WO2010029520A1

WO2010029520A1 - Primer and amplification method for forming single stranded dna amplicons

Info

Publication number: WO2010029520A1
Application number: PCT/IB2009/054027
Authority: WO
Inventors: Natela Rekhviashvili; Natasha Gous; Lesley Scott; Alexio Capovilla; Wendy Stevens
Original assignee: University of the Witwatersrand, Johannesburg
Current assignee: University of the Witwatersrand, Johannesburg
Priority date: 2008-09-15
Filing date: 2009-09-15
Publication date: 2010-03-18
Anticipated expiration: 2011-03-15
Also published as: AP2011005669A0; ZA201102807B

Abstract

This invention describes primers and an amplification method which produce single stranded DNA amplicons from either RNA or DNA. The ssDNA amplicons can be directly combined with detection systems such as NALF, flow cytometry, microarrays and the like. The invention is especially suitable for use in point of care tests. The primers include a short non-target specific sequence for preventing a displaced amplicon from self-priming; a sequence complementary to a strand extended from the primer ('overhang" sequence); a short artificial sequence that forms a link; and a sequence specific forward primer. The forward primer binds to the anti-sense cDNA strand and produces a dsDNA target. The overhang sequence is also extended and the anti-sense strand gets displaced from a nickable site, forming an amplicon with two complementary sequences. These hybridize, forming a partially looped ssDNA amplicon which is unable to bind to another forward primer and cannot be converted into a dsDNA target.

Description

PRIMER AND AMPLIFICATION METHOD FOR FORMING SINGLE STRANDED DNA

AMPLICONS

BACKGROUND OF THE INVENTION

Nucleic acid testing (NAT) for diagnosis and monitoring of infectious diseases, disease associated mutations; gene over expression and so forth is an attractive alternative to conventional testing methods and can be adapted for rapid point of care (POC) assay format. In the field of diagnostics of infectious diseases, NAT provides significantly higher analytical sensitivity and quantitative accuracy when compared to serologic or antigen-based assays.

Recently there has been an increasing interest in developing miniaturised instruments containing inexpensive, disposable test units (e.g. microfluidic cassettes) that allow amplification and detection steps of NAT to be performed in one closed system suitable for near-patient/POC testing. Nucleic acid lateral flow is a low cost, rapid detection system widely used for POC testing. Visual or instrument dependant detection of product captured on the Nucleic Acid Lateral Flow (NALF) strips is achieved by using different reporter particles, for example, colloidal gold and superparamagnetic particles.

A number of applications require ssDNA product of a defined length for further manipulation, for example, detection of nucleic acids by flow cytometry and NALF. Protocols have been developed that describe time consuming and cumbersome post-amplification treatment of dsDNA amplicons in order to convert them into ssDNA amplicons.

In "isothermal" methods, the entire amplification process is performed under one constant temperature. Even though many isothermal amplifications use an initial denaturation at high temperature or an initial reverse transcription (RT), compared to PCR amplification there is no temperature cycling, as the amplification process is performed at one uniform temperature. Therefore, an isothermal amplification does not require a PCR instrument and it can be performed in a simple heating block. This feature makes isothermal amplification easily adaptable to the POC testing format. There are a variety of published and commercially available methods for isothermal amplification of nucleic acids. Approved isothermal amplification techniques that are most commonly used for diagnostics may be classified into two major groups: signal amplification and target amplification. Among widely used signal amplification methods are branched DNA (b-DNA) amplification, hybrid capture, cycling probe technology, rolling circle amplification (RCA). Among widely used target amplification methods are nucleic acid sequence-based amplification (NASBA), transcription mediated amplification (TMA), strand displacement hybridization (SDA), loop-mediated isothermal amplification (LAMP) and helicase-dependant isothermal DNA amplification (HDA). The techniques listed above differ significantly in their design and technical characteristics, such as amplification efficiency and type of amplicons produced.

In order to be compatible with the detection formats, such as lateral flow (or any other detection format suitable for POC testing), and to be suitable for POC testing, a front end isothermal amplification technique has to meet certain requirements. Firstly, it has to provide great power of amplification so that even low concentrations of target can be amplified sufficiently in a short period of time (1 hour, or ideally even less). In other words, the method has to show good sensitivity when using a short amplification time. Secondly, an amplification method has to convert the RNA or DNA target into short (maximum 200 bases), single stranded DNA (ssDNA) or RNA amplicons in order to be readily compatible with the lateral flow detection format.

None of the developed isothermal amplification methods described above fulfils both requirements at the same time. Signal amplification techniques are usually linear and take at least a few hours to provide sufficient sensitivity. Amplicons produced with these techniques are either long stretches of DNA (e.g. RCA) or complex, "branched" molecules (e.g. b-DNA). Isothermal target amplification techniques which utilize the concept of strand displacement amplification (e.g. SDA, LAMP) show higher efficiency than the techniques based on RNA transcription (e.g. NASBA, TMA). SDA and LAMP methods have good sensitivity, but in their existing format these methods cannot be employed for POC HIV-1 RNA tests that rely on lateral flow, since the final amplification products are double stranded DNA (dsDNA) fragments and long stretches of dsDNA, respectively.

As a result, there is no known isothermal amplification method that is able to provide sufficient amounts of ssDNA amplicons within the required time without further modifications. SUMMARY OF THE INVENTION

According to a first embodiment of the invention, there is provided a nucleic acid molecule for amplifying a target region of a target nucleic acid molecule, the nucleic acid molecule comprising:

(a) a first nucleotide sequence which is non-specific to the nucleic acid molecule;

(b) a second nucleotide sequence which is complementary to at least a portion of the target region; and

(c) a primer which is capable of binding to a strand of the target molecule, wherein the first nucleotide sequence prevents an amplicon, formed by amplifying the target region with the primer, from self-priming and the second nucleotide sequence results in the amplicon having two complementary regions which hybridize to form a partial loop, thereby forming a single stranded (ss) DNA amplicon which remains in this single stranded state.

The strand to which the primer binds may be the sense strand or the antisense strand of the target molecule. In particular, a reverse primer may bind to the sense strand and a forward primer may bind to the antisense strand.

The portion of the target region to which the second nucleotide sequence is complementary may be sufficiently close to the 5' end of the target region to cause the complementary sequences to hybridise to form the partial loop.

The nucleic acid molecule according may further include a third nucleotide sequence at the end of the second nucleotide sequence, the third nucleotide sequence comprising at least two nucleotides which are complementary to at least the last two nucleotides at the 3' end of the primer.

A fourth nucleotide sequence may link the second nucleotide sequence and the primer or the third nucleotide sequence and the primer.

The nucleic acid molecule may further comprise a restriction enzyme recognition site, such as a BsoB1 recognition site. The restriction enzyme recognition site may be located within the primer or between the first and second nucleotide sequences. The target molecule may be from a pathogen, such as a virus or microbe. For example, the target molecule may be HIV-1 or a region thereof. The target region may be SEQ ID NO: 1 or a sequence which has at least 90% identity thereto.

The first nucleotide sequence may be from about 3 to about 5 nucleotides in length or from about 15 to about 18 nucleotides in length. The second nucleotide sequence may be from about 7 to 12 nucleotides in length, and may have a GC content of about 50%.

The nucleic acid molecule may have at least 90% identity to any one of SEQ ID NOs: 3 to 10.

According to a further embodiment of the invention, there is provided a primer set for amplifying a target region of a nucleic acid molecule, the primer set including a forward primer comprising a nucleic acid molecule substantially as described above and a reverse primer. The reverse primer may include:

(a) a sequence specific for the target region;

(b) a restriction enzyme recognition site; and

(c) a sequence which is non-specific to the nucleic acid molecule.

According to a further embodiment of the invention, there is provided a kit for carrying out a point of care test, the kit comprising:

(a) a nucleic acid molecule or a primer set substantially as described above;

(b) reagents for extracting nucleic acid from a patient sample;

(c) reagents for amplification of the nucleic acid; and/or (d) means for detecting the presence of a target region in the nucleic acid.

The detecting means may be a dipstick or microfluidic device.

According to a further embodiment of the invention, there is provided a method of amplifying a target region of a target nucleic acid molecule, the method including the steps of:

(a) mixing a nucleic acid sample with a nucleic acid molecule substantially as described above;

(b) allowing the nucleic acid molecule to bind to the target region of a strand of the target molecule if the target molecule is present in the sample; (c) initiating strand polymerization to generate a double stranded DNA amplicon;

(d) extending the 3' end of the strand to which the primer of the nucleic acid molecule is bound by copying the second and first nucleotide sequences of the nucleic acid molecule, thereby forming a strand with two complementary regions at its 3' end;

(e) displacing the two strands of the double stranded DNA amplicon; and (f) allowing the complementary regions of the strand in step (d) to hybridize to each other and form a partially looped structure at its 3' end, thereby maintaining the strand in a single-stranded state.

The method may further include an initial reverse transcription step (RT) if the target molecule is RNA, so as to convert RNA to cDNA. The strand of the target molecule to which the nucleic acid molecule binds may be a sense or antisense strand.

The method may further include the step of cleaving the strands of the DNA amplicon at a recognition site with a restriction enzyme after step (c).

Steps (c) and (e) may be initiated using Bst DNA polymerase and BsoBI restriction enzymes.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 : Strand Displacement Amplification (SDA) (Adapted from Schweitzer and S.

Kingsmore, 2001 ).

Figure 2: Primer design according to a first embodiment of the invention.

Figure 3: Linear amplification method according to the invention.

Figure 4: Semi-cycling amplification method according to the invention.

Figure 5: MetaPhor gel (2%) showing PCR products digested with βsoSI restriction enzyme.

On the gel: 1 - DNA ladder (50 bp); 2, 3 - PCR product amplified from ES DNA (positive control), digested product (2) and undigested (3); 4, 6 - digested PCR products amplified from the different cDNA templates; 5, 7 - the corresponding undigested PCR products.

Figure 6: Sequence alignments.

Sequences obtained from four amplification reactions carried out according to the invention performed using four different combinations of primers are aligned with the ES DNA sequence (i.e. DNA sequence equivalent of RNA sequence of the ES RNA transcript that was used as a template in the RT- LDA reactions). The sequence of ES DNA (128 bp; SEQ ID NO: 1) is highlighted in bold; sequences of primers/primer binding sites are underlined. RT-LDA1 product (-105 bp; SEQ ID NO: 15) was amplified and sequenced using primers Fw#2/Rev#1 ; RT-LDA 2 product - Fw#3A/Rev#1 (-149 bp; SEQ ID NO: 16); RT-LDA 3 product - Fw#1C/Rev#1 (-170 bp; SEQ ID NO: 17); RT- LDA 4 product was PCR amplified and sequenced using the primer set Fw#3A/Rev#1 (-120 bp; SEQ ID NO: 18). (Note: for RT-LDA 4 product Fw#3A was used instead of Fw#4a since it has a much shorter overhang for more optimal PCR and sequencing reactions).

Figure 7: MetaPhor^® gel (1%) showing PCR product amplified from the linear RT-LDA reactions.

1- DNA ladder (50 bp); 2- ES DNA directly amplified by PCR using primers Fw#1C/Rev#1 (-170 bp; positive control for this primer set); 3 - PCR product amplified from the method of the invention (RT-LDA) using primers Fw#1C/Rev#1 (-170 bp specific product and - 90 bp non-specific product); 4 - corresponding RT-LDA blank sample containing only a non-specific product; 5 - ES DNA control directly amplified with primers Fw#3A/Rev#1 (-150 bp); 6- PCR product amplified from RT-LDA using primers Fw#3A/Rev#1 (-150 bp); 7 - corresponding RT-LDA blank sample amplified by PCR (no product); 8 - PCR blank sample (no template control - NTC) showing no amplification product.

Figure 8: Melting curve analysis using FRET probes for detection of RT-LDA product.

Top graph: Melting curve profiles of both RT-LDA samples (bold lines) show the presence of the specific amplification product with a typical melting peak at -69⁰C. Bottom graph: No melting peak is observed in the RT-LDA blank sample (grey line).

Figure 9: Detection of RT-LDA product using FRET probes.

After 90 minutes of RT-LDA melting peaks (~Tm at 68-69⁰C) are observed in all samples from A1 to A6 with the corresponding concentrations: 4x10³,

4x10⁴, 4x10⁵, 4x10⁸, 4x10⁷, 4x10⁸ ES RNA copies/ml. These melting curve profiles reflect the presence of an HIV-specific amplification product detected with the FRET probes. The blank sample (bold line) shows a peak at -84⁰C, which is either due to the presence of a non-specific product or represents a detection artefact. This type of high melting peak was non-reproducible between the experiments.

Figure 10: Melting curve analysis using FRET probes for verification of PCR product.

Melting peaks at -53⁰C indicate the presence of a specific amplification product in the samples: 1 - ES DNA control; 2 and 3 - RT-LDA samples with input template concentrations of 4x10³ and 4x10⁶ ES RNA copies/ml respectively; 4 - PCR blank sample shows no detection of product. (Note: The temperature of melting peaks for PCR products is lower (-53⁰C) than for RT-LDA products (68-69⁰C) even though the same FRET probes are used for detection. This is due to different reaction mixes and salt compositions used in each type of amplification.)

Figure 11: MetaPhor^® gel (1%) showing PCR product amplified from the partially cycling RT-LDA reactions.

On the gel: 1 - DNA ladder (50 bp); 2 - ES DNA used as a positive control for PCR shows the presence of a specific product (~150 bp); 3 and 4 - PCR product obtained from further amplification of RT-LDA product (ES RNA input in the RT-LDA reaction was 4x10⁸ and 4x10⁷ copies/ml respectively); 5 and 6 - NTC (no template control) samples for RT-LDA and PCR respectively shows the presence of primer dimers only.

Figure 12: Melting curve analysis for repeated RT-LDA experiments.

RT-LDA reproducibility experiments (2 out of 5) are illustrated in figures A and B₁ respectively. Specific RT-LDA amplification product is detected using

FRET probes and melting curve analysis on the LightCycler^® platform (Tm -68.8 -71⁰C) for the ES RNA concentration range of 4x10³-4x10⁸ copies/ml. No amplification product is detected in NTC samples.

Figure 13: Detection of RT-LDA amplicons (arrow) using NALF dipsticks. From left to right: RT-LDA blank sample (lane 1 ); two positive controls containing an artificially synthesized looped amplicon (ampi29) (lanes 2 and 3); the next 6 dipsticks show the detection of diluted RT-LDA amplification products obtained from the different experiments (lanes 4-9).

Figure 14: Autoradiograph showing extracted PCR product following semi-cycling RT- LDA reaction.

Autoradiograph of PAGE gel following RT-LDA reaction and phenol/chloroform ethanol precipitation. Lane 1 : non-extracted control showing smear; Lane 2: RT-LDA extracted sample showing 173 bp amplicon; Lane 3: extracted water blank showing bands of various sizes but the correct sized amplicon is absent; Lane 4: RT-PCR control amplicon of 173 bp.

DETAILED DESCRIPTION OF THE INVENTION

This invention describes a novel primer design and amplification method utilising this primer.

The amplification method generates a ssDNA product from either RNA or DNA and can be directly combined with detection systems such as NALF, flow cytometry, microarrays and the like. Primers according to the invention can be included in a kit for carrying out the amplification method and combined with a detection system for use in an assay, such as a POC test.

The primers include a short non-target specific sequence for preventing a displaced amplicon from self-priming; a sequence complementary to a strand extended from the primer ('overhang" sequence); a short artificial sequence that forms a link; and a sequence specific forward primer. The forward primer binds to the anti-sense cDNA strand and produces a dsDNA target. The overhang sequence is also extended and the anti-sense strand gets displaced from a nickable site, forming an amplicon with two complementary sequences. These hybridize, forming a partially looped ssDNA amplicon which is unable to bind to another forward primer and cannot be converted into a dsDNA target.

In this specification, the term "amplicon" is intended to refer to any amplified product, and not only to products which have been amplified by PCR. Primers designed according to the invention are referred to as "loop mediating primers" (or "LMPs") and the amplification method of the invention is referred to as "reverse transcription loop dependent amplification" (or "RT- LDA").

Strand displacement amplification (SDA) provides great power of amplification. It consists of two phases (target generation and exponential target amplification) and employs four primers (a pair of bumping primers and a pair of primers containing a nickable site). During the second cycling phase of SDA, dsDNA amplicons accumulate exponentially. The typical target doubling time is 20-30 seconds, and a 10 billion-fold amplification of specific targets can be achieved in less than 15 minutes. SDA also makes use of two commercially available enzymes - BsoB\ restriction enzyme and Bst DNA polymerase that possesses the strand displacement ability. In the case when RNA is used as a template, the reverse transcription (RT) step is performed (RT-SDA) by a third enzyme, Avian Myeloblastosis Virus (AMV) reverse transcriptase.

This process is shown in Figure 1. In the target generation phase (top panel) of SDA, a double-stranded DNA target (1) is denatured and hybridized with two primers (2). One primer (B₁) is designated as a 'bumper' primer, and the other primer (S-O contains a BsoBl restriction enzyme sequence 5' to the target binding region. The B₁ (3) and S₁ (4) primers are simultaneously extended by the thermostable enzyme Bst DNA polymerase in the presence of thiolated dCTP. Extension from the bumper primer displaces the S₁ extended product, which can then hybridize to the opposite strand primers, B₂ and S₂ (5). Extension of both of these primers produces species 6, which is utilized in the exponential target amplification phase of the reaction (lower panel). The strand that has been extended from the S₁ primer is nicked by SsoSI (7), but the complementary strand is refractory to cleavage because of the presence of a thiolated dCTP within the restriction site. DNA polymerase binds to the nick and begins synthesis of a new strand while displacing the downstream strand (8 — 10). This recreates the double-stranded species 7, and the process is repeated. The displaced strands bind to opposite strand primers, thus producing exponential amplification.

However, as SDA results in dsDNA, it is not suitable for use in POC HIV RNA assays, described in the introduction above.

The present invention therefore describes a novel primer and its use in a new isothermal amplification technique, based on SDA, which converts RNA or DNA into short ssDNA molecules (generally less than 200bp, although this will depend on each primer design) with high efficiency.

An overview of a primer design according to the invention is shown in Figure 2: 1) Starting from the 5' end of the primer: a short artificial (i.e. non-target specific) sequence that serves to prevent a displaced amplicon from self-priming is shown by black line; a sequence complementary to a strand extended from this primer (so called "loop" overhang) is indicated in hashed line; this sequence (hashed line) also possesses at least two nucleotides that are complementary to the last two nucleotides at the 3'-end of the forward primer (i.e. a gene specific part of this forward LMP); a short artificial sequence that forms a link between the "loop" overhang and a forward primer is shown by grey line; a sequence specific forward primer is shown by black circles.

2) The forward primer with an overhang (parts indicated in black line, hashed line and grey line) binds to the anti-sense cDNA strand and gets extended. Extension of the forward primer produces a dsDNA target. An overhang sequence also gets extended by, for example, Bst polymerase from the 3' end of the cDNA strand.

3A) The anti-sense strand gets displaced from a nickable site (similarly to SDA, see Figurei). The displaced anti-sense strand, which now can be termed an anti-sense ssDNA amplicon, contains a sequence complementary to an overhang of the forward primer. 3B) The two complementary sequences within one amplicon are in close proximity and the 3' end of the anti-sense amplicon folds forward and hybridizes to its complementary sequence (indicated by hashed line). As a result, a partially "looped" ssDNA amplicon is formed. The "loop" structure at the 3' end of the amplicon prevents it from binding to another forward primer and from getting converted into a dsDNA target. Thus, most of the displaced anti-sense amplicons remain single stranded.

The "loop" structure must be sufficient to preserve the single stranded nature of the amplicons and at the same time it must not interfere with extension of the primer. This technique was designed as a front amplification suitable for NAT based POC diagnostics assays, e.g. detection of genes, pathogens and diseases, including viruses and microbes, such as HIV-1 , tuberculosis or hepatitis.

The sense strand is thiolated with dCTP and is therefore not nicked and does not form a loop.

The main features of the primer of the invention and its use, and the differences from the prior art, are:

β Reverse transcription is performed using only one, for example HIV-1, sequence specific primer and a type of AMV enzyme that has both reverse transcriptase and RNase H activity (H⁺). As opposed to SDA, the method of the present invention does not use bumping primers. cDNA (anti-sense sequence) is released from the initial RNA/cDNA complex due to RNase H activity of AMV reverse transcriptase, which also makes the method different from SDA (Figures 3 and 4).

β Amplification is achieved mainly due to the new primer (LMP) design. The primer (in this study, a forward primer) has a 5' overhang containing a sequence complementary to a sense sequence of a DNA strand extended from this primer (Figure 2). During the RT- LDA reaction, dsDNA targets (T1, T2 and T3) are formed (Figure 4) and from each template displacement of multiple anti-sense strands/amplicons begins. During strand polymerisation, these strands/amplicons copy an overhang from the forward primer. Thus, the displaced anti-sense amplicons have two regions at their 3^! end with complementary sequences, which hybridize to each other, causing a partially "looped" structure of the ssDNA amplicons (Figures 2, 3 and 4). As a result of this primer modification, the displaced amplicons assume a "loop" conformation at their 3' end. This

"looped" structure prevents the displaced anti-sense amplicons from binding to a complementary forward primer and thus prevents their conversion into the double stranded DNA amplicons (dsDNA). In other words, the primer design of the present invention preserves the single stranded structure of the DNA amplicons produced during RT-LDA.

β Due to this novel primer design, the amplification technique does not have an exponential, cycling phase, which is characteristic for SDA (Figure 1).

β The final amplification products are partially "looped", ssDNA amplicons (Figure 3 and 4) as opposed to dsDNA amplicons produced by SDA. Thus the applicant has named this new technique reverse transcription bop dependant amplification (RT-LDA). e The new primer design can be used in combination with enzymes other than those described herein (these are the enzymes used in a commercial SDA technique as described above), provided that a DNA polymerase used has strand displacement ability. If a different restriction enzyme (not SsoBI as in SDA and RT-LDA) is used, than the nickable site in the primer will have a different sequence, which would correspond to the restriction enzyme used. Different RT (reverse transcriptase) enzymes can be used instead of AMV, in combination with a separate RNase H+ enzyme.

There are three requirements for POC testing: 1) sample preparation, which would include nucleic acid extraction from blood/plasma; 2) amplification of the nucleic acid target; and 3) detection of the nucleic acid amplicon. For this invention, once combined with a suitable front-end extraction method, the RT-LDA amplification, together with a dipstick based detection format, would be performed at the bedside.

The invention also extends to a kit for use in performing the POC test. The kit would include reagents for the extraction method, amplification reagents (RT-LDA reagents as described on pg 15) and dipsticks and/or a microfluidic device for detection.

Examples

The invention will now be described in more detail below with reference to the following non- limiting examples. In these examples, HIV-1 has been used as an example of a target sequence to be amplified. A person skilled in the art, however, will understand that the primer and amplification method of the invention can be readily applied to any other target molecule, for example, other viral and microbial pathogens or genes.

Two embodiments of the RT-LDA process are described below: an initial design for linear amplification and a final design for semi-cycling amplification. Figures 3 and 4 provide detailed descriptions of linear and semi-cycling RT-LDA designs, respectively. Both designs produced the same final product - partially "looped" ssDNA amplicons with anti-sense sequence, but the amount of final RT-LDA product was much larger in the semi-cycling technique. The main difference between the linear and semi cycling RT-LDA is in the design of the forward primer (Figures 2 and 3 (linear) and 4 (semi-cycling)). This type of primer design leads to semi-cycling of the reaction via formation of three types of dsDNA targets (T1 - T3). Semi-cycling RT-LDA includes two simultaneously occurring phases: linear and semi- cycling phases (Figure 4).

In vitro transcribed RNA containing 128 bases of HIV-1 gag gene (ES RNA) was used in the examples. Detection of the amplification product was performed using FRET hybridization probes and the LightCycler^® platform. This detection format was selected because it is cost effective, sensitive and sequence specific. Visual detection of amplicons using dipsticks was performed to confirm compatibility of the amplification technique of the invention with the NALF detection format.

Example 1 : Linear amplification method

A linear amplification method performed according to the invention is shown in Figure 3, with HIV-1 RNA as an initial template for amplification. Firstly, there is a reverse primer (R) consisting of three regions. Starting from the 5' end they are as follows: an artificial (random) sequence, a recognition site sequence (so called nickable site) for the restriction enzyme BsoB\ (indicated as a white circle), and an HIV-1 sequence specific (reverse) primer. The reverse primer (R) binds viral RNA and gets extended by the RT activity of AMV reverse transcriptase (RT). A double stranded RNA/cDNA hybrid is formed. RNase H activity of AMV digests the viral RNA in the hybrid leaving the cDNA (anti-sense sequence) (AMV H⁺). The forward primer (F) converts cDNA into dsDNA target (LDA 1). The forward primer (F) includes a forward primer with a 5' overhang. From the 5' end, the primer overhang contains "an artificial" sequence (black line), a sequence complementary to an amplicon ("loop" sequence) (hashed line) and a very short linking sequence (grey line). Due to the addition of chemically modified dCTPs (dCTP-αS) in the reaction mix, the double stranded SsoBI recognition site is hemithiolated and only one strand gets nicked by the restriction enzyme (see also Figure 4). Bst DNA polymerase, which also has strand displacement ability, extends the 3'-end of the newly formed nick and at the same time displaces the existing, anti- sense DNA strand (LDA 2). During this process of DNA strand extension/displacement, the SsoSI nickable site gets regenerated and can be nicked over and over again, thus producing a large number of "looped" anti-sense ssDNA amplicons. The displaced amplicons assume a partially "looped" structure due to a carefully designed overhang of a forward primer, as described in Figure 4. Formation of partially "looped" amplicons prevents binding of the forward primer to the anti-sense amplicon and thus prevents the entire amplification process from cycling. The RT-LDA design illustrated in this figure provides accumulation of partially "looped" ssDNA anti-sense amplicons in a linear fashion.

Example 2: Semi-cycling amplification method

A primer and its use in a semi-cycling amplification method according to the invention are illustrated in Figure 4. Conversion of the initially designed linear amplification technique into semi-cycling was made possible due to another novel design applied to the forward primer.

Similarly to the first primer design (Figures 2 and 3), the second design includes an HIV-1 specific forward primer (black circles) with a 5' overhang, but the structure of the overhang is different as it incorporates a nickable site (in Figure 3, only the reverse primer included the nickable site). From the 5' end the primer overhang contains "an artificial" sequence (black line), a BsoBl recognition site (white circle) and a sequence complementary to an amplicon ("loop" sequence) (hashed line) and a very short linking sequence (grey line). The RT step for the partially cycling amplification is identical to the linear amplification (Figure 3). A modified forward primer that binds to cDNA (anti-sense) is extended by Bst polymerase, and as a result a dsDNA target is formed (T1). The strands complementary to the BsoBl restriction site are generated during the primer/strand extension and thus contain modified dCTPαS (thiolated), which prevents BsoBl from cutting dsDNA. Instead, only the unmodified nickable sites (in all targets - T1 , T2 and T3) from the primers get nicked (single stranded nick) by the enzyme. T1 dsDNA has two nickable sites - one on a sense strand (forward primer; white circle) and another one on the anti-sense strand (reverse primer; white circle). Both nickable sites are recognized and nicked (cleaved) by BsoB). Next, the 3'- end of each nicked strand (sense and anti-sense) is extended by Bst DNA polymerase and the existing strands are displaced. The displaced sense strand forms a partially "looped" amplicon that has a sequence complementary to a reverse primer. The reverse primer binds this sense amplicon and converts it into a dsDNA target (T2). T2 has only one βsoβl restriction site and a process of nicking and strand displacement/polymerization occurs on the T2 target in a similar way as described above. A cascade of processes between formations of T1 and T2 represents a linear phase of a RT-LDA. The displaced anti-sense amplicons have a sequence at their 3'- end which is complementary to the "loop" overhang of a forward primer. Thus, once displaced, these amplicons assume a partially "looped" structure. The anti-sense strand, which is displaced from T1 , possesses a 3' sequence complementary to the entire forward primer and thus a forward primer binds to the anti-sense amplicon. Extension of the forward primer converts the anti-sense strand into dsDNA target (T3). Target T3 produces "looped" sense amplicons similarly to T1. These amplicons bind to the reverse primers which lead to the formation of dsDNA targets identical to T2. A cascade of processes from the formation of T1 to T3 and then to T2 represents a partially cycling phase of this amplification technique. T2 generated during a partially cycling phase also keeps producing "looped" ssDNA anti-sense amplicons. During this type of amplification, both linear and semi-cycling phases of RT-LDA occur simultaneously and generate large amounts of a final amplification product - partially "looped" ssDNA anti-sense amplicons. The partially cycling design provides at least a 3-fold increase in magnitude of amplification.

Isothermal Amplification - RT-LDA

Loop Mediating Primers (LMPs)

Primers were designed for a sequence of an in vitro transcribed RNA that represents an external quantitation standard (ES RNA) used for a LightCycter^® viral load assay (LUX assay), described in PCT Publication No. WO 2006/082496, the entire contents of which are incorporated by reference herein. The ES RNA molecule includes a 128 base pair region of the HIV-1 gag gene:

5'- acatcaagca gccatgcaaa tgttaaaaga tagaatcaat gaggaggctg cagaatggga tagaatacat ccagtacatg cggggcctat tgcaccaggc caaatgagag aaccaagggg aagtgaca - 3' (SEQ ID NO: 1 )

The primer design of the invention (LMP) was applied only to forward primers used in the amplification method. Seven types of forward primers having different lengths and base compositions of the loop forming overhang were evaluated during this study. These seven primers were individually combined with a reverse primer (Rev#1 ) in separate amplification reactions. The sequences of primers used in this study are as follows (underline = HIV-1 forward primer; bold = sequence complementary to a region of the strand extended from the forward primer; italics = nickable site; normal = random artificial sequence; spaced sequences = an amplification stopper sequence at the 5' end and a link if in the middle of the primer sequence):

Reverse primer:

5'- GTCGACTTGC ATGCATGTCT CGGGTGTCAC TTCCCCTTGG TTCT CTCA - 3' (48 mer) (SEQ ID NO: 2)

Forward #1

5' - AAC AGCCTCC TCATTGATTG TATCTTTTAA CAT GAA ACAT

CAAGCAGCCATGCAAAT - 3' (57 mer) (SEQ ID NO: 3)

Forward #2

5' - GAT ATTCTGC AG GAA ATACA ATCAATGAGG AGGC T - 3' (35 mer) (SEQ ID NO: 4)

Forward #1A 5'-AGA ATT GAT TGT ATC TTT TAA CAT ACA TCA AGC AGC CAT GCA AAT - 3' (45 mer) (SEQ ID NO: 5)

Forward #1 B

5' - ACC TGT ATC TTT TAA CAT ACA TCA AGC AGC CAT GCA AAT - 3' (39 mer) (SEQ ID NO: 6)

Forward #3

5' - GAG GCA GCC TCC TCA TGC AAT TGT TAA AAG ATA CAA TCA AT - 3' (41 mer) (SEQ ID NO: 7) Forward #3 A (a "new" version of Fw#3)

5'- TCT CC TCC TCAT ACT TGT TAA AAG ATA CAA TCA AT- 3' (35 mer) (SEQ ID NO:

8)

Forward #1C (a "new" version of Fw#1 )

5' - GGC C TTT TAA CAT AGA ACA TCA AGC AGC CAT GCA AAT - 3' (37 mer) (SEQ ID NO: 9)

Forward #4A (a version of#3A with a nickable site for semi-cycling RT-LDA) 5'- AAC CTA TCC GGA CAA CGA TAA CCC GGG CC TCC TCA T ACT TGT TAA AAG ATACAATCAAT- 3' (59 mer) (SEQ ID NO: 10)

RT-LDA reaction conditions

RT-LDA amplification reactions were performed in a total volume of 50 μl consisting of 20 μl of template RNA and 30 μl of reaction mix. The reaction mix was assembled as follows: 35 mM potassium phosphate buffer (K₁PO₄), 5% v/v dimethyl sulfoxide (DMSO; Sigma-Aldrich, USA), 4% v/v glycerol (Promega, Madison, Wl, USA), 5 μg acetylated bovine serum albumin (AcBSA; Promega, Madison, Wl), 0.8 mM 2'-deoxycytidine 5'-O-(1- thiotriphosphate) (dCTPαS; Amersham Biosciences, Piscataway, NJ, USA), 0.6 mM dUTP, 0.2 mM each dATP and dGTP (Promega, Madison, Wl, USA), 7.5 mM magnesium acetate (MgOAc₂; Sigma- Aldrich, USA), 7.5 units AMV reverse transcriptase (USB Corp., Cleveland, Ohio, USA), 32 units BsoBl restriction enzyme (New England BioLabs Inc., USA), 20 units Bst DNA polymerase (New England BioLabs Inc.), 1 μg T4gp32 ssDNA binding protein (USB Corp., Cleveland, Ohio, USA), forward and reverse primers 0.75 μM each. All reaction components and RNA template were mixed and incubated at 53 ⁰C for 1 hour using MyCycler™ thermal cycler (BIO-RAD Laboratories, Inc., USA).

Synthetic in vitro RNA (ES RNA) was used as a template for RT-LDA reactions. The sequence was based on a reference gag sequence from the HIV-1 subtype B HXB2 isolate (http://hiv-web.lanl.gov).

Detection of RT-LDA amplification product using FRET hybridization probes and NALF

Detection of RT-LDA product using FRET hybridization probes and the melting curve analysis was performed using the LightCycler^® platform, version 1.2 and the software version 4 (Roche Applied Science, Mannheim, Germany). FRET hybridization probes were designed for the anti-sense amplicons and their sequences are: 5' - TAG AAT ACA TCC AGi ICA TGC GG - Fluorescein - 3' (SEQ ID NO: 11 ) 5'- LCRed -640 - CCT HT GCA CCA GGC CTA ATG AG - p - 3' (I = Inosine) (SEQ ID NO: 12)

Probes were manufactured by Metabion International AG (Germany). Reconstituted probes, each at 100 μM concentration, were kept at -20⁰C until use.

Post amplification detection using FRET probes was performed in 30 μl volume using 27 μl of amplification reaction and 0.5 μM of each probe. Profile of the melting curve analysis was as follows: 95⁰C for 5 s, 40⁰C for 1 min and gradual heating up to 85-9O⁰C with 01°C/s ramping rate and continuous fluorescent signal acquisition. Melting curves were viewed using a combination of channels -640/Back530.

Reagents for nucleic acid lateral flow, including buffers, magnetic conjugate, nitrocellulose DNP striped half dipsticks, capture and detection probes were provided by British Biocell International (Cardiff, UK). Sequences of the LF probes:

Ampi 2032 - Detection probe (two inosines)

5' Biotin TCTGC ACA TCC AGI ICA TGC GGG - 3' (SEQ ID NO: 13)

Arnpj 2033 - DNP - capture probe (two inosines)

5' - GCC TII TGC ACC AGG CCT - DNP- 3' (SEQ ID NO: 14)

The positive control for NALF detection was also supplied by BBI. Positive control (ampi29) was made to imitate RT-LDA amplicons and represents a synthetic ssDNA molecule that has a partially looped "structure" and the sequence similar to the real amplification product. NALF detection was performed according to the manufacturer's instructions.

Conventional PCR for detection of RT - LDA product

PCR was performed in 50 μl volume using 25 μl of ReadyMix™ Taq PCR reaction with MgCI₂ (Sigma, St. Louis, Missouri, USA), 0.5 μM forward and reverse primer, 15 μl of template (diluted RT-LDA product) and PCR grade water from the kit (Sigma).

PCR was performed using MyCycler™ thermal cycler (BIO-RAD Laboratories, Inc., USA) according to the following profile: initial denaturation at 95⁰C for 3 min and then 40 cycles of amplification with denaturation at 95⁰C for 30 s, annealing at 52⁰C for 30 s, extension at 72⁰C for 30 s and final extension at 72⁰C for 7 minutes. Detection of PCR product was performed using high resolution MetaPhor^® gels and conventional gel electrophoresis. Fifteen microliters of amplification product was mixed with 5 μl of the gel loading dye (Fermentas UAB, Lithuania) and loaded onto a 1-2% MetaPhor^® agarose (BioWhittaker Molecular Applications, Rockland, ME, USA) gel prepared with IxTBE buffer (diluted from 1OxTBE stock; Fermentas UAB, Lithuania) and 3 μl of ethidium bromide stock solution (10 mg/ml; Sigma -Aldrich., St. Louis, Missouri, USA) per 100 ml of gel. Agarose gel electrophoresis was performed at 100V for 1-2 hours. Size verification of the product was performed by running 10 μl of O'GeneRuler™ 50 bp DNA ladder (Fermentas, UAB, Lithuania) alongside the sample. The gel was viewed under a UV light.

Sequencing of RT - LDA product

PCR product amplified using the RT-LDA reaction of the invention was used as a template for sequencing. Prior to sequencing, the PCR product was excised from the MetaPhor^® gel and purified using the MinElute™ Gel Extraction kit (Qiagen, GmbH Hilden, Germany) according to the manufacturer's instructions. DNA was eluted in 10 μl of the elution buffer supplied in the kit.

The sequencing reaction was performed with the Big Dye^® Terminator version 3.1 cycle sequencing kit from Applied Biosystems (Foster City CA, USA) according to the manufacturer's instructions, on the ABI Prism 3100-Λvanf Genetic Analyzer (Applied

Biosystems). Sequence data analysis was performed using the Sequencing Analysis version

3.3 program (Applied Biosystems) and assembled using Sequencher program version 4.1.4

(Genecodes, Ann Arbor, Ml). The edited sequence was aligned and compared with DNA equivalent to the ES RNA sequence using the Clustal X program.

Autoradiography for detection of RT-LDA product

A plasmid containing the applicants 128 bp sequence of interest was manufactured by Geneart (Regensburg, Germany) and run-off transcripts of 450 bp were synthesised by in vitro transcription off the plasmid. These larger RNA transcripts were then used as templates for further RT-LDA reactions and detection with autoradiography.

Further experiments for the detection of the RT-LDA product were carried out by incorporating 0.4 mM radioactively labelled dATP (deoxyadenosine S'[α-32P]) into the RT-LDA reaction mix. The synthetic in vitro RNA of -450 bp was used as a template for the reaction. Following 1 hour isothermal reaction at 53⁰C, a standard phenol/chloroform-ethanol precipitation was performed to clean up the reaction before running it on an 8% Acrylamide/7M Urea gel for 2 hours at 150 V. The gel was then fixed for 2 hours in a 10% acetic acid/10%methanol solution and dried. The gel was then exposed to X-ray film was overnight and developed.

Results

RT-LDA - optimisation and evaluation of a novel primer design for the linear type of amplification

Initial experiments for optimisation of the RT-LDA reaction were performed using a long incubation time and a two-step temperature profile with reverse transcription at 42⁰C for 1 hour and amplification (strand displacement/polymerization) at 53⁰C for 2 hours (3 hours of

RT-LDA in total). Template nucleic acid (i.e. ES RNA) was used at this stage of optimization only at high concentrations: ~ 4x10⁷ and 4x10^s copies/ml. Five forward primers (Fw#1 ,

Fw#1 A, Fw#1 B, Fw#2 and Fw#3) combined individually with the reverse primer (Rev#1 ) were first evaluated for the linear type of RT-LDA. Three forward primers (Fw#1 , Fw1 A, Fw#1 B) out of five had the same HIV-1 specific sequence, but a different 5' overhang length. Forward primer Fw#2 is positioned more internally to Fw#1 and has the shortest "loop" overhang" with

-50% GC content. Primer Fw#3 is positioned between Fw#1 and Fw#2 and partially overlaps with primer Fw#2. Each of these primers was combined individually with the reverse primer Rev#1 in separate RT-LDA reactions. Initial RT-LDA experiments revealed no amplification product for all five primer combinations using 1% MetaPhor^® gel and FRET probes for detection.

In order to exclude failure of amplification due to non-optimal RT-LDA conditions rather than a non-optimal primer design, the individual steps of the amplification reaction were assessed. To troubleshoot the RT step, ES RNA was reverse transcribed using SKT145 forward primer (a forward primer specific for HIV-1 gag gene and used in the LUX assay for HIV-1 viral load) and Rev#1 primer. Two different reaction mixes (RT-LDA reaction mix excluding βsoSI and Bst enzymes and the reaction buffer supplied with AMV enzyme) and different concentrations of AMV enzyme were used. RT was performed for 1 hour at 42⁰C and 20 μl of undiluted RT reaction (cDNA) was than amplified by conventional PCR using primers SKT145 and Rev#1. The PCR product was confirmed using 1% MetaPhor^® gel. Troubleshooting of the RT step revealed that the 2.5 units of AMV enzyme used initially was insufficient and an optimal concentration of 7.5 units of AMV per reaction was established. Next, cDNA reverse transcribed with primer Rev#1 under optimised RT conditions was PCR amplified using the set of primers SKT145/Rev#1. PCR product was incubated with BsoB\ restriction enzyme at 53⁰C for 3 hours. Digested and undigested PCR products were verified using 2% MetaPhor^® gel (Figure 5). The presence of the digested product confirmed functional activity of BsoB\ at 53⁰C and the presence of a corresponding restriction site in the reverse primer Rev#1. Following this troubleshooting and optimisation of the RT step, RT-LDA reactions were performed using the combination of five forward primers described above and the Rev#1 primer. As with the first round of experiments, no RT-LDA amplification product was detected for all primer combinations using MetaPhor^® gel and FRET hybridization probes. Next, PCR was employed to confirm the presence/absence of RT-LDA product and to exclude insufficient sensitivity of the detection formats used. RT-LDA reactions were diluted 1 :50 and 1 :100 in molecular grade water and 15 μl of the diluted product was used for PCR. PCR was performed with the same primer combinations that were originally used for each RT-LDA reaction (e.g. Fw#2/Rev#1 , Fw#1A/Rev#1 , etc.). For each primer combination, the DNA equivalent of ES RNA (i.e. purified dsDNA amplification product obtained using ES RNA and a conventional RT-PCR) was used as a positive control and a corresponding RT-LDA no template control (NTC) sample was use as a negative control. PCR revealed the presence of RT-LDA product only in the reaction performed with a set of primers Fw#2 and Rev#1. Together with a specific amplification product, the RT-LDA reaction and the blank sample revealed the presence of a non-specific product (data not shown). Since the non-specific product was found in a blank RT-LDA sample as well as in RT-LDA reaction, it is likely to be due to the presence of partially complementary regions in primers Fw#2 and Rev#1 , causing formation of this product under the low stringency isothermal amplification conditions. The specific RT-LDA product further amplified by PCR was excised from the gel, purified and used for sequencing with a set of primers Fw#2 and Rev#1. The nucleotide sequence of PCR amplified RT-LDA aligned with the ES DNA is shown in Figure 6. These experiments demonstrated successful RT-LDA using a pair of primers Fw#2 and Rev#1. However, due to the presence of a non-specific amplification product primer design, further improvement was required.

Another two forward primers (Fw#3A and Fw#1C) were thus designed and evaluated in combination with the primer Rev#1. RT-LDA reactions were performed using a two step temperature profile and a long amplification time as described above, and two combinations of primers: Fw#1C/Rev#1 and Fw#3A/Rev#1. FRET hybridization probes revealed RT-LDA product amplified using both newly designed primers (Fw#1C and Fw#3A), but the fluorescent signals were weak (data not shown).

The presence of the RT-LDA product was confirmed with conventional PCR. RT-LDA samples amplified with primers Fw#1C/Rev#1and Fw#3A/Rev#1 and their corresponding blank samples were diluted 1 :50 and 15 μl of each diluted sample was further PCR amplified with the corresponding RT-LDA primers. ES DNA was used as a positive PCR control for both combinations of primers. A PCR product was detected in the RT-LDA samples and in two positive control samples, but not in the RT-LDA blank samples (Figure 7). Non-specific product observed using initially designed primers was not present in any of these PCR samples. Significantly less PCR product was generated using the set of primers Fw#1C/Rev#1 than using the set of primers Fw#3A/Rev#1 , which reflects lower efficiency of RT-LDA reactions performed with primers Fw#1C/Rev#1 (Figure 7). The PCR products obtained with both sets of primers were gel purified and sequenced. Sequencing revealed the presence of a HIV-1 specific sequence in PCR products amplified using both sets of primers. (Figure 6)

RT-LDA - optimisation and evaluation of a novel LMP primer design for the partially cycling type of amplification

The next phase in the development process involved a further novel modification that was applied to RT-LDA forward primers (LMPs) and converted RT-LDA into a partially cycling amplification method. Based on the results of preliminary optimisation and evaluation of the linear type of RT-LDA, the forward primer Fw#3A appeared to have an optimal design. Thus, the LMP design for a partially cycling RT-LDA incorporated the features of primer Fw#3A. In particular, primer Fw#4A for a partially cycling RT-LDA includes the same HIV-1 specific sequence and the loop overhang sequence as the primer Fw#3A and in addition it has a nickable site for SsoSI and the random primer sequence (Figures 3 and 4). New forward primer Fw#4A designed for partially cycling RT-LDA was evaluated in combination with primer Rev#1 using high concentrations of ES RNA ~ 4x10⁷ and 4x10⁸ copies/ml. Detection with FRET probes demonstrated the presence of an amplification product in RT-LDA reactions performed with primer set Fw#4A/Rev#1 and no product in the blank sample (Figure 8). The characteristic melting temperature (T_m) observed for RT-LDA amplicons using these FRET probes was ~68.8-71°C. Unambiguous detection of RT-LDA amplicons with FRET probes was achieved and allowed further optimisation of the amplification method. Thus, for optimization and preliminary evaluation of partially cycling RT-LDA reactions, only the post- amplification melting curve analysis with FRET hybridization probes was used for detection of product. RT-LDA reaction components, such as DMSO and phosphate buffer, affect the specificity and reproducibility of amplification. The concentration of DMSO was titrated from 4 to 10% v/v and the concentration of phosphate buffer from 35 mM to 50 mM. It was found that an increased concentration of DMSO requires higher concentrations of phosphate buffer to maintain sufficient amplification. Preliminary data obtained from a NALF/MAR™ (magnetic assay reader) feasibility study conducted by BBI demonstrated that higher concentrations of phosphate buffer decrease the intensity of magnetic signal detected with MAR™. Thus, high (8-10%) concentrations of DMSO that could provide greater specificity of primer annealing required higher concentrations of phosphate buffer, which interferes with the MAR™ detection.

Taking into account the desirability for compatibility of RT-LDA with the MAR™ detection platform, the most optimal concentrations of both components were established at 35mM for phosphate buffer and 5% for DMSO. Next, the concentrations of primers Fw#4A and Rev#1 were titrated in the range of 0.5 μM, 0.75 μM and 1 μM. Each concentration of primers was tested in duplicate using two ES RNA concentrations of 4x10⁷ and 4x10⁹copies/ml. A concentration of primers at 0.75 μM was found to be the most optimal for both concentrations of template ES RNA tested (data not shown). Further optimization of the RT-LDA aimed to reduce the total amplification time (< 3 hours) and perform the amplification at only one constant temperature (53⁰C). The initial experiment for time reduction was performed using a two-step temperature profile as follows: RT step at 42⁰C for 30 min and amplification step at 53⁰C for 1 hour. Total reaction time was reduced from 3 hours as previously used to 1.5 hours using this approach. A wide range of ES RNA concentrations was used: 4x10³, 4x10⁴, 4x10⁵, 4x10^δ, 4x10⁷ and 4x10⁸ copies/ml. RT-LDA amplicons were detected in all six samples using FRET (fluorescence resonance energy transfer) probes and the melting curve analysis on the LightCycler^® platform (Figure 9).

Based on this successful experiment, a further reduction in amplification time was attempted using only one constant temperature for both the RT step and amplification steps. