WO1995011316A1 - A NUCLEIC ACID DIAGNOSTIC METHOD USING RIBOPROBES, RNAse DIGESTION, AND MOLECULAR WEIGHT EXCLUSION CHROMATOGRAPHY - Google Patents

Abstract

This invention is drawn to improved methods for detecting particular nucleic acid sequences via nucleic acid hybridization. Disclosed is a diagnostic method wherein detectably labeled riboprobes are hybridized in solution to complementary nucleic acid sequences encoded by target nucleic acid molecules. Following hybridization, RNAse digestion is used to degrade unhybridized probe molecules, after which chromatographic separation is employed to isolate probe-target hybrids. This method improves the detection limit of nucleic acid hybridization and is generally useful in medical diagnostics, forensics, and molecular biology research.

Description

A NUCLEIC ACID DIAGNOSTIC METHOD USING

RIBOPROBES, RNase DIGESTION, AND

MOLECULAR WEIGHT EXCLUSION CHROMATOGRAPHY

FIELD OF THE INVENTION

The present invention relates to diagnostic methods for detecting specific nucleotide sequences. In particular, the method employs detectable, single stranded ribonucleic acid probes complementary to a target nucleotide sequence.

BACKGROUND OF THE INVENTION

Nucleic acid hybridization is one of the most important physical phenomena to be exploited by molecular biologists. The ability of a single stranded nucleic acid molecule to hybridize with another nucleic acid molecule encoding a complementary nucleotide sequence to form a double stranded hybrid was first recognized in 1953 by James Watson and Francis Crick in their model of the double helical nature of DNA. In 1961, Hal et al . [Proc . Nat ' 1 . Acad . Sci . , USA, vol . 47, p. 131] described in detail the ability of complementary nucleic acid molecules to form double-stranded hybrids with high efficiency and remarkable specificity from a mixture containing an abundance of non-complementary nucleic acids .

Since those pioneering discoveries, nucleic acid hybridization has become a fundamental technique used in some form by many, if not all, molecular biologists. For instance, nucleic acid probes are now utilized to diagnose the presence of human, plant, and animal pathogens, to identify and isolate specific genes, to study gene function and structure, to diagnose genetic disorders or a genetic predisposition to certain diseases, for forensic purposes, and to detect specific nucleic acids in tissues and cells by in situ hybridization.

A variety of different hybridization techniques exist, ranging from the use of nucleic acid molecules immobilized on solid supports to liquid phase hybridization to in situ hybridization. However, a common theme runs through all nucleic acid hybridization procedures, namely the nucleotide sequence specific base pairing between a single stranded nucleic acid molecule encoding a known sequence (the "probe" molecule or sequence) with a complementary "target" sequence encoded by a second nucleic acid molecule, i . e . the "target" nucleic acid molecule. As with any diagnostic method or scientific technique, speed, sensitivity, specificity, ease of use, and low cost are important criteria. For instance, the effective treatment of infectious disease requires efficient diagnostic methods to detect the causative agent (s) . In the case of viruses, several methods of diagnosis exist. One such method employs cell culture facilities to isolate the pathogen (s) from clinical specimens. While this detection method is extremely sensitive, it is very costly, time consuming, labor intensive, and can be performed only in specialized facilities .

Another method used to detect the presence of a specific virus is the immunoassay, a technique that enables the simultaneous testing of large numbers of specimens using standardized reagents. However, this technique depends on the availability of an antigenic viral element for either direct detection or production of a specific antibody. Clearly, when production and/or isolation of the virus has not been accomplished, or when production of viral antigens is poor, or the host's immune response is not stimulated, other methods for detecting the virus's presence are needed. Nucleic acid hybridization is one such method.

Nucleic acid hybridization can also be used to detect bacterial pathogens. Conventional bacterial diagnostic methods are based on staining and microscopic analysis of a sample. While this method is rapid, it may be insensitive and non-specific. Because of this, most samples are cultured to amplify and identify the particular bacterial pathogen. Antibody reactivity, chemical reactivity, and/or growth on selective or different media is then conducted to reach a conclusion as to the bacteria present. As these methods depend on cultivation of the bacteria, identification can require from one day to several weeks. Beyond the time and cost factors evident in such conventional bacterial diagnostic methods, failure of a particular specimen to grow or lack of a specific antibody response may lead to a misdiagnosis. Clearly, more rapid, accurate, and cost effective diagnostic methods are needed. In addition to being useful in diagnosing the presence of infectious agents, be they bacterial, fungal, or viral, nucleic acid hybridization affords the ability to detect the presence of specific nucleotide sequences in nucleic acids. Accordingly, in medicine, this technique can be used to assist in the diagnosis of genetic diseases . Such diseases arise due to alterations in an individual's DNA, ranging from single nucleotide mutations, be they deletions, insertions, or substitutions (as in sickle cell anemia) to gross chromosomal abnormalities visible under light microscopy, such as trisomy 21 (associated with Down's syndrome) or large scale translocations or deletions, which may be associated with certain neoplastic disorders, such as leukemia, and other diseases. Nucleic acid probes can, among other things, assist in the diagnosis of abnormal phenotypes, in assessing one's predisposition to certain heritable disorders, in ascertaining whether a person is the carrier of a gene or genes for a particular genetic disorder, and in prenatal diagnosis. Outside the diagnosis of disease states, nucleic acid hybridization is also a powerful research tool, enabling scientists to study the fundamental aspects of gene structure, function, and organization. This technique has enabled the identification and isolation of previously undiscovered genes as well as providing a means for the analysis of transcriptional regulation of known genes.

Presently, there are many nucleic acid hybridization methods. The kinetics of various nucleic acid hybridization reactions are well understood. See Britten et al. , (1974) Methods Enzymol . , vol . 29, p. 363; Kohne et al ., (1977) Biochemistry, vol . 16, p. 5329, and Orosz et al . (1977) Biopolymer, vol . 16, p. 1183. In short, if one of the two reacting strands (the "driver") is present in excess, the reaction rate is decreased when the driver concentration is decreased. The converse is also true, i . e . increasing the driver's concentration increases the rate of hybridization. Likewise, temperature also effects the hybridization rate . The maximum rate for sufficiently lengthy strands is usually achieved when hybridization is conducted approximately 15° to 30°C below the TM (the 'melting' temperature: the temperature at which 50% of the double-stranded molecules are dissociated into single strands) . Nucleic acid hybridization reaction rates are also affected by ionic strength below 0.4 M for electrolytes such as NaCl, while salt concentration above that level has less impact on such rates . However, extremely high salt concentrations, such as 2 - 4 M or higher, have been found to substantially increase nucleic acid hybridization rates. For instance, see European Patent Application No. 0 229 442, published 07/22/87.

Despite what is known about the kinetics of nucleic acid hybridization, a major limitation is that of the basic reaction rate. Reaction times ranging from several hours to days are frequent . Various techniques to increase the hybridization rate have been developed. These include volume exclusion in RNA:DNA and DNA:DNA hybridizations [Renz et al ., (1984), Nucl . Acid Res . , vol . 12, pp. 3435 - 3444], multi-phase emulsions for DNA:DNA reactions [Van Ness et al . , (1982) Nucl . Acid Res . , vol . 10, pp. 8061 - 8067], adjusting salt concentrations between DNA:DNA and DNA:RNA hybridizations [Kohne et al . , (1977), Biochemistry, vol . 16, p. 5329], and inclusion of one or more nucleic acid precipitating agents [European Patent Application No. 0 229 442, supra ] .

In addition to the rate of the hybridization reaction itself, other factors are also important. For example, in the typical filter hybridization procedure, where the target nucleic acid, usually DNA, is bound to a suitable substrate (such as nitrocellulose or a nylon membrane) , several additional steps requiring considerable time and skill must also be conducted. These include placing the sample containing the target nucleic acid on the filter, denaturing the target nucleic acid and permanently affixing it to the substrate, conducting the hybridization reaction with the probe nucleic acid, washing to remove unhybridized probe, and finally visualization of hybrids (typically accomplished by autoradiography when radiolabeled probes are employed) .

However, various factors contribute to the lack of reproducibility in filter hybridization methods . Such factors include the amount of sample remaining bound to the filter, heterogeneity in bonding, and the precise location of the sample. These can lead to variability in quantification and difficulty in automating the procedure. In addition, only a small fraction of target nucleic acid bound to the filter is available for hybridization to a probe, reducing sensitivity.

Many of these shortcomings have been overcome by sandwich hybridization procedures, wherein two nucleic acid probes, each complementary to a different nucleotide sequence in the target, are used. Typically, the manufacturer of the diagnostic kit supplies a substrate with a known amount of one probe already bound at specific positions. The target nucleic acid, following whatever preparatory steps must first be conducted, is then hybridized to the first, bound probe. The second probe, enabling detection of target:first probe hybrids, can either be included along with the sample containing the target or it can be added after the target has hybridized to the first probe. For greater detail, see U.S. Patent No. 4,925,785.

As an alternative to hybridizations on a fixed support, to which either a probe or target nucleic acid is bound, solution (or liquid) hybridization may be performed. Nobrega et al . [(1983) Anal . Biochem . , vol . 131 pp: 141 - 151] describe a method for detecting RNA transcripts during RNA processing by using DNA probes in solution hybridizations, after which unhybridized probe molecules are separated from double-stranded probe- target hybrids by agarose gel electrophoresis. The hybrid nucleic acids are then detected by autoradiography.

In 1987, Thompson and Gillespie [Anal . Biochem. , vol . 163, pp. 281 - 291] reported dissolving a biological sample containing a target nucleic acid molecule in a chaotropic solution followed by hybridization of a probe nucleic acid (present in excess) directly to the dissolved sample. Nucleic acids were collected by filtration onto nitrocellulose and then detected by autoradiography or scintillation counting. RNA probes were found to be advantageous because RNAse treatment following filtration could be used to eliminate complexes between probe and non- complementary target molecules.

Pellegrino et al . [(1987) BioTechniques, vol . 5, no. 5, pp. 452 - 459] describe an improvement of Thompson and Gillespie's work, supra, enabling greater detection sensitivity to be achieved. Their method employed guanidine thiocyanate dissociation of sample cells, which were then analyzed by hybridization in solution to an RNA probe in low concentration. When the target nucleic acid was RNA, hybridizations were conducted for 44 hours, followed by RNAse treatment to degrade unhybridized probe prior to precipitation of probe-target hybrid nucleic acids onto nitrocellulose or nylon membranes. In contrast, when DNA was the target, 30 hour hybridizations were performed, followed by filtration through nitrocellulose. Filters were then treated with RNAse to degrade unhybridized probe. In either case, radioactivity was then measured by scintillation counting. McKnight et al . , [(1988) Methods Enzymol . , vol . 159, pp. 299 - 311] describe the use of solution hybridization to quantitate messenger RNA (mRNA) levels. In their procedure, concentrated, previously purified total nucleic acid encoding the target sequence was hybridized overnight in solution to a specific RNA probe under conditions of target excess and low ionic strength, followed by RNAse treatment and subsequent trichloroacetic acid (TCA) precipitation onto a solid support. Radioactivity was then detected by liquid scintillation counting. European Patent Application No. 0 278 220, published on August 17, 1988, describes a quantitative nucleic acid hybridization assay for detecting a DNA or RNA target nucleic acid molecule in a biological sample. The method utilizes either a crude or purified pool of target nucleic acids. Probes of almost any specified length are prepared by the primer extension method, wherein the length of a given probe is controlled by limiting the concentration of one of the nucleotides. A particular probe molecule is then hybridized in solution to the target pool. The hybridization reaction's ionic strength ranges from 0.15 - 0.9 M and hybridizations typically run 16 hours. After hybridization, unhybridized probe is separated from probe-target duplexes by gel exclusion chromatography, permitting the differential detection of probe bound to target without unhybridized probe causing excessive background. The volume excluded from the column is then counted by scintillation to measure the measure the radioactivity present.

