WO2000065095A2

WO2000065095A2 - Methods and compositions for performing array based hybridization assays

Info

Publication number: WO2000065095A2
Application number: PCT/US2000/010894
Authority: WO
Inventors: Alex Chenchik; Alexander Munishkin
Original assignee: Clontech Laboratories, Inc.
Priority date: 1999-04-23
Filing date: 2000-04-20
Publication date: 2000-11-02
Also published as: EP1196633A2; WO2000065095A3

Abstract

Methods for performing array-based hybridization assays, as well as sets of control target nucleic acids for use in the subject methods and kits comprising the same, are provided. In the subject methods, test and control sets of target nucleic acids are contacted with an array(s), where the test and control sets of target nucleic acids may or may not be labeled differentially. The subject control sets of nucleic acids contain a plurality of nucleic acids, with at least one nucleic acid of the set binding selectively to at least a subset of the probe nucleic acids present on an array.

Description

METHODS AND COMPOSITIONS FOR PERFORMING ARRAY BASED

HYBRIDIZATION ASSAYS

INTRODUCTION Technical Field

The field of this invention is nucleic acid arrays Background of the Invention

"Biochips" or arrays of binding agents, such as ohgonucleotides and peptides, have become an increasingly important tool in the biotechnology industry and related fields These binding agent arrays, in which a plurality of binding agents are deposited onto a support surface, often a solid support surface, m the form of an array or pattern, find use in a variety of applications, including drug screening, nucleic acid sequencing, mutation analysis, and the like One important use of biochips is in the analysis of differential gene expression, where the expression of genes in different cells, normally a cell of interest and a control, is compared and any discrepancies in expression are identified In such assays, the presence of discrepancies indicates a difference in the classes of genes expressed m the cells being compared

In methods of differential gene expression, arrays find use by serving as a substrate to which is bound polynucleotide "probe" fragments One then obtains "targets" from analogous cells, tissues or organs of a healthy and diseased organism The targets are then hybridized to the immobilized set of polynucleotide "probe" fragments Differences between the resultant hybridization patterns are then detected and related to differences in gene expression in the two sources

Because of the varied and important information that arrays can provide, as well as the many potential applications of such devices, the use of arrays in research, diagnostic and related applications has grown considerably and is expected to continue to do so A variety of different array technologies have been developed in order to meet the growing need of the biotechnology industry, as evidenced by the extensive number of patents and references listed m the relevant literature section below

However, there are disadvantages with current protocols For example, the efficiency of hybridization of target nucleic acids to the array can be limited by experimental limitations, e g , different target nucleic acids can have different hybridization efficiencies to the probe nucleic acids of the array Differences in hybridization efficiency result in differences in the intensity of hybridization to different probe nucleic acids of the array, even though the targets are present in equivalent concentrations Alternatively, where two arrays are employed in a particular application, e g in gene expression analysis, variation m the quality of array (reproducibi ty of array production), and in assay conditions between the different arrays can preclude direct comparison of data obtained on the arrays, since conditions such as hybridization time, probe labeling, detection procedures may differ, and variations between the arrays may be present Furthermore, it is difficult to compare data generated using different types of o gonucleotide or polynucleotide- b.ased arrays Concentration of target nucleic acids in a sample cannot be compared between arrays produced by different methods and/or manufacturers based on intensity of signals because the set of probe sequences often differs between arrays As a result, current array technology is used mainly for discovery of differentially-expressed genes rather than for any specific quantitative assay Two formats are generally employed (a) comparison of two hybridization patterns to each other and (b) simultaneous hybridization to the same array of two different targets derived from two different biological sources and labeled by different labels In the latter approach, which is more commonly employed, fold differences in gene expression between the two samples are often measured

As such, there is a continued need for the development of additional array-based protocols Of particular interest would be the development of an array-based methodology that incorporates an internal calibration standard, where such a method would eliminate variations resulting from the quality of the array, the type of the array, the quality of the assay conditions, and the like In addition, there is a need for an array-based protocol that provides quantitative data about target concentration, and a corresponding method of quantification to allow more accurate comparison of data between arrays

Relevant Literature

Patents and patent applications describing arrays of biopolymeπc compounds and methods for their fabrication include 5,242,974, 5,384,261 , 5,405,783, 5,412,087, 5,424,186, 5,429,807, 5,436,327, 5,445,934, 5,472,672, 5,527,681 , 5,529,756, 5,545,531, 5,554,501, 5,556,752, 5,561,071, 5,599,895, 5,624,711 , 5,639,603, 5,658,734, WO 93/17126, WO 95/11995, WO 95/35505, EP 742 287, and EP 799 897

Patents and patent applications describing methods of using arrays m various applications include 5,143,854, 5,288,644, 5,324,633, 5,432,049, 5,470,710, 5,492,806, 5,503,980, 5,510,270, 5,525,464, 5,547,839, 5,580,732, 5,661,028, 5,800,992, WO 95/21265, WO 96/31622, WO 97/10365, WO 97/27317, EP 373 203, and EP 785 280

Other references of interest include Atlas Human cDNA Expression Array I (April 1997) CLONTECHniques XII 4-7, Lockhart et al , Nature Biotechnology (1996) 14 1675-1680, Shena et al , Science (1995) 270 467-470, Schena et al , Proc Nat'l Acad Sci USA (1996)93 10614- 10619, Shalon et al , Genome Res (1996) 6 639-645, Milosavljevic et al , Genome Res (1996) 6 132-141 , Nguyen et al , Genomics (1995)29 207-216, Pietu et al , Genome Res (1996) 6 492- 503, Zhao et al , Gene (1995) 166 207-213, Chalifour et al , Anal Biochem (1994) 216 299-304, Heller et al , Proc Nat'l Acad Sci USA (1997) 94 2150-2155, Schena, M , BioAssays (1996) 18 427-431 , Lander, Nature Genetics Supplement (January 1999) 21 3, Southern et al , Nature Genetics Supplement (January 1999) 21 5, Duggan et al , Nature Genetics Supplement (January 1999) 21 10, Bassett et al , Nature Genetics Supplement (January 1999) 21 51 , Debouck & Goodfellow, Nature Genetics Supplement (January 1999) 21 48, Hacia, Nature Genetics Supplement (January 1999) 21 42, Cole et al , Nature Genetics Supplement (January 1999) 21 38, Brown & Botstein, Nature Genetics Supplement (January 1999) 21 33, Bowtell, Nature Genetics Supplement (January 1999) 21 25, Lipshutz et al , Nature Genetics Supplement (January 1999) 21 20, Cheung et al , Nature Genetics Supplement (January 1999) 21 15, Ermolaeva et al , Nature Genetics (September 1998) 20 19, Rajeevan et al , J Histochemistry & Cytochemistry (1999) 47 337-342, Iyer et al , Science (January 1, 1999) 283 83, Eisen et al , Proc Nat'l Acad Sci USA (December 1998) 95 14863-14868, Trenkle et al , Nuc Acids Res (1998) 26 3883-3891 , Pmkel et al , Nature Genetics (October 1998) 20 207, Persidis, Nature Biotechnology (October 1998) 16 981 , Roth et al , Nature Biotechnology (October 1998) 16 939, Service, Science (October 16, 1998) 282 396, Geng et al , Biotechniques (September 1998) 25 434-438, Chen et al , Genomics (1998) 51 313-324, Welford et al , Nuc Acids Res (1998) 26 3059-3065, Nguyen et al , Nuc Acids Res (1998) 26 4249-4258, Hacia et al , Nature Genetics (February, 1998) 18 155, Hacia et al , Nucleic Acids Res (1998) 26 4975-4982, and Gmgeras et al , Genome Research (1998) 435 SUMMARY OF THE INVENTION Methods for performing array-based hybridization assays, as well as sets of nucleic acids for use as internal calibration standards and kits comprising the same, are provided In the subject methods, a control set of target nucleic acids is used in addition to the test set of target nucleic acids The control set of target nucleic acids employed in the subject methods contains at least one nucleic acid capable of selectively binding to at least a subset of, if not all of, the probe compositions present on the array with which it is used, e g at least a subset or portion of the probe nucleic acids present on the array are represented in the control set In the subject methods, the test and control sets of target nucleic acids may be hybridized to the same or different arrays, where the control and test set of target nucleic acids may be labeled the same or differently, depending on the particular protocol, e g whether the two sets are hybridized sequentially, whether a single array or two arrays are employed, etc Also provided are sets of control nucleic acids for use in the subject methods, as well kits for use m performing the subject methods The subject methods and compositions find particular use in quantitative differential gene expression analysis, both on a single array and for comparison of data between arrays

