WO1991012343A2 - Detection of point mutations in genes encoding gtp binding proteins - Google Patents

Abstract

The present invention provides a method for detecting whether a point mutation is present in a nucleic acid encoding a GTP bindingprotein or protein subunit. Methods are also provided for characterizing point mutations, if present. In a preferred embodiment, the method involves amplification of a nucleic acid segment followed by sequence-specific probe hybridization. The method is preferred for nucleic acids which encode a G-protein α subunit or p21 ras protein.

Description

DETECTION OF POINT MUTATIONS IN GENES

ENCODING GTP BINDING PROTEINS

This application is a continuation-in-part of copending Serial No.477,260, filed Feburary 7, 1990, which is incorporated herein by reference.

The present invention relates to the identification of point mutations within nucleic acids encoding GTP binding proteins in human samples. Point mutations within GTP binding proteins are associated with malignancies. The invention provides specific primers and probes for the detection and classification of these point mutations and potential oncogenes. The identification of oncogenes is important in the study of cell growth and cardnogenesis. The invention provides methods which relate specific point mutations to specific tumor types. In a preferred embodiment point mutations are described witiiin nucleic acids encoding G-proteins.

G-proteins function as intermediates in transmembrane signalling pathways (Gilman , 1987, Ann. Rev. Biochem.56: 615). These pathways consist of receptors, G-proteins and effector molecules and are regulated by the cyclic association of GTP and GDP with G-proteins. Each G-protein consists of three subunits: α, β, and γ. The specificity of the interaction with the effector molecule is dictated by the α subunit

Several G-proteins have been purified and characterized; Gs and Gi are involved in stimulation and inhibition, respectively, of adenylate cyclase activity. Three Giα subunits Giα1, Giα2, and Giα3 have been identified and cloned (Itoh el al, 1988, J. Biol. Chem.263:6656-6664). Gt activates cGMP phosphodiesterase in response to photosignal transductiori (Mattera et al., 1986. FEBS Lett. 206:36-42, and Didsbury et al., 1987, FEBS Lett 211:160-164). Other G-proteins have been sequenced including Go and Gz (Jones and Reed, 1987, J. Biol. Chem.262:14241- 14249, and Strathmann et al., 1989, Proc. Natl. Acad. Sci. USA: 86:407-7409). Additionally, Gk (Yatani et al., 1987, Science 235:2071 and Golf (Jones and Reed, 1989, Science 244:790) have been identified as G-proteins.

Gs activity raises the level of cAMP in cells by stimulating adenylyl cyclase. In pituitary somatotrophs, cAMP stimulates secretion of human growth hormone and causes cellular proliferation. Recently, a subset of human pituitary tumors were described having elevated levels of growth hormone and cAMP (Vallar et al., 1987, Nature 330:566-568). Landis et al., 1989, Nature 340:692-696, proposed that the abnormal cell proliferation observed by Vallar et al. was the result of a defect in the G-protein responsible for controlling cAMP levels, resulting in an accumulation of cAMP. Landis et al. identified patients harboring tumors secreting excessive amounts of growth hormone and determined that four tumors had constutively elevated levels of Gs activity. RNA was purified from fresh tissue, reverse transcribed, and cloned. The entire Gsα coding region of the cDNA was sequenced and point mutations within codons 201 and 227 were identified. These mutations are Gsp mutations. Gsp mutations are a class of mutations that activate Gs, which normally mediates stimulation by thyrotropin (TSH) of thyrocyte proliferation and production of thyroid hormones. (Lyons et al., 1990, Science.249:655-659)

Arginine 201 is a major site of ADP-ribosylation of Gsα by cholera toxin. This modification allows constitutive adenylyl cyclase activation (Lo and Hughes, 1987, FEBS Lett 224:1-31. Glutamine 227 is predicted to be a Gsα equivalent of glutamine 61 in Eas p21 proteins (Landis et al. supra.). Mutational replacement of Gln-61 in 12s p21 produces a protein that promotes malignant transformation (Der et al., 1986, Cell 44:167-176).

The ras genes encode highly related proteins approximately 21,000 daltons in molecular weight (p21s). While the exact functions of these proteins in cellular signalling pathways remains elusive, the p21s have GTPase enzymatic activities and interact with a GTPase activating protein (GAP) (Bishop, 1983, Ann. Rev. Biochem. 52:301-354, and McCormick, 1989, Cell 56:5-8).

The human ras gene family, which includes the closely related Ha-, Ki-, and N- ras genes, is one of the potential targets for mutational changes that have been implicated in the development of many human malignancies (Bos, 1988, Mutation Research 195:255-271). These alterations are either point mutations in codon 12, 13, or 61, or alternatively a 5- to 50- fold amplification of the wild-type gene. These changes convert the ras proto-oncogenes into oncogenes.

Polymerase chain reaction (PCR) methods have been used to detect known point mutations in ras oncogenes in genomic DNA isolated from tumors (Verlaan-de Vries et al, 1986, Gene 50:313-320, and Almoguera et al., 1988, Cell 5:549-554). Farr et al, 1988, Proc. Natl. Acad. Sci. USA 85:1629-1633. have combined PCR with oligonucleotide dot blot methods to examine specific ias gene point mutations in DNA isolated from patients afflicted with acute myeloid leukemia (AML).

The present invention provides methods for screening nucleic acids encoding G-proteins. Methods for screening ras genes are also provided. The nucleic acids may be RNA or DNA. Primers and probes are provided which aid in the identification of potential oncogenes and characterization of point mutations within an oncogene or potential oncogene. The invention provides primers and probes which are particularly suitable for detection of point mutations in nucleic acids encoding Gz, Gs, Go, Ga, and Gi proteins in endocrine tumors.

The present invention provides a method for detecting whether a point mutation is present in a nucleic acid encoding a G-protein α subunit, in a sample, that comprises: (a) hybridizing a G-protein α subunit probe to said sample and (b) determining whether hybridization has occurred.

In another embodiment, the method comprises a method for detecting point mutation, if present, in a nucleic acid encoding a G-protein α subunit in a sample, comprising:

(a) treating the sample with a G-protein α subunit primer pair, an agent for polymerization, and deoxynucleoside 5' triphosphates under conditions such that an extension product of each primer can be synthesized, wherein said primers are sufficiently complementary to separate strands of a nucleic acid encoding a segment of a G-protein α subunit to hybridize thereto so that the extension product synthesized from one member of said pair, when separated from its complementary strand, can serve as a template for synthesis of the extension product of the other member of said pair;

(b) separating the primer extension products from the templates on which the extension products were synthesized to form single-stranded molecules;

(c) treating the single-stranded molecules generated in step (b) with the primers of step (a) under conditions such that a primer extension product is synthesized using each of the single-stranded molecules produced in step (b) as a template;

(d) repeating steps (b) and (c) at least once to provide amplified DNA;

(e) hybridizing a G-protein α subunit probe to said amplified DNA, wherein said probe contains a nucleic acid sequence that will hybridize to a sequence, selected from a wild type and mutant nucleic acid sequence, within said amplified DNA; and

(f) determining if hybridization has occurred.

The present invention provides novel primers and probes useful for detecting potentially oncogenic point mutations within a nucleic acid encoding a G-protein α subunit.

The present invention also provides kits for amplifying and detecting point mutations, if present, within a nucleic acid encoding a G-protein α subunit

Figure 1 shows the identification of point mutations in Gsα and Giα2 genes. A region containing the indicated codons of the Gsα or Giα2 gene was amplified by PCR from genomic DNA isolated from either fresh frozen tissue or paraffin-embedded tissue. Point mutations were detected with high-stringency hybridization of sequence- specific oligonucleotides to the amplified product Each panel represents hybridization with a different oligonucleotide.

Analysis of Gsα genes is shown in Figure 1A and described in Example 2.

This analysis identified mutations in Arg 201 and Gin 227 codons of Gsα in 18 biochemically-characterized human growth hormone secreting pituitary tumors. Analysis of Giα2 genes is shown in Figure 1B and described in Example 3. This analysis identified mutations in codon 179 of Giα2 in three human adrenocortical tumors and one ovarian granulosa cell tumor.

Figure 2 provides the GVA (gel visualization assay) analysis of ias RNA/PCR products from a normal human spleen and the K562 cell line as described in

Example 5.

Figure 3 shows the results of a southern blot analysis of ias RNA/PCR products using ASO probes specific for activating point mutations in characterized cell lines. The experiment is described in Example 6.

Figure 4 provides the GVA analysis of ias RNA/PCR from alcohol-fixed, paraffin-embedded samples as described in Example 7.

Figure 5 provides the GVA analysis of ras RNA/PCR from stained and unstained microscope slides of human bone marrow as described in Example 8.

Figure 6 is schematic diagram of ras format II filters showing the positions of wild type and mutant ras oligonucleotide probes as described in Example 9.

The present invention provides a method for detecting and characterizing point mutations, if present, in a nucleic acid encoding a G-protein α subunit The point mutations detected by these methods are believed to be involved in oncogenesis. The nucleic acid is a G-protein subunit gene, RNA transcription product, cDNA product or a subsequence thereof. The method involves amplifying by a polymerase chain reaction, a segment of nucleic acid encoding at least one G-protein amino acid of interest For each nucleic acid segment suspected of comprising a potentially oncogenic point mutation, a pair of oligonucleotide primers are provided for amplification in a polymerase chain reaction. The primers will amplify wild type or mutant nucleic acid segments. Genes that encode G-proteins are proto-oncogenes in the normal somatotrophic state and are referred to as wild type. If a point mutation is present the gene is a putative oncogene. Thus, the present methods, primers, and probes allow one skilled in the art to distinguish between these two types of G-protein genes.

In the embodiment of the invention illustrated below, point mutations are detected by oligonucleotide probes. To detect an oncogenic point mutation in a nucleic acid encoding a G-protein, the probes contain either the wild type nucleic acid sequence or single nucleotide changes within codons which correspond to amino acids 49, 201, and 227 in the sequence of Gsα. The amino acid at 226 may be of interest as well. These single nucleotide changes affect the translation product of the gene. Of course, in some samples, a mutated oncogene may not be present. In such a sample only the wild type probe will hybridize to the sample. Thus, where a point mutation is not detected, the wild type probe serves to verify the-presence of the amplified product in the sample.

G-protein α subunits share a high degree of homology. Matsuoka et al., 1988, Proc. Natl. Acad. Sci. USA 85:5384-5388, provide a comparison of the amino acid sequences of several Gα subunits. Although the Gα subunits vary in length, the degree of homology between the published amino add sequences is sufficiently high such that sequence alignment is possible. In this way, the amino add corresponding to Gsα Gly 49, Gsα Arg 201, Gsα Gly 226, and Gsα Gin 227 can be determined for all Gα subunits. For example, in Giα2, Arg 179, arid Gin 205 correspond to Arg 201 and Gin 227 in Gsα.

The present invention also provides probes which detect point mutations in nucleic acid encoding as proteins. Such probes contain single base changes as well as the wild type sequences for codons encoding amino acids at p21 positions 12, 13, and 61.

The examples below describe methods to detect the presence of a point mutation within a nucleic add sequence contained in a sample. For example, the sequence to be examined can be assayed by hybridization using a wild type probe and a pool of probes, each probe containing a point mutation at a specific position. Alternatively, the methods can be used to detect and classify a point mutation. In this case, the probes, one comprising the wild type sequence and the others comprising point mutations which affect the translation product at a specific position, are used individually.

According to the method, only the probe comprising the exact nucleic acid sequence contained within the sample nucleic acid will hybridize to the nucleic acid in the sample. Those skilled in the art will recognize that the disclosed methods and probes enable new oncogenes to be detected and classified. These methods led to the discovery that for example, the gene encoding Giα2 is a proto-oncogene. The present invention has led to the discovery of new point mutations that may give rise to cancer. In addition, the invention provides primers and probes that can be used to identify oncogenic point mutations in genes encoding Giα1, Giα3, Goα, and Gzα if such mutations exist Similarly, these methods may be used to design primers and probes to identify new oncogenic point mutations in other G-protein α subunits such as Gtα, Gkα, and Golfα. These point mutations may well be related to oncogenesis and their identification lays a critical foundation for experimental research in oncogenesis.

Regardless of oncogenidty, these point mutations are useful to discriminate between individuals. Methods which discrώinate between individuals based on nucleic acid sequence differences are used in, for example, forensic medicine. The disclosed examples of the invention, relating to G-proteins, provide methods, primers, and probes to detect and classify point mutations in nucleic adds encoding Gsα and Gi2α. These methods are also suitable for detecting point mutations, if present and classifying those mutations in any nucleic acid encoding a G- protein α subunit Tables 1-4 provide primers and probes suitable for detecting and classifying point mutations in nucleic adds encoding Gsα, Giα1, Giα2, Giα3, Goα, and Gzα. These methods are directly applicable to the detection of point mutations in other Gα subunits, for example, Gtα, Gk, or Golf. Table 4 provides probes which correspond to codons 49, 201, and 227 in the Gsα subunit.

In a preferred method for detecting and classifying point mutations within oncogenes or proto-oncogenes, the nucleic acid containing the region of interest is amplified by a polymerase chain reaction prior to detection. Primers useful in these methods are suitable for amplification of proto-oncogenes as well as activated protooncogenes. Those skilled in the art will recognize that with the disclosed methods, primers, and probes, new oncogenes can be readily detected. For example, the primers and probes of the invention led to the discovery of novel point mutations and, thus, a novel putative oncogene, Giα2. Novel oncogenic genes encoding Giα1, Giα3, Goα, and Gzα may also be discovered by the methods and compositions provided. Primers of use in the present invention hybridize to genomic DNA at sites such that in a PCR reaction, the primers amplify a specific region of the G-protein DNA. The specific region amplified comprises at least one codon which corresponds to Gly 49, Arg 201, or Gin 227 in Gsα.

In one embodiment of the present invention, primers are selected such that the resultant amplified fragment comprises both codons 201 and 227 (or correspondingly 179 and 205 in Giα2). However, this is not an essential aspect of the invention. If, for example, a large intron or more than one intron exists between these codons in the genomic DNA, or if the DNA is degraded as in some paraffin embedded tissue, amplification primers are designed for the region comprising codon 201 and separately for the region comprising codon 227. Such amplification reactions are run separately or simultaneously in one reaction vessel.

