WO1987003911A1 - Displacement polynucleotide method and reagent complex - Google Patents

Abstract

Diagnostic assay methods for detecting the presence of a target nucleotide sequence (TNS) in a biological sample. A reagent complex for use in the first method is also provided which includes a labelled probe (LP) having a target binding region (TBR) and a second polynucleotide (SP) which binds to the probe in at least a portion of the target binding region. The complex is contacted with sample (G) and target displaces the second polynucleotide on the complex. Intact complex is then separated from probe/target hybrids, and the hybrids are detected. The second method involves capturing of the labeled strand after displacement from the probe which binds to the target.

Description

DISPLACEMENT POLYNUCLEOTIDE METHOD AND REAGENT COMPLEX

^*.

Background of the Invention

The present invention relates to a diagnostic assay method for detecting the presence of a target nucleotide sequence (either DNA or RNA) in a biological sample, a polynucleotide reagent complex therefor and diagnostic kits therefor.

Polynucleotide assays based upon strand displacement, reagent complexes therefor and kits for such assays including such reagent complexes have been described in European patent applications 167,238 and 164,876. Such assays employ a reagent complex of:

(a) a probe polynucleotide having a target binding region which is capable of complementary base pair binding to the target nucleotide sequence to be detected, and

(b) a labeled polynucleotide or signal strand bound in a reagent complex to the probe polynucleotide, generally by complementary base pair binding to a portion of the target binding region. The portion of the labeled polynucleotide so-bound is referred to as the pairing segment.

In the assay the target nucleotide sequence binds to the target binding region and displaces the labeled polynucleotide. The displaced labeled polynucleotide is then detected as a measure of the presence and amount of target nucleotide sequence in the sample. While in such systems, the probe and labeled polynucleotide are generally

* separate molecules, certain embodiments provide that they may be segments of a single polynucleotide strand. In any event, such assay methods all involve determining the

presence or amount of displaced labeled polynucleotide or labeled polynucleotide which is not displaced.

Two forms^'of separationof displaced labeledpolynucleo- tides from intact reagent complexes involve: (a) a reagent complex in which the probe polynucleotide is immobilized and (b) a reagent complex in which the probe polynucleotide is in solution with a pendant affinity moiety (e.g. biotin) for post-displacement immobilization. In the case of an immobilized probe or post-displacement immobilization of the probe, the displaced labeled polynucleotides remain in solution for detection. Usually, only a small proportion (1% to 0.00001%) of labeled polynucleotides are displaced. However, a variety of mechanisms such as strand breakage of reagent with detachment of immobilized reagent complexes from the solid support, or inefficient trapping of immobiliz- able complexes, can result in labeled polynucleotides remaining in solution in the absence of specific displacement events.

These mechanisms are a source of background for the assay. The level of tolerable background will depend upon both the number of reagent complexes used in the reaction and the amount of analyte to be detected. For example, to detect 10^s analytes in a reaction with 10¹¹ reagent complexes, 0.001% (10⁶) or fewer complexes must be released or not immobilized as background for any signal to be detected with a [signal plus background] background ratio of at least 2:1.

In EPA 167,238 and 164,876 embodiments are shown (e.g., in Figure 2A) wherein the labeled polynucleotide L bears an affinity moiety (biotin) for post displacement immobilization. While such technique concentrates the signal, it is unlikely to aid in background reduction, since displaced labeled polynucleotides and detached reagent complexes each have biotin and are likely to be immobilized with substantially equal efficiency. (See also, EPA 200 , 057 ) .

Brief Description of the Invention

The methods and reagents of the present invention involve diagnostic assays which may be used for detecting and determining the presence and/or concentration of an infectious agent or for distinguishing the strain, regulatory state or other state (including drug resistances) of an organism or vector. The methods of the invention are improved methods for carrying out strand displacement assays including the assays described in EPA 167,238 and 164,876.

In one aspect, the present invention provides a method ("Method I") for determining the presence of a predetermined target nucleotide sequence ("TNS") in a biological sample. This method includes the following steps: First, an inverse reagent complex ("IRC")is formed of (i) labeled probe polynucleotide ("labeled probe") having a target binding region ("TBR") which is capable of complementary base pair binding via hydrogen bonds of purine/pyrimidine bases to the target. Hereafter all references to binding shall mean complementary base pair binding unless otherwise indicated. Another portion of the complex is a second polynucleotide which is bound to the labeled probe in a region of the labeled probe at least partially coextensive with the TBR. Second, the IRC is contacted with a sample under conditions in which the target, if present, binds to the labeled probe and displaces second polynucleotide therefrom. Third, the labeled probes from which second polynucleotides have been displaced are separated from intact IRCs. Finally, the presence and/or amount of labeled probes which have been separated are determined.

In another aspect, the present invention provides a diagnostic reagent, the IRC, for use in Method I. The IRC includes the labeled probe and second polynucleotide as described above. The second polynucleotide can be immobilized or immobilizable. The potential binding between the target and the labeled probe is capable of displacing the second polynucleotide from the labeled probe. In yet another aspect, the present invention provides another diagnostic reagent, the inverse reagent complex construct ("IRCC") for use in Method I. The IRCC includes a single polynucleotide strand having (i) a TBR capable of binding to the target; and (ii) a pairing segment ("PS") bound to a portion of the TBR. Another part of the IRCC is a detectable tag which is within or adjacent to the TBR. In another aspect the present invention provides a method ("Method II") for determining the presence and/or amount of a predetermined target in a biological sample. First, a reagent complex is formed of a probe polynucleo¬ tide (hereafter "probe") which has a TBR and a labeled polynucleotide (hereafter "label strand") bound to the probe in a region of the probe at least partially coextensive with the TBR. In Method II, the reagent complex is then contacted with a sample and with a capture polynucleotide ("CP") under conditions in which the target, if present, binds to the probe displacing label strand from the reagent complex. Similarly, the CP binds selectively to the displaced label strand in a pairing segment ("PS"), i.e. the region thereof that had been bound to the probe. Displaced label strand which has bound to CP is then separated from unbound label strand. The captured displaced label strand is then detected.

In a desirable embodiment of Method II the probe polynucleotide and label strand of the reagent complex are parts of a single polynucleotide strand folded onto itself. This single strand may also include two or more stably joined polynucleotide strands, e.g., joined by covalent or non- covalent bonding stable to the conditions of hybridization and displacement. The label strand includes a PS with a tag and the probe includes a TBR. A hairpin loop may be present which divides the label strand portion from the probe portion of the single strand. [See Figs. 4A to 4D]. This reagent complex is referred to as the reagent complex construct (RCC) .

In another aspect the invention provides a kit for determining the presence and/or amount of target in a sample. The kit includes a reagent complex formed of the above-described probe and a label strand bound through its PS to a label strand binding region of the probe ("PS"'). The PS¹is at least partially coextensive with the TBR. A CP having a second binding region capable of binding selectively to a segment of displaced label strand substantially within the ps (hereafter "CS'")is also part of this kit. The third component of the it is a means for isolating the CP together with any attached displaced label strand from unbound label strand.Alternatively the kit contains an RCC and a detectable tag attached within or adjacent to the TBR.

In a further aspect, there is provided a method ("Method III") for use in diagnostic tests for disrupting the linkage between reagents which are attached by binding through the highly stable biotin-avidin or biotin- streptavidin bond. In Method III, the attached reagents are incubated in the presence of excess biotin at moderate temperatures ranging from 22-65 C. In addition. Method III can be used in a tetraethylammonium chloride ("TEAC1") salt solution. The lower temperature or the combination of lower temperature and of TEAC1 allows the use of Method III with heat labile substances such as enzymes, antibodies, antigens and other proteins in formats compatible with DNA hybridization. Additionally antibody-antigen binding may be used in Method III in the area of both DNA diagnostics and immunodiagnostics where enzymes ^'or other signals are attached to either DNA molecules, antigens or antibodies through an avidin-biotin linkage, or where reagents in a diagnostic test are attached to a support or surface through a biotin-avidin linkage. Additional uses include purification schemes utilizing binding to substances linked to a support through a biotin-avidin linkage. [See, e.g. Hofman et al, Biochem. 2.1:978-984 (1982)].

The displacement of Method III by dissolved biotin at moderate (22-65 degrees C) temperatures has significant and unexpected advantages over the prior art. T. Kempe et al. Nucleic Acids Res., 13:45-47 (1985) teaches treatment by near boiling (90 degree C) for five minutes. At these high temperatures melting of double-stranded DNA occurs and enzymes and other proteins denature. Method III allows the biotin-displacement of strands having both biotin and label under conditions suitable for releasing the immobilized complexes of CP and labeled polynucleotide or of CP and complex hybridized to target.

The methods and kits of the present invention are capable of reducing background label detection in displacement assays where the proportion of displaced label relative to intact reagent complexes is low and an amount of background label detection is attributable to detached reagent complexes, cleaved reagent complexes and/or intact reagent complexes not immobilized in a postdisplacement immobilization step. Such methods and kits are further suited to certain embodiments of strand displacement in solution in which the CP is the sole or primary means for separating displaced label strand from intact reagent complexes.

Brief Description of the Drawing's

Fig. 1 is an embodiment of an IRC (A) onto which sample polynucleotide G having target nucleotide sequence TNS has hybridized (B) r and ultimately completely displaces PS of second polynucleotide SP from labeled probe LP (C) .

Fig. 2 illustrates an IRCC (A) , which has been contacted by G (B) and which becomes bound by immobilized CP (C) .

Fig. 3 illustrates two immobilized reagent complexes on a support (A) ; sample G causing displacement of label strand LI (B) ; G bound to probe PI and LI in solution with immobilized CP having a capture segment CS (C) and finally LI immobilized on CP (D) .

Fig. 4 illustrates RCC (A) ; G binding to RCC and displacing PS (B) ; a biotinylated CP (C) , which hybridizes to RCC (D) and an intermediate structure of a hybrid between G and CP (E).

Fig. 5 illustrates a covalent complex p66d and CPs therefor.

Detailed Description of the Invention

In this application the following terms are given their generally accepted meanings in the field of molecular biology: polynucleotide or polynucleotide strand, complementary base pair binding, hybridization and other terms of art. [See also the definitions in EPA 167,238 and 164,876].

The following abbreviations used throughout the description of the invention are defined below: TNS is the target nucleotide sequence or analyte in a biological sample.

Labelled probe is the labelled polynucleotide probe employed in Method I.

Probe is the polynucleotide probe of Method II. TBR is the target binding region of the labelled probe of Method I and the probe of Method II. IBR is the initial binding region, which is a single stranded portion of the TBR to which TNS can bind without displacing the second polynucleotide of Method I or label strand of Method II.

Second polynucleotide refers to the polynucleotide strand of Method I which binds to the labelled probe.

Label strand is the labelled polynucleotide strand of

Method II which binds to the probe.

PS is the pairing segment portion on the second polynucleotide and label strand, whjich binds to the TBR on the labelled probe or probe, respectively.

PS¹ is the region of the labelled probe or probe usually located in the TBR, which binds to the PS.

CP is the capture polynucleotide strand of Method II.

CS is the region of CP capable of binding a portion of the second polynucleotide or label strand.

CS• is the portion of the second polynucleotide or label strand capable of binding to CS and located in the PS.

RCC refers to a reagent complex construct.

IRC refers to- an inverse reagent complex of Method I.

IRCC refers to an inverse reagent complex construct used in

Methods I and II.

The basic components of Method I are a labeled probe, a second polynucleotide and a biological sample containing nucleic acid, a portion of which is TNS. Other optional components of this method include a volume-excluding polymer, such as a polyether compound Additionally a support to which the reagent complex is immobilized via the second polynucleotide may be a component of the method. A support may also be employed as part of the separation step that follows displacement. In the practice of Method I, additional reagents or equipment may be required for "readout". "Readout" is the direct or indirect detection of labeled probe in one or more phases of separated reaction materials. Readout reagents may particularly be needed in the liquid phase by virtue of the displacement of the IRC and separation of labeled probe fromwhich second polynucleo¬ tide has been displaced from the solid phase containing immobilized displaced second polynucleotides and immobilized intact IRC.

The basic components of Method II include a CP, a probe, a labeled polynucleotide and a biological sample containing TNS. "Capturing" refers to hybridization of the displaced label strand to a CP which is at least partially complementary to the label strand in a region of the label strand substantially or completely within the PS that had been bound to the TBR in the reagent complex. PS is thus available for hybridization to the CP only when it has been displaced from the probe TBR.

