CA2074214C - Improvements in the in situ pcr - Google Patents

Abstract

Improvements to the in situ polymerase chain reaction (PCR), a process of in vitro enzymatic amplification of specific nucleic acid sequences within the cells where they originate, can be achieved by changing the way that the enzymatic reaction is started. Reaction initiation is delayed until the start of PCR thermal cycling, either by withholding a subset of PCR reagents from the cellular preparation until the preparation has been heated to 50° to 80°C, immediately before thermal cycling is begun, or by adding to the PCR reagents a single-stranded DNA binding protein which blocks reaction at temperatures below about 50°C. If the in situ PCR is performed on cellular preparations already attached to a microscope slide, thermal cycling also is facilitated by use of a thermal cycler sample block or compartment designed optimally to hold the microscope slide and any vapor barrier covering the slide.

Description

~G~"~~2~.

The present invention relates to novel compositions, devices and me-thods for simplifying and improving the sensitivity and specificity of the in situ polymerase chain reaction, a method of amplifying and detecting S specific nucleic acid sequences within individual cells, and may be used in the fields of cell biology, forensic science and clinical, veterinary and plant pathology.
The polymerase chain reaction (PCR) is a method for increasing by many orders of magnitude the concentration of a specific nucleic acid sequence in a test sample. The PCR process is disclosed in U.S. Patent Nos.
4,683,195, 4,683,202 and 4,965,188.
The so-called in situ nucleic acid hybridization methods have evolved to detect target sequences in the cells or organelles where they originated (for a review of the field, see Nagai et al., 1987, Intl. J. Gyn. Path. ~, 366-379).
Typically, in situ hybridization entails (1) preparation of a histochemical section or cytochemical smear, chemically fixed (e.g. with formaldehyde) to stabilize proteinaceous subcellular structures and attached to a microscope slide, (2) chemical denaturation of the nucleic acid in the cellular prepara-tion, (3) annealing of a tagged nucleic acid probe to a complementary target 2 0 sequence in the denatured cellular DNA and (4) localized detection of the tag annealed to target, usually by microscopic examination of immobilized non-isotopic (absorbance or fluorescence staining) or isotopic (autoradiographic) signals directly or indirectly generated by the probe tag. However, conventio-nal in situ hybridization is not very sensitive, generally requiring tens to 2 5 hundreds of copies of the target nucleic acid per cell in order to score the presence of target sequence in that cell.
Recently, the sensitivity enhancement associated with PCR amplifica-tion of target sequence has been combined with the target localization of in situ hybridization to create in situ PCR, wherein PCR is performed within Nt/24. Juni 1992 - Z - 2~"~~2~.4 chemically fixed cells, before the fixed cells are attached to a microscope slide and the amplified nucleic acid is located by microscopic examination of autoradiographs following isotopically tagged probing (Haase et al., 1990, Proc. Natl. Acad. Sci. USA $~, 4971-4975). The cells may be suspended S during in situ amplification.
In situ PCR requires a delicate balance between two opposite require-ments of PCR in a cellular preparation: the cell and subcellular (e.g.
nuclear) membranes must be permeabilized sufficiently to allow externally applied PCR reagents to reach the target nucleic acid, yet must remain suf ficiently intact and nonporous to retard diffusion of amplified nucleic acid out of the cells or subcellular compartments where it is made. In addition, the amplified nucleic acid must be sufficiently concentrated within its com-partment to give a microscopically visible signal, yet remain sufficiently dilute that it does not reanneal between the denaturation and probe-annea-1 5 ling steps. Haase et al., supra, relied on paraformaldehyde fixation of cells to have created sufficient but not excessive permeability.
Haase et al., supra, used a series of PCR primer pairs to specify a series of overlapping target sequences within the genome of the targeted organism to improve retention of amplified target nucleic acid within the cellular 2 0 compartment where it was made. The resulting PCR product was expected to be so large (greater than 1,000 base pairs) that its diffusion from site of ori-gin should be greatly retarded. However, the use of multiple primer pairs severely reduces the practicality of in situ PCR, not just because of the ex-pense associated with producing so many synthetic oligonucleotides, but 2 5 even more seriously because many PCR target organisms, especially patho-genic viruses, are so genetically plastic that it is hard to find even a few short sequences which are sufficiently invariant to make good primer and probe sites. Other important target sequences, such as activated oncogenes, inacti-vated tumor suppressor genes and oncogenic chromosomal translocations, 3 0 involve somatic point mutations and chromosomal rearrangements which can be distinguished from the parental sequence if relatively short PCR pro-ducts are amplified from single primer pairs. Multiple primer pairs and long structures would frustrate attainment of the specificity offien needed to distinguish cancerous cells from their normal neighbors. Multiple primer 3 5 pairs jeopardize PCR in a different way as well; they promote primer dime-rization and mis-priming, reducing sensitivity and specificity and incre-asing the likelihood of false-negative results because nonspecific amplifica-tion radically reduces the yield of amplified target sequence.

