Yl.G. 1., in~a.
Peptide detection techniques are described in Section VI.G.2., infia.

1. Detection Of The Cell Proliferation Gene And Its Transcript In one embodiment, the cell proliferation gene cDNA or fraEment~
thereof are used as a probe to detect the ~ lession of the cell proliferation gene mRNA.
35 For exarnple, sections of tissue samples may be pl~ ed and e~c~minPd by in situ hybridization with a suitable, labelled probe. Alternately, mRNA extracts may be - ~L3 -wo 98/07886 PCT/US97114514 plepared and analy_ed in Northern blot analysis. Alt~orn~tively~ synthetic oligonucleotides ~l~ociEnP~ accol~ing to the cell proliferation gene's cDNA sequence may be gell~"aLcd and used as hybridization probes. Detailed description of suitable protocols 5 can be found, for exarnple, in Sambrook et al., Molecular Clonin~: A Laboratorv Manual, 2nd Ed., Cold Spring Harbor (1989).
In one embo~lim~nt, the level of the cell proliferation gene's e~ cJ~ion is assayed by ~letectin~ and measuring its ~ sc,i~tion. For example, RNA from a cell type or tissue known, or c~ ,e.;le~l to over- or under-express the cell proliferation gene, such as 10 c~lcelous tissue, is isolated and tested utili7in~ hybridization or PCR techniques such as are described herein. The isolated cells can be derived from cell culture or from a patient. The analysis of cells taken from culture may be a nPcesc~ry step in the~C~ec~ rnt of cells .o be used as part of a cell~based gene therapy technique or, 15 alternatively, to test the effect of compounds on the e~l,lci,~ion of the cell proliferation gene. Such analyses may reveal both quantitative and qualitative aspects of the tA~,lcs~ion pattem of the cell proliferation gene, incllltling activation or inactivation of its gene e~.pl~ion.
In another emborlimpntJ probes co.lc~onding to the cell proliferation gene sequence are employed for analysis of the genomic DNA in order to identify individuals who are predisposed for, e.g., a particular type of cancer. Predisposed individuals are then monitored on â frequent basis in order to ensure early diagnosis of potential disease.
which drastically increases the likelihood of thel~ye.llical succec.~ Detailed description of suitable protocols for such Southern blot analysis can be found, among other places, in Sambrook et al., Molecular Cloniny: A LaboratorY Manuah 2nd Ed., Cold Spring Harbor (1989).
Hybridization probes for Northern blot, Southern blot, and in situ hybridization30 may be labeled by a variety of reporter groups, including radionuclides such as 32P,35S, and 3H (in the case of in situ hybridization), or enzymatic labels, such as alk~line phosphatase, coupled to the probe via avidin/biotin coupling systems, and the like. The labeled hybridization probes may be ple~Jal~d by any method known in the art for the 35 synthesis of DNA and RNA molecules. See, Section VI.H., in.fia. An additional use for nucleic acid hybridization probes involves their use as primers for polymerase chain W O 98/07886 PCT~US97/14514 reaction (PCR). PCR is described in detail in U.S. Patents 4,965,188, 4,683,195, and 4,800,195.
In still other embodimP~tc, mutations within the cell proliferstion gene can be detected by lltili7ing a nurnber of techniques. Nucleic acid from any nllrle~tPd cell can be used as the starting point for such assay techniques, and may be isolated acco,dillg to standard nucleic acid ple~alion procedures which are well known to those of skill in the art.
DNA may be used in hybridization or amplification assays of biological samples 10 to detect ~bnt nn~lities involving gene ~L~u~;lule, including point mutations, insertions, deletions and chromosomal rearrangements. Such assays include, but are not limited to, Southern analyses, single stranded conformational polymorphism analyses (SSCP), and PCR analyses.
Diagnostic methods for the detection of cell proliferation gene-specific mutations can involve for example, cont~ctine and incl~b~ting nucleic acids including recombinant DNA molecules, cloned genes or degenerate variants thereof, obtained from a sample, e.g., derived from a patient sample or other applo~.;ate cellular source, with one or more 20 labeled nucleic acid reagents including recombinant DNA molecules, cloned. genes or deg~.leldle variants thereof, under conditions favorable for the specific ~nnP~linp of these reagents to their complementary sequences within the cell proliferation gene. Preferabl~, the lengths of these nucleic acid reagents are at least 15 to 30 nucleotides. After incllb~tion, all non-annealed nucleic acids are removed from the nucleic acid molecule hybrid. The presence of nucleic acids which have hybridized. if any, is then detected.
Using such a detection system, the nucleic acid from the cell type or tissue of interest can be immobilized, for example, to a solid support such as a membrane, or a plastic surface such as that on â microtiter plate or polystyrene beads. In this case, after 30 inrubation, non-annealed, labeled nucleic acid reagents are easily removed. Detection of the rem~ining, ~nn.-~lerl labeled cell proliferation gene's nucleic acid reagents is accomplished using standard techniques well-known to those in the art. The cell proliferation gene sequences to which the nucleic acid reagents have annealed is35 conll)aled to the ~nne~lin, pattern expected from a norrnal gene sequence in order to determine whether a gene mutation is present.

.

CA 02263744 l999-02-l8 W Ot~t~tb~ PCTAUS97/14514 Alternative t~ noStiC methods for the deteetion of the cell proliferation gene'sspecific nucleic acid molecules, in patient samples or other applopl;ate cell sources, may involve their ~mplific~tion, e.g., by PCR (the e,A~ .ental embodiment set forth in 5 Mullis, K.B., 1987, U.S. Patent No. 4,683,202, see, supra), followed by the deteetion of the amplified molecules using techniques well known to those of skill in the art. The res-llting amplified sequenees can be c~lnpal~,d to those which would be ~e~,ted if the nucleic acid being amplified contained only normal copies of the cell proliferation gene in order to dete~rnine whether a gene mutation exists.
Additionally, well-known genotyping techniques can be pc.roll.led to identify individuals carrying mutations in the cell proliferation gene. Such techniques include, for example, the use of restriction fragment length polymorphisms (RFLPs), whichinvolve sequence variations in one of the recognition sites for the specific restriction 15 enzyme used.

2. Detection Of The Cell Proliferation Gene Product Antibodies directed against wild type or mutant cell proliferation 2 0 gene products or conserved variants or peptide fr~gmPntc thereof, may also be used as cell growth disorder r~ Tlostics and prognostics, as described herein. Such diagnostic mPtho-lc, may be used to detect abnormalities in the level of the gene's ex~"es~ion, or abnormalities in the structure and/or temporal, tissue, cellular, or subcellular location of its gene product, and may be perforrned in vivo or in vitro, such as, for example, on biopsy tissue.
The tissue or cell type to be analyzed will generally include those which are known, or suspected, to aberrantly express the cell proliferation gene, such as, for exarnple, cancerous tissue. The protein isolation methods employed herein may, for 3 ~ example. be such as those described in Harlow and Lane (Harlow, E. and Lane, D., 1988, "Antibodies: A Laboraton~ Manual", Cold Spring Harbor Laboratory Press, Cold Sprin~ Harbor, New York). The isolated cells can be derived from cell culture or from a patient. The analysis of cells taken from culture may be a nPce~,., y step in the 3 5 ~ccPccm~nt of cells that are used as part of a cell-based gene therapy technique or, W 098/07886 PCTrUS97/14514 alternatively, to test the effect of compounds on the ~A,~,ies:,ion of the cell proliferation gene.
For example, antibodies, or fr~gm~ntc of antibodies useful in the present 5 invention, such as those described in Section VI.I., infra, may be used to ~-~An~ ely or qualitatively detect the plesence of the cell proliferation gene products or conserved variants or peptide fr~gm~nte thereof. This can be accomplished, for example, byimmunofluo~escence techniques employing a fluol~,scelllly labeled antibody (see, this Section, infra) coupled with light microscopic, flow cytometric, or fluolill,~ll,c detection.

The antibodies (or fr~gm~ntc thereof) or fusion or conjugated proteins useful inthe present invention may, additionally, be employed histologically, as in immlInofluorescenr~e, imm~Inoelectron microscopy or non-immI~no assays, for in situ 15 detection of the cell proliferation gene products or conserved variants or peptide fra~m~ntc thereof, or for catalytic subunit binding (in the case of labeled catalytic subunit fusion protein).
In situ detection may be accomplished by removing a histological specimen from 20 a patient, and applying thereto a labeled antibody or fusion protein of the present invention. The antibody (or fragmert) or fusion protein is preferably applied byoverlaying the labeled antibody (or fragment) onto a biological sample. Through the use of such a procedure, it is possible to determine not only the presence of the cell proliferation gene product, or conserved variants or peptide fr~gmentc~ but also its distribution in the ex~minPd tissue. Using the present invention, those of ordinary skill will readily perceive that any of a wide variety of histological methods (such as st~ininy procedures) can be modified in order to achieve such in situ detection.
Immnnoa~c~ys and non-immuno~cc~ys for cell proliferation gene products or 30 conserved variants or peptide fr~gm~ntc thereof will typically comprise incub~ting a sample, such as a biological fluid, a tissue extract, freshly harvested cells, or Iysates of cells which have been incubated in cell culture, in the presence of a detect~bly labeled antibody capable of identifying the cell proliferation gene products or conserved variants 35 or peptide fr~gm~ontc thereof, and detecting the bound antibody by any of a number of techni4ues well-known in the art.

W O 98/07886 PCTrUS97/14S14 The biological sample may be brought in contact with and immobilized onto a solid phase support or carrier such as nitrocellulose, or other solid support which is capable of immobilizing cells, cell particles or soluble proteins. The support may then be washed with suitable buffers followed by tre~tmPnt with the ~letpct~hly labeled antibody or fusion protein. The solid phase support may then be washed with the buffer a second time to remove unbound antibody or fusion protein. The amount of bound label on solid support is then det~Pcte~l by convPntion~l means.
By "solid phase support or carrier" is int~nd~Pd any support capable of binding an 10 antigen or an antibody. Well-known ~ulJp~ or carriers include glass, polystyrene, polypropylene, polyethylene, dextran, nylon, amylases, natural and modified celluloses, polyacrylamides, gabbros, and ...~g.-~ I;te. The nature of the carrier can be either soluble to some extent or insoluble for the purposes of the present inven~ion. The support 15 material may have virtually any possible structural configuration so long as the coupled molecule is capable of binding to an antigen or antibody. Thus, the support configuration may be spherical, as in a bead, or cylindrical, as in the inside surface of a test tube, or the external surface of a rod. Alternatively, the surface may be flat such as 20 a sheet, test strip, etc. Preferred :tU~ j include polystyrene beads. Those skilled in the art will know many other suitable carriers for b nding antibody or antigen, or will be able to readily ascertain the same.
The binding activity of a given lot of antibody or fusion protein is determined according ~o well kno~n methods. Those skilled in the art will be able to readily determine operative and optimal assay conditions.
With respect to antibodies, one of the ways in which the antibody can be detect~bly labeled is by linking the same to an enzyme and use in an enzyme immnno~cc~y (EIA) (Voller, 1978, Diagnostic Horizons 2:1-7, Microbiological 30 Associates Quarterly Publication, Walkersville, MD); Voller et al., 1978, J. Clin. Pathol.
31:507-520; Butler, 1981, Me~h. Enzymol. 73:482-523; Maggio, E. (ed.), 1980, EnzYme Imm-lno~c~y, CRC Press, Boca Raton, FL,; Ishikawa et al., (eds.), 1981, Enzvrne Irnmulloassa~ Kgaku Shoin, Tokyo). The enzyme which is bound to the antibody will 35 react with an app~ ,iate ~ubs~ale~ preferably a chromogenic substrate, in such a manner as to produce a chemical moiety which can be detecte-l for example, by WO ~ a6 PCTrUS97/14514 s~,e~llophotometric, fluo~ ic or by visual means. Enzymes which can be used to lt~tect~kly label the antibody include, but are not limited to, malate dehydrogenase, staphylococcal nuclease, delta-5-steroid isolll~"dse, yeast alcohol dehydrogenase, alpha-glyc~.ophosphate, dehydrogenase, triose phosrh~te ison.cla3e, horseradish peroxidase, alkaline phosFh~t~ce7 aspar~gin~cp7 glucose oxidase, beta-galactosidase, ribonl~cle~e, urease, cat~l~ce7 glucose-6-phnsph~tr dehydrogenase, glucoarnylase and acetylcholinesterase. The detection can be accomplished by colo,h,lcL,;c methods which employ a chromogenic substrate for the enzyme. Detection may also be accomplished by 10 visual col,ll,~ison of the extent of enzymatic reaction of a substrate in culllpalison with similarly p,~e~d,~d standards.
Detection may also be accomplished using any of a variety of other immunoassays. For example, by radioactively labeling the antibodies or antibody 15 fragments, it is possible to detect the cell proliferation gene product through the use of a radioirnmunoassay ~RIA) (see, for example, Weilllldub, B., Principles of Radioimmnno~c~vs. Seventh Training Course on Radioligand Assay Techniques, The Endocrine Society, March, 1986). The radioactive isotope can be cletected by such 20 means as the use of a gamma counter or a scintill~tion counter or by autoradiography.
It is also possible to label the antibody with a fluor~scent compound. When the fluorescently labeled antibody is exposed to light of the proper wave length, its presence can then be detected due to fluorescence. Among the most commonly used fluorescent labeling compounds are fluorescein isothiocyanate, rho~lAmine, phycoerythrin, phycocyanin, allophycocyanin, o-phth~ ehyde and fluoresc~mine.
The antibody can also be detect~bly labeled using fluorescence emitting metals such as '52Eu, or others of the l~nth~nide series. These metals can be ~tt~rhrd to the antibody using such metal chelating groups as diethylenetriaminepentacetic acid (DTPA) 3 0 or ethylenr~ r inrtnraacetic acid (EDTA).
The antibody also can be detectably labeled by coupling it ~:o a chemill~min~scent compound. The presence of the chemihlmin~sce~t-tagged antibody is then detPrminrd by detecting the ple~ellce of lumh~eice .cc that arises during the course of a chemical 35 reaction. Examples of particularly useful chemilnm;llesce..l labeling compounds are W 098/07886 PCTrUS97/14514 I1lmino1, isoluminol, theromatic acridinium ester, imi~1~7l-1e, acridinium salt and oxalate ester.
Likewise, a biol~ osce~l compound may be used to label the antibody of the present invention. Biolllminescen~e is a type of chemil~ scr .ce found in biological systems in, which a catalytic protein incleases the efficiency of the chernih~ n-~sce-,1 reaction. The plese.lce of a biolllminescent protein is det~ d by ~letectine the;,el~ce of l ..;nfsce .~e. Illlpol~ biol~ sc~ compounds for ~ oses of l~beling are luciferin, luciferase and aequorin.

H. Use Of The Identified Cell Proliferation Gene Sequences For Development Of Antisense Al)~..r ' -- And Ribozymes Also within the scope of the subject invention is the use of oligonucleotide 15 or oligoribonucleotide sequences comprising ~nticen~e DNA or RNA molecules orribozymes that function to inhibit the translation of the cell proliferation gene mRNA.
For example, ~nti~nce DNA or RNA molecules act to directly block the translation of the cell proliferation gene by binding to the targeted mRNA and thus preventing protein translation.
Ribozymes are enzymatic RNA molecules capable of catalyzing the specific cleavage of RNA. The mtorh~ni~m of ribozyme action appears to involve site specific hybridization of the ribozyme molecule to complementary sequences of the target RNA, followed by a endonucleolytic cleavage. In one embodiment of the invention, ribozyme 25 molecules are engineered that specifically catalyze endonucleolytic cleavage of mRNA of the cell proliferation genes identified with the selection systems of the invention.
Suitable target sites for ribozyme activity are i~ntifi~d by first sc~nning the target molecule for potential ribozyme cleavage motifs, second by evaluating the30 structural features of the about 15 to 25 amino acids co,lespollding to the region of the target molecule co~ ;t-;~.g the identified cleavage recognition site. Further. the suitability of the c~n.li~l~te targets may also be evaluated by testing their accessibility to hybridization with comple ..~ y oligonucleotides, using ribonuclease protection assays.
Bordonaro et al., 1994, Biotechniques 16:428 430.
The labeled hybridization probes, see, Section VI.G.l., supra, ~nti~ence DNA andRNA oligonucleotides and ribozymes of the subject invention are plepaled by any - SO -wo 98/07886 PCT/USg7/14514 method known in the art for the synthesis of DNA and RNA molecules. For example,oligonucleotides may be sy~shtosi7Pd çllrmic~lly using c~ cially available DNA or RNA synthPi~7f,s like m~r~in~s sold by Applied Biosystems. Alternatively, RNA
5 molecules may be gel-e.dted by in vitro and in vivo L,~ulsc~ tion of DNA sequences encoAing the Anti~rnce RNA molecule. Such DNA sequences may be illcc,ll.oldIed into a wide variety of vectors which comprise suitable RNA polymerase promoters such as the T3, T7, or the SP6 polymerase promoters. Alternatively, ~nti~rnce cDNA constructs that synthesi7~ ~nti~en.~e RNA consliluli~ ely or inducibly, may be introduced stably into cell 1 ~ lines.
Various moAific~tions to the DNA and RNA molecules may be introduced as a means of increasing the intracellular stability and half-life. For example, fl~nkin~
sequences of ribo- or deoxy- nucleotides may be added to the 5' and/or 3' ends of the 15 molecule, or phosphorothioate or 2' O-methyl rather than phosphodiester linkages may be used within the oligonucleotide backbone. Xu et al., 1996, Nucleic Acid Res.
24: 1 602- 1 607.

I. Generation And Use Of Cell Proliferation Gene ~r~tiho~
Various procedures known in the art may be used for the producti~ n of antibodies to epitopcs of the recombinantly produced cell proliferation genes identified and isolated employing the selection systems of the present invention. Such antibodies include but are not limited to polyclonal, monoclonah chimeric, single chain~ Fab fr~gmentc and fragments produced by an Fab expression library. Such antibodies may be useful, e.g., as Ai~gnostic or theldp~u~ic agents. As th~ldpt;ulic agents, neutralizing antibodies, i.e., those which compete for binding with a ligand, substrate or adapter molecule, or interfering with the cell proliferation genes activity, are of especially 3 0 plef ,..~d interest.
For use as diagnostic agents, monoclonal antibodies that bind to the cell proliferation gene are r~Aioartively labeled allowing detection of their location and distribution in the body after injection. Radioactivity tagged antibodies may be used as a 35 non-invasive diagnostic tool for im~ging in vivo the plesence of a tumors and met~t~ce~
associated with the ~xl~ie;,~ion of said cell proliferation gene.

W O 98/07886 PCTrUS97/14~i4 lmmlmotoxins may also be de~ n~l which target ~,~lotoxic agents to specific sites in the body. For example, high affinity monoclonal antibodies may be covalently complexed to bacterial or plant toxins, such as f~iphthPria toxin, abrin, or ricin. A
general method of ~lepa~a~ion of antibody/hybrid molecules may involve use of thiol-cro~linkin,~ reagents such as SPDP, which attack the primary amino groups on theantibody and by ~ --lfide exchange, attach the toxin to the antibody. The hybridantibodies may be used to specifically elimin~t~ cells c~ e;~aing the cell proliferation gene.
For the production of antibodies, various host ~nim~lc are immllni7~d by injection with the cell proliferation gene protein including, but not limited to, rabbits, mice, rats, etc. Various adjuvants may be used to incl~ase the immllnological res~,ollse, depending on the host species, including but not limited to Freund's (complete and incomplete), 15 mineral gels such as all~minl~m hydroxide, surface active al~b~ ces such as Iysolecithin, pluronic polyols, polyanions, peptides, oil emulsions, keyhole limpet hemocyanin, di~ ul)h~llol, and potentially useful human adjuvants such as BCG (bacille C~lm~tte-Guerin) and Corynebacterium parvum.
Monoclonal antibodies to the cell proliferation gene may be prepared using any technique which provides for the production of antibody molecules by continuous cell lines in culture. These include, but are not limited to, the hybridoma technique originally described by Kohler and Milstein, 1975, Nature 256:495-497, the human B-cell hybridoma technique (Kosbor et al., 1983, Immunology Today 4:72; Cote et al., 1983, Proc. Natl. Acad Sci. U.S.A. 80:2026-2030) and the EBV-hybridoma technique (Cole et al., 198~, Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc., pp. 77-96). In addition, techniques developed for the production of "chimeric antibodies" (Morrison et al., 1984, Proc. Natl. Acad Sci. U.S.A. 81:6851-68S5; N~ub~lgcl et al., 1984, Nature 312:604-608; Takeda et al., 1985, Nature 314:452-454) by splicing the genes from a mouse antibody molecule of approl,liate antigen specificity together v~ith genes from a human antibody molecule of applo~liate biological activity can be used. Altematively, techniques described for the production of single chain antibodies (U.S. Patent 4,946,778) can be adapted to produce cell proliferation gene-specific single chain antibodies.

W O 98/07886 PCT~US97/14514 Antibody fi~g...- .-t~ which contain specific binding sites of the cell proliferation gene may be geneldled by known techniques. For example, such fr~mPr~t~ include, but are not limited to, F(ab')2 r~AI~...c ~ which can be produced by pepsin digestion of the antibody molecule and the Fab fra~mr1-tc which can be generated by redllcing the~iculfi~e bridges of the F(ab')2 r.~..,. I,tc ~lte~n~qtively, ~ab eA~re~ion libraries may be constructed (Huse et al., 1989, Science 246:1275-1281) to allow rapid and easy identification of monoclonal Fab rrA~n~f~ with the desired specificity to the cell proliferation gene.

J. Use Of Revertant Cells Or The ~-o'-~e~l Cell Proliferation Genes For The Identifi~ti~n Of Compounds Useful For The Tr~at~l~t Of Disease Related To Uncontrolled Cell Proliferation 1. Id~ r~ on Of Compounds The revertant cells identified using the selection system process of the invention, may be used directly, i.e., without isolation of the relevant cell proliferation gene, for the identification and isolation of compounds inhibiting aberrant cellproliferation. Alternatively, the cell proliferation genes identified by the process of the 20 invention may be isolated and used for in vitro or in vivo assays for the identification and isolation of compounds specifically interfering with their activity.
More specifically, the identified revertant cells may be exposed to chemical compounds or compound libraries, and compounds exhibiting growth inhibition may be 25 identified. Alternatively, the identified cell proliferation genes may be cA~uressed in suitable eA~ression systems, design~d to allow for high-throughput testing of compounds from any source to identify molecules having an inhibitory effect on the cell proliferation genes.
Nucleotide sequences encoding the cell proliferation genes identified and isolated using the selection systems of the invention may be used to produce the coll~sl~onding purified protein using well-known methods of recombinant DNA technology. Among the many publications that teach methods for the cAlJlession of genes after they have been isolated is Gene Expression Technology. Methods and EnzvmologY. Vol.:18S~ edited by 35 Goeddel, Academic Press, San Diego, California (1990).

W O 3~~JOD6 PCTrUS97/14514 The cell proliferation genes may be cA~ ,s:ied in a variety of host cells, either prokaryotic or eukaryotic. In many cases, the host cells would be eukaryotic, more preferably host cells would be m~mm~ n. ~lost cells may be from species either the same or dirf~lent than the species from which the cell proliferation gene encoding nucleotide sequences are naturally present, i.e., endogenous. Advantages of producing the cell proliferation genes by recombinant DNA technology include obt~ hlg highly en,;ched sources of the plot~ins for purification and the availability of simplified purification procedures. Methods for recombinant production of yu~olt;ins are generally 10 very well established in the art, and can be found, arnong other places in Sambrock et al., supra.
In one embodiment of the invention, cells transformed with ~Ayiession vectors encoding the cell proliferation gene are cultured under conditions favoring eAyie;~ion of 15 the cell proliferation gene sequence and the recovery of the recombinantly-produced protein from the cell culture. A cell proliferation gene produced by a recombinant cell may be secreted or may be cont~in~od intracellularly, dc~ ding on the nature of the cell proliferation gene and the particular genetic construction used. In general, it is more 20 convenient to prepare recombinant proteins in secreted form. Purification steps will depend on the nature of the production and the particular cell proliferation gene produced. Purification methodologies are well established in the art; the skilled artisan will know how to optimize the purification conditions. General protocols of how to o~lh~ the purification conditions for a particular protein can be found, among other places, in Scopes in: Protein Purification: PrinciPles and Practice. 1982, Springe-Verlag New York, Heidelberg, Berlin.
In addition to recombinant production, cancer peptide fr~gm~nt~ may be produced by direct peptide synthesis using solid-phase techniques. See, Stewart et al., Solid-Phase 30 Peptide Svnthesis (1969), W. H. Freeman Co., San Francisco; and Merrifield, lg63, J.
Am. Chem. Soc. 85:2149-2154.
~ n vitro polypeptide svnthesis may be performed using manual techniques or by automation. Automated synthesis may be achieved, for example, using Applied 35 Biosystems 431A Peptide Synth~i7~r (Foster City, California) following the instructions provided in the instruction manual supplied by the m~nl~f~c~rer.

W O 98/07886 PCTrUS97/14514 In one embodiment of the invention, the cell proliferation genes and/or ~ ,leising cell lines e~ e~ing the cell proliferation gene are used to screen for antibodies, peptides.
organic molecules or other ligands that act as agonist or antagonists of the cell proliferation gene activity. For example, antibodies capable of hlle.r~ g with the activity, e.g., enzymatic activity of the cell proliferation gene, or with its interaction with a ligand, adapter molecule, or substrate are used to inhibit the cell proliferation gene function. In cases where amplification of the cell proliferation gene function is desired, antibodies which mimic, e.g., a ligand, an adapter molecule or ~ub~lldle of the 10 co"~s~ollding the signal tr~ncd~rtion pathway may be developed. Obviously, if desired, antibodies may be generated which modify the activity, function, or specificity of the cell proliferation gene.
Alternatively, screelling of peptide libraries or organic compounds with 15 recombinantly e~l,ressed cell proliferation gene protein or cell lines e~.e~illg the cell proliferation gene may be useful for identification of th~ ;uLic molecules that function by inhibiting, enh~nring, or modifying its biological activity.
Synthetic compounds, natural products, and other sources of potentially 20 biologically active materials can be screened in a number of ways. The ability of a test compound to inhibit, ~nh~n~e or mo~ te the function of the cell proliferation gene may be determined with suitable assays measuring the cell proliferation gene function. For example, responses such as its activity, e.g., enzymatic activity, or its ability to bind its ligand, adapter molecule or substrate may be determined in in vitro assays. Cellular assays may be developed to monitor a modulation of second mecc~nger production~
changes in cellular metabolism, or effects on cell proliferation. These assays may be performed using conventional techniques developed for these purposes. Finally, the ability of a test compound to inhibit, enh~nre or mo~ te the function of the cell 30 proliferation gene will be measured in suitable animal models in vivo. For example, mouse models will be used to monitor the ability of a compounds to inhibit the development of solid tumors, or effect reduction of the solid tumor size.
In one embodiment of the invention, random peptide libraries consisting of all 3 5 possible combinations of amino acids ~tt~clled to a solid phase support are used to identify peptides that are able to interfere with the function of the cell proliferation gene.

WO 98/07a86 PCT/US97/14514 For example, peptides may be identified binding to a ligand-, adapter molecule- or substrate binding site of a given cell proliferation gene or other filnrtiorl~l dorn~in~ of the cell proliferation gene, such as an enzymatic (l~m~in Accordingly, the scleening of peptide libraries may result in col.lpou~lds having thc,~ ic value as they interfere with its activity.
Identification of molecules that are able to bind to the cell proliferation gene may be ~cornrli~h~d by scl~e~ g a peptide library with recombinant soluble cell proliferation gene protein. Methods for ~ esaion and purification of the selPcted cell 10 proliferation genes are described in Section VI.F., supra, and may be used to express recombinant full length cell proliferation gene protein or fr~Jnentc thereof, depenAing on the functional domains of interest.
In order to identify and isolate the peptide/solid phase support that interacts and 15 forms a complex with the cell proliferation gene, it is l-~cessi~y to label or "tag" the cell proliferation gene molecule or fragment thereof. For example, the cell proliferation gene may be conjugated to enzymes such as ~ lin~ ph~sph~t~ce or horseradish peroxidase or to other reagents such as fluolescelll labels which may include fl~lorescein isothyiocynate 20 (FITC), phycoerythrin (PE) or rho~l~mine. Conjugation of any given label to the cell proliferation gene may be performed using techniques that are routine in the art.
In addition to using soluble cell proliferation gene molecules or fr~ment~ thereof, in another embodiment, peptides that bind to the cell proliferation gene may be identified using intact cells. The use of intact cells is preferred for use with cell proliferation genes which comprise cell surface receptors, which require the lipid domain of the cell membrane to be functional. Methods for ge~ aLing cell lines e~,iea~ing the cell proliferation genes identified with the selection systems of the invention are described in Secfion Vl.F., supra. The cells used in this technique may be either live or fixed cells.
3 ~ The cells are incubated with the random peptide library and will bind to certain peptides in the library. The so formed complex between the target cells and the relevant solid phase support/peptide may be isolated by standard methods known in the art, including differential centrifugation.

Wo 98/07886 PcTrusg7/145l4 As an alternative to whole cell assays for membrane bound lCC~.~)tOI~ or lcc~ptols that require the lipid domain of the cell ~ lbl~c to be functional, the receptormolecules can be fccon~ u~ed into liposomes where a label or "tag" can be att~rh~
In another embo~lim~nt, cell lines that express the cell proliferation gene or, AltçmAtively isolated cell proliferation gene protein or Lag~ Ic thereof, are used to screen for molecules that inhibit, enhAnl~e, or modulate the cell proliferation gene activity or signal tr~ncduction. Such molecules may include small organic or inor~ànic cc,l"poul,ds, or other molecules that effect the cell proliferation gene activity or that 10 promote or prevent the complex formation with its ligand, adapter molecules, or substrates. Synthetic compounds, natural products, and other sources of potentially biologically active materials can be screened in a number of ways, which are generally known by the skilled artisan.
Fo~ example, the ability of a test molecule to interfere with the cell proliferation gene function may be measured using standard biochemical techniques. Alternatively, cellular r~ OnSe5 such as activation or ~plession of a catalytic activity, phosphorylation or dephosphorylation of other proteins, activation or mo~inlAtion of 20 second mesc~n~er production, changes in cellular ion levels, association, dissociation or translocation of cignAIIing molecules, or ~-ans~ tion or translation of specific genes may also be monitored. These assays may be performed using conventional techniques developed for these purposes in the course of scleel.i"g.
Further, effects on the cell proliferation gene function may, via signal transduction pathways, affect a variety of cellular processes. Cellular processes under the control of the cell proliferation gene signAIIing pathway may include, but are not limited to, normal cellular functions, proliferation, differentiation, mAintenAnre of cell shape, and adhesion, in addition to abnormal or potentially deleterious processes such as unregulated 30 cell proliferation, loss of contact inhibition and, blocking of differentiation or cell death.
l he qualitative or q~A~ e observation and measurement of any of the described cellular processes by techniques known in the art may be advantageously used as a means of scoring for signal trAnc~lnrtion in the course of s~;fec~lillg.
Various technologies may be employed for the screening, identification, and evaluation of compounds which interact with the cell proliferation gen_s of the invention.

W O 98t07886 PCT~US97/14~14 which c~lllpvullds may affect various cellular ~l1oce;,ses under the control of said cell proliferation gene.
For example, the cell proliferation gene or a functional derivative thereof, in pure or semi-pure forrn, in a membrane pfepdl~lLion, or in a whole live or fixed cell is ed with the cG111poll..d. Subsequently, under suitable conditions, the effect of the compound on the cell proliferation gene function is S~lu~ ;7~1 e.g., by measuring its activity, or its signal transduction, and co111~ing the activity to that of the cell proliferation gene, inr.ub~ted under same conditions, without the compound, thereby 10 dct~-....l,;..~ whether the compound stimulates or inhibits the cell proliferation gene's activity.
in addition to the use of whole cells t;~l.1essillg the cell proliferation gene for the s.;1~ lg of compounds, the invention also includes methods using soluble or 15 immobilized cell proliferation gene protein. For example, molecules capable of binding to the cell proliferation gene may be identified within a biological or chemicalion. For example, the cell proliferation gene, or functional fr~gmPnts thereof, e.g., fr~gmentc co-,t;.;~ g a specific domain of interest, is immobilized to a solid phase 20 matrix, subsequently a chemical or biological ~ ,ala~ion is contacted with the immobilized cell proliferation gene for an interval sufficient to allow the compound to bind. Any unbound material is then washed away from the solid phase matrix, and the p1~,sence of the compound bound to the solid phase is detected, whereby the compound is identifie~ Suitable means are then employed to elute the binding compound.

2. Small Molecule Di~pl-ccm; t Assay In a specific embodiment of the invention a system has been developed for As.~es~ protein-protein interactions and their inhibition in a cell in vivo, 30 e.g., a fungal, bacterial, ~ n~ n cell, or in vitro. Those systems, referred to as small molecule displ~rrment assays, can be used to screen libraries of small molecules to identify specific cG111pow1ds that disrupt such protein-protein interaction.
The small molecule displacement assay has several advantages over traditional 35 assays used for the identification of small molecule inhibitors. First, if the assay is performed in vivo, each compound must be able to penetrate the cell membrane to carry W O 98t07886 PCTrUS97/14514 out its di~lu~ e activity, and would therefore be preselected for membrane pe~ hility, which is a desirable or even crucial ph~.llacological p.op~.ly. Moreover, the screen has general applicability since it can be used against any protein-protein interaction which can be recapitulated within a cell. Furthermore, the screen is effi~ient because the cells can be gridded out in wells into which compounds are applied, either individually or in pools, and the r~lJu~lel construct can be assayed independently in each well. The assay might consist of a colo,il,l~,h;c output to report the pl~3ence or ~bs~nre of the hll,,lacliOn, which may be performed in vivo or in vitro, or an in vivo cell growth assay.
Generally, in a first step, the protein-protein int~ ;lion is cl~ ed or verified, and in a second step, inhibitors of the i~ l..clion are i~entifiP~3 Assays For The ~dentiScation And Determination Of rr- ~ ~ Protein Inte~D~tial.s. Any method suitable for detecting protein-protein hlleld-,lions may be 15 employed for identifying intracellular proteins that interact with the cell proliferation gene product. Among the traditional methods which may be employed are co-imml~nop~icipi~lion, crosclinking and co-purification through gradients or chromatographic columns of cell Iysates or proteins obtained from cell Iysates to identify 20 proteins in the Iysate that interact with the cell proliferation gene product. For these assays, the cell proliferation gene product used can be a full length gene product, or a truncated peptide. Once isolated, such an interacting protein can be identified and can, in turn, be used, in conjunction with standard techniques, to identify proteins with which it interacts. ~or example, at least a portion of the amino acid sequence of an intracellular protein which interacts with cell proliferation gene product, can be ascertained using techniques well known to those of skill in the art, such as via the Edrnan degradation technique. (See, e.g., Creighton, 1983, "P.oteil.s. Structures and Molecular Principles", W.H. Freeman & Co., N.Y., pp.34-49). The amino acid sequence obtained may be used 3 ~ as a guide for the generation of oligonucleotide I~ lules that can be used to screen for gene sequences enco~1ing such intracellular proteins. Screening may be accomplished, for exarnple, by standard hybridization or PCR techniques. Techniques for the generation - and s~l~e.fing of oligonucleotide mixtures are well-known. (See, e.g, Ausubel, supra., 35 and PCR Protocols: A Guide to Methods and Applic~tion~, 1990, Innis, M. et al., eds.
Ac~demic Press, Inc., New York).

W O ~8~/a~6 PCTAUS97/14514 ~ rlrlition~lly, metho-lc may be employed which result in the ~im-llt~n~ol-e identification of genes which encode the intracellular proteins hll~,ra~ g with the cell proliferation gene product. These methods include, for c~llplc, probing t;~ ion libraries, in a manner similar to the well known technique of antibody probing of Agtl l libraries, using cell proliferation gene protein, or cell proliferation gene derived peptide or fusion protein, e.g., a domain fused to a marker (e.g., an el~yll,c, fluor, l~ sc~
protein, or dye), or an Ig-Fc domain.
One method which detects protein inh.a~;lions in vivo, the two-hybrid system, is10 described in detail for illustration only and not by way of limitation. One version of this system has been described (Chien et al., l991, Proc. Natl. Acad. Sci. USA, 88:9578-9582) and is collull~lcially available from Clontech (Palo Alto, CA).
Briefly, lltili7ing such a system, plasmids are con~ ;led that encode two hybrid15 proteins: one plasmid consists of nucleotides enro-~ing the DNA-binding domain of a transcription activator protein fused to a nucleotide seguence encoding the cellproliferation gene product, or a fragment or fusion protein thereof, and the other plasmid consists of nucleotides encoding the l-~lscli~tion activator protein's activation domain 20 fused to a cDNA encoding an unknown protein or the E~ ulllably illtc.a~,lillg protein of interest (e.g, the pe.lull,ag~n) which has been recombined into this plasmid (or can be a part of a cDNA library). The DNA-binding domain fusion plasmid and the plasmid enco-~ing the presumably interacting protein (or the cDNA library) are transformed into a strain of the yeast Saccharol.lvces cerevisiae that contains a reporter gene (e.g., HBS or lacZ) whose regulatory region contains the lldnscli~lion activator's binding site. Either hybrid protein alone cannot activate transcription of the repolle. gene; the DNA-binding domain hybrid cannot because it does not provide activation function, and the activation domain hybrid cannot because it cannot localize to the activator's binding sites.
3 ~ Interaction of the two hybrid ploteh~s lecon~lilules the functional activator protein and results in e~ ession of the l~pollcr gene, which is ~etected by an assay for the reporter gene product.
The two-hybrid system or related methodology may be used to verify any protein-35 protein interaction identified by the present invention using the p~llulbagen approach, butalso to screen activation domain libraries for proteins that interact with the "bait" gene W O 98/07886 PCTrUS97/14~14 product. By way of e~mple, and not by way of limitation, the cell proliferation gene product may be used as the bait gene product. Total genomic or cDNA sequences are fused to the DNA encoding an activation clorn~in This library and a plasmid encoding a hybrid of a bait cell proliferation gene product gene product fused to the DNA-binding domain are cotran~rc,..,ed into a yeast ~ ,o.L~r strain, and the res~llting transformants are s~ encd for those that express the reporter gene. For example, and not by way oflimitation, a bait cell proliferation gene sequence, such as the open reading frame of the cell proliferation gene product or a domain thereof, is cloned into a vector such that it is 10 l,allslalionslly fused to the DNA encoding the DNA-binding domain of the GAL4protein. These colonies are purified and the library plasmids le~orlsible for reporter gene ~ .,ei.~ion are isolated. DNA sequencing is then used to identify the proteins encoded by the library pl~cmi~
A cDNA library of the cell line from which prcteins that interact with bait cellproliferation gene product are to be detected can be made using methods routinely practiced in the art. According to the particular system described herein, for example, the cDNA fr~gm~ntc can be inserted into a vector such that they are translationally fused 20 to the lrallsc~;ptional activation domain of GAL4. This library can be co-transfected along with the bait cell proliferation g~ ne product gene-GAL4 fusion plasmid into a yeast strain which contains a lacZ gene driven by a promoter which contains GAL4 activation sequence. A cDNA encoded protein, fused to a GAL4 transcriptional activation domain.
that interacts with bait cell proliferation gene product will reconstitute an active GAL4 protein and thereby drive ~A~-es~ion of the HIS3 gene. Colonies which express HIS3 can be ~et~octed by their growth on petri dishes cont~ining semi-solid agar based media lacking histidine. The cDNA can then be purified from these strains, and used toproduce and isolate the bait cell proliferation gene-interacting protein using techniques 3 ~ routinely practiced in the art.
SmaU Molecule D~rlr~c~ t Assay l o Identify l-thi~ Of The Pro~ein-Protein Inte~a_tion. The macromolecules that interact with the cell proliferation gene product are referred to, for purposes of this discussion, as "binding p~ le~ These 35 binding pdll~ltl:~ are likely to be involved in the cell proliferation gene product signal tr~n~ ction pathway, and therefore, in the role of the cell proliferation gene product's W O ~51'~O~6 PCTrUS97/14514 cell activation regulation. Therefore, it is desirable to identify cc,--",-)ullds that h~hlrclc with or disrupt the il-t~aCliOn of such binding pa~ e~ with the cell proliferation gene product which may be useful in regulating the activity of the cell proliferation gene product and thus control cell proliferation disorders ~ori~t~d with the cell proliferation gene product's activity.
The basic principle of tne assay systems used to identify compounds that interfere with the interaction between the cell proliferation gene product and its binding partner or partners involves plcpallng a reaction ~.lixlu,c c~ cell proliferation gene product, 10 polypeptide, peptide or fusion protein, and the binding partner under conditions and for a time sufficient to allow the two to interact and bind, thus forming a complex. In order to test a colllpou.ld for inhibitory activity, the reaction nli~lule is prcp~ed in the presence and absence of the test compound. The test compound Inay be initially 15 included in the reaction mixture~ or may be added at a time subsequent to the addition of the cell proliferation gene product and its binding partner. Control reaction mixtures are in~ b~ted without the test compound or with a placebo. The formation of any compl~xes between the cell proliferation gene product and the binding partner is then detecte~l The 20 formation of a complex in the control reaction, but not in the reaction mixture co.~ i.,e the test compound, inriir~tPs that the compound i~"~lr~.t s with the interaction of the cell proliferation gene product and the interactive binding partner. Additionally, complex formation within reaction mixtures cont~inin~ the test compound and normal cell proliferation gene product may also be compared to complex formation within reaction ixlwes cont~ining the test compound and a mutant cell proliferation gene product.
This co~llpa.;son may be hllpc"~lt in those cases wherein it is desirable to identify cc,l"pollllds that disrupt interactions of mutant but not normal cell proliferation gene products.
The assay for compounds that interfere with the int.,.ac~ion of the cell proliferation gene product and binding partners can be con(lucted in a heterogeneous or homogeneous format. Heterogeneous assays involve anchoring either the cell proliferation gene product or the binding partner onto a solid phase and detPctine 35 complexes anchored on the solid phase at the end of the reaction. In homogeneous assays, the entire reaction is carried out in a liquid phase. In either approach, the order W O 98/07886 PCTnUS97/1~514 of addition of re~ c can be varied to obtain dirf~ information about the compounds being tested. For ex~mrle, test cGll.poullds that hll~,.r~.~ with the hl~e~a.;lion by coll,~clilion can be i~le--tified by conducting the reaction in the ~resellce of the test sub~lce, i.e., by adding the test subst~nce to the reaction mixture prior to or ~im~ usly ~,vith the cell proliferation gene product and i,.l~,.acli~e binding partner.
Alte.llalively, test compounds that disrupt pl~,fo-llled complexes, e.g., compounds with higher binding co~ that displace one of the co,llponc.lls from the complex, can be tested by adding the test co~ )oulld to the reaction llli~ e after complexes have been 10 formed. The various formats are described briefly below.
In a heterogeneous assay system, either the cell proliferation gene product or the illt~a~,live binding partner, is anchored onto a solid surface, while the non-anchored species is labeled. either directly or indirectly. In practice, microtiter plates are 15 conveniently utili7P~l The anchored species may be immobilized by non-covalent or covalent ~ hr..~ c Non-covalent ~ rhm~nt may be accomplished simply by coating the solid surface with a solution of the cell proliferation gene product or binding partner and drying. Alternatively, an immobilized antibody specific for the species to be 20 a~lcholed may be used to anchor the species to the solid surface. The surfaces may be prepared in advance and stored.
In order to conduct the assay, the partner of the immobilized species is exposed to the coated surface with or without the test compound. After the reaction is complete.
unreacted components are removed (e.g., by washing) and any complexes formed will remain immobilized on the solid surface. The detection of complexes anchored on the solid surface can be accomplished in a number of ways. Where the non-immobilizedspecies is pre-labeled, the detection of label immobili~d on the surface indicates that complexes were formed. Where the non-immobilized species is not pre-labeled, an 3 ~ indirect label can be used to detect complexes anchored on the surface; e.g., using a labeled antibody specific for the initially non-immobilized species (the antibody, in turn, may be directly labeled or indirectly labeled with a labeled anti-Ig antibody). Depending upon the order of addition of reaction components, test compounds which inhibit 3 5 complex forrnation or which disrupt preformed complexes can be detecte~

