TNF receptor associated	AA456295	5.25
factor 6
Human COP9	AA489699	3.81
Antithrombin III	T62060	3.21
Mucin 1, transmembrane	AA488073	2.81
Casein Kinase,alpha 1	AA62578	2.79
Adenosine receptor A3	AA863086	2.47
calcium/calmodulin-	AA056626	2.42
dependent protein kinase II
Human protein immuno-	AA088258	2.34
reactive with anti-PTH
antibodies
Retinoic acid receptor,	AA496438	2.11
gamma 1

The present invention also provides for a customized kit to practice the method of the present invention.[0111]

The kit would be assembled to include at least an expression cDNA library constructed for specified cells as they are expressing the phenotype. Further a culture of cells of the requested phenotype could also be provided in the kit.[0112]

The above discussion provides a factual basis for the method of identifying genes that are essential for the maintenance of specific cell phenotypes. The methods used are shown below and can be shown by the following non-limiting examples and accompanying figures.[0113]

Throughout this application, various publications, including United States patents and applications, are referenced by author and year and patents and applications by number. The disclosures of these publications and patents and patent applications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art to which this invention pertains.[0114]

EXAMPLESGeneral Methods

General methods in molecular biology: Standard molecular biology techniques known in the art and not specifically described were generally followed as in Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Springs Harbor Laboratory, New York (1989, 1992), and in Ausubel et al., Current Protocols in Molecular Biology, John Wiley and Sons, Baltimore, Md. (1989).[0115]

Polymerase chain reaction (PCR) was carried out generally as in PCR Protocols: A Guide To Methods And Applications, Academic Press, San Diego, Calif. (1990).[0116]

Reactions and manipulations involving other nucleic acid techniques, unless stated otherwise, were performed as generally described in Sambrook et al., 1989, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, and methodology as set forth in U.S. Pat. Nos. 4,666,828; 4,683,202; 4,801,531; 5,192,659 and 5,272,057 and incorporated herein by reference.[0117]

Recombinant Protein Purification is undertaken as generally set forth in Marshak et al, “Strategies for Protein Purification and Characterization. A laboratory course manual.” CSHL Press, 1996 unless otherwise specified.[0118]

Vectors are constructed containing the cDNA of the present invention by those skilled in the art and should contain all expression elements necessary to achieve the desired transcription of the sequences (see below in specific methods for a more detailed description). other beneficial characteristics can also be contained within the vectors such as mechanisms for recovery of the nucleic acids in a different form. Phagemids are a specific example of such beneficial vectors because they can be used either as plasmids or as bacteriophage vectors. Examples of other vectors include viruses such as bacteriophages, baculoviruses and retroviruses, DNA viruses, cosmids, plasmids, liposomes and other recombination vectors. The vectors can also contain elements for use in either procaryotic or eucaryotic host systems. One of ordinary skill in the art will know which host systems are compatible with a particular vector.[0119]

The vectors are introduced into cells or tissues by any one of a variety of known methods within the art (calcium phosphate transfection; electroporation; lipofection; protoplast fusion; polybrene transfection).[0120]

The host cell can be any eucaryotic and procaryotic cells, which can be transformed with the vector and which will support the production of the enzyme. Methods for transformation can be found generally described in Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Springs Harbor Laboratory, New York (1992), in Ausubel et al., Current Protocols in Molecular Biology, John Wiley and Sons, Baltimore, Md. (1989), Chang et al., Somatic Gene Therapy, CRC Press, Ann Arbor, Mich. (1995), Vega et al., Gene Targeting, CRC Press, Ann Arbor, Mich. (1995) and Gilboa, et al. (1986) and include, for example, stable or transient transfection, lipofection, electroporation and infection with recombinant viral vectors. In addition, see U.S. Pat. No. 4,866,042 for vectors involving the central nervous system and also U.S. Pat. Nos. 5,464,764 and 5,487,992 for positive-negative selection methods.[0121]

General methods in immunology: Standard methods in immunology known in the art and not specifically described were generally followed as in Stites et al.(eds), Basic and Clinical Immunology (8th Edition), Appleton & Lange, Norwalk, Conn. (1994) and Mishell and Shiigi (eds), Selected Methods in Cellular Immunology, W.H. Freeman and Co., New York (1980).[0122]

Immunoassays: In general, ELISAs are the preferred immunoassays employed to assess a specimen. ELISA assays are well known to those skilled in the art. Both polyclonal and monoclonal antibodies can be used in the assays. Where appropriate other immunoassays, such as radioimmunoassays (RIA) can be used as are known to those in the art. Available immunoassays are extensively described in the patent and scientific literature. See, for example, U.S. Pat. Nos. 3,791,932; 3,839,153; 3,850,752; 3,850,578; 3,853,987; 3,867,517; 3,879,262; 3,901,654; 3,935,074; 3,984,533; 3,996,345; 4,034,074; 4,098,876; 4,879,219; 5,011,771 and 5,281,521 as well as Sambrook et al, Molecular Cloning: A Laboratory Manual, Cold Springs Harbor, N.Y., 1989.[0123]

Polyclonal and Monoclonal Antibody Production: Antibodies may be either monoclonal or polyclonal. Conveniently, the antibodies may be prepared against a synthetic peptide based on the sequence, or prepared recombinantly by cloning techniques or the natural gene product and/or portions thereof may be isolated and used as the immunogen. Such proteins or peptides can be used to produce antibodies by standard antibody production technology well known to those skilled in the art as described generally in Harlow and Lane, Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1988 and Borrebaeck, Antibody Engineering—A Practical Guide, W.H. Freeman and Co., 1992.[0124]

For producing polyclonal antibodies a host, such as a rabbit or goat, is immunized with the protein or peptide, generally with an adjuvant and, if necessary, coupled to a carrier; antibodies to the protein are collected from the sera.[0125]

For producing monoclonal antibodies the technique involves hyperimmunization of an appropriate donor with the protein or peptide fragment, generally a mouse, and isolation of splenic antibody producing cells. These cells are fused to a cell having immortality, such as a myeloma cell, to provide a fused cell hybrid which has immortality and secretes the required antibody. The cells are then cultured, in bulk, and the monoclonal antibodies harvested from the culture media for use.[0126]

The antibody can be bound to a solid support substrate or conjugated with a detectable moiety or be both bound and conjugated as is well known in the art.[0127]

(For a general discussion of conjugation of fluorescent or enzymatic moieties see Johnstone & Thorpe, Immunochemistry in Practice, Blackwell Scientific Publications, Oxford, 1982.) The binding of antibodies to a solid support substrate is also well known in the art. (see for a general discussion Harlow & Lane Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Publications, New York, 1988 and Borrebaeck, Antibody Engineering—A Practical Guide, W.H. Freeman and Co., 1992) The detectable moieties contemplated with the present invention can include, but are not limited to, fluorescent, metallic, enzymatic and radioactive markers such as biotin, gold, ferritin, alkaline phosphatase, 0-galactosidase, peroxidase, urease, fluorescein, rhodamine, tritium, “C and iodination.[0128]

Specific Methods

Construction of Anti-sense Expression Vector:[0129]

This method is not limited to any specific vector system. The actual requirements are that the vector system express at high levels anti-sense molecules and that they can be identified. In a preferred embodiment the Epstein Barr Virus (EBV) Episomal vector is used [Deiss et al, 1991].[0130]

The EBV episomal vector consists of DNA segments that are necessary for the episomal maintenance of the episome both in bacteria ([0131]E. coli) and in human cells (this include an origin of replication and a trans-acting factor (EBNA-1). The episome also includes genes encoding resistance markers for selection either in bacteria or in human cells. Finally the vector contains a transcription cassette. Initially it will be based on the vector as described in Deiss et al. [1991], but the present invention contemplates any transcription cassette that produces high levels of anti-sense expression. The EBV episomal vector contains a RNA Polymerase II promoter and or enhancer driving the transcription of a synthetic transcript containing a set of cloning sites, a splice donor and acceptor site and a polyadenylation signal, followed by a second set of enhancers. This vector can be efficiently shuttled from animal cells to bacteria and vice versa. one procedure that allows for rapid shuttling is using the method of Hirt [1967] to extract episomal vectors from animal cells and using this preparation to transformE.coli. Applicants have observed that, on average, cells transfected with a library cloned into the vector contain only one expressing vector.