DEFINITIONS

The term "nucleic acid" as used herein means a polymer composed of nucleotides, e g deoxyπbonucleotides or πbonucleotides The terms "πbonucleic acid" and "RNA" as used herein means a polymer composed of πbonucleotides

The terms "deoxyπbonucleic acid" and "DNA" as used herein means a polymer composed of deoxyribonucleotides

The term "o gonucleotide" as used herein denotes single stranded nucleotide multimers of from about 10 to 100 nucleotides m length

The term "polynucleotide" as used herein refers to single or double stranded polymer composed of nucleotide monomers of greater than about 120 nucleotides in length up to about 1000 nucleotides m length

The term "array" refers to a plurality of nucleic acids stably associated with the surface of a solid support, where at least a portion of the stably associated nucleic acids are probe nucleic acids

The term "probe nucleic acid" refers to nucleic acids stably associated with the surface of a solid support which correspond to a target gene of interest in a sample Probe nucleic acids are not random nucleic acids or nucleic acids that corresponding to genes that are not of interest in a sample, e g housekeeping genes, genes that are widely expressed among tissues, etc

The term "target nucleic acid" refers to a nucleic acid isolated from a physiological sample which comprises a gene(s) of interest The term "control target nucleic acid" refers to a nucleic acid that is complementary to a probe nucleic acid on an array, where the control target nucleic acid is part of a set of target nucleic acids in which the amount of each nucleic acid in the set is known and in which each probe nucleic acid corresponding to a gene of interest present on an array is represented Control target nucleic acid is (in a preferred embodiment) synthetic RNA made artificially in vitro and not isolated from a biological source As such, control target nucleic acid is distinguished from test target and test control, which are derived from a biological source Control target nucleic acids are not the same as test control nucleic acids, which are derived from the same physiological sample as the test target nucleic acids and are therefore unique and specific for that sample A further contrast been test control nucleic acids and the subject control target nucleic acids is that test control nucleic acids do not correspond, e g they are not complementary to, probe nucleic acids on an array, but instead correspond to non-probe nucleic acids on the array, e g housekeeping genes, etc

DESCRIPTION OF THE SPECIFIC EMBODIMENTS Methods and compositions for performing quantitative array-based hybridization assays are provided In the subject methods, both a test set of target nucleic acids and a control set of target nucleic acids are employed, where the control set of target nucleic acids is characterized by including at least one nucleic acid that binds selectively to probe nucleic acids in at least a subset of the probe nucleic acids present on the array, I e at least a subset of, if not all of, the probe nucleic acids present on the array employed in the method is represented m the control set

Depending on the protocol employed, e g whether a single array is employed, whether the control and test sets are hybridized sequentially or simultaneously to the same array, etc , the control and test sets of target nucleic acids may be labeled with the same label or differentially labeled, l e labeled such that they are simultaneously distinguishable from each other Also provided are sets of control target nucleic acids for use m the subject methods, as well kits comprising the sets of control target nucleic acids for use in performing the subject methods The subject invention finds use in a variety of different applications, including gene expression analysis In further describing the subject invention, the methods will be described first, followed by a description of the subject kits and a discussion of representative applications in which the subject invention finds use

Before the subject invention is described further, it is to be understood that the invention is not limited to the particular embodiments of the invention described below, as variations of the particular embodiments may be made and still fall withm the scope of the appended claims It is also to be understood that the terminology employed is for the purpose of describing particular embodiments, and is not intended to be limiting Instead, the scope of the present invention will be established by the appended claims

In this specification and the appended claims, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood to one of ordinary skill in the art to which this invention belongs

METHODS As summarized above, the subject invention is directed to array-based hybridization assays in which a test set of labeled target nucleic acids is hybridized to an array of probe nucleic acids A critical feature of the subject invention is the use of a control set of target nucleic acids, where at least a subset of the probe nucleic acids, and m certain preferred embodiments all of the probe nucleic acids, present on the array are represented in the control set, l e the control set includes a nucleic acid capable of hybridizing to each different probe nucleic acid of at least a subset of all of the different probe nucleic acids of the array with which it is employed In general, the subject methods include the following steps (1) procurement of the test and control target nucleic acids, (2) hybridization to an array(s), (3) washing, and (4) detection Each of these steps of the subject methods is described in greater detail below

Procurement of Control and Test Target Nucleic Acids

The first step in the subject methods is the procurement of the labeled control and test sets of target nucleic acids Both the control and test sets of nucleic acids are pools, mixtures or collections of a plurality of distinct nucleic acids that differ by sequence The number of distinct nucleic acids in the control set of target nucleic acids is at least about 50, usually at least about 100 and more usually at least about 200, where the number of distinct nucleic acids in a given set may be as high as 20,000 or higher, but will typically not exceed about 10,000 and usually will not exceed about 5,000 The number of distinct nucleic acids in the test set is generally at least about 1000 and generally less than about 50,000, where the number generally ranges from about 5,000 to 20,000 The control and test sets of target nucleic acids are now described separately in greater detail

Control Sets of Target Nucleic Acids and Sets Thereof

The control sets of target nucleic acids are collections or pools of control target nucleic acids, as mentioned above A control target nucleic acid may be any nucleic acid capable of hybridizing selectively to its respective probe nucleic acid on an array As such, the nucleotides that make up the control target nucleic acid molecule may be made up of nucleotides that are naturally occurring or nucleotides that are synthetically produced analogues that are capable of forming base-pair relationships with naturally occurring base pairs

As such, the control target nucleic acids may include polymers ot ribonucleotides and deoxyribonucleotides, with the ribonucleotide and/or deoxyribonucleotides being connected together via 5 ' to 3 ' linkages. Control nucleic acids of the invention may be ribonucleic acids, for example sense or antisense ribonucleic acids, full-length or partial fragments of cRNA, full-length or partial fragments ot mRNA, and/or ribo-oligonucleotides Alternatively, control target nucleic acids of the invention may be deoxyribonucleic acids, preferably single-stranded full-length or fragments of sequences encoding the corresponding mRNAs. The form of the control and target nucleic acids should be chosen so that they are complimentary to and form appropriate Watson-Crick hydrogen bonds with probes present in an array. For example if probe sequences correspond in sequence to mRNA, then target sequences should be complementary, e.g. antisense or complementary RNA (cRNA)

As mentioned above, the control target nucleic acids may be polymers of synthetic nucleotide analogs Such control target nucleic acids may be preferred in certain embodiments because of their superior stability under .assay conditions Modifications m the native structure, including alterations in the backbone, sugars or heterocychc bases, have been shown to increase intracellular stability and binding affinity Among useful changes in the backbone chemistry are phosphorothioates, phosphorodithioates, where both of the non-bridgmg oxygens are substituted with sulfur, phosphoroamidites, alkyl phosphotπesters and boranophosphates Achiral phosphate derivatives include 3'-O'-5'-S-phosphorothιoate, 3'-S-5'-O- phosphorothioate,

3'-CH₂-5'-0-phosphonate and 3'-NH-5'-O-phosphoroamιdate Peptide nucleic acids replace the entire πbose phosphodiester backbone with a peptide linkage Locked nucleic acids give additional conformational stability of sugar moiety due to additional bonds between 2'-carboxιl and 5'- carboxil or 4'-carboxιl groups of deoxynbose Sugar modifications are also used to enhance stability and affinity The α-anomer of deoxynbose may be used, where the base is inverted with respect to the natural β-anomer The 2'-OH of the πbose sugar may be altered to form 2'-O-methyl or 2'-O-allyl sugars, which provides resistance to degradation without comprising affinity Modification of the heterocychc bases that find use in the method of the invention are those capable of appropriate base pairing Some useful substitutions include deoxyuπdme for deoxythymidme, 5-methyl-2'-deoxycytιdιne and 5-bromo-2'-deoxycytιdme for deoxycytidme 5- propynyl-2'-deoxyuπdιne and 5-propynyl-2'-deoxycytιdιne have been shown to increase affinity and biological activity when substituted for deoxythymidme and deoxycytidme, respectively Examples of non-naturally occurring bases that are capable of forming base-pairing relationships include, but are not limited to, aza and deaza pyπmidine analogues, aza and deaza purine analogues, and other heterocyclicbase analogues, wherein one or more of the carbon and nitrogen atoms of the purine and pyrimidme rings have been substituted by heteroatoms, e.g. , oxygen, sulfur, selenium, phosphorus, and the like. In many preferred embodiments, the control target nucleic acids are structurally as similar as possible to the test target nucleic acids that are employed in the assay, e.g. , both sets of target and control nucleic acids are labeled cDNAs. In other words, the structure of the control target nucleic acids should be similar to that of the test target nucleic acids in order to maximally imitate the hybridization of the test target nucleic acids Each target nucleic acid of the control set should bind to its corresponding probe nucleic acid with selectivity and sensitivity A nucleic acid that selectively binds with its corresponding probe nucleic acid is at least 10 times, preferably at least 100 times, and more preferably at least 1000 times more likely to bind with its designated probe nucleic acid than to a non-specific nucleic acid, and preferably any other sequence present on the array Non-specific nucleic acids include those of random sequence, coding sequences found in a particular array other than the designated probe nucleic acid, and coding sequences of non-probe sequences specific to the organism from which the probe nucleic acids are derived