. .

wo 98/07886 Pcr/uS97/14514 Alt,_.,lalivcly, the reaction can be co~ cled in a liquid phase in the ~.esellce or absence of the test col,.polu~d, the reaction products se~ ed from ul--ea~;Led COlll~Ol1f nl~, and cûr,plcAes ~letPcte~l; e.g., using an immobilized antibody specific for 5 one of the binding co~ o~ s to anchor any complexes formed in solution, and a labeled antibody specific for the other partner to detect anchored colllplexes. Again, ~f p" .~.I;.~g upon the order of Q~lition of ree~ ~ n~ to the liquid phase, test co-l-~,oul-ds which inhibit complex or which disrupt preformed colnpll Aes are identified.
In an alternate embodiment of the invention, a homo~,~nf ous assay can be used.
10 In this approach, a preformed complex of the cell proliferation gene product and the i.,lc.~ e binding partner is p,~ ed in which either the cell proliferation gene product or its binding ~,~L"e.~ is labeled. but the signal gc~lc.~lcd by the label is q~lenthPd due to formation of the complex (see, e.g., U.S. Patent No. 4,109,496 by Rubenstein which 15 utilizes this approach for immlmn~cs~ys) The addition of a test s~bstQncP that competes with and displaces one of the species from the p.erc,-l..cd complex will result in the generation of a signal above background. In this way, test Sub:il~lces which disrupt cell proliferation gene product/intracellular binding partner hll~.a.;lion are identifie(l SmanMoleculeD:rl7~mentAssayFollowingIJduc~o Of RevertontsUsing P~,th~ .S. In a particular embodiment of the invention, following i.olation of the p~,.lu bag~ sequence, it is relatively straightforward to define its target in the cell (~c,~ the target is a cellular protein) using yeast two-hybrid analysis. In one formulation of the cApe~in~ent, the perturbagen is fused to the GAL4 DNA bindingdomain and introduced into HlS3-yeast cells. A second fusion construct is also introduced that contains the G~L4 activation domain fused to a random-primed library of cDNA inserts, prcfc,ably constructed using mRNA from the cell originally used to define the p~ bagen. Selection for HIS plùlollul)y (cA,ulessiOn of a HIS3 gene under GAL4 3 ~ control) and lacz ~A~,ression (also under GAL4 control) permits identification of sequences from the library that provide l~con~lilulion of GAL4 llansc-;lJtional activity;
that is, the ples~;.lce of the p~ bagen/DNA binding domain fusion along with the GAL4 activation domain fusion in the same cell results in GAL4 function. This result is 35 nornally obtained when the pcllu~bagen and a sequence from the cDNA library encode proteins that interact, bridging the two halves of the bisected G~L4 factor. Further tests W O 98/07886 PCTrUS97/14514 can be ~.~lllled, if desired, to ensure that the library se.~ e is indeed the binding partner (ie., the target) of the p~ bagen in vivo.
Once the p~,lu.l,agen and its t~arget are identified, it is possible to reconfigure the 5 two-hybrid interaction so that screens for small molecules can be undertaken. Such screens take advantage of the protein-protein interaction b~wt;ell the l,e.lull,agen and its target. They seek out small molecules that are capable of tlicpl~ring the protein-protein h~t~,,a~liOn. Teçhnir~ly, such a screen could be carried out in yeast cells, in m~mm~ n cdls in which the h~ ion has been reco~ l or, perhaps best of all, in a test10 tube. Such a screen is csnfig-l~ed by fusing one of the binding p~lll.,lS (e.g., the p~tull~agen) to a convenient l~l,olt~l molecule such as Green Fluol~scent Protein (GFP).
The other binding partner (e.g., the target) is fused to a second protein that can be absoll,ed onto a solid support via a biotin bridge or an antibody or some other ligand.
15 The h~te.~clion between the perturbagen and its target must be m~ ined in the new fusion setting. The release of GFP fluorcsc~lce signal from the solid support (i.e., into the S~ ~ "s~t~t) is then detected after addition of test compounds. Compounds that are able to ~iicrl~ce the GFP/perlurbagen fusion are c~n~lid~tes for pcllull,agen mimics.
20 Some of these may bind the p~ .l,agen, while others may bind the target. These two classes are readily distinguished by subsequent tests with the perturbagen and the t~rget.
In general, the displ~c~ assay must utilize a reporter construct in the cell that is not too sensitive to Aict~nces or geometries between the two protein partners. It can be applied in numerous different cell systems a few examples are described in the followmg.
Yeast. The traditional two-hybrid system in yeast may be applied in both the GAL4 version and the lexA forrn~ tion. Bartel et al., 1995, Methods Enzymol. 254:241-263; Men~el~ohn et al., 1994, Curr. Opin. Biotechnol. 5:482-486. Both systems take 3 0 advantage of the bipartite nature of yeast ~ sclil tion factors. The DNA binding c~ .onent can be separated from the activation component and each fused to different proteins. If the proteins interact strongly enough with each other, a functionaltranscription factor is recon~titl~te(l and the reporter gene(s) are turned on. In the GAL4 35 version, the ~ O,t~,.S are HIS3 and lacz. These genes are engi~ ed to contain GAL4 binding sites ~ ,~ , in a suitable position to provide activation if and only if an . . , Wo ~ o86 PCT/US97/14514 activation domain is also supplied, either directly or via a protein-protein i~ .dclion.
HIS3-positive cells are selecte~ for in a HIS3-mutant strain ~ g growth selection.
Lacz serves as a cololllllcLIic, reasonably quarllila~ e, independent measure of5 recoll~ led GAL4 activity.
E. Coli In bacteria, several ~ , systems can be envisioned. These might involve Nut sites that function only when the DNA binding colnl)ollellt is fused to the Nut protein anti-tc....i~.A~;..g sequences. Alternatively, the bip~lile nature of the lambda phage re~ ;ssol (cl) could be used in a way similar to the yeast kalls~ tional system.
10 In this case, however, a protein-protein hl~e.a,lion would recGIlslilule a lel)res~r of llallscl;l~tion, not an activator. Thus, when the process is dislu~ttd, transcription ensues.
M. rlirn Cells. In ~" -~ Ali~n cells, a system similar to the yeast two hybrid system is developed, because the lld.ls~l;plion process is relatively similar. An Upstream 15 Activa~or Sequence (UAS) could be positioned U~SI~ ll of a lcl~ulkl gene such as Green Fluolescellt Protein or lacz so that a l~co..~ d protein-protein interaction brought in the domain from the l~,l,o,lel gene results in its ~I,leçsion. Alternatively, the function of an adapter protein is replaced by a two-hybrid protein i~ .aclion.
Similar systems in other fungal, bacterial, or m~mm~ n cells are col~le.ll~lated.

3. Source Of Candidate Test Compounds The test compounds employed for such assays are obtained from any cGll~ tlcial source, including Aldrich (1001 West St. Paul Ave., Milwaukee, WI
53233), Sigma Chemical (P.O. Box 14508, St. Louis, MO 63178), Fluka Chemie AG
(Indu~l~icstrasse 25, CH-9471 Buchs, Switzerland (Fluka Chemical Corp. 980 South 2nd Street, Ronkonkoma, NY 11779)), F~ctm~n Chemical Company, Fine Chemicals (P.O
Box 431, Kingsport, TN 37662), Boehringer l~rmheim GmbH (Sandhofer Strasse } 16,3~ D-68298 I~A~nnheim)~ Takasago (4 Volvo Drive, Rockleigh, NJ 07647), SST Corporation (635 Brighton Road, Clifton, NJ 07012), Ferro (111 West Irene Road, Zachary, LA
70791), Riedel-deHaen Aktiengesellschaft (P.O. Box D-30918, Seelze, Germany), PPG
Industries Inc., Fine Chemicals (One PPG Place, 34th Floor, Piu~burgh, PA 15272).
3 5 Further any kind of natural products may be sc~ ed using the assay cascade of the invention, including microbial, fungal or plant extracts.

wO 98/07886 PCTIUS97/14514 a. Indications For The Use Of Compounds IDterfering With The Cell Proliferation Genes Of The Invention The compounds identified by the methods of the present invention are modulators of a cell proliferation activity in general, or a cell proliferation 5 gene in particular. As such, the compounds produced by the processes and assays of the invention are useful for the ~ f-l~t of disease related to ab~ , uncontrolled orina~propliate cell proliferation.
A large number of disease states involve excess or ~liminich~d cell proliferation.
10 Generally, many of these ~lislo~ces may be treated with DNA sequences, proteins, or small molecules that influence cell proliferation. In some inct~nceS the goal is to stimnl~t~
proliferation; in others, to prevent or inhibit proliferation of cells. The list of dise~cec directly involving cell growth includes. but is not limited to, cancer, psoriasis, 15 infl~mm~tory ~lise~cec, such as rhel-m~toid arthritis, restenosis, immnnological activation or suppression, including tissue rejection, neurodege.lel~ion or expansion of neuronal cells and viral infection.
Accordingly, phann~re~lticRl compositions comprising a therapeutically effectiveamount of a compound identified by the methods of the invention will be useful for the llc~ l"~nt of diseases driven by unregulated or inapp,o~,l;ate cell proliferation, including cancer, such as glioma. melanoma, Kaposi's sarcoma, psoriasis, h~m~ngioma and ovarian, breast, lung, pancreatic, prostate, colon and epidermoid cancer, rheumatoid arthritis, psoriasis, restenosis, immunological activation or suppression, including tissue 25 rejection, neurodegeneration or expansion of neuronal cells.

K. Formulations/Route Of AdminiJt,.-liol-The identified compounds can be ~lminictPred to a human patient alone or in 30 pharrn~relltic~l compositions where they are is mixed with suitable carriers or excipient(s) at therapeutically ~ffective doses to treat or ameliorate a variety of disorders.
A therape~ltic~lly effective dose further refers to that amount of the compound sufficient to result in amelioration of symptoms as determined in a decrease of cell proliferation.
Techniques for forrnulation and a~lmini~tration of the compounds of the instant application may be found in "Remington's Pharmaceutical Sciences," Mack Publishing Co., Easton, PA, latest edition.

, CA 02263744 l999-02-l8 W O 98107886 PCTrUS97/14514 l. Routes Of Administration.
Suitable routes of ~ ninietration may, for example, include oral, rectal, tr~n~m-lcosal, or intestin~ mini~tration; pa~ al delivery, including intr~muccul~r, subcut~n~ous, i~ ..rd~ y injections, as well as intrathecal, direct intraventricular, intravenous, inlla~ oneal, inl~anasal, or intraocular injections Alternately, one may ~rimini~ter a coIllpo~uld of the invention in a local rather than systemic manner, for example, via injection of the compound directly into a solid tumor, often in a depot, or in a ~ ed release formulation.
Ful~he.l~lore, one may a~lmini~t~r the drug via a targeted drug delivery system, for example, in a liposome coated with tumor-specific antibody. The liposomes will be targeted to and taken up selectively by the tumor.

2. Compositioll/Formulation The pharm~ceutiG~l compositions of the present invention may be m~mlfactIlred by means of conventional mixing, dissolving, gr~m-I~ting, dragee-m~king~
levigating, emulsifying, en~psul~ting, elllra,~,~)lhlg or IyophiIi7ing l,~ocesses.
Pharmaceutical compositions for use in accoIdallce with the present invention thus may be formlII~ted in a conventional manner using one or more physiologically acceptable carriers comprising excipients and auxiliaries which facilitate processing of the active compounds into ylcp~lions which can be used pharm~eutically. Proper formulation is dependent upon the route of ~flminictration chosen.
For injection, the agents of the invention may be formulated in aqueous solutions.
preferably in physiologically collIp~lible buffers such as Hanks's solution, Ringer's solution, or physiological saline buffer. For tr~rl~m.-lcos~l a-lminictration, p~n~
appl." liate to the barrier to be permeated are used in the formulation. Such penellallts 30 are generally known in the art.
For oral arlmini~tration, the compounds can be formulated readily by combining the active compounds with pharmaceutically acceptable carriers well known in the art.
Such carriers enable the colllpollllds of the invention to be form~ t~d as tablets, pills, 35 dragees, capsules, liquids, gels, syrups, slurries, suspensions and the like, for oral ingestion by a patient to be treated. Ph~nn~reutical pIe~)~d~ions for oral use can be W 098/07886 PCTrUS97/14514 obtained as a solid excipient, optionally grinding a resulting llliXlU~G, and ploces~il,g the mixture of granules, after adding suitable auxiliaries, if desired, to obtain tablets or dragee cores. Suitable excipients include fillers such as sugars, including lactose, sucrose, mannitol, or sorbitol; cellulose ~ palalions such as, for example, maize starch, wheat starch, rice starch, potato starch, gelatin, gum tra~r~nth, methyl cellulose, hydroxy,u,o~yl,,,cthyl-cellulose, sodium ca.l,.~y",el},ylcellulose, andlor polyvinyll,y"olidone (PVP). If desired, t~ glalillg agents may be added, such as the cross-linked polyvinyl pyrrolidone, agar, or alginic acid or a salt thereof such as sodium 10 ~Igin~tP, Dragee cores are provided with suitable coatings. For this ~ ose, collcGllllaledsugar solutions may be used, which may optionally contain gurn arabic, talc, polyvinyl pyrrolidone, carbopol gel, polyethylene glycol, and/or li~iulll dioxide, lacquer solutions, 15 and suitable organic solvents or solvent nli~ s. Dyestuffs or pigm~ntc may be added to the tablets or dragee co~ting~ for identification or to characterize different combinations of active colllpou-ld doses.
Pharm~eutic~l l,rGp~alions which can be used orally include push-fit capsules 20 made of gelatin, as well as soft, sealed c~ps--les made of gelatin and a plasticizer, such as glycerol or sorbitol. The push-fit capsules can contain the active ingredients in ~rimixtllre with fillers such as lactose, binders such as ~ hes, and/or lubricants such as talc or m~gnlocjum stearate and, optionally, stabilizers. In soft capsules, the active compounds may be dissolved or suspended in suitable liquids. such as fatty oils~ liquid paraffin, or liquid polyethylene glycols. In addition, stabilizers may be added. All formulations for oral ~rlminictration should be in dosages suitable for such .
~rlmlnlctratlon.
For buccal ~lmini~tration,the conlpo~ilions may take the form of tablets or 30 lozenges forrn~ ted in conventional manner.
For ~rlminictration by inhalation, the compounds for use accolding to the present invention are conveniently delivered in the form of an aerosol spray plesenlalion from ized packs or a nebulizer, with the use of a suitable propel}ant, e.g, 35 dichlorodifluoromethane, trichlorofluorom~oth~n~ dichlorotetrafluoroethane, carbon dioxide or other suitable gas. In the case of a pressurized aerosol the dosage unit may be .. .

CA 02263744 1999-02-18 .

W O 98/07886 PCT~US97/14514 d~ t~-...;nPcl by providing a valve to deliver a metered ~molmt C~psl~lps and cartridges of, e.g., gelatin, for use in an inhaler or incllM~tor, may be forrnulated cullli~;nil~g a powder mix of the compound and a suitable powder base such as lactose or starch.The compounds may be formtll~ted for pa~ rlminictration by injection, e.g., by bolus injection or continuous infusion. Formulations for injection may be ,.csented in unit dosage form, e.g., in ~mpolllPs or in multi-dose cont~i~prs~ with an added preservative. The compositions may take such forms as sllcp~pncions~ solutions or emulsions in oily or aqueous vehicles, and may contain form~ tory agents such as10 sucpen~ing stabilizing and/or dispersing agents.
Ph~rm~reutic~l formulations for p~. l~t~ ,iminictration include aqueous solutions of the active compounds in water-soluble form. Additionally, sucpçncions of tne active compounds may be prepared as a,vpro~liate oily injection ~u~ellsions.15 Suitable lipophilic solvents or vehicles include fatty oils such as sesame oil, or synthetic fatty acid esters, such as ethyl oleate or triglycerides, or liposomes. Aqueous injection ~u~ n~ions may contain subst~ncPs which increase the viscosity of the suspension, such as sodium carboxymethyl cellulose, sorbitol, or dextran. Optionally, the sllcpencion may 20 also contain suitable stabilizers or agents which increase the solubility of the compounds to allow for the p.~ lion of highly concentrated solutions.
Alternatively, the active ingredient may be in powder form for constitution with a suitable vehicle, such as sterile pyrogen-free water, before use.
The compounds may also be forrnul~ted in rectal compositions such as suppositories or retention enern~c e.g, cont~ining conventional suppository bases such as cocoa butter or other glycerides.
In addition to the formulations described previously, the compounds may also be formulated as a depot p-e~ lion. Such long acting formulations may be ~-iminict.ored 30 by implantation (for example suhcut~n~ously or intr~mncclll~rly) or by intr~rnnccul~r injection. Thus, for example, the compounds may be forrnl-l~ten with suitable polymeric or hydrophobic materials (for exarnple as an emulsion in an acceptable oil) or ion excll~nge resins, or as sparingly soluble derivatives, for example, as a sparingly soluble 3 5 salt.

CA 02263744 l999-02-l8 W 098/07886 PCT~US97/14514 A pli .~ r~ti~ i carrier for the hydrophobic compounds of the invention is a cosolvent system comprising benzyl alcohol, a nonpolar surfactant, a water-miscible organic polymer, and an aqueous phase.
The cosolvent system may be the VPD co-solvent system. VPD is a solution of 3% w/v benzyl alcohol, 8% w/v of the nonpolar ~ulr~ct~ult polysorbate 80, and 65% w/v polyethylene glycol 300, made up to volume in absolute ethanol. The VPD co-solvent system (VPD:SW) consists of VPD diluted l:l with a 5% dc,~llose in water solution.
This co-solvent system dissolves hydrophobic compounds well, and itself produces low 10 toxicity upon systemic ~l...;t~ .dlion. Naturally, the proportions of a co-solvent system may be varied considerably without destroying its solubility and toxicity ch~a~;le~ ics.
Furthermore, the identity of the co-solvent components may be varied: for example, other low-toxicit~ nonpolar surfactants may be used instead of polysorbate 80; the fraction size 15 of polyethylene glycol may be varied; other biocompatible polymers may replace polyethylene glycol, e.g., polyvinyl pyrrolidone: and other sugars or polysaccharides may be s~lbstit~tçd for dextrose.
Alternatively, other delivery systems for hydrophobic ph~...~r~-ltical compounds2 o may be employed. Liposomes and emulsions are well known examples of deliveryvehicles or carriers for hydrophobic drugs. Certain organic solvents such as dimethylsulfoxide also may be employed. although usually with a greater toxicity.
Additionally, the compounds may be delivered using a sustained-release system, such as se"lip.,ll.lcable matrices of solid hydrophobic polymers cont~ining the therapeutic agent. Various s~-ct~in~od-release materials have been established and are well known by those skilled in the art. Sll~t~ined-release capsules may, dep~n(iin~ on their chemical nature, release the compounds for a few weeks up to over 100 days.
Depen-ling on the chemical nature and the biological stability of the the.apc~ll,c 3 ~ reagent, additional strategies for protein stabilization may be employed.
The pharrn~reutical compositions also may comprise suitable solid cr gel phase carriers or excipients. Examples of such carriers or excipients include but are not limited to calcium c&,l,onate, calciu n phosphate, various sugars, starches, cellulose derivatives, 35 gelatin, and polymers such as polyethylene glycols.

CA 02263744 l999-02-l8 W O 98/07886 PCTrUS97/14514 Many of the eell proliferation inhibiting eolllpoullds of the invention may be provided as salts with ph~rm~reutic~lly colll~alible counterions. Ph~lllac~ulieally colll~alible salts may be formed with many acids, ineluding but not limited to hydrochloric, sulfuric, aeetic, laetie, tartaric, malie, suecinic, ete. Salts tend to be more soluble in aqueous or other protonic solvents that are the coll~ onding free base forms.

3. Effective Dosage.
Pharm~r~e~tic~l eo,llposilions suitable for use in the present 10 invention ine}ude compositions wherein the aetive ingredients are colll~ined in an effective amount to achieve its int~nr~d purpose. More specifieally, a the. -~ ;r~lly effective amount means an amount effeetive to prevent development of or to alleviate the existing symptoms of the subject being treated. Determination of the effeetive amounts 15 is well within the eapability of those skilled in the art, especially in light of the ~let~iled disclosure provided herein.
For any compound used in the method of the invention, the the.à~ ulically effeetive dose can be çstim~tçd initially from cell culture assays. For example, a dose 2 0 can be form~ t~d in animal models to achieve a cireul~ting concentration range that ineludes the IC50 as determined in cell culture (i.e., the conc~.llldLion of the test compound which achieves a half-maximal inhibition of the cell proliferation activity).
Sueh information ean be used to more accurately determine useful doses in hl-m~nc A the.al)~ulically effective dose refers to that amount of the compound that results in amelioration of symptoms or a prolongation of survival in a patient. Toxicity and th~la~ lic effieaey of sueh eompounds ean be determined by standard ph~rm~celltir:ll proeedures in eell eultures or e~p~ ental ~nim~lc, e.g., for determining the LD50 (the dose lethal to 50% of the population) and the ED50 (the dose the.ap~.llieally effective in - 30 50% of the population). The dose ratio belwt;en toxic and thelap~.lLic effects is the Ih~a~ lic index and it can be e~l,lessed as the ratio between LD50 and ED50.
Compounds which exhibit high therapeutic indices are pre~led.
The data obtained from these cell eulture assays and animal studies can be used in 35 form~ ting a range of dosage for use in human. The dosage of such compounds lies preferably within a range of cireul~ting concentrations that inelude the ED50 with little or wo 98/07886 PCT/USg7/14514 no toxicity. The dosage may vary within this range ~lepen~ling upon the dosage form employed and the route of ~ ;cllalion utili7p~ The exact fonn~ tion, route of a~lminictration and dosage can be chosen by the individual physician in view of the patient's condition. (See, e.g., Fingl e~ al., 1975, in "The Ph~rrn~rological Basis of The,dl)culics'', Ch. I pl).
Dosage amount and interval may be adjusted individually to provide plasma levelsof the active moiety which are sufficient to ~ the kinase mod~ ting effects, or minim~l effective concentration (MEC). The MEC will vary for each compound but can 10 be estim~ted from in vitro data; e.g., the concelllralion npces~ly to achieve 50-90%
inhibition of the kinase using the assays described herein. Dosages nrce~ y to achieve the MEC will depend on individual characteristics and route of ~ministration. However, HPLC assays or bioassays can be ~Ised to determine plasma conce~ dlions.
Dosage intervals can also be determined using MEC value. Compounds should be ~innini~tçred using a regimen which m~int~inc plasma levels above the MEC for 10-90% of the time, preferably between 30-90% and most preferably between 50-90%. In cases of local arlminictration or selective uptake, the effective local conce~ dtion of the 20 drug may not be related to plasma conc~ alion.
The amount of composition ~lnnini~t~red will, of course, be dependent on the subject being treated, on the subject's weight, the severity of the affliction, the manner of a~mini.ctration and the jur1~mPnt of the prescribing physician.

4. P~ in~
The compositions may, if desired, be presented in a pack or tli~pen~er device which may contain one or more unit dosage forms co~ in;~lg the active ingredient. The pack may for example comprise metal or plastic foil, such as a blister 3 ~ pack. The pack or dispe.lsel device may be accompanied by instructions for mini~tration. Compositions comprising a compound of t~he invention formulated in a comp~tible pharm~eu1ical carrier may also be pl~dled, placed in an app,o~lia~e c~mt~iner, and labelled for tre~tmPnt of an in~lic~ted condition. Suitable conditions 35 inrijc~te~i on the label may include inhibition of cell proliferation, tre~tmlont of a tumor, tre~tmPnt of arthritis, and the like.

. . .

wo ~ 6 PCT/US97/14514 The following examples for the generation and use of the selection systems of the invention are given to enable those skilled in the art to more clearly understand and to practice the present invention. The present invention, however, is not limited in scope by the exemplified embodiments, which are inten~ed as illustrations of single aspects of the invention on}y, and methods which are functionally equivalent are within the scope of the invention. Indeed, various modifications of the invention in addition to those described herein will become appalent to those skilled in the art from the foregoing description and accompanying drawings. Such modifications are intended to fall within 10 the scope of the appended claims.

VII. EXAMPLES
A. Example 1: Arrest Of Melanoma Cells By Expression Of P16 In this example. the generation of a growth-arrested melanoma cell line is described as a selection system of the present invention. The obtained growth-arrested melanoma cell line may be used for the selection and isolation of growth-proficient revertants. Analysis of these revertants may result in the identification and isolation of 2 0 novel cell proliferation genes related to the development of diseases related to unregulated or inappropriate cell proliferation, for exarnple, cancer.
The melanoma cell line HS294T, which lacks endogenous pl6. was used to create a cell that could be forced into Go/G~ arrest by introduction of the inducible pl6 expression construct pOPRSVI.pl6 (FIGURES 3A and 3B) into the cells. The promotor of the inducible pl6 construct contains sequences from the Rous Sarcoma Virus (RSV) long terminal repeat (LTR) that act as a potent transcriptional initiator located upstream of the complete pl6 coding sequence. Between the pl6 translational start site and the RSV LTR are operator sequences derived from the E. coli lac operon. These sequences 3 ~ are sufficient for binding of the lac repressor. In the presence of functional lac repressor, transcription from the RSV LTR is dramatically reduced by the lac operator sequences.
However, when IPTG is added to the culture media, the lac repressor molecules are prevented from blocking the transcription of pl6; pl6 mRNA is synthesized andpl63 5 protein is produced.

SUBSTITUTE SHEET (RULE 26) W 098/07886 PCTAUS97/r~514 As a colls~ucnce, the cells, termed HS294T/pl6+, respond to IPTG by induction of pl6 and cell cycle arrest. Death of arrested HS294T/pl6 cells after addition of IPTG
occurred over a period of several days during the second week of arrest. By day fifteen 5 (15), no viable cells were present and the vast majority of the adherent cells had disappc~d from the bottom of the culture dish.

B. Esample 2: Selection Of Gr~ .. lh F'n,rt ~ent Revertants In this example, the selection of growth-proficient g~.le.ated as revertants 10 derived from the growth suy~ 3sed H2594T/pl6+ melanoma cells ~elle.dted as described in Example 1, supra, is described. Further analysis of the rev~,.~l~s will reveal the identity of cell proliferation genes useful for the diagnosis and prognosis of ~ e~ees related to uncontrolled or ina~,.o~,;ate cell proliferation, and for the development of 15 targeted drugs for the Ll~n~ el-t of disease related to uncontrolled cell proliferation.
To select revertants from the population of plharrested cells, HS294Tlpl6~ cellswere plated in microtiter wells at a density of 2000 cells/well in the p~e~ ce of IPTG.
As a control, parental HS294T cells that con~ ue to grow in the presence of IPTG were 20 seeded at different densities among arrested HS294T/pl6 cells in a s~udle set of microtiter wells. As expected, these wells gave rise to growing clones of cells that spread over the well bottom.
By day twenty (20) after plating, 11/96 microtiter wells clearly ct~nt~inPcl growing cells. ~snming that a single progenitor cell spawned the colony in each of the eleven wells, this implies a reversion rate of ap~roximately one per 20,000 arrested cells (11/96(2000)).
Materials And Method$ The melanoma cell line HS294T was eng;l-P~ ~ed to contain an IPTG-inducible pl6 gene in the PopRSV vector (str~t~g~np~ San Diego, CA) 30 as described by Stone e~ al, 1996, Cancer Res., in press. The res~ ing cell line, HS294T/pl6, was arrested by addition of O.lmM IPTG after s~lhcultl~ring 2,000 cells per well of a 96-well culture plate (Falcon). Fresh medium (DMEM, non~ssPnti~l aminoacids, gl~ e (2 mM), sodium pyruvate (100 mg/ml), hygromycin (30 llg,ml), 35 geneticin (34 ~g/ml), IPTG (0.1 mM) was added every four (4) to five (5) days. After twenty (20) days. eleven (11) wells were judged to contain growing cells. six (6) of Wo 98/07886 PCT/US97/14514 which survived subcloning. These ~ growth-proficient cells were grown up inflasks for subsequent analysis in the ~bsen~e of IPTG.
Cells from the eleven positive wells were l.~,s~ ,d to larger culture dishes. The cells were allowed to grow in the absen~e of IPTG to mitigate against toxicity of the IPTG which over long periods of exposure impairs HS294T viability. The value of this p,~,c~ ion was co"r.. ed by subcloning the parental HS294T control cells in the esence of IPTG, a process that killed the cells. Despite IPTG withdrawal, only six e.~t lines survived the transfer procedure. These lines were grown up and 10 characterized in several ways.

C. Example 3: Ch~r_~t- ~lion Of The Revertants To ensure that the rev~ cell lines of Exarnple 2 were still resis~ t to 15 IPTG-inrlllred pl6, the lines were re~wlled to IPTG-cont~inine media. In collL,a-al to the original HS294T/pl6+ cells which enter GJG, arrest within twenty four (24) hours, all six revertant cell lines contin~le~l to grow. The pe.c~ll~ges of cells in Go/GI in the pl~,a~nce and absence of IPTG were measured and co",yalcd to the distributions in 2 0 various control cell lines. The r~ t lines had largely similar growth profiles to the parental lines. The rev4 and rev6 lines al)pealed tc have slightly lower G,tG2 ratios indicating more significant changes to the cellular signal transduction (TABLE III) co"~ d to the parentel H5294T/pl6+ line. The revl line appealed to possess some residual pl6 sensitivity based on its partial arrest in response to IPTG.

TABLE III
Line G,/G7 (-IPTG) G~/G, (+IPTG) POP 2.6 2.7 POP/pl 6 3.5 45.0 revl 3.5 6.6 rev2 2.2 3.1 rev3 3.6 4.0 rev4 1.9 1.2 rev5 3.8 4.4 rev6 1.7 1.8 The exl-~es~.on status of the pl6, Rb, and CDK4 ~ene products was ex~min~d in the revertant line by western blot analysis. See, FIGURES 4 and 5. Four of the six lines had lost ex~iession of the inducible pl6 construct. A fifth line had no detect~ble Rb protein. while a sixth line, rev6, appeared to have the expected levels of pl6, Rb, CDK4 and cyclinD1.
2 ~ Flow Cytomet~. Revertant and control cell lines were grown to about 70%
confluency and treated with 0.1 mM IPTG for twenty four (24) hours. The cells were imme~ tely harvested, fixed in ethanoh and stained with propidium iodide prior to analysis on a FACscan flow cytometer (Becton-Dickinson). F.etim~t~s of cells in Gl and 25 G2 were made by fitting Gallcci~n curves to the fluorescence data and integrating the curves using the program (Modfit; Verity House Software).
Western Blot$ The revertant and control cell lines were treated with 0.1 mM
IPTG for twenty four (24) hours prior to making total cell Iysates. Ix107 cells were washed and resuspended in Iysate buffer (0.1 M NaCI, 0.01 M TrisCI pH 7.6, 1 mM
EDTA pH 8.0), boiled, and frozen at -80~C. Approximately equal amounts of thawedtotal protein were run on SDS polyacrylamide gels and transferred using the semi-dry method (Hoeffer) onto nitrocellulose membranes. Blocking and antibody treatment of the blots ~as according to standard procedures (BioRad). Primary antibodies were35 obtained from various sources: anti-pl6, anti-CDK4, and anti-cyclin-Dl where obtained from the ICRF (London, UK); anti-RB was obtained from Santa Cruz Biotechnology W O 98/07886 PCT~US97/14514 (Santa Cruz, California). The blotted l,loteills were vi~u~li7ed using ~lk~lin-- pho~yh~ e (BioRad).

D. Esample 4: Determination Of The General Efficacy Of A Screen For Perturbagen Molecules In the following example, the phcloll.olle le~,onse pathway of the budding yeast Saccharomyces cerevisiae was employed to ~termine the general efficacy of a screen for ~e.Lul~agen molecules.
By way of background, haploid yeast responds to pheromones see.c:led by cells ofthe opposite mating type in a variety of ways in ple~d~ion for mating and diploid formation. For reviews, see, Sprague and Thorner, 1992, The Molecular Biology of the Yeast Saccharomyces cerevisiae: Gene Expression. Broach, J., and Pringle J.R. (eds), 15 Cold Spring Harbor, New York: Cold Spring Harbor Laboldlc,ly Press, pp. 657-744; and Kurjan, 1992,Annu. Rev. Biochem. 61:1097-1129. TheseresponsesincludeG,-phase cell cycle arrest and changes in cell morphology and cell wall composition. The G,-phase cell cycle arrest re~onse can be exploited to find yeast h~.lJo~ g mutations which inhibit the pheromone res~,onse process since escape from cell cycle arrest results in growth and hence colony formation. For several reasons, this system is attractive for testing p~.~u,l,agen libraries as "mutagenic" agents. First, yeast grow rapidly and are easily transformed. Second, due to the extensive study of this pathway, manv of the genes encoding proteins involved in re~onse to pheromones have been identified and 25 characterized. Finally, the complete yeast genome has been sequenced. For all these reasons, the yeast system is prone for the rapid identification of the targets of agens, and a determination of whether the p~.lu,l,agens themselves are derived from proteins involved in the response. Because a wide variety of genetic strategies have 3 0 been applied to the study of this pathway (Mackay and Marmey, 1974, Genetics 76:255-271; Mackay and Manney, 1974, Genetics 76:273-288; Hartwell, 1980, Journ. Cell Biol.
85:811-822; Dietzal and Kurjan, 1987, Cell 50:1001-1010; Blinder et al., 1989, Cell 56:479-486; Stevenson, 1992, Genes and Dev. 6:1293-1304; and Ramer et al., 1992,Proc. Natl. Acad. Sci. U.S.A. 89: 11589-11593), the identification of heretoforemimrlit ~ted genes would indicate that this strategy complements other types of genetic oaches in "genetic" systems.

wo ~~~ ~6 Pcr/usg7/l45l4 A large-scale screen was carried out for both random peptides and fr~gmPntc of yeast genomic DNA that cause escape from a-factor-in~ ced cell cycle arrest. Fourteen different rr~gJ~ tc of yeast genomic DNA and two randomly generated peptides were identified which, when e,.~,essed, promoted escape from cell cycle arrest. Of the fourteen (14) genomic fragmPntc, nine (9) are predicted to encode portions of yeast proteins, inrh~ ng portions of the STE11 and STES0 proteins, two genes involved in the pheromone-rt, I,onse p~LL~ y. Hartwell, 1980, J. Cell. Biol. 85:811-822; and Rad et al., 1992, Mol. Gen. Genet. 236:145-154. The r~ g five (5) fr~m~nt~ are predicted to 10 express relatively short peptides not found in any known or predicted yeast coding sequence. Thus, genetic screens employing pc~ ag~"l libraries replcse.,l an effective means for identifying genes involved in illlpol~ll cellular pIocesses. In addition, for pathways relevant to human tli~ .os such as cancer, pc.iu.l.agen-based genetic methods 15 may lead to novel thc~ ic agents and targets.
Strains And Media. The Saccharomyces cerevisiae strains used in the screen for a-factor-resistant colonies was yVT12 (MATa, ura3-1, leu2-3, 112, Iys2, sst2~, hmla, hmra, mfal~::hisG, mfa2::hisG, ade2-1, STE::GAL1-STE3::HIS3, strain JRY5312.
20 Boyartchuk e~ al., 1997, Science 275:1796-1800. Yeast strains were transfo~ned by the method of Gietz and Schiestle (Gietz and Schiestl, 1995, Methods in Molecular and Cellular Biology 5:255-269f) and plasmids were m~ ed by growth in standard media. Isolation of plasmids from yeast was accomplished by harvesting cells from 2 ml overnight cultures by centrifugation. discarding the Sll~ , and resuspending cells in 200 ~l of extraction buffer (2% Triton; 1% SDS; 100 mM NaCl; 10 mM Tris, pH 8; and 1 mM EDTA). 200 Ill of phenol:chloroform (1:1) equilibrated with TE (100 mM Tris, pH 8; and 10 mM EDTA) and a small volume of 425-600 micron acid-washed glass beads (Sigma) were then added and the mixture vortexed for one (1) to two (2) Ininut~os, 30 Organic and aqueous phases were sepa,a~d by centrifugation and the aqueous phase removed ~nd extracted with phenol. 1 1ll of the aqueous phase was used to transform DH5-a E. coli cells by ele~;l,opol~ion.

- 7g -, _ .. . .

WO 3~ PCT/USg7/14514 1. Construction Of Peptide And Genomic Fragment Libraries A t~,vo-step approach was used to establish the efficacy of a broad genetic search for inhibitory peptides and gene fragm~ntc The first step was the creation of two eA~ie~ion libraries, one comprised of randomly gcn~ ed short peptides, displayed within the context of a larger protein scaffold (the green fluo,escent protein, GFP), and the second compri~ed of small r~ t~ of the yeast genvlllc ~;AI,lcssed as carboxy-terrninal fusions with GFP. Prasher et aL, 1992, Gene 111:229 233. The second step was to cull from these libraries clones that were able to confer resict~n~ e to 10 a-factor-in~luced cell cycle arrest in haploid yeast a cells. For review, see, Sprague, G.F., Jr., and Thorner, J. (1992), The Molecular Biology of the Yeast Saccharomyces cerevisiae: Gene Expression, Broach, J., and Pringle, J.R. (eds), Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press, pp. 657-744; and Kurjan, 1992, 1 5 supra.
The first library p~ led the ~;A~leJ~ion and ~l~sc;,l~lion of randomly generatedfifteen (15) amino acid peptides from within a red-shifted variant of GFP. Heim et al., 1995, Nature 373:663-664. This library conlained approximately 6.5x 106 individual 20 clones, roughly 40% of which encoded properly folded, full-length GFP molecules as judged by fluorescence sc~nning (FIGURE 6A).
The second library consisted of size-selected DpnII fr~gmentc (100-2500 base pairs in length) generated from yeast genomic DNA inserted into the BglII site of the vector pVT21 (FIGURE 6B). This library containcd ~I,loxilnately 6.5x 105 different clones with an average insert size of 440 base pairs, eAplessed as C-terminal protein fusions with GFP. The rationale behind cA~.ressing the genomic fragments as fusion proteins with GFP rather than ~A~ g them alone was to ill~,lease ~Apres~ion levels by stabilizing both the mRNA (from early-nonsense-codon mP~ t~d mRNA decay (Losson 30 and Lacroute, 1979, Proc. Natl. Acad. Sci. U.S.A. 76:5134-5137) and the protein. Both the peptide and yeast genomic fragment libraries were under the tl~lsc,i~lional control of the conditional GAL1 UAS which allowed for high eA~.re.,~ion of library clones in the plesence of galactose and l,d.~s~ lional repression of library clones in the presence of 35 glucose. For review, see, Johnston! 1987, Microbiol. Rev. 51:458-476. The peptide and W O 98N7886 PCT~US97/l~S14 genomic fragment libraries were introduced into yeast strain yVT12 by s~ndald techniques ge~laling yeast libraries of 3x 10' and 1x 10' Lldllsf."..-- lt~ a~e.,li~rely.
Me~hods. The constructs employed for the generation of the peptide display 5 library are depicted in FIGURE 6A. In brief, 45 residue oligonucleotides of the col,lpo~ilion (NNG/T/C)15 were h~se.led into pVT21 using Xhol and BamHI restriction enzyme sites which had previously been ~lgin~ered into the green fluorcsce.ll protein (GFP) at nucleotide position 468. The yeast genomic fragment library was co~ cl~d by digesting genomic DNA from strain yVT5 (MATa, leu2-3, 112, trpl-l, ura3-1, his3-10 I l, 15, ade2-1, canl-100 [strain JRY2334] with DpnII (New Fngl~n-l Biolabs, Beverly, MA), isolating digested DNA 100-500 base pairs in length from a 1% agarose gel using the gene clean 111 kit (BIO 101), and ligating the purified DNA to pVT21 that had previously been digested with BglII (New Fngl~n-l Biolabs, Beverly, MA), treated with 15 calf intPstin~l phosph~t~se (New Fngl~nd Biolabs, Beverly, MA), and purified.Following ligation, DNA was introduced into E. coli strain DHS-~ by ele-;L,opoldlion, and the resulting amplified library purified using the Qiagen "maxi-prep" kit (Quiagen.
San Diego, CA).
The library size was ectim~ted via serial dilutions of the primary E. coli transformations. The average size of library inserts was determined by isolating DNA
from twenty (20) individual library colonies and determining insert sizes by restriction digest analysis.

2. Identification Of Library Clones That Confer cr-Factor Resistance To identify library clones that promoted recict~nce to a-factor-ind~lced cell cycle arrest, a primary screen for a-factor-resistant colonies wac carried out.
30 Aliquots co..L~ g on average about two (2) to three (3) yeast cells of each primary transformant derived from both the peptide and genomic fiagll.~ libraries were grown for six (6) hours in rich media co~ ing galactose and r~mnose as carbon sources (YEPGR) in order to induce e,.~ieasion of the library pl~cmi~c Following this induction, cells were spread onto 150 x 15 mm YEPGR plates (lx 106 cells per plate) co.~ ing lx 10 8 M a-factor, which is the minim~l concel.l.dlion of ~-factor required to arrest yVT12 cells at this plating density. See, infra. 1750 a-factor-resistant colonies W O ~ o~6 PCTrUS97/14514 derived from the peptide library and 520 colonies derived from the genomic-fragment library were picked from the plates over the course of four days following the initial plating, after which time a-factor escapels became in~3ictin~ujshable from the background growth of cells. Pheromone-resistant colonies were ~ re~led to selective plates cr...lA;.~ g glucose in order to ~..~;..l~;n the library pl~mj~le and repress ~ .sc-;ption of the library clones.
Resistance to a-factor-in~ red cell cycle arrest could either arise from ~Aylession of a library plasmid or from acquisition of a chromosom~l mutation. To distinguish 10 between these possibilities, a secon~l~ry screen was performed. This secondary screen took advantage of the fact that ~A~Jlei,~ion of library clones was depenll~nt on the presence of ~ tose, and therefore those yeast cells which esc~pe~l arrest due toeA~uiession of a library clone would grow in the ~l~se,lce of ~-factor in a galactose-15 dependent manner.
Colonies initially transferred to selective media co..lz~ g glucose as a carbonsource were replica-plated to selective plates co.-t~;..i.-g either glucose or ~ tose and r~ffinose as carbon sources. After twenty four (24) hours of growth following replica 20 plating, cells grown on selective glucose plates were replica-blocked to YEPD, and cells grown on selective galactose/r~ffinose plates to YEPGR plates, each co~ g lo-6 M a-factor (a higher concentration of a-factor was required due to the increased cell density, see, infra). Twenty one (2l) of the 1750 colonies isolated from the peptide library, and 85 of the 520 colonies derived from the genomic fragment library escaped a-factor-induced cell cycle arrest exclusively on plates cont~ining galactose. The number of different peptide library plasmids was further reduced to fourteen (14) and the number of genomic fragment plasmids to nin-oteen (l9) by grouping the clones into classes based on their insert sequences. See, TABLE IV.
3 0 Titration Of Critical o~-Factor Levels. Minimal a-factor levels sufficient to arrest strain yVTl2 at various cell densities were del~l.llhlcd by plating various numbers of cells on YEPD (yeast extract, peptone, dextrose) and YEPGR (yep, galactose and ~.r~.~ose) plates that contained concellllations of a-factor (Sigma) ranging from 10-6 M
35 to 10-1~ M. Growth of cells on plates cont~ining a-factor was then co,~ d to growth on plates lacking a-factor. Minimal concentrations of a-factor required to arrest cells at W O ~e~ O~6 PCTrUS97/14514 densities of l-Sx lo2 and lx 105 cells on 100 x 15 mm plates were 5x 10'~~ M and lx 10-8 M, ~e~pc~ rely. For Ix l06 cells plated on a 150 x 15 mm plate, 1-5x 10 8 M a-factor was ~ i.ed and conce~ alions between 10-7 M and 10~ M were ie.~ ed to arrest 5 thicker patches ~ r~ d by replica plating. Min~mal concentrations of a-factor re~uired to alTest cells were the sarne on ~oth YEPGR and YEPD plates.

TABLE IV

Slr~ h of Cell Cycle Arrest Escape Phenotypes Number of Colonies Pt~ g of Cells LlbraryPer Plsteb F.-c . ~ g from Plasmid aFactor Arrest YEPGR
-~F +aF -~F +~FDe~trose G~ tose pVT21 204 0 100 ~ < 005 <.01 1 242 1 158 155 .004 98 2 432 1 262 158 .002 60 3 302 0 116 62 <.003 53 4 231 0 240 47 <.004 20 420 0 376 23 <.002 6 6 400 0 240 146 <.003 61 7 386 0 64 29 <.003 45 2 5 8 412 0 382 42 <.002 11 9 500 0 376 30 <.002 8 366 0 ND 34 <.003 ND
11 696 0 404 57 <.001 14 3 0 12 936 4 449 47 .009 10 13 440 0 444 2~8 <.002 S4 14 696 0 227 77 < 001 34 a. The numbers 1-14 refer to each of the founeen (14)p~ u~ pJasrnids, pVT21 is the parental vec~or for the library.
3 5 Colonies counts were p~.r~).. ed five (5) days after the initial platin~.

W O~ /o86 PCTrUS97/14S14 3. Plasmid Linkage Analysis To firmly establish whether library plasmids identifiPd in the seco,ldaly screen were l~;,pollsible for the obsel~ed le~ e to pheromone-ind~re(l cell 5 cycle arrest, linkage be.~.~ell the library plasmids and ~-factor arrest was tested. The 33 dirr.,.el,L library plasmids were isolated from yeast, ~acs~gP~ through baclelia, and reintroduced into strain yVT12. Following leu,L,ocluclion, the ability of each of these p!~cmi~c to confer l~s;~ re to phelu.l,one was tested by comparmg the growth of roughly lx 10~ cells of each l,al~rull"ant on YEPGR plates (plus or minus 10-8 ~-factor) 10 to that of yVT12 ll~rolllled with the parental plasmid pVT21. Of the 33 pl~smirls eY~min~d, two (2) of the foul~,en (14) peptide library plasmids and fourteen (14) of the n;~ e~l- (19) genomic fragment plasmids, co,~.l~d l~ nre to cell cycle arrest caused by ~-factor.
4. Sequence Of Library Clones Sequenres of the insens of each of the fourteen (14) plasmids cont~ining genomic DNA were determined by first se~uPnring the 5' and 3' ends of20 each insert and then COIIlpalillg these sequenres to the complete DNA seq--Pnre of the Saccharomyces cerevisiae genome. See, infra. Based on their sequences, the fourteen (14) clones can be divided into two general categories (TABLE V). The first group was colll~lised of five (5) members (Plamid Numbers 1 through 5) which contained insert DNA that did not create translational fusions between GFP and the coding regions of any known or hypothetical yeast open reading frames (ORFs). Rather, they were predicted to give rise to short peptides (21-5g amino acids in length, see, TABLE V) ap~,ended to the C-terminus of GFP. In COllllaSL, the second group (Plasmid Numbers 6 through 14) encoded portions of nine (9) dirr.,.ent ORFs fused in frarne to GFP (see, 3 ~ TABLE V). Thus, belwt;~;.- the peptide and genomic fragment libraries, seven (7) random peptides and nine (9) gene fragmPntc were i~lentifiPd that, when expressed, co,~l,~d recict~nt~e to ~-factor-in~hlred cell cycle arrest.
It should be noted that two of the library clones (Plasmid Numbers 7 and 9, see,35 TABLE V) encode portionc of the STE11 and STES0 genes, lespecLi~ely, two (2) genes previously known to be involved in the pathway. Hartwell, lg80, J. Cell. Biol. 85:811-W 098/07886 PCTrUS97/14514 822; Rhodes et al., 1990, Genes and Dev. _:1862-1874; Rad et al., 1992, Mol. Gen.
Genet. 236:145-154; and Xu et al.,1996, Molec. Microbiol. 20:773-783. Indeed, u~.,.c~l res~ion of either the N-terminal half of the protein enro~ed by the STEl l gene 5 or a C ~ 1 truncation allele of STE50 (ste50-2), both of which are similar to the regions o~leAylei,sed in these two library clones, have been previously reported to dcclease sensitivity to ~h~lolllolle to varying degrees (Ste~,ensoll, 1992, Genes and Dev 6:1293-1304; and Rad et al., supra. Thus, one class of ~llulbagen molecules that can be i~ "ir~ in such broad screens are pollions of pro~ci~s that are themselves directly 10 involved in the process under study. F~ ion of the roles played by the ~I'O~c~ll5 en~oded by six (6) of the seven (7) rem~ining ORFs (Nuull~ls 6, 8, and 10-14) in the pheromone-response pathway, and the identification of the targets inhibited by all the pelLull,agerls, may resolve how often pcllulllagens themselves are portions of pro~eil s 15 involved in the process under study, and how wide the po~ell~ial range of pelLull,agen targets is.
PCR Arnplification And Seqv~n~ing Of Library Clone DNA. Whole-colony PCR was performed by ~ Ç~l,ing yeast cells from single colonies to PCR vessels, 20 Illicrowaving the cells for one minute at full power, and imm~ tely cooling the cells on ice. After cooling, PCR reactions were performed using standard reagents and protocols. Ausubel et al., (eds) Current Protocols in Molecular Biology, John Wiley and Sons, New York (1996). Primers used to amplify the genomic inserts were oVT201 (5'-ATT TTA GCG TAA AGG ATG GGG-3'), which is homologous to a region within the PGK1 3' untr~n.~ ed region (3'UTR), and oVT326 (S'-TGA GAA
TTC GGA TCC AAG AGA GAC CAC ATG GTC C-3'), part of which is homologous to a region within the GFP codin~ region. Sequencing of the 5' and 3' ends of genomic inserts present in both PCR-amplified products and plasmid DNA was accomplished 30 with primers oVT326 and oVT201, and seqoen~e data was ol~ ed using an ABI373A
DNA sequenrer (Applied Biosystems Division, Perkin-Elmer, Inc., Foster City, CA).

TABLE V
Library Plasmid Sequence Infolmation Plasmid Ch,~ . ~ SPTlrnr~ GFP Fusion Partnerb 5Number Number C~.~
12 34S,284-345,535 59 amino acid peptide 2 2 17,605-17,785 21 amino acid peptide 3 14/5 409,769409,846/ 48 amino acid peptide' 81,490-81,418 4 2 390,347-390,624 20 amino acid peptide 7 954,846-955,084 31 amino acid peptide 6 5 408,993408,253 arnino acids 23-269 of YER124c 7 12 849,840-850,463 arLuno acids 14-221 of STEll 8 2 600,538-600,774 amino acids 11-89 of YBR186w 9 3 63,438-64,244 amino acids 32-279 of STE50 425,915424,661 amino acids 1512-1753 of YER132c 2 011 2 329,957-329,565 amino acids 32-160 of GIPl 12 7 854,410-854,195 amino acids 161-231 of YGR179c 13 2 357,343-355,292 amino acids 934-1108 of YBR059c 14 13 441,164443,186 arnino acids 653-960 of YMR086w a. Shown are the il_ol positions of the first and las~ P~ C Of each DNA fragment on their respective cL.l b. Shown are the predicted GFP-fused translation products encoded by each fr~gm.~nt c. The ".~dicled peptide is encoded by tandemly ligaled genomic 3 ~ r ~ -- r.l~ from ORF YNL116w and RAD23.

5. P~. lu- L~g~.. ~.~elral-c~
The pc,.u~l,agens isolated through the e~c"-l,c.l~ described above 3 5 must compete with the wild type function of some cellular component to overcome arrest. Thus, the question arises as to how effective individual p~l~u~l~agens are in . sshlg the response to ~-factor. The term "pcn~llance" is used here to describe CA 02263744 l999-02-l8 W O 98107886 PCTrUS97/14514 the ~llc~ of the p."lull~agens in a simple colony ro"..~l;r,n assay. To d~t~ P the nce of each p.,~lulbagen plasmid, yeast hall,u.ing either one of each of the fou-l~en (14) pe.lu,l,agen plasmids or the parental vector pVT21 were grown overnight to mid-log phase in selective liquid media co~ either glucose or galactose/r~rfil~se as a carbon source. 250 ~1 of ~ tionc cont~ining 1000 cells/ml of each o~.,..l,gllL culture were then plated onto either YEPGR or YEPD media (dc~,e~ g on the carbon source present in the media in which they were grown) that either cor t~in~od or lacked 5x 10-1~ M a-factor (the lowest co~e~ ation of a-factor ll~cess~ y 10 to arrest strain yVT12 at this cell density). Colonies on the various plates were counted after five days and the fractions of the total nurnber of cells plated in the p.~,sence of a-factor able to form colonies for each plasnlid were der~ d The results of this analysis clearly reveal difr~.~nccs in the ~ nglh of the p.,.~ulbagens. Some have 15 pe.lctl~nce of 100% in the assay; o~ers are less than 10% penetrant. The basis for these dirr~ pcn~ll~ces is not clear. It may involve dirr~,.e--ces in Ki's among the various pc.~ullagens, dirr~,ences in their e~ ;,sion levels, and/or dirÇ~.~nces among their targets.

E. Exalnple 5: Selection Systerns Based On Expl.- ;(!n 0~ The R~t~qob~t~. la Gene ~udu~l In analogy to the pl6-arrest eAl~c~illlents~ the rb gene may be eAI,rcssed in tumor cells to select for the ide.-lirr~tion of novel cell proliferation genes. The so 25 obtained selection systems may be used for the selection of random revertants or for the isolation of revertants obtained upon inrh1ction with pe.~ull,agenS. See, supra.
Re~,~.~llL~ of rb-arrested cells are expected contain alterations in a set of genes that overlaps considerably with the pl6-arrested revertants because rb acts dowl~llcalll in the 30 same signal tr~ncdllrtion pathway as pl6.
Further analysis of the rb-lc~ s will reveal the identity of cell proliferation genes useful for the di~gno~sic~ progllosis, and for the development of ~Igeted drugs for the ~ t~ of dicp~ces related to unregulated or ~,~vl)ropliate cell proliferation related to the rb signal tr~ncd~ction pathway.

W O 98/07886 PCT~US97114514 F. Example 6: Selection Systems Based On The Expression Of Genes In The PS3 And The P21 Pathway Selection systems are designlod which involve the p53/p21 pathway. pS3 or p21 are employed in selection e~ lel-~ analogous to the pl6-arrest e~t~c"ll-ents 5 described in Exarnple 1, supra, to select for random revertants or for the isolation of e~ obtained upon inrll)rtion with p.,~lull~agens.
Further analysis of thepS3 orp21 re~e.L~.~ will reveal the identity of cell proliferation genes useful for the ~ gnocic~ prognosis and for the development of 10 ~-~;~,t~,d drugs for the tre~ nt of ~licP~ps related to unregulated or i.~l,~lopliate cell proliferation associated with the p53 or p21 pathway.

G. Example 7: S~i~ctir~n Systems ~ased On E~ si~ Of The BRCAI
Gene In analogy to the pl6-arrest e~eli.. ,.-l~, the BRCAI gene may be expressed in tumor cell lines to select for the ide~,~ir~c~lion of novel cell proliferation genes. Specifically, BRCAI is ove,c~ ssed in the breast cancer cell line MCF-7. The so obtained selection systems may be used for the selection of random .e~ L~ or for 20 the isolation of revertants obtained upon induction with pellu~bagens.
Revertants of BRCA1-arrested cells are analyzed to identify do~l sl-~a--l medi~ors of BRCAI tumor ~u~piessor function, which may be useful for diagnosis, prognosis, and the development of drugs for the tre~ment of lii~;P~ces related to 25 unregulated or hlapplo~--ate cell proliferation ~ssoci~t~d with BRCAl, such as breast cancer.

