WO2000028084A1 - Isothermal nucleic acid amplification methods - Google Patents

Abstract

Methods are provided for the rapid, substantially isothermal, amplification of the sequence information of a target nucleic acid positioned within a single- or double-stranded polynucleotide fragment. The methods are based on the serial generation of double-stranded DNA engineered to contain terminal nicking sites, nicking of at least one of those sites, and extensions from the nick(s), thereby displacing any existing polynucleotides. A kit combining the components commonly used in practicing methods of the invention is also provided.

Description

ISOTHERMAL NUCLEIC ACID AMPLIFICATION METHODS

FIELD OF THE INVENTION

The present invention generally relates to nucleic acid biochemistry. In particular, the invention is drawn to isothermal nucleic acid amplification technologies.

BACKGROUND OF THE INVENTION

Nucleic acid amplification is required in a wide variety of molecular techniques to provide sufficient substrate for manipulation. A traditional approach to amplifying target nucleic acids involves cloning. By placing a desired nucleic acid into a vector and introducing that recombinant molecule into a host organism, one may ultimately isolate many copies of the target. This approach, however, requires co-amplification of vector sequences, thereby reducing the efficiency of target amplification. Moreover, cloning requires the maintenance, and manipulation, of a host, adding to the time and cost of nucleic acid amplification.

Cell-free amplification systems avoid the time, effort and cost of maintaining and manipulating hosts. The Polymerase Chain Reaction (i.e. , PCR) is a prominent cell-free method for amplifying a target sequence found in a DNA molecule. Saild et al, Science 259/487-491 (1989) and U.S. Patent Nos. 4,683,195, 4,683,202 and 4,800,159. Although the PCR method is cost- effective in only using two primers, PCR is limited to amplifying regions of a target DNA that are defined in terms of sequence. PCR also requires temperature manipulation, creating a costly need for a thermocycling apparatus to achieve practical implementation. Reverse Transcription-Polymerase Chain Reaction (i.e.,

RT-PCR) is a modification of PCR that permits the amplification of a target sequence found in an RNA molecule. Myers et al. , Biochemistry 50:7661-7666 (1991) and U.S. Patent Nos. 5,310,652 and 5,407,800. In RT-PCR, the PCR cycle is preceded by the reverse transcription of an RNA target, thereby generating a cDNA product, which can then be amplified by PCR. Like PCR, RT-PCR only becomes practical with the automation of the requisite temperature changes, using a thermocy cling apparatus.

Another ex vivo method of nucleic acid sequence amplification is the Ligase Chain Reaction (LCR), developed by Wu et al., Genomics 4:560-569 (1989); a thermophilic version was introduced by Barany, PCR Methods and Applications 1:5-16 (1991). LCR employs two pairs of primers. In LCR, one pair of primers anneals to adjacent positions on a target nucleic acid sequence. The other pair of primers anneals to adjacent positions on the complement of the target nucleic acid sequence region that binds the first set of primers. Although the target in LCR is amplified exponentially, the method exhibits low fidelity because of a tendency to produce target- independent Mgations. Moreover, the amplified products consist almost entirely of costly synthetic primers. In addition, LCR, like PCR, requires continual thermocycling and the thermocycling instrumentation that brings practicality to the method.

Another amplification method, Strand Displacement Amplification (i.e., SDA), is described in Walker et al, Proc. Natl. Acad. Sci. (USA) 89:392-396 (1992), Walker et al , Nucl. Acids Res. 20(7):1691-1696 (1992), and U.S. Patent Nos. 5,270,184, 5,422,252, 5,455,166, 5,470,723, 5,536,649, and 5,547,861, as well as European Patent Application No. EP 0 624 623 A2. SDA is an isothermal amplification technique that generates DNA copies of a single- or double-stranded target polynucleotide fragment. The original SDA methodology described in Walker et al. , Proc. Natl. Acad. Sci. (USA) 59:392-396 (1992) and U.S. Patent No. 5,455,166 amplifies the entire sequence of a single- or double-stranded target polynucleotide fragment (i.e. , the target spans the entire fragment containing it). The technique employs two primers with each primer exhibiting a target binding region at its 3' terminus. Disposed toward the 5' terminus of each primer is a single-stranded sequence corresponding to a nickable restriction endonuclease recognition site. These recognition sequences are not necessarily complementary to any sequence in the target polynucleotide fragment, and are designed to overhang the 3' ends of an original double-stranded target polynucleotide fragment, or the 3' ends of an original single-stranded target polynucleotide and its complement. In these positions, the unbound sequences overhanging the 3' ends of the target polynucleotide fragment and primer serve as templates for enzymatic extensions from the free 3' ends of both the primer and its target polynucleotide, respectively. Because these primers must overhang the 3' ends of a polynucleotide containing the target in order to create the required nickable site or sites (which must be double stranded to be functional), the SDA method requires knowledge of the nucleotide sequence at both 3' ends of any double- stranded target polynucleotide fragment to be amplified. For the amplification of any single-stranded target polynucleotide, the nucleotide sequences of the 5' end and the 3 ' end must be known. This is a difficult task if the sample to be analyzed contains a mixture of polynucleotide fragments from a variety of sources and in various stages of degradation or fragmentation, such as found in biological fluids, tissues, foods, and water supplies.

Variations on the basic SDA method are described in U.S. Patent Nos. 5,270,184 and 5,422,252. The '184 patent describes a method of amplification that involves four primers. In addition to the amplification primers (S_ and Sj) that are used in the original SDA method, the method of the '184 patent includes two bumper primers (B_j and B₂). To design the sequences of these four primers, the '184 method requires knowledge of the polynucleotide sequence at each end of an internal target nucleic acid sequence and, additionally, at regions flanking the target sequence. These requirements present obstacles to amplification in terms of the time and effort required to sufficiently characterize the polynucleotide sequence of a targeted nucleic acid such as a diagnostic portion of a pathogenic virus or a disease-linked genetic allele in humans. Additionally, the '184 method suffers from a heightened potential for artifactual and misleading results produced by spurious annealing of members of the required set of four primers.

Another variation on the original SDA method involves "multiplex" amplification as described in the '252 and '723 patents. The method described in these patents, while allowing the simultaneous amplification of more than one target nucleic acid sequence, requires six primers (the two original SDA primers, two bumper primers, and two adapter primers) and a correspondingly greater characterization of the polynucleotide sequence of a target genome to construct the six primers than is demanded by the original SDA method or the variant described in the '184 patent. Further, the greater number of primers in the reaction mixture increases the likelihood of spurious amplification reactions.

A variant on "multiplexing" SDA involves the serial amplification of targets from a single sample, made practical by introducing a decontamination step between each round of SDA. This development of SDA is described in U.S. Patent No. 5,536,649 and European Patent Application No. EP 0 624 623 A2. The SDA methodology remains conventional, requiring [αS] dATP to produce a nickable site, as noted in EP 0 624 623 A2 at page 3. During SDA, however, dUTP is also incorporated into amplicon for the express purpose of producing an amplified polynucleotide susceptible to degradation involving uracil-N-glycosylase. In this manner, the amplification product of one round of SDA may be destroyed prior to subjecting that same sample to another and different SDA reaction (i.e., the serial amplification of two targets in one sample). This variation on the standard SDA reaction, however, involves the costly use of two modified nucleotides- the α-thiolated derivative for SDA, and dUTP, required for decontamination. In a related variation of SDA described in U.S. Patent No. 5,547,861, the combination of an α-thiolated nucleotide and dUTP is used in an SDA reaction designed to produce both primary and secondary target amplifications (i.e. , a target amplification internal to the primary target amplification), for purposes of detecting, monitoring or immobilizing the primary amplicon. Again, the advantage offered by the variation comes at the cost of using two modified nucleotides.

Another isothermal nucleic acid amplification methodology is known as Rapid AMPlification, or RAMP. RAMP is described in International Patent Application No. PCT/US97/04170. RAMP is similar to SDA in the cyclic amplification stage of the reaction (see below), which in both methodologies is characterized by nicking a segregated, hemi-modified, duplex target nucleic acid, followed by polymerase-mediated primer extension and concomitant downstream strand displacement. RAMP, like SDA, must generate a hemi-modified restriction endonuclease recognition site flanking the target sequence. Generating that structure by chemical modification of sites flanking a target is difficult insofar as chemical reactions typically discriminate at the fundamental level of nucleotide structure. Because polynucleotides are typically composed of a small set of nucleotides found throughout its structure, a chemical approach to nucleotide amplification would be expected to lack sufficient specificity to provide the type of hemi-modified polynucleotides amenable to nicking. In contrast, enzymatic approaches to generating these hemi-modified sites are technically feasible, requiring enzyme-mediated incorporation of a modified nucleotide during synthesis of at least one strand ofthe site. Conventional modified nucleotides, such as thiolated nucleotides, alter the phosphodiester backbone of polynucleotides, lowering the rate of strand synthesis. This constraint, in turn, limits both the quantity and length of amplification product being produced. Thus, the convenience of using isothermal amplification methods such as SDA and RAMP is partially compromised by the untoward effect of conventional modified nucleotides, such as thiolated nucleotides, on polynucleotide polymerization, slowing the rate of the process and ultimately limiting the quantity of amplified product.