A. The Probes of Method I and II.

The probe employed in the present invention may be a linear or circular polynucleotide (RNA or DNA) capable of binding in at least one region of its purine/pyrimidine base sequence to specific TNS of a sample. TNS which can be analyzed include viral or bacterial nucleic acid sequences, DNA or RNA sequences of plants, animals (including humans) or microorganisms (including plasmids) and rRNA and other non-protein coding sequences. Thus, binding may be between DNA and RNA, between DNA and DNA or between RNA and RNA. It is generally only a specific region of the probe which binds selectively to the TNS. Other regions of the probe may be various naturally occurring or synthesized sequences, which do not participate in the hybridization reaction with the target, but which may play an important role in the present invention, e.g., by serving as a site for attachment to a label or by providing some degree of separation between the label and the region to which the target binds, if desired. Sizes, types and geometries of probe, TBR, second polynucleotide and label strand may be as set forth in the published patent applications referred to herein.

The region of the probe to which TNS will specifically bind is the TBR. The binding at the TBR may be, and preferably is, perfect. Each nucleotide in the probe finds its correct complementary binding partner (e.g., dA to dT) in TNS. Alternatively, the hybridization may contain some mismatches, as described in the referenced published applications. At least one portion of the TBR is preferably single-stranded (i.e., it is not complementary to second polynucleotide sequences nor self-complementary) . This single-stranded region is called the initial binding region ("IBR") , because TNS can bind to this region of bases without displacing any of the second polynucleotide. The IBR of the probe is at least fifteen bases in length, and is preferably at least fifty bases in length.

The complete TBR includes the IBR and preferably all of the PS• , the region which binds to the second polynucleo¬ tide or label strand. Alternatively the complete TBR may include less than the complete PS' . While such pairing to the second polynucleotide or label strand is normally confined to a portion of the TBR, some limited number of base pairs outside of the TBR (e.g. , up to 15 such nucleotide pairs) is permitted, but not preferred. The effect of such an overhang or residual binding region corresponds to the effect of the RBR discussed in relation to Fig. IG of EPA 167,238. The length of the complete TBR is not independently critical, but can be considered a function or sum of the preferable lengths of the IBR and PS¹ portions. Base lengths of the IBR above five hundred are generally not required but can be used. A suitable lower limit on the length of the IBR depends upon the base sequence of the TBR and base composition and other physical factors, including the conditions and kinetics of the intended hybridization, and the readout system employed. In Method I the probe also contains a detectable tag. One or more tags may be located, using conventional techniques, at one of several points along the labeled probe, especially if the tag is a radionuσlide or biotin or the like. Alternatively one or more tags may be located only at one end or only at one specific internal location on the labeled probe (e.g., at a purine or pyri idine base not involved in base pairing to the second polynucleotide) . In the labeled probe, the tag is preferably located or concentrated in a probe region outside of the TBR.

Preferred forms of detectable tags, especially if remote from the probe TBR, should have little or no effect on the strength of base pairing between the labeled probe and either the second polynucleotide or TNS. In practice this may be determined by little or no diminution of the IRC melting temperature and, more importantly, by negligible effects on the hybridization reaction between TNS and the labeled probe.

B. Second Polynucleotide of Method I and Label Strand of Method II.

The second polynucleotide or label strand (DNA or RNA) includes a pairing segment PS bound in the reagent complex to the labeled probe or probe. The length of such PS corresponds to the length of the probe PS' , mentioned above. The pairing between the second polynucleotide and the labeled probe or the label strand and the probe can be perfect or can include a limited number of mismatches. The second polynucleotide or label strand can also be bound in the reagent complex via multiple PS, each to a portion of a unique TBR of the labeled probe. Such second polynucleotide or label strand would be fully displaced either by multiple targets of the sample, or by one or more sample targets and one or more selected nucleotide sequences of a reagent polynucleotide. Additionally, the selected reagent polynucleotide may bind to the second polynucleotide to displace a PS from a portion of a TBR in an inverse fashion.

In addition to the PS, the second polynucleotide can also contain a moiety capable of achieving physical separation of labeledprobes fromwhich secondpolynucleotide has been displaced from intact IRC (e.g. through immobilization, size separation, phase separation, gel electrophoresis and the like) . Such a physical separating moiety is exemplified by biotin, wherein the post- displacement separation would involve immobilized avidin, streptavidin or anti-biotin antibody. Alternatively, the second polynucleotide can be immobilized by a solid support in the IRC. For ease of reference, the physical separating moiety and the act of physical separation shall be referred to as "immobilizable". Covalent linkages are preferred when second polynucleotides are immobilized to a solid support in the IRC.

C. The CP of Method II.

In addition to containing the sequence CS, the CP also should have some property, or modification to make this capturer separable from intact labeled or reagent complexes as well as from other sources of background, such as detached tags. Suchmodifications include attachment of the CP to a solid support, such as colloidal particles, filters or membranes, and attachment of affinity reagents such as biotin to the CP. Alternatively, the CP can be significantly larger than the unreacted displacement complex to allow a size separation after hybridization.

Method II is applicableto displacement assays performed both on a solid support and solely in solution. For solid support assays, capturing reduces background signal in the assay which may result from the detachment of unreacted reagent complexes from the support due to breakage of the polynucleotide, failure of the attachment reaction, degradation of the solid support or other sources of non- specifically detached label. A second benefit is that the captured label strand can be removed from the displacement reaction and transferred to a position or condition which is more advantageous for readout. The combination of solid support displacement reactions with the use of CP has the advantage of potentially higher signal to noise ratios, due to two separation steps, and may be simpler to use.

For solution assays, capturing provides a method for separating displaced label strand from unreacted reagent complexes. Captured label strands from solution assays can also be transferred to more advantageous positions for readout. Solution displacement reactions have the potential advantage of better hybridization kinetics than reactions on solid supports, and allow the use of polymers such as polyethylene glycol (PEG) to enhance these reaction rates.

The use of a CP bearing an affinity reagent such as biotin permits both the initial displacement reaction and the capturing reaction to be carried out in solution. Separation of the displaced probe is achieved by trapping captured label strand on an avidin support (alternatively, an immobilized streptavidin or anti-biotin antibody) . An additional advantage is that the displacement reaction can be carried out using an easily prepared complex which contains the label strand and TNS in a single polynucleo¬ tide strand. This "covalent displacement complex" (RCC) simply unfolds after hybridization and strand displacement by the target, making the complement of the CP (CS' in Figure 4B hereof) single stranded and available for hybridization. Only RCC which have hybridized to TNS can be captured.

The CP can be added after the displacement reaction has been completed, or can be present during the displace¬ ment reaction. In general, the concentration of CP will be greater than the concentration of target in the sample and can be greater or smaller than the concentration of reagent complexes. Since the target and CP will be complementary over at least a portion of each sequence, target can first hybridize to the CP (Fig 4C) . Since the target is still available for hybridization to the TBR of the reagent complex, this CP-target hybrid (the "first intermediate complex") can now hybridize to the labeled reagent complex orming a "second intermediate complex". The second complex may be stable in this configuration, or may resolve by strand exchange and be recaptured. In either event, the displacement complex will only be captured when it has hybridized to target. If, however, second intermediate complexes are assayed as positive, then some of the characteristics of a sandwich assay are imparted. Therefore, some sample polynucleotides may be considered targetwhichhave regions complementaryto differentportions of the TBR spaced too far apart to cause displacement with high efficiency or which are incapable of complete displacement for a variety of reasons.

The capturing reaction is compatible with many readouts including enzymes, polyA tails (see EPA 200,057 and 200,056) fluorescent latex beads and enzyme substrates. In addition, capturing can occur in a defined space on a surface when the CP is already attached to a solid support. When a biotinylated (or other immobilizable) CP is used, label strand can be trapped at a particular position on a surface coated with avidin. An avidin coated "dipstick" could be used to trap captured label strand. Alternatively, avidin could be attached to surfaces to allow trapping in a predefined position for automated signal readout.

In a solution system, the CP provides the only means for separation (or the primary means if other steps are included) , and must bind to displaced label strand with high selectivity (not binding to undisplaced reagent complexes) . If the CP is immobilized or immobilizable onto a support, then the nonspecific binding of intact or broken reagent complexes to the support must be rare. The fraction of nondisplaced reagent complexes adhering to the support, either through nonspecific capturing or nonspecific binding to the support that would be per issable in the reaction is a function of both the initial number of reagent complexes and the number of target molecules to be detected. The relationship between these parameters can be described as follows. In a reaction with 10⁸ reagent complexes, 10^ target molecules can only be detected with a (signal and background)/background ratio of at least 2/1 if background nonspecific binding is less than 1%. As the fraction of nonspecifiσally bound complexes decreases, the signal to noise ratio increases, with a resulting increase in assay sensitivity.

In many cases, it is expected that capture alone can provide adequate specificity to a solution displacement assay. However, in some cases, capturing can be combined with both probe immobilization and/or release of captured tag from a support.

D. Tags for the Labelled Probe of Method I and the Label Strand of Method II.

Directly detectable tags for use in the methods, kits and reagents of the present invention include radionuclides and fluorescein compounds commonly employed in diagnostic probes attached to the free end of a labeled probe or to one or more of the bases of a labeled probe. Moieties directly detectable by other means including being cleaved off, being detectable σolorimetriσally or otherwise, like nitrophenol, may also be used as tags.

Indirectly detectable tags for use in the invention include those modifications that can serve as antigenic determinants, affinity ligands, antigens or antibodies recognizable through immunochemiσal or other affinity reactions, such as described inpublishedpatent applications EPA 63,879, WO83/02277, and EPA 97,373. Other such tags include biotinylated nucleotides present in or added onto the labeled probe (e.g., by the enzyme terminal deoxynucleotidyl transferase, which will add multiple nucleotides at the 3 ' end of the labeled probe in the absence of a template strand) . Other indirect tags are enzymes attached to a labeled probe (especially at a free end remote from the TBR) whose presence can be determined after the displacement and separation steps of the methods by addition of the substrate for the enzyme and quantifi¬ cation of either the enzymatic substrate or, preferably, the enzymatic reaction product. Similarly, the tag may be an apoenzyme, co-enzyme, enzymatic modifier, enzymatic cofactor or the like, with the other necessary reagents usually added after displacement and separation, along with the appropriate enzymatic substrate. If the enzymatic reaction cannot occur with all but one component present (e.g., the substrate), these other reagents may be present in solution during the contacting or displacement steps.

Other tags which may be employed in the present invention are the ribonucleotide segment tags described in European patent applications 200,056 and 200,057. Thus, e.g., a polyriboadenosine -("poly A") segment may be formed on the 3' end of a DNA portion of the probe including the TBR. After displacement and separation, the poly A segment can be digested with polynucleotide phosphorylase to riboadenosine diphosphate, which can then be phosphorylated by phosphoenol pyruvate in the presence of pyruvate kinase to adenosine triphosphate ("ATP") . This ATP can then be detected (e.g. , with luciferin/luciferase) as an amplified signal indicative of the number of such labeled probes so separated.

Multiple detectable tags can be added in the process of manufacturing the labeled probe or label strand of these methods by enzymes, such as terminal deoxynucleotidyl transferase, DNA ligase, polynucleotide phosphorylase, and the like. Multiple labeled probes or strands, each containing a signalling moiety or detectable tag, can also be used. One form of attachment of an enzyme to the labeled polynucleotide is via affinity reagents, e.g, streptavidin to biotin. Such a binding form can be used in various embodiments, (e.g. , where the complex is prepared by hybridizing a biotin-labeled polynucleotide to the probe and then binding a streptavidin-enzyme conjugate to the biotin prior to the contacting/displacement step described above. Additionally, a moiety interacting with the detectable tag in the complex may be present on the second polynucleotide.

The tags have been illustrated in the Figs, as on an end of a nucleic acid strand. However, other forms of single or multiple tags incorporated in or on the appropriate polynucleotide segment or strand can be used. The important features of the tag are that it be directly or indirectly detectable. Additionally, for Method II, it is important that the tag be positioned or attached to remain after displacement and capture with the appropriate strand or segment (PS in most cases) .

Method III involves disrupting the affinity linkage between biotin and avidin or biotin and streptavidin in a manner that avoids denaturing or disrupting the binding forces associated with diagnostic reagents attached to these affinity linkers. This method of disrupting the biotin/avidin linkage is particularly adaptable to Method II herein, where the CP is immobilized by such a linkage. Method III is a method for releasing a diagnostic reagent bound to a- support (or other substance) by a biotin/avidin or biotin/streptavidin linkage comprising incubating the bound diagnostic agent in the presence of an excess concentration of free biotin to the concentration of avidin or streptavidin present in the bound reagent-support complex. In a desirable embodiment of the method, the biotin concentration can be lmM or greater. Similarly, Method III is most preferable where the reagent or support is singly biptinylated. The incubation is performed at temperatures of between 22 to 65 degrees centigrade. The time required for the release is dependant on the temperature employed.