The specific choice of promoters and enhancers are dependent on the exact selection condition and the cell line used. This must be empirically determined for each selection condition as is known to those skilled in the art.[0132]

In an embodiment, the EBV vector can also contain an inducible expressed promoter such that the expression of the anti-sense library would be inducibly expressed by a specific inducer. This will allow additional flexibility in designing selection protocols.[0133]

Construction of Anti-sense cDNA Library:[0134]

There are several methods available to construct directional cDNA libraries. Any of these methods would be sufficient since they result in the production of a directionally identified cDNA library and the practitioner can use the method they are most familiar with. The directional cDNA is then cloned into the expression cassette in the anti-sense orientation. A method that may be used is detailed in Deiss et al, 1991.[0135]

Briefly, it consists of making cDNA by the method of Gubler and Hoffman [1983] and making the cDNA directional by the method of Meisner et al. [1987].[0136]

The mRNA is extracted from cells that have been cultured under a variety of conditions that mimic the actual selection conditions. This is designed to ensure that the library will include all the messages that are expressed in the target cell under selection conditions.[0137]

RNA is prepared at time points that will contain messages that are always present as well as messages that are induced by the selection procedure. This is achieved by extracting RNA previous to the selection and at times during the selection. The various pools of RNA are then mixed together so that all possible RNA molecules are present [Deiss and Kimchi, 1991].[0138]

An alternative method that can be used consists of deriving a library of genomic DNA fragments cloned into the expression cassette. Since all the transcribed messages are derived from genomic DNA (with the exception of RNA edited messages; this will actually include mitochondrial DNA as well) this method would generate all possible messages. The directionality would be lost so the library would be only half anti-sense. Since the sense fragments are unlikely to frequently encode full length proteins or have biological activity the anti-sense fragments would still likely produce the most frequent biological effects. The genomic fragments are produced by restriction enzyme cleavage of genomic DNA.[0139]

It would be necessary to produce only one library per species, since with the exception of the B and T cell receptors, genomic DNA does not differ in different cell types at least in mammals (again, erythrocytes or any cells that lack nuclei are an exception).[0140]

In the case of the genomic library is be necessary to determine whether any expressed fragments express a sense or an anti-sense message. This is done by using the insert as a strand specific probe, both from the expressed and non-expressed strand in a Northern analysis. This indicates if the expressed fragment is sense or anti-sense in relation to the endogenously expressed gene.[0141]

Some sequence will match, some will not match, genes already deposited in the various databases. In the case where the identified gene matches the sequence of a gene already in a database this information would then enable the determination if the insert is a sense or anti-sense insert.[0142]

Transfection of the Anti-sense Library:[0143]

There are a large variety of methods to transfect DNA into cell lines and cell cultures. The most efficient method for each selection is determined empirically based on experience and the known relative efficiency of each method.[0144]

The method selected must both efficiently delivery DNA into cells and not effect the biological responses that will be selected following the transfection. Viral vector system can also be used and this would entail producing infectious virus and infecting the target cells. Applicant has found electroporation to be an efficient method, but other methods can be used as are known in the art.[0145]

Identification of Differentially Expressed Antisense Messages:[0146]

The methods of identifying the differentially expressed antisense messages in a preferred embodiment includes the methods described in Braun [1996] and as described in Diatchenko et al.[1996]. These methods include a PCR amplification of subtracted populations.[0147]

Appropriate restriction sites are included in the expression vector so that following PCR amplification of the cDNA inserts, the inserts are flanked by appropriate restriction sites. Restriction digestion is then used to produce templates that are useful for these techniques.[0148]

Another method that is used is the “GEM” gene expression microarray as described in Schena et al. [1995]. In this technique, PCR fragments corresponding to a set of specific plasmids (in this case antisense cDNA inserts contained in antisense expressing vectors or other DNAs as appropriate) are fixed to a glass template and this is hybridized with two fluorescently labeled probes. In this specific case, the probes are reverse transcribed antisense transcripts derived from cells transfected with the antisense expression library either before or following a selection.[0149]

Generation of Efficient Antisense Inhibitor:[0150]

In addition to assaying expression cassettes where all the transcripts are directionally cloned in the antisense orientation, another strategy employed in the present invention is to generate randomly primed cDNA and cleave the cDNA with two restriction enzymes (hereby represented by “X” and “Y”) and clone the resulting mixture into two different expression cassette. In the first cassette site, X would transcriptionally precede Y, and the second cassette site, Y would transcriptionally precede X. In this arrangement, the cDNA is divided into sections that may have different abilities to serve as an efficient antisense inhibitor. The strongest differential signal is likely to be produced by the fragment that is the most efficient antisense inhibitor. Thus, the screening is more likely to produce a meaningful differential signal.[0151]

SPECIFIC EXAMPLESExample 1Determination of Combination of Methods for Preferred Embodiment

The method requires two distinct major steps as described herein above. It is greatly advantageous if this method can be applied to a wide variety of cells and situations.[0152]

In the first step genes are inactivated in order to determine whether individual genes are essential for a specific phenotypic change. It is advantageous if these inactivation will have a phenotype both in haploid cells and in diploid cells. Since many cells of interest are diploid in nature. Furthermore, it is also an advantage if the inactivation method allows for the rapid identification of the inactivated genes. This can be achieved in a variety of manners. The inactivation methods are generally based on one of three different principles.[0153]

The first principle is that genes can be functionally inactivated by expressing mRNA that is derived from the anti-sense strand of the sense message.[0154]

This allows for inactivating the mRNA in the cell and does not require a specific gene dose. This can work for single copy or multiple copy genes either from haploid or diploid organism. It has been shown that anti-sense inactivation can be effective in a wide range of organisms including bacterial, plant and animal.[0155]

Applicants and others have extended the original observation by generating anti-sense expressing cDNA libraries. Applicants termed this method Technical Knock Out (TKO). These libraries contain collections of many (usually 100,000 to 1,000,000) different anti-sense expression constructs that will individually express a single anti-sense RNA molecule when transfected into appropriate target cells. Since these libraries contain large collections of these vectors they in effect can express anti-sense RNA to virtually all expressed mRNA molecules. Several investigators have used these type of libraries to inactivate genes and change the phenotype of cells. Once an altered cell is identified the expression cassette contained in the cells can be identified since the expression cassette DNA sequence is known. Subsequently, the anti-sense expressed cDNA molecule that is contained in the expression cassette is identified. This can be achieved by a variety of methods; applicants have used two methods. The first method involves shuttling the vector from animal cells into bacterial cells. Once the vector is in bacterial cells it is easy to produce large amounts of the vector for further analysis. The second method we have employed involves PCR amplification of the cDNA inserts by designing PCR primers that flank the cDNA cloning site on the vector. The flanking vector sequences are known so it is easy to chose appropriate primers. PCR amplification with these primers amplifies any cDNA molecules that were present between the two primers. The anti-sense approach also allows for tagging of the inactivation event. That is the identity of the sense message generally can be determined by sequencing the anti sense construct. This construct can then be identified and isolated from the phenotypically altered cells.[0156]

The second gene inactivation method that fits these requirement is an inactivation method that relies on production of “dominant negative” fragments of genes from an expression cDNA library. This method is called the Genetic Suppressor Element method (GSE). It is based on the observation that small fragments of a gene when is expressed may interfere with the normal function of the full length gene product and in fact interfere with the normal function. In this manner these gene fragments were called “dominant negatives”. A GSE library thus consists of fragmented cDNA molecules which are cloned into an expression cassette. When expressed from translation initiation signals in the cDNA molecule, or from translation initiation signals present in the expression cassette, these gene fragments can interfere with gene function. In addition the libraries used in the GSE method also include some anti-sense fragments and therefore gene inactivation can occur either by anti-sense or by dominant negative inactivation of gene function.[0157]

The third method of gene inactivation that could be used is called “Random Homozygous Knock-Out” (RHKO). In this method gene inactivation is achieved in two steps.[0158]

A retroviral vector is used to infect target cells. The integration of the retroviral element itself can lead to inactivation of one copy of a gene if the integration event itself functionally disrupts the normal transcription or activity of the gene in which it integrates. The retroviral vector used has an additional property that it encodes a transcription element that should transcribe into the chromosomal location in which it has integrated. In the case that this generates and anti-sense RNA transcript, additional copies of the gene could be inactivated. Thus this method also relies on anti-sense inactivation.[0159]

The TKO method was chosen for generating inactivations in these examples as the preferred embodiment because the other methods described above are not as compatible with the second step in the method of the present invention as the TKO procedure. However, as improvements become available in these methods they could be used.[0160]

In the GSE method both sense and anti-sense gene fragments are generated which are expected to have different biological activities. It is difficult to distinguish closely related gene fragments of this sort by the methods that will be used in the second part of the gene identification method of the present invention.[0161]

Thus the rare molecules that cause biological changes when expressed would be very difficult to distinguish from many similar molecules that would not have biological effects. The molecules that do not have effects would in essence mask the active molecules.[0162]

The RHKO method was also difficult to adapt to a high throughput subtractive procedure. The potential anti-sense fragments generated in this procedure must be cloned out individually and this is a process that is hard to adapt to subtraction.[0163]

The TKO method was easily adapted to subtraction.[0164]

The cDNA inserts contained in expression vectors should be all or at least mostly anti-sense in nature. The cloning procedure outlined in Deiss and Kimchi was used.[0165]

This generates anti-sense cDNA libraries and results in libraries that are biased to be anti-sense. It is possible to obtain some sense cDNA inserts with this method. Thus since most of the fragments are antisense the subtraction step will be mainly between different cDNA fragments that were expressed as anti-sense constructs. Again the principle of the present invention is that the abundance of an anti-sense construct(s) that induces a disadvantageous phenotype will be reduced after a biological selection. In order to identify these constructs,the TKO method was chosen for the gene inactivation step of the present invention.[0166]

It can be used in a variety of cell populations, both in haploid, diploid and aneuploid cells. It can be easily scaled up to involve 100,000 events or more without undue expense, and it can be easily adapted to the subtractive methods that are needed in the second part of the method of the present invention.[0167]

The second step in the gene identification method requires identification of the loss of specific anti-sense gene constructs from a large population of anti-sense constructs that are not lost. This can be accomplished in a variety of different ways.[0168]

Because it is a great advantage to be able to identify specific losses in the presence of large numbers of molecules that are not lost we needed a method that has a high throughput capacity. One method that fits this requirement involves using high density arrayed chips such as the GEM chips. These are arrayed dots containing specific DNA molecules corresponding to genes. The dots are arrayed at high density on a glass coverslip with the position of each dot and the identity of the DNA molecule fixed on each dot precisely determined. Two probes derived from different population of DNA or RNA molecules are labeled with two different fluorescent dyes and hybridized to the arrays. After appropriate washing the relative binding of the dyes at each dot is determined.[0169]

The amount of dye bound at each spot reflects the abundance of the gene fixed on that dot relative in the whole population. Thus when two populations of DNA molecules are labeled with different dyes one can accurately determine whether there has been a change in relative abundance of individual molecules in the population. If there is no change the ratio of the two dyes will be one. If there has been a change in abundance then the ratio of the two different dyes will also change. This method can rapidly measure the changes of large numbers of genes in a large population. A copy of the DNA fix on each dot is stored and can be retrieved for further analysis. Although in the following example the GEM method was not used to measure the loess of anti-sense constructs it is a method that can be used in the practice of the present invention.[0170]

A second method used to identify the loss of specific anti-sense gene constructs from a large population of anti-sense constructs that are not lost is called “Subtraction”. This involves manipulating two populations of DNA or RNA molecules so that only molecules that are present in one population and not in another are recovered. The version that was actually used is called PCR-Select which is a commercially available kit from CLONTECH. Briefly 1. Two populations of double stranded DNA are generated. It is assumed that some of the dsDNA molecules are present in only one of the populations (or at least much more abundant in one population).[0171]

2. These populations are separately processed. The population that is assumed to have extra species of molecules is called the tester sample. The second population that is assumed not to have these specific species is called the driver. The tester population is separately ligated to two different linkers. This generates[0172]

tester population

1 andtester population2.

The driver is left without linkers.[0173]

3. A series of manipulations including denaturation and renaturation of the driver and tester in various combinations is used. This results in generating a series of DNA molecules that have different set of linkers at their ends. The only product that can be effectively PCR amplified at the end of the manipulations are those that are present in the tester and absent (or reduced) in the driver. These molecules are easily isolated after the last PCR step.[0174]

The net result of this method is that a population of gene fragments that are present in one population and lost or absent in another population is rapidly isolated.[0175]

This is exactly what is needed in the gene identification method. These PCR products can be isolated by standard techniques and be used for further analysis as shown in Example 2.[0176]

Example 2Identification of Genes in HeLa Cells that are Involved in Fas Antibody Sensitivity

The method of the present invention was applied to HeLa cells treated with anti-Fas antibody in order to identify genes that when knocked-out cause sensitization of HeLa cells to the action of anti-Fas antibodies.[0177]

HeLa cells are derived from a human cervical carcinoma and were used in the original TKO [Deiss and Kimchi, 1991]. HeLa cells were used as an exemplar of the method of the present system as they are easily grown in culture, are easily transfected and respond to anti-Fas antibody treatment.[0178]

Anti-Fas antibody (Kamiya Biomedical Company, Seattle, Wash., catalog number: MC-060) is directed against Fas/CD95/Apo-1, a transmembrane receptor that is known to signal a death response in a variety of cell types. This antibody is an activating antibody, that is, the binding of the antibody mimics the effects of binding of ligand. Applying the appropriate dose to responding cells has been shown to lead to induction of cells death (Deiss et al., 1996). HeLa cells respond to this treatment.[0179]

In this exemplar genes are identified that regulate the sensitivity of HeLa cells to killing by anti-Fas antibody. Specifically, genes are identified whose loss sensitizes HeLa cells to anti-Fas treatment.[0180]

The outline of the procedure is as follows:[0181]

1. HeLa cells were transfected with an anti-sense cDNA library.[0182]

2. Cells containing anti-sense expression vectors were isolated by selection with Hygromycin. Since the vector contains the Hygromycin resistance marker, the selection of the transfected cultures with Hygromycin generated a population of cells which contain the anti-sense expression cassettes.[0183]

3. Aliquots of this pool of cells were treated with anti-Fas antibody under two different experimental conditions. It should be noted that more conditions could be screened at the same time.[0184]

a. Treatment with a sub-lethal dose of anti-Fas antibody (10 ng/ml). Cells that are super-sensitive to treatment with anti-Fas antibody were killed whereas the majority of the population which is resistant to the treatment proliferated.[0185]

b. In the second condition, the cells were treated with a lethal dose of anti-Fas antibody (100 ng/ml). The cells were harvested at 24 hours, before the majority of cells had been killed. In this case, applicants were looking for anti-sense events that accelerate the killing associated with anti-Fas treatment as another type of sensitization.[0186]

4. Aliquots of the cells just before the treatment with anti-Fas antibody and just after the treatment with anti-Fas antibody were harvested. The DNA contained in each cell population was extracted.[0187]

5. The anti-sense cDNA inserts contained in these DNA samples were preferentially amplified through the use of PCR (see details below).[0188]

6. The pools of anti-sense cDNA fragments that were derived from cells after treatment were subtracted from those before treatment(see details below). This generated a set of cDNA fragments that were present in cells before treatment but were absent after treatment.[0189]

These fragments are good candidates for sensitizing cDNA fragments. In other words, it is likely that expression of some of these fragments leads to the inactivation of genes which causes cells to become super-sensitive to anti-Fas antibody treatment. These super-sensitive cells are quickly killed at a lower dose of anti-Fas antibody or more rapidly than the majority of cells. These cells are therefore lost from the treated cultures but are present in the untreated population. Likewise, the plasmids inducing this super-sensitivity are present in the cells before treatment but are absent from the cell sample taken after treatment. Thus, these fragments are identified during the subtraction.[0190]

7. The cDNA fragments generated by the subtraction were cloned into the original expression vector.[0191]

Appropriate restriction enzyme sites were generated or maintained during the subtraction procedure so that the recloned construct is exactly identical to the construct in the originally transfected cells. The sequence of the isolated cDNA fragments was determined.[0192]

8. The anti-sense expression plasmids containing the cDNA inserts that were identified in the method of the present invention were individually re-transfected into HeLa cells and the transfectant cells were assayed for sensitivity to anti-Fas antibody treatment.[0193]

Specific Materials and Methods[0194]

HeLa cells were transfected with anti-sense cDNA library cloned in the episomal vector, anti-sense expression vector pTKO-1. This is the same library described in Deiss and Kimchi [1991]. One million cells plated in a 100 mm dish were transfected with 15 μg of DNA containing the anti-sense cDNA library, by using the Superfect reagent (Qiagen, Santa Clarita, Calif.) as suggested by the manufacturer. Two days following transfection, cells were treated with Hygromycin B (200 [μg/ml) (Calbiochem-Novabiochem Corporation, La Jolla, Calif.). Following two weeks of selection, the population of cells was completely resistant to Hygromycin B.[0195]

These cells were plated in triplicate at a density of 2.5×10[0196]⁶cells per 150 mm dish in the absence of Hygromycin B. One plate was treated with anti-Fas antibody at 10 ng/ml (clone CHI-11 Kamiya Biomedical Company, Seattle, Wash.) for five days, the second plate was treated with 100 ng/ml of anti-as antibody for 24 hours and the third plate was UN-treated for 24 hours.

Following the treatments, the cells were harvested by washing twice with ice cold PBS (NaCl 8 g/liter; KCl 0.2 g/liter; Na[0197]₂HPO₄1.44 g/liter; KH₂PO₄0.24 g/liter; final pH of solution adjusted to pH 7.4 with HCl) and concentrated by centrifugation (15,000×g for 15 seconds). DNA was extracted by using solutions P1, P2 and P3 from the Qiagen Plasmid Purification Kit (Qiagen, Santa Clarita, Calif.). The cell pellet was resuspended in 200 μl of solution P1 (50 mM Tris-HCl, pH 8.0; 10 mM EDTA; 100 μg/ml RNase A) then mixed with 200 μl of solution P2 (200 mM NaOH, 1% SDS) and incubated five minutes at room temperature. 200 μl of solution P3 (3.OM Potassium Acetate, pH 5.0) were added and incubated two minutes at room temperature, followed by a ten minute centrifugation at 15,000×g. The clear supernatant was mixed with an equal volume of isopropanol and centrifuged at 15,000×g for ten minutes. The precipitated DNA was resuspended in 100 μl of water and stored frozen until use.

For PCT amplification of the cDNA inserts contained in these DNA preparations, the following reaction was set in a total volume of 100 μl: 1 μl of the DNA, 200 μM of dATP, dGTP, dCTP, dTTP, 500 ng each of primers; 10 mM Tris-HCl pH 9.0; 0.1% Triton X-100; 1.0 mM MgCl and 1 unit of Taq DNA polymerase (Gibco/BRL, Gaithersburg, Md.). This reaction was incubated in a Thermocycler 2400 (Perkin-Elmer, Foster City, Calif.) according to the following protocol: First, the reaction was heated to 94° C. for five minutes, then was cycled 25 times using the following three temperatures: 58° C. for one minute, 72° C. for five minutes, 94° C. for one minute. After 25 cycles, the reaction was incubated at 72° C. for seven minutes.[0198]

This resulted in amplification of the cDNA inserts. Primers were designed such that the end of the cDNA insert that is proximal to the promoter in the pTKO-1 vector is exactly flanked by a HindlIl restriction site (this site is present in the vector) and the end of the cDNA that is distal to the promoter in pTKO-1 vector contains a BamHI restriction site. The BamHI site was created by altering a single base in the sequence immediately adjacent to the distal cDNA insert site , by PCR. When the library was generated [Deiss and Kimchi, 1991], this site distal to the promoter was generated by the fusion of a BamHI restriction site (derived from the cDNA fragments) and a BgIII site (derived from the vector). This fused site is resistant to cleavage by either enzymes, but a single base change restored the cleavage by BamHI. Thus, the amplified cDNA fragments are flanked by a HindIII restriction site on the promoter proximal side of the cDNA and by a BamHI site on the promoter distal side.[0199]

This allowed the exact re-cloning of the fragments into the pTKO-1 expression vector with exact conservation of sequence and orientation.[0200]

Following the PCR reaction, the mixture was cleaved with BamHI and HindIII (Gibco/BRL, Gaithersburg, Md.) as described by the manufacturer. The digestion products were purified using the Wizard PCR Prep Kit (Promega, Madison, Wis.). This generated cDNA inserts with HindIII and BamHI ends.[0201]

These nucleic acid fragments were subjected to subtraction using the PCR-Select Kit (Clontech, Palo Alto, Calif.) according to the instructions of the manufacturer with the following modifications. The driver was the PCR products derived from the untreated samples and two testers were used. The first tester was derived from cells treated with 10 ng/ml anti-Fas antibody and the second tester was derived from cells treated with 100 ng/ml of anti-Fas antibody. First modification: the subtraction was done between dsDNA pools so no cDNA synthesis is required. The fragments generated from the previous step were used directly in the subtraction. Thus, applicants began at Step IV F3 in the instructions (preparation of the adapter ligated tester cDNA). The second modification was the replacement of the blunt end ligation of[0202]

adapter

1 and adapter2R with cohesive end adapters. These cohesive end adapters were ligated to the BamHI and HindIII cleaved PCR fragments generated in the step above. The cohesive ligation is usually more efficient than blunt end ligation and since applicants use cDNA flanked by different restriction sites allowing the orientation of the fragments to be maintained when recloning the subtracted products. If the blunt end ligation is used, it would not allow distinguishing one end from the other and applicants would not be able to determine the relative orientation of the cDNA in the original expression cassette. Thus,adapter1 was replaced by an equal mixture of the appropriate primers. The same applies for adapter2R. The other primers were of identical sequence as described in the kit. The manual supplied by the manufacturer with the kit was followed from the point of ligation of the adapters to the tester (Section IV F3 in the Manual). 0.3 μg of the tester was taken for adapter ligation. The initial hybridization included 0.9 μg of the driver and 0.03 μg of the adapted ligated tester. At the conclusion of the subtraction, a final PCR reaction is done using nestedPCR primer1 and nested PCR primer2R. This material contains the cDNA fragments that were present in the untreated sample but absent from the treated samples. The product of this PCR reaction were re-cloned into the anti-sense expression vector.

Re-cloning of the subtracted fragments was accomplished by cleaving the subtracted population with BamHI and HindIII and purifying the cleaved products with the Wizard PCR Prep Kit (Promega Madison, Wis.).[0203]

The cleaved products were then directly cloned into the pTKO1-DHFR vector between the HindIII and BgIII sites.[0204]

This replaced the DHFR sequences with the cDNA. This is precisely the procedure that was used to generate the anti-sense cDNA expression library. Thus, the fragments that were generated by the subtraction were exactly re-cloned into the original anti-sense expression vector that was used to transfect cells at the beginning of the procedure. The re-cloned constructs exactly duplicate the constructs that were present in the library. The re-cloned constructs were introduced into bacteria and DNA was extracted from the bacteria following conventional methods. These DNA preparations were used as a template for sequencing in order to determine the nucleotide sequence of the isolated cDNA inserts.[0205]

In addition, plasmids carrying the re-cloned inserts were transfected into HeLa cells to confirm their ability to induced super-sensitization to anti-Fas antibody treatment in HeLa cells.[0206]

HeLa cells were transfected with 15 μg of plasmids or control vectors as described for transfection of the original library. The cells were selected for two weeks for resistance to Hygromycin B treatment (200 μg/ml).[0207]

This selects for cells which contain expression cassettes. One million cells were plated in a 100 mm dish and treated with anti-Fas antibody. Effects of anti-Fas antibody on the transfected cultures were quantified by MTT assays as described by the manufacturer (Sigma, St. Louis, Mo.)[0208]

Analysis of the Isolated DNA Sequences

The clones were sequenced using a primer which anneals close to the edge of the cDNA which is distal to the promoter in the antisense expression cassette. Thus, in the case that the sequence matches the sense strand of a known gene then the insert is in the antisense orientation. Sequences were compared to the combined nonredundant database and the dbest compiled at the NCB[0209]1 using the Blastn program with default parameters. New and known sequences were determined by the method of the present invention, some of which are detailed in Table A; see also genes discussed in Example 3.

As described above, the isolated fragments were recloned and then reassayed for sensitivity to treatment with anti-Fas antibody (50 ng/ml for 72 hrs) using the MTT assay. These assays showed that expression of two of the fragments resulted respectively in a 1.6 and 2.0 fold increase in sensitivity to anti-Fas antibody treatment whereas expression of CrmA (a protective protein) resulted in a 2.3 fold reduction in sensitivity. This demonstrates that the method of the present invention can be successfully used to identify genes based on a positively selected phenotype.[0210]

For further information on the methods of the present invention and the fragments and genes isolated thereby, see co-assigned PCT publication No. WO 98/21366, co-assigned U.S. Pat. No. 6,057,111, and co-assigned PCT publication No. WO 01/57189 which are hereby incorporated by reference in their entirety.[0211]

Example 3

The Achilles Heel Method utilizes functional profiling as diagrammed in FIG. 3. The first step consists of introducing an anti-sense expression library (Deiss and Kimchi, 1991) into target cells to generate a pool of cells, each expressing a different anti-sense fragment (Pool[0212]1). Then, the transfectants are treated with a sub-optimal dose of a PCD inducer and the surviving cells are collected (Pool2). Cells containing inactivation events that sensitize the cells to killing are preferentially lost fromPool2. Consequently, the anti-sense cDNAs contained in the sensitized cells are depleted fromPool2. The “sensitizing” cDNA inserts that are present inPool1 but depleted fromPool2 are identified by two methods, subtraction or hybridization to cDNA microarray. Following the subtraction ofPool2 cDNAs fromPool1 cDNAs, the potentially sensitizing cDNAs are cloned in an anti-sense orientation in an episomal expression vector. The anti-sense cDNA containing episomes are individually transfected into target cells in order to confirm their ability to render the cells more sensitive to the killing inducer. Alternatively,Pool1 andPool2 cDNAs are labeled and used as probes for hybridization of cDNA microarray filter. Computer analysis identifies the cDNAs depleted fromPool2. In both cases “function profiling” is being employed to identify signal pathway inhibitors. Recently, similar “function profiling” methods have been described for genetic analysis ofS.cerevisae(Pat Brown and Ron Davis). These methods are well suited to yeast since they require prior knowledge of gene sequence and the ability to generate haploid cells. By contrast, AHM does not require a priori knowledge of any gene sequence or haploid cells. Thus, AHM is a powerful genetic tool for “function profiling” in mammalian cells. Moreover, AHM can be easily scaled up to generate “function profiles” of all expressed human genes.

AHM is used to identify inhibitors of the Fas induced PCD pathway. Fas is a trans-membrane death receptor of the TNF super family. The binding of Fas ligand to Fas results in the cascade of events that lead in most cell types to apoptosis. Fas induced killing is utilized in different physiological processes as follows: (for review see Nagata, Gloldstein etc): elimination of auto-reactive T-cells, tumor induced immune suppression and destruction of virally infected cells, transformed cells and b-cells in cases of Insulin Dependent Diabetes Melitus (IDDM). In addition, activation of the Fas pathway has been suggested to play a role in liver damage, brain damage, arteriosclerosis and tumor suppression. Modulation of the Fas pathway has clinical implications in animal models: inhibition of Fas induced PCD by caspase inhibitors limits liver damage in mice and acceleration of Fas induced killing ameliorate the auto-immune phenotype of gld mice. Thus, identifying regulators of the Fas pathway that can be used as targets for drug development will have great clinical impact.[0213]

For the identification of inhibitors of Fas induced cell death, AHM was applied to HeLa cells that were treated with sub-lethal dose of Fas agonistic antibody. The latter mimics the binding of Fas ligand to Fas and induces apoptosis. “Function profiling” was performed to identify “sensitizing” cDNA fragments by using subtraction and gene array analysis. cDNA inserts from[0214]

Pool

2 were subtracted fromPool1 cDNAs and the recovered cDNAs were further analyzed. Sequencing of 226 fragments revealed 168 unique sequences, of which 53% are novel and 47% correspond to known genes. Six out of seven randomly chosen cDNAs that were individually transfected into HeLa cells conferred increased sensitivity to Fas induced killing cells, ranging between 2.9 to 5.3 fold. These fragments include three novel sequences and three fragments of previously described genes. One of the cDNA inserts is an anti-sense fragment of human Basic Fibroblast Growth Factor (FGF-2, bFGF) and the other is an anti-sense fragment of the cap-n-collar b-zip transcription factor NF-E2 related factor 2 (NRF2).

bFGF is a potent survival factor that plays a role in development, angiogenisis and in cell migration. Previous reports show that down regulation of bFGF by anti-sense expression or by blocking antibodies result in loss of a transform phenotype, reduced tumor growth and reduced angiogenesis. Five different polypeptides of 34 kD, 24 kD, 22.5 kD, 22 kD and 18 kD are translated from the human bFGF gene, initiating at different sites and terminating at the same position. The anti-sense cDNA fragment isolated in the subtraction is 295 nucleotides long and corresponds to nucleotide 890 to 1184 of the bFGF gene. It spans the last 60 nucleotides of the coding region (shared by all bFGF polypeptides) and a portion of the 3′ untranslated region.[0215]

In order to confirm that anti-sense bFGF confers sensitivity to Fas, pools of cells transfected with control vector (harboring no insert) or with anti-sense bFGF were generated and treated with sub-optimal dose of anti-Fas antibody. Analysis of two independent pools of transfectants demonstrates that under conditions that results in 59% and 29% killing of the vector transfected cells, anti-sense bFGF transfected cells are 3.7 and 4.4 fold more sensitive to killing (FIG. 4A). This significant increase sensitivity of anti-sense bFGF transfected cells was reproducible in six independent pools of transfectants. It is not due to altered growth rate of the anti-sense bFGF transfected cells (compare the number of untreated cells in the control vector transfected pools to the number of untreated cells in the anti-sense bFGF transfected pools) or to a non-specific increase in sensitivity of anti-sense transfected cells to Fas induced killing since anti-sense cDNAs were previously isolated that render transfected cells resistant to Fas induced apoptosis. Complementary to the bioassay experiments, quantitative Southern analysis of[0216]

Pool

1 andPool2 indicates that the abundance of anti-sense bFGF cDNA is reduced by 1.9 fold inPool2 of cells surviving sub-lethal dose of anti-Fas antibody, compared toPool1.

Western blot analysis of control vector transfected cells as well as anti-sense bFGF transfected cells revealed four polypeptides of 24 kD, 22/22.5 kD and 18 kD, while the 34 kD form is not detected by the antibody used (FIG. 4B). Quantitative analysis of the relative level of bFGF forms revealed that in the absence of anti-Fas antibody, expression of anti-sense bFGF results in reduction of approximately 25%-30% in the levels of each of the detected forms (FIG. 4C). Interestingly, in cells treated with anti-Fas antibody a more significant reduction in the levels the 24 kD form and 18 kD is observed, 57% and 66% respectively, while the reduction of the levels of the 22/22.5 kD is not altered. Selective reduction in the level of some of the bFGF forms by an anti-sense fragment that overlaps the coding region of all the bFGF polypeptides can be due to a network of feedback regulation loops as previously reported for some bFGF forms.[0217]

While previous studies has shown that over-expression of the 34 kD form protects cells from serum deprivation induced killing and over-expression of the 24 kD form protects cells from ionizing radiation, here it is demonstrated that bFGF is an inhibitor of Fas induced apoptosis, as identified by AHM.[0218]

The second inhibitor of the Fas pathway that was identified by AHM is the cap-n-collar b-zip transcription factor NF-E2 related factor 2 (NRF2). Nrf2 activates the transcription of phase II detoxifying enzymes such as NAD(P)H quinone oxireductase (NQO1) and Glutathione S-transferase (GST) by direct binding to the Antioxidant Response Element (ARE) in the promoter of these genes. Studies of nNrf2 null mice indicate that Nrf2 is essential for the transcriptional activation of phase II enzymes. NQO1 and GST act in concert with phase I detoxifying enzymes (such as cytochrome p-450 monooxygensase) to mediate the cellular detoxification of xenobiotics. In the absence of Nrf2, this coordinated detoxification is impaired and toxic products from phase I reactions can accumulate. In the AHM screen an anti-sense fragment of Nrf2 corresponding to nucleotide 147 to 970 of the human Nrf2 (gbS74017, Moi et al, 1994,[0219]PNAS 21, 9926) was recovered.

Bioassays of two pools of HeLa cells transfected with anti-sense Nrf2 clearly demonstrates that anti-sense Nrf2 render the cells 4.1 and 5.4 fold more sensitive to Fas induced apoptosis (FIG. 5A). Again, this increased sensitivity is not a result of impaired growth, since there is only limited alteration in the growth rate of anti-sense Nrf2 transfected cells (FIG. 5A). Sensitization by anti-sense Nrf2 was reproducible in seven independent pools of transfectants. Western blot analysis indicated a significant 3.8 fold reduction in the level of Nrf2 protein in the anti-sense Nrf2 transfected cells (FIG. 5B).[0220]

The role of Nrf2 as an inhibitor of the Fas pathway was further validated by pharmacological agents. It is predicted that treatment of cells with Dicumarol, an inhibitor of GST and NQO1 will sensitize cells to Fas induced apoptosis, since Nrf2 up-regulates the levels of GST and NQO1. As shown in FIG. 5C, HeLa cells treated with 100 mM Dicumarol are 2.8 fold more sensitive to Fas induced killing compared to cells treated with vehicle (fold sensitization was calculated as described in FIG. 4A).[0221]

Since down regulation of Nrf2 sensitizes cell to Fas induced apoptosis, it was questioned whether increasing any of the activities induced by Nrf2 will protect HeLa cells from apoptosis. Nrf2 up-regulates GST that conjugates glutathione to the reactive products of phase I detoxification. Increased activity of GST will protect cells from Fas induced apoptosis. GST activity was elevated by treating HeLa cells with the glutathione precursor N-acetyl Cysteine (NAC) that increases the glutathione pool.[0222]

As shown in FIG. 5D, NAC strongly protects HeLa cells from Fas induced apoptosis as previously reported for microglia, neutrophils and T-cells.[0223]

Thus, by using AHM Nrf2 was identified as an inhibitor of Fas induced PCD in HeLa cells and this result was validated by genetic and pharmacological approaches. Interestingly, Ohtsubo et al has recently reported that Nrf2 is cleaved by caspase-3, producing a fragment that acts as a dominant negative fashion and is lethal. The mechanism underlying the mode of action of Nrf2 in regulating apoptosis deserves further investigation.[0224]

A technically simpler alternative to “function profiling” by subtraction is analysis of cDNA microarray. The relative abundance of cDNAs was measured in[0225]

Pool

1 and inPool2 by radio-labeling each pool and hybridizing each of the probes to a cDNA microarray containing approximately 4,000 different known human genes. FIG. 6A is a pseudo colored image of the microarray filter representing for each cDNA spot the ratio of signal generated by hybridization to Pool2 probe to that generated byPool1 probe. Dark green spots indicate cDNAs absent inPool2, representing “sensitizing” anti-sense cDNAs. The corresponding genes are predicted to be survival factors that inhibit Fas induced apoptosis. Dark red spots indicate cDNAs that are enriched inPool2. These genes are positive mediators of killing and their inactivation by anti-sense results in resistance to PCD. The abundance of such anti-sense cDNAs is therefore increased inPool2 that is comprised of cells that survived Fas induced apoptosis. Most of the spots are of intermediate color indicating only modest changes in abundance inPool2 relative toPool1. A histogram representation of the results is shown in FIG. 6B. As seen, the abundance of the majority of cDNAs is not changed. However, a small number of cDNAs are depleted fromPool1 by 2 folds or more. A partial list of these genes is presented in Table A. As predicted, these genes include survival factors. For example, the most depleted cDNA (5.6-fold) corresponds to TNF receptor associated factor 6 that relays a strong survival signal via the activation of NFKB and AKT. Inaddition casein kinase 1 alpha and adenosine A3 receptor have been shown to be survival factors.

In summary, this is a novel powerful tool for identifying signaling inhibitors in human cells. Thus, a large gap in the genetic analyses of mammalian cells has now been filled. AHM can be broadly used to identify inhibitors of any given selectable pathway for the purposes of basic research or clinical applications. Moreover, since it does not require previous knowledge of any sequences, AHM can be employed as a high throughput method of gene discovery and “function profiling” as part of the ongoing effort of deciphering the human genome.[0226]

Materials and Methods:[0227]

AHM: HeLa cells (10[0228]⁶cells/100 mm plate) were transfected with 15 ug of anti-sense cDNA library in pTKO-1 (Deiss and Kimchi 1991) by Superfect reagent (Qiagen). Two days later cells were treated with 200 ug/ml Hygromycin B (Calbiochem-Novabiochem) for two weeks. 2.5×10⁶Hygromycin^Rcells were plated in a 150 mm plate 24 hours prior to treatment with 10 ng/ml anti-Fas antibody (clone CH-11, Kamiya Biomedical Company) (Pool2). Five days post treatment approximately 30-40% of the cells were killed as estimated by microscopic examination. A parallel culture was grown in the absence of anti-Fas antibody (Pool1). After five days, cells were washed twice with PBS, scraped off the plate and stored as pellets at −80°. 100 ul of frozen pellet were lysed by addition of 200 ul of solution P1, followed by 200 ul of solution P2. After the lysate sat on ice for 5minutes 200 ul of solution P3 was added(Qiagen plasmid purification kit). Following 5 minutes incubation on ice, the lysate was centrifuged for 10 minutes at 15,000×g, the supernatant was mixed with an equal volume of isopropanol and centrifuged at 15,000×g for 10 minutes. The DNA pellet was rinsed with 70% ethanol and resuspended in 100 ul of water. The cDNA inserts were amplified by PCR in a 100 ul reaction containing: 1 ul DNA, 200 uM of dATP, dGTP, dCTP, dTTP; 10 mM Tris-HCl pH9.0; 0.1% Triton X-100; 1.0 mM MgCl; 1 unit Taq DNA polymerase (Gibco BRL) and 500 ng each of primers:. These primers are derived from the sequences that flank the cDNA insertion site in the pTKO-1 anti-sense expression vector. The primers are designed to restore a HindlIl restriction site on the promoter proximal side of the cDNA and a BamHI site on the promoter distal side to conserve the orientation of the cDNA fragments upon their cloning in pTKO-1. The reaction was incubated 94° C. for 5 minutes; subjected to 25 cycles of: 94° C. for one minute, 58° C. for one minute and 72° C. for five minutes; followed by 72° C. for seven minutes. The PCR products were cleaved by BamHI and HindIII, purified (Wizard PCR Prep Kit, Promega) and used in subtraction (PCR-Select kit, Clontech). The driver for the subtraction was the product of the PCR reaction derived from the untreated cells (Pool1) and the tester was derived from treated cells (Pool2). The following modifications to the manufacturer's instructions were made: 1. The first step wasIV F 3, since no cDNA synthesis is required. 2. The blunt endsadapters1 and2R were replaced with cohesive ends adapters.

Cohesive end adapters ligate more efficiently to the cDNA and permit the directional cloning of the cDNA inserts. 0.3 ug of the tester was used for adapter ligation. 3. The initial hybridization included 0.9 ug of the driver and 0.03 ug of the adapted ligated tester. The products of the subtraction were cleaved with BamHI and HindIII, purified and cloned into the pTKO-1 between BgIII and HindIII sites. Individual clones were sequenced and transfected into HeLa cells.[0229]

Transfection and Bioassays: HeLa cells (2×10[0230]⁶cells/100 mm plate) were plated 20 hours prior to transfection with either 17 ug of either anti-sense expressing vector or control vector harboring no cDNA insert, by calcium phosphate. Forty eight hours post transfection cells were treated with 200 ug/ml Hygromycin B (Calbiochem-Novabiochem) for two weeks. For bioassays, anti-sense transfected cells or control vector transfected cells (1.6×10⁵cells/ well in 6 wells plates) were plated 20-24 hours prior to the treatment with 200 ng/ml anti-Fas antibody (clone CH-11, Kamiya Biomedical Company). The number of viable, trypan blue (Gibco/BRL) excluding cells that remained attached to the plate following rinsing with PBS was counted 24 hours post treatment

Western analysis: Anti-sense transfected cells or control vector transfected cells (2.5×10[0231]⁶cells/150 mm plate) were plated 24 hours prior to treatment with 200 ng/ml anti-Fas antibody (clone CH-11, Kamiya Biomedical Company). 24 hours post treatment cells were washed with PBS and lysed in RIPA buffer (1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 1 mM PMSF, 2 mg/ml aprotonin and 2 mg/ml pepstatin in PBS). Samples containing 50 mg protein were separated by SDS-PAGE and transferred to nitrocellulose membranes. The immunoblots were probed with either anti-Nrf2 antibody (1:100, Santa Cruz, sc722) or anti-bFGF-2 antibody, (1:200, Santa Cruz, sc 079), incubated with goat anti rabbit conjugated to horseradish peroxidase (Pierce) followed by incubation with SuperSignal substrate (Pierce). Following autoradiography, the probes were stripped off (Amersham, ECL Western blotting protocols) and the membranes were hybridized with anti-actin antibody, (1:100, Sigma A4700 or A2066). The intensities of the bands were quantified by the National Institute of Health Image program.

Treatment with N-acetyl cysteine: HeLa cells (8.3×10[0232]⁴cells /well in 6 wells plates) were plated 20-24 hours prior to treatment with various concentrations of NAC (Sigma/Aldrich,) in the presence or absence of 50 ng/ml anti-Fas antibody (clone CH-11, Kamiya Biomedical Company). The number of viable, trypan blue (Gibco/BRL) excluding cells that remained attached to the plate following rinsing with PBS was counted 5 days post treatment.