Therefore, a need continues to exist in the art for versatile isothermal nucleic acid amplification methods involving polymerase-mediated strand syntheses that exhibit improved thermodynamic and kinetic properties, while minimizing cost.

SUMMARY OF THE INVENTION

The present invention provides isothermal nucleic acid amplification methods that rely on modified nucleotides capable of being incorporated into nucleic acid amplification products such that the phosphodiester polynucleotide backbone is preserved. The use of these modified nucleotides (e.g. , dUTP, uracil derivatives of dUTP, 7-deaza dGTP, dlTP, and their corresponding ribonucleotide analogs) in methods according to the invention optimizes the interaction between the modified nucleotide and the two enzyme activities required to practice the invention: a restriction endonuclease and a polynucleotide polymerase. The amplification methods of the invention provide a set of nucleotide substrates particularly suited for the polymerase-mediated phosphodiester bond formation underlying polynucleotide synthesis. At the same time, the modified nucleotide interferes with restriction endonuclease cleavage, thereby promoting nicking at hemi-modified sites. Consequently, the methods of the invention exhibit improved thermodynamics and kinetics resulting in increased sensitivity (i.e. , target quantity required to generate detectable amplification products) and polynucleotide length, thus broadening the both range of target sources (e.g. , forensic samples containing trace quantities of nucleic acids) and the range of target lengths (i.e. , extending the methods to longer targets). Moreover, these advantages are frequently realized by substituting less costly modified nucleotides (e.g. , dUTP) for the more costly modified nucleotides (e.g. , [α-S] dNTPs) used in conventional isothermal amplification methods.

In one aspect of the invention, an isothermal method for amplifying a target nucleic acid is provided that comprises the following steps: providing a reaction mixture comprising a sample containing or suspected of containing a target nucleic acid; first and second amplification primers, wherein each primer has a restriction endonuclease recognition sequence for forming a hemi-modified restriction endonuclease recognition site, the sequence disposed 5' relative to a target binding site; a collection of nucleotides consisting essentially of conventional nucleotides and one or more modified nucleotides capable of phosphodiester bond formation, wherein the nucleotides of the collection are collectively capable of base pairing with each of dGMP, dAMP, IMP, and dCMP; and a polymerase lacking an effective 5' — > 3' exonuclease activity; contacting the reaction mixture with a restriction endonuclease for nicking at least one of two hemi-modified restriction endonuclease recognition sites formed by hybridization of the amplification primers to the target nucleic acid; and incubating the reaction mixture under substantially isothermal conditions, thereby amplifying the target nucleic acid.

Another aspect of the invention provides an isothermal method for amplifying a target nucleic acid comprising the following steps: providing a reaction mixture comprising a sample containing or suspected of containing a target nucleic acid; first and second amplification primers, wherein each primer has a restriction endonuclease recognition sequence for forming a hemi- modified restriction endonuclease recognition site, the sequence disposed 5' relative to a target binding site; a collection of nucleotides comprising conventional nucleotides and one modified nucleotide, wherein the modified nucleotide is capable of phosphodiester bond formation, and further wherein the nucleotides of the collection are collectively capable of base pairing with each of dGMP, dAMP, IMP, and dCMP; and a polymerase lacking an effective 5' — > 3' exonuclease activity; contacting the reaction mixture with a restriction endonuclease for nicking at least one of two hemi-modified restriction endonuclease recognition sites formed by hybridization of the amplification primers to the target nucleic acid; and incubating the reaction mixture under substantially isothermal conditions, thereby amplifying the target nucleic acid. Yet another aspect of the invention is directed to a kit for isothermal nucleic acid amplification comprising first and second amplification primers, each primer having a restriction endonuclease recognition sequence for forming a nickable restriction endonuclease recognition site disposed 5' relative to a target binding sequence; a collection of nucleotides consisting essentially of conventional nucleotides and a single modified nucleotide capable of phosphodiester bond formation, wherein the nucleotides of the collection are collectively capable of base pairing with each of dGMP, dAMP, TMP, and dCMP; a polynucleotide polymerase lacking an effective 5' -- > 3' exonuclease activity; and a restriction endonuclease for nicking at least one of two restriction endonuclease recognition sites.

Numerous other aspects and advantages of the present invention will be apparent upon consideration of the following drawings and detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1A and IB schematically illustrate a method according to the invention.

Figure 2 shows the sensitivity of isothermal nucleic acid amplification reactions using either [α-S] dNTP or dUTP. DETAΠ,ED DESCRIPTION OF THE INVENTION

The present invention provides improved methods for amplifying nucleic acids under essentially isothermal conditions. The improvement involves the use of modified nucleotides that function as enzymatic polymerization substrates capable of forming a phosphodiester polynucleotide backbone, thereby minimizing any interference with the enzyme-catalyzed nucleotide polymerizations underlying amplification, such as results from use of thiolated nucleotides. At the same time, modified nucleotides capable of phosphodiester backbone bond formation do interfere with the double-stranded cleavage activity of restriction endonucleases, thereby promoting the nicking of hemi-modified duplex target nucleic acids.

Conventional SDA and RAMP use modified nucleotides such as thiolated nucleotides to generate nickable target polynucleotides. The thiolated nucleotides generally substitute a sulfur atom for the phosphorus atom of conventional unmodified nucleotides. Consequently, polynucleotide strands containing the thiolated derivatives are held together, in part, by thioether bonds rather than phosphodiester bonds, thereby disrupting the conventional phosphodiester backbone structure of polynucleotides. Such strands, when hybridized to strands having unmodified nucleotides, create hemi-modified duplex polynucleotides that are nickable because a restriction endonuclease is able to cleave the unmodified strand but is unable to cleave the modified strand. However, the required incorporation of a thiolated nucleotide comes at the expense of forcing a polymerase to use an unnatural substrate and catalyze a reaction that differs from the one it catalyzes in vivo (i.e. , formation of a thioether bond rather than a phosphodiester bond).

In contrast, the methods of the present invention use modified nucleotides that retain the capacity to form phosphodiester bonds. In conjunction with the unmodified nucleotides used in the methods, these nucleotides present substrates compatible with polynucleotide synthesis via phosphodiester bond formation, the reaction catalyzed by polymerases in vivo. Nevertheless, these modified nucleotides, while retaining the capacity to form phosphodiester bonds, do alter the structural context of a restriction endonuclease recognition site, thereby inhibiting a restriction endonuclease from catalyzing the double-stranded cleavage of a hemi-modified polynucleotide. Instead of double-stranded cleavage, the hemi-modified duplex target nucleic acids are nicked, a required step in achieving isothermal amplification. Thus, the methods of the invention amplify target nucleic acids, and do so at a higher rate than is achieved using conventional isothermal amplification methods such as SDA or RAMP.

Some embodiments of the present invention are improvements over the SDA methods disclosed in U.S. Patent No. 5,455,166, the disclosure of which is incoiporated herein by reference in its entirety. Other embodiments of the invention are improvements over variant SDA methods, as disclosed in U.S. Patent Nos. 5,270,184, 5,422,252, 5,470,723, 5,536,649, and 5,547,861, as well as European Patent No. EP 0 624 623 A2, each of the disclosures of which are incorporated herein by reference in their entireties. Still other embodiments of the invention constitute improvements over the RAMP methods disclosed in International Patent Application No. PCT/US97/04170, the disclosure of which is also incorporated herein by reference in its entirety.