In Method III, the diagnostic reagent and support may be similar or different entities, and may include single or double stranded DNA or RNA, individual nucleic acids, oligo or polynucleotides, hybrids thereof, enzymes, cofactors, antigen, antibody, signalling or labelling moieties, tags, and derivatives and modifications thereof. Similarly, the support can be a "standard" support such as latex, agarose, filter, membrane, or natural, synthetic or semi-synthetic polymers.

Several desirable specific embodiments of Method III are where the reagent is a double stranded polynucleotide, which remains undenatured by the disrupting step. Addition¬ ally the reagent may be a biotinylated nucleic acid and the support is an avidin or streptavidin-associated nucleic acid. When used in conjunction with Method II, the diagnostic reagent of Method III can be a capture strand. Another desirable embodiment is where one reagent is a biotinylated DNA and the support is avidin conjugated to a signalling moiety, such as an avidin-enzyme. Similarly, the reagent can be a biotinylated antigen and the support an avidin-antibody.

This biotin displacement method is compatible with the use of TEAC1 buffer at 55 degrees C. The extent of displacement is equivalent to that observed with standard NaCl and Tris buffer solvents. In the absence of free biotin, the avidin/biotin linkage is still stable in TEAC1 buffer. Because TEAC1 is a hybridization solvent that can be used at 55 degrees C, there is a much improved chance of retaining enzyme activity and stable hybrid polynucleotide strands. The primary advantage of Method III for use in Method II is that it allows removal from solid support using conditions that do not denature polynucleotide hybrids. Method III is also useful for releasing signals from solid phase to solution phase, and also as a method of reducing background since background remains stuck to support while signal is specifically released into solution.

The operation of the methods of the present invention can best be described by resort to the attached Figures 1 through 5. Figs. 1A through 1C illustrate one embodiment of Method I, a displacement assay employing the IRC according to the present invention. Second polynucleotide (SP) is immobilized at one end by attachment to a support (SU) and has a PS at the opposite end. The labeled probe (LP) has a detectable tag (T) at one end and a TBR including about half of the complete length of the LP, including the end opposite the T. PS' of TBR is bound to PS of SP. The interior half of TBR is a single-stranded IBR. Fig. IB illustrates the IRC after a sample polynucleotide (G) having the target (TNS) has hybridized to the IBR. The portion complementary to IBR (IBR') forms a double-stranded portion (IBR/IBR¹) joining G to LP. Since LP remains bound to segment PS of SP, all three strands remain attached to SU.

Branch migration may now occur in which PS and the remainder of TNS zip and unzip within the end half of TBR. As described in EPA 167,238 this phenomenon is very rapid. Within several minutes or less, the displacable strand (in this illustration, SP) is displaced. As shown in Fig. 1C, TNS has now completely displaced the PS of the second polynucleotide from the TBR of the labeled probe. The hybrid G/LP which includes a tag T can now migrate away from the support and be separated therefrom in a liquid phase. All intact IRCs which have not been contacted by TNS should remain on support SU, as illustrated in Fig. 1C. Thus the tags detected directly or indirectly in the separated liquid phase give a qualitative and a quantitative measure of the TNS in the sample. Departures from the geometry of Figs. 1A to 1C within the scope of the invention can be understood by reference to various Figs of the strand displacement assay of EPA 167,238. In Fig. 1A herein, the second polynucleotide PS binds only to a portion of TBR. There is no residual binding region of labeled probe bound to PS but not part of TBR. Some number of such nucleotides are permissible; but in general such a residual binding region is preferably no greater than 15 nucleotides in length. [See Fig IG and Example 13 of EPA 167,238]. If such a residual binding region is present at the stage of complete displacement (Fig. 1C) wherein TNS is fully hybridized with TBR, such residual binding region of the probe could still be bound to PS. The reaction conditions (e.g., temperature and salt concentration) could be changed or adjusted by conventional means at the end of the reaction, to conditions at which labeled probe bound only by the residual binding region melts off the PS.

In other unillustrated embodiments of Method I, the end of the second polynucleotide opposite to PS has a moiety (e.g., biotin, as in a poly-bio-dϋtail) immobilizable by an immobilized avidin or streptavidin. In such cases, the displacement is preferably conducted in solution. Thereafterthe reactionmixture is contactedwithimmobilized avidin or streptavidin to immobilize intact IRCs and displaced second polynucleotides. Sample strand/labeled probe hybrids are left in solution for subsequent direct or indirect detection. A homopolynuσleotide tail (e.g., poly-dC) can be present on the end of SP opposite PS. The post-displacement immobilization step would employ an immobilized complementary homopolynucleotide (e.g, oligo- dG-cellulose to immobilize the poly-dC tails of SP) .

When the labeled probe/sample polynucleotide hybrid is in a liquid phase for subsequent detection, intermediate treatments can be conducted to obtain further information (e.g., restriction fragment size or sequence information) about sample polynucleotide. Thus, the liquid phase may be subjected to gel eleσtrophoresis [See, S. G. Fisher and L. S. Lerman, Proc. Natl. Acad. Sci., USA. 72, 989-993 (1983); R. M. Myers et al, Nature, 313: 495-498 (1985)], restriction endonuclease treatment, SI nucleasedigestion or ribonuclease digestion [R. M. Myers, et al.. Science. 230: 1242-1246 (1985)] to develop information beyond merely the number of target sequences.

A special embodiment of the diagnostic reagents of the present invention is the IRCC of Fig. 2A, which has a single polynucleotide strand including a PS near one end and a TBR adjacent to the opposite end. This opposite end also bears a detectable tag T. PS (containing a portion labelled CS') . is bound by complementary base pairing to the half of the TBR nearest to T, leaving the interior half of the TBR single-stranded as an IBR. Similarly, in Fig. 2B an IRCC has been contacted by a sample polynucleotide strand G containing the target TNS. TNS has bound to all of the TBR, displacing PS. Here PS is in single-stranded form, remaining attached to TBR and T by the covalent phosphate/sugar backbone. Alternatively, other stable attachments could be employed for this purpose.

The hybrid of Fig. 2B is illustrated in Fig. 2C as subsequently bound by an immobilized CP which contains a portion CS complementary to a portion on the IRCC, designated CS' . CP will capture the hybrid, but not the intact IRCC of Fig. 2A. CP can also hybridize to sample polynucleotide G directly, forming an intermediate (CP-G) which can then hybridize to the IRCC probe complex. Although such intermediates can function in the assay, it may be desirable in using the IRCCs of Fig. 2A, to employ proportions, orders of addition or other parameters either to avoid or to fully resolve any such intermediate structures as may form.

Modifications can be made in the IRCC of Fig.2Awhereby the IRCC is initially immobilized. Initial immobilization of the IRCC requires an immobilized third polynucleotide having a binding segment complementary to and bound to a portion of TBR other than the portion of TBR bound to PS. Either binding segment would be bound to a portion of the IBR segment shown in Fig 2A or alternatively, the TBR would extend to the right of the PS' portion and the binding segment would be bound to all or a portion of TBR to the right of PS* . Displacement of such third polynucleotide would release IRCC from the support. In Fig 2C, the tag T can be detected on the solid phase or can be introduced into a new liquid phase after the liquid phase containing intact IRCCs has been removed.

Figs. 3A through 5 illustrate Method II. Fig. 3A shows two reagent complexes on a support, having probes Px and P₂ with identical TBR. A label strand or signal strand L and L₂) is bound to a portion of each TBR. The portion of L_]_ so-bound is PS. The portion of TBR not bound to PS is the IBR. In Fig. 3B, sample strand G containing TNS has bound to the IBR of Pχ« The reagent complex including probe P₂ is intact. At this point, the top reagent complex is subject to strand migration and ultimately displacement of segment PS from TBR. The result, shown on the left of

Fig. 3C, is that G is bound to P^, via TNS and TBR and that

L-_L is displaced into solution. Intact reagent complexes

(such as that containing P in Fig. 3C) will normally be present at a 100-fold, 1000-fold or greater excess relative to displaced L-_]_ in Fig. 3C. If any such intact reagent complex separates from the support at the point of attachment, or if separation occurs at any point along the strand to the left of TBR, or if the tag T detached from L₂, then detectable tag other than the displaced Li will be in solution.

On the right of Fig. 3C, a portion CS* of the PS of the label strand is shown. CP is shown in immobilized form, with a capture segment CS which is complementary to segment CS¹ of L^. After the displaced L-^ is in solution, it can hybridize to CP via CS'/CS, forming the captured structure shown in Fig. 3D.

Comparing the intact reagent complex at the bottom left of Fig. 3C with the capture structure shown in Fig. 3D, it can be seen that most forms of background signal which may be non-specifiσally released into solution will not specifically bind to CP. If P₂ either detaches from its support or breaks near its left end, then the hybrid released into solution does have a segment homologous to CS' of Li within L₂, but such segment is in the middle of a double-stranded segment and is thus not available for hybridization to CS. If tag T breaks off, it has no polynucleotide sequence specific for CS. The only form of non-specific signal likely to bind to CP with high efficiency is any intact or nearly intact L₂ which melts from probe P₂. This is not a likely source of significant background; however, such melting should be avoided, as should the presence of label strands which never bind to probes (e.g., those adsorbed non-specifically to the support) .

Fig. 4A illustrates a RCC having a TBR near one end and a PS near the other end, with a tag T on the end near PS. PS is hybridized to the end-most portion of TBR, leaving the interior portion IBR of TBR single-stranded. No optional hairpin structure separating TBR from PS is shown in Fig. 4A. In Fig. 4B, a sample polynucleotide G having a TNS has bound to IBR (see Fig. 4A) and displaced PS from TBR, forming a complete TNS/TBR double-stranded segment. The PS, adjacent to the tag T, is now in single-stranded form. An interior (intermediate) portion of the PS is designated CS' . In Fig. 4C CP is shown with a capture segment CS complementary to portion CS' of PS. CP also has a series of attached biotin moieties (shown as B's) . Biotin may be attached enzymatically, chemically, or photochemically.

In Fig. 4D, the hybrid between the opened reagent complex of Fig. 4B and the CP of Fig. 4C is shown. The pendant biotins can now bind to immobilized avidin (shown as Av on support IM) . Comparing Fig. 4A to Fig. 4D, it is seen that segment CS• is unavailable in the intact RCC for hybridization to CP. Accordingly, CP can efficiently separate RCCs which have been hybridized to sample polynucleotides G having TNS from those reagent complexes which have not been contacted by TNS.

Once the CP has bound to the support IM via biotin/ avidin, the solid phase can be separated from the liquid phase. Tag T attached to the solid phase can now be determined. Alternatively, tag T may be specifically released into a fresh solution phase in a variety of ways: (1) displacement with dissolved biotin, (2) melting the double-stranded segment CS/CS' if it is sufficiently short, (3) displacement with either (a) a strand complementary to more of PS than is CS' or (b) a strand complementary to more of CP than is CS or (4) specific chemical cleavage of the link between tag T and segment PS. Modes of release are preferred which do not operate on reagent complexes non-specifically bound to support phase IM. Thus, in many cases, displacement modes (1) and (3) are preferred.

The scheme shown in Figs. 4A-4D is available whether CP was present during hybridization of G to IBR or not. Nevertheless., it is not the only mechanism available. especially when RCC, CP and sample are all admixed together. It should be apparent from Figs. 4A, 4B and 4C that the target can, and usually does, contain a segment identical to segment CS¹ and thus complementary to segment CS, or functionally identical in ability to form a stable hybrid with segment CS under the conditions employed. Thus, especially where CP is provided in excess relative to RCC (each is generally in excess relative to target) , CP may bind to part of TNS, forming a first intermediate structure shown in Fig.. 2E. Most of TNS remains single-stranded in this structure, including a portion ("IBR"¹) of TNS which is complementary to IBR or TBR.

The first intermediate, shown in Fig. 4E, may still bind to the RCC shown in Fig. 4A, with IBR' binding to IBR, forming a second intermediate structure. Strand migration at this point will normally lead to formation of a completely double-stranded TNS/TBR segment. Thus, the structure of Fig. 4B should form with both PS displaced from TBR and CS displaced from the CS' portion of TNS. The same CP, or a different such molecule, may now bind to portion CS' of PS. If, however, the CP/TNS/RCC second intermediate formed when IBR' binds to IBR is stable (strand migration does not displace CP and PS) , then such second intermediate may be a second source of signal specifically bound to support IM. It may, however, be desirable to avoid the formation of first and second intermediates or allow such second intermediates ample time to resolve by branch migration before the original reaction mixture (liquid phase) is removed from support IM. Any such intermediates present at the time when the original reaction mixture is removed from support IM may subsequently complete strand migration and thus detach the signal, unless the particular displaced CP molecule (or an adjacent one on the solid phase also bound to avidin) can find the simultaneously displaced PS segment. Such a second intermediate in a reagent complex having distinct probe strands and label strands may resolve in such fashion to a major extent under certain conditions.

The lengths of TBR, IBR, PS and heterologous portions of the probe and of the label strand can, in general, be any of the suitable or preferred lengths indicated in EPA167,238 and 164,876. Thus, for example, preferred ranges include about 50-1000 nucleotides for IBR, 20-1000 nucleotides for PS (equal in size to LB in the Figs, of these applications) 70-2000 nucleotides for TBR and 0-15 nucleotides of PS binding to the probe outside of TBR.

The portion CS' of PS to which CP can specifically bind once PS is displaced from TBR can be all of PS or most of PS or at one end of PS. It is preferred, however, that each end of CS' be located at least 25, and preferably at least 50 nucleotides from each end of PS (that is, be an intermediate segment of PS) . Such an intermediate segment has reduced susceptibility to becoming momentarily single- stranded under circumstances (sometimes called "breathing") in which the PS/TBR duplex reversibly unwinds (only to subsequently rewind and thus not separate) . Since such breathing is more likely to occur at an end, PS which have CS^f segments that include nucleotides at an end of PS may be captured by CP even without displacement by TNS having occurred. Such non-specific capture could result in non¬ specific signal detection unless the rewinding of TBR and PS was effective to displace such CPs which had been non- specifically bound to PS.

The length of segment CS' within segment PS (and thus also the length CS within CP) is preferably at least 25 nucleotides and more preferably 50-500 nucleotides. At the lower end of the more preferred range for CS ' (50 nucleotides) , it is preferred that PS be at least 100 nucleotide in length (to permit 25 nucleotides on each side of CS¹)^' and more preferred that PS be at least 150 nucleotides in length (to permit 50 nucleotides on each side of CS•) .

Random introduction of breaks ("nicks") into the normally intact label strand either prior to or during the capture event, can be a source of non-specific signal. Such non-specific signal can also be reduced by the portion of the PS complementary to the CP being internally located within the PS. The preferred segment lengths and positions of the PS and CS described above reduce or eliminate both non-specific capture due to breathing and non-specific signal detection due to such nicking.

Multiple PS segments binding to portions of the same TBR and multiple TBR regions of the probe, with a label strand having a separate (and different) PS bound to each are also embodiments of this invention. In most cases, the CP need bind specifically only to the PS of a label strand which had been displaced by sample TNS when that displaced PS has been separated from the immobilized or immobilizable probe.

Embodiments are also contemplatedusingmultiplereagent complexes to simultaneously contact a sample to assay for multiple TNS. In such case, one or more CPs can be used to separately isolate each group of displaced label strands for separate detection of tags, indicative of the presence and amount of each target. Such separate reagent complexes can be replaced by a probe having multiple label strands attached, each to a portion of a specific TBR. Additionally, the several label strands can be captured simultaneously, but then released for detection sequentially.

The CS/CS^» , TBR/TNS and PS/PS* binding described herein need not be perfect, but may include a limited number of mismatches (or deletions or other loop-forming sequences) , so long as such mismatches are insufficient in number to prevent the desired specific binding or (considering the topology and size of PS and TBR) to inhibit the desired hybridization of CS to CS' or to cause non-specific melting of the CS/CS¹ duplex. One example where mismatches may be present is where the binding of TNS to TBR is perfect and the binding of CS to CS' (a portion of PS) is perfect, but the binding of PS to PS' (a portion of TBR) contains mismatches. Such example may prevent rehybridization of CS* to TBR leading to reversal of displacement and loss of signal. This approach may be especially useful for σovalent reagent complexes where following displacement the displaced pairing segment is held in close proximity to the TBR. Concerning PS/PS' pairing, no regions of the label strands should be in a single-stranded form (e.g., in loops) that can hybridize to capture strand without displacement.

It is also desirable to avoid complementarity of significant size (in general, greaterthan 10-25 nucleotides depending on sequence, melting temperatures and reaction conditions) between the CP (either region CS or heterologous regions outside of CS) and regions of the label strand outside of PS or between the CP and the probe (either within TBR and especially IBR, or outside of TBR) . Such complementarity can result in hybridization of labeled reagent complex to the CP in the absence of displacement, resulting in increased background.

Of thevarious permutations of reagent complextopology and CP, it should be appreciated that either immobilized or immobilizable CPs may be used for either solution or solid phase displacement assays employing either: (1) covalent reagent complexes, (2) reagent complexes with distinct label strands andprobe or (3) covalent IRCs. While covalent attachment has been used herein as an exemplary means for attaching probes or CPs to solid supports or to attach PS to TBR, other forms of attachment such as stable hydrogen bonding (e.g., complementary base pairing) can be employed. Such other forms of attachment, such as biotin avidin attachment and especially complementary base pairing, can also be employed for post-capture immobilization of an immobilizable CP (e.g., poly-dC on CP immobilized by oligo- dG-cellulose, or similar immobilization by sequences of CP outside of segment CS binding to complementary sequences on a strand immobilized to a solid support or otherwise separable, e.g., by virtue of size).

Thevarious reaction conditions (e.g., temperature, salt concentration, presence of agents such as recA protein and co actor or polymers such as PEG can be as described in EPA167,238 and 164,876, with the additional proviso of not destabilizing the CS/CS* pairing, once formed.

In general, the number of CPs should exceed the number of expected displaced PS. Accordingly, in the usual case where the number of reagent complexes exceeds the anticipated level of TNS, the number of CPs should also exceed anticipated levels of TNS, preferably by a factor of at least 10, more preferably by a factor of at least 100. Thus with 10^β expected analyte strands having TNS (whether there are 10⁷, 10⁸, 10⁹ or 10¹⁰ reagent complexes employed) , there should be at least 10⁶, preferably at least 10⁷, more preferably at 10⁸ and commonly 10⁹-10¹⁰ molecules of CP employed. In such cases, the molar ratio between reagent complexes and CP used in the same assay method (or present in the same kit) does not have independent significance, but rather varies within the range 10⁴:1 to 1:10⁴ reagent complexes to CP or even outside of such broad range. However, the minimum number of CP that may be present is determined by the minimum number required to hybridize a sufficient fraction of displaced label strands in order to result in a signal of the desired intensity within the desired period of time.

The following examples illustrate the components and operation of the methods of the present invention.

EXAMPLE 1 Constuction of IRC model target and capture strand. A. IRCs were prepared from a nucleic acid construct as follows: Two E. coli plasmids, pMLC12 and pMLC13, each containing ah M13 origin of replication, were prepared by partial Hindlll digestion and Klenow end fill of plasmids pSDL12 and pSDI.13 [A. Levinson et al., J. Mol. Appl. Genet.. 2.: 507-517 (1984)]. pMLC12 and pMLC13 were digested to completion with EcoRI. The large fragment from pMLC12 and the small fragment from pMLC13 were combined, ligated and transformed into MC1061 (F-;hsdR, delta ara-len 7697, araD139, delta lac X 74, galU, galK, rpsL (str^r) ) [See, Y. Casadaban, J. Mol . Biol .. 138 : 179-207 (1980)]. Chloramphenical resistant colonies were picked and the correct plasmid, designated pMLC12/13 delta, was identified on the basis of loss of the BamHI cleavage site seen in both pMLC12 and pMLC13. pMLC12/13delta was partially digested with EcoRI so that only one of the two EcoRI sites is digested in most molecules. The partially cut, linearized pMLC12/13delta was isolated following gel electrophoresis. Mp7 DNA was digested with PvuII and a 383 bp fragment was isolated following gel electrophoresis. The PvuII fragment was digested with EcoRI to produce the 52 bp EcoRI fragment containing the Mp7 polylinker and two other PvuII/EcoRI fragments (about 123 bp and 208 bp) . The EcoRI digested PvuII fragment from Mp7 and the linearized, partially EcoRI digested pMLC12/13 elta were ligated, transformed into MC 1061 and chloramphenicol resistant cells were selected. Individual colonies were then grown and DNA was prepared. Plasmids which had correctly incorporated the EcoRI polylinker from Mp7 were identified by the acquisition of a BamHI site and designated pMLC12/13deltaM7. pMLC12 was digested to completion with Hindi. The plasmid pAlbB6 which contains a portion of a human albumin cDNA clone [Lawn et al. , Nucleic Acids Res. , 9_:6103-6114 31

(1980)] was digested to completion with PvuII and Hindi. The 915 bp PvuII/HincII fragment and the pMLC12/HincII vector and were ligated together and transformed into MC1061. Clones which had incorporated the correct albumin fragment were identified by colony hybridization [Grunstein and Hogness, Proc. Natl. Acad. Sci., U.S.A., 72.:3961 (1976) using the alb 32 mer (5^»ACATCCTTTGCCTCAGCATAGTTTTTGCAAAC3') as a hybridization probe. Positive colonies were picked and grown up; and DNA from individual colonies was digested with Hindlll and Pstl to determine the orientations. Plasmids F31a and F31c containing the albumin fragment insert in opposite orientations were selected for further steps.

F31a was digested to completion with EcoRI and BamHI and the two large fragments were isolated. F31c was digested to completion with EcoRI and Bglll and the small (approximately 550 bp) fragment (between the EcoRI and Bglll sites) was gel isolated. The gel isolated fragments from F31a and F31c were ligated together and transformed into DH1 (ATCC #33849) a recA" bacterium. (The recA~ host was used to reduce the possibility of deletion of one or both copies of the inverted repeat through a recA mediated mechanism. Subsequent experiments have shown that these inverted repeat clones are stable even in the absence of recA" mutation.) DNA was then prepared from .individual chlora phenicol resistant colonies and digested separately with Pstl, EcoRI plus Hindlll, or Bglll plus Bam HI to identify clones with the correct structure. One such clone F41a was used for further analysis.

F41a was digested with EcoRI and Hindlll and blunted with the Klenow fragment of DNA Polymerase I. The approximately 1500 bp fragment was isolated following gel electrophoresis. pMLC12/13deltaM7 was digestedto completion with Ace I and blunted with the Klenow fragment of DNA Polymerase I. The pMLC12/13deltaM7/Acd and Klenow fragment and^*the gel-isolated fragment from F41a were joined by DNA ligase and transformed into MC1061. Plasmids which had incorporated the gel isolated fragment were identified by hybridization to the albumin 32 mer and were verified by digestion with Pstl, BamHI, or Xbal. This plasmid was designated pMLC12/13deltaM71VR.

Plasmid MpTL poly was prepared by phosphorylating the lower oligonucleotide strand of:

EcoRI

5••-HO-TCACGAATTCCATCTGTCAAGG-OH-3^« 3'HO-GTGCTTAAGGTAGACAGTTCC-OH-3'

After ligating, a dimer was isolated having the middle portion:

STU I -ATCTGTCAAGGCCTTGACAGAT- -TAGACAGTTCCGGAACTGTCTA- recognizable because of the Stu I site that had formed. The dimer was filled with Klenow fragment of DNA polymerase I so as to have the end portions:

Eco Rl Eco Rl

5^» CACGAATTCC- -GGAATTCGTGA-3 '

3'AGTGCTTAAGG- -CCTTAAGCACT-5*

This duplex was now cloned into Mp8/Hinc II and a clone picked up on the basis of acquisition of a Stu I site. Plasmid MpTL poly was confirmed upon sequencing. pMLC12/13deltaM71FR was partially digested with Xbal (which cuts four times within the plasmid) , blunted as above, and full length linearized plasmid DNA was isolated by gel electrophoresis. The plasmid MpTL poly was digested to completion with BamHI and Hindlll and was blunted as above. The 80 bp blunted fragment was isolated, linearized, Xbal partially cut, blunted pMLC12/13deltaM71VR. The DNA was transformed into MC1061 and screened with the oligonucleotide which is complimentary to the TL polylinker. Positive colonies were picked and plasmid which had incorporated the TL polylinker at the correct Xbal site was identified by digestion with Stul and Bgl II and designed pMLC12/13 delta M7IVRTL. pMLC12/13deltaM7IVRTL thus contains fragments from the human albumin gene cloned as inverted repeats in an M13 origin plasmid. Single stranded forms of the construct fold up into a stem loop structure. Cleavage of the mp7 polylinker with a restriction enzyme releases the stem-loop structure from the single-stranded vector backbone. This insert can be used directly as an approximately 1.6 kb nucleotide covalent displacement complex with an approxi¬ mately 0.5kb Bgl II-Hinc II albumin fragment as the signal strand, and an approximately 1 kb Bgl II-Pvu II albumin fragment nucleotide TBR. In single-stranded complexes, the TBR is located 22 nucleotides from the 5* end and the label strand 41 nucleotides from the 3* end. pMLC12/l^'3deltaM7IVRTL contains an additional inverted repeat ("IVRTL") located at the inside edge of the label strand, which forms a 52 nucleotide (26 base pair) hairpin containing a double-stranded Eco Rl cleavage site in single stranded forms of this construct. Cleavage at both the Bam HI site in the mp7 polylinker and at the IVRTL Eco Rl site results in the formation of a displacement complex in which the label strand and target strand are held together only by base pairing. Two forms of the construct in which the IVRTL hairpin is or is not cleaved by Eco Rl are distinguished by referring to them as the "non-covalent complex" and the "covalent complex", respectively. p66b was constructed by gel-isolating the double- stranded PvuII fragment containing the sequence for the entire displacement complex from pMLC12/13deltaM7IVRTL and ligating it to the gel-isolated PvuII backbone of the M13 origin plasmid pUC119. Single-stranded forms of the construct were produced as previously described for pMLC12/13deltaM7IVRTL, except that the DNA was transformed into the E. coli host strain MV1193 [obtained from Dr. Michael Volkert; JM101 del(srIR-reσA) 3OG: : nlO] . Superinfections are routinely done with baσteriophage M13K107. The resulting displacement complexes are identical to those produced by pMLC12/13deltaM7IVRTL; this method is preferred only due to higher yields of the displacement complex. IRCs used in Examples 2 and 3 below result from labeling of the 5* end of p66b.

Model target was constructed by gel purifying a 2 kb Hindll-Eco Rl fragment from a plasmid pAllAlb which contains the entire cDNA sequence of human albumin. The Hindlll site is the Hindlll site in the 3 ' end of the albumin cDNA. The EcoRI site is present in adjacent vector sequences. The Hindlll-EcoRI fragment was ligated into Hind III-EcoRI digested M13mp8 to give mp8AHAlb. Single stranded DNA was purified from phage containing this construct, and was partially digested with Haelll to linearize these targets. There are no Haelll sites within the albumin cDNA sequence. mpSAllAlb template DNA is complementary to the TBR of p66b displacement complexes.

Capture strand mpl9AlbTaqPst was constructed by ligating a 280bp Pstl-Taql segment isolated from a 350bp Bglll-Pstl fragment of human albumin cDNA into Accl/Pstl digested mpl9RfDNA. mp7deltaAlbXbal+ was made by digesting mpl9AlbTaqPst Rf DNA with Xba I, and end filling and gel purifying the resulting 300bp fragment, and ligating it to the 680bp gel purified PvuII vector backbone fragment of mp7. One of the two resulting phage isolates containing single stranded albumin DNA complementary to the labeled polynucleotides of p66b displacement complexes is labeled 1+. mp7deltaAlb Xbal+ differs from mpl9AlbTaqPst in that a portion of the lac gene and all polylinker cloning sequences are deleted from the mp7delta backbone. Further, the albumin insert in mp7delta AlbXbal+ is complementary to a more interior portion of the pairing segment.

Capture strand mp7deltaAlbXbal+ was biotinylated using Photoprobe Biotin (Vector Laboratories) essentially as described by the manufacturer.

Successful reaction was monitored by taking an aliquot of the biotinylated DNA and hybridizing a 32-P labeled oligonucleotide (σALB 32-mer) complementary to a 32 base segment of the capture strand. One-half of the sample (control) was then electrophoresed directly on an agarose gel. The other half was mixed with 10 microliter of streptavidin latex beads (Pandex Laboratories) in 0.2 M NaCl, 20 mM Tris-HCl, pH 8.0, 0.1% NP-40 for 10-20 minutes at room temperature. After the binding step, the beads were removed from the solution by centrifugation (2 minutes, Eppendorf centrifuge) and the combined solution phases were electrophoresed in a parallel lane to the control sample. Following electrophoresis and autoradio- graphy, nearly all the 32-P labeled oligonucleotide sample that was hybridized to mp7deltaAlbXbal+ was removed from the sample that was exposed in the streptavidin agarose. These results indicated that the majority of capture strands had at least one biotin group attached.

EXAMPLE 2 Displacement and Capturing

In previous capture procedures employing capturers complementary- to the entire PS segment (labeled probe) of the complex, a small amount of background capturing (about 0.5%) was observed. Use of ^'controls eliminated the possibilities that background was derived from M13 cross hybridization or strand exchange. Therefore, it is likely that a small percent of these complexes are being nicked near enough to the 3' end to allow dissociation of these labeled nicked fragments from the complex andrehybridization to-the capturing strand. This example demonstrates that background can be eliminated by using a complex and capturer combination wherein the capturer can only hybridize to an internal portion of the displaced label strand, and small nicked fragments would then not be captured.

Single stranded p66b DNA was digested to completion with Bam HI, and the covalent complex was isolated by gel purification. The complex was labelled by ligating a 32P- kinasedoligonucleotidewiththesequence5*GATCCGCGGCGGTAC3* to the 3* end of the complex [See T. Maniatis et al. Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory (1982) forkinase reaction conditions] as follows: The labeled oligonucleotide was purified away from unincorporated 32P -rATP by centrifuging the reaction twice at 6000 rpm for 30 minutes in a total volume of 500 ul of TE in a Centricon 10 filtration device from Amicon. 10pm kinased oligonucleotide and 2.4pm of Bam HI cut p66b gel purified complex were precipitated together with ethanol, and resuspended in 14 ul of TE. 4 ul of 5X ligase buffer [Maniatis et al supra] and 2 ul of DNA ligase were added and the reaction was incubated at 15"for 4 hours. The reaction was diluted to 50 ul with TE, heated to 42 for 10 minutes to melt any non-ligated oligonucleotides and electrophoresed on a 1% agarose gel. Following electrophoresis, the gel was stained with ethidium bromide and viewed with a uv light box. Approximately 66% of the complexes ligated to each other, so that only 0.33 pm of complex were available for ligation to the kinased oligonucleotide. The kinased complex was purified and had an estimated specific activity of about 8 x 10⁴ per pm.

Either 0.02 pm (1000 cpm) or 0.03 pm (3000 cpm) were incubated in the presence or absence of 0.01 pm Hae III cut mpδ.AllAlb target with 0.1 pm of capture DNA in a final volume of 20 ul of hybridization buffer (0.3M NaCl, 0.1 M Tris HC1, pH8.0, and 10 mM EDTA) for 60 minutes at 65°C. Reactions were analyzed by gel electrophoresis and autoradiography.

In the absence of target, no detectable background capturing was observed with any of the following capturers: mp7, mpl9, mp7deltaAlbXbal+, mp7delta AlbaXba3- or mp7deltaAlbXba4-. The addition of 0.01 pm target to reactions with 0.01 pm complex resulted in 100% displacement; no capturingwas observed with the mp7deltaAlbXba4- capturer, and 100% capturing was observed with the mp7deltaAlbXbal+, capturer. The addition of 0.01 pm target to reactions with 0.03 pm complex resulted in approximately 50% displacement and no capturing with mp7delta AlbXba4- and complete capturing of the displaced complexes with mp7delta AlbXbal+.

Capturing of displaced label strands can also be accomplished when the CP is immobilized, e.g. , by attachment to latex particles by conventional means.

EXAMPLE 3 Capturing and trapping of displaced complex withbiotinylated capture strand p66b Bam complex was labeled at the 3' end by ligation of a 32P-labeled oligonucleotide and purified as described in Example 2. Biotinylated mp7deltaAlbXbal+ was used as the capture strand. Four reactions were set up which included combinations of 0.03pm complex with the following combinations of capture and the Haelll cut mpδ.AllAlb target: Reaction 1 with 0.01 pm target and 0.08 pm capturer; reaction 2 with 0.08 pm capturer only; reaction 3 with 0.01 pm target only; and reaction 4 with neither target nor capturer. All reactions were performed in 50 ul of hybridization buffer, and incubated at 65°C for 60 minutes.

The reactions were then bound to about 100 ul packed volume streptavidin agarose (BRL) which had been washed twice in 500 ul of binding buffer (0.2 M NaCl, 0.05 M Tris HCl, pH8.0, 100 ug/ml sonicated salmon sperm DNA, 1 mM EDTA and pelleted in an Eppendorf tube. Samples were diluted to 38

100 ul with binding buffer, transferred to the tube containing the streptavidin agarose pellet, mixed in an Eppendorf shaker for 1 minute and allowed to bind without shaking for an additional 2 minutes. The samples were then diluted to 500 ul with binding buffer, mixed for 2 minutes in the shaker, and centrifuged for 3 minutes to pellet the streptavidin agarose. The supernatant was removed and counted for radioactivity. A 3 minute incubation of the reaction with the streptavidin agarose was insufficient to give complete trapping.

The pellet was washed three more times with 500 ul binding buffer and all supernatants and the pellet were counted. The first rinse was then rebound to a fresh 100 ul packed volume aliquot of streptavidin agarose (prewashed with binding buffer as before) in a 5 ml Sarstedt centrifuge tube which was rotated end over end for 15 minutes during binding. The agarose was pelleted by centrifugation, and the supernatant saved and counted. The agarose was washed with 500 ul binding buffer for 15 minutes as above, followed by centrifugation and counting of this supernatant and pellet. Substantially maximal trapping occurred after 15 minutes of incubation. The pellet was then reincubated with the supernatant for 45 minutes to determine if any additional binding would occur with a longer incubation. The agarose was then centifuged, the pellet rewashed, and the pellet and two supernatants were counted. The longer incubation did not result in any significant further trapping. Reaction 1 had a % binding of 35%; reaction 2, 3.0%; reaction 3, 1.4% and reaction 4, 1.5%. This experiment demonstrates the use of a biotinylated capturer with a streptavidin agarose trap to collect the captured molecules. Only the reaction which included all three components showed a significant amount (35% of total counts) of label bound to the support. EXAMPLE 4 Large scale displacement and capture with trapping on streptavidin agarose.

The Bam p66b covalent displacement complex was labeled to a specific activity of about 10⁶ cpm/pm by ligating a 32P-kinased oligonucleotide to the 5' end of the complex with the use of a 21 base splint. 10 pm of the kinased lδmer (indicated below by the asterisk) , 10 pm of splint, and 1 pm of p66b Bam (underlined below) were incubated together at 22^βC for 15 minutes in 10 ul of IX ligase buffer. 1 ul of ligase was added and the reaction incubated for an additional 30 minutes. The three molecules form the joint diagrammed below.

*CGAAGCTTGGATCCGCGATCCGTCAGCTT...p66b GAACCTAGGCGCTAGGCAGT (SPLINT) Four reactions were set up [Reaction 1 with 0.1pm complex only; Reaction 2 with complex plus 0.16 pm capture strand; Reaction 3 with complex, capture strand and 0.05 pm target; and Reaction 4 with complex, capture strand and 0.01 pm target] in a total volume of 50 ul of hybridization buffer (0.3M NaCl, 0.1 M Tris HCl, pH8.0, and 10 mM EDTA) and incubated for 30 minutes at 65"C. Haelll-cut mpSAllAlb and biotinylated mp7deltaAlbXbal+ were target and capturer, respectively. 25 ul of each reaction was then analyzed by gel electrophoresis and 25 ul by binding to streptavidin agarose as follows. 100 ul packed volume of streptavidin agarose was washed twice in 500 ul binding buffer in a 5 ml Sarstedt tube rotated end over end for 15 minutes, and pelleted by centrifugation. The 25 ul reaction aliquots were diluted to a total of 500 ul binding buffer, and incubated, rotating as above, for 15 minutes. The sample was transferred to an Eppendorf tube for centrifugation and the supernatant was saved. The pellet was rewashed as above, once at room temperature for 15 minutes, then twice at 65"C for 15 minutes, then for 60 minutes at room temperature and finally for 15 minutes at room temperature with lOmM Tris HCl pH8 ImM EDTA [TE] . The final pellet and all supernatants were counted. The percent cpm bound to agarose for each reaction are 1.7% (Reaction 1); 2.5% (Reaction 2); 26.3% (Reaction 3); and 8.7% (Reaction 4). These results show that binding of complex to the support is dependent upon the presence of capturer and target and upon the amount of target present.

Gel analysis of the same reactions indicated that there is less than 0.05% non-specificcapturing in these reactions. Specific capturing was more efficiently analyzed by gel separation, in that the presence of target resulted in capturing of approximately 80% and 20% of the complex in reactions 3 and 4, respectively.

EXAMPLE 5 Prehybridization ofcomplex and target, followedbycapturing Two additional reactions were done using the Bam p66b complex described in Example 4. In these reactions, 0.1 pm complex alone (reaction 1) or 0.1 pm complex and 0.05 pm Haelll cut mpδAllAlb target (reaction 2) were incubated in 50 ul of hybridization buffer for 30 minutes at 65^βC. 0.16 pm of biotinylated mp7delta.AlbXbal+ was then added to both reactions. Both reactions were then divided and treated as in Example 4, except that all rinses were at room temperature with binding buffer. By gel analysis, less than 0.05% non¬ specific capturing, and approximately 40% specific capturing was observed. The results of analysis on streptavidin agarose showed 1.9% cpm bound to support for reaction 1 and 13.5% for reaction 2. These results indicate that capturing and trapping occur with approximately equal efficiencies whether capture DNA is added after (Example 8) or is present during (Example 4) the target-dependent displacement reaction. EXAMPLE 6 Comparison of displacement and capturing with covalent and non-covalent p66d complexes.

Covalent p66d displacement complexes were prepared and labeled at the 5' end by ligation of the kinased 16mer using the EF21 splint as described in Example 4. The specific activity of the resulting complexes was about 1 x 10⁶ cpm/pm.

Non-covalent p66d complexes were produced by complete digestion of approximately 50 ug of single stranded templated DNA with Bam HI and Eco Rl. Complete digestion was ascertained by the appearance of equimolar amounts of three bands, corresponding to vector backbone, target strand, and signal strand, after electrophoresis of a small aliquot of the digest on an alkaline gel. Non- covalent p66d complexes were labeled at the 5' end of the signal strand as described for covalent complexes, by EF21 splint ligation of a kinased 16mer, with a resultant specific activity of 3 x 10⁶ cpm/pm.

Four reactions were set up [reactions 1 and 2 contained 0.2 pm target and 0.8 pm capturer with 0.2 pm of either covalent or non-covalent complex, respectively; reactions 3 and 4 were identical to 1 and 2 except they contained no capturer] in a total volume of 50 ul of hybridization buffer [See Example 4] and incubated for 60 minutes at 65 using Haelll digested mpll.AllAlb DNA as target and biotinylated mp7deltaAlbXba3- capturer. 10 ul of each reaction were analyzed by gel electrophoresis nd autoradiography, and the remaining 40 ul by binding to streptavidin agarose. 200 ul packed volume streptavidin agarose was used per reaction. Binding and washing was as described in Example 5 below, except that, after binding, the pellet was rinsed 3 times for 30 minutes at room temperature and once for 60 minutes at 65^βC. The final pellet and all supernatants were counted. The final cpm bound to agarose are 55.6% (reaction 1), 62.1% (reaction 2), 3.2% (reaction 3), 1.0% (reaction 4). These results show that displacement and capturing are approximately equally effective for covalent and non-covalent complexes. Gel analysis of the same reactions, as well as two reactions in which only displacement complex and capturer were included, demonstrated that 100% of the complexes were displaced by target, and when capturer was included, 100% capturing occurred. In the absence of target, no capturing was observed. In addition, since complexes which have hybridized both to capturer and target migrate differently from capturers which have hybridized only to label strand displaced from nonσovalent complexes, label capture intermediates ("second intermediates") which contain capturer hybridized to target which in turn is hybridized to the TBR of the displacement complex, can be distinguished from those non-covalent complexes which have resolved to contain only capturer and displaced label strand. In this experiment, approximately 90% of the captured signal is present in the resolved form, despite the fact that capture DNA was present in excess over target and complex, and would be likely to form the intermediate structure before displacement.

EXAMPLE 7 Release of captured strands by biotin displacement

Biotinylated nucleic acid strands immobilized by the binding of biotin on the nucleic acid to avidin on a support may be released by displacement with free biotin.

A 33bp DNA segment with the sequences

5*CGAAGCTTGGATCCGCGATCCGTCGACCTGCAG 3 * 3'GCTTCGAACCTAGGCGCTAGGCAGCTGGACGTC 5* was cloned into the Smal site of the M13 phage vector mpll to produce phage RM11. A 225 base 32P labelled single stranded DNA fragment biotinylated at its 5• end was prepared by primer extension using a biotinylated primer, RM16, and the single stranded template DNA from phage RM11. The sequence of RM16 is 5'CGAAGCTTGGATCCGC 3^» and the biotin was attached as follows:

12-Hydroxydodeσanoic acid (10.82g, 50mmol) was added to 3% HCl in CH₃OH (210ml) . The reaction was stirred at 22^βC for 3.5 hours and concentrated on a rotary evaporator. The residue was taken up in ethyl acetate (200ml) and washed with IM K₂C0₃ (2 x 150ml) , once with saturated aqueous NaCl (150ml), dried over Na₂S0₄, concentrated and vacuum distilled to yield 10.3g of methyl 12 hydroxydo- decanoate in 90% yield; mp 33.5-35.0., NMR (CDCI₃) delta 3.66 (s, 3H) , 3.62 (t, 2H) , 2.88 (t, 2H) , 1.28 (br s, 18H) .

To a rapidly stirred solution of diisopropylethylamine (3.4ml, 19.5mmol) andchlorodiisopropylaminomethoxyphosphine (2ml, 9.75mm.ol) in CH₂C1₂ (15ml) was added dropwise a solution of methyl 12-hydroxydodecanoate (1.73g, 7.5mmol) in CH₂C1₂ (7.5 ml) over 15min. at 22'C. The reaction was stirred for an additional 10 minutes, diluted with ethyl ether (100ml) and extracted with saturated aqueous NaCl (with 5% NaHC0₃, 4x75ml) . The ethyl ether layer was dried over Na₂S0₄, filtered, concentrated and distilled under vacuum to yield 1.4g of 12-methylσarboxydodecylmethyl-N,N- diisopropyl phosphoramidite in 60% yield; NMR (CDCI₃) delta 3.65 (s, 3H) , 3.39 (d, 3H) , 2.32 (t,2H), 1.22 (br s, 18H) , 1.21 (d, 12H) ; ³¹P NMR (CH₃CN/C₆D6) delta 148.9 with a minor (approximately 5%) contaminant at delta 10.7.

The fully protected RM16 was prepared on a 1.0 or more commonly a 7.5 umole scale. With the automated procedure, a 0.2 M solution of the phosphoramidite in dry CH₂C1₂ was loaded on the synthesizer and the program modified for the addition of this reagent after the detritylation of the last base. In the manual procedure, the support was removed from the synthesis column, dried in vacuo and treated as described below.

Acid 16mer; d[H0₂C(CH₂)_X1 OpCGAAGCTTGGATCCGC], was prepared by a combination of automated and manual solid phase methodologies (see, e.g., Oligonucleotide Synthesis: a Practical Approach, Gait, M.J. ed., IRL Press 1984) on either an Applied Biosystems Model 380A or Beckman Instruments System 1 DNA synthesizer.

After the fully protected acid 16 mer attached to controlled pore glass (8.5 umol) was removed from the columns, it was treated with thiophenol: triethylamine: dioxane (1:2:2, 12 ml, 90 minutes), and then 12.5% trimethylamine: in water (12 ml) for 48 hours at 37^βC. Evaporation and standard aqueous ammonia treatment afforded the crude acid 16-mer (713 A₂go units) . Preparative HPLC purification gave pure acid 16-mer (130 A₂gø units, 812 nmol? 9.6% yield based on starting 3'-nucleoside attached to the solid support) .

Acid 16-mer (3.0 A₂g_Q units, 18 nmol), l-ethyl-3- (3-dimethylaminopropyl) carbodii ide HCl (900 nmol, pH4) and biotin hydrazide (500 nmol) were combined (50 ul) and heated at 37"C with shaking for 1 hour. The reaction mixture was chromatographed on a G-50 Sephadex column to afford 2.9 A₂go units of the acyl-biotinylated RM 16-mer (98%) . HPLC analysis of the crude product showed that it was greater than 90% pure and contained a small amount of N-acyl urea 16-mer, an adduct of DEC with the acid 16-mer.

The RM16 primer was annealed to RM11 template DNA and extended in the presence of excess dCTP,^"dGTP and dTTP and 25uCi [alpha-32P]-dATP withKlenow fragment of DNApolymerase for 15 minutes. Excess unlabeled dATP was then added and the reaction was incubated for another 15 minutes. After digestion with the restriction enzyme Avail, the 225 base fragment was purified after electrophoresis on an alkaline agarose gel. This fragment is referred to as labeled 225- 45 mer.

Biotinylated labeled 225-mer was bound to avidin in solution at room temperature. Buffer BB+ (Example 8) or excess free biotin in BB+ buffer was then added and the samples were incubated at 65 C for various periods of time. At each time point, the samples were rapidly chilled to 4°C. At the completion of the experiment, all samples were electrophoresed on a 5% non-denaturing acrylamide gel. When the bromophenol blue dye is run far enough into gel, the biotinylated labeled 225-mer separates well (approximately 1cm) from the same biotinylated labeled 225-mer bound to avidin. At time 0, all biotinylated labeled 225-mer was bound to avidin. In the absence of free biotin at 65C, there was no change in the mobility of the biotinylated, labeled 225-mer at all time points (5,10,30,60,120 minutes) indicating that under these conditions, no separation of biotin from avidin could be demonstrated. The same result was observed if excess free biotin was added to the samples at the time of chilling to 4C. However, when free biotin was present during the incubation at 65C, after 5 minutes more than 80% of the biotinylated labeled 225-mer migrated at the position of DNA not bound to avidin. After 10 minutes more than 95% of the biotinylated labeled 225-mer had been released from the avidin and by 30 minutes no biotinylated labeled 225- mer still bound to avidin could be detected. This result indicates that free excess biotin can effectively displace biotinylated DNA from an avidin-DNA complex in solution. Control experiments demonstrated that addition of excess free biotin to the avidin before labeled DNA was added completely blocked the avidin-DNA interaction. Addition of free biotin to the avidin-DNA complex and incubation at room temperature (22C) for 10 minutes resulted in partial displacement (10%) . EXAMPLE 8 Quantitative displacement kinetics

Streptavidin-agarose (BRL) or streptavidin-latex (Pandex Laboratories) were washed 4X with BB+ (0.2M NaCl, 10 mM Tris-HCl, pH 8.0, 0.01% NP-40) and then resuspended in 1 ml of BB- To each sample was added approximately 40,000 cpm of labeled, biotinylated 225-mer as described above. Following binding for 10 minutes at room temperature, the agarose or latex bound DNA was separated from unbound DNA by centifugation and washing in BB+. The washes included 3 room temperature washes and two washes for 10 minutes each at 65°C. Approximately 75% of the counts bound to the support under these conditions. Each avidin support-DNA complex was then aliquoted into 9 equal reactions and either 800 ul BB (without NP40) or 800 ulBB (without NP-40) containing 1 M free biotin were added to each tube. Following mixing, the samples were placed at 65"C for 1, 3, 10, or 30 minutes. At each time point, the sample was removed from the 65^βC bath, centrifuged immediately to separate the support from the solution and the supernatant was removed. The pellet was washed twice more with 1 ml BB+ and the supernatants were combined. The agarose or latex pellet was then resuspended in 3 ml BB+ and the pellets and combined supernatants were then counted by Cerenkov counting. The percent of displaced counts (supernatant/[supernatant + pellet]) was then determined for each time point. Similar experiments were also carried out involving displacements at 45°C and 22"C for longer periods of time (2, 10, 28.5, 45 and 150 minutes). Displacement from the agarose support in the absence of free biotin ranged from 0.5% to about 2% at 22^βC, 45°C or 65°C for time points up to 150 minutes. The addition of free biotin resulted in the displacement of up to about 85% of the 255 mer at 65 "C in 30 minutes. At 45°C, about 29% displacement was observed in 30 minutes, with this value increasing to about 70% in 150 minutes. At 22°C, approximately 19% of the 225 mer was released after 150 minutes. Background displacement from the latex support was higher, with about 12% displacement in 30 minutes at 65"C. Addition of free biotin to the latex band sample resulted in displacement of about 80% of the 225 mer in 30 minutes at 65."C.

EXAMPLE 9 Background reduction using biotin displacement

M13 DNA (1 ug) was annealed to the M13 Hybridization probe primer (New England Biolabs, 50 pmoles) and extended as described above except that the synthesized DNA was not digested with a restriction endonuclease, and was purified by G-50 spin column chromatography. Approximately 80 x 10⁶ cpm of labelled DNA were synthesized and used as probe molecules.