Treatment with Dicumarol: HeLa cells, 1.6×10[0233]⁵cells /well were plated in 6 wells plates. 20-24 hours later cells were treated with various concentrations of dicumarol (Sigma/Aldrich,) in 0.2 mM NaOH for 15 minutes prior to the addition of 200 ng/ml anti-Fas antibody (clone CH-11, Kamiya Biomedical Company). The number of viable, trypan blue (Gibco/BRL) excluding cells that remained attached to the plate following rinsing with PBS was counted 17 hours post treatment.

cDNA microarray analysis:. Approximately 500 ng of the PCR products of[0234]

Pool

1 and Pool2 (same preparations that were used for the subtraction, before their cleavage by BamHI and HindII) were labeled with 100 mCi of [³³P] dCTP (3000 Ci/mmole, ICN,) by the random primers DNA labeling system (Gibco/BRL), purified (Aershm/Pharmacia, ProbeQuant G50 micro columns) and individually hybridized to Human GeneFilters (GF211, Research Genetics). The filter was pre-hybridized for 40-60 minutes at 68° C. in ExpressHyb Hybridization solution (Clontech), followed by hybridization for 3-5 hours at 68° C. The filter was washed in 2×SSC, 0.05% SDS at room temperature 3-5 times for 10-15 minutes each time followed by 2 washes for 15 minutes each in 0.1×SSC, 0.1% SDS at 55° C. The image was generated by Molecular Dynamics phospho-imager. In between hybridizations, the probe was stripped off by adding boiling solution of 0.5% SDS and incubating at room temperature for 1 hour. Successful removal of probe was confirmed by phosphor-imager analysis. Images processing and calculation of the ratio of the signals ofPool2 probe to Pool1 probe were performed by Pathways II software (Research Genetics). All the spots that showed significant differential abundance were visually inspected.

Throughout this application, various publications, including United States patents, are referenced by author and year and patents by number. Full citations for the publications are listed below. The disclosures of these publications and patents in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art to which this invention pertains.[0235]

The invention has been described in an illustrative manner, and it is to be understood that the terminology which has been used is intended to be in the nature of words of description rather than of limitation.[0236]

Obviously, many modifications and variations of the present invention are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described.[0237]

References

Altschul et al. (1990) Basic local alignment search tool.J. Mol. Biol., 215:403-410.[0238]

Braun et al. (1996) Molecular and Cellular Biology, (8):4623-4630.[0239]

Cregg et al. (1993) Recent Advances in the Expression of Foreign Genes in Pichia[0240]pastoris. Bio/Technology 11:905-910.

Deiss et al. (1996) Cathepsin D protease mediates programmed cell death induced by interferon-y, Fas/APO-1 and TNF-a. EMBO 15(15):3861-3870.[0241]

Deiss et al. (1995) Identification of a novel serine/threonine kinase and a novel 15-kD protein as potential mediators of the T interferon-induced cell death.[0242]

Genes & Develop. 9:15-30.[0243]

Deiss and Kimchi (1991) A Genetic Tool Used to Identify Thioredoxin as a Mediator of a Growth Inhibitory Signal.[0244]

Science 252:117-120.[0245]

Diatchenko et al. (1996) Proc. Natl. Acad. Sci.USA., 93:6025-6030.[0246]

Gilboa et al. (1986) Transfer and expression of cloned genes using retroviral vectors.[0247]

BioTechniques 4(6):504-512.[0248]

Gubler and Hoffman (1983) Gene (Amst.) 25:263.[0249]

Gudkov et al. (1994) Cloning mammalian genes by expression selection of genetic suppressor elements: association of kinesin with drug resistance and cell immortalization.[0250]

Proc. Natl. Acad. Sci. U.S.A.[0251]_—, 91:3744-3748.

Hirt (1967) Selective extraction of polyoma DNA from infected mouse cell cultures. J. Mol. Biol., pp. 365-9.[0252]

Holzmayer et al. (1992) Isolation of dominant negative mutants and inhibitory antisense RNA sequences by expression selection of random DNA fragments. Nucleic Acids Res . . . , 20:711-717.[0253]

Izant and Weintraub (1984) Inhibition of thymidine kinase gene expression by anti-sense RNA: a molecular approach to genetic analysis. Cell, 36: 1007-1015.[0254]

Kissil et al. (1995) “Isolation of DAP3, a Novel Mediator of Interferon-T-induced Cell Death” J. Biol. Chem. 270(46):27932-27936.[0255]

Levy-Strumpf et al. (1997) DAP-5, a novel homolog of eukaryotic translation initiation factor 4G isolated as a putative modulator of gamma interferon-induced programmed cell death. Mol. Cell. Biol.[0256]_—, 17: 1615-1625.

Li and Cohen (1996) TsglOl: A novel tumor susceptibility gene isolated by controlled homozygous functional knockout of allelic loci in mammalian cells. Cell, 85:319-329.[0257]

Meissner et al. (1987) PNAS (USA) 84:4171.[0258]

Roninson et al. (1995) Genetic suppressor elements: new tools for molecular oncology—thirteenth Cornelius P.[0259]

Rhoads Memorial Award Lecture. Cancer Res., 55:4023-4028.[0260]

Schena et al., 1995, Aiello et al., 1994, Shen et al., 1995, Bauer et al., 1993, Liang and Pardee, 1992, Liang and Pardee, 1995, Liang et al., 1993, Braun et al., 1995, Hubank and Schatz, 1994 Smith et al. (1995) genetic footprinting: a genomic strategy for determining a gene's function given its sequence. PNAS 92:6479-6489.[0261]

Su et al. (1993) LYAR, a novel nucleolar protein with zinc finger DNA-binding motifs, is involved in cell growth regulation. Genes Dev., 7:735-748.Schena et al. (1995) Science, 270:467-470.[0262]

Agrawal, 1996. Antisense oligonucleotides: towards clinical trials, TIBTECH, 14:376.[0263]

Calabretta, et al, 1996. Antisense strategies in the treatment of leukemias. Semin.[0264]

Oncol. 23:78.[0265]

Crooke, 1995. Progress in antisense therapeutics, Hematol.[0266]

Pathol. 2:59.[0267]

Felgner, 1997. Nonviral Strategeies for Gene Therapy.[0268]

Scientific American. June, 1997, pgs 102-106.[0269]

Gewirtz, 1993. Oligodeoxynucleotide-based therapeutics for human leukemias, Stem Cells Dayt. 11:96.[0270]

Hanania, et al 1995. Recent advances in the application of gene therapy to human disease. Am. J. Med. 99:537.[0271]

Lefebvre-d'Hellencourt et al, 1995. Immunomodulation by cytokine antisense oligonucleotides. Eur. Cytokine Netw. 6:7.[0272]

Lev-Lehman et al., 1997. Antisense Oligomers in vitro and in vivo. In Antisense Therapeutics, A. Cohen and S. Smicek, eds (Plenum Press, New York) Loke et al, 1989. Characterization of oligonucleotide transport into living cells. PNAS USA 86:3474.[0273]

Wagner et al., 1996. Potent and selective inhibition of gene expression by an antisense heptanucleotide. Nature Biotechnology 14:840-844.[0274]

Wagner, 1994. Gene inhibition using antisense oligodeoxynucleotides. Nature 372:333.[0275]

Radhakrishnan et al., 1990. The automated synthesis of sulfur-containing oligodeoxyribonucleotides using 3H-1,2-Benzodithiol-3-[0276]One 1,1Dioxide as a sulfur-transfer reagent. J. Org. Chem. 55:4693-4699.