In contrast to the above, the present invention provides a method to sensitively detect a target nucleic acid, be it DNA or RNA, in a crude, unpurified biological sample in one day or less, using a specific RNA probe in a solution hybridization reaction, followed by RNAse digestion, column chromatography, and detection of the probe-target hybrids.

Definition of Terms The following terms are used throughout the specification. Unless otherwise indicated, these terms are defined as follows: "Hybridization" generally concerns the annealing of any two nucleic acid molecules comprising complementary nucleotide sequences, each nucleic acid molecule having originated from a different source. "Renaturation, " in contrast, refers to the reassociation of two complementary, single stranded nucleic acid molecules denatured previously. "Stringent" hybridization refers to hybridization under conditions promoting the annealing of nucleic acid molecules having complementary nucleotide sequences but retarding the annealing of noncomplementary molecules. Factors influencing hybridization include probe size and nucleotide composition, temperature, salt, ionic strength, pH, reactant concentration, the presence of other molecules, including chaotropic agents, etc.

"Hybrid" refers to a double stranded nucleic acid molecule formed as the result of the hybridization a "target" nucleic acid molecule and a "probe" nucleic acid molecule. As used in this invention, "hybrid" includes RNA:DNA and RNA:RNA duplexes unless otherwise specified. The target nucleic acid molecule may be comprised either DNA or RNA. The probe nucleic acid molecule is comprised of RNA and can also include a number of nucleotide analogs and/or non-nucleotide adducts. A "pool" of nucleic acid molecules is a collection of nucleic acid molecules encoding one or potentially many more nucleotide sequences. The "pool" may or may not contain the sought after, or "target," nucleotide sequence. It is only after hybridization with a particular probe nucleic acid molecule and subsequent assay steps that the presence, and, if desired, quantity, of a particular target nucleic acid molecule in the pool can be deduced.

SUMMARY OF THE INVENTION

It is the object of this invention to provide improved methods for the detection of specific nucleic acid sequences. Generally, the invention involves a method comprising: (1) combining, in solution, a pool of nucleic acid molecules and an RNA probe designed to hybridize to molecules encoding the target nucleic acid sequence, under stringent conditions so as to produce double stranded probe-target hybrid nucleic acid molecules; (2) degrading unhybridized probe molecules with RNAse; (3) separating the hybrid nucleic acids from other solution components; (4) and recovering and detecting probe-target hybrids.

One aspect of the invention concerns the type of target nucleic acid molecules which are detected. In one embodiment, the target nucleic acid sequence may be encoded by DNA, while in another, it may be encoded by RNA. In either case, the pool of nucleic acid molecules may be derived from a biological sample . In one particular embodiment, the target nucleic acid sequence indicates the presence of a pathogen, particularly a pathogen of viral, bacterial, or fungal origin.

In another embodiment, a the presence or absence of a particular target nucleic acid sequence enables detection of a neoplasm, be it benign or malignant. Furthermore, the quantity of target nucleic acid present can be assayed. In another embodiment, a particular target nucleic acid sequence might indicate the existence of a heritable genetic disorder, or a genetic marker. Further, a specific genetic marker may be linked to a particular heritable genetic disorder. Yet other embodiments involve using the methods of the invention in forensic analysis and in analyzing gene expression, in terms of both location and quantity. Another aspect of the invention concerns covalently attaching a detectable label substance to the RNA probe. The presence of probe.'target hybrids can then be detected by detecting the label substance. Suitable detectable label substances include, but are not limited to: radioisotopes; fluorescent molecules; chemiluminescent molecules; and members of specific binding pairs . One embodiment of the invention entails - li ¬

the use of a radioisotope as the detectable label substances. Radioisotopes useful in accordance with the disclosed methods include ³H, ¹²⁵I, ³²P and ³⁵S. In a preferred embodiment, ³⁵S serves as the detectable label, while in another such embodiment, ³²P may be used. In contrast, an alternative embodiment may utilize a specific binding pair, especially avidin or streptavidin, which may then be complexed with biotin.

An additional aspect of this invention relates to the conditions under which the solution hybridization reaction are conducted. Conditions such as temperature, pH, and electrolyte concentration play an important role. In a preferred embodiment, the temperature of the solution hybridization reaction ranges from about 15° C to 30° C below the melting temperature of the probe- target hybrid, the pH ranges from about 6.6 to 7.2, and the electrolyte concentration is from about 2.0 M to 3.5. In a particularly preferred embodiment, the pH is about 6.9 ± 0.2, especially about 6.9 ± 0.05, and the electrolyte concentration is approximately 2.75 ± 0.2 M for RNA, especially 2.75 ± 0.05 M. When the target nucleic acid is DNA, the electrolyte concentration is often somewhat lower, generally 2.5 ± 0.2 M. In another preferred embodiment, the temperature of the solution hybridization is about 84 "C ± 4^*C when the target nucleic acid comprises RNA, and when DNA comprises the target nucleic acid, about 72"C ± 4°C.

Yet another aspect of the invention involves the time period required for the solution hybridization reaction. In one embodiment, time period ranging form approximately one to eight hours, and particularly about two hours, is employed.

Another aspect of the invention relates to separation of the probe-target hybrid nucleic acids from other solution components. In a preferred embodiment, molecular weight exclusion chromatography is employed to perform this separation.

A further aspect of the invention concerns the method employed to detect the probe-target hybrids. In one embodiment, liquid scintillation is used.

BRIEF DESCRIPTION OF THE FIGURES

FIGURE 1 shows a typical limit of detection calculation. Particularly illustrated is the detection limit of 0.04 amol of target nucleic acid pSP65:HPV16 when probed with HPV16 B fragment riboprobe according to the methods described herein.

FIGURE 2 graphically represents the effect of varying concentrations of Na₁.₅H₁.₅PO₄ on the efficiency of detection of a DNA target nucleic acid.

FIGURE 3 depicts the elution profile of a riboprobe-column assay using the column matrix S200 to separate probe-target hybrids, wherein the target nucleic acid comprises RNA, from other solution components, including degraded, unhybridized probe fragments.

FIGURE 4 illustrates the elution profile of a riboprobe-column assay using CL-4B resin to separate probe-target, wherein the target nucleic acid comprises DNA, hybrids from other solution components, including degraded, unhybridized probe fragments.

FIGURE 5 graphically depicts PF 4 mRNA expression levels in HEL cells treated with various concentrations of PMA. FIGURE 6 shows PF 4 mRNA expression levels at different times following PMA induction.

Numerous aspects and advantages of the invention will be apparent to those skilled in the art upon consideration of the following detailed description which provides illumination of the practice of the invention in its preferred embodiments.

DETAILED DESCRIPTION

This invention provides a rapid, sensitive, quantitative assay for detecting the presence of a target nucleic acid in a biological sample using a detectably labeled RNA probe in a solution hybridization procedure. The biological sample, which contains the pool of target nucleic acid molecules, need not be purified prior to performing the described methods . However, purified or partially purified samples are also suitable for these methods. In the event the sample is crude or only partially purified, it is treated with detergent (s) and/or protease (s) to solubilize and release the contained nucleic acids.

If the target nucleic acid sequence is double stranded, it must be denatured prior to hybridization to the probe nucleic acid. For instance, denaturation may be performed under alkaline conditions or by heating. Once the target nucleic acid is rendered single stranded, it is hybridized in solution to a specific probe nucleic acid molecule comprised of RNA. General hybridization considerations and methods are described in Molecular Cloning, Second Edition, Sambrook et al . eds., Cold Spring Harbor Laboratory, 1989, as are numerous other techniques used in this invention. Following hybridization, unhybridized single stranded RNA probe molecules can be degraded by use of RNase. When the target nucleic acid is RNA, single- stranded regions of the resulting hybrid are likely to be degraded. As a result of such degradation, the RNA:RNA hybrids so obtained may be relatively short, in contrast to situations where the same target nucleic acid sequence is comprised of DNA, which is resistant to degradation according to the instant methods . After degradation of unhybridized RNA probe molecules, the solution is subjected to molecular weight exclusion chromatography to separate double stranded probe-target hybrid nucleic acid molecules from degraded probe fragments. The probe-target hybrids, if any, may then be recovered, detected, and quantified.

This nucleic acid diagnostic method is extremely sensitive. Furthermore, these methods are extremely rapid, enabling the entire procedure to be performed by one person in less than one working day. As a result of its sensitivity and speed, this invention has many potential uses. For instance, it can be used to ascertain the presence or quantity of a particular pathogen, be of bacterial, fungal, or viral origin, or group of pathogens, in a tissue sample; to diagnose genetic or neoplastic disorders or predispositions to certain diseases; and for forensic purposes. Additionally, this method can be used in molecular biological research to detect and measure the level of expression of a specific gene in tissues or cell lines, and to study regulation of gene expression.

An example of how the instant invention can be used in medical diagnosis involves human papilloma viruses. In 1976, papilloma viruses were reported to be associated with genital cancer in humans [zur Hausen, H. (1976) Cancer Res . , vol . 36, p.794; see also zur Hausen, H. (1977) Curr. Top . Microbiol . Immunol . , vol . 78, p. 1; Kreider et al . , (1985) Nature, vol . 317, pp. 639 - 641] . Previously, human papilloma viruses (HPVs) had been known to induce mainly benign tumors, including warts of the skin and genital mucosa. It is now known that HPVs are small, naked viruses which contain a circular, episomal genome of about 8,000 base pairs. A number of HPV subtypes are known to exist. HPV classification is based on nucleic acid homology to other subtypes of known DNA sequence. A new HPV type is defined when it cross hybridizes (DNA:DNA) with known subtypes to less than 50% under stringent conditions. More than 50 HPV subtypes have been identified, and complete nucleotide sequences have been obtained for at least eight of these.

Development of an antibody-based HPV diagnostic assay has proved elusive, due largely to the inability to propagate the virus in cell cultures. As a result of the lack of specific antigen detection assays, nucleic acid hybridization is the preferred HPV diagnostic technique. In the past, spot blots, sandwich blots, in situ hybridization, and Southern blots have been used. However, each of these methods required filter hybridization. In addition, while in situ hybridization allows for histological analysis as well, the technique is laborious and sometimes encounters serious background problems. It is known that HPV types 6, 11, 16, 18 and

31 occur in genital lesions. See Gissmen et al . , (1983), Proc . Nat 'l . Acad. Sci . , USA, vol . 80, pp. 560 - 563; Durst et al . , (1983), Proc . Nat 'l . Acad. Sci, USA, vol . 80, pp. 3812 - 3815; Bushart et al. (1984), EMBO, vol . 3, p. 1151; and Lorinez et al . , (1986), J. Virol . , vol . 58, p. 225. HPV types 16 and 18 occur in about 80% of cancers of the cervix, vulva, and penis, while types 6 and 11 are more frequently associated with genital warts and benign lesions [Hypia et al . , supra ] . In addition, hybridization studies reveal that HPV types 6 and 11 are closely related, whereas types 16, 18, 31, and 33 are more distantly related and cross-hybridize poorly. Whether or not the infecting HPV strain has transforming potential depends largely on whether or not the HPV DNA becomes integrated or remains episomal. Benign lesions are most often associated with episomal HPV DNA, whereas invasive genital cancers correlate with integrated forms of the virus [Dubt et al . , (1985), J. Gen . Viral . , vol . 66, p. 1515] .

Present therapy for HPV infection includes surgical, i . e . , cryogenic, laser, etc . , removal of the affected areas. When performed at an early stage, the prognosis is good. However, recurrence happens frequently and thus determination of the HPV type present is important for proper treatment. Clearly, an accurate, sensitive, and inexpensive diagnostic test is needed to assist in the diagnosis and treatment of individuals afflicted with HPV infection.