Control target nucleic acids of the invention also display sufficient sensitivity upon bindmg with their designated probe nucleic acids By "sufficient sensitivity" is meant that binding of the probe nucleic acid is significantly greater than the binding of background nucleic acids of random sequence, where the strength of binding is at least 10 times, preferably at least 100 times, and more preferably 500 times greater than the recognition of background nucleic acids of random sequence In many preferred embodiments, the nucleotide sequences of the subject control target nucleic acids are chosen with algorithms, where such algorithms are described in detail m PCT publication WO 97/10365 and PCT/US96/14839, the disclosures of which are herein incorporated by reference

A critical feature of the control sets of target nucleic acids is that they include at least one target nucleic acid complementary to each probe nucleic acid present in at least a subset of the probe nucleic acids present on the array with which they are used In other words, at least a subset of the probe nucleic acids present on a given array are represented in the control set intended for use with the given array, where the probe members of the subset preferably correspond to genes for which quantitative analysis is desired By at least a subset is meant that at least 20, usually at least 30 and more usually at least 50 of the probe nucleic acids present on the array are represented in the control set In certain embodiments, at least 20%, usually at least 30% and more usually at least 50%) of the probe nucleic acids present on the array are represented in the array In many preferred embodiments, all of the probe nucleic acids present on the array are represented in the control set For example, where a given array includes 500 distinct probe nucleic acids which are distinct from each other based on sequence, a control set for use with this particular array includes at least 500 different target nucleic acids — one for each probe nucleic acid on the array

Non-probe sequences on the array may not have a target nucleic acid in the control set, e g array sequences such as orientation sequences, negative and positive control sequences, etc that may be present on an array In general, control target nucleic acids are not necessary for sequences on an array which do not require quantification, where a particular protocol is intended to provide qualification data only

The number of unique control target nucleic acids in the set or pool of control target nucleic acids will, in most embodiments, be at least about 20, usually at least about 50, more usually at least about 100, where the number may be as high as 1000 or higher, but m many embodiments will not exceed about 800, and usually will not exceed about 750

Of particular interest in many embodiments of the invention are control sets that include a representative or representational number of target nucleic acids As the subject control sets comprise a representational number of target nucleic acids, the total number of different target nucleic acids in any given set will be only a fraction of the total number of different or distinct RNAs in the sample from which the test set of target nucleic acids (described infra) is derived, where the total number of target nucleic acids in the control set will generally not exceed 80%, usually will not exceed 50% and more usually will not 20% of the total number of distinct RNAs in the original sample from which the test set of target nucleic acids is derived, e g the total number of distinct messenger RNAs (rnRNAs) m the original sample Any two given RNAs m a sample will be considered distinct or different if they comprise a stretch of at least 100 nucleotides m length in which the sequence similarity is less then 98%, as determined using the FASTA program (default settings) As the sets of control target nucleic acids comprise only a representational number of target nucleic acids compared to the mRNA population of the sample from which the test set is derived, with sources comprising from 5,000 to 50,000 distinct RNAs, the number of different target nucleic acids m the control set typically ranges from about 20 to 10,000, usually from about 50 to 2,000 and more usually from about 75 to 1500 In the preferred embodiment the representative number of target nucleic acid can be generated using a mixture of gene-specific primers, as described in 08/859,998, 08/974,298, 09/225,998, the disclosures of which are herein incorporated by reference

Control target nucleic acids can be the same length, shorter or longer than their corresponding probe sequences on the array or test nucleic acid in the solution (if present) However, each control target nucleic acid should have a least partial complementarity to its corresponding probe nucleic acid and at least partial sequence identity with its corresponding test target nucleic acids (if present) In addition, the control target nucleic should have structural and hybridization characteristics very similar to its corresponding test target nucleic acid, l e , it should have similar hybridization efficiencies, similar kinetics with complementary probe sequences, similar background hybridization with other sequences, etc For example, where the control set of target nucleic acids comprises labeled cDNAs reverse transcribed from a control set of a representative pool of synthetic RNAs, the test target nucleic acids will also generally be labeled cDNAs reverse transcribed from rnRNAs, e g synthetic mRNAs

Each target nucleic acid in the control set may be the same length as its corresponding probe nucleic acid, longer than its corresponding probe nucleic acid or shorter than its corresponding probe nucleic acid In general, the length of each target nucleic acid m a given control set is at least about 25 nucleotides, usually at least about 50 nucleotides, and more usually at least about 100 nucleotides, where the length could be as a long as 2 kb or longer, but will generally not exceed about 1 kb and more usually will not exceed about 800 nucleotides

A critical feature of control sets of target nucleic acids is that the concentration of each control target nucleic acid present in the set be known

The control sets of target nucleic acids are further characterized m that at least two different gene functional classes are represented in a given control set, where the number of different functional classes of genes represented in the a given control set will generally be at least 3, and will usually be at least 5 In other words, the sets of control target nucleic acids comprise nucleotide sequences complementary to RNA transcripts of at least 2 gene functional classes, usually at least 3 gene functional classes, and more usually at least 5 gene functional classes Gene functional classes of interest include oncogenes, genes encoding tumor suppressors, genes encoding cell cycle regulators, stress response genes, genes encoding ion channel proteins, genes encodmg transport proteins, genes encoding intracellular signal transduction modulator and effector factors, apoptosis related genes, DNA synthesis/recombination/repair genes, genes encoding transcription factors, genes encoding DNA-binding proteins, genes encoding receptors, including receptors for growth factors, chemokines, mterleukms, interferons, hormones, neurotransmitters, cell surface antigens, cell adhesion molecules etc , genes encoding cell-cell communication proteins, such as growth factors, cytokines, chemokines, mterleukms, interferons, hormones etc , and the like

Of particular interest are control sets of target nucleic acids in which each of the genes collectively listed in the tables of the following applications are represented in the control set U S Patent Application Serial No 08/859,998, U S Patent Application Serial No 08/974,298, U S Patent Application Serial No 09/225,998, U S Application Serial No 09/221,480, U S Application Serial No 09/222,432, U S Application Serial No 09/222,436, U S Application Serial No 09/222,437, U S Application Serial No 09/222,251 , U S Application Serial No 09/221,481 , U S Application Serial No 09/222,256, U S Application Serial No 09/222,248, and U S Application Serial No 09/222,253, the disclosures of which are incorporated herein by reference

Another critical feature of the control target nucleic acids is that they are synthetic nucleic acids and not isolated from a biological source The control target nucleic acids may be generated using any convenient protocol, including reverse transcription protocols (e g using AMV or MoMLV reverse transcriptase), bacteriophage RNA polymerase (T7 RNA polymerase, T3 RNA polymerase, etc ) mediated transcription, PCR protocols, ohgonucleotide synthesis protocols (I e nucleotide chemistry), and the like In a preferred embodiment, the control target nucleic acid sequences are generated using cDNA fragments cloned into appropriate expression vectors using a set of a representative number of gene specific primers, as described U S Application Serial nos 08/859,998, 08/974,298,09/225,998, the disclosures of which are incorporated herein by reference These cloned cDNAs are then used to produce RNA control targets using techniques such as PCR and/or bacteriophage RNA polymerase mediated transcription Of particular interest are applications m which the gene specific primers used to generate the control sets are the same as the gene specific primers used to generate the probe nucleic acids on the array with the control set is employed

After synthesis, each control target nucleic acid is quantitated using procedures such as spectrophotometry, fluorescence me<asurement, etc Known quantitative amounts of each control target nucleic acid are then mixed together for use in the hybridization assays, described in greater detail infra In a preferred embodiment, the control target nucleic acids are mixed together in equal molar amounts, at predetermined ratios, at equal weight amounts, etc, where m many embodiments they will be mixed together in equal weight amounts, such that the amount of each individual target nucleic acid m the control set is the same as any every other individual target nucleic acid m the set

Test Target Nucleic Acids and Sets Thereof

Turning now to the test target nucleic acids, the test target nucleic acids are generally isolated trom a biological sample (cells, tissues, organs, etc ) preparation, and then converted to other nucleic acids using known in the art technology, such as PCR, reverse transcription, etc , e g mRNA, cDNA, PCR products, cRNA, and the like The target nucleic acids may be isolated from a tissue or cell of interest using any method known in the art Total RNA or its trans cπptionally active fraction mRNA can be isolated from a tissue and labeled and used directly as a target nucleic acid, or it may be converted to a labeled cDNA, cRNA, etc via methods such as reverse transcription, transcription and/or PCR Generally, such methods will employ the use of ohgonucleotide primers, and the primers can be anchored by bacteriophage RNA polymerase promoter The primers may be designed to copy a large spectrum of RNA species, e g olιgo(dT) primers or random hexamers, or designed specifically to copy a subset of genes of interest After the copying step, I e conversion of mRNA to cDNA, cDNA can be amplified by PCR or by linear amplification using bacteriophage RNA polymerase mediated transcription As with the control target nucleic acids, in a preferred embodiment the test target nucleic acid sequences are generated using a set of a representative number of gene specific primers, as described U S Application Serial nos 08/859,998, 08/974,298, 09/225,998, the disclosures of which are incorporated herein by reference

Additional Features of the Control and Test Sets of Target Nucleic Acids

Depending on the particular assay protocol with which the subject test and control sets of target nucleic acids are employed, the test and control sets of target nucleic acids may be labeled with the same label, such that the test and control sets cannot be distinguished from one another, or the test and control sets of target nucleic acids may be differentially labeled, such that the two sets are readily distinguishable from each other

As such, in certain embodiments, the test and control sets of target nucleic acids are differentially labeled By "differentially labeled" is meant that the test and control sets of target nucleic acids are labeled differently from each other such that they can be simultaneously distinguished from each other For example, where one has a control set of target nucleic acids and test set of target nucleic acids, each target nucleic acid in the test set will be labeled with the same first label and each target nucleic acid in the control set will be labeled with the same second label that is different and distinguishable from the first label Likewise, where two control sets are employed in the method, each target nucleic acid in the second control set will be labeled with a third label different and distinguishable from both the first and second label

In yet other embodiments, the test and control sets of target nucleic acids are labeled with the same label, so as to be indistinguishable from each other When the test and control sets of target nucleic acids are labeled with the same label, each target nucleic acid of each set is labeled with the same label

A variety of different protocols may be used to generate the labeled target nucleic acids, as is known in the art, where such methods typically rely on the enzymatic generation of labeled target nucleic acid using an initial primer and template nucleic acid Labeled primers can be employed to generate the labeled target Alternatively, label can be incorporated into the target nucleic acid during first strand synthesis or subsequent synthesis, labeling or amplification steps m order to produce labeled target Label can also be incorporated directly to mRNA using chemical modification of RNA with reactive label derivatives or enzymatic modification using labeled substrates Representative methods of producing labeled target are disclosed m U S Application Serial nos 08/859,998, 08/974,298, 09/225,998, the disclosures of which are incorporated herein by reference.

A variety of different labels may be employed, where such labels include fluorescent labels, lsotopic labels, enzymatic labels, particulate labels, etc For example, suitable labels include fluorochromes, e g fluorescem lsothiocyanate (FITC), rhodamine, Texas Red, phycoerythrm, allophycocyanin, 6-carboxyfluoresceιn (6-FAM), 2',7'-dιmethoxy-4',5'- dιchloro-6-carboxyfluorescem (JOE), 6-carboxy-X-rhodamme (ROX), 6-carboxy-2',4',7',4,7- hexachlorofluorescein (HEX), 5-carboxyfluoresceιn (5-FAM) or N,N,N',N'-tetramethyl-6- carboxyrhodamine (TAMRA), cyanme dyes, e g Cy5, Cy3, BODIPY dyes, e g BODIPY 630/650, Alexa542, etc Suitable isotopic labels include radioactive labels, e g ³²P, ^3jP, ^3:>S, ³H other suitable labels include size particles that possess light scattering, fluorescent properties or contain entrapped multiple fluorophores The label may be a two stage system, where the target DNA is conjugated to biotm, haptens, etc having a high affinity binding partner, e g avidm, specific antibodies, etc The binding partner is conjugated to a detectable label, e g an enzymatic label capable of converting a substrate to a chromogemc product, a fluorescent label, and isotopic label, etc

Any combination of labels, e g first and second labels, first, second and third labels, etc , may be employed for the test and control target sets, provided the labels are distinguishable from one another Examples of distinguishable labels are well known in the art and include two or more different emission wavelength fluorescent dyes, like Cy3 and Cy5, or Alexa 542 and Bodipy 630/650, two or more isotopes with different energy of emission, like ³²P and ³³P, labels which generate signals under different treatment conditions, like temperature, pH, treatment by additional chemical agents, etc , and labels which generate signals at different time points after treatment Using one or more enzymes for signal generation allows for the use of an even greater variety of distinguishable labels based on different substrate specificity of enzymes (e g alkaline phosphatas e/p eroxidas e)

Array Hybridization The next step in the subject methods is the hybridization of the test and control sets of target nucleic acids to a nucleic acid array(s) The nucleic acid arrays employed in the subject hybridization assays have a plurality of nucleic acid spots, and preferably in many embodiments polynucleotide spots, stably associated with a surface of a solid support, where the solid support may be rigid, e g glass, or flexible, e g nylon membrane At least a portion of the nucleic acid spots on the array are made up of probe nucleic acids Arrays with which the subject methods find use include nucleic acid biochips, e g cDNA biochips, RNA biochips, polynucleotide biochips, ohgonucleotide biochips, and the like Of particular interest are the arrays described in U S Patent Application Serial No 08/859,998, U S Patent Application Serial No 08/974,298, U S Patent Application Serial No 08/859,998, U S Patent Application Serial No 08/974,298, U S Patent Application Serial No 09/225,998, U S Application Serial No 09/221,480, U S

Application Serial No 09/222,432, U S Application Serial No 09/222,436, U S Application

Serial No 09/222,437, U S Application Serial No 09/222,251 , U S Application Serial

No 09/221,481 , U S Application Serial No 09/222,256, U S Application Serial No 09/222,248, U S Application Serial No 09/222,253, U S Application Serial No 60/104,179, the disclosures of which are incorporated herein by reference

As mentioned above, in practicing the subject methods the test and control sets of target nucleic acids are hybridized to an array, where the sets of target nucleic acids may be hybridized to the same array or different arrays, where when the sets of target nucleic acids are hybridized to different arrays, all of the different arrays will at least share common arrays of probe nucleic acids, l e they will be identical with respect to their probe nucleic acids

In certain preferred embodiments, the test and control sets of target nucleic acids are hybridized to the same array In such embodiments, the array is hybridized with a test set of labeled target nucleic acids and at least one control set of labeled target nucleic acids In those embodiments where more than one control set of target nucleic acids is employed, the number of different control sets may range from 2 to 6, usually 2 to 4 and more usually 2 to 3 Of particular interest are those embodiments in which 1 or 2 different control sets of target nucleic acids are employed The probe and control sets of target nucleic acids may be hybridized to the array and/or detected simultaneously or sequentially Thus, where a control set and target set are employed, the two sets of target nucleic acids may be combined prior to hybridization and the array hybridized to both simultaneously to minimize potential variability in hybridization conditions For example, a known amount of labeled sets of test target and control target nucleic acids can be added to the same hybridization buffer, and then contacted with one or more arrays simultaneously under hybridization conditions In another example, a known amount of labeled sets of test target and control target nucleic acids are added to the same hybridization mix, and this buffer ahquoted for the separate hybridization of different arrays By storing a quots of the hybridization mix (e g storage at -20 °C or -70 °C), different arrays may be hybridized at different times with approximately the same amounts of target nucleic acid sequences

In the above embodiments where the test and control target nucleic acids are hybridized simultaneously to a given array, labeled test and control target nucleic acids are premixed or pooled prior to contact with the array In a preferred embodiment, mixtures of test and control target nucleic acids have amounts of control and target nucleic acids which are sufficient to generate signals that are at least 10 fold, usually at least 20 fold and more usually at least 50 fold higher than background signals observed with the array The relative amounts of control and test target nucleic acids in the mixture are selected to be sufficient to allow reliable detection of the test sequences complimentary to the probe nucleic acid while at the same time allowing complete binding of the test target nucleic acids with a nofold excess of unbound probe nucleic acid on the array The amount of test nucleic acid present m the mixture is usually determined by available amount of sample and sensitivity of technology employed in a particular protocol Preferably, the amount of test nucleic acid present in the mixture ranges from about 0 01 -100 μg of nucleic acid. e g cDNA, and more usually from about 0 1-10 μg of nucleic acid, e g cDNA In many embodiments, the amount of control target nucleic acid employed in the hybridization protocol is about the same or less than the amount of test target nucleic acid that is employed, where less than typically means 10 fold less, usually 100 fold less and more usually 1000 fold less Of interest are mixtures of labeled nucleic acids that provide for an intensity of signal from each probe nucleic acid in the control detection channel that ranges from about 0 001 -0 1%, usually from about 0 001 to 0 01%) abundance level

Alternatively, one or more arrays may be hybridized with the control and test sets of target nucleic acids sequentially For example, arrays may be hybridized with a hybridization mix containing the labeled test target nucleic acids to allow these molecules uninhibited access to the probe sequences of the array Following this hybridization, control target nucleic acids could then be exposed to the array for use as an internal control The hybridization of the control target nucleic acids may be completely separate from the hybridization of the test target nucleic acids, e g using different hybridization mixes at different times, or the control target sequences may be added to the hybridization buffer containing the test target nucleic acids following an incubation period with the test target nucleic acids The latter is especially appropriate when the test target nucleic acids require a longer hybridization incubation period than the control test nucleic acids When used sequentially, the control and test target nucleic acids may be differentially labeled or labeled with the same label, since detection occurs separately

In yet other embodiments, the test and control sets of target nucleic acids are hybridized to different arrays, where each of the different arrays has an identical population of probe sequences, l e the different arrays do not vary with respect to their probe sequences In such methods, the control and test target nucleic acids may be labeled with the same label so as to be indistinguishable from one another, and discussed above In certain embodiments, the test and control target nucleic acids are hybridized to high throuhput array devices, as described in 5,545,531 and PCT/US99/00248, the disclosures of which are herein incorporate by reference

The control and test sets of target nucleic acids are hybridized to the array(s) by contacting the control and test sets with the array(s) under hybridization conditions By "hybridization conditions" is meant conditions sufficient to promote Watson-Crick hydrogen bonding to occur between the target and probe nucleic acids The hybridization conditions, such as hybridization tme, temperature, wash buffers used, etc can be altered to optimize the efficient and specific binding of the target sequences Test target nucleic acids having sequence similarity to the probes may be detected by hybridization under low stringency conditions, for example, at 50°C and 6^χSSC (0 9 M sodium chlorιde/0 09 M sodium citrate, 1% SDS) and remain bound when subjected to washing at 55°C m 1 xSSC (0 15 M sodium chlorιde/0 015 M sodium citrate, 1% SDS) Test target sequences with sequence identity may be determined by hybridization under stringent conditions, for example, at 60°C or higher and 6^χSSC (15 mM sodium chloride/01 5 mM sodium citrate, % SDS) Preferably, the control target nucleic acids have a region of substantial identity to the provided probe sequences on the array, and bind selectively to their respective probe sequences under stringent hybridization conditions Other suitable hybridization conditions for various nucleic acid pairs are well known to those skilled in the art and reviewed m Mamatis et al , and m PCT WO 95/21944, the disclosure of which is herein incorporated by reference Analysis of the differences m signal generated by two or more sources may be carried out by using multiple arrays with the same or similar probe compositions, one for each set of test target nucleic acids Each array is then hybridized with a labeled set of control target nucleic acids and a labeled set of test target nucleic acids Preferably, the labeling efficiency and amount of control target sequences and test target sequences is approximately equivalent between arrays, e g an equal amount of labeled control target nucleic acids is used to hybridize to each array This is not essential, however, since hybridization of the set of labeled control target nucleic acids functions as an independent internal control for each probed array

Levels of hybridization of test target RNA to the probe compositions can be standardized by comparing the hybridization signal of the test with control target sequences on each array Differences in hybridization of the control target sequences allows a comparison of relative hybridization levels between arrays

Washing

Following hybridization, non-hybridized labeled nucleic acid is removed from the support surface, conveniently by washing, generating a pattern of hybridized nucleic acid on the substrate surface A variety of wash solutions and protocols are known to those of skill m the art and may be used See Sambrook, Fπtsch & Mamatis, Molecular Cloning A Laboratory Manual (Cold Spring Harbor Press)(l 989) Detection

The resultant hybridization patterns of labeled nucleic acids may be visualized or detected in a variety of ways, with the particular manner of detection being chosen based on the particular label of the target nucleic acid, where representative detection means include scintillation counting, autoradiography, fluorescence measurement, colorimetπc measurement, light emission measurement, light scattering and the like

Following detection or visualization, the hybridization patterns generated by control and test target nucleic acids may be compared to identify differences between the signals Where arrays in which each of the different probes corresponds to a known gene are employed, differences in signal intensity can be related to a different target concentration of a particular gene The comparison of the intensity of binding of a test target nucleic acid to a probe sequence can be compared to the intensity of the binding of the corresponding control target sequence, and the measurement converted to a quantitative RNA concentration for that target sample The quantitative RNA levels of the test target can be compared between arrays to identify or confirm differential expression of genes in particular samples

UTILITY

The subject methods find use in, among other applications, standardization of differential gene expression assays Thus, one may use the subject methods in the differential expression analysis of (a) diseased and normal tissue, e g neoplastic and normal tissue, (b) different tissue or tissue types, (c) developmental stage, (d) response to external or internal stimulus, (e) response to treatment, and the like The methods of the subject invention therefore find use in broad scale expression screening for drug discovery, diagnostic and research, such as the effect of a particular active agent on the expression pattern of genes in a particular cell, where such information can be used to reveal drug toxicity, carcmogenicity, etc , environmental monitoring, disease research and the like A number of different tasks can be accomplished with the subject invention, which tasks mclude, but are not limited to detecting relative hybridization of target sequences, calibrating a hybridization assay, harmonizing data between hybridization assays, and testing reagents used in a hybridization assay The subject methods m which control and test sets of target nucleic acids are employed can also be used in the generation of gene expression databases, as the data generated from the subject methods are quantitative, reflect the real RNA concentration rather than intensity of signal, and are independent of the type of array Each of these different aspects of the invention is discussed separately below Detecting Relative Hybridization of Target Sequences

The methods of the present invention are useful in detecting relative levels of hybridization of different genes in a sample by providing a set of internal hybridization controls Since the control set of nucleic acids are of a known sequence, m a known quantity, and of a known specific activity (where in a preferred embodiment the control and test target are labeled with the same specific activity), the level of hybridization of the control nucleic acids can be used to determine the level of expression of each gene m a test sample based on its level of binding to a probe sequence The fact that each probe sequence has its own internal control also allows for the detection of potential expression differences between samples and differences in binding affinities between probe sequences, both on a single array and between arrays Thus, the intensity level of hybridization of a control sequence can be used to calculate the expression level of a gene in a sample based upon the intensity of the test target hybridization to the corresponding probe sequence

Calibrating a Hybridization Assay

The methods of the subject invention also find use in the calibration of hybridization assays Using known concentrations of probe nucleic acid, test target nucleic acids, and control target nucleic acids allows one to optimize the hybridization conditions for a particular use, such as increasing stringency to allow better detection of nucleic acids with some level of sequence homology (e g differential expression between genes from a single family or alternative splice forms for the same gene) The use of the internal standards of the method of the subject invention allows hybridization, labeling procedures, and the like to be optimized for a particular use, which is especially valuable for standardization of large scale of hybridization assays, such as high- throughput screening of biological samples Optimization thus means that one can change hybridization conditions in order to achieve maximal intensity of specific hybridization signals with complimentary probe sequences and minimal level of non-specific hybridization with non- complementary probe sequences

Harmonizing Data Between Hybridization Assays The methods of the subject invention also find use m the harmonization of data between hybridization assays, thus allowing for a direct comparison of expression levels despite potential differences due to variables such as differences in hybridization conditions, differences m sample preparation and even between different types of arrays, differences in quality and performance withm and between different arrays, differences in specific activity of the labeled target sequences, and the like Because each hybridization assay has its internal control for at least a subset of the probe sequences on the array, the data can be compared using ratios of the intensity of the control target nucleic acids and the intensity of the test target nucleic acids Thus, the use of simple mathematical formulations to correct for differences between assays allows the levels of gene expression in these different assays to be adjusted to the same level and then compared in a biologically relevant fashion

Testing Reagents and Equipment. Optimizing Hybridization Wash Conditions, Used in a Hybridization Assay

The methods of the present invention are also useful in determining the efficacy of hybridization reagents Such reagents may be, for example, new reagents, e g different buffer solutions for prehybπdization and hybridization, or established reagents, e g a new batch of a known, commercially available reagent The internal control of the methods of the subject invention provide for two levels of quality assurance upon testing the reagents, basically providing an extra control for determining the efficacy of a reagent in a single hybridization Efficiency means maximum specific signal with minimal level of non-specific signal and background binding to solid surface Other parameters such as temperature, buffer composition, length of hybridization and/ washing times, etc , may be optimized using calibration controls Also, the same calibration target nucleic acids can be used routinely to test and calibrate detection equipment to expected level intensity of signals, thus limiting variability due to functionality of the equipment, and may be used to test and calibrate the quality of arrays for control procedures

Building Gene Expression Databases The subject methods in which control and test target nucleic acids are employed can be used to determine real RNA concentrations in sample This information can m turn be used to compile a gene expression database Use of subject methods in building a database has advantages over databases derived from images or signal intensities Such advantages include the generation of more compact information (number/versus image file), the identification of expression levels that are not dependent on type of array, hybridization conditions, lot of array, etc These advantages are significant m that expression data obtained with the subject methods does not need annotation to be meaningful, and the database generated from the data can be universal, l e it can be generated using data generated in different labs, or at different times, or even using different types of arrays

KITS The present invention also provides kits for performing the subject array-based hybridization assays The subject kits at least include a control set of target nucleic acids, as defined above, or a precursor thereof By "nucleic acid precursor" is meant any nucleic acid from which with the control set may be prepared, e g a set of RNAs encoding the nucleic acids of the control set, plasmids containing nucleic acids for generation of the control set, and the like Labeled cDNA can be derived from these precursors by enzymatic synthesis, or ohgonucleotides chemically synthesized based on sequence information of these precursors The set may contain RNAs that recognizes each probe composition on an array, and such RNAs may be pre-labeled, may be labeled for use with the test target nucleic acids, or may be converted to labeled cDNA for hybridization Kits of the present invention may also contain cDNA or o gonucletodies that selectively binds to the probe compositions of the array to be screened The cDNAs or ohgonucleotides may be pre-labeled, or may be labeled by the user through any convenient protocol, such as the protocol used to generate the labeled test target nucleic acids A kit containing a set of control target RNAs may further contain ohgonucleotides for the production of cDNA In a preferred embodiment, these ohgonucleotides are gene specific primers, particularly gene specific primers that have sequence identical to those that were used in the production of the probe compositions on the array to be used in the particular assay In another embodiment, primers can be oligo dT or random primer, if these primers are used for making test sample target

Kits for carrying out differential gene expression analysis assays are preferred Such kits according to the subject invention will at least comprise the subject sets of nucleic acids The kits may further comprise one or more arrays corresponding to the set of control target nucleic acids Of particular interest are the arrays disclosed in U S Patent Application Serial No 08/859,998, U S Patent Application Serial No 08/974,298, U S Patent Application Serial No 08/859,998, U S Patent Application Serial No 08/974,298, U S Patent Application Serial No 09/225,998, U S Application Serial No 09/221,480, U S Application Serial No 09/222,432, U S Application Serial No 09/222,436, U S Application Serial No 09/222,437, U S Application Serial No

09/222,251 , U S Application Serial No 09/221,481 , U S Application Serial No 09/222,256, U S Application Serial No 09/222,248, U S Application Serial No 09/222,253, U S Application Serial No 60/104, 179, the disclosures of which are incorporated herein by reference The kits may further comprise one or more additional reagents employed m the various methods, such as primers for generating target nucleic acids, dNTPs and/or rNTPs, which may be either premixed or separate, one or more uniquely labeled dNTPs and/or rNTPs, such as biotinylated or Cy3 or Cy5 tagged dNTPs, or other post synthesis labeling reagents, such as chemically active derivatives of fluorescent dyes, enzymes such as reverse transcriptases, DNA polymerases, RNA polymerases and the like, various buffer mediums, e g hybridization and washing buffers, prefabricated probe arrays, labeled probe purification reagents and components, like spin columns, etc , signal generation and detection reagents, e g streptavidm-alka ne phosphatase conjugate, chemifluorescent or chemilummescent substrate, and the like In addition to the sets of nucleic acids, arrays and other components described above in the general description of kits, the assay kit may further include a set of gene specific primers that are employed to generate labeled targets In many embodiments, the set of gene specific primers will be the same primers used to generate the polynucleotide probes that are present on the array to be screened Of particular interest m certain embodiments are kits comprising a set of primers selected from the primers identified as SEQ ID NO 01 - 1372, where in these kits of particular interest, at least twenty, usually at least 50 and more usually at least 100 of the gene specific primers in the kit will be selected from this group of primers identified as SEQ ID NO 01-1372, as shown in Table 1 of U S Patent Application Serial No 08/859,998, the disclosure of which is herein incorporated by reference

The following examples are offered by way of illustration and not by way of limitation

EXPERIMENTAL Example 1 - Identification of differentially expressed genes

An assay to determine relative levels of gene expression in a sample is conducted as follows

A Preparation of human stress cDNA array

236 cDNA fragments corresponding to 236 different human genes were amplified from quick-clone cDNA (CLONTECH) m 236 separate test tubes using a combination of sense and antisense human stress gene-specific primers as described in U S Patent Application Serial No 09/222,256, the disclosure of which is herein incorporated by reference Amplification was conducted in a 100-μl volume containing 2 μl of mixture of 10 Quick-clone cDNAs from placenta, brain, liver, lung, leukocytes, spleen, skeletal muscle, testis, kidney and ovary (CLONTECH), 40 mM Tπcine-KOH (pH 9 2 at 22 °C), 3 5 mM Mg(OAc)₂, 10 mM KOAc, 75 μg/ml BSA, 200 μM of each dATP, dGTP, dCTP and dTTP, 0 2 μM of each sense and antisense gene-specific primers and 2 μl of KlenTaq Polymerase mix Temperature parameters of the PCR reactions were as follows 1 mm at 95 °C followed by 20-35 cycles of 95 °C for 15 sec and 68°C for 2 mm, followed by a 10-mιn final extension at 68 °C PCR products were examined on 1 2% agarose/E Br gels m lx TBE buffer As a DNA size marker a 1 Kb DNA Ladder was used Double stranded (ds) cDNA was then precipitated by addition of a half volume of 4M ammonium acetate (about 35 μl) and 3 7 volumes of 95% ethanol (about 260 μl) After vortexmg, the tube was immediately centπfuged at 14,000 r p m in a microcentπfuge for 20 mm The pellet was washed with 80% ethanol without vortexmg, centπfuged as above for 10 mm, air dried, and dissolved in 10 μl of deiomzed water Yield of ds cDNA after the amplification step was about 5 μg The ds cDNA fragments for all 236 genes were cloned into pBR322 derived vectors using convention blunt end cloning procedure and identity of the clones was confirmed by sequence analysis The ds cDNA inserts with the sequence corresponding 236 genes were amplified by PCR using a combination of antisense and sense gene-specific primers, as described above The ds cDNA was denatured by boiling in alkaline buffer (pH 8-10)(sodιum bicarbonate 300 μM), or DNA was dissolved in 5 M guamdmium isothionate All cDNA probes were transferred m a 384-well plate and deposited onto a glass slide using process analogous to that described in PCT Application Serial No PCT/US99/00248, the disclosure ot which is herein incorporated by reference.

B fsolation and preparation of test target ribonucleic acids from human tissue sample

As described m greater detail below, the above described human array is used to screen a human tissue sample to determine gene expression of each of the 236 genes in a human sample

The tissue sample is tested for expression by isolating mRNA from the sample and subsequently transcribing the mRNA into labeled cDNA This labeled pool of cDNA is then used as test target nucleic acids for hybridization against the probe nucleic acids on the array

Total RNA is isolated from homogenized human tissue using a ATLASPURE™ RNA isolation kit and mRNA is isolated from the total RNA using an ohgo-(dT)-cellulose spin column (both from CLONTECH, Palo Alto CA) according to the manufacturer's protocols.

The mRNA of the tissue sample is used to generate target cDNAs for hybridization to the probe array C Preparation of control target ribonucleic acids

A control set of ribonucleic acids was synthesized by T7 transcription from cDNA fragments corresponding to the 236 genes on the Human Stress array prepared as described above, where each of the cDNA fragments included a T7 promoter The fragments were amplified by PCR in 236 separate tubes as described in section A using a combination of sense and antisense gene specific primers Antisense primers carry a T7 promoter for following transcription resulting in RNA fragments mimicking the corresponding mRNA fragment T7 transcription was accomplished m 236 separate tubes in a 200 μl volume containing 200 ng of the individual cDNA fragment, 80 mM HEPES-KOH (pH 7 5 at 22 °C), 15 mM MgCl₂, 2 mM spermid e, 10 mM DTT, 3 mM each ATP, GTP, CTP, UTP, and 200 units of T7 RNA polymerase (Epicentre Tech , Madison, WI) Incubation of reaction mixture was performed at 37 °C for 2 hours RNA products where examined on a 2% agarose/EtBr gel in 1 xTAE buffer RNA was then precipitated by addition of equal volume of 4M ammonium acetate and 2 volumes of 95% ethanol After vortexmg, the tubes were immediately centnfuged at 14,000 rpm for 20 mm The RNA pellet was dissolved in 100 μl of water and purified by column chromatography using a CHROMA SPIN- 200 column (CLONTECH, Palo Alto, CA) according to the manufacturer's instructions

After purification, the concentration of each RNA fragment sample was determined on a spectrophotometer by UV absorption with 1 optical density unit corresponding to 40 μg of RNA The adjusted equal molar weight concentration for each fragment was then calculated A length of 2200 nucleotides (which is equal to the average length of mRNA) was used as a standard For example, if the real concentration for a particular fragment with a size of 200 nucleotides was lng/μl, its adjusted equal molar weight concentration was 11 ng/μl Then all 236 synthetic RNAs were mixed in equal molar ratio and this mixture of synthetic RNAs was used as a standard, as described below

D cDNA Synthesis

2 μl of RNA dissolved in water at 0 5-1 μg/μl (or appropriate concentration for the control set of RNA) was mixed to 1 μl of l OxCDS primer mix (0 2 mM each, comprising mixture of antisense primers complementary to 236 rnRNAs related to the Stress/toxicology array) The sample m the tube was then mixed by vortexmg and spun down by microcentrifuge The tubes were then incubated in a prewarmed PCR thermal cycler at 70 °C for 2 mm The temperature was then reduced to 50 °C and incubation was continued for another 2 mm The remaining reagents were then added and incubation was continued for 20 mm The final reaction conditions were 50 mM Tπs-HCl pH 8 3, 75 mM KC1, 3 mM MgCl₂, 0 5 mM each of dGTP, dCTP, dATP, 0 1 mM of dTTP and 0 1 mM of allylamine-dUTP, 5 mM DTT, 50 units of MMLV Reverse Transcriptase Reaction was stopped by the addition of 1 μl of termination mix (0 1 M EDTA pH 8 0. 1 mg/ml Glycogen) cDNA was precipitated by the addition of an equal volume of 4M ammonium acetate and 2 volumes of ethanol The pellet was collected by centrifugation, washed with 70% ethanol, dried and dissolved in water

E Labeling of control target and tes t target nucleic acids

The control target nucleic acids and the test target nucleic acids are labeled using differentially detectable fluorescent labels For labeling of the test target nucleic acids, following conversion of the target ribonucleic acids to ammo-modified cDNAs, 1 mg of Cy3 succimmide ester is dissolved in 10 μl of dimethyl sulfoxide, and 10 μl of the test target cDNAs are added to it For labeling of the control target set, 1 mg of Cy5 succimmide ester is dissolved in 10 l of dimethyl sulfoxide and 10 μl of control target cDNAs are added to it Both mixtures are incubated at room temperature overnight

Each labeled set of target nucleic acids is purified separately by column chromatography using a CHROMA SPLN-200 column (CLONTECH, Palo Alto.CA) according to the manufacturer's instructions See also U S Patent Application Serial No 08/859,998, the disclosure of which is herein incorporated by reference

F Contact of the target sequences to the array

The control target and test target cDNAs are then pooled for hybridization to the array To prepare the mixture of labeled control and test target cDNAs, 1 μg of the Cy3 labeled test target cDNA and 1 μg of the Cy5 labeled control target prepared above are combined

G Hybridization and detection of probe sequences on array

A solution of ExpressHyb™ (CLONTECH) and sheared salmon testes DNA (Sigma) is prepared as follows 5 ml of ExpressHyb™ is prewarmed at 50-60°C 0 5 mg of the sheared salmon testes DNA is heated at 95-100 °C for 5 mm, and then chilled quickly on ice Heat- denatured sheared salmon testes DNA is mixed with prewarmed ExpressHyb™ The human cDNA array is placed in a hybridization container, and 1 ml of the ExpressHyb™/salmon DNA solution is added Prehybπdization is conducted for 5 min with continuous agitation at 60 °C Labeled cDNA test and control target nucleic acids, as prepared above, (about 200 μl) are mixed with 2 μl ( 1 μg/μl) of human Cot- 1 DNA , and denaturated at 99 °C for 2 min The mixture is added to the hybridization solution and mixed together thoroughly The container is sealed by sealmg tape Hybridization is allowed to proceed overnight with continuous agitation at 60 °C The hybridization solution is carefully removed and discarded in an appropriate container, and replaced with 10 ml of Wash Solution 1 (2 x SSC, 0 1% SDS) The array is washed for 10 mm with continuous agitation at 65 °C The step is repeated two times Additional 10-mm washes are performed in 10 ml of Wash Solution 2 (0 1 x SSC, 0 1% SDS) with continuous agitation at 65 °C Using forceps, the cDNA array is removed from the container, briefly washed in 0 1 xSSC and excess buffer is removed from surface by centrifugation in a Beckman CS-6R centrifuge at 2000 φm Glass arrays are scanned using a confocal scanning microscope (General Scanning, Watertown, MA) Images are scanned at a resolution of 10 μm per pixel

Example 2 - Determination of differential expression between samples

A method to determine differentially expressed genes in normal and cancerous cells, or treated v untreated cells, is conducted as follows

Two identical human cDNA arrays and a set of control target cDNAs are prepared using the methods as described m Example 1

Two separate sets of test target nucleic acids are produced, one set isolated from a cancerous human tissue, and the second set isolated from the corresponding physiologically normal human tissue Each tissue sample is tested for differential expression by isolating mRNA from each sample and subsequently transcribing the mRNA into labeled cDNA Each labeled pool of cDNA is then used as test target nucleic acids for hybridization against the probe sequences of one of the two arrays, and the levels of expression between the two arrays is compared to determine relative differences in expression between the normal and the cancerous tissues Following preparation, both the set of control target nucleic acids and the two sets of test target nucleic acids, corresponding to the normal and the cancerous tissue, are labeled using detectable fluorescent labels Each set of test target nucleic acids is labeled in a separate reaction with Cy3 For each labeling procedure, an aliquot of test cDNAs is added to a 0 5 ml Eppendorf tubes containing 1 mg of Cy3 succimmide ester dissolved in 10 μl of dimethyl sulfoxide For differential labeling of the control target set, 1 mg of Cy5 succimmide ester is dissolved in 10 μl of dimethyl sulfoxide and 10 μl of cDNA generated from the control set PCR reaction is added to it All of these mixtures are incubated at room temperature overnight Each labeled set of target nucleic acids is purified separately by column chromatography Each array is hybridized using one of the two labeled sets of test target nucleic acids, and one-half of the labeled set of control target nucleic acids A solution of ExpressHyb™ (CLONTECH) and sheared salmon testes DNA (Sigma) is prepared and prewarmed at 55 °C 1 mg of the sheared salmon testes DNA is heated at 95-100 °C for 5 mm, and then chilled quickly on ice Heat-denatured sheared salmon testes DNA is mixed with the prewarmed ExpressHyb™, and 1 ml of the ExpressHyb™/salmon DNA solution is added to each of two separate hybridization containers Prehybndization is conducted for 5 mm with contmuous agitation at 60°C

Each hybridization mixture is prepared by mixing about 200 μl of one set of test target cDNAs synthesized from 1 μg of polyA RNA, 100 μl of the set of control target cDNAs synthesized from lμg of control RNA mix containing 0 01%) of each synthetic control RNA for each probe on the array (the size of each synthetic RNA is adjusted to average size of mRNA, 1 e , 2 2 kb mixed together and then equal molar weight % was adjusted to 0 01%), and 2 μl ( 1 μg/μl) of human Cot-I DNA The mixture is denatured at 99 °C for 2 min prior to hybridization The hybridization mixture containing the test nucleic acids from the cancerous tissue is added to the prehybndization solution of the first array, and the hybridization mixture containing the test nucleic acids from the normal tissue is added to the prehybndization solution of the second array Each array is also contacted with the labeled control set of target nucleic acids Each container is sealed, and hybridization conditions and detection are carried out as described m Example 1 Direct comparisons of the amounts of expression on each array are carried out by determining the ratio of the intensity of the test target hybridization with the intensity of the control target hybridization Once determined, these ratios allow for correction of experimental variation between the two arrays, and thus allow a direct comparison of the levels of gene expression in the cancerous and non-cancerous samples A set of housekeeping control probe signals is used in order to adjust the intensity of both channels to each other

Example 3

Synthetic RNA corresponding to the Human Stress array was synthesized for 236 fragments arrayed on membrane or glass RNA fragments were mixed in equal molar ratio and this mixture was used for further experiments cDNA from a control mixture of RNA was labeled with Cy5 fluorescent dye as described earlier cDNA from placenta poly(A)+ mRNA or liver poly(A)+ mRNA was labeled with Cy3 dye Hybridization was performed on the glass slides The first probe contained labeled control cDNA Hybridization of this probe to a glass Human Stress array was done at different concentrations relative to amount of poly(A)+RNA Hybridization was to 5 different glass slides at concentrations of 0 01%, 0 04%, 0 1%, 0 4%, and 1%, corresponding to 0 1 ng, 0 4 ng, 1 ng, 4 ng, 10 ng of adjusted weight amount of individual RNA species in the mixture At the same time, 0 1 ng of each individual control Cy5 labeled cDNA (corresponds to 1%) was mixed with the Cy3 labeled cDNA from 1 μg poly(A)+RNA Hybridization of this sample was performed on a 6^th glass slide array Hybridization conditions were as described earlier The first five glass slides were used for plotting the calibration curve and the 6^th glass slide was use for determination of concentration of RNA in the sample based on calibration curve and intensity of Cy3 and Cy5 dyes m the same spot The corresponding ratio of intensity of Cy5 to Cy3 was determined for all other spots

In the range of synthetic RNA concentrations used (from 0 01% to 1%), the intensity of the signal in the spot increased proportionally (and linearly) with an increasing amount of RNA The linear change in intensity of the signal from synthetic RNA in each spot allowed the direct comparison of the abundance of an mRNA of interest on two arrays with liver and placenta poly(A)+ RNA samples using the intensity of the Cy5 channel as a normalization value

The type of calibration curve is not dependent upon the method of synthesizing the sample or the complexity of the probe (control RNA mix may be used alone or added to poly(A)+ RNA before labeling This was demonstrated using a membrane based array with radioactive labeling instead of fluorescent dyes Design of this experiment was similar to the glass experiment descnbed above Five different concentrations of control RNA were used (from 0 01% to 1%) Labeling of these probes was done separately and after the mixing to 1 μg poly(A)+ RNA

It is evident from the above results and discussion that the subject invention provides a significant contribution to the field of array-based gene expression analysis With the subject invention, quantitation data is obtained from a single array and therefore errors due to variable in array quality, array type, assay conditions, etc , that are seen m situations involving two or more arrays, e g a control array and a test array, are avoided Furthermore, data received from different arrays, even different types of arrays, may be compared In addition, because a known amount of control target nucleic acids is employed in the subject methods and comparison of two different signals from a single array spot is performed, concentration determinations may be made that do not depend on efficiency of hybridization or efficiency of reverse transcription All publications and patent applications cited in this specification are herein incoφorated by reference as if each individual publication or patent application were specifically and individually indicated to be incoφorated by reference The citation of any publication is for its disclosure prior to the filing date and should not be construed as an admission that the present invention is not entitled to antedate such publication by virtue of prior invention

Although the foregoing invention has been described in some detail by way of illustration and example for puφoses of clarity of understanding, it is readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims

Claims

WHAT IS CLAIMED TS

1 An array-based hybridization assay comprising the steps of contacting an array(s) with a test pool of target nucleic acids and a control pool of target nucleic acids, wherein at least a subset of the probe nucleic acids present on said array are represented m said control pool of target nucleic acids

2 The assay according to Claim 1, wherein said test pool of target nucleic acids is labeled with the same label as said control pool of target nucleic acids

3 The assay according to Claim 1, wherein said test pool of target nucleic acids is labeled with a different label from said control pool of target nucleic acids

4 The assay according to Claim 1, wherein said test and control pools of target nucleic acids are contacted with the same array

5 The assay according to Claim 1, wherein said test and control pools of target nucleic acids are contacted with different arrays

6 A hybridization assay comprising the steps of contacting an array of probe nucleic acids with a pool of test target nucleic acids and at least one pool of control target nucleic acids, wherein each of said pools is differentially labeled so as to be distinguishable from each other, wherein said pool of control target nucleic acids comprises at least one nucleic acid that binds selectively to each probe nucleic acid of said array, and detecting hybridization of said test and control target nucleic acids to said array

7 The assay according to Claim 6, wherein said assay further comprises generating said labeled control set of target nucleic acids

8 The assay according to Claim 6, wherein said control set of target nucleic acids comprises a representative number of target nucleic acids 9 The assay according to Claim 6, wherein said control set of target nucleic acids comprises at least 50 target nucleic acids

10 A hybridization assay comprising the steps of contacting an array of probe nucleic acids with a first pool of test target nucleic acids labeled with a first label and a second pool of control target nucleic acids labeled with a second label, wherein said pool of control nucleic acids is comprised of at least one nucleic acid that binds selectively to each probe nucleic acid, detecting hybridization of said labeled target and control nucleic acids to said array, and comparing said hybridization

11 The assay according to Claim 10, wherein said test target nucleic acids and said control target nucleic acids are contacted to said array simultaneously

12 The assay according to Claim 10, wherein said test target nucleic acids and said control target nucleic acids are contacted to said array sequentially

13 The assay according to Claim 10, wherein said assay further comprises using a third pool of control nucleic acids labeled with a third label

14 The assay according to Claim 10, wherein said assay further comprises generating said labeled control target nucleic acids

15 The assay according to Claim 10, wherein said assay further compnses generating said labeled test target nucleic acids

16 The assay according to Claim 10, wherein said assay further comprises calculating individua testl target nucleic acid concentrations present in said pool of test target nucleic acids

17 A pool of control target nucleic acids for use in hybridization to an array of probe nucleic acids, wherein said pool comprises at least one nucleic acid that binds selectively to at least a subset of the probe sequences on said array 18 The pool of control target nucleic acids according to Claim 17, wherein said control target nucleic acids are selected from the group consisting of ribonucleic acids and deoxyribonucleic acids

19 The pool of control target nucleic acids according to Claim 17, wherein said pool comprises a representative number of control target nucleic acids

20 A kit for use in an array based hybridization assay, said kit comprising a control set of target nucleic acids or nucleic acid precursors thereof, wherein said confrol set of target nucleic acids comprises a representative number of target nucleic acids

21 The kit according to Claim 20, wherein said nucleic acids are ribonucleic acids

22 The kit according to Claim 20, wherein said kit further comprises primers

23 The kit according to Claim 22, wherein said primers are gene specific primers

24 The kit according to Claim 20, wherein said kit further comprises a label

25 The kit according to Claim 20, wherein said kit further comprises an array