Amplification requires the use of primer pairs that will amplify a discrete region of DNA present in a sample. These primer pairs are oligonucleotides. The PCR products generated from these primers are then analyzed by hybridization with sequence specific probes. Sequence specific probes may also be allele-specific probes. An allele-specific probe (ASO) will hybridize to an allele-specific sequence in a nucleic acid within a sample. An allele-specific sequence is a component of an individual's genotype. A sequence-speάfic probe comprises a specific sequence which may or may not exist as an allele. Thus, until a sequence is identified in at least one individual, probes which are sequence-specific are not necessarily allele-specific. However, as the term allele-specific probe is used by those of skill in the art these terms are used interchangeably herein.

Amplification of DNA by PCR is disclosed in U.S. Patent Nos.4,683,195 and

4,683,202 (both of which are incorporated herein by reference). Methods for amplifying and detecting nucleic adds by PCR using a thermostable enzyme are disclosed in U.S. Patent No. 4,965,188, which is incorporated herein by reference. PCR amplification of DNA involves repeated cycles of heat-denaturing the DNA, anneahng two ollgonucleotide primers to sequences that flank the DNA segment to be amplified, and extending the annealed primers with DNA polymerase. The primers hybridize to opposite strands of the target sequence and are oriented so that DNA synthesis by the polymerase proceeds across the region between the primers, effectively doubling the amount of the DNA segment Moreover, because the extension products are also complementary to and capable of binding primers, each successive cycle essentially doubles the amount of DNA synthesized in the previous cycle. This results in the exponential accumulation of the spedfic target fragment, at a rate of approximately 2n per cycle, where n is the number of cycles.

In the disclosed embodiment, Taq DNA polymerase is preferred although this is not an essential aspect of the invention. Taq polymerase, a thermostable polymerase, is active at high temperatures. Methods for the preparation of Taq are disclosed in U.S. Patent No.4,889,818 and incorporated herein by reference.

The choice of primers for use in PCR determines the specificity of the amplification reaction. In the present invention, primers are used that will amplify G- protein or ras p21 sequences present in a sample. The primers of the invention can include degenerate primers. These are mixtures of oligonucleotides synthesized to have any one of several nucleotides incorporated at a selected position during synthesis. For example, the primers may be sufficiently complementary to all types of ras genes to amplify a DNA sequence of any rasDNA present in the sample. Illustrative primers of this type are referred to, for example, as "pan" ras primers. The primers are designed to amplify a region of DNA, cDNA, or RNA that contains sequences specific to any ias p21 gene: c-N-ras, c-Ha-ras, or c-Ki-ras. Because the "pan" ras primers span large intron sequences, cDNA and RNA templates are preferred. The amplified DNA can therefore be used to classify the ras gene present in the sample.

Alternatively, it may be desirable to use primer pairs which are specific for each gene to be detected. Such a primer pair will amplify a specific DNA or RNA segment encoding, for example, c-Ki-ras, c-N-ras, c-Ha-ras, Giα1, Giα2, Giα3, Gsα, Goα, or Gzα. In one embodiment, three separate pairs of ras primers are included in one PCR reaction such that each pair will specifically amplify either c-N-ras, c-Ki-ras, or C- Ha-ras. The primers are designed so that each PCR product has a discrete size. Thus, three primer pairs are used to simultaneously amplify several DNA or RNA segments. The identity of the segments) amplified can then be determined by, for example, gel electrophoresis and size determination. The presence of point mutations and the classification of such mutations can be subsequently determined by the methods provided.

When more than one nucleic acid segment is characterized, it is not essential that primers are designed such that the resulting amplification products are of different sizes. Amplification reactions using different primer pairs can be run independently of one another and analyzed simultaneously, for example, using individual lanes on an acrylamide gel. Alternatively, several primer pairs can be used simultaneously in one reaction, and the amplification products divided and analyzed to characterize the sample by, for example, separate probe hybridizations.

In another embodiment of the present invention, nested primers are used (Mullis et al., 1986, Cold Spring Harbor Symposium on Quantitative Biology 51:263, incorporated herein by reference). This method may be preferred when the amount of nucleic add in a sample is extremely limited, for example, where archival, paraffin embedded samples are used. When nested primers are used, the nucleic acid is first amplified with an outer set of primers. This amplification reaction is followed by a second round of amplification cycles using an inner set of primers.

Once a sample has been treated with a primer pair under conditions suitable for PCR, the method of the invention requires in a preferred embodiment that the amplified product is characterized. It may be preferred, but is not essential in the practice of the invention, to determine whether amplification has occurred. The use of an internal amplification control to assure the competency of a sample for PCR is within the scope of the invention and reduces the likelihood of false negative results. There are a variety of ways to determine whether amplification has occurred. A portion of the reaction mixture can be subjected to gel electrophoresis and the resulting gel stained with ethidium bromide and exposed to ultraviolet light to observe whether a product of the expected size is present. Labeled primers or deoxyribonucleotide 5'-triphosphates can be added to the PCR reaction mixture, and incorporation of the label into the amplified DNA measured to determine if amplification occurred. Another method for determining if amplification has occurred is to test a portion of PCR reaction mixture for ability to hybridize to a labeled ollgonucleotide probe or mixture of probes designed to hybridize to only the amplified DNA. Alternatively, the determination of amplification and characterization of a point mutation can be carried out in one step by testing a portion of the PCR reaction mixture for its abillty to hybridize to one or more specific probes.

Due to the enormous amplification possible with the PCR process, small levels of DNA carryover from samples with high DNA levels, positive control templates or from previous amplifications can ϊesult in PCR product, even in the absence of purposefully added template DNA. If possible, all reaction mixes are set up in an area separate from PCR product analysis and sample preparation. The use of dedicated or disposable vessels, solutions, and pipettes (preferably positive displacement pipettes) for RNA/DNA preparation, reaction mixing, and sample analysis will minimize cross contamination. See also Higuchi and Kwok, 1989, Nature.339:237-238 and Kwok, and Orrego, in: Innis et al. eds., 1990 PCR Protocols: A Guide to Methods and Applications, Academic Press, Inc., San Diego, CA, which are incorporated herein by reference.

One particular method for minimizing the effects of cross contamination of nucleic acid amplification is described in U.S. Serial No. 609,157, filed

November 2, 1990, which is incorporated herein by reference. The method involves the introduction of unconventional nucleotide bases, such as dUTP, into the amplified product and exposing carryover product to enzymatic and/or physical-chemical treatment to render the product DNA incapable of serving as a template for subsequent amplifications. For example, uracil-DNA glycόsylase will remove uracil residues from PCR product containing that base. The enzyme treatment results in degradation of the contaminating carryover PCR product and Serves to "sterflize" the amplification reaction.

The present invention has led and will continue to lead to the discovery of many previously unknown or uncharacterized oncogenes. For example, using the methods, primers, and probes of die invention, four clinical samples examined were discovered to contain two different Giα2 oncogenes. These new oncogenes are an important aspect of the present invention, as are the point mutations that distinguish these from wild type, and the probes that hybridize to these gene sequences in specific fashion.

In one embodiment, the present invention provides a number of probes for use in detecting and characterizing potential oncogenes. These probes are set forth in Table 2, below. Those skilled in the art will recognize that although the specific primers and probes of the invention exemplified herein have a defined number of nucleotide residues, one or more nucleotide residues may be added or deleted from a given primer or probe typically without great impact on the suitability of that primer or probe in the present methods. The essential aspect of these probes is their abillty to discriminate between wild type and mutant sequences. When a portion of the PCR reaction mixture contains DNA that hybridizes to a probe, the sample contains DNA comprising the wild type or mutant sequence according to the specific sequence of the probe.

An important aspect of the present invention relates to detecting the novel probes provided for use in the present methods. There are a number of ways to determine whether a probe has hybridized to a DNA sequence contained in a sample. Typically, the probe is labeled in a detectable manner, the target DNA (i.e., the amplified DNA in the PCR reaction buffer) is bound to a solld support, and

determination of whether hybridization has occurred simply involves determining whether the label is present on the solld support This procedure can be varied, however, and works just as well when the target is labeled and the probe is bound to the solid support.

The hybridization probes disclosed herein are sequence-specific ollgonucleotide probes for each G-protein α subunit codon to be characterized. As described above, sequence-specific probes are similar to allele-specific probes in use and utillty.

Methods for utillzing ASOs are described in Saiki el al, 1986, Nature 324: 163-166, incorporated herein by reference. The probes may be used individually for detecting, for example, a wild type sequence. Alternatively, the probes may be used in a panel format for characterizing a tumor genotype. The tumor genotype may be compared to, for example, somatic tissue or other tumor types.

The probes can be used in a variety of different hybridization formats.

Although solution hybridization of a nucleic add probe to a complementary target sequence is clearly within the scope of the present invention, commerdalization of the invention will likely result in the use of immobilized probes and thus a quasi "solid- phase" hybridization. In this format the probe is covalently attached to a solld support and target sequences are hybridized with the probe. A preferred method for immobilizing probes on solld supports is disclosed in U.S. patent application S.N. 347,495, filed May 4, 1989, incorporated herein by reference. According to this method, sequence-specific probes are attached to a solid support by virtue of long stretches of T residues which are added during probe synthesis on an automated synthesizer after the hybridizing sequence is synthesized.

For example, in a fixed probe format, the following probes are useful for detecting point mutations in a gene encoding Gsα at amino add positions 201 and 227.

A fixed probe format is suitable for detecting other gsp mutations, as well, and is demostrated in Example 5 for detecting point mutations in as gene PCR products.

Many methods for labeling nucleic acids, whether probe or target, are known in the art and are suitable for purposes of the present invention. In one embodiment illustrated below the probes were labeled with radioactive phosphorous ³²P, by treating the probes with polynucleotide kinase in the presence of radiolabelled ATP. However, other non-radioactive labeling systems may be preferred, i.e., horseradish peroxidase- avidin-biotin systems. Horse-radish peroxidase (HRP) can be detected by its abillty to covert diaminobenzidine to a blue pigment A preferred method for HRP-based detection uses tetramethyl-benzidine (TMB) as described in Clin. Chem.33:1368 (1987). An alternative detection system is the enhanced chemiluminescent (ECL) detection kit commercially available from Amersham. The kit is used in accordance with manufacturer's instructions. A variety of alternative dyes and chromogens and corresponding labels are available for nucleic acid detection systems (see, e.g., U.S. patent appllcation S.N. 136,166, filed December 18, 1987).

Another non-radioactive alternative detection method uses terminal transferase (Tdt) and biotinylated dUTP to add homopolymer tails to the oligonucleotide probes. Biotin serves as the detectable moiety. Following probe hybridization, the filters are washed as usual. Hybridized biotin is detected with strep-avidin conjugated HRP (Se-eqence^® available from Cetus) according to manufacturer's instructions. The ECL system is then used to visualize the biotin-HRP product.

Probes are typically labeled with radioactive phosphorous ³²P, by treating the probes with polynucleotide kinase in the presence of radiolabeled ATP. However, for commercial purposes non-radioactive labeling systems may be preferred, such as, horseradish peroxidase-avidin-biotin or alkaline phosphatase detection systems. HRP can be used in a number of ways. For example, if the primer or one or more of the dNTPs utilized in a PCR amplification is labeled (for instance, the biotinylated dUTP derivatives described by Lo et al., 1988, Nuc. Adds Res.16:8719) instead of the probe, then hybridization can be detected by assay for the presence of labeled PCR product. In a preferred embodiment, probes are biotinylated and detected with the ECL system described above. For example, biotinylated probes are prepared by direct biotinylation of the ollgonucleotide rather than incorporation of biotin-dUTP during PCR. For 5' biotinylation of ollgonucleotides direct solld phrase synthesis using biotin containing phosphoramidites is done according to Alves et al., 1989, Tetra. Let 30:3098; Cocuzza, 1989, Tetra Let.30:6287; and Barabino et al., 1989, EMBO J. 8:4171. Solid phase synthesis of biotinylated ollgonucleotides at any internal or terminal (5' or 3') position is also suitable for preparing biotinylated primers and probes (Pieles et al., 1989, NAR 18:4355, and Misiura et al., 1989, NAR 18:4345). Alternatively, probes and primers are conjugated to HRP, for example, by the method disclosed in WO89/2932, and Beaucage et al., 1981, Tetra. Lett.22:1859-1862. These references are incorporated herein by reference.

Those skilled in the art will recognize that with the above description, primers, and probes for amplifying, detecting, and characterizing new G-protein point mutations can be readily obtained. It will also be readily apparent to those skilled in the art that the specific primers and probes provided in the examples are merely illustrative of the invention. Primers and probes of the invention can also be prepared to amplify and detect sequence variations within areas G-protein sequences other than those specifically exemplified herein, for example, codons corresponding to Gsα 49.

The method of the present invention is appllcable for detecting a point mutation within a gene encoding a GTP binding protein or the expression product of that gene. Thus, the method can detect specific point mutations in a sample containing RNA, or DNA, or both. If the sample contains RNA, the nucleic acid will be reverse transcribed, providing a double-stranded DNA template prior to amplification.

Procedures for reverse transcribing RNA are known (see Maniatis et al., 1982, Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY). In one embodiment, the methods are used to detect point mutations in RNA.

Alternatively, if RNA is abundant, the probes (or the complement of a probe sequence) are suitable for direct detection and characterization of an oncogene or protooncogene mRNA suspected of being present in the sample or in the reverse transcribed cDNA product Similarly, if DNA is abundant, for example, as in a fresh tissue sample, the probes are useful for direct detection of gene sequences. Thus, amplification by PCR is not an essential component of the present invention.

Samples suitable for analysis by the methods described may be fresh or archival. Fresh samples may be, for example, biopsied tumor, tissue samples, or blood. Archival samples may be, for example, frozen or paraffin embedded. In one embodiment of the invention, paraffin-embedded samples are analyzed using G-protein primers and probes. In another embodiment, methods are provided for ias oncogene detection in RNA purified from surgical biopsy samples or cultured cells. Methods are also provided for extracting RNA from air-dried bone marrow sides and alcohol-fixed paraffin embedded tissues. For paraffin embedded tissues containing intact DNA Table 2 provides primers and conditions for amplifying potential oncogenic sites in Gi₃, Gi₂, Gsα, Gi₁, Gz, and Go.

In some instances DNA samples are less well preserved and, consequendy, amplification of long segments is problematic. Example 1 describes primers for amplifying short segments of DNA that include oncogenic regions of gsp.

Thus, the present invention provides methods for detecting activated oncogenes at the genomic (DNA) level. The methods are also suitable for monitoring the expression of proto-oncogenes and oncogenes. The level of oncogene expression can be monitored during and following, for example, chemotherapeutic treatment

Remission and recurrence of tumors can be monitored as well. Comparing the activation of proto-oncogenes into oncogenes at the genomic level, with the expression of activated oncogenes at the mRNA level, provides valuable information for analysis of cardnogenesis and therapies. These methods may also provide information regarding a predipostion to specific mallgnanάes.

The invention provides needed tools useful in analysis of cardnogenesis events. Probes are provided for determining allellc dominance for a specific oncogene. For example, probes specific for Giα2 alleles can be used to analyze somatic cells as well as tumor cells. These probes can also distinguish, in the case of a tumor comprising one mutant and one wild type allele, the relative abundance of the mRNA products of these alleles. One skilled in the art will recognize the utillty of such analyses following study of and studying mutagenic events.

In another aspect, the primers and probes described provide a method for phenotyping a cell. The cell may be a tumor cell or a somatic cell. Analysis using the methods of the invention provides information relating to the proto-oncogene and oncogene profile of a cell. In this way, events related to the presence of, for example, more than one G-protein point mutation, may be discerned which were previously undetectable.

These methods provide a means for associating a specific mallgnancy with a specific oncogene or point mutation. For example, of 306 samples for 15 tumor types analyzed for Gsα, 18 point mutations were detected; all 18 Gsα oncogenes were detected in pituitary adenomas and thyroid tumors. In another example using Giα2 primers and probes, 254 samples representing 14 different tumor types were analyzed. Four tumors of two types had Giα2 oncogenes; one ovarian granulosa cell tumor and three adrenal cortical tumors. Analyses using the methods and probes provided has led to the discovery that the proto-oncogenes encoding G-protein α subunits are activated in endocrine tumors. These studies suggest that distribution of Gsα and Giα2 oncogenes is restricted among specific endocrine target cells.

An analysis of G-protein oncogenes and tumor specificity provides information of useful in determining the role of G-proteins, where that role is as yet undefined. For example, the data suggests that corticotropin (ACTH), which stimulates cσrtisol secretion via Gs, adenylyl cyclase, and cAMP, does not utillze Gs and cAMP to stimulate proliferation of ACTH target cells. Tumors derived from adrenal cortical cells (the target cell of ACTH) do not harbor the Gsα oncogene. Those skilled in the art will recognize that the methods of the present invention provide essential tools for identifying new oncogenes, such as Giα2 (alternatively referred to herein as gip2) and exploring the diverse mix of signalling pathways that mediate regulation of prolif eration in endocrine target cells.

The present invention also provides a number of previously unknown point mutations. These sequences encode corresponding mutant proteins and can be used to synthesize novel, mutant, proteins. Such proteins, or protein subunits or

subsequences, can be used to generate antibodies useful in the detection of mutant G- proteins. These antibodies would provide important tools for screening, for example, biopsied tissue to detect mutant G-proteins, and thus provide important information regarding the genetic make-up of an individual or the carcinogenic state of the sampled tissue.

A mutant G-protein may be dominated in vivo by a normal G-protein unless carcinogenesis is triggered by other events. Thus, antibodies enabled by the present invention would also find use in, for example, screening transplantation tissue as an indicator of potential oncogenic compllcations.

It will be apparent to those skilled in the art that the method of the present invention is amenable to commercialization as a kit for the quantitation of one or more nucleic acids in a sample. For example, in its simplest embodiment such a kit would provide an ollgonucleotide primer pair for amplification of a G-protein α subunit segment and corresponding wild type ollgonucleotide probe. another e Imnbodiment, a kit may contain an array of G protein probes fixed onto a solld support and a corresponding primer pair for amplifying and detecting oncogenic point mutations in genes encoding G protein subunits. anoth Ienr embodiment, a kit may contain an ollgonucleotide primer pair, corresponding G-protein α subunit wlld type and mutant probes, a DNA polymerase, a RNA polymerase, a reverse transcriptase, nucleotide triphosphates, restriction enzymes, and buffers for carrying out cDNA synthesis, restriction enzyme digests, and amplification by PCR. Further, the kits may contain a thermostable DNA polymerase; for example, the thermostable DNA polymerase Taq isolated from Thermus aquaticus as an agent of polymerization.

To facilltate the understanding of the invention, brief definitions are provided below. "Proto-oncogene" refers to the wlld type form of gene encoding a protein in which a point mutation affecting the amino add sequence encodes a protein which may have a carcinogenic or tumorigenic effect.

"Oncogene" refers to a proto-oncogene containing a point mutation and encoding protein which may have a carcinogenic or tumorigenic effect. Oncogenes may alternatively be referred to herein as "activated proto-oncogenes."

"G-protein α subunit primers" refer to primer pairs which hybridize to complementary strands of a nucleic add encoding a G-protein α subunit and will function in a PCR reaction to amplify a nucleic add segment comprising one or more codons suspected of harboring a point mutation. In the disclosed embodiment, these codons encode amino acids corresponding to Gsα amino acids 49, 201, and 227. However, those skilled in the art wffl recognize that potentially oncogenic point mutations may exist at other positions as well. G-protein α subunit primers can readily be designed to amplify such regions according to the methods described herein.

"G-protein α subunit probe;" "G-protein probe," or "probe" as used herein refers to an ollgonucleotide probe designed to characterize the nucleic add sequence encoding an amino acid at a position suspected of containing a point mutation. In the preferred embodiment of the present invention, individual probes comprise the wild type nucleic acid sequence, or a nucleic add containing a point mutation, within a specific codon. Specific codons include any codon within a nucleic add encoding a G- protein, which when replaced with a non-wild type sequence results in a G-protein oncogene. In the present examples, the specific codons include for each G-protein α subunit, codons encoding the amino acids corresponding to Goα amino acid positions 49, 102, and 227.

For any specific codon and any specific G-protein, one wlld type probe and nine point mutation probes may bfc designed and used in the present methods.

However, it is not an essential aspect of the invention that probes be included for practice of these methods. In fact, the use of only one probe may be sufficient to discriminate between two individuals or between the presence and absence of a point mutation.

The following examples provide an illustration of the present invention. They are not a limitation to the scope of the invention. Those skilled in the art will recognize that the primers and probes disclosed can be modified, for example, by altering the length of an ollgonucleotide without altering the purpose and effectiveness of the described invention. Example 1

Materials and Methods

A. Source of Tumors

Human pituitary tumors specimens were supplied by Charles Wllson

(University of Callfomia, San Frandsco) and Anna Spada (Milan, Italy) provided biochemically characterized pituitary samples. Additional samples were provided by Hans Feichtinger and Kurt Grϋnewald at the University of Innsbruch, Austria, and Claudia Landis, Griffith Harsh, Quan- Yang, Dun and Orlo Clark at the University of California, San Frandsco. B. Sample Preparation

For isolation of genomic DNA from paraffin-embedded tissue, 3-5 adjacent 5 mm sections were cut from paraffin blocks and mounted on glass slldes (Wright et al. PCR Protocols: A Guide to Methods and Applications eds. M. Innis, D. Gelfand, J. Sninsky, and T. White, Academic Press, San Diego, pp. 153-158, incorporated herein by reference). One sllde was stained with hematoxylin and eosin and used as a guide to select a region composed entirely of tumor on the other slldes. With a razor blade, excess paraffin and unwanted tissue were removed from the unstained slldes, and the remaining tumor tissue was scraped into a sterile 1.5 ml microcentrifuge tube. To remove contaminating paraffin, the sample was incubated with 5 mis of octane (anhydrous, Aldrich) or Hemo-De (Fischer) at room temperature for 30 minutes with shaking. The tissue sample was pelleted by centrifugation (5 minutes, 1000X g) and the supernatant was discarded. The tissue sample was extracted 2X with 500 mis absolute ethanol to remove traces of octane and then vacuum dried. To digest the tissue and release the genomic DNA, the sample was treated with 0.2 mg/ml proteinase K in 100 mis digestion buffer (50 mM Tris, pH 8.5; 1 mM EDTA; 0.5% Tween 20) at 37°C overnight The sample was centrifuged to remove undigested debris and the

DNA-containing supernatant was incubated at 95°C for 8 minutes to denature proteolytic enzymes and nucleases.

Fresh frozen samples were prepared according to Verlaan-de Vries et al., 1986, Gene 50:313.

C. Amplification Procedure

Nested amplification primers were used in the PCR amplification procedure to improve specificity and yield (Mullls et al., supra). Genomic DNA (100-500 ng DNA from fresh tissue or 10 mis of the DNA solution from paraffin-embedded tissue) was first amplified with 30 pmols of the outer primers (see Table 1) in 100 mis of 0.1 mM dNTPs; 50 mM KCl; 20 mM Tris, pH 8.3; 2.5 mM MgCl2; 100 g/ml BSA; and 1.5 units Taq polymerase (Perkin-Elmer Cetus). A thermocycler (Peririn-Elmer Cetus) was used for amplification. The amplification program was: 5 minutes at 95°C followed by 30 cycles of 1 minute at 95°C, 2 minutes at 50°C, and 2 minutes at 72°C. A second amplification reaction with 30 pmols of the inner primers was done using 2 μls of the initial amplification mixture in the same dNTP and buffer conditions as above and 0.5 units Taq polymerase. The program for the 2nd amplification reaction was: 30 seconds at 94°C, 40 seconds at 57°C and 45 seconds at 72°C. Five μls of the final product was sized on a 2% Nusieve, 1% Seakem agarose gel and visuallzed by ethidium bromide staining. A 526 base pair fragment was obtained for Gsα and a 504 bp fragment for Giα2.

P. Dot Blot Procedure

Nylon filters (Pall Biodyne-B, 0.45 mm) were briefly rinsed in water and mounted on a Bio-RAD dot-blot apparatus. Four mis of each final amplification product were denatured in 0.4 N Na OH and 25 mMEDTA for 5 minutes and spotted on the filter. The DNA was crosslinked to the filter using a Stratalinker (Stratagene) set at auto crosslink. The filters were prehyhridized in 5X SSPE and 0.5% SDS at 50°C for 30 minutes. One ng of a ³²P end-labeled ollgonudeotide was added to the prehybridization solution and incubated at 50°C for 45-60 minutes. The hybridized filters were washed briefly in 2X SSPE and 0.1% SDS at room temperature, followed by a 10 minute incubation in 3 M tetramethyl ammonium chloride, 0.2% SDS, and 50 mM Tris, pH 8.0 (TMACI) at the following temperatures: for Gsα codon 201, 64.5°C; Gsα codon 227, 67°C; Giα2 codon 179, 61.5°C; Giα2 codon 205, 67.5°C. When other probes are employed, hybridization is carried out as described above. The filters are then washed in TMACI at 58°C for 10 minutes. The wash temperature is adjusted in 1°C increments until only the wild type and mutant signals can be detected. In this way, the appropriate wash temperature is determined. Under these conditions, the chosen temperature allows only fully complementary hybrids to stay formed, resulting in a positive dot on a filter. The filters were exposed to Kodak X-AR film for 2-6 hours at -70°C with intensifying screens. For subsequent hybridization with different ollgonucleotides, the nylon filters were stripped by boiling the filters for five minutes in 2X SSPE, 0.1% SDS and then processed as described above.

E. Sequencing

For double-stranded sequencing of PCR product a single band of appropriate size was exdsed from an ethidium bromide-stained agarose gel under UV llght The excised band was placed into a Costar spin-X 0.22 μm cellulose acetate filter unit frozen at -70°C for 15 minutes and spun in an Eppendorf microcentrifuge for 15 minutes at full speed. The DNA-containing eluate was transferred to a microcentrifuge tube, 20 μg glycogen was added, and DNA was predpitated with 0.2 volumes of 3 M sodium acetate and 0.3 volumes ISO propanol. DNA was pelleted in a

microcentrifuge, washed with 70% ethanol, vacuum dried, and resuspended in 20 μl double distilled water. The sample was sequenced according to the Sequenase (United States Biochemical) protocol using 7.75 μl DNA solution and 2.5 pmol sequencing primer.

F. Ollgonucleotides

Table 3 provides primer pairs for amplification of nucleic acid segments possibly containing oncogenic point mutations. If primer sequence is within the coding region (exon) of the α subunit the primer pair is suitable for amplification of either a DNA or cDNA template. The nucleic acid sequence of the G-protein α subunits are publlshed for Giα (see Bray et al., 1987, Proc. Nad. Acad. Sci. USA 84:5115-5119), Gsα (see Kozasa et al., 1988, Proc. Natl. Acad. Sci. USA 85:2081-2085), Goα (see Lavu et al., 1988, Biochem. Biophys. Res. Comm. 150:811-815). and Gzα (see Fong et al., 1988, Proc. Nad. Acad. Sci. USA 85:3066-3070). For each primer pair, the size of the amplification product from a genomic DNA template is shown. Note that for Giα1, Giα2, Giα3, Goα, and Gzα the designation 201 and 227 refers to the codon encoding the amino acid corresponding to position 201 or 227 in amino add sequence of Gsα.