In general terms, the present invention contemplates improved isothermal nucleic acid amplification methods that involve the incorporation of one or more modified nucleotides capable of forming a phosphodiester polynucleotide backbone. Typically, the modified nucleotide or nucleotides are incorporated by annealing and extending two oligonucleotide primers that have been hybridized to each of the two ends of a nucleic acid target. To hybridize, the primers must be substantially complementary to a portion of either a target sequence or its complement. A primer is substantially complementary if it hybridizes under standard stringent hybridization conditions. Exemplary hybridization conditions involve hybridization at 42° C in 50% formamide, 5X SSC, 20 mM Na»PO₄, pH 6.8 and washing in 0.2x SSC at 55°C. It is understood by those of skill in the art that variation in these conditions occurs based on the length and GC nucleotide content of the sequences to be hybridized. Formulas standard in the art are appropriate for determining exact hybridization conditions. See Sambrook et al., 9.47-9.51. For example, stringent hybridization conditions requiring at least 90 % , and preferably 95 % similarity, can be calculated for a given polynucleotide based on a determination of its T_M, calculated using the following equation: T_M = 81.5°C + 16.6(log₁₀[Na⁺] + 0.41 (fraction G-C) - 0.63 (% formamide) - (600/1), where T_M is the melting temperature of a perfect hybrid, Na⁺ is the sodium ion concentration, fraction G-C is the G-C content, and 1 is the length of the hybrid. Armed with the T_M of a perfect hybrid, the T_Ms of hybrids mismatched to varying degrees can be calculated using the knowledge that the T_M of a duplex DNA decreases by 1-1.5 °C for every 1 % mismatch. Similar considerations apply to RNA-containing hybrids, as well as short ( < 100 bp) DNA hybrids, as is known in the art (see Sambrook et al. §§ 9.47-9.51).

Targets suitable as amplification substrates include single- and double-stranded DNA, RNA, and mixtures or composites thereof. A preferred target is a single-stranded RNA, as illustrated in Example 4 below. Single- stranded RNA targets facilitate the separation of the target from a first nascent strand, a step that is required to segregate the target. Single-stranded RNA targets are amenable to separation using an RNase such as RNase H. Targets used in the methods of the invention also may be asymmetrically located within a larger duplex polynucleotide. This arrangement facilitates target segregation by generating a single-stranded polynucleotide containing the target using an exonuclease to digest the strands of a duplex polynucleotide. For example, Exonuclease HI may be used to processively degrade, in a 3' - > 5' direction, each strand of a duplex DNA containing a target. By locating the target sequence towards the 5' end of one of the two strands of the duplex DNA and controlling the extent of Exonuclease HI degradation, some of the molecules surviving Exonuclease HI treatment will be single-stranded molecules containing the target sequence. Subsequent hybridization of an amplification primer and extension therefrom with a modified nucleotide provides a duplex modified polynucleotide that is amenable to amplification using the inventive methods.

Alternatively, a target-containing duplex DNA fragment may be manipulated such that only the non-target strand is phosphorylated at the 5' end. For example, the manipulation may involve dephosphorylation of both strands of a duplex polynucleotide containing a target, followed by restricting the fragment on one side of the target (i.e. , 5' or 3' of an arbitrarily chosen single strand of the target) to effectively re-introduce one of the 5' phosphates. Subsequently, λ Exonuclease may be used to selectively degrade the 5' phosphorylated strand, yielding a single-stranded target nucleic acid.

The collection of nucleotides used in the methods and kits of the invention includes a modified nucleotide that retains the capacity to form a phosphodiester bond, the bond characteristic of polynucleotide backbones. Preferred modified nucleotides are unconventional nucleotides capable of forming phosphodiester bonds. Examples of preferred modified nucleotides are the deoxyribose triphosphate derivatives of uracil, uracil derivatives such as 5- Br-uracil, 7-deaza-guanine, 7-methyl guanine, the base moieties of Queuosine, Wyosine or Inosine, N°-methyladenine, N°-isopentenyladenine, 3- methylcytosine, 5-methylcytosine, dihydrouracil, pseudouracil, and 4- thiouracil, as well as the ribose analogs of all of these modified nucleotides except for dUTP. In general, the invention contemplates modified nucleotides that are derivatives of conventional unmodified nucleotides. In particular, modified nucleotides include conventional unmodified nucleotides that have been derivatized at their base and/or sugar moieties such that the capacity to form phosphodiester bonds is effectively retained. A preferred modified nucleotide is dUTP.

The collection of nucleotides also includes conventional unmodified nucleotides. As defined herein, conventional unmodified nucleotides are selected from the set of typical DNA nucleotides (i.e. , dGTP, dATP, TTP and dCTP) and RNA nucleotides (i.e. , rGTP, rATP, UTP and rCTP). The selection of conventional unmodified nucleotides is designed to provide, in conjunction with any modified nucleotides, a set of substrates that will enable a polynucleotide polymerase to extend a suitable primer using a target nucleic acid template. Of course, during incorporation into a polynucleotide, the nucleotides undergo a change in phosphorylation state. Thus, preferred collections of nucleotides are triphosphorylated nucleotides, which function as substrates for polynucleotide syntheses. The collection of nucleotides may include four or more types of nucleotides, or members, provided that the collection includes at least one member capable of base pairing (i.e. , complementary) to each of the conventional unmodified nucleotides found in either DNA or RNA. Thus, the nucleotides of the collection are collectively capable of base pairing with each of the conventional unmodified nucleotides that may be found in a target nucleic acid. For example, one collection of nucleotides may have four members, wherein each member uniquely base pairs with one of the four conventional unmodified nucleotides of either DNA or RNA, depending upon the type of target for which the collection of nucleotides was designed. Alternatively, the collection may include five or more nucleotides, wherein more than one nucleotide is capable of base pairing with a given conventional unmodified nucleotide, provided that the collection includes at least one nucleotide capable of base pairing with each one of the four conventional unmodified nucleotides of DNA or RNA. A modified nucleotide may be combined with various conventional unmodified nucleotides to produce collections of nucleotides according to the invention. As noted above, the collection of nucleotides must include at least one member capable of base pairing to each of the conventional unmodified nucleotides of DNA or RNA. Thus, the selection of nucleotides to include in a collection is guided by an understanding of nucleic acid base pairing rules. The pairing rules for conventional unmodified nucleotides are known (dG-dC, dA-T; rG-rC, rA-U). For modified nucleotides, base pairing tendencies can be predicted from an assessment of the hydrogen bonding characteristics associated with the chemical structure of the nucleotide' s base and the type and location of chemical substituents attached thereto. For some modified nucleotides, such as dUTP, the base moiety is found in rUTP, a conventional unmodified nucleotide; the base pairing properties of the two nucleotides are essentially the same. Thus, dUMP base pairs with dAMP or rAMP. The base pairing properties of some other modified nucleotides according to the invention are: 5-Br-dUMP pairs with dAMP or rAMP; 7- deaza-dGMP pairs with dCMP or rCMP; and dIMP pairs with any one of dUMP, dCMP, dAMP, rCMP, or rAMP. Hence, one collection of nucleotides includes dlTP, dATP and dCTP because dIMP base pairs with both dCMP and dAMP, eliminating the need for dGTP and TTP in order to provide a polymerase with the required nucleotide substrates for polynucleotide synthesis. A preferred collection of nucleotides includes dGTP, dATP, dUTP and dCTP. A related collection of nucleotides further includes TTP, such that dUTP and TTP compete for inclusion in the amplification products produced by the methods of the invention. Of course, in the case of this type of collection of nucleotides, routine experiments are performed to optimize the quantities of the competing nucleotides such that nickable restriction endonuclease recognition sites are produced by the methods of the invention. In particular, the relative quantity of modified nucleotide may be reduced to a level such that only some of the relevant restriction endonuclease recognition sites are hemi-modified, provided that enough of these hemi-modified sites exist to support the production of detectable amplicon.

A variety of hemi-modified restriction endonuclease recognition sites are also contemplated by the invention. Preferred sites do not have a nickable strand which could contain a modified nucleotide within that portion ofthe restriction endonuclease recognition site that is downstream (i.e. , 3') of a nicking site, such that subsequent nicking would be inhibited. Suitable restriction sites include, but are not limited to, the sites listed in Table I of Example 5 that show a relative inhibition of cleavage. Preferred restriction sites include those sites capable of accommodating more than one modified nucleotide or accommodating a single modified nucleotide at a position such that cleavage would require destruction of a bond formed by the modified nucleotide. Also preferred are restriction sites that would accommodate two modified nucleotides wherein at least one of those modified nucleotides would form a bond that would not be destroyed by a restriction endonuclease, thereby providing an opportunity for nicking. Beyond the preceding guidance, one of ordinary skill will recognize that the placement of modified nucleotides in positions closely related to a restriction endonuclease recognition site may promote nicking per se, or expand the conditions under which a nickable site may be nicked, and that such placement is guided by the relationship between physical proximity to a site and the effect on the cleavage characteristics of such sites. Placement of modified nucleotides in the vicinity of a restriction endonuclease recognition site, for example within 40 nucleotides of a site or, more particularly, within 10 nucleotides of such a site, and routine assaying for subsequent nickability, are within the skill in the art.