The riboprobe methods of the instant invention are much more specific and have lower backgrounds than corresponding DNA probe-based diagnostics. In addition, the detection limit, or sensitivity, of an HPV diagnostic employing the instant methods is less than 0.1 copies of HPV per human genome. And because of regions of relatively low sequence homology, i.e., high type-specificity, in the HPV genome, it is possible to generate HPV type-specific nucleic acid probes. Thus, an initial HPV diagnosis may be made by employing either a mixture of probes specific for particular HPV types or probes which hybridize to highly conserved regions of the HPV genome. Identification of the particular HPV type present can then be made by employing type-specific probes directed to HPV genome regions of high type- specificity.

As is clear from the above description, generation of highly specific and sensitive nucleic acid assays may be developed according to the methods provided herein for a multitude of infectious agents. In addition, a fast, efficient, and inexpensive mechanism is described enabling the development of diagnostics capable of distinguishing between subtypes of a given pathogen. Furthermore, the invention has applicability outside the arena of medical and/or forensic diagnostics. For instance, these methods can also be applied in studies of gene expression and regulation. Because many genes are expressed at different times and/or at different rates during an organism's lifetime, much interest is focused on what factors and conditions affect and regulate this differential expression of genes. For instance, all mammals possess a hematopoietic system that replaces the multiplicity of all blood types found in a healthy animal, including white blood cells (neutrophils, macrophages, and basophil/ ast cells) , clot forming cells (megakaryocytes, platelets) , and red blood cells (erythrocytes) . Presently, it is believed that combinations of specific polypeptides (growth factors and colony stimulating factors) direct the proliferation, differentiation, and maturation of each of the various components of the hematopoietic system. As mentioned above, megakaryocytes are components of the hematopoietic system. These cells undergo progressive development from progenitor cells until, by an ill-defined process of controlled disintegration, they terminally differentiate into thousands of anuclear platelets . During this maturation process, megakaryocytes produce and/or display on their surfaces a number of platelet-related proteins . Among these proteins are platelet factor 4 (PF 4) [Ryo et al . , (1983), J. Cell, Biol . , vol . 96, p. 515] and platelet- derived growth factor (PDGF) [Ryo et al . , supra . ] . Megakaryocyte maturation and platelet formation are a physiologically controlled processes. Maturation is stimulated in response to thrombocytopenia [Corash et al . , (1987), Blood, vol . 70, p. 177] and inhibited by thrombocytosis [Jackson et al . , (1984)

Blood, vol . 63, p. 768) . These processes are apparently regulated humorally, as administration of interleukin-6 increases platelet counts in vivo [Ishibashi et al . , (1989), Blood, vol . 74, p. 1241] . Erythropoietin, interleukin-3, and granulocyte-macrophage colony stimulating factor have positive effects on various in vitro megakaryocyte systems [Bruno et al . , (1989) Blood, vol . 73, p. 671) .

The scarcity and fragility of megakaryocytes make it difficult to obtain pure populations of these cells for study. Additionally, megakaryocyte maturation is broken into stages based on expression of platelet- specific proteins or morphological features, although these stages are not discrete, thereby making the study of their progressive development difficult. As a result, immortalized cell lines, such as human erythroleukemic cells (HEL) [Martin et al . , (1982), Science, vol . 216, p. 233] that express phenotypic and biochemical markers for megakaryocytic lineages have been used for in vitro study. It is known that in HEL cells treated with phorbol esters, there is a dramatic increase in PF 4 expression [Tebilio et al . , (1984), EMBO, vol . 3, p. 453] .

PF 4 is a protein synthesized in megakaryocytes and packaged into platelet alpha granules sometime prior to platelet release. Because megakaryocyte maturation is apparently regulated by polypeptide growth factors, PF 4 mRNA expression levels were examined following treatment with a variety of growth factors and phorbal 12-myristate 13-acetate

(PMA) . PF 4 mRNA expression levels were monitored using the nucleic acid hybridization methods of the instant invention.

Analysis of PF 4 mRNA expression levels showed that induction of mRNA synthesis was not rapid, as PF 4 mRNA did not reach its maximum level until 24 - 48 h post-induction, suggesting induction of PF 4 mRNA synthesis is an indirect effect of PMA stimulation, since some genes are turned on within minutes of PMA addition. Expression levels for PDGF were also monitored using the instant diagnostic techniques following PMA stimulation. The results indicated the expression of the gene coding for PDGF was unaffected, even under conditions where PF 4 mRNA expression was maximal [Hunt et al . (1991), Exp . Hematol . , vol . 19, pp: 779 - 784] .

As can be seen from the above, the nucleic acid hybridization methods described herein have wide applicability throughout the field of molecular biology, from medical diagnostics for pathogenic infection and heritable diseases to more research-oriented studies of gene expression. The following procedures may be used in conjunction with the instant invention.

Probe Characterization A probe used in accordance with the methods of this invention is comprised of RNA and is complementary to the target nucleic acid to be identified. Probes may be prepared according to any number of techniques. See Sambrook et al . , supra . In a preferred embodiment, RNA polymerase is used to transcribe a DNA molecule, either cloned or resulting from polymerase chain amplification, encoding a probe nucleotide sequence. As several bacteriophage RNA polymerases and their specific promoters are known, a variety of transcription cloning vectors harboring these sequences are now commercially available or otherwise known in the art . Such vectors may contain promoters for a single bacteriophage RNA polymerase, or for two different polymerases, adjacent to a multiple cloning site into which the DNA encoding the probe nucleotide sequence is cloned. Transcription with the appropriate RNA polymerase in the presence of ribonucleotides, one or more of which is labeled with a detectable label substance, enables the production of a large quantity of high specific activity riboprobe corresponding to either the coding or non-coding strand of the inserted DNA molecule.

The following procedures were used to generate each of the riboprobes used in the examples below. The plasmid pGEM-3 (Promega, Madison, WI, 1989/90 cat. no. P2131) was used as the transcription vector and T7 RNA polymerase was used to transcribe each riboprobe.

However, as noted above, alternate systems are available to generate probes useful in accordance with this invention.

Probe Synthesis.

Each T7 RNA polymerase transcription reaction described herein occurred under the following conditions: 40 mM Tris-HCl, pH 7.5; 6 mM MgCl₂; 2 mM spermidine; 10 mM NaCl, 10 mM DTT; 15 μM ³⁵S-UTP (New England Nuclear, cat. no. NEG-039H) ; 500 μM CTP, ATP, and GTP; and 75 - 100 ng of isolated DNA template with a functionally associated T7 promoter isolated from the transcription vector following digestion with an appropriate restriction endonuclease and agarose gel electrophoresis. It should be noted that the RNA synthesis rate rapidly decreases when the UTP concentration is less than 12 μM. Additionally, the size of the RNA transcripts produced is dependent on the concentration of nucleotide precursors. The typical transcription reaction can be performed as follows: In a 500 μ Eppendorf tube, 120 pmol (picomoles) of ³⁵S-UTP was completely dried. On the bench top (no ice) , the ³⁵S-UTP was resuspended in a premix comprising: 3.3 μL¹ DePC-treated H₂O; 1.6 μL 5X transcription buffer (200 mM Tris-HCl, pH 7.5 (measured at 37°C) , 30 mM MgCl2, 10 mM spermidine, and 50 mM NaCl); 0.8 μL 100 mM DTT; and 0.4 μL of 40 units/μL RNasin (Promega) . Next, 1.2 μL of a solution containing 10 mM each of CTP, ATP, and GTP was added, followed by 0.4 μL (75-100 ng) of template DNA and 0.2 μL of 20 units/μL T7 RNA polymerase. This solution was then incubated at 40°C for 30 min., after which time an additional 0.2 μL (4 units) of T7 RNA polymerase was added, followed by another 30 min., 40°C incubation.

After riboprobe synthesis completion, the DNA template was digested by adding 2.7 μL (2.7 units) of RNase-free DNase (Promega Biotech) and 0.8 μL of 5X transcription buffer, supra, and incubating the reaction at 37°C. After 20 min., 0.6 μL 2% SDS was added, followed by a 20 min. incubation at 72°C. At that point, 34 μL of TE + 10 mM NaCl, 4 μL RNasin, and 0.8 μL of 500 mM DTT was added to complete the final reaction mixture.

Radioisotope Incorporation Into Probe. To determine the percent incorporation of

³⁵S-UTP into a riboprobe, a small volume, typically 1 - 10%, of a runoff transcription reaction may be diluted and chromatographed. Frequently, 1% of this diluted solution is employed in the chromatography procedure by spotting it onto a particular location on the chromatography matrix. Suitable chromatography matrices include paper, such as DE81 (Whatman, Clifton, NJ) , or silica gel on glass plates (thin layer chromatography, TLC) .

¹ Adjust volume as necessary to arrive at a total volume of 7.9 μL before the second addition of T7 RNA polymerase. Chromatography is conducted by using a solvent which causes unincorporated ribonucleotides to migrate near the solvent front, while polynucleotides, such as riboprobes, remain where they were spotted (the origin) . Following chromatography, the matrix is cut into sections and the radioactivity present on each section is measured. After subtraction of background, the results obtained for each section are summed and the percent incorporation of radioactive nucleotide into polynucleotide is calculated by dividing the radioactivity of the section containing the origin by the total radioactivity and multiplying by 100.

Typical ³⁵S-UTP incorporation into riboprobe using the herein-described methods is about 80%. Multiplication of the calculated incorporation rate by the measured radioactivity of a particular quantity of a given riboprobe synthesis reaction reveals the amount of the signal attributable to the riboprobe itself, exclusive of unincorporated label. This procedure may be used to quantitate ³⁵S-UTP incorporation into riboprobes that are either unpurified or purified following synthesis. Purification, if conducted, may involve column chromatography to remove unincorporated, labeled ribonucleotides. See Berger et al . , (1987) Methods in Enzymology, vol . 152, pp. 638 - 639.

In the examples below, either of the following procedures were used to calculate the percent of radioisotope added actually incorporated into probe . In the first such procedure, the amount of ³⁵S-UTP (or other radioisotope) incorporated in each riboprobe synthesis can be calculated by removing 1 μL of the above final reaction mixture and adding it to 49 μL TE + 10 mM NaCl. 0.5 μL of this solution is then spotted onto a DE81 paper strip approximately 15 cm x 1 cm. The DE81 paper is then chromatographed in 0.3 M ammonium formate until the solvent front migrates about 80 - 90% of the length of the paper. The strip is next cut into several pieces and each piece is placed in a separate scintillation vial. 15 mL of Liquiscint® (National Diagnostics) is then added to each vial, the vials are counted in a Beckman liquid scintillation counter, and the percent incorporation calculated from the results.

An alternative percent incorporation determination can be performed using thin layer chromatography. Here, a piece of PEI-Cellulose F (E. Merck reagent 5579-7) approximately 8 cm x 4 cm is initially soaked in 100% ethanol and then dried. 0.5 μL of a 1/50 dilution of the final reaction mixture (as described in the previous percent incorporation procedure) is spotted adjacent to a 0.5 μL spot of 1/250 dilution of the ³⁵S-UTP label used in the transcription reaction. The samples are then chromatographed using a 1 M LiCl solution. After the solvent front migrates approximately 3 cm, the cellulose is dried and autoradiographed. Scanning the resultant film with a densitometer (Gilford Response II, Ciba-Corning Diagnostics Corp., Pleasanton, CA) enables a determination of the amount of ³⁵S-UTP incorporation into the probe.

Standardization.