Table 3

Amplification Primers for Detection of G-Protein Point Mutations

Primer

Gsα 201/227 Outer Primers Position

JFL69 5' GCG CTG TGA ACA CCC CAC GTG TCT intron JFL70 5 ' CGC AGG GGG TGG GCG GTC ACT CCA intron

Gsα 201/227 Inner Primers (526 bp)

JFL135 5 ' GTG ATC AAG CAG GCT GAC TAT GTG exon JFL136 5 ' GCT GCT GGC CAC CAC GAA GAT GAT exon Primer

Gsα -201 (289 hp) Position JFL228 5' AAG AAA CCA TGA TCT CTG TTA TAT intron JFL135 5' GTG ATC AAG CAG GCT GAC TAT GTG exon

Gsα -227 (263 hp)

JFL229 5' CCC CAG TCC CTC TGG AAT AAC CAG intron

JFL136 5' GCT GCT GGC CAC CAC GAA GAT GAT exon

Gsα -49 (171 bp)

JFL226 5' AAC AGC AGA CCT CCC TGC CCA AAG intron JFL227 5' CCC CCC TGC ACA GAT TTG ACA CTT intron

Giαl -201/227

JFL223 5' TTG GAC AGA ATA GCT CAA CCA AAT exon JFL224 5' TAG AAC CAG GTC GTA GTC ACT exon

Giα2 - 201/227 Outer Primers

JL54 5' CCC CCC ATC CCC AGC TAC CT exon/intron JL57 5' TCT CAC CAT CTC CTC GTC CTC exon/intron

Giα2 - 201/227 Inner Primsrs (504 bp)

JL55 5' ATT GCA CAG AGT GAC TAC ATC CCC exon JL56 5' GGC GCT CAA GGC TAC GCA GAA exon

Giα3- 201 Outer Primers

JFL109 5' TGT CTT TTA TTT AGT ATC AA exon/intron JFL110 5' GAT CTG GAT AGA ATA TCC CAG exon/intron

Giα3 - 201 Inner Primers (99 bp)

JFL110 5' GAT CTG GAT AGA ATA TCC CAG exon JFL112 5' GGT GAA ATG TGT TTC TAC AAT exon

Giα3- 227 Outer Primers

JFL113 5' TTC CCC TTG CGC AGG ATG TTT exonAmtron JFL114 5' ACA TAC CAT CTC CTC GTC CTC exon/intron

Giα3 - 207 Inner Primers (120 bp)

JFL115 5' CAG AAC AAG GTC ATA ATC ACT exon JFL113 5' TTC CCC TTG CGC AGG ATG TTT

Goα - 201 (87 bp)

SP9 5' CTG GAC AGC CTG GAT CGG ATT GGG exon SP10 5' GAG GTT CTT GAA TGT GAA GTG GGT exon

Goα -227(129 hp)

JFL139 5' AAC CTC CAC TTC AGG CTG TTT exon SP11 5' GTG GAG CAC CTG GTC ATA GCC GCT exon

Gzα -201/227 (-200bp)

JFL201 5' TGT GAC GCC CTC GAA GCA GT exon SP15 5' AAC GAC CTG GAG CGC ATC GCC exon Table 4

G-Protein Probes for the Detection of Point Mutations and Wild Tvne Alleles

Gsα - 49

5 ' TA GGT GCT GGA GAA TCT GGT 3 ' Gly

AGA Arg CGA Arg GAA Glu GCA Ala GTA Val

Gsα - 201

5 ' TT CGC TGC CGT GTC CTG ACT 3 ' Arg

TGT Cys GGT Gly AGT Ser CAT His CCT Pro CTT Leu

Gsα - 227

5 ' GTG GGT GGC CAG CGC GAT GA 3 ' Gln

TAG Thr GAG Glu AAG Lys CGT Arg CCG Leu CTG Pro CAC His CAT His

Giα1 - 201

5 ' TC AGA ACT AGA GTG AAA ACT 3 ' Arg

AGC Ser AGT Ser GGA Gly ACA Tar AAA Lys ATA lle

Giα1 - 227

5 ' TG GGA GGT CAG AGA TCT GAG 3 ' Gln

GAG Glu AAG Lys CTG Leu CCG Pro CGG Arg CAT His CAC His Giα2 - 201

5' TA CGG ACC CGC GTA AAG ACC 3' Arg

AGC Ser

GGC Gly TGC Cys

CAC His CCC Pro

CTC Leu

Giα2 - 227

5' GTG GGT GGT CAG CGG TCT GA 3' Gln

CTG Leu CCG Pro CGG Arg AAG Lys GAG Glu CAT His CAC His

Giα3-201

5' TT CGG ACG AGA GTG AAG ACC 3' Arg

AGC Ser

AGT Ser

GGA Gly

ATA lle

ACA Thr

AAA Lys

Giα3-227

5' GTA GGT GGC CAA AGA TCA GA 3" Gln

CAT His CAC His CTA Leu CCA Pro CGA Arg AAA Lys GAA Glu

Goα - 201

5' CTC CGA ACC AGG GTC AAA AC 3' Arg

GGG Gly TGG Trp AAG Lys ACG Thr ATG Met AGC Ser AGT Ser Goα - 227

5 ' GTC GGA GGC CAG CGA TCT GA 3 ' Gln

AAG Lys

GAG Glu

CCG Pro CGG Arg

CTG Leu

CAT His

CAC His

Gzα - 201

5 ' TG CGC TCC CGG GAC ATG ACC 3 ' Arg

GGG Gly

TGG Trp

CAG Gin

CCG Pro CTG Leu

Gzα - 227

5 ' TG GGG GGG CAG AGG TCA GAG 3 ' Gln

AAG Lys

GAG Glu CCG Pro

CTG Leu

CAC His

CAT His

CGG Arg For each G-protein codon to be characterized, ollgonucleotide probes are shown in Table 4 as follows. For each position, i.e., Gsα 201, the full probe sequence is shown for the wlld type allele. The wild type codon at the potentially oncogenic site is underlined, and the translated amino add is shown to the right A set of probes is provided where the sequence is identical to the wlld type except at the codon to be characterized. Thus, for non-wlld type probes only that codon and the translated amino acid product is shown in the table. Only those point mutations encoding amino adds different from the wlld type amino acid are shown.

Table 5 shows stretches of Gsα sequence surrounding the arginine-201 and glutamine-227 codons which are highly conserved in G-protein α chains of vertebrates, yeast, and slime mold. Publlshed sequences include rat Gsα, Giα2, Giα3, and Goα, human Gzα (Katada and Vi, 1982, Proc. Natl. Acad. Sci. USA 79:3129), Gtα of bovine retinal rod cells (Chambard et al., 1987, Nature 326:800), the α chain of the G- protein (called GPAl/SCGl) that mediates pheromone signalling in Saccharomyces cerevisiae (Corven el al., 1989, Cell 59, and Itoh et al., 1988, J. Biol. Chem.

263:6656). and an α chain (Gαl) from Dictyostelium discoideum (Zarbl et al., 1985, Nature 315:382). The number is parenthesis following each sequence is the actual amino acid position of the last amino acid in the sequence shown. In the table, a one letter amino acid code is used where:

D = Asp K = Lys

L = Leu G = Glu

R = Arg M = Met

C = Cys A = Ala

V = Val Q - Gln

T- Thr F = Phe

S = Ser E = Glu

I = lle

To determine the frequency of mutations in codons 201 and 227 of Gsα genes in pituitary and other tumors, the polymerase chain reaction (PCR) was used to amplify a specific region of genomic DNA and high-stringency hybridization of

sequence-specific ollgonucleotides to detect point mutations in the amplified product To detect mutations in the Gsα gene, a single region including both codons 201 and 227 and an intervening intron was- amplified from tumor genomic DNA prepared from fresh frozen or paraffin embedded samples as described in Example 1. The

10 amplification primers shown in Table 3, designated "Gsα 201/227 Outer Primers" and "Gsα 201 227 Inner Primers," were used according to the method described above. Ollgonucleotides specific for wlld type or single-base mutations at codon 201 (6 possible missense mutations) or codon 227 (7 possible missense mutations, 1 nonsense mutation) were hybridized to the a pllfied product (Table 4). Genomic DNA from

15 more than 300 tumors was analyzed either in the form of high molecular weight DNA prepared from fresh tissue or as obtained from paraffin-embedded tissue. Group 1 tumors had low basal adenylyl cyclase activity thatresponded normally to stimulatory agents, group 2 tumors had marked elevation of basal adenylyl cyclase activity that responded poorly to stimulatory agents.

20 The hybridization results are shown in Figure 1 A. The hybridization probes shown in Table 4 were used as fottows: R201 indicates the wlld type probe for Gsα Arg 201; R201C indicates that the probe used contained a point mutation (CGT to TGT) encoding cysteine; and R201H indicates that the probe used contained a point mutation (CGT to CAT) encoding histidine. The fourth panel was probed with the

25 probe corresponding to Gin 227 containing a point mutation (CAG to CGG) encoding arginine.

in all tumors where a mutant allele was detected, suggesting that mutations that activate Gs are dominant.

Of 42 GH-secreting pituitary tumors, 16 were biochemically characterized in terms of adenylyl cyclase activity. Eight tumors showing elevated adenylyl cyclase were predicted to harbor an activated Gsα; each of these tumors contained a mutation in codon 201 or codon 227 (Figure 1A). No mutations were detected in eight tumors that showed normal adenylyl cyclase activity. Although Gsα mutations in other codons may also inhibit GTPase, the strong concordance between elevated adenylyl cyclase activity and a mutation in codon 201 or codon 227 indicates that activating mutations at other sites are relatively infrequent

Table 6 provides a summary of human tumors screened for mutations in codons 201 and 227 for Gsα and codons 179 and 205 for Giα2. DNA for PCR amplification Mullis et al., supra., was isolated from either fresh tissue, Verlaan-de Vries et al., 1986, Gene 50:313, or paraffin-embedded tissue Kozasa ei al., 1988, Proc. Natl. Acad. Sci. USA 85:2081. Eighteen growth hormone (GH) secreting pituitary adenomas contained a mutation in either Gsα codons 201 or 227. One ovarian granulosa cell tumor and three adrenal cortical tumors (two adenomas, one carcinoma) contained a mutation in Giα2 codon 179.

Table 7 provides a llst of wlld type codons for the conserved arginine and conserved glutamine in Gsα and Giα2 genes. The table also shows single-nucleotide base changes (in bold) and the resulting amino acid changes. Oligonucleotides specific for wild type and each missense or nonsense single-base change listed were used to screen human tumors, with the exception of base changes that would be silent (marked with asterisks). Mutations detected in the tumors listed in Table 6 are underlined. "Term" indicates termination or stop signal.

Example 3

Identification of G-Protein Point Mutations in Giα2 Genes In order to investigate the possibillty that mutational activation of signaling pathways mediated by other G-proteins might lead to abnormal proliferation and tumor formation, a large panel of human tumors was screened for mutations in a Giα gene. In Giα2 the coding sequence and intron between the two codons to be tested, arginine-179 (corresponding to Gsα Arg 201) and glutamine-205 (corresponding to Gsα Gin 227), is short enough to allow PCR amplification of a single genomic DNA fragment containing both codons (Itoh et al., 1988, J. Biol. Chem.263:6656).

The primers and probes used for detection and characterization of Giα2 are shown in Tables 3 and 4. Samples were prepared and analyzed as described in Example 1. The hybridization results are shown in Figure 1B. In the figure, the first two rows of each panel were probed with the wild type probe for codon Arg 179 (R179). The third and fourth rows were hybridized to the Giα2/201 probe of Table 2 containing a point mutation (CGT or TGT) encoding cysteine (R179C). The last two rows of each panel were hybridized to the Giα2/227 probe shown in Table 4 comprising a point mutation (CGT or CAT) encoding histidine (R179H). The amplification methods are described in Example 1.

Table 6 summarizes the hybridization results. Mutations in codon 179 of Giα2 were detected in two different endocrine tumor types 3 of 11 tumors of the adrenal cortex and one of 6 ovarian granulosa cett tumors. The adrenal tumor lacking a wlld type allele was an adenocarcinoma; the other 2 adrenal tumors were adenomas. No mutations were found in codon 205. The high frequency of codon 179 mutations in tumors of two related cell types suggests that these mutations converted the Giα2 gene into an oncogene, referred to herein as gip2 (for Gi protein-2).

Strikingly, the amino acids that replaced arginine-179 in Giα2, cysteine and histidine, were the same as those that replaced the cognate arginine at position 201 in Gsα oncogene products found in pituitary tumors (Landis et al.). It is possible that, of the six possible missense mutations that can result from single-base changes in these codons of Gsα and Giα2, only these mutant proteins are biologically active. Also, all mutations found so far, in either Gsα or Giα2, are transition mutations (Table 7); consequently, these mutations may reflect a common mutagenic mechanism.

In one tumor of the adrenal cortex, a normal allele of Giα2 was not detected (Figure 1B). Sequence analysis of PCR products revealed a single sequence, corresponding to the codon 179 mutation. This result makes it likely that the normal allele was missing, although the possibillty that both alleles contain the same mutation cannot be excluded. Loss of the normal allele suggests that its protein product interferes with the oncogenic effect of the mutant protein, so that fallure to express normal Giα2 confers an additional selective advantage on cells carrying an activating Giα2 mutation in the other allele.

Example 4

Detection of gsp Mntationsin Paraffin-Embedded Samples by Microdissection

Changes in the two Gsα codons affected by gsp. mutations were detected in formalin-fixed, paraffin-embedded tissue blocks of human tumors by isolating genomic DNA, amplifying appropriate Gsα sequences with the polymerase chain reaction (PCR), and screening the amplified products for their abillty to hybridize with allele- specific ollgonucleotides. In Examples 1-3 DNA from all cells in a 5 mm tissue section was analyzed. DNA from one thyroid tumor had two different gsp mutations, in addition to the wild type allele. To ask whether the two mutations were located in different regions of the 5 m tissue slice, an adjacent 5 mm section from the same tumor was divided into several smaller fragments; genomic DNA was isolated from each fragment separately and subjected to PGR amplification and screening with allele- specific ollgonucleotides. The two gsp mutations were detected in different fragments; in addition, several fragments contained only the wlld type codons at positions 201 and 227 of Gsα.

The latter observation suggested that the heterogeneous distribution of mutations in 5 mm sections may go undetected due to dilution by wild type DNA sequences from cells in parts of the sections that contained no mutations. Accordingly, this modified microdissection approach was applied to additional tumors. Identification of a high prevalence of gsp mutations some in tumors earller tested as negative indicated that this approach is more sensitive at detecting gsp mutations. A. Mahgnant Thyroid Tumors

To explore the roles of gsp mutations in pathogenesis of mallgnant thyroid tumors, the microdissection technique to tissue blocks of primary tumor or lymph node metastases from 37 patients with differentiated thyroid carcinoma. These tissue fragments were also tested for mutations in specific codons of the three human ras genes.

Stained sections from these surgical specimens were examined and each microdissected fragment was classified as mallgnant or benign thyroid tissue; in every case, the pathologic diagnosis applled to at least 90% of the cells in the fragment. Both the pathologic examination and the DNA analysis were performed in a blinded fashion: The pathologist diagnosed histology of tissue fragments without knowledge of the presence or absence of mutations, and the DNA analyses were performed on coded samples.

B. Sample Preparation

Multiple 5 mm sections were cut from each tissue block and one was stained with hematoxylin and eosin. Regions of the stained sllde, 30-100 mm² in area, were demarcated with a pen and designated by number, when possible, demarcations separated histologically distinct regions. Using the stained sllde as a template, an adjacent unstained 5 mm section was divided into corresponding fragments with a razor blade. These fragments were transferred into separate tubes and each was processed for PCR and hybridization screening, exactly as described in Example 1.

C. PCR Amplification

Using 20-25 base pair (bp) ollgonucleotide primers upstream and downstream of codons 201 and 227 in the human Gsα gene, templates ranging from 165 to 1,200 bp in length. Primers chosen for PCR amplification of regions around codons 12, 13 and 61 of the human H-ras, Ki-ras and N-ras genes yielded products of 112-117 bp. Amplification of formalin-preserved, paraffin-embedded tissues was most effective with relatively short PCR products. Amplification with nested primers was required when the first PCR of larger DNA fragments produced an unsatisfactory amount of amplified DNA, as judged by agarose gel electrophoresis and ethidium bromide staining.

PCR conditions were as generally discribed in Example 1, however, the oligonucleotides and cycling temperatures used were as were as follows.

For nested primer amplification of Gsα codons 201-227: Outer sense, (JFL69) 5'GCG CTG TGA ACA CCC CAC GTG TCT; outer antisense (JFL70), 5'CGC AGG GGG TGG GCG GTC ACT CCA; product 1,200 bp; optimal PCR condition, 30 cycles of 1, 2, and 2 min at 95°/50°/72°; inner sense (JFL135), 5'GTG ATC AAG CAG GCT GAC TAT GTG; inner antisense (JFL136), 5'GCT GCT GGC CAC CAC GAA GAT GAT; product 526 bp; optimal PCR condition, 30 cycles 30, 40, and 45 sec at 95°/57°/72°.

For Gsα codon 201: sense (JFL135), 5'GTG ATC AAG CAG GCT GAC

TAT GTG; antisense (JFL286),5' TA ACA GTT GGC TTA CTG GAA; product 222 bp; optimal PCR conditions: 40 cycles of 30, 30, and 30 sec at 95°/55°/72°. For Gsα codon 227: sense (JFL229),5' CCC CAG TCC CTC TGG AAT AAC CAG; antisense (JFL136), 5'GCT GCT GGC CAC CAC GAA GAT GAT; product 165 bp; optimal PCR condition, 50 cycles of 60, 30, and 30 sec at 95°/55°/72°. For ias gene ampllcations the following standard PCR conditions of 50 cycles of 60, 30, and 30 sec 95°/55°/72° were used. Ras primers were as follows.

Ha-ras codons 12 and 13: sense (JFL243), 5'AGA CCC TGT AGG AGG ACC CCG GGC C; antisense (JFL244), 5'ATA GTG GGG TCG TAT TCG TCC ACA A; product 150 bp.

For Ha-ras codon 61: sense (JFL252), 5'GTC ATT GAT GGG GAG ACG TG; antisense (JFL253), 5'ACA CAC ACA GGA AGC CCT CC; product 112 bp; for Ki-ras codons 12 and 13: sense (EK371), 5'CCT GCT GAA AAT GAC TGA ATA TAA A; antisense (EK372), 5'T ATT GTT GGA TCA TAT TCG TCC ACA; product 118 bp;

for Ki-ras codon 61: sense (JFL248), 5'GTA ATT GAT GGA GAA ACC TG; antisense (JFL249), 5'ATA CAC AAA GAA AGC CCT CQ product 112 bp.

For N-ras codons 12 and13: sense (JFL216), 5'CTT GCT GGT GTG AAA TGA CT; antisense (JFL257), 5;GGT GGG ATC ATA TTC ATC TA; product 150 bp.

For N-ras codon 61: sense (JFL218), 5'GTT ATA GAT GGT GAA ACC TG; antisense (JFL242), 5'GGC AAA TAC ACA GAG GAA GCC TTC; product 112 bp.

Hybridization dot blots like that shown were scored by counting in an AMBIS radioanalytic imaging system as described above. Repllcate PCR amplifications and analyses of adjacent 5 mm sections produced similar results, indicating that the procedure was accurate and reprodudble as well as sensitive.

To avoid false positives by nonspecific hybridization of mutant probes to the PCR product in the absence of mutations, a level of hybridization was set, below which a sample would be considered negative for a particular mitation. This level was set at 20% of the signal detacted by hydridization of the wild type probe to unmutated DNA in the sample. Hybridization signals were determined to be reproducible at this level, but not below it. Consequently, a positive result (20% or more of the amplified DNA contains a mutation) indicates that at least 40% of the cells in the assayed tissue fragment contain the mutation, if— as expected for dominant somatic mutationseach cell has one wlld type and one mutant allele. D. Hybridization and Detection of Point Mutations

For mutation-specific olinucleotide hybridization, the PCR product was spotted (dot-blot apparatus, Bio-Rad) and cdvalently bound to a nylon filter (Pall Biodyne-B, 0.45 um) using UV light at the auto-crosslink setting (Stratallnker, Stratagene).

Hybridization with [³²P]-radiolabdid mutation-specific ollgonucleotides 20 bp long were performed exactly as described in Example 1. E. Criterion for Presence of a Mutation

To verify the presence of point mutations we scanned all a_s hybridization reactions with an AMBIS radioanalytic imaging system (Ambis Co., San Diego, CA), which measures the emission of radioactivity from each mutant (CPM_m) or wild type (CPM_wt) dot on the filter. The [³²P] b radioactivity of mutant ollgonucleotides nonspecifically bound to the same filters after high stringency washing was termed background activity (CPM_b), and ranged from 5 to 10 % of the wlld type hybridization signal. A dot was considered to represent a gsp mutation if (CPM_m - CPM_b) divided by (CPM_wt - CPM_b) was greater than or equal to 0.2. This criterion thus required that ampllfied DNA samples judged as positive for a mutation must exhibit a mutant signal 20% of that observed with wlld type. Applying this criterion also minimized the chance that a spuriously positive result could result from contamination of a gsp-negative sample by DNA from a gsp-positive sample.

F. Controls

The microdissection procedure gave negative results for Gsα and ias mutations in sections of normal connective tissue and human thyroid removed dining

parathyroidectomy. To confirm positive results, PCR products from six fragments for which hybridization results indicated R201C mutations at levels near the demonstrated cutoff point (i.e., 22-35%) were sequenced. Genomic DNA was ampllfied using one biotinylated and one non-biotinylated primer, to generate umlateral biotinylated PCR products. After binding the biotinylated PCR product to streptavidin coated beads (Dynabeads, Dynal) the -complementary strand was denatured and aspirated, leaving single stranded DNA. Sequencing was performed using the Sequenase kit (Sequenase, USB). In all six cases the sequencing gels showed both the wlld type and the mutant 201 codon, confirming the results of dot blot hybridization.

G. Number and Distribution of Mutations

Gsp mutations were found in surgical spedmens from 24 of 37 patients (65 %) with differentiated thyroid cancer. Gsp mutations in these patients were

heterogeneously distributed among microdissected fragments; overall, 81 of 266 thyroid tissue fragments (30.5 %) contained detectable gsp mutations. Among the 24 surgical specimens with at least one gsp positive fragment, a gsp mutation was found in 60 of 120 histologically mallgnant fragments (50 %). The mutations were not confined to fragments with obviously mallgnant tissue, however, indeed, in the same tumors 21 of 60 histologically benign fragments (36.2 %) also contained gsp mutations. Of the 24 patients in whom the microdissection technique revealed gsp

mutations, only five were originally suspected of harboring gsp on the basis of the previous testing procedure, which used the entire 5 m section as a single sample. In each of these five previously positive cases, more than half of the fragments tested with the new procedure were positive for gsp mutations, a frequency found in only one other case. These observations indicate that without microdissection the dilution of samples by DNA from regions without mutations can cause a substantial proportion— perhaps more than 75%— of thyroid gsp mutations to be missed.

Specimens from six of the 37 patients harbored two different gsp mutations, and the tumor of one patient (number 22, Table 8) contained three different gsp

mutations. These multiple mutations were found in separate fragments of these specimens: Of the 31 gsp-positive fragments tested in patients with more than one gsp mutation, only one fragment contained two mutations. Multiple gsp mutations in a single tumor did not correlate with clinical aggressiveness or other characteristics of the tumor (Table 8).

Of the 37 specimens examined, 15 were obtained from tumor metastases to areas physically separate both from the original primary tumor and from normal thyroid tissue— i.e., spread beyond the strap muscles to lymph nodes in 13 cases and to lung in two cases. Of these 15 specimens, 11 harbored gsp mutations. In three cases (patients 12, 19 and 24), these metastases revealed two different gsp mutations.

Occasional reports have documented more than one ias mutation in a tumor (e.g., in human cutaneous melanoma, see Van't Veer et al., 1989, Mol. Cell Biol., 2:3114-3116). The present study is the first to find a high incidence of multiple mutations of a single oncogene in individual tumors. The microdissection technique will likely find a similar multipllcity of oncogenes in other tumors as well.

H. Ras Mutations

Mutational substitutions of amino acid residues in three key positions of p21^ras — glycine-12, glycine-13, and glutamine-61— inhibit GTP hydrolysis and promote its abillty to trigger neoplastic transformation. Other investigators have reported prevalences of ras mutations in thyroid cancer ranging from 17 to 60 % (LeMoine et al., 1989. Oncogene, 4:159-164 and Wright et al., 1989. Brit. J. Cancer, 60:576-5771 Consequently, the 37 surgical specimens from patients with thyroid cancer were tested for oncogenic mutations affecting the three key codons of the three human ras genes. Microdissected fragments from 12 of 37 patients (32 %) contained N-ras mutations. These included 12 fragments with codon-61 mutations (Q61R), 18 codon-13 mutations (nine G13C, nine G13D), and a single fragment with a codon-12 mutation (G12C). N-ras mutations resembled gsp mutations in several respects: Both were heterogeneously distributed, sometimes multiple in a single patient, and present in benign as well as mallgnant thyroid tissue (Table 8). Of 117 microdissected fragments tested in the 12 ras-positive patients, 31 (26 %) contained an N-ras mutation. Two patients had more than one different ras mutation.

Detection of ras and gsp mutations in the same microdissected fragment does not mean that both mutations are present in the same cell; indeed, further

microdissection would probably segregate different cell populations containing each mutation. I. New Gsp Mutations

The thyroid tumors exhibited four previously unreported mutations, including substitutions of protine or serine for arginine-201 and substitutions of histidine or proline for glutamine-227.

J. Conclusions

The microdissection technique for finding point mutations greatly extends the practical resolution for detecting certain oncogenes, to the point that it can detect a mutation present in 40% or more of the few thousand cells in a 5 m x 30-100 mm² fragment of tissue. Microdissection may serve to uncover heterogeneously distributed oncogenes in non-thyroid tumors also. The increased sensitivity of this technique also shows that a conclusion that a particular oncogene mutation is not present in a tumor can be wrong. If based upon PCR amplification of large fragments of tumor, such a negative conclusion must be qualified; in fact, a negative result only indicates that the mutation is not present in a substantial proportion of the cells in the tumor fragment. The results of this analysis are presented in Table 8.

K. Table 8

Patients numbered 1-27 and 28-37 (left column) underwent surgery in Dϋsseldorf and San Francisco, respectively. Gap and ras mutations are listed as present or absent in microdissected fragments of thyroid tissue; fragments containing 5 no thyroid tissue are not included. Each fragment tested was classified as "carcinoma" or "benign"; for each, 90% or more of the cells seen in the corresponding part of the stained section were histologicall 'mallgnant or benign, as indicated. Mutations are enumerated as a fraction of n mutations detected in N fragments tested; ND indicates that no fragment of the particular classification (carcinoma or benign thyroid) were

10 present in the material available for analysis. Although Harvey- and Kirsten-rαs

mutations were also sought, only N-ras were found.

Clinical characteristics of patients: The "HistoJTNM" column (third from the left) indicates whether the tumor was diagnosed as a papillary (P) or folllcular (F) thyroid carcinoma, and provides the numerical cllnical classification (TNM) of thyroid

15 cancer devised by the World Health Organization²!, in which T, N, and M are numbers referring to different characteristics: T (varying from 1 to 4) indicates increasing size and extent of the tumor mass; N (varying from 1 to 3) indicates increasing numbers of lymph node m

Example 5

Materials and Methods

A. Human Tissues

Human tissues were obtained from surgical biopsy and either RNA immediately extracted or rapidly frozen to -70°C until use. All human tissues were obtained by informed consent in approval with the University of California, Davis Human Subjects Review Committee. Tissues were also supplled by the Tissue Bank of the Department of Pathology University of California, Davis) under direction of Dr. Robert D. Cardiff.

Microscope slldes with bone marrow smears on them were analyzed. One was hematoxylin and eosin (H+E) stained sllde, whlle the other was merely air dried bone marrow. Both slldes were stored at room temperature for several months prior to RNA extraction.

Other archival samples were either methanol- or ethanol-fixed prior to being embedded in paraffin. B. Cell Lines

Human tumor cell lines previously characterized to have known activating point mutations in various ras alleles were used. Cell lines EJT24 (bladder transitional cell carcinoma) (Tabin et al., 1982, Nature 300:143-149, and Reddy et al., 1982, Nature 300:149-152) and SK-N-SH (neuroblastoma) (Taparowsky et al., 1983, Cell 344:581- 586) were a gift of Dr. R. Cardiff. Calu-1 (lung carcinoma), SW 480 (colon carcinoma, and PA-1 (teratocarcinoma) were obtained from the American Type Culture Collection (ATCC) (Capoh et al., 1983, Nature 304:507-513, and Tainsky et al., 1984. Science 225:643-6451 HL-60 (promyelocytic leukemia) was a gift of Dr. J. Lawrence (Murray et al., 1983, Cell 33:749-757).

Other cell lines were used as negative controls for ras mutations because they are not known to contain activated alleles. K562 (erythroleukemia) was supplled by the Cetus Tissue Culture Collection (CTCC) (Lozzio and Lozzio, 1975, Blood 45:321- 334). Cell line G-2101 (renal clear cell carcinoma) was originated in our lab

(Gumerlock et al., 1988, In Vitro Cell Devel. Biol.24:429-434). The cell strain T- 3891 (fetal lung) is a normal, nonimmortallzed fibroblastic culture (Rossitto et al., 1988, J. Virol.140:431-435). All of the above cell llnes were maintained according to the instructions of the suppller.