The choice of a restriction endonuclease recognition site for use in practicing the invention may also be influenced by the properties of a cognate restriction endonuclease. Preferred restriction endonucleases are Type π restriction endonucleases that are thermostable, preferably being active at 50 °C and more preferably being active at 65-70° C. Preferred restriction endonucleases will not nick a fully modified site, although enzymes that do nick fully modified sites may be used with targets that lack internal recognition sites, which may be determined using techniques that are routine in the art. The methods of the invention also require a polynucleotide polymerase activity. The polymerase activity may be provided by any template-based polynucleotide polymerase, such as a DNA-dependent DNA polymerase, a DNA-dependent RNA polymerase, or an RNA-dependent DNA polymerase (i.e. , a reverse transcriptase), provided that the polymerase does not contain a 5' — > 3' exonuclease activity effective in degrading existing strands at the expense of their displacement. A preferred polymerase activity is provided by a thermostable DNA-dependent DNA polymerase such as Bst DNA polymerase. Also preferred is a processive DNA-dependent DNA polymerase such as T7 DNA polymerase.

With respect to buffers suitable for use in the methods of the invention, a variety of buffers is expected to be compatible with practice of the invention. A preferred buffer is described in Example 3, although one of ordinary skill in the art would be able to adjust the components of a buffer through routine efforts to optimize the amplification reaction. Similarly, the reaction time, temperature, and other variables associated with the methods of the invention may be varied using no more than routine optimization procedures, as would be understood in the art. With respect to temperature in particular, a preferred temperature is the highest temperature compatible with 1) retention of stable hybrid formation between a portion of a target and the target-binding portion of an amplification primer, and 2) the thermostability of enzymes such as restriction endonucleases, exonucleases, and polymerases.

In general terms, and without placing limitations on the invention as claimed, the isothermal amplification methods of the invention may be understood as operating in the following manner. The segregation/amplification scheme implemented by the methods of the present invention is shown in Figures 1A and IB. The reaction steps schematically shown in Figure 1 A generally correspond to that stage of isothermal nucleic acid amplification methods that is responsible for the segregation of a target flanked by one or more hemi-modified nickable restriction endonuclease recognition sites; the steps shown in Figure IB generally correspond to the cyclic amplification stage of isothermal nucleic acid amplification methods. For purposes of illustration, an embodiment involving a target nucleic acid sequence in a single-stranded polynucleotide 1 is exemplified (Figure 1A). Initially, sequence data deduced from the target nucleic acid sequence (e.g. , RNA or DNA) is used to design two primers, (+) primer 2 and (-) primer 5. Each primer contains, towards its 5' end, a restriction endonuclease recognition sequence for a restriction endonuclease which nicks the double-stranded, hemi-modified DNA produced during segregation and amplification. The 3' end of (+) primer 2 contains a target binding region exhibiting a sequence complementary to the 3 ' end of the target nucleic acid sequence within the polynucleotide 1. The 3' end of (-) primer 5 contains a target binding region exhibiting a sequence which is complementary to the 3' end of the complement of the target nucleic acid sequence within the polynucleotide 1, a sequence that is found in a complementary (+) strand 3. Within the polynucleotide 1, it is the complement of the target binding region of (+) primer 2 and the target binding region of (-) primer 5 that delimit the target nucleic acid sequence because it is the polynucleotide portion between the two primers, as well as the primer sequences themselves, that will be segregated and amplified.

The segregation/amplification scheme begins with a separating step (not shown), for example by polynucleotide denaturation, if the target nucleic acid sequence is in a double-stranded form. Denaturation can be accomplished by, for example, the application of heat or a change in pH using techniques standard in the art. In a preferred embodiment, strand separation is accomplished by enzymatic means. For example, single-stranded RNA targets are initially rendered double-stranded by annealing a DNA primer such as (-) primer 5 and extending the primer in the presence of a DNA polymerase and deoxy nucleotides, including a modified nucleotide capable of forming phosphodiester bonds. The isothermal amplification methods of the invention are improved over known isothermal amplification methods in using such modified nucleotides during all extension reactions. These extension reactions are catalyzed by polymerizing activities such as polymerases, which catalyze the linking of nucleotides into nucleic acid strands by forming phosphodiester bond linkages between the nucleotides. In using modified nucleotides that retain the capacity to form phosphodiester backbone bonds, the methods of the invention provide a substrate that permits polymerases to catalyze the reaction that they naturally catalyze. In contrast, known isothermal methods using modified nucleotides that do not retain the capacity to form phosphodiester bonds force polymerases to catalyze an unnatural reaction for which they may reasonably be expected to be less well suited. For example, alteration of the phosphoryl groups of a nucleotide, such as by substituting sulfur for phosphorus as in thiolated nucleotides, precludes the subsequent formation of phosphodiester backbone bonds. These modified nucleotides are incorporated by recruiting polymerases to catalyze unnatural bond formations such as thioether linkages, a task that polymerases would be expected to perform less efficiently than forming phosphodiester bonds. Subsequently, the RNA of the RNA: DNA hybrid is selectively degraded by an RNase such as RNase H and an isolated, double-stranded DNA containing the target sequence is generated by annealing a DNA primer such as (+) primer 2 and extending it by a polymerase-mediated reaction, facilitated by the use of modified nucleotides capable of forming a phosphodiester backbone (e.g. , dUTP, derivatives of dUTP comprising uracil derivatives, 7- deaza dGTP, and dITP).

Segregation of a target from a double-stranded DNA may be accomplished by controlled enzymatic degradation of single DNA strands, preferably by an exonuclease. This degradation may be accomplished, e.g., by Exonuclease LU, or by dephosphorylation, restriction and λ Exonuclease degradation of the single DNA strand bearing a 5' phosphate. (Step / in Figure 1A, not shown).

For a polynucleotide 1 that is an RNA molecule, step II in Figure 1 A is driven by a polymerizing enzyme capable of template-directed nucleic acid polymerization. An example of such a polymerizing enzyme is an RNA- dependent DNA polymerase such as a reverse transcriptase. This enzyme synthesizes a complementary (+) strand 3 having a typical phosphodiester backbone structure by extension of a first amplification primer, (+) primer 2, using polynucleotide 1 as a template. When polynucleotide 1 is RNA, reverse transcriptase is a preferred enzyme for catalyzing the synthesis of complementary (+) strand 3, which, in this case, is a cDNA. The reaction is performed in the presence of deoxyribonucleotides improved by the addition of at least one modified deoxyribonucleotide capable of forming phosphodiester bonds. Extension of (+) primer 2 by a polymerizing enzyme such as a DNA polymerase or a reverse transcriptase with such substrates efficiently forms complementary (+) strand 3, which is a modified strand (except for the portion contributed by (+) primer 2) containing the complement of the target sequence.

As noted above, the methods of the invention differ from known methods in using modified nucleotides having structural modifications confined to their base and/or sugar moieties, thus improving the selective interference with restriction endonuclease cleavage that avoids or minimizes interference with polymerizing activities. The hybridization of complementary (+) strand 3 to polynucleotide 1 yields product 4, a partially duplex (i.e., partially double- stranded), and partially hemi-modified, polynucleotide. The complementary (+) strand 3 is then separated from polynucleotide 1. When the polynucleotide 1 is an RNA molecule, separation is preferably achieved using an enzyme which selectively degrades the RNA polynucleotide 1 participating in an RNA/DNA duplex, as illustrated by the dashed line depicting partially degraded polynucleotide 1' in step /// of Figure 1A. Suitable enzymes are those which selectively act on the RNA strand of an RNA/DNA duplex and include enzymes which comprise an RNase H activity, as described above. For example, this step may be catalyzed by a reverse transcriptase which contains RNase H activity or by a separate RNase H enzyme. A preferred enzyme is E. coli RNase H. Other preferred methods to effect strand separation include the use of an exonuclease, the application of heat, or alteration of the pH of the mixture. Whether proceeding through a stage involving a partially degraded polynucleotide 1' or not, eventually the complementary (+) strand 3 is rendered functionally single stranded as illustrated by step TV.

In step V, complementary (+) strand 3 is contacted with a second amplification primer, (-) primer 5. The (-) primer 5 anneals to the region of complementary (+) strand 3 that contains sequence complementary to the sequence defining one end of the target nucleic acid sequence, this target being essentially the same as the sequence of the 3 ' binding region of (-) primer 5. As defined herein, RNA and DNA sequences of essentially the same informational content are essentially the same sequences and constitute copies of one another.