200 ul of avidin agarose from BRL was washed in BB+ and transferred to a small disposable column with a filter containing pores small enough to prevent the passage of agarose beads. 80 x 10^β cpm of the labeled probe were then passed through this column and reloaded twice. The eluate was collected. The column was then washed successively with 5 ml of BB+ at 65°C, for approximately 3 minutes each wash. Each wash was collected separately. After 7 washes, 5 ml of BB+ was added and the elution continued for 5 minutes. This sample was then collected and 5ml of BB+ including 1 mM free biotin was added and the incubation continued for an additonal 5 minutes. The counts which are eluted during the plus or minus biotin elutions represent the background which could be expected using this approach.

After these extensive washes, 0.00314% of the total counts remained bound to the column, representingnonspecific background on the support. Only 0.000236% or 0.000229% of the total counts were released into solution during the 5 minute incubations with and without biotin, respectively. Thus, in this experiment, about a 14X decrease in background signal was observed with biotin displacement.

EXAMPLE 10 Strand displacement, capturing and biotin displacement.

The non-covalent p66d Bam HI/EcoRI cut displacement complex described in Example 6 was used. Hae III cut mpll.AllAlb DNA was used as the target. A 300 base capture which has a single biotin at the 5' end was synthesized by primer extension of a 5' biotinylated M13 sequencing primer hybridized to^" mpl9.AlbTaqPst as follows:

20 ug of mpl9.AlbTaqPst template, and 20 pm of biotinylated primer in 100 ul of 50 mM NaCl, 10 mM Tris HCl, pH8.0, 10 mM MgC12 were boiled for 1 minute in a water bath and allowed to cool to room temperature for 30 minutes in the bath. 3 ul of 5 mM dGTP, dCTP, dATP, dTTP and 2 ul of Klenow fragment Pol I were added, and the reaction incubated 30 minutes at room temperature. A second 1 ul aliquot of Klenow was added and the reaction incubated for 30 additional minutes. The DNA was then digested with Hind III for 2 hours at 37°C to cut out the 300 base primer extended fragment which is complementary to the insert in mpl9.AlbTaqPst. The DNA was denatured by adding 2 ul of 5 M NaOH and incubating it for 10 minutes at 65^βC. The primer extended fragment was purified after separation by electrophoresis on a 1% alkaline agarose gel with NA45 paper (Schleicher and Schuell) . DNA yield was estimated by comparison of an aliquot of the capturer with standards on an ethidium stained gel.

Three reactions were set up containing 0.2 pm non- covalent complex alone (reaction 3), with 0.5 pm capturer (reaction 2) and with capturer and 0.2 pm target (reaction 1) . Reactions 1 and 2 were incubated for 60 minutes in 50 ul of hybridization buffer at 65^βC (reaction 3 was not 49 incubated) . 5 ul of reactions 1 and 2 were then removed for gel analysis. The rest of reactions 1 and 2, and reaction 3 were then bound as described in Example 6 to approximately 200 ul packed volume streptavidin agarose. In order to try and minimize sample agitation during the rinses, which may be responsible for a portion of the captured DNA separating from the support, the samples were rinsed by adding 1 ml of binding buffer to the pellet in an eppendorf tube, inverting the tube five times, and centifuging it for 3 minutes. Five rinses were done at room temperature, and a final rinse was done at 65^βC for 30 minutes with no shaking after the initial 5 inversons. All supernatants and the final pellet were counted. The % cpm band in reaction 1 was 43.6%, in reaction 2, 0.9% and in reaction 3, 2.3%. The gentler washing protocol, or the use of the smaller and singly biotinylated capturer promotes more stable binding of the captured complex to the support. By cutting out and counting the appropriate bands from the gel analysis, it appears that approximately 36% of the captured complexes resolve to form capture-signal strand hybrids, while 64% are apparently present as analyte-complex-capturer intermediates under these reaction conditions.

To illustrate biotin displacement of captured displaced strands, reactions 1 to 3 as described above were used. One ml BB containing 0.1% NP-40 (BB+) was added to each pellet at room temperatue, shaken briefly by inversion and centrifuged to separate the phases (RT wash) . One ml BB+, (at 65°C) was then added and the samples were incubated for 5 minutes at 65^βC. The samples were then centrifuged to separate the phases and washed twice with one ml BB+ at room temperature. The combined supernatant phases were then pooled (65^βC/5 minutes/-bio) . The -biotin, 65"C wash was repeated once more for sample 3 only. One ml of BB+ containing 1 mM biotin was then added to all samples and these were incubated for 5 minutes, centrifuged and washed as above (65°C/5 minutes/+bio) . The final pellet was resuspended in 3 ml BB+ and all samples were counted by Cerenkov counting. The percentage of total counts recovered in each sample (after 30 background cpm were subtracted) were for reaction 1, 85%; for reaction 2, 54%, and for reaction 3, 6%.

The signal to noise ratio before the biotin displacement (i.e., the ratio of counts in the reaction 1 pellet/reaction 2 pellet) was 48:1. The improvement brought about by the biotin displacement can be measured by the ratio of % counts released by biotin in fraction 1 over the percent counts released by biotin in either reaction 2 or 3. Thus by comparison with reaction 2 the improvement is 1.55. For reaction 3 the improvement is 14 X. The poor improvement seen in the reaction 2 sample is likely due to the fact that the reaction 2 capturer contains a small amount of M13 polylinker sequence which does result in some capturing by the biotinylated capturer in the absence of target. This capturing, though small, would lead to counts released by biotin. The reaction 3 sample (complex only) represents the type of background most likely to be found in an actual displacement measurement and therefore gives a better representation of the background improvement expected.

Numerous modifications may be made by one skilled in the art to the^' methods and components of the present invention in view of the disclosure herein. Such modifications are believed to be encompassed in the appended claims.

Claims

WHAT IS CLAIMED IS:

1. A method for determining the presence and/or amount of a predetermined target nucleotide sequence in the nucleic acid of a biological sample which comprises the steps:

(a) providing a reagent complex of (i) labeled probe polynucleotide having a target binding region which is capable of base pair binding via hydrogen bonds of purine/pyrimidine bases to said target sequence, and (ii) a second polynucleotide which is bound by base pair binding via hydrogen bonds of purine/pyrimidine base pairs to said labeled probe polynucleotide in a second polynucleotide binding region of said labeled probe polynucleotide at least partially coextensive with said target binding region;

(b) contacting said reagent complexwith a sample under conditions in which the target nucleotide sequence, if present, binds to the labeled probe polynucleotide and displaces second polynucleotide from the labeled probe polynucleotide;

(c) separating labeled probe polynucleotides from which second polynucleotide has been displaced from intact reagent complexes; and

(d) determining the presence and/or amount of labeled probe polynucleotides which have been separated.

2. The method according to claim 1, wherein said second polynucleotide binding region is wholly within the probe target binding region.

3. The method according to claim 2, wherein the second polynucleotide binding region is partially within and adjacent to one end of the target binding region.

4. The method according to claim 3, wherein the second polynucleotide binding region is at least about 15 to about 1000 nucleotides in length. 5. The method according to claim 4, wherein the portion of the target binding region excluding the second polynucleotide binding region is at least about 100 nucleotides in length, and is at least as large as the second polynucleotide binding region.

6. The method according to claim 4, wherein the portion of the target binding region excluding the second polynucleotide binding region is at least about 50 nucleotides in length.

7. The method according to claim 1, wherein the labeled probe polynucleotide further comprises a target binding region containing at least a 15 nucleotide portion of a second polynucleotide binding region bound by purine/pyrimidinebase pairing to the second polynucleotide; and no more than about 15 nucleotides of the second polynucleotide binding region being outside the target binding region.

8. Themethod according to claim 1, whereinthe second polynucleotide is immobilized to a solid support in the complex.

9. The method according to claim8, whereinthe second polynucleotide is covalently linked to the solid support in the complex.

10. The method according to claim 8, wherein the separating step (c) comprises separating a liquid phase comprising labeled probe polynucleotide bound to target nucleotide sequence from a solid phase comprising intact reagent complex. 11. The method according to claim 1, wherein the second polynucleotide and labeled probe polynucleotide form one single polynucleotide strand, or more than one strand joined by linkage stabled to hybridization.

12. The method according to claim 11, wherein the portion of the second polynucleotide which is adjacent to one end of the nucleic acid strand is bound by complementary base pair binding in the complex to the portion of the labeled probe polynucleotide which is adjacent to the other end of said single polynucleotide strand.

13. The method according to claim 12, wherein a detectable tag is on said other end of the nucleic acid strand.

14. The method according to claim 1, wherein the reagent complex is in a solution phase during the contacting step.

15. The method according to claim 14, wherein the separating step (c) comprises contacting the reaction mixture with an immobilized affinity reagent for an affinity moiety contained in the second polynucleotide to immobilize intact reagent complex, and separating a liquid phase containing labeled probe polynucleotide from which second polynucleotides have been displaced.

16. The method according to claim 14, wherein the intact reagent complexes are immobilized after the contacting step.

17. A diagnostic reagent for determining the presence of a predetermined target nucleotide sequence in the nucleic acid of a biological sample comprising: 54

(i) a labeled probe polynucleotide comprising a target binding region which is capable of base pair binding via hydrogen bonds of purine/pyrimidine base pairs to the target nucleotide sequence, and

(ii) an immobilized or immobilizable second polynucleotide which is bound by base pair binding via hydrogen bonds of purine/pyrimidine base pairs to the labeled probe polynucleotide in a region of the labeled probe polynucleotide at least partially coextensive with the target binding region; the potential base pair binding between the target nucleotide sequence and the labeled probe polynucleotide being capable of displacing the second polynucleotide from the reagent complex.

18. The reagent according to claim 17, wherein the labeled probe polynucleotide further comprises a second polynucleotide binding region bound by purine/pyrimidine base pairing to bases of the second polynucleotide; said second polynucleotide binding region being contained within the target binding region.

19. The reagent according to claim 18, wherein the second polynucleotide binding region is adjacent to one end of the target binding region.

20. The reagent according to claim 19, wherein the second polynucleotide binding region is within a range of at least about 15 nucleotides in length to about 1000 nucleotides in length.

21. The reagent according to claim 20, wherein the portion of the target binding region excluding the second polynucleotide binding region is at least about 50 nucleotides in length. 22. The reagent according to claim 18, wherein the second polynucleotide is attached to a solid support adjacent to one end of the second polynucleotide and the portion of the second polynucleotide bound to the target binding region of the labeled probe polynucleotide is adjacent to the opposite end of the second polynucleotide.

23. The reagent according to claim 17, wherein the second polynucleotide contains an affinity moiety capable of binding to an immobilized affinity reagent to immobilize intact reagent complexes.

24. A reagent complex construct for determining the presence of a predetermined target nucleotide sequence in the nucleic acid of a biological sample which comprises:

(a) a polynucleotide strand having:

(i) a target binding region segment which is capable of base pair binding via hydrogen bonds of purine/pyrimidine bases to the target nucleotide sequence, and

(ii) a pairing segment which is bound by base pair binding via hydrogen bonds of purine/pyrmidine base pairs to a portion of the target binding region segment; and

(b) a detectable tag which is within or adjacent to the target binding region segment.

25. The reagent according to claim 24, further comprising

(c) an immobilized or immobilizable polynucleotide strand having a binding segment bound to the target binding region in a region thereof which is not bound to said pairing segment, which immobilized polynucleotide is capable of displacement by the target nucleotide sequence. 26. A method for determining the presence and/or amount of a predetermined target nucleotide sequence in the nucleic acid of a biological sample which comprises the steps of: a) providing a reagent complex construct which comprises (i) a target binding region which is capable of complementary base pair binding to the target sequence stably joined to (ii) a pairing segment which is bound by complementary base pair binding to a portion of the target binding region, and (iii) a detectable tag which is attached within or adjacent to the target binding region; b) contacting the sample with the reagent complex construct and with a capture polynucleotide under conditions in which the target sequence displaces pairing segment from ^" target binding region, and in which the capture polynucleotide binds by complementary base pair binding selectively to the displaced pairing segment; c) separating polynucleotides which have bound to capture polynucleotides from unbound polynucleotides; and d) detecting the detectable tag which is located on polynucleotides which have bound to capture polynucleotides and have been separated.

27. The method according to claim 26, wherein the capture polynucleotide is immobilizable or immobilized.

28. The method according to claim 26, wherein the pairing segment comprises at least a portion of a sequence of nucleotides capable of binding to the capture polynucleotide.

29. A kit for determining the presence of a predetermined target nucleotide sequence in the nucleic acid of a biological sample which comprises: a) a reagent complex construct comprising (i) a TBR capable of complementary base pair binding to the TNS and (ii) a pairing segment bound by complementary base pair binding to a portion of the TBR and (ii) a detectable tag attached within or adjacent to the TBR; b) a capture polynucleotide capable of complementary base pair binding to the pairing segment selectively when the pairing segment is displaced from the target binding region segment; and c) means for separating capturepolynucleotide, together with attached polynucleotide strand, from unbound polynucleotides.

30. The kit according to claim 29, wherein the capture polynucleotide comprises an affinity moiety, and the separating means comprises an immobilized affinity reagent for the moiety.

31. A method for determining the presence or amount of a predetermined target nucleotide sequence in the nucleic acid of a biological sample which comprises the steps:

(a) providing a reagent complex comprising:

(i) a probe polynucleotide having a target binding region capable of complementary base pair binding to said target sequence and (ii) a labeled polynucleotide having a pairing segment bound to a region of the probe polynucleotide at least partially coextensive with the target binding region;

(b) contacting the complex with a sample and with a capture polynucleotide under conditions in which the target sequence binds to the probe polynucleotide and displaces labeled polynucleotide from the complex, and in which the capture polynucleotide binds selectively to at least a portion of the pairing segment of the displaced labeled polynucleotide;

(σ) separating the bound displaced labeled polynucleotide/capture polynucleotide entity from labeled polynucleotidewhichhas notboundto capturepolynucleotide; and

(d) detecting the entity of step (c) .

32. The method according to claim 31, wherein said labeled polynucleotide pairing segment comprises at least a portion of the region to which the capture polynucleotide binds.

33. The method according to claim 31, wherein said probe polynucleotide and labeled polynucleotide of the reagent complex form one single, polynucleotide strand, or more than one strand joined by a linkage stable to hybridization and capture conditions.

34. The method according to claim 33 wherein the label is in or adjacent to the target binding region.

35. The method according to claim 31, wherein the capture polynucleotide is immobilizable or immobilized. 36. The method according to claim 35, wherein the separating step (c) further comprises immobilizing the capture polyn σleotide/labeled polynucleotide entity on an immobilized affinity reagent capable of binding to an affinity moiety associated with said capture polynucleotide.

37. The method according to claim 36, wherein the affinity moiety is biotin.

38. The method according to claim 36, further comprising detecting the labeled polynucleotide in said immobilized entity on a solid phase or after release into a liquid phase.

39. The method according to claim 38 wherein said release results from the dissociation of the affinity moiety from the affinity reagent or the displaced labeled polynucleotide from the capture polynucleotide.

40. The method according to claim 39, wherein said release comprises redisplacing said labeled polynucleotide from the immobilized capture polynucleotide by contact with a selected polynucleotide comprising a segment complementary to said capture polynucleotide or a segment complementary- to a portion of the labeled polynucleotide, said capture polynucleotide including an initial- binding region specific for the selected polynucleotide or said labeled polynucleotide including an initial binding region specific for the selected polynucleotide.

41. The method according to claim 35, further comprising admixing the capture polynucleotide with reaction mixture after the sample has been incubated with the reagent complex.

42. The method according to claim 31, whereintheprobe polynucleotide of the reagent complex is immobilized.

43. The method according to claim 31, wherein said reagent complex is in solution.

44. The method according to claim 43, further comprising immobilizing the probe polynucleotide with attached, undisplacedlabeledpolynucleotideaftercontacting step (b) .

45. The method according to claim 43, further comprising immobilizing capture polynucleotide with bound, displaced labeled polynucleotide during or after contacting step (b) .

46. The method according to claim 45, wherein said contacting step further comprises simultaneously contacting reagent complex with sample and capture polynucleotide.

47. The method according to claim 45, wherein the contacting step further comprises contacting reagent complex with sample; allowing hybridization to occur; and subsequently contacting the reaction mixture with capture polynucleotide.

48. The method according to claim 31, further comprising contacting the sample simultaneously with capture polynucleotide and reagent complex.

49. A kit for determining the presence and amount of a predetermined target nucleotide sequence in the nucleic acid of a biological sample which comprises:

(a) a reagent complex comprising (1) a probe polynucleotide having a target binding region capable of complementary base pair binding to the target nucleotide sequence; and (2) a labeled polynucleotide having a pairing segment bound by complementary base pair binding to a region of the probe polynucleotide which is at least partially coextensive with the target binding region;

(b) a capture polynucleotide having a second binding region capable of complementary base pair binding selectively to a first sequence of displaced labeled polynucleotide substantially within the pairing segment; and

(σ) means for separating the capture polynucleotide and bound displaced labeled polynucleotide from unbound labeled polynucleotide.

50. The kit according to claim 49, wherein said capture polynucleotide is immobilized or immobilizable.

51. The kit according to claim 49, wherein the labeled polynucleotide and probe polynucleotide form one single polynucleotide strand or more than one strand joined by a linkage stable to the hybridization and capture conditions.

52. The kit according to claim 51, wherein the linkage is a phosphodiester linkage.

53. The kit according to claim 49, wherein the probe polynucleotide is immobilized.

54. The kit according to claim 49, wherein the reagent complex is in solution or in soluble form.

55. The kit according to claim 51, wherein the label is in or adjacent to the target binding region.

56. A method for releasing a diagnostic reagent bound to a support by a biotin-avidin or biotin-streptavidin linkage comprising: incubating said bound diagnostic agent in the presence of a concentration of free biotin in excess over the concentration of avidin or streptavidin present in said bound reagent support at temperatures of between 22 to 65 degrees centigrade.

57. The method according to claim 56 wherein said biotin concentration is about ImM or greater.

58. The method according to claim 56 wherein said reagent or support is singly biotinylated.

59. The method according to claim 56 wherein said reagent is selected from the group consisting of single or double stranded DNA or RNA, oligo- orpolynucleotide, enzyme, cofactor, antigen, antibody, signalling moiety, and derivatives and modifications thereof.

60. The method according to claim 56 wherein said support comprises a second diagnostic reagent selected from the group consisting of single or double stranded DNA or RNA, oligo- or polynucleotide, enzyme, cofactor, antigen, antibody, signalling moiety, and derivatives and modifications thereof. ₆₁. The method according to claim 56 wherein said support is selected from the group consisting of latex, agarose, filter, membrane, natural, synthetic or semi- synthetic polymers.

62. The method according to claim 56 wherein said reagent is a double stranded polynucleotide, which remains undenatured by said disrupting step.

63. The method according to claim 56 wherein said reagent is a biotinylated nucleic acid.

64. The method according to claim 56 wherein said support is an avidin or streptavidin-associated nucleic acid.

65. The method according to claim 56, wherein said diagnostic reagent is as biotinylated capture strand.

[received by the International Bureau on 29 May 1987 (29.05.87); original claims 1, 17 amended; claims 56-65 replaced by amended claims 56-63 (5 pages)]

1. A method for determining the presence and/or amount of a predetermined target nucleotide sequence in the nucleic acid of a biological sample which comprises the steps:

(a) providing a reagent complex containing (i) labeled probe polynucleotide having a target binding region which is capable of base pair binding via hydrogen bonds of purine/ pyrimidine bases to said target sequence, and (ii) a second polynucleotide which is bound by base pair binding via hydrogen bonds of purine/pyrimidine base pairs to said labeled probe polynucleotide in a second polynucleotide binding region of said labeled probe polynucleotide at least partially coextensive with said target binding region;

(b) contacting said reagent complex with a sample under conditions in which the target nucleotide sequence, if present, initially binds to the labeled probe polynucleotide in a portion of said target binding region adjacent to said second polynucleotide binding region and displaces second polynucleotide from the labeled probe polynucleotide;

(c) separating labeled probe polynucleotides from which second polynucleotide has been displaced from intact reagent complexes; and

(d) determining the presence and/or amount of labeled probe polynucleotides which have been separated.

2. The method according to claim 1, wherein said second polynucleotide binding region is wholly within the probe target binding region.

3. The method according to claim 2, wherein the second polynucleotide binding region is partially within and adjacent to one end of the target binding region.

4. The method according to claim 3 , wherein the second polynucleotide binding region is at least about 15 to about 1000 nucleotides in length. ₁₁. _The method according to claim 1, wherein the second po_lynuc_leotide and labeled probe polynucleotide form one single polynucleotide strand, or more than one strand joined by linkage stabled to hybridization.

12. The method according to claim 11, wherein the portion of the second polynucleotide which is adjacent to one end of the nucleic acid strand is bound by complementary base pair binding in the complex to the portion of the labeled probe polynucleotide which is adjacent to the other end of said single polynucleotide strand.

13. The method according to claim 12 , wherein a detectable tag is on said other end of the nucleic acid strand.

14. The method according to claim 1 , wherein the reagent complex is in a solution phase during the contacting step.

15. The method according to claim 14 , wherein the separating step (σ) comprises contacting the reaction mixture with an immobilized affinity reagent for an affinity moiety contained in the second polynucleotide to immobilize intact reagent complex, and separating a liquid phase containing l ab el ed p rob e p o l ynucl e o t ide f r om which s econd polynucleotides have been displaced.

16. The method according to claim 14 , wherein the intact reagent complexes are immobilized after the contacting step.

17. A diagnostic reagent for determining the presence of a predetermined target nucleotide sequence in the nucleic acid of a biological sample comprising: (i) a labeled probe polynucleotide comprising a target binding region which is capable of base pair binding via hydrogen bonds of purine/pyrimidine base pairs to the target nucleotide sequence, said target binding region having an initial binding region to which said target sequence is capable of initially binding prior to a displacement reaction; and

(ii) an immobilized or immobilizable second polynucleotide which is bound by base pair binding via hydrogen bonds of purine/pyrimidine base pairs to the labeled probe polynucleotide in a region of the labeled probe polynucleotide at least partially coextensive with the target binding region; the potential base pair binding between the target nucleotide sequence and the labeled probe polynucleotide being capable of displacing the second polynucleotide from the reagent complex.

18. The reagent according to claim 17, wherein the labeled probe polynucleotide further comprises a second polynucleotide binding region bound by purine/pyrimidine base pairing to bases of the second polynucleotide; said second polynucleotide binding region being contained within the target binding region.

19. The reagent according to claim 18, wherein the second polynucleotide binding region is adjacent to one end of the target binding region.

20. The reagent according to claim 19, wherein the second polynucleotide binding region is within a range of at least about 15 nucleotides in length to about 1000 nucleotides in length.

21. The reagent according to claim 20, wherein the portion of the target binding region excluding the second polynucleotide binding region is at least about 50 nucleotides in length. substantially within the pairing segment; and

(c) means for separating the capture polynucleotide and bound displaced labeled polynucleotide from unbound labeled polynucleotide.

50. The kit according to claim 49, wherein said capture polynucleotide is immobilized or immobilizable.

51. The kit according to claim 49, wherein the labeled polynucleotide and probe polynucleotide form one single polynucleotide strand or more than one strand joined by a linkage stable to the hybridization and capture conditions.

52. The kit according to claim 51, wherein the linkage is a phosphodiester linkage.

53. The kit according to claim 49, wherein the probe polynucleotide is immobilized.

54.' The kit according to claim 49, wherein the reagent complex is in solution or in soluble form.

55. The kit according to claim 51, wherein the label is in or adjacent to the target binding region.

56. A method for releasing a nucleic acid containing diagnostic reagent bound to a support by a biotin-avidin or biotin-streptavidin linkage comprising: incubating said bound diagnostic agent in the presence of a concentration of at least about ImM free biotin which is in excess over the concentration of avidin or streptavidin present in said bound reagent support at temperatures of between 22 to 65 degrees centigrade. 57. The method according to claim 56 wherein said reagent or support is singly biotinylated.

58. The method according to claim 56 wherein said reagent is selected from the group consisting of single or double stranded DNA or RNA, oligo- or polynucleotide, enzyme, cofactor, antigen, antibody, signalling moiety, and derivatives and modifications thereof.

59. The method according to claim 56 wherein said support comprises a second diagnostic reagent selected from the group consisting of single or double stranded DNA or RNA, oligo- or polynucleotide, enzyme, cofactor, antigen, antibody, signalling moiety, and derivatives and modifications thereof.

60. The method according to claim 56 wherein said support is selected from the group consisting of latex, agarose, filter, membrane, natural, synthetic or semi- synthetic polymers.

61. The method according to claim 56 wherein said reagent is a double stranded polynucleotide, which remains undenatured by said disrupting step.

62. The method according to claim 56 wherein said support is an avidin or streptavidin-associated nucleic acid.

63. The method according to claim 56, wherein said diagnostic reagent is a biotinylated capture strand.