C. RNA Extractions from Fresh or Frozen Tissues and Cell Lines

Total cellular RNA was extracted from tissues and cell llnes using modifications of the previously described guanidinium-isothiocyanate-phenol-chloroform methdds (Maniatis et al., 1982, In. Molecular Cloning, New York, Cold Spring Harbor Page 190, and Chirgwin et al., 1979, Biochem.18:5294-5299). Guanidmium

isothiocyanate solution (5 M guanidinium isothiocyanate, 25 mM sodium citrate, 0.5% sarcosyl, pH 7.0) (GITC) was prepared to 5% β-mercaptoethanol (GITC-ME) just prior to use. Tissue bits were powdered in liquid nitrogen in a mortar, further ground in the mortar upon additional of GITC-ME and 1.5 ml of the slurry was layered onto a cesium chloride (CSCL) density gradient in 13 x 51 mm polyallomer tubes (Beckman Laboratories). The CsCl density gradient was prepared by layering 1.5 ml of a 40% CsCl density gradient was prepared by layering 1.5 ml of a 40% CsCl solution in 20 mM Tris-HCl, 2 mM EDTA, pH 7.5 (TE) onto 2.0 ml of 5.7 M CsCl in TE. RNA was pelleted through this density gradient by ultracentrifugation at 40,000 rpm in an SW-50.1 rotor at room temperature for 16 to 19 hours.

The RNA pellet was suspended in 50 μl TE-SDS (10 mM Tris-HCl, 1 mM EDTA, pH 7.4 wit 0.5% SDS) in a microcentrifuge tube for phenol-chloroform extraction. TE saturated phenol was mixed 1:1 (v/v) with a chloroform:isoamyl alcohol (24:1) solution. An equal volume of this phenol:chloroform solution was added to the RNA solution, vortexed vigorously for 10 seconds and phase separated in a microfuge for two minutes. This extraction was repeated, and the aqueous phase containing the RNA was placed in a 2 ml microcentrifuge tube for predpitation. The RNA was precipitated by addition of 5 M NaCl to create a final concentration of 0.3 M NaCl followed by addition of two volumes of ice-cold 100% ethanol. This solution was placed at -70°C for a minimum of one hour. The tube was then warmed to room temperature to melt the ice and spun in a microfuge at 4°C for 15 minutes to pellet the RNA predpitate. The supernant was decanted and residual llquid was removed by vacuum desiccation. When nearly dry, the RNA pellet was redissolved in TE (without SDS) and precipitated a second time. This time the RNA pellet was redissolved in 50- 100 μl of 0.2X TE. RNA concentrations were determined by reading optical density at 260 mm in a spectrophotometer and calculated by setting 1.0 O.D. equivalent to 40 μg per ml RNA. D. RNA Extraction from Air-dried Bone Marrow Slides

Microscope slides of human bone marrow were extracted for RNA. One was stained with H+E while the other was merely air-dried and left unstained. The cells on these slides were scraped with razor blades into microcentrifuge tubes. To the tubes was added 1 ml of GITC-ME buffer, and the tubes were shaken vigorously on a rotary shaker for 60 minutes to dissolve the cells. The solution was then put in a 2 ml microcentrifuge tube. To precipitate DNA away from the RNA in solution, 0.1 ml of 2 M sodium acetate (pH 4.8) was added to each tube. The DNA precipitation and an extraction were performed by adding 1 ml of phenolxhloroform to the tube, inverting the tubes multiple times, and placing the tubes on wet ice for 15 minutes. This method of quick RNA extraction including the precipitation of DNA away from the RNA is modification of that described by Chomezynski and Sacchi, 1987, Anal. Biochem. 162:156-159. Following the incubation of ice, DNA remains at the interface between the organic and aqueous phases. The tubes were spun in a microfuge at 4°C for 20 minutes, and the aqueous phase containing the RNA was removed and transferred to a new 2 ml tube for precipitation of the RNA. to each tube was added 750 μl of isopropanol, and the tubes were inverted several times before placing at -20°C for one hour. Precipitated RNA was pelleted by spinning in a microfuge at °C for 20 minutes. The RNA was redissolved in 300 μl of GITC-ME and precipitated a second time by addition of 300 μl of isopropanol and placing at -20°C for one hour. RNA was pelleted again as above, supernatant was discarded, and the RNA pellets washed with 1 ml of 70% ethanol. Once again, the RNA was pelleted, supernatant discarded, and the pellet dried under vacuum desiccation. The RNA was finally redissolved in 50 μl of 0.2 X TE and quantitated. Both slides each yielded over 15 μg of RNA.

E. RNA Extraction from Alcohol-fixed Paraffin-Embedded Tissues

Fifty micron sections of paraffin blocks were cut and deparaffinized in 1 ml of xylenes by vigorous shaking in a microfuge tube for 30 minutes. Tissue bits were pelleted by microfuging for five minutes and the xylenes decanted. Residual xylenes were removed by washing with 100% ethanol andrepelleting the tissue bits. To the tissue 1 ml of the GTTC-Me solution described above was added. Tubes were vigorously shaken on a rotary shaker for one hour to dissolve the tissues. All subsequent steps in the RNA isolation were the same as those described above for the bone marrow slldes where DNA was predpitated away from RNA. Each fifty micron section yielded approximately 25 μg of RNA.

F. RNA/PCR Procedure

The strategy for amplification of mRNA sequences in the polymerase chain reaction (RNA/PCR) is based on that previously described (Kawasaki et al., 1988, Proc. Natl. Acad Sci. USA 85:5698-5702). Total cellular RNA was converted to a pool of cDNAs by reverse transcription. This'cDNA was then subjected to PCR using gene-specific primer pairs. The primer pairs werd designed to be homologous to exon sequences separated by one or more introns. Primer A, the upstream primer of the pair, comprised same sequence (5'--->3') as that determined for the non-template strand used in RNA transcription of the gene (the same sequence as the RNA itself with thymine substituted for uradl). Primer B, the downstream primer, was designed to be complementary to and anti-parallel to the non-template strand. Primers were generally 21 to 24 bases in length (21- to 24-mers) with a GC content of 40-60%.

With the primer pair spanning an intron, PCR amplification of spllced mRNA results in ampllfied products of a predicted length containing only the exon sequences of the gene. Amplification of unspllced RNA or any contaminating genomic DNA in the RNA preparation yields products of a larger size including the intron sequences. The smaller product predicted from the spllced mRNA sequence will only be produced if spllced mRNA transcribed from the gene of interest is present Therefore, the present of the predicted band on an ethidium bromide stained gel is an unequivocal assay for that gene's transcription or expression. This assay is referred as the Gel Visuallzation Assay (GVA) for gene expression.

The predicted ampllfied products were confirmed by use of internal probe hybridization to the PCR products. For this reason, ollgonucleotides were prepared in sets of three: two as the ampllmer pair and a third internal to both ampllmers to be used as a probe for the resulting products of PCR. However, with confirmation of the predicted band, GVA can be used to screen for gene expression in a extremely rapid and sensitive fashion. G. Ras Primers and Probes

Seven sets ofras-specific RNA/PCR primers were designed and are llsted in Table 6. An upstream primer specific for exon 1 in each of the three human ras genes, c-N-ras (primer EK 221), c-Ha-ras-1 (EK 222), and c-Ki-ras-2 (EK 223) was prepared, these were each used separately in combination with a generic exon 2 downstream primer (EK 225) targeted to a conserved region. In addition, a generic upstream primer (EK 224) was used, paired with EK 225 to detect "pan"-ras

expression. The products of each of these primer sets were predicted to be a size of 200 bp.

Primer pairs were also designed for each of the three genes to yield different- sized RNA/PCR products. Using GVA to score gene expression, the production of different-sized products allowed all three gene expression assays to be run

simultaneously in the same reaction mixture and then viewed in a single lane of a gel. The result saved in enzyme expenses and allowed scale-up of the number of samples to be screened by a factor of three.

RNA/PCR products from c-N-ras (primers EK 365 and EK 366), c-Ha-ras- 1

(EK 367 and EK 368), and c-Ki-ras-2 (EK 369 and EK 370) messages of 299 bp, 259 bp, and 234 bp, respectively. Each upstream primer was designed to have a greater than six base mismatch with the other two upstream primers to prevent cross amplification of the other ias messages. These three sets of primers were able to be used simultaneously in the same reaction mixture.

In these experiments, allele-specific ollgonucleotide (ASO) probes hybridization was used to detect point mutations at the 12th and 61st codons to screen for activating point mutations in ias alleles. The 20-mer probes used are llsted in Table 7. These probes were used according to the procedure described in Fair et al., 1988, Proc. Natl. Acad. Sci. USA 85:9268-9272. H. Synthesis of cDNA from RNA

Complementary DNA (cDNA) was synthesized from the extracted total cellular RNA essentially as previously described (Kawasaki et al., 1988, Proc Natl. Acad. Sci. USA 85:5698-5702). One microgram of total RNA was reverse transcribed with Moloney Murine Leukemia Virus (Mo-MuLV) reverse transcriptase (Bethesda Research Laboratories) in a 20 μl reaction in 1 X PCR buffer (50 mM KCl, 20 mM Tris-HCl pH 8.3, 2.5 mM MgCl₂, and 0.01% BSA) containing 20 U RNAsin (Promega), 1 μl of a 10 mM each stock of nucleotide triphosphates (dATP, dCTP, dGTP, and dTTP), and 100 pmoles of random hexamer primers. The change to random hexamer primers rather than ollgo-dT primers was based cm a recent report showing better yield of RNA/PCR products (Noonan and Roninson, 1988, Nucl. Acid Res.16:10366).

Reverse transcription reactions were incubated at room temperature for 15 minutes, 42°C for 30 minutes, and 95°C for five minutes. The 95°C incubation was done to heat denature the Mo-MuLV reverse transcriptase enzyme. These cDNAs were stored at 4°C until use in the PCR reaction. I, PCR Reactions Using cPNA

The PCR method was used according to Saiki et al., as described for DNA using recombinant thermostable DNA polymerase originating from Thermus aquaticus (rTaq) (Perkin Elmer-Cetus). Sllght modifications included reducing the amount of primers to 10 pmoles each, reducing the dNTPs to 1 μl of the 10 mM each stock described above, and using the rTaq enzyme at 2 U per reaction in a total 100 μl reaction. The substrate for RNA/PCR was 2 μl of the 20 μl cDNA described above. This amount corresponds to 100 ng of the initial 1 μg of total cellular RNA used in the reverse transcription reaction. Where three gene reactions were run simultaneously, this corresponds to a minimum of 30 gene expression studies from 1 μg of total cellular RNA. PCR thermal profiles of 95°C for one minute (denaturing of double strands), 55ºC for 30 seconds (annealing of amplimers), and 72°C for 30 seconds (synthesis of DNA) were performed in a programmable heat block (Perkin Elmer-Cetus) for 30-50 cycles. A 30 second synthesis step at 72°C was suffident to produce RNA/PCR products of at least 935 bp.

J. Gel Visuallzation Assav (GVA) for mRNA Expression

Discrete gene expression was scored by the gel visuallzation assay (GVA) following RNA/PCR. RNA/PCR products were screened by running 9 μl of the reaction mixture in 2% NuSieve (FMC, Rockland, MD), 1% agarose gels in Tris- borate EDTA buffer (TBE). For size markers, the 123 bp DNA ladder (Bethesda Research Laboratories) was used. Gels of 75 ml were run in wide mini-sub cells (Bio- Rad Laboratories) in TBE at a constant 100 volts for approximately 90 minutes. Gels were stained in an ethidium bromide solution (0.5 μg per ml) for 30 minutes, detained for 30 minutes, and photographed under ultraviolet llght with a Polaroid Land camera. K. ASO Probing of RNA/PCR Products

RNA/ PCR products to be probed were run in 2% agarose gels in a Tris-borate EDTA electrophoresis buffer (TBE) in a mini-gel system, alkaline transfer to Zeta- Probe nylon filters (Bio-Rad Laboratories) in a wick-action transfer was done with a 0.4 N NaOH solution in water. Transfers were allowed to proceed for 90 minutes. Following transfer, blots were neutrallzed in 2X SSC for 5 minutes. Blots were prehy bridized in a solution of 3 M tetra-methyl ammonium chloride (TMAC), 50 mM Tris-HCl pH 7.5, 2 mM EDTA, 5X Denhardt's solution, and 0.3% SDS at 55°C for one hour with circular agitation. Hybridization with ASO probes that were kinase- labeled with gamma-³²P-ATP was done in 5 ml of the TMAC buffer llsted above with 2 x 10⁶ cpm per ml of probe added. Hybridization continued at 55°C for one hour. The hybridization buffer and the first wash of 2X SSC with 0.1 % SDS (50 ml) at room temperature were discarded as radioactive waste. A second wash was done at room temperature with 2X SCC with 0.1% SDS. Blots were quickly rinsed in the TMAC buffer minus the Denhardt's solution and extensively washed in the same buffer at 61°C for one hour. This wash was the critical wash for allowing the discriminatory abillty of the ASO probes to distinguish point mutations. Blots were then blotted dry with Whatmann 3 MM paper and exposed to Kodak XAR film for at -70°C for one hour. Film was then developed in an automatic processor and data interpreted.

Example 6

Results

The results of GVA of RNA/PCR amplification of human ras family mRNAs are shown in Figure 2. The samples used were a normal spleen and the cell llne K562. Lanes 1 and 14 contain the 123 bp DNA ladder. Negative controls (no RNA) for each reaction are shown in lanes 4, 7, 10, and 13. Lanes 2-4 display the RNA/PCR products utilizing the "pan" ras primers EK 224 and EK 225 on the RNA from the normal human spleen, the K562 cell llne, and the negative control of no RNA, respectively. As predicted, die 200 bp ampllfied product is present Lanes 5-7 display the results using the c-N-ras-specific primers EK 365 and EK 336. The samples are displayed in the same sequence and as predicted, a 299 bp product is present indicating expression a the c-N-ras gene is both samples of human cells. c-Ha-ras- 1 expression is shown in lanes 8-10. The primers EK 367 and EK 368 produce a 259 bp product and that is clearly seen in lane 9 (K562 cells). No product is seen from RNA isolated from the normal human spleen (lane 8). The lack of a product is interpreted to reflect lack of c-Ha-ras-2 mRNA or levels of the message below that detectable after 30 cycles of RNA/PCR in the normal human spleen. This result shows the utillty of the GVA for the detection of mRNA after RNA/PCR.

Lanes 11-13 contain the RNA/PCR products using the c-Ki-ras-2 primers EK 369 and EK 370. The predicted product of 234 bp is present in both lanes 11 (normal human spleen RNA) and 12 (K-562 RNA) indicating expression of the gene in both samples; however, the abundance of message is less in the normal human spleen from that in the K562 cell llne.

Example 7

Utilization of the ras RNA/PCR Products in the Screening for Activating Point

Mutations

ASO probe hybridization with a different point mutation specific probe was conducted as described in Example 5. The results are shown in Figure 3.