In step VI, (-) primer 5 is extended using complementary (+) strand 3 as a template and deoxyribonucleotides, wherein at least one of the deoxyribonucleotides is modified in a manner that preserves its capacity to form a phosphodiester bond. Yet again, interference with polymerizing activities is minimized by providing a modified nucleotide suitable for use in the reaction underlying nucleic acid polymerization, which is the formation of phosphodiester backbone bonds. Not surprisingly, both the kinetics of polymerization and the sensitivity of the reaction are improved. The product of this extension reaction generates modified (-) strand 6 participating, with complementary (+) strand 3, in a partial duplex polynucleotide 7 (step VI). Partial duplex polynucleotide 7 contains two incompletely modified polynucleotide strands. The 5' terminus of each strand, modified (-) strand 6 and complementary (+) strand 3, is provided by an unmodified primer sequence, including an unmodified restriction endonuclease recognition sequence. The 5' end of (-) strand 6 is contributed by (-) primer 5; the 5' end of complementary (+) strand 3 is contributed by (+) primer 2. Consequently, the single engineered restriction endonuclease recognition site in the partial duplex polynucleotide 7 is a substantially hemi-modified recognition site because of the participation of unmodified (+) primer 2 in forming that site. In Figures 1A and IB, nickable restriction endonuclease recognition sites are illustrated by juxtaposed boxes of different patterns. The boundary between adjacent boxes represents the cleavage site.

In step VII, the hemi-modified restriction endonuclease recognition site in partial duplex polynucleotide 7 is nicked by restriction endonuclease (e.g. , Hpal) scission of the single unmodified strand of the restriction endonuclease site. The nick produces upstream (+) strand 8 and downstream (+) strand 9, both annealed to modified (-) strand 6 which is not nicked. The upstream (+) strand 8 has a free 3' terminus available to prime the synthesis of another polynucleotide strand complementary to modified (-) strand 6. In step VIII, the 3' end of upstream (+) strand 8 is extended to form modified (+) strand 10, which is complementary to modified (-) strand 6. Again, the extension is accomplished using an improved polymerization reaction that uses modified nucleotides capable of phosphodiester bond formation as substrates that allow a polymerase to catalyze phosphodiester bond formation, the reaction for which nature designed the enzyme. This step displaces (+) strand 9 in the process. The annealing of modified (+) strand 10 to modified (-) strand 6 yields a double-stranded polynucleotide fragment 11 that is substantially free of non-target DNA. The non-target DNA in double- stranded polynucleotide fragment 11 consists primarily of the two introduced restriction endonuclease recognition sites. Therefore, double-stranded polynucleotide fragment 11 is a double-stranded form of a segregated copy of the target nucleic acid sequence suitable for amplification in the method of the present invention. Double-stranded polynucleotide fragment 11 contains two substantially hemi-modified, and nickable, restriction endonuclease recognition sites, one located at each end of the fragment. The two nickable restriction endonuclease recognition sites in double-stranded polynucleotide fragment 11 may be the same or different.

In Figure IB, the double-stranded polynucleotide fragment 11 cycles through a process of nicking at the hemi-modified restriction endonuclease recognition sites (step IX) and extensions of the 3' termini produced by the nicks, resulting in displacements of the downstream strands containing the target nucleic acid sequence or its complement (step X). Importantly, these cyclic extension reactions, improved by matching the substrate (i.e. , modified nucleotides capable of phosphodiester bond formation) to the enzymatic activity acting upon it (i.e. , a polymerase activity), build upon the kinetic and sensitivity improvements of prior cycles, geometrically amplifying the magnitude of the improvements. With specific reference to Figure IB, nascent (+) strand 15 is formed by extending upstream (+) strand 8 in the presence of the aforementioned modified nucleotides. The synthesis of nascent (+) strand 15 displaces the pre-existing displaced (+) strand 12. Although not shown in Figure IB, displaced (+) strand 12 binds (-) primer 5. Extension of (-) primer 5 in the presence of those same modified nucleotides yields nascent (-) strand 16 (not shown). As synthesized, nascent (+) strand 15 is hybridized to displaced (-) strand 13, forming nascent duplex 17 as shown in step XIa. Nascent duplex 17 contains one regenerated nickable restriction endonuclease recognition site. Nicking that site cleaves nascent (+) strand 15, forming upstream (+) strand 8 and minimal downstream (+) strand 19 (step XUa). In step Xllla, extension of upstream (+) strand 8 using the previously described modified nucleotides generates another copy of nascent (+) strand 15 while displacing minimal downstream (+) strand 19. In step XTVa, the extension reaction is completed, yielding a complete copy of nascent (+) strand 15 hybridized to displaced (-) strand 13, thereby regenerating nascent duplex 17. Nascent duplex 17 recycles through steps Xlla, Xllla, and XTVa. The displaced minimal downstream (+) strand 19 (step XVa) hybridizes to (-) primer 5 (step XVIa). Extension of (-) primer 5 produces a copy of nascent (-) strand 16; extension of minimal downstream (+) strand 19 using the sequence of (-) primer 5 as a template yields a copy of displaced (+) strand 12 (step XVIIa). Again, each of the extension reactions is improved by the use of modified nucleotides capable of forming the phosphodiester bond naturally catalyzed by polymerase activity, in contrast to the forced catalysis of thioether bonds when using thiolated nucleotides. As synthesized, displaced (+) strand 12 is hybridized to nascent (-) strand 16, thereby creating a copy of nascent duplex 18 which serves as a substrate in step Xllb.

The preceding discussion traced the fate of upstream (+) strand 8 and displaced (+) strand 12 from step X. An analogous series of reactions is involved in manipulations directed to upstream (-) strand 14 and displaced (-) strand 13 (step X). As explained above, each one of the extension reactions occurring during each cycle of the reaction is improved by using modified nucleotides capable of phosphodiester bond formation as substrates better suited to polymerase-mediated extension reactions than the backbone-altering modified nucleotides typified by thiolated nucleotides. In step Xlb, extension of upstream (-) strand 14 generates a copy of nascent (-) strand 16 and displaces the previously annealed displaced (-) strand 13. Displaced (-) strand 13 hybridizes to (+) primer 2; extension of (+) primer 2 yields nascent (+) strand 15 which, hybridized to displaced (-) strand 13, generates a copy of nascent duplex 17 (shown as the product of step XVIIb). In step Xllb, nascent (-) strand 16, participating in nascent duplex 18, is nicked, thereby creating upstream (-) strand 14 and minimal downstream (-) strand 20. Extension of upstream (-) strand 14 generates nascent (-) strand 16 and displaces minimal downstream (-) strand 20 (step Xlllb). In step XlVb, the synthesis of a copy of nascent (-) strand 16 is complete and, because that strand is hybridized to displaced (+) strand 12, a copy of nascent duplex 18 is created. Nascent duplex 18 then cycles back through steps Xlb, Xlllb, and XTVb. The displaced minimal downstream (-) strand 20 (step XVb) hybridizes to (+) primer 2 (step XVIb). Extension of (+) primer 2 yields a copy of nascent (+) strand 15, and the extension of minimal downstream (-) strand 20 using (+) primer 2 as a template produces displaced (-) strand 13 (step XVIIb). Because nascent (+) strand 15 is hybridized to displaced (-) strand 13, a copy of nascent duplex 17 is generated, which feeds into step Xlla. These steps are repeated during the amplification process.

The (+) primer 2 and (-) primer 5 used in the methods generally have lengths of about 40 nucleotides each. The (+) primer 2 and (-) primer 5 target binding sites exhibit sequences that are substantially complementary to the target or the target's complement.

Aspects of the invention disclosed in the preceding discussion may be better understood upon consideration of the following Examples, wherein Example 1 discloses a method for preparing a double-stranded DNA target; Example 2 illustrates the preparation of an RNA target; Example 3 is a comparative example (in conjunction with Example 4) which describes the conventional isothermal amplification of an RNA target using TTP and [αS] dCTP; Example 4 discloses the isothermal amplification of an RNA target using dUTP and dCTP; Example 5 illustrates an assay for, and identification of, candidate sites for nicking; and Example 6 discloses a direct assay for nicking activity.

Example 1 An RNA target was designed to contain sequence corresponding to a 159 bp region (nucleotides 937-1095 of SEQ ID NO:l) of the 1,750 bp 18S ribosomal DNA of Cryptosporidium parvum, the sequence of which is available from the GenBank database under accession number LI 6997. The sequence of the C. parvum 18S rRNA gene is presented in SEQ ID NO: 1. The target was prepared by PCR amplification of chromosomal DNA, with the concomitant introduction of a flanking T7 promoter sequence. Subsequently, the PCR product was used as a template in an in vitro transcription reaction (see Example 2) to generate the RNA target subjected to an isothermal amplification method according to the invention. C. parvum chromosomal DNA was isolated using a standard

DNA isolation protocol. Sambrook et al. , Molecular Cloning: A Laboratory Manual, §§ 9J6-9.22 (Cold Spring Harbor Laboratory Press, Cold Spring Harbor N.Y. 1989). A first PCR primer was PCR F, having the sequence 5'- ACGAAAGTTAGGGGATCGAAGA-3', as set forth in SEQ ID NO:2; a second PCR primer was PCR R Pro, having the sequence 5'- TGTAATACGACTCACTATAGGGCGATAAGTTTCAGCCTTGCGACCAT- 3', as set forth in SEQ ID NO: 3. The 20 nucleotides at the 5' end of PCR R Pro are derived from a T7 promoter sequence and are not complementary to any region of the 159 bp 18S rDNA target. PCR was performed using the commercially available Qiagen PCR Optimization Kit according to the supplier's recommendations. Qiagen, Inc., Valencia, CA. PCR was conducted in a final reaction volume of 100 μl; ascertainable final concentrations of reaction components were: lx PCR buffer (Tris HCl j H 8.7], (NH₄)₂SO₄ (proprietary quantity), and 2 mM MgCl₂), 200 μM of each standard deoxynucleotide (dGTP, dATP, TTP, and dCTP), lx Q solution, 0.5 M (50 pmol) each of PCR primers PCR F and PCR R Pro, and 2.5 units of Taq DNA polymerase. Qiagen, Inc. The concentrations of reagents in the PCR buffer and the Q solution were retained by the supplier as proprietary information. A typical protocol was used to amplify the 159 bp C. parvum target using PCR. In particular, the double-stranded target was denatured by incubation at 80 °C for 5 minutes, followed by 30 seconds at 98 °C. The reaction mixture was then incubated at 55 °C for 30 seconds to promote primer hybridization. Extension reactions were conducted by elevating the temperature to 72 °C for one minute, followed by a denaturation step at 94 °C for 30 seconds. A total of 40 cycles of primer hybridization, extension, and denaturation were typically used. The final cycle differed from all preceding cycles in having a 10 minute extension period at 72 °C followed by an immediate reduction in the temperature to 4°C for storage.

PCR products were chromatographically purified using the QIAquick PCR Purification Kit (Qiagen, Inc.) according to the supplier's instructions. Typically, 95% recovery of PCR products was achieved, amounting to approximately 10 μg of product.

Example 2

PCR products purified as described in Example 1 were subjected to in vitro transcription reactions using T7 RNA polymerase to generate single- stranded RNA targets. In vitro transcription reactions were performed using the Ambion MEGAscript Kit containing T7 RNA polymerase and an RNase inhibitor according to the supplier's instructions. Ambion Inc., Austin, TX. Reactions were routinely performed in 20 μl, containing lx proprietary reaction buffer; 75 mM each of GTP, ATP, UTP and CTP; 5 μl (approximately 0.5 μg) of PCR-amplified target prepared, e.g. , in accordance with Example 1; proprietary quantities of T7 RNA polymerase and an RNase inhibitor, and diethylpyrocarbonate (i.e., DEPC)-treated water. Reactions were performed in sterile 0.5 ml microfuge tubes at 37 °C for 4-6 hours. RNA product was purified using a standard technique involving phenol/chloroform extraction followed by precipitation with 0.1 volumes of 3 M ammomum acetate and 2.5 volumes of ethanol (-20 °C). Sambrook et al , App. E.10-11. Precipitated RNA (approximately 0J-0.5 mg) was dissolved in 100 μl of DEPC-treated water, divided into sample aliquots (approximately 10 μl each at 1-5 μg/μl) and stored at -80°C.

Example 3

An isothermal rapid amplification reaction was performed without the use of a modified nucleotide, such as dUTP, to serve as a control. The reaction mixture contained 1 μl (approximately 1-5 μg) of C. parvum 18S rRNA prepared as described in Example 2. The RNA target was added to a final volume of 50 μl of reaction mixture also containing 35 mM K»PO₄ (pH 7.6); 0.5 mM each of dGTP, dATP, and TTP; 1.4 mM [αS] dCTP; 0.5 μM each of primers PI and P2; 35 μg BSA; 10.2 mM MgCl₂; 0.7 mM Tris-HCl (pH 7.9); 3.4 mM KC1; 2% maltitol; 1.34% trehalose; 4 units (31 U/μl) AMV reverse transcriptase (Chimerx, Madison, WI); 32 units of recombinant Bst DNA polymerase (520 U/μl; Chimere); and 150 units of flwHKCI (1500 U/μl; Chimerx) Primer PI has the following DNA sequence: 5 -ACCGCATCGAATGCATGTTCTCGGGTCGTAGTCTT AACCAT-3' (SEQ ID NO:4). The DNA sequence of Primer P2 is 5'- CGATTCCGCTCCAGACTTCTCGGGTGCTGAAGGAGTAAGG-3' (SEQ ID NO:5).

The amplification reaction mixture was incubated at 60 °C, typically for 30 minutes. The reaction was stopped and double-stranded DNA was denatured by incubation at 95 °C for 10 minutes, followed by rapid cooling in an icewater bath for 5 minutes. Single-stranded amplicon was then added to a microtiter well containing an immobilized capture oligonucleotide perfectly complementary to a portion of the single-stranded amplicon. In particular, the capture oligonucleotide has the sequence 5 '- CTATGCCAACTAGAGATTGGAGGTTGTT-3' (SEQ ID NO: 6; Chimerx) Immobilized amplicon strands were then detected by hybridizing a labeled oligonucleotide, designated P2 Comp, to an amplicon strand portion that differed from the portion to which the capture oligonucleotide bound. P2 Comp, also obtained from Chimerx, has the sequence 5'- CCTTACTCCTTCAGCACCCGAGAAGTCTGGAGCGGAATCG-3'(SEQ ID NO: 7); its 5' end was attached to horseradish peroxidase (i.e., HRP; Roche Molecular Biochemicals, Boehringer Mannheim Corp., Indianapolis, IN), according to the supplier's instructions.

Hybridization reactions for target detection were carried out in 100 μl volumes of 2x SSC for 60 minutes at room temperature, a standard procedure in the art. Following hybridization, immobilized amplicon was indirectly detected as the required intermediate in a hybridization sandwich that resulted in the immobilization of HRP-labeled P2 Comp. The activity of the HRP attached to P2 Comp was assayed using tetramethylbenzidine (i.e. , 1MB; Moss, Inc. , Pasadena, MD) according to the supplier's instructions; reaction products were detected using a microplate spectrophotometer (Microplate Autoreader EL 309; BioTek Instruments, Winooski, NT). Immobilization of horseradish peroxidase in a microtiter well resulted in the presence of an enzyme capable of converting the colorless substrate to a blue-colored product. Subsequent addition of 100 μl of 0.5 Ν HCl changed the product's color to yellow, which was then quantitated at 450 nm using the microplate spectrophotometer. One O.D.₄₅₀ equals about 33 nanograms of a 240 bp polynucleotide. Typically, these amplification reactions generated a total of about five micrograms (μg) of polynucleotide product in approximately 15 minutes.

Example 4

An improved isothermal RAMP reaction was performed with the use of dUTP as a modified nucleotide. The target was single-stranded 18 S rRNA of C. parvum, produced by PCR amplification of genomic DNA as described in Example 1 , followed by in vitro transcription of the amplified DNA as described in Example 2. Isothermal amplification of the single- stranded RNA target according to the invention was performed as described in Example 3, with the following modifications. The modified nucleotide described in Example 3, [αS] dCTP (1.4 mM), was replaced by 0.5 mM dCTP. The 0.5 mM TTP of Example 3 was replaced by 1.4 mM dUTP as a modified nucleotide. The 150 units of BsiHKCI were replaced by 30 units of Hpal. The amplification primers, PI and P2, were replaced by Hpal PI and Hpal P2. The Hpal PI and Hpal P2 primers differ from the primers disclosed in Example 3 in specifying a different restriction endonuclease recognition sequence (i.e. , Hpal) towards their five prime ends. In particular, the sequence of Hpal PI is 5'-ACCGCATCGAATGCATGTTAACGGGTCGTAGT- CTTAACCAT-3' (SEQ ID NO: 8); Hpal P2 has the following sequence: 5'- CGATTCCGCTCCAGAGTTAACGGGTGCTGAAGGAGTAAGG-3' (SEQ ID NO:9). Beyond these differences, the isothermal amplification of the single- stranded RNA target was performed as described in Example 3.

Amplification reactions conducted in the presence of modified nucleotides capable of forming phosphodiester bonds, such as dUTP, typically produced approximately five micrograms of product in less than two minutes.

Thus, the use of modified nucleotides such as dUTP resulted in the production of amplicon in considerably less time than has been required using conventional modified nucleotides such as [αS] dCTP. (A two-minute reaction time with dUTP compared to a 15-minute reaction time for [αS] dCTP, see Example 3 above.)

Beyond the benefit in kinetics, the isothermal amplification methods of the invention offer the additional benefit of improved sensitivity. In Fig. 2, a comparison of the relative levels of target nucleic acid required to support the production of detectable levels of amplicon using conventional amplification methodology versus the methods of the invention is presented. Fig. 2 shows the relative quantity of amplicon produced from varying quantities of target nucleic acid using either a conventional isothermal amplification method (open circles) or a method according to the invention (open triangles). Not even a sample target nucleic acid dilution of 10^"10 was limiting in methods according to the present invention. In contrast, a 10^"8 dilution of target nucleic acid resulted in a marked decrease in amplicon yield using conventional isothermal amplification technology. Thus, the methods of the invention are at least 100-fold more sensitive than conventional amplification methods.

Example 5

A variety of dUTP-modified restriction endonuclease recognition sites were tested for their ability to undergo cleavage when contacted by a cognate restriction endonuclease. Each one of the examined sites (see Table I below) was present in pSL1180 DNA (Amersham Pharmacia Biotech, Piscataway, N.J.) The modified nucleotide, dUTP, was incorporated (as dUMP) independently into each restriction endonuclease recognition site during PCR amplification of the pSL1180 template.

In particular, the PCR reactions contained 0.5 μg of pSL1180, 0.5 μM primers PI (SEQ ID NO:4) and P2 (SEQ ID NO:5), PCR reaction buffer (10 mM Tris HCl |pH 8.4], 50 mM KC1, 1.5 mM MgCl₂, 0.1 % Triton X-100, and 200 μM dGTP, dATP, dCTP, and either one of TTP or dUTP), and 2.5 Units of Taq DNA polymerase in a final volume of 100 μl. The reactions were incubated at 68 °C for three minutes, 72 °C for four minutes and 95 °C for one minute. Thirty-five cycles were performed for each reaction, aided by use of a Perkin-Elmer thermocycling apparatus. PCR reaction products were subsequently centrifuged through Sephacryl 400 spin columns (Amersham Pharmacia Biotech) according to the supplier's instructions.

Modified sites were assayed for cleavage by combining 0J μg of PCR product with a commercial enzyme and reaction buffer, in accordance with the supplier's recommendations (enzyme activities are indicated in Table I). Final reaction volumes were 20 μl, and the reactions were incubated at the recommended temperature for one hour. Subsequent assay for cleavage was performed by fractionating digestion products using 1 % agarose gel electrophoresis in a standard TBE buffer (Tris, borate, EDTA; see Sambrook et al., 1989), followed by visual examination and sizing of ethidium bromide- stained restriction fragments. The results of this experiment are also presented in Table I.

Apparent from an inspection of the results is the generally high level of enzyme required to cleave a modified site. Thus, modified sites are relatively poor substrates for restriction endonuclease cleavage, regardless of the site/enzyme pair being investigated. Table I further reveals that, while the effect of site modification on cleavage is not site/enzyme-specific, sites exhibiting structural similarities behave similarly in terms of their relative susceptibility to cleavage. For example, those sites lacking a modified nucleotide (i.e. ,dUMP-free Apal, Narl, Notl, and SαcII sites), exhibit essentially no cleavage inhibition (dU/T enzyme ratio of 4 for Apal, Narl and S cII) . In contrast, the BamHl and Nrul sites, representative of those sites (Mel, PvuU, Sphl, Pstl, Mlul, BamHl, Ncol, Nrul, Stul, and Kpnl) having a single dUMP at a position that is not directly involved in cleavage, had an average dU/T enzyme ratio of 32.5. Those sites having more than one dUMP nucleotide, with one of those nucleotides directly involved in cleavage (Hpal, Xbal, EcoRV, Seal, and Clal), remained refractory to cleavage at all enzyme levels tested, with the exception of the Clal site; this group of sites also had an average dU/T enzyme ratio of > 180. A final category of sites contained more than one dUMP, wherein none of the modified nucleotides was directly involved in cleavage. None of these sites (EcόBI, Bg , and HindM) was cleaved at tested levels ranging up to 50- 100 units per μg DNA, and these sites exhibited an average dU/T enzyme ratio of > 116. The dU/T enzyme ratios reveal that dUTP substitution in a DNA containing a restriction endonuclease recognition site inhibits cleavage of that site, including a site lacking any dUMP, but the inhibition is much greater as the number of dUMP residues within the restriction endonuclease recognition site increases from zero to two.

Table I

Site Modified dT dU dU/T sequence (enzyme units) (enzyme units) enzyme ratio

Apal GGCC/C 0J 0.5 5

(SEQ ID NOJO) Narl GG/CGCC 0.5 2.5 5

(SEQ ID NO: 11)

Noil GC/GGCCGC 0J5 -

(SEQ ID NO: 12)

Sacll CCGC/GG 0.5 1.0 2

(SEQ ID NO: 13)

Sail G/UCGAC 5 10 2 (SEQ ID NO: 14)

Xhol C/UCGAG 0.5 15 30

(SEQ ID NO: 15) Sad GAGCU/C 0.5 30 60

(SEQ ID NO: 16) Nhel G/CUAGC 1.0 - (SEQ ID NO: 17)

Pvull CAG/CUG 10 - (SEQ ID NO: 18)

Sphl GCAUG/C 5 -

(SEQ ID NO: 19)

Pstl CUGCA/G 1 - (SEQ ID NO:20)

MM A/CGCCU 5 - (SEQ ID NO:21)

BamHl G/GAUCC 1 25 25

(SEQ ID NO:22)

Ncόl C/CAUGG 2.5 -

(SEQ ID NO:23)

Nrul UCG/CGA 0.5 20 40

(SEQ ID NO:24)

Stul AGG/CCU 5 - (SEQ ID NO.-25)

Kpnl GGUAC/C 2.5 - (SEQ ID NO:26)

Hpal GUU/AAC 0.5 >25 >50 (SEQ ID NO: 27)

Xbal U/CUAGA 0.25 >50 >200 (SEQ ID NO:28)

EcoRV GAU/AUC 0J >50 >500 (SEQ ID NO.J9)

Seal AGU/ACU 0.5 >50 > 100 (SEQ ID NO:30) Clal AU/CGAU 0.5 25 50 (SEQ ID NO:31)

EcoRl G/AAUUC 0.5 > 100 >200 (SEQ ID NO:32)

BgM A/GAUCU 0.5 >50 > 100 (SEQ ID NO:33)

Hinffll A/AGCUU 1 >50 >50 (SEQ ID NO:34)

Although the data contained in Table I address the cleavability of modified restriction endonuclease recognition sites, the relative susceptibility of a fully modified site to double-stranded cleavage is inversely correlated to its relative susceptibility to single-stranded nicking when the site is hemi-modified. Assays designed to confirm this correlation, or to directly assay for nicking activity, are within the skill in the art. For example, denaturing gel electrøphoresis could be used to separately monitor the integrity of each strand of a duplex polynucleotide. Alternatively, the technique described in Example 6 may be used to directly assay for nicking activity. One of ordinary skill in the art would also understand that the methods involve the nicking of restriction sites and that nicking activity may be provided by the conventional cognate restriction endonuclease or an isoschizomer thereof.

Example 6

A direct assay for nicking activity involves a double-stranded nucleic acid having at least one restriction endonuclease recognition site of interest. In the experiment described below, a hemi-modified duplex oligonucleotide of 42 bp that contained internal Hpal (GTT/AAC) and SαcII (CCGC/GG) sites was generated using techniques known in the art. The template strand of the duplex nucleic acid, having the sequence 5'- TTI rTGAATTCGTTAACCCGCGGGATATCTGATCATTTTTT-3 ' (SEQ ID NO: 35), had biotin attached at its 5' end and digoxigenin attached at its 3' end, thus placing the restriction endonuclease recognition site between the immobilizing and detecting agents. The modified strand substituted dUMP for TMP. The biotin served as an immobilizing agent in binding to the streptavidin-coated wells of a microplate. The digoxigenin was used in an indirect colorimetric detection assay also requiring anti-digoxigenin antibodies conjugated to horseradish peroxidase.