To quantitate the amount of target in a given clinical sample, a standard curve can be developed by conducting a series of hybridization (or titration) assays employing a constant amount of radiolabeled probe (measured in cpm) and varying amounts of target nucleic acid of known size, concentration, and sequence. A convenient source of DNA target molecules complementary to a given riboprobe is the DNA clone or fragment that serves as the template for synthesis of the riboprobe. The target DNA concentration can be quantitated by spectroscopic or fluorescent methods prior to dilution into the hybridization solutions.

The preferred method for generation of RNA target molecules complementary to the riboprobe is to transcribe the opposite strand of the same DNA molecule that served as the riboprobe template in the presence of trace quantities of label, such as the radiolabel ³⁵S- UTP. This may be accomplished either by using a clone harboring the template DNA molecule in the opposite orientation from that used to generate the probe (and thereby use the same type of RNA polymerase as was used to make the probe) or to utilize the same clone and two different types of RNA polymerase . In the latter system, one strand of the DNA template is transcribed by one RNA polymerase that recognizes a promoter adjacent to the one end of the template, while the other strand is transcribed by a different type of RNA polymerase that recognizes and binds to a different promoter adjacent to the other end of the inserted DNA template. The plasmid vector pGEM-3 and other vectors of similar makeup afford the ready implementation of this latter approac .

Irrespective of the system used to synthesize target RNA, the concentration of target RNA produced can be calculated from the amount and specific activity of the input UTP, the measured fraction of UTP incorporated into target, and the number of U residues in each RNA target molecule. The concentration of incorporated ³⁵S- UTP in cpm (counts per minute) per μL can be calculated by taking the product of the measured ³⁵S-UTP percent incorporation and the number of cpm per μL of reaction. This result is then divided by the specific activity of target RNA, in cpm/amol (1 amol = 10~¹⁸ mol, or 6.023 x 10⁵ molecules) , calculated by multiplying the specific activity of the input UTP in cpm/amol by the number of uridines in the target RNA, to yield the number of amol of target per μL of transcription reaction.

The sensitivity (in cpm/amol) of the probe assay may then be determined by hybridizing a fixed amount of probe (as measured by cpm) with varying amounts (including none) of target (the probe complement) . Following hybridization, RNAse digestion, and probe-target purification of each sample, the signal obtained in the assays lacking target is subtracted from the signal detected for each sample containing target . This corrected signal is then divided by the number of amol of target in each sample to give the signal per amol of target. Alternatively, a standard curve is generated by plotting corrected signal versus amol of target; or slope and intercept are calculated by use of linear regression on logarithms of the two quantities.

In the examples below, the various riboprobes were standardized against 2.0, 0.5, and 0.0 amol of target, namely the plasmid vector from which the various probes were derived. The various pools of target molecules were prepared in 1.5 mL Sarstedt centrifuge tubes as follows: The 2.0 amol pool comprised 24 μL of TE(+) (5.0 μL 20% SDS, 8.9 μL 2.25 M DTT, 986 μL TE) ; 6.0 μL of 1.0 amol/μL plasmid target; and 120 μL of HP + TE(+) [62.5 μL of 2 μg/μL HPDNA (human placental DNA, Sigma, St. Louis, MO) resuspended in H2O and then sheered by passing several times through a 22 gauge needle, followed by sonication, phenol extraction, chloroform extraction, ethanol precipitation, and resuspension in TE) and 937.5 μL TE(+)] . The 0.5 amol target pool comprised 28.5 μL TE(+), 1.5 μL of 1.0 amol/μL plasmid target stock, and 120 μL of HP + TE(+); while the 0.0 amol target pool was made up of 30 μL TE(-f) and 120 μL of HP + TE(+) . Next, the target DNA was degraded and denatured by adding 15 μL of 2 M HC1 to each tube. The solutions were incubated at room temperature for 1 min. before 15 μL of 4M NaOH was added, after which the solutions were incubated for 2 min. in a boiling water bath. The tubes were then cooled for 2 min. before a 10 sec. centrifugation. Duplicate 60 μL aliquots of each sample were then removed and placed into fresh, separate tubes for hybridization.

Limit of Detection A sensitivity limit (limit of detection) for a given riboprobe assay performed in accordance with this invention can be calculated using various known quantities of the target nucleic acid in conjunction with a fixed amount (measured in cpm) of probe. As an example, 0.0 amol, 0.5 amol, and 2.0 amol of target may be useful target quantities for determining the detection limit of the assay. Initially, the background reading of the counting device employed, such a liquid scintillation counter, must be determined along with the standard deviation of that background signal. This may be accomplished by taking the average of several scintillation fluid-only cpm measurements. Next, the corrected signal for each sample is calculated by subtracting the counter background (cB) from the signal obtained for each sample (preferably the average signal of two equivalent samples) . The normalized signal for the samples is then determined by subtracting the corrected signal without target (containing 0.0 amol target) from the corrected signal with target and then dividing that result by the amount (amol) of target.

The limit of detection for a given amount of target may then be deduced by taking the sum of the corrected signal without target and the counter's standard deviation multiplied by a factor of three (selected for statistical purposes to achieve a confidence level of greater than or equal to 99%) and dividing that sum by the normalized signal obtained for a given amount of target. An example of a such a detection limit calculation for 0.5 amol of target is provided in FIG. 1. This information, coupled with the knowledge of how many cells were used in an assay, can then be used to calculate the minimum number of DNA or mRNA molecules/cell that can be detected using the procedures described herein. Finally, the signal to noise ratio for a given amount of target can be ascertained by dividing the corrected signal for a given amount of target by the corrected signal measured for 0.0 amol of target. Limits of detection ranging down to as few as 0.2 amol of target nucleic acid molecules when RNA is the target, and 0.03 amol when DNA harbors the target sequence, are readily achievable when the instant methods are employed.

Hybridization.

Each hybridization performed in accordance with the present invention is a solution hybridization. The rate of hybridization is dependent upon temperature, salt and ionic strength, pH, reactant concentration, and the presence of other molecules, such as dextran sulfate, polyethylene glycol, etc . The effects of varying Na₁.₅H₁.₅PO₄ concentrations on the practice of the invention are provided in FIG. 2. Each standardization hybridization was conducted by adding 86 μL of 4.8 M Na₁.5H₁.₅PO₄ to each 60 μL aliquot of denatured target DNA, followed by vortexing. 4 μL of the appropriate ³⁵S-labeled riboprobe (approximately 15,000 cpm) was then added to each tube, the contents were mixed and the solutions briefly (10 sec.) centrifuged, followed by a 2 hr. incubation at 72°C. RNAse Digestion.

As RNA is used as the probe in the practice of the invention, digestion with RNase may be utilized following hybridization to degrade unhybridized probe molecules and thus reduce background attributable to unhybridized probe molecules . Because of the hardy nature of many RNase enzymes, numerous reaction conditions are possible for the RNase digestion reaction. In the examples, RNase digestions were conducted by adding 12 μL of RNase Mix [10 mM Tris, pH 8, 1 mM EDTA, 5700 units/ L RNase Tl (Boehringer Mannheim, cat. no. 109 207), and 40 μg/mL RNase A (Sigma Type 1-A, cat. no. R 4875)] to each sample following hybridization. The contents of the tubes were then vortexed, centrifuged for 10 sec, and placed at 37°C for 10 min. After the 10 min. RNase treatment, 195 μL of Formamide-dye mix (99.9% formamide (reagent grade) and 0.1% bromphenol blue) was added to each reaction, followed by vortexing. Each sample was then ready for molecular weight exclusion chromatography.

Molecular Weiσht Exclusion Chromatography.

Molecular weight exclusion chromatography enables the separation of compounds based upon differences in molecular weight. In this procedure, a granular matrix, composed of polyacrylamide, sepharose, cross-linked sepharose, agarose, cross-linked agarose, dextran, or other similar materials, is used to separate probe-target hybrids from ribonucleotides generated by the RNase-mediated degradation of unhybridized single stranded riboprobes. This separation occurs due to the porosity of the matrix material utilized. A matrix having small pores allows small molecules, such as free nucleotides, to enter the matrix, while larger molecules, such as probe-target hybrids, are excluded. When the matrix is packed in a chromatography column, this excluded volume passes through quickly. In contrast, smaller molecules in solution enter the matrix and their movement is delayed as an inverse function of their effective molecular weight.

Products suitable for use in this invention as the column matrix include Sephadex® G50, G100, and G200 (Pharmacia, Piscataway, NJ) , Sepharose® CL-2B, CL-4B, CL-6B, S-200, S-400, and S-1000 (Pharmacia), and P-20, P-60, P-100, and P-200 (BioRad Corp., Hercules, CA) .

The matrix selected for any particular experiment will depend upon the size of the expected probe-target hybrids. Hybrid size is often determined by whether the target nucleic acid is comprised of RNA or DNA. When RNA comprises the target and the riboprobe employed is shorter than the target, overhanging, unhybridized, single stranded portions of the target, along with unhybridized probe molecules, will be degraded during the RNAse treatment. An equivalent (in terms of nucleotide sequence) but DNA-based target will not be so degraded, leaving a larger probe-target hybrid. As a result of this, different column matrices may need to be employed to achieve proper separation between probe- target hybrids and unhybridized, degraded probe molecules. As an example, FIG. 3 shows the results of a PF 4-specific riboprobe comprised of 163 ribonucleotides hybridized to its complementary RNA target. After RNAse digestion, the solution components were separated using an S200 column. The S200 column had a bed volume of 4.4 mL in a 0.7 cm x 11.5 cm column. In contrast, FIG. 4 shows the results obtained when a HPV18-specific riboprobe comprised 1944 ribonucleotides was hybridized to its complementary DNA target. After RNAse treatment, the solution components were separated using a 10.6 mL CL-4B bed in a 1 cm x 16 cm column. To maximize the invention's efficiency, a column containing the desired molecular sieving matrix may be prepared for each sample as follows during the 2 hr. hybridization: 22 mL of a 50% slurry of Sepharose CL-4B [prewashed in Column Buffer (TE + 10 mM NaCl) ] is poured into a 1 cm x 20 cm column (BioRad, cat. no. 737- 1020) . Each column is washed with 12 mL of Column Buffer and allowed to drain until the Column Buffer reaches the upper surface of chromatography matrix, at which point the column bottoms are capped.

To run the columns, 3.2 mL of Column Buffer is carefully loaded onto the top of each column bed. The more dense nucleic acid-containing hybridization samples are then carefully layered under the aforementioned 3.2 mL volume at the interface of the Column Buffer and chromatography matrix. The caps are removed from the column bottoms, thereby allowing the columns to run. Each column is drained until the initially loaded Column Buffer drops to the upper surface of the column bed. The initial 3.2 mL of column effluent is discarded. An additional 2.5 mL of Column Buffer is then added to each column. Each column is run to dryness and the effluent, the void volume, is collected.

Following collection, the presence of probe- target hybrids is assayed and/or quantitated. The detection method employed depends upon the detectable label utilized. When the label is a radioisotope, gamma or photographic detection, autoradiography, and particularly liquid scintillation counting, may be used. When one member of a specific binding pair (including antibodies and their specific antigens) is covalently linked to the probe, the second member of the pair can be used for detection and quantitation. If the label is an enzyme or enzyme inhibitor, reactions assaying for the presence or absence of enzymatic activity may be used. In addition, if other detectable label substances are used, such as fluorescent or chemilumenescent markers, etc. , other appropriate means of detection and quantitation can be utilized.

Liquid scintillation is the preferred detection and quantitation method when radioisotopes are employed as the detectable label substance. For example, when a sample comprising the 2.5 mL void volume is collected from a column in a 20 mL glass liquid scintillation vial, 5 mL of a fluor, preferably Liquiscint® (National Diagnostics, Manville, NJ) , is added to each vial (when ³⁵S is the radiolabel) . The vials are then capped, the contents vortexed, and the samples counted for 20 minutes in a Beckman LS 100 scintillation counter with a window setting of 100 - 1000.

The following examples are offered to more fully illustrate the present invention. In addition, the examples provide preferred embodiments of the present invention but are not meant to limit the scope thereof. Examples 1 and 2 are specifically directed to diagnostic methods for detecting the presence of HPV in clinical samples using riboprobes, while Example 3 illustrates application of the invention to methods of assaying mRNA levels in an in vitro model of gene expression in megakaryocytes.

EXAMPLE 1 HPV Probe Preparation

The detection of HPV infection, coupled with identification of the particular HPV type present, is important in the diagnosis and treatment of certain tissue lesions and cancers. The following procedures describe the analysis of clinical samples for the presence of HPV and identification of HPV types using RNA probes complementary to HPV regions of high type- specificity. Four RNA probes, each specific to a particular HPV type (6, 11, 16 and 18), were employed.

Probe Preparation

The complete nucleotide sequences are available for HPV 6 [Schwartz et al . , (1983) EMBO J. , vol . 2, pp. 2361 - 2368], HPV 11 [Dartmann et al . , (1986) Virol . , vol . 151, pp. 124 - 130], HPV 16 [Seedorf et al . , (1985) Virol . , vol . 145, pp. 181 - 185], and HPV 18 [Cole et al . , (1987) J. Mol . Biol . , vol . 193, pp. 599 - 608] . To generate vectors from which the specific RNA probes used herein could be made, HPV DNA from clones harboring the nucleotide sequences of the various HPV subtypes was cloned into plasmids capable of directing the synthesis of mRNA from inserted DNA molecules as described below.

HPV 6

For HPV 6, 10 μg of total cellular DNA containing HPV episomal monomers was obtained from a clinical sample comprising a genital wart. This DNA was cleaved with Bam HI and cloned into λL47 [Loenen et al . , (1980) Gene, vol . 10, p. 249] that had been similarly restricted. λL47 was selected for cloning because it accepts Bam HI inserts of about 5 kb to 19 kb in length. As human DNA is mostly cut into fragments smaller than can this lower insert-size threshold, a large (> 10- fold) enrichment for HPV DNA can be accomplished using this vector. Following ligation, the reaction products were packaged into infectious phage particles using the Gigapack® (Vector Research Systems) in vitro packaging system. Using this system, packaging efficiencies of 0.2 x 10⁸ to 1.0 x 10⁸ plaque forming units (pfu) per μg λ DNA were typically obtained. The packaged phage particles were then used to transfect E . coli strain LE392 (A.T.C.C. accession no. 33572) harboring integrated P2 phage.

An aliquot containing no more than 50,000 phage particles in 50 μL was taken from each packaging reaction and combined with 0.3 mL of fresh, log phase LE392 grown in LB (Sambrook et al . , supra) plus 0.2 % maltose. Phage particles were allowed to adsorb to the cells by incubating the sample 20 min. at 37^*C without shaking. Following adsorbtion, each sample was combined with 6.5 mL of molten 0.7% top agarose plus 10 mM MgS0₄ that had been maintained at 47^*C, gently mixed, and then poured onto dry, pre-warmed 150 mm LB agar plates. After the top agarose solidified, the plates were placed at 37"C until plaques were just beginning to overlap. The plates were then placed at 4°C for 1 hr. Plaque lifts were performed by overlaying nitrocellulose filters (Scheichen and Schuell) for 60 sec. After keying, each filter was removed and placed, DNA-side up, on Whatman 3MM paper barely saturated with 0.5 N NaOH,

1.5 M NaCl to denature the phage particles. After 1 - 5 min. of denaturation, each filter was transferred to a piece of Whatman 3MM paper barely saturated with neutralizing solution (1.5 M NaCl, 0.5 M Tris-HCl, pH 7.4) for 5 min. Following neutralization, the filters were rinsed in a shallow tray containing 2X SSC (0.3 M NaCl, 0.03 M sodium citrate) and then laid out, DNA-side up, to dry an Whatman 3MM paper. Duplicate filters were generated for each plate following the same procedure, except that each duplicate filter was allowed to whet for 3 min. on each plate.

After all the filters had been air dried, they were sandwiched between layers of Whatman 3MM paper and baked at 80'C under vacuum for 2 hr. Next, the filters were allowed to cool to room temperature, after which they were completely wetted in a shallow tray containing 2X SSC. The filters were then prehybridized together in about 2 mL of prehybridization buffer (6X SSC, 0.1% SDS, 5X Denhardts, 50 μg/mL sheered salmon sperm DNA) per filter at 55^*C for 4 hr. with gentle agitation. The prehybridization buffer was then removed and replaced with hybridization buffer, which comprised the same components as the prehybridization buffer. To this, approximately 10⁶ cpm/mL of a ³²P-endlabeled oligonucleotide probe ( [SEQ ID NO: 1] : 5' - CAC TGC TGG ACA ACA TGC ATG GAA G - 3 * ) , prepared according to the method of Caruthers et al . , [(1985) Science, vol . 230, pp. 281 - 285] (as were all other oligonucleotides used in this invention) was added. Hybridization was conducted for 16 hr. at 60"C, after which the filters were removed and washed in about 200 mL 2X SSC, 0.1% SDS with gentle shaking at 60°C for 30 min. This washing step was repeated twice.

To identify plaques containing DNA molecules complementary to the probe, autoradiography of the stringently washed filters was performed. After developing the exposed X ray film, the position and orientation of the filters marked onto the film. Comparison of the duplicates for each plate revealed a number of putative positive plaques. Of these potentially positive plaques, six were pulled using small, disposable glass pipettes to isolate those regions of the plate corresponding to the signal on the film. Each plug was placed in 1 mL of SM [per L: 5.8 g NaCl, 2 g gSθ₄-7H2θ, 50 mL 1M Tris-HCl, pH 7.5, 5 mL 2% gelatin, H2O to 1 L, sterilized by autoclaving] containing a drop of chloroform.

Each plug was then titrated to determine the number of plaque forming units present. Titration was accomplished by serially diluting 100 μL of each plug in 1 mL of SM. 50 μL of each serial dilution was then used to transfect 0.3 mL of fresh LE392 as described above. Following adsorbtion, each solution was combined with 3 mL of top agarose and plated on 90 mm LB plates. Outgrowth at 37^*C for 12 hr. enabled the determination of the number of plaque forming units (pfu) in each plug. Using this number, another transfection designed to produce 500 well separated plaques on a 90 mm plate was conducted according to the previously described method. Plaque lifts from these plates were performed as before, followed by filter hybridization using the same probe. This plaque purification process resulted in four purified λL47 clones possibly containing the HPV 6 genome.

To determine which clone (s) actually contained HPV 6 DNA, DNA from the various clones was purified according to the plate lysate method [Sambrook et al . , supra] . The resultant DNAs were digested with Bam HI and electrophoresed on a 0.7% agarose gel. A Southern blot [Sambrook et al . , supra; Southern, E.M. (1975) J. Mol . Biol . , vol . 98, p. 503] was then performed. The same oligonucleotide probe as was used to determine which λL47 clones contained HPV 6 DNA was hybridized to the blot, and autoradiography revealed each of the four clones contained HPV 6 DNA.

DNA from one of the HPV 6 DNA-containing clones, designated λ47HPV6a-l, was then digested with Bam HI and electrophoresed on a 0.7% agarose gel. The 7.9 kb band, corresponding in size to the HPV 6 genome, was excised and the DNA extracted according to the methods described in Sambrook et al . , supra; Parker et al . (1980) Methods Enzymol . , vol . 65, pp. 358-363. The HPV 6 DNA was then ligated into pGEM-3 that had been previously digested with Bam HI and treated with bacterial alkaline phosphatase (BAP) to prevent reannealing of uninserted vector. These ligation products were transformed into

E. coli strain DH1 (A.T.C.C. accession no. 33849) and plated onto LB media plates containing 100 μg/mL ampicillin. One of the clones, pGEM-3:HPV6 (+) , was selected for further use. Subclones of pGEM-3:HPV6 (+) were prepared by digestion with Sau 961. Sticky ends were blunted with T4 DNA polymerase [Sambrook et al . , supra] . These fragments were then ligated into Sma I- digested pGEM-3, which contains the T7 and SP6 promoter sequences at opposite ends of the multiple cloning site. The various portions of the HPV genome represented in these subclones were then deduced by restriction mapping and confirmed by DNA sequencing [Sanger et al . (1977), Proc . Nat 'l . Acad. Sci . , vol . 74, p. 5463] .

Riboprobes were then prepared from the various subclones. Initially, plasmid DNA was cleaved with Pvu II and Hind III to release inserts comprising the HPV 6 DNA in functional association with the T7 RNA polymerase promoter. The digestion products were then electrophoresed through a 1% agarose gel. High specific activity RNA-based probes were prepared after isolating (by electrolelution) and purifying the desired PvuII - Hind III fragments and transcribing 0.2 - 1.0 μg of DNA at room temperature as previously described.

The various HPV 6-specific riboprobes were then hybridized to human placental DNA using the Slot- Blot procedure [Sambrook et al . , supra] . Several of the probes cross-hybridized with human DNA, but the riboprobe generated from subclone pGEM-3:HPV6.1 [A(+) ] harboring base pairs 1072 - 2214 (as reported by Schwartz et al . , supra) of the HPV 6 genome (designated Fragment A) , did not cross-hybridize and was adopted for use in the column and blot assays .

b. HPV 11

A procedure similar to that described above was employed to generate the vector used to make the riboprobe specific for HPV 11. Initially, 10 μg of total cellular DNA was extracted from a vulvar condyloma tissue sample. The DNA so isolated was then digested with Bam HI and ligated into λL47 which had been similarly cleaved. The ligation products were then packaged into functional phage particles using the Gigapack® system (as described above for HPV 6) and transfected into LE392P2 as had been done for HPV 6.

After plating and outgrowth, plaque lifts were performed as described above. However, in this case, one duplicate from each plate was to be screened with a probe specific for HPV "early" genes, while the other duplicate was to be screened with a probe specific for HPV11 "late" genes. Those filters to be screened with the "early" probe were prehybridized and then probed separately from the duplicates to be probed with "late" probe. In each instance, the filter bound DNA was prehybridized overnight in 50 mL of prehybridization buffer with gentle agitation at 55^*C. After prehybridizing, 6 mL of hybridization solution, comprising about 916,000 cpm/mL of 5' ³²P-endlabeled oligonucleotide "late" probe [SEQ ID NO: 2; 5' - CTG TGG TAG ATA CCA CAC GCA GTA C - 3 ' ] , was exchanged for the prehybridization solution for the filter set to be probed with the "late" probe. In the case of the filters to be screened with a labeled "early" probe, 6 mL of hybridization solution comprising about 683,000 cpm/mL of 5' ³²P-endlabeled oligonucleotide "early" probe [SEQ ID NO: 3; 5' - CAC TGC TGG ACA ACA TGC ATG GAA G - 3'] . Each filter set was hybridized for approximately 9 hr. at 55°C, after which time the hybridization temperature was elevated to at 60"C for an another 6 hr. Following the hybridizations, all of the filters were washed twice for 45 min. in 6X SSC, 0.1% • SDS, first at room temperature, then at 60°C. The filters were then washed twice (45 min./wash) at room temperature in 2X SSC, 0.1% SDS. Autoradiography revealed more stringent washing conditions were required, so the filters were rewashed with gentle agitation in 2X SSC for 4 hr. at 60^*C, followed by a 10 hr. wash at the same temperature in 0.2X SSC.

Potential positive plaques (those which hybridized to both the "early" and "late" probes) were identified by autoradiography. Seven of the potential positives were pulled for further analysis. Plaque purification (performed according to the same procedures as for HPV 6, supra) confirmed that all of the original λL47 isolates harbored the HPV 11 genome. DNA from each of these clones was then grown up and isolated according to standard plate lysate and liquid culture procedures. See Sambrook et al . , supra . These DNAs were then subjected to restriction analysis to insure the presence of the entire HPV 11 genome. DNA from one clone, λL47.11.1, was digested with Bam HI. The restriction products were separated by agarose gel electrophoresis. The 7.93 kb band corresponding in size to the viral genome was cut from the gel, eluted as described previously, and cloned into a Bam Hi-cleaved, BAP- treated pSP65 plasmid vector (Promega; Melton et al . , (1984) Nucl . Acids Res . , vol . 12, pp. 7035 - 7056) . Following transformation of competent DHl cells, a HPV 11-containing clone designated pSP65.11.5 (+) , was retained. pSP65.11.5 (+) was then digested with Fok I. The 3' exonuclease activity of T4 DΝA polymerase was used to blunt the 3' overhangs resulting from the Fok I digestion. Subgenomic clones were generated in pGEM-3 from two regions of relatively high type specificity. One clone, designated pGEM-3:11.1.3, contained a Fok I - Hpa I fragment spanning nucleotides 2415 - 4157 (as reported by Dartmann et al . , supra) of the HPV 11 genome. pGEM-3:ll .1.3 yielded a highly specific riboprobe (prepared by digesting pGEM-3 : 11.1.3 with Pvu II and Hind III, isolating the resultant DNA fragment, and transcribing the DNA molecule as described, supra) used in all further experiments. RNA probes prepared from the full length HPV 11 clone (pSP65-ll .5 (f) ) cross- hybridized more strongly to human DNA.

HPV 16

The vector used to synthesize HPV 16-specific riboprobes was generated in a fashion similar to those for the HPV 6 and HPV 11 vectors. First, 10 μg of total cellular 16 DNA was isolated from an oral lesion tissue sample having a morphology consistent with HPV infection. This DNA was then restricted with Bam HI and cloned into Bam Hi-digested λL47. The resultant DNA was then packaged into infectious phage particles as described above, followed by infection of LE392P2. After plaque formation, plaque lifts were performed. The filters were then hybridized at 60"C with two oligonucleotide probes specific for "early" and "late" regions of the HPV16 genome. In the hybridization with the HPV16 "early" probe (5' - AAG CAG AAC CGG ACA GAG CCC ATT A - 3'; SEQ ID NO: 4), 0.5 pmol (1 x 10⁷ cpm) of "early" probe was used, while 0.1 pmol (0.2 x 10⁷) of the HPV16 "late" probe (5' - CTT GCC TCC TGT CCC AGT ATC TAA G - 3'; SEQ ID NO: 5) was used to screen the duplicate filters . After hybridization, the filters were washed twice for 20 min.⁷at room temperature and one time at 55"C in 6X SSC, 0.1% SDS. Autoradiography of these filters revealed ten putative positive plaques (those hybridizing to both of the HPV16 "early" and "late" probes), each of which was isolated and plaque purified. Seven of these plaque-purified clones were found to carry the HPV 16 genome through restriction analysis. Each of these clones was grown up and the DNA purified therefrom. Aliquots of DNA from each isolate was then cleaved with Bam HI to release the 7.9 kb HPV 16 genomic fragment, and DNA from one isolate, designated λL47.16.2, was then purified by agarose gel electrophoresis and the DNA eluted from excised gel slices. This DNA was then cloned into pSP65 that had been digested previously with Bam HI and BAP-treated. After transformation, a clone harboring the HPV 16 genome, designated pSP65.16.8 (+) , was retained for further use.

Once again subgenomic clones spanning the HPV genome were generated by Fok I digestion of pSP65.16.8 (+) . Following restriction, the 3' exonuclease activity of T4 DNA polymerase was used to blunt the 3 ' overhangs resulting from the Fok I digestion. The resulting blunt-ended DNA fragments were then ligated into the Sma I site of pGEM-3. Several clones containing large HPV 16 subclones were tested for their extent of cross-hybridization to human DNA on Southern blots and in solution hybridization. One subclone comprising 1211 nucleotides, designated pGEM- 3:HPV16B(-), spanning from nucleotide position 683 through nucleotide position 7376 of the 7904 bp HPV 16 genome as reported by Seedorf et al . , supra, was selected for use in synthesizing riboprobes from the adjacent T7 promoter.

RNA probes were generated from pGEM:HPV16B(-) by digesting the plasmid with Pvu II to release a fragment containing 1081 base pairs of the HPV 16 genome in functional association with the T7 promoter from pGEM-3. After isolating this fragment, RNA probes were synthesized as described, supra . d . HPV 18

Riboprobes specific for an HPV subtype, HPV 18, which often integrates into the human genome, as opposed to remaining in episomal form, were also prepared. HeLa cells (A.T.C.C. accession no. CCL 2) are known to contain several integrated subgenomic copies of HPV 18 DNA in Hind III fragment of three different lengths. Sequence and mapping information about HPV 18 in HeLa cells presented at the Papilloma Virus Workshop (Cold Spring Harbor, N.Y., September 1986) was used to generate and screen a library of HeLa DNA in λL47 as follows .

DNA was isolated from HeLa cells according to methods in Molecular Cloning, Maniatis et al . , Cold

Spring Harbor Laboratory, 1982. 30 μg of HeLa DNA was Hind Ill-digested in a total volume of 50 μL. After complete digestion, the reaction was first extracted with phenol, then with chloroform, and finally with ether. DNA was then precipitated with ethanol. Following precipitation, the DNA was resuspended overnight in 25 μL TE. Next, 6 μL of this Hind Ill- digested HeLa DNA was ligated to 1 μg of Hind Ill- prepared λL47 arms. Inserts ranging from about 6.73 kb to 21.6 kb can be efficiently packaged using λL47 arms prepared with Hind III. 4 μL of the ligation reaction was next packaged into infectious phage particles using the Gigapack® system, supra . After titering the packaging reaction, ten transfections were performed. In each transfection, 50 μL of the packaging reaction was used to transfect 100 μL of fresh log phase VCS257 (supplied with the Gigapack® kit) cells, followed by plating in 7 mL top agarose onto 150 mm LB agar plates. After plaque formation, plaque lifts in duplicate were performed as described previously, followed by denaturation, neutralization, and baking. Ten filters were prehybridized together in 100 mL of prehybridization buffer, supra, at 55°C. The filters were then hybridized in 50 mL of hybridization buffer for about 60 hr. Those filters that had been exposed to the plates initially were probed with a mixture of two ³²P-kinased, HPV 18-specific oligonucleotides (96-21 [SEQ ID NO: 6] and 96-22 [SEQ ID NO: 7]) at 42°C. The duplicate filters were probed with the HPV 18-specific oligonucleotides 96-19 (SEQ ID NO: 8) and 96-23 (SEQ ID NO: 9) at 38 ^*C. After hybridization, the filters were washed once at room temperature and twice at their respective hybridization temperatures in 6X SSC, 0.5% SDS, followed by two 4 hr. washes in 6X SSC, 0.1% SDS at their respective hybridization temperatures. Subsequent autoradiography of the filters lead to the identification of five clearly duplicated signals. The corresponding plaques were pulled, plaque purified, and grown up. Clones containing portions of the HPV 18 genome were obtained and verified by restriction enzyme mapping. Southern blotting with labeled HPV 18 DNA as a probe and DNA sequence analysis was also performed. One clone, designated λHPV-18.3, was chosen for further study. The HPV 18 DNA portion from λHPV-18.3 was excised by digestion with Hind III. A fragment comprising 4298 base pairs of the HPV 18 genome was released. Following gel purification and elution, this fragment was ligated into Hind Ill-cleaved pGEM-3. Only one resulting HPV-18 clone, designated pHPV18H, contained the desired portion of HPV 18 genome.

Interestingly, pHPV18H also comprised two copies of pGEM-3. pHPV18H was subjected to a variety of restriction enzyme digests and the various fragments cloned into Hind III/Pst I digested pGEM-3. Several subclones were identified as useful by restriction mapping and by sequence homology to corresponding regions of HPV 18. One subclone, designated pHPV18-92 and comprising about 3.5 kb of the HPV 18 genome, was found to yield RNA probes of superior specificity (lowest human DNA cross-hybridization) when the Pvu II fragment (comprising the T7 promoter and 1944 bp of HPV 18, spanning nucleotides 6766 to 853 of the 7857 bp HPV 18 genome as reported by Cole et al . , supra) was used to generate riboprobes by the procedures described previously for the HPV 6, HPV 11, and HPV 16 RNA probes. pHPV18-92 also contained no human DNA.

EXAMPLE 2 Diagnostic Screening for HPV

Having type-specific riboprobes available for each of HPV 6, 11, 16, and 18, various tissue samples could be screened to determine if HPV was present. Any sample shown to contain HPV could then be rescreened with individual type-specific probes for further characterization and quantitation. In this example, tissue samples from obtained during cervical exams of 172 women were assayed for HPV infection using slot blots and the riboprobe-column assay of this invention. 98 of the samples were obtained from women at high risk for HPV infection. These women were classified as "high risk" because they were patients at the University of Alabama, Birmingham's sexually transmitted disease clinic. The other 74 samples studied were obtained from women assessed as being at low risk for HPV infection, i.e., members of the general public. In the riboprobe- column assays, the analyzes were performed according to the first of the two alternative riboprobe-column assay methods described below. However, either method is suitable for practicing this invention. a. Riboprobe-Column Assay Method 1.

Initially, each of the 172 samples were split into two parts. One group was used for slot blot analysis. The other was assayed using the riboprobe- column assay described of the invention. Both sets of samples were lysed as follows . Each clinical sample was placed into a 1.5 mL Eppendorf tube and centrifuged 30 sec. at 12,000 x g to pellet the cells. Supernatants were discarded and each pellet was resuspended in 1.0 mL of PBS (10 mM Na₂P0 , 138 mM NaCl, pH 7.4) . The cells were again pelleted by a 30 sec. centrifugation at 12,000 x g and the supernatants discarded. The pellets were resuspended in 300 μL Lysis Buffer (10 mM Tris, pH 8, 1 mM EDTA, 20 mM DTT, 0.1% SDS, and 50 μg/mL

Proteinase K) by vortexing. The cells were lysed by allowing the samples to incubate 15 min. at room temperature. Cellular debris was pelleted by centrifugation at 12,000 x g for 1 min. For the samples to be assayed by slot blot, subsequent to pelleting cell debris, each supernatant was extracted two times with phenol, once with phenol/chloroform, and once with chloroform. After transferring the aqueous phases to new tubes, DNA was precipitated by adding 0.1 volume 3 M sodium acetate and two volumes of ethanol. After being mixed and stored on dry ice for 20 min., precipitated nucleic acids were pelleted at 12,000 x g for 15 min., rinsed once with 70% ethanol, dried, and resuspended in 30 μL TE. 5 μL of each sample was then spotted onto GeneScreen Plus®. The spotted DNA was then denatured and neutralized, followed prehybridization at 56^*C in approximately 1.5 mL/blot. After 4 hr., about 8 x 10⁶ cpm/mL (a mixture of approximately 2 x 10⁶ cpm/mL of each of the four riboprobes for HPV 6, 11, 16, and 18, prepared as described above) was added. Following an overnight hybridization, the blots were then washed for a total of 1.5 hr. in 0.IX SSC, 0.1% SDS at 56^*C, after which they were autoradiographed. Of the 74 low risk samples, four hybridized to the probe mixture in this slot blot assay, while in the high risk group, 26 samples out of 98 samples hybridized.

In those portions of each clinical sample to be analyzed by the riboprobe-column assay, after pelleting cellular debris following treatment with Lysis Buffer, a 50 μL aliquot from each sample was removed and transferred to a fresh 1.5 mL Sarstedt screw cap tube for hybridization. The remainder of each lysed sample was placed at -80°C.

Prior to hybridization, DNA present in each of the 50 μL aliquots of the various samples was denatured by the addition of 10 μL of 1 M NaOH, followed by a 5 min. room temperature incubation. Following denaturation, 86 μL of 4.8 M Na₁.₅H₁.₅PO₄ was added to each tube, the contents of which were then vortexed. Next, 4 μL of ³⁵S-labeled RNA probe, made of riboprobes specific for HPV 6, 11, 16, and 18 (approximately 20 amol and 15,000 cpm of each), was added and mixed with each sample, followed by a 10 sec. centrifugation. The hybridization reactions were conducted for 2 hr. at 72°C.

After the hybridization reactions were completed, unhybridized riboprobes were degraded by the addition of 12 μL RNase mix, supra, to each sample. Again the samples were briefly (10 sec.) centrifuged before being placed at 37°C for 10 min. The RNase incubation was halted by adding 195 μL Formamide-dye mix, supra, and vortexing. The samples were then loaded onto Sepharose® CL-4B columns (prepared as described above) . These columns were run and samples collected and counted as described, supra . The results are shown in Table 1. Table 1 Riboprobe-Column Assay Results

ςpm # of samples cpm # of samples CP # of samples

0 2 18 1 96 1

1 3 19 0 99 1

2 2 20 1 106 1

3 5 21 1 118 1

4 13 24 1 132 1

5 14 25 1 239 1

6 19 27 1 244 1

7 25 37 1 277 1

8 8 38 1 368 1

9 7 42 1 399 1

10 11 50 1 421 1

11 2 55 1 549 1

12 8 56 1 672 1

13 11 59 1 870 1

14 2 62 1 935 1

15 0 64 1 1680 1

16 2 85 1 4200 1

17 4 90 1

The limit of detection of the riboprobe-column assays was determined to be 21 cpm. Thus, those samples producing 22 or more cpm were considered positive for HPV infection. In the 74 low risk patients, four women were identified as having been infected with HPV. In contrast, in the high risk group, 27 out of 98 women were found to be infected with HPV. In all cases where HPV infection had been determined by slot blot, infection was also detected by the riboprobe-column assay. Importantly, the instant method also identified one additional patient as having an HPV infection that was not detected by the slot blot assay. This result was subsequently verified (see below) . As these results indicate, the described riboprobe-column diagnostic procedures are an improvement over existing nucleic acid hybridization-based diagnostic methods, such as slot blotting, in terms of both speed and sensitivity. In those samples testing positive for HPV infection, rescreening was done with riboprobes specific for HPV 16 and 18 due to the correlation between the location of the infection and HPV subtypes. Duplicates of each HPV-positive sample were rescreened by the riboprobe-column method as described above. However, in these assays, each sample was screened using only one HPV type-specific riboprobe, either for HPV 16 or HPV 18. Of the 31 women above testing positive for HPV infection by solution hybridization, 25 were found to be infected with HPV 16, four were infected with HPV 18, and one woman had simultaneous HPV 16 and HPV 18 infections. In addition, the one patient who tested negative for HPV infection via slot blot screening but who was identified as HPV-infected using solution hybridization was found to be infected by the closely related subtype HPV 33 [Cole et al . , (1986) J. Vir . , vol . 58, no. 3, pp. 991 - 995] upon subsequent isolation, cloning, and sequencing of viral DNA obtained from a clinical sample .

Riboprobe-Column Assay Method 2.

The second riboprobe-column assay HPV detection method, infra, is a simplification of the first method. Its purpose is to lessen both the total of time and the hands-on time required to conduct the assays. Normally, the first method, supra, requires about 290 min. from start to finish (when conducted by one familiarized with the procedure) , with approximately 130 min. of that requiring actual participation by the individual conducting the assays. In contrast, the simplified method shortens total assay time to about 235 min. and hands-on time to 85 min., thereby saving about 55 min. in total time and reducing hands-on time by nearly 0.75 hr. In the improved method, a clinical sample is collected directly in 500 μL of Lysis Buffer, supra, minus Proteinase K. Cells are lysed by adding Proteinase K to a final concentration of 50 μL/mL (through the addition of a stock solution comprising 10 μg/mL Proteinase K) and vortexing. Cell lysis occurs during the subsequent 15 min. incubation at room temperature. Cell debris is then pelleted by centrifuging the samples 1 min. at 12,000 x g. Next, 150 μL of each sample's supernatant is transferred to a 1.5 mL screw cap Sarstedt centrifuge tube and the remainder of each sample stored at -80°C.

Hybridizations are conducted by adding 270 μL of a premix comprised of 258 μL of 4.8 M Na₁.₅H₁.₅PO₄ and 12 μL of ³⁵S-labeled riboprobe (approximately 15,000 cpm) . After mixing the solutions, the hybridization reactions are conducted for 2 hr. at 72°C, followed by the addition of 12 μL of RNAse mix, supra, a 10 sec. centrifugation and a 10 min. incubation at 37°C, followed by the addition of 460 μL Formamide-dye mix, supra, to each sample. These samples, despite the slightly larger hybridization volume, are then chromatographed through Sepharose® CL-4B and counted as described in the first method.

EXAMPLE 3

Analysis of PF 4 mRNA Levels

To examine the differential expression of the gene encoding the platelet-related protein PF 4, HEL cells, which express phenotypic and biochemical markers for megakaryocytic cell lineages, among others, were studied by treating the cells with chemicals and various growth factors. Basal, low-level PF 4 mRNA expression is known to be significantly increased by phorbal esters [Tabilio et al . , (1984) Embo, J. , vol . 3, p. 445] . The HEL cells used in the following experiments were maintained at 37°C and 5% CO₂ in RPMI-1640 medium supplemented with 10% fetal calf-serum, 200 μM glutamine, and the antibiotics penicillin (100 units/mL) and streptomycin (100 μg/mL) . PF 4 mRNA induction was studied using phorbol

12-myristate 13-acetate (PMA; Sigma, cat. no. P 8134) . The PMA used in these experiments was previously prepared as a 10^-3 M solution in ethanol and stored at 80°C. Initially, 2 x 10⁵ HEL cells were inoculated into 2 mL of RPMI-1640 supplemented with 200 μM glutamine and the antibiotics penicillin (100 units/mL) and streptomycin (100 μg/mL) , 2% fetal calf serum, and 0, 10^-7, 10^-8, 10^-9, and 10^_10 M PMA. The cultures were incubated at 37°C in 5% CO2 for 72 hr. , after which time the cells were harvested and placed in microfuge tubes. Any cells adhering to the plates were released with a solution of 0.05% (w:v) trypsin. Viable cells were enumerated by trypan blue exclusion. The cells were then pelleted at room temperature and the supernatant discarded. Based on the trypan blue exclusion cell counts, the cell pellets were resuspended in a volume of lysis buffer (0.2% SDS, 10 mM Tris, pH 8.0, 1 mM EDTA, 20 mM DTT, and 100 μg/mL Proteinase K) so that an anticipated cell concentration of 3 x 10³ cells/μL was obtained. Lysates were either used immediately or stored at -80°C until further use.

To calculate the actual number of cells in the lysate, in some samples a 50 μL aliquot of the lysate was diluted to 1 mL in water and analyzed spectrophotometrically at 260 n against a blank of similarly diluted lysis buffer. The resulting absorbance was compared to the absorbance of a standard lysate containing a known number of cells . A linear relationship exists between the A26O ^{and the} number of cells in the lysate. To quantitate PF 4 mRNA levels following PMA induction, solution hybridization with PF 4-specific RNA probes was performed. The RNA probes utilized were prepared as follows: A double stranded DNA molecule encoding a nucleotide sequence identical to that for base pairs 106 - 266 of the published human PF 4 DNA sequence [Poncz et al . , (1987) Blood, vol . 69. pp. 219- 223] was generated by preparing two oligonucleotides having complementary regions of 22 bp at their 3' termini. The hybrid was made completely double-stranded by treating it with the Klenow fragment of E. coli DNA polymerase I in the presence of all four nucleotides according to procedures in Sambrook et al . , supra .

This blunt-ended DNA fragment was then ligated into the plasmid vector pSP65, supra, which had previously been cleaved with S a I and treated with bacterial alkaline phosphatase to remove the 5 ' phosphates to prevent religation of the vector itself. pSP65 contains the bacteriophage SP6 promoter immediately adjacent to the multiple closing site into which the 160 bp fragment was cloned. Following ligation, two clones were selected, R12 and R4. DNA sequence analysis of each clone revealed that R12 would be used to produce RNA probe molecules, while R4 would be used to produce RNA molecules complementary to the probe.

The RNA probes were generated by runoff transcription as follows. Initially, R12 was digested to completion with Hind III to linearize the plasmid just 3' to the insert. The reaction was then extracted with phenol:chloroform to remove the restriction enzyme. After the phenol :chloroform treatment, the solution was extracted with chloroform. The DNA was then precipitated and resuspended in water (or 10 mM Tris, 1 mM EDTA) at a concentration of 0.25 μg/μL.

Each probe-generating run-off transcription reaction occurred in either 100, 20, or 3 μL total volume depending upon the amount of probe needed. Solutions and reagents were added in the following order and the reaction maintained at room temperature, not 4°C, during the component additions. A 100 μL transcription reaction was conducted as follows: In a sterilized 0.5 mL Eppendorf tube, DePC-treated water [prepared as a 100 mL stock solution by adding 10 mL of 10% DePC (diethylpyrocarbonate) in absolute ethanol to 100 mL H₂O in a sterile container, allowing the solution to stand at room temperature for 5 min., followed by heating to 90°C for 5 min. and then cooling to room temperature] was added in an amount sufficient to render the final reaction volume 100 μL. 20.0 μL of 5X transcription buffer (200 mM Tris-HCl, pH 7.5 at 37°C, 30 mM MgCl₂, 10 mM spermidine, and 50 mM NaCl) was added next, followed by 10.0 μL of 100 mM DTT. 4.0 μL of RNasin™ ribonuclease inhibitor was then added to achieve a final concentration of 1 unit/μL. After adding RNasin™, 20.0 μL of a freshly prepared ribonucleotide stock solution comprised of 2.5 mM ATP, CTP, GTP (from 10 mM stocks neutralized to pH 7) and 25 μL radiolabeled ³⁵S-UTP (New England Nuclear, cat. no. NEG 039H) was added. Next, 2 μL of linearized R12 (2 - 5 μg) was added, followed by 10 - 50 units of SP6 RNA polymerase. The reaction was then incubated 60 - 120 min. at 37°C - 40°C. Typically, 5 - 10 μg of RNA probe was obtained per μg of DNA template.

As an alternative to the RNA probe synthesis procedure described above, another procedure may be used. Again, solution and reagents were added at room temperature and the reaction mixture is not kept on ice. Initially, DePC-treated water was added to a sterile reaction tube so that a final volume of 20 μL was attained. 4.0 μL of 5X transcription buffer was then added, followed by 2.0 μL 100 mM DTT. 0.8 μL of RNasin™ was then added. Next, 4.0 μL of a solution containing 2.5 mM ATP, CTP, and GTP (prepared by mixing equal volumes of 10 mM, pH 7, stock solutions of each of these ribonucleotides) was added, after which 2.4 μL of 100 mM UTP (final concentration of 12 μm) was added. As template, 0.2 - 1.0 μg R12 DNA was used (in a volume of 1.0 μL) . 5.0 μL of ³⁵S-labeled UTP (New England Nuclear, cat. no. NEG 039H) was then added, followed by 0.5 - 1.0 μL (5 - 15 units) of SP6 RNA polymerase. The reaction was incubated at 37°C - 40°C for 60 min. After the RNA synthesis reactions were completed by either of the above-described methods, the R12 DNA template was degraded by adding RNase-free DNase to a concentration of 1 unit/μg DNA. The DNA digestion reaction was placed at 37°C for 15 min., after which it was subjected to phenol:chloroform extraction, followed by a chloroform-only extraction. The RNA probes were then precipitated by adding 0.1 volume of 3 M sodium acetate and 2.5 volumes of ethanol and in incubating at -20°C for 1 hr. Probe was recovered by centrifugation at 12,500 rpm for 15 min. in a microfuge. The pelleted riboprobe was washed twice with 70% ethanol and then dried in a speed vac (Savant) . The riboprobe-containing pellet was next resuspended in 50 μL DePC-H₂0 in the presence of 20 mM DTT 50 μg/μL partially hydrolyzed polyadenylic acid, and 1 unit/μL RNasin™. At this point, the solution typically contained 2.5 x 10⁶ cpm/μL.

To quantitate the amount of probe generated by the run-off transcription reaction, another single- stranded RNA molecule having a ribonucleotide sequence complementary to that produced from R12 was made by run- off transcription of R . The same procedures as described above for producing the RNA probe were used to generate the quantitation probe, except that only trace amounts of ³⁵S-UTP were employed in conjunction with 0.2 mM (final concentration) unlabeled UTP. Quantitation of the R12-derived probe was conducted by performing a series of titration assays. Each titration assay employed the same amount of probe, namely about 20,000 cpm, and increasing amounts of target (here the quantitation probe derived from R4) . In each assay, probe having a specific activity of 2 x 10⁹ cpm/μg was present in excess. Each titration solution hybridization was conducted in a final volume of 100 μL and contained the following components: 2.7 M Na₁.₅H₁.5PO₄, 50 μg/mL partially hydrolyzed polyadenylic acid, and 20 mM DTT. The hybridizations were conducted for 2 hr. at 84°C, after which time the reactions were cooled and RNase digested for 20 minutes at 37°C by adding 10 μL of a solution comprising 500 units RNase Ti and 3 μg boiled RNase A (from fresh stock solutions of 5xl0⁵ units/mL RNase Ti and 10 μg/μL RNase A) .

Upon completion of the RNase digestions, 240 μL formamide was added to each titration assay. The 350 μL reactions were then each loaded onto separate 4.1 mL bed volume Sephacryl® S-200 (Pharmacia) columns that had been equilibrated previously in Column Buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDTA and 50 mM NaCl, and capped to prevent sample flow-through. 2.0 mL of Column Buffer was then added to each column, which were then allowed to run by removing the caps. The 2.35 mL excluded volume from each column was collected in a scintillation vial. The hybrid-containing samples were then subjected to liquid scintillation counting after the addition of 5 mL (approximately 2 volumes) of Liquiscint®. This titration data was used to produce a standard curve showing the amount of signal generated per given mass of target mRNA. And because the target and probe were of known size, the signal generated could be correlated with the number of target RNA molecules complexed in probe-target hybrids . Cellular PF 4 mRNA levels were quantified by adding 70 μL of probe mix to 30 μL of cell lysate so that the solution hybridization conditions were equivalent to those used in the titration assays. The hybridizations, followed by RNase treatment and then molecular exclusion chromatography, were conducted as described above. This procedure allowed the detection of as few as two PF 4 mRNAs per cell.

The results in FIG. 5 indicate that PF 4 mRNA expression in HEL cells was stimulated by PMA. PF 4 mRNA expression increased linearly with logarithmic increases in PMA concentration. The kinetics of PMA induction of HEL cell PF 4 mRNA expression was also examined by conducting solution hybridizations as described above in HEL cells stimulated with PMA for various time periods. Induction of PF 4 mRNA expression was initiated and reached maximal levels between 24 and 48 hr. after stimulation. See FIG. 6. The kinetics of the response were not influenced by PMA concentration. This delayed expression indicated that PF 4 mRNA synthesis was possibly a secondary consequence of PMA treatment. This hypothesis was confirmed by stimulating HEL cells with 10"⁸ M PMA in the presence (2 μg/mL) or absence of cycloheximide. Table 2 PF 4 mRNA Levels

PF 4 mRNA molecules [PMA] Cycloheximide (2 μg/mL) per cell

0 - 48

0 + 29

10~⁸ - 1174

10^-8 + 24

De novo protein synthesis was required for PF 4 mRNA expression to be induced by PMA in HEL cells, as PF 4 mRNA levels were not increased above constitutive levels when protein synthesis was inhibited by cycloheximide.

While the present invention has been described in terms of preferred embodiments, it is understood that variations and modifications will occur to those skilled in the art in light of the above description. Therefore, it is intended that the appended claims cover all such variations which come within the scope of the invention as claimed. SEQUENCE LISTING (1) GENERAL INFORMATION:

(i) APPLICANTS: Amgen Inc.

(ii) TITLE OF INVENTION: A Nucleic Acid Diagnostic Method Using Riboprobes, RNase Digestion, and Molecular Weight Exclusion Chromatography

(iii) NUMBER OF SEQUENCES: 9

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Amgen Inc.

(B) STREET: Amgen Center

1840 Dehavilland Drive

(C) CITY: Thousand Oaks

(D) STATE: California

(E) COUNTRY: USA

(F) ZIP: 91320-1789 (V) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Diskette, 3.5 in., DS, 1.4 MB

(B) COMPUTER: Apple Macintosh

(C) OPERATING SYSTEM: Macintosh OS 7.0.

(D) SOFTWARE: Microsoft Word Version 5.1a (vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER:

(B) FILING DATE:

(C) CLASSIFICATION:

(2) INFORMATION FOR SEQ ID NO:l: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 25 nucleotides

(B) TYPE: nucleic acid (C) STRANDEDNESS: single stranded

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:l: CACTGCTGGA CAACATGCAT GGAAG 25

(3) INFORMATION FOR SEQ ID NO:2: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 25 nucleotides

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single stranded

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2: CTGTGGTAGA TACCACACGC AGTAC 25

(4) INFORMATION FOR SEQ ID NO:3: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 25 nucleotides

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single stranded

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3: CACTGCTGGA CAACATGCAT GGAAG 25

(5) INFORMATION FOR SEQ ID NO:4: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 25 nucleotides

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single stranded

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4: AAGCAGAACC GGACAGAGCC CATTA 25 (6) INFORMATION FOR SEQ ID NO:5: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 25 nucleotides

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single stranded

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5: CTTGCCTCCT GTCCCAGTAT CTAAG 25

(7) INFORMATION FOR SEQ ID NO:6: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 nucleotides

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single stranded

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6: ACATGAGCTG GGCACTATAG 20

(8) INFORMATION FOR SEQ ID NO:7 : (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 nucleotides

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single stranded

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:7: GTTGCAGCAC GAATGGCACT 20

(9) INFORMATION FOR SEQ ID NO: 8 : (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 nucleotides

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single stranded (D) TOPOLOGY: unknown (xi) SEQUENCE DESCRIPTION: SEQ ID NO:8: TCTAGAATTA GAGAATTAAG 20

(10) INFORMATION FOR SEQ ID NO:9: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 nucleotides

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single stranded

(D) TOPOLOGY: unknown

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:9: GGGCTCTAAA TGCAATACAA 20

Claims

WHAT IS CLAIMED IS:

1. A method for detecting a target nucleic acid sequence using an RNA probe, the method comprising:

a) combining in solution a pool of nucleic acid molecules and an RNA probe capable of hybridizing to a target nucleic acid molecule encoding the target nucleic acid sequence under stringent conditions to form a probe-target hybrid nucleic acid molecule;

b) following hybridization, treating the solution of part (a) with RNAse;

c) separating the probe-target hybrid nucleic acid molecules from other solution components; and

d) recovering and detecting the probe-target hybrid nucleic acid molecules from part (c) .

2. A method according to Claim 1 wherein the target nucleic acid sequence is encoded by a DNA molecule.

3. A method according to Claim 1 wherein the target nucleic acid sequence is encoded by an

RNA molecule .

4. A method according to Claim 2 wherein the pool of nucleic acid molecules is produced from a biological sample.

5. A method according to Claim 4 wherein the target nucleic acid sequence is indicative of a pathogen selected from the group consisting of viral, bacterial, and fungal pathogens in the biological sample.

6. A method according to Claim 5 wherein the pathogen is a virus.

7. A method according to Claim 5 wherein the pathogen is a bacteria.

8. A method according to Claim 5 wherein the pathogen is a fungus.

9. A method according to Claim 4 wherein the target nucleic acid sequence is indicative of the presence of a neoplasm.

10. A method according to Claim 9 wherein the neoplasm is benign.

11. A method according to Claim 9 wherein the neoplasm is malignant.

12. A method according to Claim 4 wherein the target nucleic acid sequence is indicative of a heritable genetic disorder.

13. A method according to Claim 4 wherein the target nucleic acid sequence is a genetic marker.

14. A method according to Claim 13 wherein the genetic marker is associated with a heritable genetic disorder.

15. A method according to Claim 13 wherein the genetic marker is associated with the future development of a neoplasm.

16. A method according to claim 13 wherein the genetic marker is used for forensic analysis.

17. A method according to Claim 1 wherein the RNA probe comprises a detectable label substance selected from the group consisting of radioisotopes, fluorescent molecules, chemiluminescent molecules, and members of specific binding pairs .

18. A method according to Claim 17 wherein the detectable label substance is a radioisotope selected from the group consisting of ³H, ^{1 5}I, ³²P, and ³⁵S.

19. A method according to Claim 18 wherein the radioisotope is ³²P.

20. A method according to Claim 18 wherein the radioisotope is ³⁵S.

21. A method according to Claim 17 wherein the specific binding pair is avidin:biotin.

22. A method according to Claim 1 wherein the pool of nucleic acid molecules and RNA probe are combined for not less than 15 minutes but not more than eight hours .

23. A method according to Claim 22 wherein the pool of nucleic acid molecules and RNA probe are combined for about two hours .

24. A method according to Claim 22 wherein the quantity of the RNA probe exceeds that of the target nucleic acid.

25. A method according to Claim 1 wherein the method is used to quantitate the amount of a target nucleic acid sequence present .

26. A method according to Claim 1 wherein stringent conditions comprise:

(a) a temperature of about 15°C to 30 ^*C below the melting temperature of the probe- target hybrid;

(b) a pH of about 6.6 to 7.2; and

(c) an electrolyte concentration of about 2.0 M to 3.5 M.

27. A method according to Claim 26 wherein the target nucleic acid comprises RNA and the temperature is about 84 "C ± 4°C.

28. A method according to Claim 26 wherein the target nucleic acid comprises DNA and the temperature is about 72 °C ± 4°C.

29. A method according to Claim 26 wherein the pH is about 6.9 ± 0.2.

30. A method according to Claim 26 wherein the pH is about 6.9 ± 0.0.05.

31. A method according to Claim 26 wherein the electrolyte concentration is about 2.75 M ± 0.2 M.

32. A method according to Claim 31 wherein the electrolyte concentration is about 2.75 M ± 0.05 M.

33. A method according to Claim 26 wherein the electrolyte concentration is about 2.5 M ± 0.2 M.

34. A method according to Claim 33 wherein the electrolyte concentration is about 2.5 M ± 0.05 M.

35. A method according to Claim 1 wherein the probe-target hybrid nucleic acid molecules are separated from other solution components by molecular weight exclusion chromatography.

36. A method according to Claim 1 wherein the probe-target hybrids are detected by liquid scintillation.