The cell line samples on each blot are the same: lane 1, EJ/T24 RNA amplified with primers EK 222 and EK 225 (c-Ha-ras- 1); lane 2, EJ/T24 RNA amplified with primers EK 224 and EK 225 ("pan" jas); lane 3, PA-1 RNA ampllfied with primers EK 221 and EK 225 (C-N-ras); lane 4, SW-48- RNA ampllfied with primers EK 223 and EK 225 (c-Ki-ras-2); lane 5, SW-480 RNA amplified with primers EK 224 and EK 225 ("pan" ras); lane 6, HL-60 RNA amplified with primers EK 221 and EK 225 (c-N- ras); lane 7, Calu-1 RNA ampllfied with primers EK 223 and EK 225 (c-Ki-ras-2); lane 8, Calu-1 RNA amplified with primers EK 224 and EK 225 ("pan" ras); lane 9, G2101 RNA ampllfied with primers EK 224 and EK 225 ("pan" ras); lane 20, 123 bp DNA ladder. The blot in Panel A has been probed with a pool of ollgonucleotides specific for activating point mutations at the second nucleotide of the 12th codon of c-Ha-ras-1 (JN 03).

As predicted, lanes 1 and 2 containing the EJ/T24 RNA amplified both the c-Ha- ras-1 primers and the "pan" ras primers are positive for the characterized EJ/T24 mutation. Panel B blot has been probed with a pool of ollgonucleotides specific for activating point mutations at the second nucleotide of the 12th codon of c-N-ras (JN 17). The PA-2 cell line is known to contain a mutation at this position, and lane 3 is positive as expected. Panel C blot has been probed with ollgonucleotide pool JN 09 targeted to mutations at the first nucleotide of codon 12 in c-Ki-ras-2.

The SW-480 cell llne contains one of those mutations and lanes 4 and 5 containing RNA/PCR products for that cell line are positive. Because the signal in lane 5 is quite weak, it may indicate that the mutant allele's message is in low abundance with respect to all other ras messages in the cell as the "pan" cas primers were used for that lane or that the "pan" ras primers are less efficient at amplifying c-Ki-ras-2 messages with respect to the other two ras genes.

Panel D blot has been probed with pool JN 22 specific for mutations at the second position of the 61st codon of c-N-ras. Cell llne HL-60 has a mutation at the position and is positive in lane 6. This panel, in combination with Panel B, illustrates that the RNA/PCR products amplified by the primers EK 221 and EK 225 (c-N-ras) contain sequences of both the 12th, 13th, and 61st codons of that gene. This is a significant advance for ias point mutation screening in that a single PCR reaction allows one to screen both activating hotsppts in a single product as opposed to the need for two reactions when screening genomic DNA due to those two spots being present in exons 1 and 2 separated by a largeiutron. There is also the additional value of screening for expressed mutations in mRNA as opposed to the possibillty of detecting mutations in non-expressed alleles. Example 8

RNA/PCR products from alcohol-fixed paraffin-embedded samples were analyzed by GVA (Figure 4). Lanes 1, 5, and 12 contain the 123 bp DNA ladder. Samples in lanes 2, 6, and 9 have been amplified with primers EK 365 and 366 (c-N- ras: 299 bp), those in lanes 3, 7, and 10 with primers EK 367 and EK 368 (c-Ha-ras- 1: 259 bp), and those in lanes 4, 8, and 11 with primers EK 369 and EK 370 (c-Ki-ras- 2: 234 bp). Lanes 2-4 are the negative controls with no RNA added to the reverse transcriptase reaction of RNA/PCR. No products are seen in those lanes other than the primer dimers mentioned above. Lanes 6-8 contain RNA/PCR products from the Calu- 1 cell llne and the products corresponding to as messages from all three genes are present. Lanes 9-11 contain the RNA/PCR products from the cell llne G-2101. In this case, there is a lack of any signal from c-Ha-ras-1 messages indicating lack of expression.

These reactions were run for 50 cycles of RNA/PCR which increases the presence of background bands, but also the technique for precipitating DNA away from RNA which was used in the preparation of these RNAs is not totally effident and can result in contaminating genomic DNA in the RNA preparations. The contaminating genomic DNA can result in specific amplification of the ras genes containing the intron between exons 1 and 2 but it will be a much larger size than that predicted for the spliced mRNA.

Example 9

The "pan" ras primers were used to amplify reverse transcribed RNA products. Sample preparation and the amplification procedure were as described in Example 4.

The results of "pan" ras amplification of RNA isolated from the air-dried stained microscope sllde preparations of human bone marrow are shown in Figure 5. The predicted 200 bp RNA/PCR products using primers EK 224 and EK 225 are shown for the unstained sllde (lane 3) and the stained sllde (lane 4) adjacent to the negative control (no RNA) in lane 2 and the 123 bp DNA ladder in lane 1. These are the results of a 50 cycle RNA/PCR run. Example 10

Detection of ras Mutations by Format II

The synthesis of biotinylated primers, poly T tailing of the probes and preparation of the format II filters are as described in commonly assigned, copending U.S. Serial No. 197,000 incorporated herein by reference (also see Chiang et al., 1989. BioTechniques 7(41:3601. Three sets of filters representing 21 mutation specific ollgo probes for N-, H-, and K-ras, respectively, are shown in Figure 6. The sequences of the ias probes are llsted in Table 12. All probes were designed to have approximately the same melting temperature (~50 -52°C) so that hybridization conditions could be standardized for all the ollgonucleotides. Five pmoles of each talled probe were spotted onto Biodyne Nylon membranes (Pall Biosupport, NY) and UV immobilized.

The ias ollgonucleotide bound filters were hybridized in 5X SSPE, 0.5% SDS with alkall denatured PCR products for 60 minutes at 42°C. Washing was done in 3M tetramethylammonium chloride to minimize the influence of base composition among the various nucleotides. The filters were briefly rinsed with 2X SSPE, 0.1% SDS, then incubated in the same buffer wtth 2 μg/ml srreptavidin-horse radish peroxidase conjugate ("Sequence," Cetus Corporation) for 30 minutes at room temperature. The filters were then washed for five minutes with the same buffer without the conjugate. Reagents of the ECL gene detection system (Amerrsham) were added and incubated for one minute at room temperature. Filters were then wrapped in Saran wrap and the light signal produced was detected by exposing Kodak XRP film to the filters for 20 seconds to one minute.

N-ras codon 16 YZ4 AGCTGGACAAGAAGAGT

YZ30 AGCTGGAGAAGAAGAGT

YZ5 CAGCTGGAAAAGAAGAG

YZ31 AGCTGGACGAGAAGAG

YZ6 AGCTGGACTAGAAGAGT

YZ32 AGCTGGACCAGAAGAG

YZ51 AGCAGGACTGAAGAGT

YZ33 AGCAGGACACGAAGAG

H-ras codon 12 YZ7 GGAGCCGGCGGTG

YZ34 GGCGCCAGCGGTGT

YZ35 GGCGCCTGCGGTGT

YZ36 GGCGCCCGCGGTG

YZ37 GGCGCCGACGGTGT

YZ8 GGCGCCGTCGGTGT

YZ38 GGCGCCGCCGGTG

H-ras codon 13 YZ39 GCCGGCAGTGTGGG

YZ9 GCCGGCTGTGTGGG

YZ40 GCCGGCCGTGTGGG

YZ41 GCCGGCGATGTGGG

YZ42 GCCGGCGCTGTGGG

YZ43 GCCGGCGTTGTGGG

H-ras codon 61 YZ10 GCCGGCCAGGAGGA

YZ11 GCCGGCGAGGAGGA YZ44 CGCCGGCAAGGAGG YZ45 GCCGGCCGGGAGG YZ46 GCCGGCCTGGAGGA YZ47 GCCGGCCCGGAGG YZ48 CGCCGGCCATGAGG YZ49 GCCGGCCACGAGGA

K-ras codon 12 YZ52 GGAGCTGGTGGCGTA

YZ53 GGAGCTAGTGGCGTAG YZ54 GGAGCTTGTGGCGTAG YZ55 GGAGCTCGTGGCGTA YZ56 GGAGCTGATGGCGTAG YZ57 GGAGCTGCTGGCGTA YZ58 GGAGCTGTTGGCGTAG

K-ras codon 13 YZ59 GCTGGTAGCGTAGGC

YZ60 GCTGGTTGCGTAGGC YZ61 GCTGGTCGCGTAGGC YZ62 GCTGGTGACGTAGGC YZ63 GCTGGTGTCGTAGGC YZ64 GCTGGTGCCGTAGGC K-ras codon 61 YZ65 AGCAGGTCAAGAGGAG

YZ66 AGCAGGTGAAGAGGAG

YZ67 AGCAGGTAAAGAGGAGT

YZ 68 GCAGGTCGAGAGGAG

YZ69 AGCAGGTCTAGAGGAG

YZ 70 GCAGGTCCAGAGG AG

YZ 71 AGCAGGTCATGAGGAG

YZ72 GCAGGTCACGAGGAG

Claims

IntheClaims

1. AmethodfordetectingaG-proteinαsubunitpointmutationinanucleic addsegmentpresentinabiologicalsample,saidmethodcomprising:

(a) treatingthesamplewithaG-proteinαsubunitprimerpair, anagentfor polymerization, anddeoxynucleoside5'triphosphatesunderconditions suchthatan extensionproductofeachprimercanbesynthesized,whereinsaidprimers are suffidentiycomplementarytoseparatestrandsofanucleicacidencodingasegmentofa G-proteinαsubunittohybridizetheretosothattheextensionproductsynthesizedfrom onememberofsaidpair,whenseparatedfromitsconφlementarystrand,can serveasa templateforsynthesisoftheextensionproductoftheothermemberofsaidpair,

(b) separatingtheprimerextensionproductsfromthetemplatesonwhichthe extensionproductsweresynthesizedtoformsingle-strandedmolecules;

(c) treatingthesingle-strandedmoleculesgeneratedinstep (b)withthe primersofstep (a)underconditionssuchthataprimerextensionproductissynthesized usingeachofthe single-strandedmoleculesproducedinstep (b)asatemplate;

(d) repeatingsteps (b) and(c)atleastoncetoprovideampllfiedDNA;

(e) hybridizingaG-proteinαsubunitprobeto saidampllfiedDNA,wherein saidprobecontainsanucleicacidsequencethatwillhybridizetoasequenceselected fromawildtypeandamutantnucleicaddsequencewithinsaidamplifiedDNA; and (f) determiningifhybridizationhasoccurred.

2. ThemethodofClaim 1,whereinsteps(b)and(c) arerepeatedatleastfive timesandsaidagentofpolymerizationisathermostableDNApolymerase.

3. ThemethodofClaim2,whereinsaidthermostableDNApolymeraseis Taqpolymerase.

4. ThemethodofClaim 1,whereinthesampleisremovedfromahuman tumor.

5. ThemethodofClaim4,wherein saidhumantumorisanendocrinetumor.

6. ThemethodofClaim 1,wherein saidprimerpairwillamplify anucleic acid segmentencodingasubsequenceofaG-proteinαsubunitselectedfromthegroup consisting ofGiα1, Giα2, Giα3, Gzα, Goα, andGsα.

7. The method of Claim 6, wherein said nudeic acid segment encodes at least one amino acid selected from the group consisting of the amino acids

corresponding to Gsα 49, 201, and 227.

8. The method of Claim 7, wherein said primer pair is selected from the group consisting of: JFL69 and JFL70; JFL135 and JFL136; JFL228 and JFL135;

JFL229 and JFL136; JFL226 and JFL227; JFL223 and JFL224; JL54 and JL57; JFL109 and JLF110; JFL110 and JFL112; JFL113 and JFL114; JFL115 and JFL113; SP9 and SP10; JFL139 and SP11; JFL201 and SP15; JL55 and JL56; JFL223 and SP33; and JFL224 and SP34; JFL235 and JFL237; JL55 and JEL212; JFL215 and JL56; and JFL135 and JFL286.

9. The method of Claim 1, wherein said probe comprises a sequence which hybridizes to DNA encoding a subsequence of a G-protein subunit selected from the group consisting of Giα1, Giα2, Giα3, Goα, Gsα, and Gzα.

10. The method according to Claim 1, wherein said probe hybridizes to a G- protein subsequence encoding the amino acid corresponding to the Gsα amino acid at position 49, 201, or 227.

11. The method of Claim 10, wherein said probe is selected from die group consistingg of

5' TA GGT GCT GGA GAA TCT GGT 3'

5' TA GGT GCT AGA GAA TCT GGT 3'

5' TA GGT GCT CGA GAA TCT GGT 3'

5' TA GGT GCT GAA GAA TCT GGT 3'

5' TA GGT GCT GCA GAA TCT GGT 3'

5' TA GGT GCT GTA GAA TCT GGT 3'

5' TT CGC TGC CGT GTC CTG ACT 3'

5' TT CGC TGC TGT GTC CTG ACT 3'

5' TT CGC TGC GGT GTC CTG ACT 3'

5' TT CGC TGC AGT GTC CTG ACT 3'

5' TT CGC TGC CAT GTC CTG ACT 3'

5' TT CGC TGC CCT GTC CTG ACT 3'

5' TT CGC TGC CTT GTC CTG ACT 3' 5' GTG GGT GGC CAG CGC GAT GA 3';

5' GTG GGT GGC TAG CGC GAT GA 3';

5' GTG GGT GGC GAG CGC GAT GA 3';

5' GTG GGT GGC AAG CGC GAT GA 3';

5' GTG GGT GGC CGT CGC GAT GA 3';

5' GTG GGT GGC CCG CGC GAT GA 3';

5' GTG GGT GGC CTG CGC GAT GA 3';

5' GTG GGT GGC CAC CGC GAT GA 3';

5' GTG GGT GGC CAT CGC GAT GA 3';

5' TC AGA ACT AGA GTG AAA ACT 3';

5' TC AGA ACT AGC GTG AAA ACT 3';

5' TC AGA ACT AGT GTG AAA ACT 3';

5' TC AGA ACT GGA GTG AAA ACT 3';

5' TC AGA ACT ACA GTG AAA ACT 3';

5' TC AGA ACT AAA GTG AAA ACT 3';

5' TC AGA ACT ATA GTG AAA ACT 3';

5' TG GGA GGT CAG AGA TCT GAG 3';

5 TG GGA GGT GAG AGA TCT GAG 3';

5' TG GGA GGT AAG AGA TCT GAG 3';

5' TG GGA GGT CTG AGA TCT GAG 3';

5' TG GGA GGT CCG AGA TCT GAG 3';