A duplex nucleic acid having the sequence set forth in SEQ ID NO: 35 and containing two unmodified strands, as well as a corresponding duplex nucleic acid containing one strand modified by substituting dUMP for TMP, were synthesized using conventional techniques. Thus, one duplex nucleic acid contained unmodified Hpal and SacU sites; the other duplex contained hemi-modified Hpal and SαcII sites. Restriction digestions were conducted in 10 μl volumes containing 10 mM Tris HCl (pH 7.5), 10 mM MgCl₂, 50 mM KC1, 1.0 mM dithiothreitol, 100 μg bovine serum albumin, 0.1 μg of duplex nucleic acid (10 pmol), and 1 μl (approximately 15 units) of restriction endonuclease. (Hpal and SacH were obtained from Chimerx) Reactions were typically incubated at 37 °C for one hour. Heat denaturation was effected by incubation at 100° C for five minutes, followed by quick chilling in ice water.

Indirect detection of nicking activity was carried out by the dilution of 1 μl of the digestion products in 100 μl total volume of TBST buffer (20 mM Tris HCl [pH 8.5], 0J5 M NaCl, 0.5% triton X-100). The diluted reaction products were then added to streptavidin-coated wells of a microplate and incubated at 37 °C for one hour, followed by three washings using TBST. Antibody conjugate (anti-digoxigenin attached to horseradish peroxidase; DIG Detection ELISA [TMB], Roche Molecular Biochemicals), diluted 1:100 in TBST, was then added to each well in a volume of 100 μl. Again, the microplate was incubated at 37 °C for one hour. Wells were then washed 5x with 100 μl of TBST per wash. Finally, 100 μl of 1MB, a horseradish peroxidase substrate, was added to initiate the colorimetric reaction; the reaction was stopped after about five minutes with the addition of 100 μl of 0.5 N HCl. A product of the reaction was then quantitated spectrophotometrically (450 nm) and the data were preserved with a microplate recorder, as described in Example 3.

The unmodified duplex nucleic acid showed an optical density (O.D.₄₅₀) of 0.798 (approximately 26 ng) in the absence of any restriction endonuclease, but when exposed to Hpal (GTT/AAC), or SacU (CGCG/GG), the O.D.₄₅₀ fell to 0J56 (5 ng) or 0.085 (3 ng), respectively. A control reaction lacking duplex nucleic acid yielded an O.D.₄₅₀ of 0.071. Assays performed on the hemi-modified duplex nucleic acid resulted in O.D.₄₅₀ readings of 0.418 (no restriction endonuclease), 0.420 (Hpal), 0J02 (SacU), and 0.079 (no duplex nucleic acid). Thus, Hpal did not fully cleave the duplex nucleic acid that had been hemi-modified by substituting dUMP for dTMP; SacU did cleave both strands of the hemi-modified duplex nucleic acid. These results are consistent with the substitutable presence of "T" residues in the Hpal, but not SacU, site. Finally, the experiment described above was repeated with the addition of a thermal denaturation step prior to detection. The following results were obtained: 0.242 (no restriction endonuclease), 0.062 (Hpal), 0.046 (SacU), and 0.071 (no duplex nucleic acid). As expected, there was little immobilized digoxigenin after SacU treatment and heat denaturation. However, the relatively high level of immobilized digoxigenin (O.D.₄₅₀ of 0J56) following exposure to Hpal alone was lost upon exposure to heat denaturation (O.D.₄₅₀ of 0.062). These results indicate that the digoxigenin was detectably retained by hybridization of the digoxigenin-linked nicked strand to the intact hemi-modified complementary strand, which was, in turn, hybridized to the other half of the nicked strand attached to biotin. The biotin served to anchor the entire hybridization sandwich by interaction with the streptavidin that was coated on the wells of the microplate.

While the present invention has been described in terms of specific embodiments, it is understood that variations and modifications will occur to those skilled in the art. Accordingly, only those limitations appearing in the appended claims should be placed upon the invention.

Claims

What is claimed is:

1. An isothermal method for amplifying a target nucleic acid comprising the following steps:

(a) providing a reaction mixture comprising a sample containing or suspected of containing a target nucleic acid; first and second amplification primers, wherein each primer has a restriction endonuclease recognition sequence for forming a hemi-modified restriction endonuclease recognition site, said sequence disposed 5' relative to a target binding site; a collection of nucleotides consisting essentially of conventional nucleotides and one or more modified nucleotides capable of phosphodiester bond formation, wherein the nucleotides of said collection are collectively capable of base pairing with each of dGMP, dAMP, TMP, and dCMP; and a polymerase lacking an effective 5' — > 3' exonuclease activity;

(b) contacting said reaction mixture with a restriction endonuclease for nicking at least one of two hemi-modified restriction endonuclease recognition sites formed by hybridization of said amplification primers to said target nucleic acid; and

(c) incubating said reaction mixture under substantially isothermal conditions, thereby amplifying said target nucleic acid.

2. The method according to claim 1 wherein said one or more modified nucleotides is a single modified nucleotide capable of phosphodiester bond formation.

3. The method according to claim 1 wherein said collection of nucleotides consists essentially of dGTP, dATP, dUTP and dCTP.

4. The method according to claim 1 wherein said modified nucleotide is selected from the group consisting of dUTP, derivatives of dUTP comprising uracil derivatives, 7-deaza dGTP, and dITP.

5. The method according to claim 4 wherein said modified nucleotide is 5-Br-dUTP.

6. The method according to claim 1 wherein said hemi-modified restriction endonuclease recognition sites each comprise a modified nucleotide having an atom that participates in a covalent bond that is nicked.

7. The method according to claim 1 wherein said hemi-modified restriction endonuclease recognition sites each comprise at least two modified nucleotides.

8. The method according to claim 1 wherein said target nucleic acid is single-stranded.

9. The method according to claim 1 wherein said restriction endonuclease is selected from the group consisting of Hpal, Xbal, EcoRY, Seal, Clal,

EcoRl, BglU, HindUl, Sail, Xhol, Sacl, Nhel, PvuU, Sphl, Pstl, MM, BamHl, Ncol, Nrul, Stul, Kpril, HincU, and isoschizomers thereof.

10. The method according to claim 1 wherein at least one of said restriction endonuclease recognition sites has a sequence selected from the group consisting of the sequences set forth as SEQ ID NO: 14 to SEQ ID NO: 34 and their complementary sequences.

11. The method according to claim 1 wherein said restriction endonuclease recognition sites have the same sequence.

12. The method according to claim 1 wherein said restriction endonuclease recognition sites each comprise a deoxyadenosine nucleotide.

13. An isothermal method for amplifying a target nucleic acid comprising the following steps:

(a) providing a reaction mixture comprising a sample containing or suspected of containing a target nucleic acid; first and second amplification primers, wherein each primer has a restriction endonuclease recognition sequence for forming a hemi-modified restriction endonuclease recognition site, said sequence disposed 5' relative to a target binding site; a collection of nucleotides comprising conventional nucleotides and one modified nucleotide, wherein the modified nucleotide is capable of phosphodiester bond formation, and further wherein the nucleotides of said collection are collectively capable of base pairing with each of dGMP, dAMP, TMP, and dCMP; and a polymerase lacking an effective 5' -> 3' exonuclease activity;

(b) contacting said reaction mixture with a restriction endonuclease for nicking at least one of two hemi-modified restriction endonuclease recognition sites formed by hybridization of said amplification primers to said target nucleic acid; and

(c) incubating said reaction mixture under substantially isothermal conditions, thereby amplifying said target nucleic acid.

14. A kit for isothermal nucleic acid amplification comprising:

(a) first and second amplification primers, each primer having a restriction endonuclease recognition sequence for forming a nickable restriction endonuclease recognition site disposed 5' relative to a target binding sequence; (b) a collection of nucleotides consisting essentially of conventional nucleotides and a single modified nucleotide, wherein said modified nucleotide is capable of phosphodiester bond formation, and further wherein the nucleotides of said collection are collectively capable of base pairing with each of dGMP, dAMP, TMP, and dCMP;

(c) a polynucleotide polymerase lacking an effective 5' — > 3' exonuclease activity; and

(d) a restriction endonuclease for nicking at least one of two restriction endonuclease recognition sites.

15. The kit according to claim 14 wherein said modified nucleotide is dUTP.