207~+~1~

The present invention increases the convenience, se~t9iti~rit'~-~a-spec~-ficity of in situ PCR, also eliminating any need for multiple primer pairs to detect a single target sequence. In doing so, it also allows in situ PCR to discriminate among alleles.
S In a first aspect, the invention provides an improved method of in situ polymerase chain reaction (PCR) with increased amplification specificity and sensitivity. This improvement involves withholding at least one PCR
reagent from a preparation comprising fixed cells, PCR reagents and optio-nally a single-stranded DNA binding protein until the preparation has been heated to a temperature, in the approximate range of 50° to 80°C, where non-specific reactions of the nucleic acid polymerase are disfavored. The method applies equally whether nucleic acid amplification is performed before or after the fixed cells have been attached to a microscope slide.
In a second aspect, the improved in situ PCR method relates to the 1 S better specificity and sensitivity that result by including in the reaction mix-ture a single-stranded DNA binding protein (SSB) at a concentration which interferes with nonspecific polymerase reactions without blocking specific target amplification. A variety of naturally occurring, genetically engineered, or totally synthetic polypeptides with SSB activity can benefit in 2 0 situ FCR. This second aspect also is independent of the temporal order of nucleic acid amplification and cell attachment to slides.
In a third aspect, the invention relates to modified thermal cyclers used to automate PCR amplification, wherein the sample compartment used to transfer heat rapidly to and from the reaction holds microscope slides. In 2 5 one embodiment, the sample compartment comprises a metal block which has a horizontal flat surface dimensioned to hold one or several microscope slides with their largest dimensions oriented horizontally. The flat surface may lie at the bottom of a well suitable for holding a shallow mineral oil va-por barrier which prevents drying of the in situ PCR preparation during 3 0 thermal cycling. In another embodiment, the compartment comprises a me-tal block containing one or more slots which substantially and closely en-close microscope slides with their largest dimensions oriented in an appro-ximately vertical plane. Such orientation substantially increases the number of slides which can be analyzed at one time. In a third embodiment, the 3 S compartment holds a moving heattransfer fluid and contains holders for se-curing microscope slides in the fluid flow. The third embodiment also com-_ 2~''~~2~.~
prises plastic envelopes which encase the microscope slides and protect them from desiccation or PCR reagent wash-out.
The first two aspects of the present invention improve the specificity and sensitivity of in situ PCR; they reduce the chance of false negative results be-cause even cells containing only a single copy of target nucleic acid sequence can confidently be detected. The increased specificity simplifies the detection of amplified nucleic acid. Whereas in situ nucleic acid analysis traditionally has required annealing of tagged probe nucleic acid containing a sequence complementary to the target sequence, high amplification specificity allows 1 0 confident detection of tagged primers which have been incorporated into lon-ger nucleic acids, with decreased concern for false positive results which might arise from primer incorporation into nonspecifically amplified nucleic acids. Therefore, an additional probing step is no longer needed but still can be used. The increased sensitivity also simplifies detection of ampli-1 5 fled nucleic acid after in situ PCR, by generating so much analyte that non-isotopic signals can replace autoradiographically recorded isotopic signals.
Absorbance, fluorescence and chemiluminescence signals are faster, simpler and safer to record than is radioactive decay. Adoption of nonisoto-pic detection should greatly increase the appeal of in situ PCR to clinical pa-t 0 thologists and other practitioners of routine analysis (as opposed to biological and medical research).
The first two aspects of the invention also greatly increase the practi-cality and generality of in situ PCR by eliminating the need for multiple primer pairs for sensitive detection of a single target sequence. f~uite apart 2 5 from the expense, multiple primer pairs are hard to apply to highly poly-morphic target organisms, like many retroviruses, or to allele-specific am-plification such as is required for PCR detection of many oncogenic somatic mutations. Now that single primer pairs suffice for in situ PCR, the method will have the same breadth of application as conventional PCR. Such special 3 0 adaptions as multiplex PCR, degenerate priming, nested priming, allele-specific amplification, one-sided PCR and RNA PCR can be tried in situ with increased confidence in method transfer.
The second aspect of the invention is also a significant improvement.
HotStartTM methods block only pre-amplification aide reactions which yield 3 5 nonspecific products; SSBs also appear to reduce mis-priming which occurs during thermal cycling. Therefore, SSBs more effectively reduce nonspecific amplification. Too, inclusion of an SSB in the PCR reagent mixture elimina-- 5 - ~t~'~42:~4 tes the need to perform a manual Hot Start"H procedure, which requires some operator skill to effect a closely timed addition of the missing PCR re-agent without damaging or desiccating the in situ PCR preparation. A me-thod where all components of the assay are assembled at room temperature and covered with a vapor barrier before heating is begun is more reliable than one which requires manipulation of hot materials and vapor barrier addition to a hot system.
By facilitating routine application of in situ PCR, the first two aspects of the invention extend ultra-sensitive nucleic acid detection to new markets and practical problems, such as are presented by clinical, veterinary and plant pathology. These professional fields often rely on information regar-ding analyte location in biological samples to make critical judgments; con-ventional PCR does not easily yield that information. Furthermore in situ PCR is practically immune to the creation of falsepositive results by conta-1 5 urination of reactions with amplified target from previous reactions, because the analyte shows subcellular localization, usually in the nucleus. In addi-tion, multiple staining, for example, for cell-surface antigens, permits di-sease diagnosis and prognosis based on infection rates of cellular subpopula-tions. In situ PCR applied to blood or biopsy samples from patients believed to 2 0 be infected by a lymphotrophic retrovirus, such as HIV-1, should yield va-luable prognosis information such as the fraction of CD4 (surface antigen) plus cells carrying integrated viral genomes or viral particles.
The instruments of modified heat blocks of this invention will increase the speed and reliability of in situ PCR performed on microscope slides by 2 5 accelerating and rendering more uniform the heat transfer which occurs during thermal cycling.
To promote understanding of the invention, definitions are provided be-low for the following terms.
"PCR" refers to a process of amplifying one or more specific nucleic 3 0 acid sequences, wherein (1) oligonucleotide primers which determine the ends of the sequences to be amplified are annealed to single-stranded nucleic acids in a test sample, (2) a nucleic acid polymerase extends the 3' ends of the annealed primers to create a nucleic acid strand complementary in se-quence to the nucleic acid to which the primers were annealed, (3) the resul-3 S ting double-stranded nucleic acid is denatured to yield two single-stranded nucleic acids and (4) the processes of primer annealing, primer extension and product denaturation are repeated enough times to generate easily iden--s-2~D'~~2'~~
tified and measured amounts of the sequences defined by the primers.
Practical control of the sequential annealing, extension and denaturation steps is exerted by varying the temperature of the reaction container, normally in a repeating cyclical manner. Annealing and extension occur optimally in the 40° to 80°C temperature range (exact value depending on primer concentrations and sequences), whereas denaturation requires tem-peratures in the 80° to 100°C range (exact value depending on target sequence and concentration).
Such "thermal cycling" commonly is automated by a "thermal cycler"
1 0 an instrument which rapidly (on the time scale of one to several minutes) heats and cools a "sample compartment," a partly or completely enclosed container holding the vessel in which nucleic acid amplification occurs and the heat-transfer medium directly contacting the PCR vessel. Moat com-monly the sample compartment is a "sample block," normally manufactu-1 S red out of metal, preferably aluminum. Conventional sample blocks contain wells designed to fit tightly the plastic microcentrifuge tubes in which PCR
amplification normally is performed. The sample block of the present inven-tion replaces some or all of these conical wells with flat surfaces or slots de-signed to optimize heating and cooling of microscope slides. Less commonly, 2 0 the sample compartment is a chamber through which a hot or cold heat-transfer fluid, such as air or water, moves past reaction tubes bathed by the fluid.
"PCR reagents" refers to the chemicals, apart from test sample nucleic acid, needed to make nucleic acid amplification work. They consist of five 2 5 classes of components: (1) an aqueous buffer, (2) a water-soluble magnesium salt, (3) at least four deoxyribonucleoside triphosphates (dlVTPs), (4) oligo-nucleotide primers (normally two for each target sequence, with sequences which define the 5' ends of the two complementary strands of the double-stranded target sequence), and (5) a polynucleotide polymerase, preferably a 3 0 DNA polymerase, most preferably a thermostable DNA polymerase, which can tolerate temperatures between 90° and 100°C for a total elapsed time of at least 10 minutes without losing more than about half of its activity.
The four conventional dNTPa are thymidine triphosphate (dTTP), de-oxyadenosine triphosphate (dATP), deoxycytidine triphosphate (dCTP) and 3 5 deoxyguanosine triphosphate (dGTP). They can be augmented or sometimes replaced by dNTPs containing base analogues which Watson-Crick base-pair like the conventional four bases. Examples of such analogues include de-~~ _ 20 74 2 1 4 oxyuridine triphoaphate (dUTP) and dUTP carrying molecular tags such as biotin and digoxigenin, covalently attached to the uracil base via spacer arms.
Whereas a "complete set" of PCR reagents refers to the entire combina-S tion of essential reactants except test sample nucleic acid, a "subset" of PCR
reagents lacks at least one of the essential reagents other than the aqueous buffer. The "complement" or "complementary subset" to a first PCR reagent subset consists of all reagents missing from the first subset. PCR
"reactants" refers to the PCR reagents plus test sample nucleic acid.
1 0 "Hot StartTM PCR" refers to PCR amplification in which a subset of re-agents is kept separate from its complement and the test sample until the latter components have been heated to a temperature between about 50°
and about 80°C, hot enough to minimize nonspecific polymerase activity.
After all PCR reactants have been mixed, thermal cycling is begun, with reaction 1 5 temperature controlled so that it never drops below about 50°C
until amplifi-cation is completed.