5' TG GGA GGT CGG AGA TCT GAG 3';

5' TG GGA GGT CAT AGA TCT GAG 3';

5' TG GGA GGT CAC AGA TCT GAG 3';

5' TA CGG ACC CGC GTA AAG ACC 3';

5' TA CGG ACC AGC GTA AAG ACC 3';

5' TA CGG ACC GGC GTA AAG ACC 3';

5' TA CGG ACC TGC GTA AAG ACC 3';

5' TA CGG ACC CAC GTA AAG ACC 3';

5' TA CGG ACC CCC GTA AAG ACC 3';

5' TA CGG ACC CTC GTA AAG ACC 3';

5' GTG GGT GGT CAG CGG TCT GA 3';

5' GTG GGT GGT CTG CGG TCT GA 3';

5' GTG GGT GGT CCG CGG TCT GA 3';

5' GTG GGT GGT CGG CGG TCT GA 3';

5' GTG GGT GGT AAG CGG TCT GA 3';

5' GTG GGT GGT GAG CGG TCT GA 3';

5' GTG GGT GGT CAT CGG TCT GA 3';

5' GTG GGT GGT CAC CGG TCT GA 3';

5' TT CGG ACG AGA GTG AAG ACC 3';

5' TT CGG ACG AGC GTG AAG ACC 3';

5' TT CGG ACG AGT GTG AAG ACC 3';

5' TT CGG ACG GGA GTG AAG ACC 3';

5' TT CGG ACG ATA GTG AAG ACC 3';

5' TT CGG ACG ACA GTG AAG ACC 3';

5' TT CGG ACG AAA GTG AAG ACC 3'; 5' GTA GGT GGC CAA AGA TCA GA 3';

5' GTA GGT GGC CAT AGA TCA GA 3';

5' GTA GGT GGC CAC AGA TCA GA 3';

5' GTA GGT GGC CTA AGA TCA GA 3';

5' GTA GGT GGC CCA AGA TCA GA 3';

5' GTA GGT GGC CGA AGA TCA GA 3';

5' GTA GGT GGC AAA AGA TCA GA 3';

5' GTA GGT GGC GAA AGA TCA GA 3';

5' CTC CGA ACC AGG GTC AAA AC 3';

5' CTC CGA ACC GGG GTC AAA AC 3';

5' CTC CGA ACC TGG GTC AAA AC 3';

5' CTC CGA ACC AAG GTC AAA AC 3';

5' CTC CGA ACC ACG GTC AAA AC 3';

5' CTC CGA ACC ATG GTC AAA AC 3';

5' CTC CGA ACC AGC GTC AAA AC 3';

5' CTC CGA ACC AGT GTC AAA AC 3';

5' GTC GGA GGC CAG CGA TCT GA 3';

5' GTC GGA GGC AAG CGA TCT GA 3';

5' GTC GGA GGC GAG CGA TCT GA 3';

5' GTC GGA GGC CCG CGA TCT GA 3';

5' GTC GGA GGC CGG CGA TCT GA 3';

5' GTC GGA GGC CTG CGA TCT GA 3';

5' GTC GGA GGC CAT CGA TCT GA 3';

5' GTC GGA GGC CAT CAC TCT GA 3';

5' TG CGC TCC CGG GAC ATG ACC 3';

5' TG CGC TCC GGG GAC ATG ACC 3';

5' TG CGC TCC TGG GAC ATG ACC 3';

5' TG CGC TCC CAG GAC ATG ACC 3';

5' TG CGC TCC CCG GAC ATG ACC 3';

5' TG CGC TCC CTG GAC ATG ACC 3';

5' TG GGG GGG CAG AGG TCA GAG 3':

5' TG GGG GGG AAG AGG TCA GAG 3':

5' TG GGG GGG GAG AGG .TCA GAG 3':

5' TG GGG GGG CCG AGG TCA GAG 3':

5' TG GGG GGG CTG AGG TCA GAG 3':

5' TG GGG GGG CAC AGG TCA GAG 3':

5' TG GGG GGG CAT AGG TCA GAG 3':

5' TG GGG GGG CGG AGG TCA GAG 3';

5' TCG CTG CCG TGT CCT GGA C 3':

5' TCG CTG CAG TGT CCT GGA CT 3':

5' TCG CTG CGG TGT CCT GGA C 3':

5' TCG CTG CTG TGT CCT GGA CT 3':

5' TCG CTG CCA TGT CCT GGA CT 3':

5' TCG CTG CCT TGT CTT GGA CT 3':

5' TGG GTG GCC AGC GGC GAT GA 3':

5' TGG GTG GCC TGC GCG ATG A 3':

5' TGG GTG GCC CGC GCG ATG 3':

5' TGG GTG GCC GGC GCG ATG 3':

5' TGG GTG GCG AGC GCG ATG A 3' 5' TGG GTG GCT AGC GCG ATG A 3';

5' TGG GTG GCC ATC GCG ATG A 3';

5' TGG GTG GCA AGC GCG ATG A 3; and

5' TGG GTG GCC ACC GCG ATG 3'.

12. The method of Claim 1, wherein said nucleic acid is RNA and is reverse transcribed to provide a double-stranded cDNA copy prior to step (a).

13. The method of Claim 1, wherein said sample is paraffin embedded tissue.

14. An ollgonucleotide primer pair for the amplification of a subsequence of a nucleic acid encoding a G-protein α subunit wherein said subsequence comprises a nucleic add sequence encoding an amino add corresponding to the Gsα amino acid at a position selected from the group consisting of 49, 201, and 227.

15. An ollgonucleotide primer pair according to Claim 14, wherein said G- protein α subunit is selected from the group consisting of Gsα, Goα, Giα1, Giα2, Giα3, and Gzα.

16. An ollgonucleotide probe for distinguishing between oncogenes and protooncogenes encoding G-protein α subunits, wherein said probe hybridizes to a region of nucleic acid encoding an amino add selected from the group consisting amino acids corresponding to the Gsα amino acid at positions 49, 201, and 227.

17. An ollgonucleotide primer pair for the amplification of a subsequence of a nucleic add encoding a G-protein α subunit wherein the primer pair is selected from the group consisting of: JFL69 and JFL70; JFL135 and JFL136; JFL228 and JEL135;

JFL229 and JFL136; JFL226 and JFL227; JFL223 and JFL224; JL54 and JL57;

JFL109 and JLF110; JFL110 and JFL112; JFL113 and JFL114; JFL115 and JFL113;

SP9 and SP10; JFL139 and SP11; JFL201 and SP15; JL55 and JL56; JFL223 and SP33; and JFL224 and SP34; JFL235 and JFL237; JL55 and JFL212; JFL215 and

JL56; and JFL135 and JFL286.

18. A probe for the detection of a point mutation in a nucleic acid encoding ; segment of a G-protein α subunit wherein the probe is selected from the group consisting of:

5' TA GGT GCT GGA GAA TCT GGT 3';

5' TA GGT GCT AGA GAA TCT GGT 3';

5' TA GGT GCT CGA GAA TCT GGT 3';

5' TA GGT GCT GAA GAA TCT GGT 3';

5' TA GGT GCT GCA GAA TCT GGT 3';

5' TA GGT GCT GTA GAA TCT GGT 3';

5' TT CGC TGC CGT GTC CTG ACT 3';

5' TT CGC TGC TGT GTC CTG ACT 3';

5' TT CGC TGC GGT GTC CTG ACT 3';

5' TT CGC TGC AGT GTC CTG ACT 3';

5' TT CGC TGC CAT GTC CTG ACT 3';

5' TT CGC TGC CCT GTC CTG ACT 3';

5' TT CGC TGC CTT GTC CTG ACT 3';

5' GTG GGT GGC CAG CGC GAT GA 3';

5' GTG GGT GGC TAG CGC GAT GA 3';

5' GTG GGT GGC GAG CGC GAT GA 3';

5' GTG GGT GGC AAG CGC GAT GA 3';

5' GTG GGT GGC CGT CGC GAT GA 3';

5' GTG GGT GGC CCG CGC GAT GA 3';

5' GTG GGT GGC CTG CGC GAT GA 3';

5' GTG GGT GGC CAC CGC GAT GA 3';

5' GTG GGT GGC CAT CGC GAT GA 3';

5' TC AGA ACT AGA GTG AAA ACT 3';

5' TC AGA ACT AGC GTG AAA ACT 3';

5' TC AGA ACT AGT GTG AAA ACT 3';

5' TC AGA ACT GGA GTG AAA ACT 3';

5' TC AGA ACT ACA GTG AAA ACT 3';

5' TC AGA ACT AAA GTG AAA ACT 3';

5' TC AGA ACT ATA GTG AAA ACT 3';

5' TG GGA GGT CAG AGA TCT GAG 3':

5 TG GGA GGT GAG AGA TCT GAG 3';

5' TG GGA GGT AAG AGA TCT GAG 3':

5' TG GGA GGT CTG AGA TCT GAG 3';

5' TG GGA GGT CCG AGA TCT GAG 3':

5' TG GGA GGT CGG AGA TCT GAG 3';

5' TG GGA GGT CAT AGA TCT GAG 3';

5' TG GGA GGT CAC AGA TCT GAG 3';

5' TA CGG ACC CGC GTA AAG ACC 3';

5' TA CGG ACC AGC GTA AAG ACC 3';

5' TA CGG ACC GGC GTA AAG ACC 3';

5' TA CGG ACC TGC GTA AAG ACC 3';

5' TA CGG ACC CAC GTA AAG ACC 3';

5' TA CGG ACC CCC GTA AAG ACC 3';

5' TA CGG ACC CTC GTA AAG ACC 3'; 5' GTG GGT GGT CAG CGG TCT GA 3';

5' GTG GGT GGT CTG CGG TCT GA 3';

5' GTG GGT GGT CCG CGG TCT GA 3';

5' GTG GGT GGT CGG CGG TCT GA 3';

5' GTG GGT GGT AAG CGG TCT GA 3';

5' GTG GGT GGT GAG CGG TCT GA 3';

5' GTG GGT GGT CAT CGG TCT GA 3';

5' GTG GGT GGT CAC CGG TCT GA 3';

5' TT CGG ACG AGA GTG AAG ACC 3';

5' TT CGG ACG AGC GTG AAG ACC 3';

5' TT CGG ACG AGT GTG AAG ACC 3';

5' TT CGG ACG GGA GTG AAG ACC 3';

5' TT CGG ACG ATA GTG AAG ACC 3';

5' TT CGG ACG ACA GTG AAG ACC 3';

5' TT CGG ACG AAA GTG AAG ACC 3';

5' GTA GGT GGC CAA AGA TCA GA 3';

5' GTA GGT GGC CAT AGA TCA GA 3';

5' GTA GGT GGC CAC AGA TCA GA 3';

5' GTA GGT GGC CTA AGA TCA GA 3';

5' GTA GGT GGC CCA AGA TCA GA 3';

5' GTA GGT GGC CGA AGA TCA GA 3';

5' GTA GGT GGC AAA AGA TCA GA 3';

5' GTA GGT GGC GAA AGA TCA GA 3';

5' CTC CGA ACC AGG GTC AAA AC 3';

5' CTC CGA ACC GGG GTC AAA AC 3';

5' CTC CGA ACC TGG GTC AAA AC 3';

5' CTC CGA ACC AAG GTC AAA AC 3';

5' CTC CGA ACC ACG GTC AAA AC 3';

5' CTC CGA ACC ATG GTC AAA AC 3';

5' CTC CGA ACC AGC GTC AAA AC 3';

5' CTC CGA ACC AGT GTC AAA AC 3';

5' GTC GGA GGC CAG CGA TCT GA 3';

5' GTC GGA GGC AAG CGA TCT GA 3';

5' GTC GGA GGC GAG CGA TCT GA 3';

5' GTC GGA GGC CCG CGA TCT GA 3';

5' GTC GGA GGC CGG CGA TCT GA 3';

5' GTC GGA GGC CTG CGA TCT GA 3';

5' GTC GGA GGC CAT CGA TCT GA 3';

5' GTC GGA GGC CAT CAC TCT GA 3';

5' TG CGC TCC CGG GAC ATG ACC 3';

5' TG CGC TCC GGG GAC ATG ACC 3';

5' TG CGC TCC TGG GAC ATG ACC 3';

5' TG CGC TCC CAG GAC ATG ACC 3';

5' TG CGC TCC CCG GAC ATG ACC 3';

5' TG CGC TCC CTG GAC ATG ACC 3';

5' TG GGG GGG CAG AGG TCA GAG 3';

5' TG GGG GGG AAG AGG TCA GAG 3'; 5' TG GGG GGG GAG AGG TCA GAG 3';

5' TG GGG GGG CCG AGG TCA GAG 3';

5' TG GGG GGG CTG AGG TCA GAG 3';

5' TG GGG GGG CAC AGG TCA GAG 3';

5' TG GGG GGG CAT AGG TCA GAG 3';

5' TG GGG GGG CGG AGG TCA GAG 3';

5' TCG CTG CCG TGT CCT GGA C 3';

5' TCG CTG CAG TGT CCT GGA CT 3';

5' TCG CTG CGG TGT CCT GGA C 3';

5' TCG CTG CTG TGT CCT GGA CT 3';

5' TCG CTG CCA TGT CCT GGA CT 3';

5' TCG CTG CCT TGT CTT GGA CT 3';

5' TGG GTG GCC AGC GGC GAT GA 3';

5' TGG GTG GCC TGC GCG ATG A 3';

5' TGG GTG GCC CGC GCG ATG 3';

5' TGG GTG GCC GGC GCG ATG 3';

5' TGG GTG GCG AGC GCG ATG A 3';

5' TGG GTG GCT AGC GCG ATG A 3';

5' TGG GTG GCC ATC GCG ATG A 3';

5' TGG GTG GCA AGC GCG ATG A 3'; and

5' TGG GTG GCC ACC GCG ATG 3'.

19. A method for detecting a point mutation, if present in a nucleic acid encoding a G-protein α subunit in a sample comprising:

(a) hybridizing a G-protein α subunit probe to said sample, and

(b) determining whether hybridization has occurred.

20. A kit for detecting point mutations in a nucleic acid encoding a G-protein α subunit, comprising, in a separate container

(a) a G-protein primer pair suitable for providing an ampllfied G-protein DNA segment in a PCR reaction;

(b) a G-protein probe comprising a wfld type sequence which will hybridized to said DNA segment if it is a wild type; and

(c) a G-protdn probe comprising a sequence containing a point mutation for detecting a point mutation if present in said DNA segment.