H. Example 8: Selection Systems Based On Expression Of CDK
Inhibitors CDK inhibitors, inrhlflin~ plS, pl6, pl8, p21, p27, pS7 are e~l,.. ,ssed in Rb+ cells to select for random revertants or for the isolation of lI,V~ obtained upon induction with p.,llu-l,agens.
Revc~ of the CDK inhibitor-based selection systems are isolated and 35 analyzed to identify upstream mP~ ors of CDK inhibilol~; the inforrnation obtained will be useful for the ~ gnosi~, prognosis and for the development of targeted drugs for - s8 -the treatment of diseases related to unregulated or inappropriate cell proliferation associated with CDK inhibitors.
I Example 9: Selection Systems Based On Components Of Oncogene Pathways In order to identify the components of oncogene pathways, dominant-negative oncogenes or oncogene fragments of interest are expressed ectopically in a transformed cell such that growth is inhibited or apoptosis is induced. The dominant-negative oncogenes and cell systems employed in this experiment are listed in TABLE I, supra. The transformed cell lines may be used for the selection of random revertants or for the isolation of revertants obtained upon induction with perturbagens.
Revertant cells are isolated and analyzed to identify altered proliferation genes downstream in the oncogene's growth control pathway. These proliferation genes may be useful for the diagnosis, prognosis and for the development of targeted drugs for the treatment of diseases related to unregulated or inappropriate cell proliferation associated with oncogenes.
J. Example 10: Selection Systems Based On Tumor Formation And Metastasis In Vivo Genes that render tumorigenic cells non-tumorigenic are overexpressed in tumor cell lines. The non-tumorigenic cells are injected into immono-compromised mice, e.g., nude mice, followed by the isolation of clonal tumor variants. Revertant cell lines may be induced by introduction of perturbagens.
Analysis of these revertant cells permits the isolation of important cell proliferation genes that contribute to tumor formation, and genes that contribute to tumor formation in vivo may be directly analyzed and recovered. The so obtained genes may be used for the diagnosis, prognosis, and for the development of targeted drugs for the treatment of diseases related to unregulated or inappropriate cell proliferation associated with aberrant expression or control of these cell proliferation genes.

K. Example 11~ ection Systems Based On Ar~tosis Lymphocytes or cells derived from a lylllyhocyle cell line are cultured in tissue culture flasks to subconfl~Pnre. Anti-FAS antibody is added to the media in order to stimnl~te the FAS l~,cc~lol, resnlting in the inAllction of a~optosis; surviving le-e.~ll cells that fail to die are then isolated. These survivors which have lost, by mutation, key fllnrtion.c in the apol)tolic l)atll~.ay under study are idPntifiPd and analyzed and the unde.lyillg genes ~;,I onsible for al)ol)tosis or loss of apvlJts;s recovered.
Revertants may also be in~llced by the introduction of pe.lulbagens.
Analysis of these le~l~nt cells allow the isolation of cell proliferation genes that are involved in apoptopic ~a~ways, which may be useful for the diagnosis, prognosis and ~ -P ll of tli~P~eS related to unregulated or i~ lu~liaLe cell proliferation.

L. Example 12: SelectiQr~ Systems Based On C~r~, ~t L~lhibition A human nlelanol.la cell line which generally is contact inhibited~ such as HT-144, is grown in tissue culture flasks. Non-contact inhibited revertant cells that have lost i llpol~l.l growth regulatory signals are id~PntifiP~l by the formation of foci or 20 their ability to grow in soft agar and isolated.
Revcl~t~ may also be in~llced by the introdlction of l~cllulbagens. Gene e~l~n,;,sion is co...paled with that of the non-revertant parent cell line, and dirrc.enlially expressed genes in the le~l~nl-cells are identified and recovered. The so obtained genes may be used, e.g., for diagnosis, prognosis and the development of targeted drugs for the tleaL",enL of, e.g., cancer, in particular the treatment of melanoma.

M. Example 13: Sel~ticr~ Systems Based On The Growth Factor Requirement Of Non-Transformed Cells Non-trau~ru~lRd cells such as melanocytes are cultured in tissue culture flas~s in culture ...~ u~l~ rd with all required factors, inr~ ing phorbol ester, FGF, MSH-~, insulin/IGF-1. When the cells are sC~llicûllnl~ent~ a selec~e~ growth factor is removed from the media, reslllting in death of the vast majority of cells.
Su~sequ-ontly, I~ t cells which continue to grow in the ~hsenre of the factor are selecte~.

W 098/07886 ~CTrUS97/14514 Reversion of the cells may also be inAneed by il~ d~ n of pellusbagens~ The mutations that have eli...i..~t~d the function of the regulatory pa~ y that p~ sgrowth in the absence of the factor are identified and the c~.ll.,s~ond~g genes 5 recovered. The so obtained genes can have ll.lmc.uus medic~l applications, including gnr~sic and prognosis of di.ce~ces related to ullcolllrolled cell proliferation, and the development of drugs for the 1~ nl of such rli.ce~ces.

N. Example 14: Selection Systems Based On The Inability Of Non-Transformed T-Cells To Grow In lsolation Many non-~r~Çolll,cd T-cell lines can only be cloned, i.e., grown in isolation from other neighbors when the individual cells are placed on a "feeder layer"
of other cells.
Non trancformed T cell lines are diluted to a conce.ltl~.tion of 100 cells/ml; 10 ml of this cell suspension are then seeded on a ten (10) cm tissue culture plate.
Re~ s cells which do grow at low density in colonies are selected. Such revertants are ple~ullRd to contain alterations in genes involved in a pathway of growth depe~en~e on neighbors, and hence, depend on secreted factors. Rc~ cells are sel-octed and isolated, and the coll~ponding genes are recovered. The so identifipd cell proliferation genes may be used, e.g., for rli~gnosic, prognosis, and the development of targeted drugs for cancer therapy.

O. Example 15: Selecti~ Systems Based On I,.. ,.,o,lalization Of Primary Cells Freshly isolated human primary epithelial cells are cultured in suitable media; the vast majority of the cells has a finite lifespan and die after a certain number of cell doublings. Revertants which survive the "crisis phase" are selectPd. These revertant cells have undergone ch~lges that lead to immortalization and contain for mutations in genes that normally limit life span. S~seqmPntly, the dif~.en~ially~ ,l.,ssed or initi~tPd genes from these revertant cells when co~ ;d to noImal primary cells are recovered.

W O ~ o~6 PCTrUS97/14514 All lcf~ c~is cited within the body of the instant sr~ecific~tion are hereby in~ol~ol~ted by lei~ "ce in their entirety.

Claims (31)

WHAT IS CLAIMED IS:

1. A process for identifying a cell proliferation gene comprising the steps of:
(a) selecting a growth proficient revertant cell from a plurality of cultured cells arrested for growth, said growth arrested cells transformed with a library comprising a plurality of nucleic acid sequence inserts, wherein at least one insert from the library encodes a perturbagen within said revertant cell that results in said reversion to growth proficiency; and (b) identifying one or more genes or gene products in said revertant cell that cause said reversion to growth proficiency.

2. The process of Claim 1, wherein the cell proliferation gene is selected from the group consisting of an oncogene, a dominant transforming gene, a tumor suppressor gene and a gene involved in the control of apoptosis.

3. The process of Claim 1, wherein growth arrest of the cultured cells is caused by expression of a tumor suppressor gene or a dominant negative oncogene.

4. The process of Claim 1, wherein growth arrest of said cultured cells is caused by the expression of a gene inducing apoptosis in said cells.

5. The process of Claims 3 or 4, wherein said gene has been introduced on an expression plasmid under the control of a promoter.

6. The process of Claim 5, wherein said gene is expressed under the control of an inducible promoter.

7. The process of Claim 6, wherein the promoter is an IPTG inducible promoter.

8. The process of Claim 3, wherein said gene is a dominant negative oncogene selected from the group consisting of cJUN,, EGF-R, GRB2, RAF, MAX, RAS, SRC, and tyrosine kinase receptor mutants.

9. The process of Claim 3, wherein said gene is a tumor suppressor gene selected from the group consisting of P16, P53, RB1, WT1, BRCA1, BRCA2, NF1, NF2, P15, P18, P19, P21, P27, P57 and VHL.

10. The process of Claim 9, wherein the tumor suppressor gene is p16.

11. The process of Claim 1, wherein the cultured cells are selected from the group consisting of cells derived from primary tumors, cells derived from metastatic tumors, primary cells, cells which have lost contact inhibition, immortalized primary cells, transformed primary cells, cells which may undergo apoptosis, and cell lines derived therefrom.

12. The process of Claim 11, wherein the cultured cells express rb.

13. The process of Claim 11, wherein the cultured cells are derived from a melanoma cell line.

14. The process of Claim 13, wherein the melanoma cell line is HS294T.

15. The process of Claim 1, wherein the perturbagen library is introduced into said cells using a retroviral vector.

16. The process of Claim 1, further comprising the steps of:
(a) measuring differences in gene expression between said revertant cells and said growth arrested cultured cells to identify differentially expressed genes; and (b) isolating genes identified in step (a) to identify a differentially expressed cell proliferation gene.

17. The process of Claim 1, wherein the perturbagen is DNA encoding an RNA or polypeptide product which upon expression confers growth proficiency on said revertant cells.

18. The process of Claim 1, further comprising the steps of isolating the perturbagen present in said revertant cell and identifying the sequence of said perturbagen.

19. The process of Claim 1, wherein the perturbagen is DNA encoding a cell proliferation gene, a gene product thereof, or an active fragment of said gene or gene product.

20. The process of Claim 1, further comprising the step of identifying at least one cell component affected by said perturbagen.

21. The process of Claim 20, wherein said cell component is selected from the group consisting of a cell proliferation gene and a gene product thereof.

22. The process of Claim 1, wherein the perturbagen is or encodes a dominantly active peptide sequence that disrupts the action of an endogenous gene in a growth control pathway.

23. The process of Claim 22, wherein the endogenous gene is a tumor suppressor gene or a dominant-negative proto-oncogene.

24. The process of Claim 22, wherein the perturbagen disrupts the action of a cellular tumor suppressor gene, or one or more downstream targets, said process further comprising:
(a) identifying said tumor suppressor gene;
and (b) isolating said tumor suppressor gene.

25. The process of Claim 22, wherein the perturbagen disrupts the action of a cellular proto-oncogene, or one or more downstream targets, said process further comprising:
(a) identifying said cellular oncogene; and (b) isolating said cellular oncogene.

26. A process for identifying a compound that inhibits cell proliferation comprising the steps of:
(a) incubating a revertant cell obtained by the process of Claim 1 with one or more test compounds;
and (b) selecting from said test compounds a candidate compound that inhibits the growth of said revertant cell.

27. The process of Claim 26, wherein the inhibition of growth in step (b) is determined by an assay that tests a physiological response selected from the group consisting of inhibition of foci formation, inhibition of cell growth, inhibition of DNA replication and inhibition of tumor formation.

28. A compound obtained by the process of Claim 26 or 27.

29. A process for identifying a compound that inhibits cell proliferation comprising the steps of:
(a) exposing a polypeptide encoded by a cell proliferation gene obtained by the process of Claim 16, 18, 24 or 25 to one or more test compounds; and (b) selecting from said test compounds at least one candidate compound that inhibits a cell proliferative effect of said polypeptide.

30. The process of Claim 29, wherein the compound that inhibits cell proliferation is selected from the group consisting of an inhibitor of enzymatic activity, inhibitor of binding of a substrate molecule, inhibitor of phosphorylation, inhibitor of protein/protein interactions, inhibitor of protein/DNA interactions, and an inhibitor of protein/RNA interactions.

31. A compound obtained by the process of Claim 29 or 30.

32. A pharmaceutical composition comprising a therapeutically effective amount of a compound of Claim 28.

33. A pharmaceutical composition comprising a therapeutically effective amount of a compound of Claim 31.

34. The use of a compound of Claim 28 or 31 for the preparation of a pharmaceutical composition useful for treating a disease associated with aberrant cell proliferation.

35. The use according to claim 34, wherein the disease is selected from the group consisting of cancer, arteriosclerosis, psoriasis, rheumatoid arthritis and retenosis.

36. A method for identifying the expression in a tissue sample of a cell proliferation gene identified by the process of Claim 1, 16, 18, 20, 24 or 25, comprising the steps of:
(a) exposing nucleic acid derived from mRNA of said tissue sample to a labeled oligonucleotide probe comprising a sequence complementary to a fragment of said cell proliferation gene; and (b) identifying specific hybridization of said oligonucleotide probe with said nucleic acid.

37. An antibody against a cell proliferation gene or gene product obtained by the process of Claim 16, 18, 20, 24 or 25.

38. A method for identifying the expression in a tissue sample of a cell proliferation gene identified by the process of Claim 1, 16, 18, 20, 24 or 25, comprising the steps of:
(a) contacting a tissue sample from a patient with an antibody against said cell proliferation gene, and (b) identifying a specific interaction between said antibody and said tissue sample.

39. A method for identifying an individual predisposed to cancer comprising the steps of:
(a) exposing nucleic acid derived from chromosomal DNA from said individual to a labeled oligonucleotide probe comprising a sequence complementary to a fragment of a cell proliferation gene, said cell proliferation gene identified according to the process of Claim 1, 16, 18, 20, 24 or 25; and (b) identifying specific hybridization of said probe with said nucleic acid.

40. A nucleic acid which encodes a cell proliferation gene, said nucleic acid isolated by the process of Claim 16, 18, 20, 24 or 25.

41. A recombinant DNA molecule comprising nucleic acid sequences selected from the sequences which comprise the nucleic acid of claim 40.

42. An expression vector comprising nucleic acid sequences selected from the sequences which comprise the nucleic acid of claim 40.

43. A host cell comprising the recombinant DNA
molecule of claim 41.

44. A host cell comprising the expression vector of claim 42.

45. A diagnostic kit comprising a nucleic acid which is complementary to the nucleic acid of claim 40.

46. A diagnostic kit comprising an antibody against a cell proliferation gene or gene product according to

claim 31.