"Fixed cells" refers to a sample of biological cells which has been che-mically treated to strengthen cellular structures, particularly membranes, against disruption by solvent changes, temperature changes, mechanical 2 0 stresses and drying. Cells may be fixed either in suspension or while contai-ned in a sample of tissue, such as might be obtained during autopsy, biopsy or surgery. Cell fixatives generally are chemicals which crosslink the pro-tein constituents of cellular structures, most commonly by reacting with pro-tein amino groups. Preferred fixatives are buffered formalin, 95% ethanol, ~ 5 formaldeh~~de, paraformaldehyde or glutaraldehyde. Fixed cells also may be treated with proteinases, enzymes which digest proteins, or with surfactants or organic solvents which dissolve membrane lipids, in order to increase the permeability of fixed cell membranes to PCR reagents. Such treatments must follow fixation to assure that membrane structures do not completely 3 0 fall apart when the lipids are removed or the proteins are partially cleaved.
Protease treatment is preferred following fixation for more than one hour and is less preferred following shorter fixation intervals. For example, a ten-minute fixation in buffered formalin, without protease treatment, is stan-dard after suspended cells (e.g. from blood) have been deposited centrifugally 3 5 on a slide by cytospin procedures standard in the cytochemical art.
"Histochemical section" refers to a solid sample of biological tissue which has been frozen or chemically fixed and hardened by embedding in a - s - ~~"~~~~.4 wax or a plastic, sliced into a thin sheet, generally several microns thick, and attached to a microscope slide.
"Cytochemical smear" refers to a suspension of cells, such as blood cells, which has been chemically fixed and attached to a microscope slide.
"In situ PCR" refers to PCR amplification performed in fixed cells, such that specific amplified nucleic acid is substantially contained within the cell or subcellular structure which originally contained the target nucleic acid sequence subjected to specific amplification. The cells may be in aqueous suspension or may be part of a histochemical section or cytochemi-1 0 cal smear. Preferably the cells will have been rendered permeable to PCR
reagents by proteinase digestion or by lipid extraction with surfactant or or-ganic solvent. An "in situ PCR preparation' consists of a combination of fi-xed cells with a subset or complete set of PCR reagents.
"Vapor barrier" refers to an organic material, in which water is inso-1 5 luble, which covers a PCR reaction or preparation in a way which substanti-ally reduces water loss to the atmosphere during thermal cycling. Preferred vapor barrier materials are liquid hydrocarbons such as mineral oil or pa-raffin oil, although some synthetic organic polymers, such as fluorocarbons and silicon rubber, also may serve as effective PCR vapor barriers. Waxes 2 0 which are solid at temperatures below about 50°C and liquid at higher tem-peratures also make convenient vapor barriers. The vapor barrier may be a thin plastic film, fabricated into an envelope completely enclosing the in situ PCR preparation or glued to a microscope slide carrying an in situ PCR pre-paration in such a way as to isolate the reaction from the atmosphere.
2 5 "Single-stranded DNA binding protein" (SSB) refers to a polypeptide which binds to single-stranded DNA more tightly than to double-stranded DNA. Naturally occurring SSBs include the bacteriophage T4 gene 32 pro-tein, the filamentous bacteriophage gene 5 protein, the SSB from E. coli with a subunit molecular weight of 19 kilodaltons, the 30 kilodalton movement 3 0 proteins of tobamoviruses and Agrobacterium tumefaciens vir E2 protein.
"Detection" of PCR-amplified nucleic acid refers to the process of obser-ving, locating or quantitating an analytical signal which is inferred to be specifically associated with the product of PCR amplification, as distingui-shed from PCR reactants. The analytical signal can result from visible or 3 5 ultraviolet absorbance or fluorescence, chemiluminescence or the photogra-phic or sutoradiographic image of absorbance, fluorescence, chemilumi-_ 9 _ 2~"~4~9~.4 nescence or ionizing radiation. Detection of in situ PCR products involves microscopic observation or recording of such signals. The signal derives directly or indirectly from a molecular "tag" attached to a PCR primer or dNTP or to a nucleic acid probe, which tag may be a radioactive atom, a chromophore, a fluorophore, a chemiluminescent reagent, an enzyme ca-pable of generating a colored, fluorescent, or chemiluminescent product or a binding moiety capable of reaction with another molecule or particle which directly carries or catalytically generates the analytical signal. Common binding moieties are biotin, which binds tightly to streptavidin or avidin, di-goxigenin, which binds tightly to anti-digoxigenin antibodies, and fluo-rescein, which binds tightly to anti-fluorescein antibodies. The avidin, strep-tavidin and antibodies are easily attached to chromophores, fluorophores, radioactive atoms and enzymes capable of generating colored, fluorescent or chemiluminescent signals.
"Nucleic acid probe" refers to an oligonucleotide or polynucleotide con-taining a sequence complementary to part or all of the PCR target sequence, also containing a tag which can be used to locate cells in an in situ PCR pre-paration which retains the tag after mixing with nucleic acid probe under solvent and temperature conditions which promote probe annealing to speci-2 0 fically amplified nucleic acid.
A preferred mode of fixing cell samples for in situ PCR according to the present invention is to incubate them in 10°!o fornaalin, 0.1 M Na phosphate, pH 7.0, for a period of 10 minutes to 24 hours at room temperature. The cells may be a suspension, as would be obtained from blood or a blood fraction 2 5 such as buffy coat, or may be a solid tissue, as would be obtained from biopsy, autopsy or surgical procedures well known in the art of clinical pathology. If PCR is to be performed in cell suspension, suspended cells preferably are centrifuged after forrmalin fixation, resuspended in phosphate-buffered sa-line and re-centrifuged to remove the fixative. The washed, pelleted cells 3 0 may be resuspended in PCR buffer and added directly to a PCR tube. If PCR
is to be performed on a microscope slide, suspended cells preferably are de-posited on the slide by cytospin, fixed 10 minutes in buffered formalin, was-hed 1 minute in water and washed 1 minute in 95% ethanol. Alternatively, suspended cells can be pelleted in a centrifuge tube and the pellet can be em-3 5 bedded in paraffn and treated like a tissue specimen. Tissue samples may be processed further and then embedded in para~n and reduced to serial 4-5 ~m sections by microtome procedures standard in the art of clinical patho-logy. Histochemical sections are placed directly on a microscope slide. In -10 - ~0'~4~1.4 either case, the slide preferably will have been treated with 2% 3-aminopro-pyltriethoxysilane in acetone and air dried. After smears or sections have been applied to slides, the slides are heated at about 60°C for about 1 hour.
Paraffinembedded sections can be deparaffinized by 2 serial 5 minute washes in xylene and 2 serial 5 minute washes in 100% ethanol, all washes occurring at room temperature with gentle agitation.
Choosing PCR primer sequences, preparing PCR reagents and reaction mixtures and designing and rurming PCR reactions are well known proce-dures in the PCR art. In the event that nucleic acid amplification is perfor-med on suspended cells in a standard PCR tube, the cells are treated like any conventional PCR test sample: diluted into reaction mixture shortly before amplification is started, at a total cell number ranging from approximately 100 to approximately 106. For carrying out the first aspect of the invention in a reaction tube, the only change from conventional PCR practice is that at 1 5 least one reagent, preferably enzyme but quite possibly primers, dNTPs or MgCl2, is omitted from the reaction mixture. After 50 to 100 N.1 of mineral oil have been added to the reaction tube, the tube is placed in a thermal cycler, many versions of which are commercially available, and heated to a tempe-rature between about 50° and about 80°C, preferably between 70° and 80°C.
2 0 While the tube is held at that temperature, the missing reagent is delivered beneath the vapor barrier with a standard manual sampler, preferably in a 5 to 15 N,1 volume of PCR buffer. If multiple samples are amplified simulta-neously in different tubes, a fresh sampler tip is used to add the missing re-agents) to each tube, to prevent cross-contamination. After all tubes have 2 S been prepared and capped, the standard three-temperature thermal cycle program of denaturation, annealing and extension for approximately 10 to 40 cycles is performed under thermal cycler microprocessor control. Alter-natively, and often preferably, a series of two-temperature cycles can be run wherein annealing and extension are performed at a single temperature, 3 0 normally optimized for stringent annealing of primer to template. Because reaction rates may be somewhat retarded with cellular preparations as compared to cell-free nucleic ands, it may be necessary to increase the du-rations of the denaturation, anneal, extend or anneal-extend cycle segments as much as several-fold from values standard when the test sample contains 3 5 cell-free nucleic acid. This adjustment easily is performed by trial and error, looking for conditions which maximize the intensity of the signal seen du-ring amplified nucleic acid detection or which minimize the number of cycles needed to reach a given signal intensity. A similar optimization procedure can be used for MgCl2, dNTP, primer and enzyme concentrations 2~?'~4~14 in the reaction mixture; these parameters of~,en show different optima for different targets.
For carrying out the second aspect of the invention in a reaction tube, several changes from the preferred mode just described are needed. There is no need to carry out the manual Hot StartTM procedure described above; all reactants can be mixed and the vapor barrier added at room temperature be-fore thermal cycling is started.
However, it is important that reagents be mixed in an order such that the SSB not be added last. Preferably, SSB will be pre-mixed with primers and the two reagents added together to the remaining reactants. The optimal quantity of SSB will vary with the identity of the SSB and with the quantity of single-stranded DNA in each reaction, and can be determined by trial and error according to the criteria given above for optimizing thermal cycle pa-rameters. Because cellular preparations normally should contain little 1 5 single-stranded DNA, the amount of primers in a reaction permits appro-ximation of the optimal quantity of SSB, in the following way: (1) calculate the total moles of all primers from the number of primers used, their con-centrations, and the reaction volume; (2) divide the average primer length by the lmown value for the footprint of the particular SSB used and round off to 2 0 the nearest integer; (3) multiply this integer times the total moles of primer to get the total moles of SSB needed to react with all of the primer and (4) add an amount of SSB equal to 0.5 to 2 times the calculated minimal amount of SSB. Further adjustment can be done by trial and error. The SSB footprint is the number of nucleotides occupying one SSB binding site. The following ap-t 5 proximate SSB footprints have been reported in the research literature: 8 nucleotides for bacteriophage T4 gene 32 protein; between 33 and 65 nucleoti-des per E. coli SSB tetramer; 5 nucleotides per filamentous phage gene 5 pro-tein monomer; 4 to 7 nucleotides per 30 kilodalton tobamovirus movement protein monomer; 30 nucleotides per 64 kilodalton Agrobacterium tumefa-3 0 ciens vir E2 protein monomer.
Whether the first or second aspect of the invention is applied to fixed cells suspended in a standard PCR tube, the preferred post-PCR treatment required for microscopic analysis is the same. The cells are pelleted and washed once in phosphate-buffered saline before deposition on organosilane-3 5 treated slides as described above, either as a smear or as a microtome sec-tion through a paraiBn-embedded pellet. Heating at 60°C for one hour im-proves cell adherence to the slide.

-12 - ~4~~~- _4 In the event that the first or second aspect of the invention is applied to histochemical sections or cytochemical smears attached to microscope sli-des, amplification procedures differ somewhat from those described above for reactions in PCR tubes. A preferred mode of effecting the first aspect of the invention on microscope slides is to cover the section or smear with ap-proximately 5 to 10 ~tl of a PCR reagent mixture lacking at least one reagent, such as enzyme. Then a plastic cover slip is placed over this preparation, the microscope slide is placed inside an aluminum foil boat, about 5 to 10 mm deep, the bottom of which is slightly larger than the slide, and the boat is placed on a metal thermal cycler sample block. After the sample block is brought to about 80°C and held at that temperature, the cover slip is lifted and 2 to 10 E.tl of PCR buffer containing the missing reagents) are distributed across the surface of the reagent mixture. The cover slip is replaced before the in situ PCR preparation dries out, a drop of nail polish is applied to one 1 5 corner of the cover slip to anchor it to the slide, and the slide is covered with enough mineral oil to assure that all cover slips, including their edges, are protected from the atmosphere. Preferably the oil has been pre-heated, so that its addition does not transiently reduce the temperature of the in situ PCR preparation. Then a standard two-temperature or three-temperature 2 0 thermal cycle is run for about 40 cycles. As above, cycle parameters, number of cycles and PCR reagent concentrations may need optimization to compen-sate for the abnormal heating and cooling kinetics of the oil-covered micro-scope slide and for possible reaction rate changes caused by the cellular na-ture of the test sample. After amplification, the mineral oil is removed from 2 5 the slide with an organic solvent such as xylene, and the slides are dried with 100°!o ethanol or a graded series of ethanol concentrations. The oilfree preparation is incubated for approximately 15 minutes at about 50°C in 0.15 M NaCl, 0.015 M Na citrate, pH 7.0 to remove unreacted PCR reagents. This step is most useful if primers or dNTPs have been tagged.
3 0 A preferred mode of effecting the second aspect of the invention on microscope slides is to follow the procedure recommended above for the first aspect, with a few changes. The manual Hot Starts method is not neces-sary, although it can still be used. A quantity of an SSB is added to the re-agent mixture which is estimated to suffice to bind all of the single-stranded 3 5 DNA in the in situ PCR preparation. This estimation is performed as de-scribed above for in situ PCR performed in reaction tubes. As before, it is im-portant that the SSB not be the last reagent added; preferably it is pre-mixed with primers, the major form of single-stranded DNA. If the manual Hot StartTM method is not used, the entire preparation is assembled, covered 2Q'~4~~.4 with a plastic cover slip (which is anchored to the microscope slide with a drop of nail polish), placed in the foil boat, and covered with mineral oil at room temperature; a normal aeries of thermal cycles is used, without hol-ding initially at 70° to 80°C. Post-PCR oil removal and preparation drying are as above.
The detection phase of in situ PCR is performed the same way, whether following the first or the second aspect of the invention and whether PCR is performed in a reaction tube or on a microscope slide. There are two basic detection strategies. The first strategy involves tagging either the PCR pri-mers or at least one of the dNTPs with a radioisotope or with a binding moiety such as biotin, digo~genin or fluorescein or with another lluoro-phore. In this case, tag incorporated into amplified nucleic acid can be ana-lyzed directly, provided that the unreacted tagged reagent has been washed out post-PCR and provided that the washing and drying procedure has not 1 5 mobilized the amplified nucleic acid from its point of synthesis. The analyti-cal validity of this simple detection strategy requires that the invention has increased in situ PCR specificity sufficiently that negligible nonspecific pro-ducts have been made which are large enough to resist washing from the preparation. To test and validate this consequence of the first two aspects of 2 0 the invention, appropriate control reactions can be performed. The logically most compelling control reaction is to perform the procedure on cells known to lack the target sequence; validation of the simplified detection strategy re-quires that no signal be generated in the control cells. Often such control cells are present in a histochemical or cytochemical preparation, so that the 2 5 standard analysis contains its own control. A less compelling control is to use primers which differ sufficiently from the optimal primers for the target sequence that they will not amplify the target sequence under the specified annealing and extension conditions.
The second strategy involves detecting amplified nucleic acid by in situ 3 0 hybridization to a tagged nucleic acid probe: an oligonucleotide or poly-nucleotide with a sequence complementary to at least part of the amplified nucleic acid sequences (preferably excluding the primer sequences). In situ hybridization, well known in the hiatochemical and cytochemical art, has four basic steps: denaturation of DNA in the test sample, annealing of probe 3 S to teat sample nucleic acid under stringent conditions, wash of the micro-scope slide with a solvent under stringent conditions to remove unhybridized probe and detection of the probe which has been retained on the slide.

20,~2~ 4 Regardless of which detection strategy is used, the methods for obser-ving and recording the presence and location of tag on the microscope slide are the same. If the tag is a radioisotope (preferably a strong beta radiation-emitter, such as 32P or 125I), the microscope slide is coated with nuclear S track emulsion such as NTB-2~from Eastman Kodak Co (Rochester, 1V~, in-cubated at 4°C for an interval determined by trial and error and developed by standard methods to leave microscopically detectable silver grains in the vi-cinity of immobilized tags. Procedures for 1251 tagging probe or PCR product are described by Haase et al., supra. If the tag is a fluorophore, it may be ob-1 0 served directly in a fluorescence microscope with excitation and emission fil-ters optimized for the particular fluorophore. This detection method is parti-cularly suitable for multiplex in situ PCR with different primer pairs for dif ferent target nucleic acid sequences. Either different fluorophores can be at-tached to primers of different specificity or different fluorophores can be at-1 5 tacked to probes of different specificity. Methods of attaching fluorophores to oligonucleotides and polynucleotides, preferably at their 5' ends, are well known in the nucleic acid chemistry and PCR arts. If the tag is a binding moiety such as biotin or digoxigenin, it is incorporated directly into PCR
product (via primers or dNTPs) or into probes by essentially the same me-t 0 thoda used to attach other tags. However, in this case, signal generation re-quires additional detection steps. Preferably, the microscope slide is incuba-ted in buffered aqueous solvent containing a covalent conjugate of a detection enzyme and a binding protein specific for the tag (avidin or stregtavidin for biotin, an anti-digoxigenin antibody for digoxigenin, an anti-fluorescein an-t 5 tibody for fluorescein). The preferred detection enzyme is horseradish pero-xidase or alkaline phosphatase. After unbound enzyme conjugate is remo-ved by washing in a buffered aqueous solvent, the microscope slide is im-mersed in a solution containing a chromogenic substrate for the enzyme used. After an insoluble dye, product of the enzyme reaction, has been depo-3 0 sited at points on the microscope slide where enzyme conjugate has been bound, unreacted substrate is washed away in water or buffered aqueous solvent to prevent the buildup of nonspecific background stain over time. The preferred chromogenic substrates which generate insoluble products are well known in the histochemical and cytochemical art, as are the methods 3 S for staining and for enzyme conjugate incubation and washing. The sub-strates and enzyme conjugates are commercially available from a wide va-riety of sources well known to histochemists and cytochemists.
A preferred companion procedure in the detection steps of the present invention is counterstaining of the microscope slide with fluorescent dyes Trademark -15 - ~0'~4~~.4 (for fluorescent tags) or chromophoric dyes (for radio-autoradiographic de-tection or enzymatic generation of insoluble chromophores) which emit or absorb with different spectral characteristics than the analyte-specific si-gnals and which highlight cell structures, especially in cells which lack target nucleic acid sequence. Especially preferred for examination of insolu-ble blue dye deposits by transmission microscopy is counterstaining by nuclear fast red, standard in the histochemical and cytochemical art. The methods for examining stained in situ PCR preparations by transmission or fluorescence microscopy are well known in the histochemical and cytoche-mical art, as are methods of recording permanently the microscopic image photographically or via digitized video images.
When the first or second aspect of the invention has been applied to fi-xed cells suspended in a PCR tube, an alternative detection mode to attach-ment to a slide for microscopic examination is direct flow cytometry of the 1 5 suspended cells. Flow cytometry is best adapted to fluorescent signals, whether incorporated into amplified nucleic acid during in situ PCR or attached to amplified nucleic acid by probe hybridization post-PCR. In either case, it is important that the cells be washed by sedimentation and resus-pension in tag-free buffered aqueous solvent to assure that tag not associated 2 0 with amplified nucleic acid is completely removed. Flow cytometric me-thods, well known to cell biologists, are useful primarily for counting the proportions of cells containing and lacking tag, although they also can re-cord the quantitative distribution of tag among cells.
The preferred mode of effecting the sample block and instrument 2 5 aspects of the invention is to modify the manufacturing procedure for con-ventional thermal cycler sample blocks to change just the top surface so that it is optimized for heat flow to and from microscope slides. Two very distinct designs are provided. One, for in situ PCR applications where very few slides are to be run simultaneously, the top surface is designed to create flat hori-3 0 zontal areas large enough to hold slides so that the large dimensions (height and width) are horizontal. These flat areas may be recessed in shallow wells which hold a mineral oil vapor barrier covering the slides. The areas must be at least about 16 mm wide and 77 mm long to fit conventional glass microscope slides. The wells must be at least about 2 mm deep to fit a slide 3 5 plus coverslip plus vapor barrier. This design is compatible with either the first or second aspect of the invention.

- is - 2~'~~2~
Too, for in situ PCR applications where many slides are to be run si-multaneously, the block is designed to contain many narrow, deep, vertical or approximately vertical slots, sized to hold slides inserted edgewise with minimal space separating the slide from metal surfaces facing its top and bottom surfaces. The intervening space normally will be filled with mineral oil or another nonvolatile liquid to provide a vapor barrier and efficient heat transfer during thermal cycling. The plane of a slot may be inclined from the vertical by as much as about 45° in order to use the force of gravity to assure that one surface of the slide touches the metal of the sample block.
1 0 Slots must be about 15 mm deep, at least ?7 mm long, and at least 2 mm wide to fit a conventional slide plus a cover slip. This design is compatible with the second aspect of the invention but is not preferred with the first because it blocks rapid access to the in situ PCR preparation for cover slip removal, manual addition of the missing PCR reagents) and cover slip replacement.
Many different thermal cyclers are commercially available, each with distinct sample block design. However, these sample block designs can be described in terms of several general features: (a) composition: practically all are made of metal, preferably aluminum, to promote durability and rapid heat transfer; (b) shape and overall dimensions of length, width and thick-2 0 ness; (c) bottom and occasionally aide surfaces designed to integrate with the heating and cooling mechanisms which determine block temperature when the thermal cycler is operating; (d) a top surface containing many wells di-mensioned to hold tightly the small plastic microcentrifuge tubes, preferably of about 0.5 ml capacity but occasionally holding about 1.5 ml, which are 2 5 commonly used to hold nucleic acid amplification reactions; (e) occasionally one or a few small wells in one surface designed to hold tightly a thermocou-ple or thermistor probe which feeds back the sample block temperature to the thermal cycler control circuitry.
A preferred mode of realizing the third aspect of the invention is to 3 0 change only the top surface of the sample block, leaving the other design fea-tures (except possibly block thickness) substantially unchanged in order to minimize the impact of the invention on thermal cycler manufacture and performance. Also preferred is to render the sample block of the invention equal in mass to the conventional sample block of the thermal cycler in 3 5 question, to minimize impact on heating and cooling kinetics.
Thermal cycler sample blocks most commonly are manufactured by machining into a single metal block, for example with a rotary mill, exact 24~'~~~1~
dimensions, wells and other contours needed to integrate with the rest of the thermal cycler. Holes for bolting the block to the rest of the thermal cycler may be made with a drill press. The same manufacturing procedures are suitable for the sample block of the present invention. However, the rectili-near shape of wells adapted to fit microscope slides tightly is also easily pro-duced by stamping or machining (including laser and water jet cutting) of relatively thin sheets of metal which are bolted together to create a laminated assembly. The entire block may be laminated; or just the top portion, holding the microscope slide wells, can be laminated and bolted to a solid bottom por-tion which contains the features of the block which integrate with the rest of the thermal cycler.
Also preferred for the third aspect of the invention is a thermal cycler sample block design which includes both wells optimized for microscope sli-des and wells designed to hold conventional nucleic acid amplification reac-t 5 tion tubes. Preferably the reaction tube wells will occupy one or several rows along the edges of the sample block, reserving the central area of the sample block for microscope slide wells. This mode also is best realized by leaving the other sample block features, including mass of metal, unchanged.
Manufacture is most simply performed by machining, because of the 2 0 cylindrical symmetry of reaction tube wells.
A few commercially available thermal cyclers and published thermal cycler designs avoid metal sample blocks and immerse conventional PCR
tubes in a rapidly moving stream of hot or cold sir, water, or other heat-transfer fluid. Such designs are easily adapted to microscope slides by repla-2 S sing the tube holders with a metal wire or plastic lattice which holds slides firmly in the stream of heat-transfer fluid. Preferably, the slides are oriented so that their smallest dimension (thickness) faces the fluid flow and the do-minant fluid flow vector lies in a plane which parallels the plane of their larger dimensions (width and length). Slight canting of the microscope 3 0 slides to the dominant fluid flow vector can create mild turbulence which helps to ensure uniform heat transfer.
In the event that the thermal cycler contacts microscope slides directly with a moving heat transfer fluid, it is necessary to isolate the slides from the heat-transfer fluid by a thin barrier which blocks material transfer 3 5 between the in situ PCR preparation and the heat-transfer fluid. Otherwise the preparation may be desiccated or experience wash-out of PCR reagents.
Preferred barriers are envelopes of a thin, waterimpermeant plastic with 2O'~~4~~.4 - is -high thermal conductivity, such as a fluorocarbon, a polyurethane, a polyolefin, a polyimide or a polyamide. The envelopes must be sealed in a way which prevents leakage of fluid or water vapor into or out of them.
Either a water-resistant adhesive or a tight clip may serve adequately to seal the envelope. If the heat-transfer fluid is a liquid, one edge of the envelope may project above the liquid into the vapor space over it. As an alternative to an envelope, the vapor barrier may comprise a thin sheet of plastic with ap-proximately the length and width of a microscope slide, carrying hot-water-resistant adhesive applied in a narrow strip around all edges on one face.
1 0 The sheet is pressed tightly to the top face of the microscope slide before thermal cycling is started and can be peeled off afterward for detection pro-cessing. It may even replace the coverslip.
From the above description and the following examples, one of ordinary skill in the art can appreciate the many diverse aspects of the present inven-t 5 tion as encompassed by the following claims.
Cells of the stable human cervical cancer cell line, SiHa (ATCC HTB
2 0 35), containing one integrated copy of human papilloma virus (HPV) type 16 genome per human genome, were grown to density of about 105 cells/mL in Eagle's minimal essential medium with non-essential amino acids, sodium pyruvate, and 15°lo fetal bovine serum, washed two times in Tris-buffered sa-line, adjusted to an approximate density of 104 cellsJmL, and stirred over-t 5 night at room temperature in 109'0 (vol/vol) formaldehyde in phosphate buf fer. The formaldehyde-fixed cells were centrifuged at 2,000 rpm for 3 minu-tes, and the pellet was embedded in para~n. Microtome sections (4 ~.m thickness) of the para~n block were attached to glass microscope slides which had been dipped in 29'0 3-aminopropyltriethoxysilane (Aldrich 3 0 Chemical Co.) in acetone by floating the sections in a water bath. After at-tachment, sections were depara~nized and proteolytically digested with re-agents from the Viratype~ in situ Tissue Hybridization Kit (Life Technolo-gies, Inc., Gaithersburg, MD) following the manufacturer's instructions.
(The equivalent reagents and methods of the Oncor S6800kit, Oncor, Inc., 3 5 Gaithersburg, MD, could have been used instead). Slides were placed in hand-made aluminum foil boats, approximate dimensions of 8 x 3 x 1 mm;
and each set of four sections (per slide) was overlaid with 5 to 10 E,~l of PCft 2~'~4~1.4~

solution (see below). A plastic coverslip then was placed over each four section in situ PCR preparation. For conventional in situ PCR, the coverslip was anchored to the slide with a drop of nail polish, the slide was covered with approximately 1 ml of mineral oil, the foil boat was laid on top of the aluminum sample block of a PCR thermal cycler and thermal cycling was started. For manual Hot StartTM in situ PCR, the boat containing a slide (with coverslip) was heated to 82°C and held at that thermal cycler tempera-ture while the coverslip was lifted, 2 N.1 of the missing PCR reagents (see be-low) were distributed over the surface of the preparation, the coverslip was 1 0 replaced and attached to the slide with a drop of nail polish and approxi-mately 1 ml of mineral oil pre-heated to 82°C was laid over the slide and co-verslip in the boat. Then the normal thermal program was resumed.
The pH 8.3 PCR solution contained 10 mM TrisCl, 50 mM KCl, 4.5 mM
MgCl2, 20 mM of each dNTP, 0.2 unit/~.L of AmpliTaq~ DNA polymerase 1 5 and 6 ~M of each primer. For "single primer pair" experiments, the primers were PV1 and PV2, dictating a 449 by product from the HPV type 16 genome.
For "multiple primer pair" experiments, primers PVl to PV7, dictating a series of overlapping approximately 450 by PCR products covering a total se-quence length of 1247 bp, were used. All primer sequences are given in the 2 0 Table below.
Position of First PV1 (5') 1 110 5'-CAGGACCCACAGGAGCGACC

PV2 (3') 2 559 5'-TTACAGCTGGGTTTCTCTAC

2 5 PV3 (5') 3 501 5'-CCGGTCGATGTATGTCTTGT

PV4 (3') 4 956 5'-ATCCCCTGTTTTTTTTTCCA

PV5 (5') 5 89B 5'-GGTACGGGATGTAATGGATG

PV6 (3') 6 1357 5'-CCACTTCCACCACTATACTG

PV7 (5') 7 1300 5'-AGGTAGAAGGGCGCCATGAG

3 0 For conventional in situ PCR, all of the components listed above were present in the PCR solution initially added to the histochemical sections.
For manual Hot Start.'' in situ PCR, the solution initially added to the sec-tions lacked primers and Taq polymerase. These reagents were added sepa-rately in 2 N.l of 10 mM TrisCl, 50 mM KCI, pH 8.3, after the slide had been 3 5 heated to 82°C. For the first thermal cycle, denaturation was performed for 3 minutes at 94°C, and annealing/extension was performed for 2 minutes at _~_ 2 0 7 4 2 1 4 55°C; the :remaining 39 cycles consisted of 1 minute denaturation at 94°C and 2 minutes annealing/extension.
After DNA amplification, mineral oil was removed by dipping in xy-lene, the cover slip was removed, and the mounted sections were dried in S 100% ethanol. Each slide was incubated with 10 ~.1 of a 500 ng/ml solution of biotinylated HPV type l6specific polynucleotide probe (Viratype Kit *Life Technologies, Inc.) in 0.03 M Na citrate, 0.30 M NaCI, pH 7:0, S% dextran sulfate, 50% formamide at 100°C for 5 minutes and then 37°C for 2 hours;
then the slide was treated with an alkaline phosphatase-streptavidin conju-1 0 gate and the phosphatase substrates, 5-bromo-4-chloro3-indolyl phosphate (BCIP) and vitro blue tetrazolium (NBT), according to the instructions of the supplier of the S6800 Staining Kit (Oncor, Gaithersburg, MD). After enzyma-tic detection of biotinylated probe captured on the sections, the sections were counterstained with nuclear fast red for 5 minutes. The following results 1 S were obtained in this experimental system, when the stained slides were examined by transmission microscopy under 40-400 X magnification. In conventional in situ PCR, single-copy HPV targets in SiHa cells were not de-tectable with a single primer pair but showed up clearly in most nuclei with multiple primer pairs. In manual Hot Starts in situ PCR, a single primer 2 0 pair stained about 80% of the cell nuclei more strongly than did multiple primer pairs in the conventional method. The other 20% may have been damaged during sectioning. The previously published conclusion that in situ PCR requires multiple primer pairs specifying overlapping targets is thus invalid. The practical, and in fact improved, performance of a single 2 5 primer pair greatly increases the utility of in situ PCR.
Example 2 In Situ PCR F~vbrid,'_zation Detection of HIV I Integrated into Human G enom,_' c DNA
The human T lymphocytic cell line, H9 (ATCC CRL 8543), was grown to 3 0 a density of about 106 cells/ml in complete RPMI medium, infected with HIV-1 as described in Basic Virological Techniques, pp. 66-69, and incubated for four days at room temperature. Approximately 104 cells from this incuba-tion were formaldehyde fixed, paraffin-embedded, sectioned (4 m thickness), mounted on glass slides and proteolytically permeabilized as in Examplel.
3 5 Conventional and manual Hot StartTM in situ PCR were performed as in Example 1 except that a single primer pair, SK38 and SK39 (Perkin Elmer Cetus Instruments, Norwalk, CT) specifying a 115 by target from the HIV-1 "~ Trademark -21- 2p74214 gag region, was used. Post-PCR processing was as in Example 1, except that the probe, SK19 (Perkin Elmer Cetus Instruments), was labeled with di-goxigeninll-dUTP using random primers, using the reagents and following the instructions of Boehringer Mannheim (Indianapolis, IN), manufacturer of the tagged dNTP and the GeniusTM labeling kit. Staining of the probed microscope slide was with an alkaline phosphatase-anti-digoxigenin conju-gate and BCIP/NBT chromogens, also as directed by Boehringer Mannheim.
Microscopic examination of the slide showed that about 90% of cell nuclei were BCIP/NBT stained after manual Hot StartTM in situ PCR; con-1 0 ventional in situ PCR yielded no stained nuclei. Even a 115 by product appears to be detectable nonisotopically by (and only by) the manual Hot Start's methodological improvement.
T_n Situ PCR Snecificitv Improv .
1 5 Rresul_ting from the Hot Start Met_ho Microscope slides carrying histochemical sections of embedded fixed SiHa cells were prepared as in Example 1. The sections were augmented with approximately 50 ~1 of human peripheral leukocytes (approximately 5,000 cells/ml) from an HPVnegative donor, prepared from buffy coat and 2 0 deposited on the slide by cytospin. The added cells were fixed for 5 minutes at room temperature in 10% formaldehyde in phosphate buffer. The slides were subjected to conventional or manual Hot StartTM in situ PCR as described in Example 1, except that the dIVTPs were augmented with 5 mM digoxigenin-11-dIJTP (Boehringer Mannheim). HPV primer pairs PV1 and PV2 were 2 5 used.
After DNA amplification, all digoxigenin-tagged DNA which was not removed during washing and dehydration was stained with an alkaline phosphatase-antidigoxigenin conjugate and phosphatase substrates, BCIP
and NBT, as recommended by Boehringer Mannheim, supplier of the stai-3 0 ring reagents, except that reagent volumes were scaled down approximately 95%a to accommodate histochemical sections rather than Southern blotting membranes. After staining of the amplified DNA, the leukocytes were im-munohistochemically stained by a pair of mouse monoclonal primary anti-bodies against leucocyte common antigen (DAKO-LCA, containing antibo-3 5 dies pD7/26 and 2BN; DAKO-PATTS) and a Histostain-SP kit for detecting mouse primary antibody (Zymed Laboratories Inc., South San Francisco, " Trademark -~- 207414 CA). This kit uses a biotinylated anti-mouse secondary antibody, horsera-dish peroxidasestreptavidin, and the chromogenic peroxidase substrate, aminoethylcarbazole. Both the primary antibodies and the staining lrit were used according to the manufacturer's instructions.
S Microscopic examination showed that manual Hot StartTM in situ PCR
stained about 80% of SiHa cell nuclei and no leucocyte nuclei, demonstrating the specificity and sensitivity of the Hot StartTM procedure, even with a single primer pair. In contrast, conventional in situ PCR was so nonspecific that all cells, both SiHa and leucocyte, were stained, indicated that considerable 1 0 non-target-directed amplification occurs when the Hot StartTM procedure is not used. Although target-specific probes can distinguish specific and non-specific amplified DNA after in situ hybridization (see Example 1), the pre-sent Example demonstrates that the Hot StartTM method can render probing unnecessary, greatly simplifying detection and thereby enhancing the prac-1 5 ticality of in situ PCR even more.
L

~uence Listing INFORMATION FOR SEQ ID NO: 1:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 bases (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: Other Nucleic Acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:

INFORMATION FOR SEQ ID NO: 2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 bases (B) TYPE: nucleic acid 1 5 (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: Other Nucleic Acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:

2 O INFORMATION FOR SEQ ID NO: 3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 bases (B) TYPE: nucleic acid (C) STRANDEDNESS: single 2 5 (D) TOPOLOGY: linear (ii) MOLECULE TYPE: Other Nucleic Acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3:

INFORMATION FOR SEQ ID NO: 4:

3 O (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 bases (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear 3 S (ii) MOLECULE TYPE: Other Nucleic Acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

INFORMATION FOR SEQ ID N0: 5:

(i) SEQUENCE CHARACTERISTICS:

4 O (A) LENGTH: 20 bases (B) TYPE: nucleic acid ~~'~~~~.4 (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: Other Nucleic Acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5:

INFORMATION FOR SEQ ID NO: 6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 bases (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: Other Nucleic Acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6:

I S INFORMATION FOR SEQ ID NO: 7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 bases (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: Other Nucleic Acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7:

Claims (16)

1. A process for in situ PCR amplification of a target nucleic acid sequence, which comprises (a) incubating an in situ PCR preparation comprising fixed cells obtained after chemically treating with a fixative which crosslinks the protein constituents of cellular substructures, a first subset of PCR reagents comprising a portion of the PCR reagents necessary to allow PCR to proceed at a temperature between about 50 ° and about 80 °C;
(b) adding to the preparation of step (a) a PCR reagent subset which complements said first PCR reagent subset and is necessary to allow PCR to proceed; and (c) subjecting the preparation of step (b) to thermal cycling sufficient to amplify the target nucleic acid sequence specified by the complete set of PCR reagents.

2. The process of Claim 1, wherein said fixed cells are attached to a microscope slide following step (c).

3 . The process of Claim 1, wherein (1) said fixed cells are attached to a microscope slide and said preparation is covered with a cover slip before the incubation of step (a), (2) the cover slip is lifted for an interval of between about 5 seconds and about 15 seconds in order to perform step (b) and then is replaced before performing step (c) and (3) a vapour barrier is placed over the cover slip before performing step (c).

4. A process for in situ PCR amplification of a target nucleic acid sequence, which comprises (a) providing an in situ PCR preparation comprising fixed cells obtained after chemically treating with a fixative which crosslinks the protein constituents of cellular substructures, a complete set of PCR
reagents and a single-stranded DNA binding protein; and (b) subjecting the preparation of step (a) to thermal cycling sufficient to amplify the target nucleic acid sequence specified by the complete set of PCR reagents.

5. The process of Claim 4, wherein said fixed cells are attached to a microscope slide following step (b).

6. The process of Claim 4, wherein (1) said fixed cells are attached to a microscope slide, and (2) a vapour barrier is placed over said preparation before step (b).

7. The process of any one of Claims 1 to 6, wherein said fixed cells have been rendered permeable to PCR reagents.

8. The process of any one of Claims 1 to 6, wherein said fixed cells reside within a histochemical section or cytochemical smear.

9. The process of any one of Claims 1 to 3, wherein said first subset of PCR reagents consists of all PCR reagents except a nucleic acid polymerase.

10. The process of claim 1 comprising the further step of detecting the amplified nucleic acid sequence in a manner which locates it in the individual cells originally containing the target nucleic acid sequence.

11. The process of claim 4 comprising the further step of detecting the amplified nucleic acid sequence in a manner which locates it in the individual cells originally containing the target nucleic acid sequence.

12. The process of claim 1, wherein in step (a) the first set of PCR
reagents further comprises a single-stranded DNA binding protein.

13. The process of Claim 12, wherein said single-stranded DNA
binding protein is bacteriophage T4 gene 32 protein.

14. The process of Claim 12, wherein said single-stranded DNA
binding protein is the 19 kilodalton SSB from E. coli.

15. The process of Claim 3, wherein said vapour barrier is placed over the cover slip before performing step (c) of Claim 1 and said vapour barrier and cover slip are removed before performing the further step defined in Claim 10.

16. The process of Claim 6, wherein said vapour barrier is placed over said preparation before performing step (b) of Claim 4 and said vapour barrier is removed before performing the further step defined in Claim 11.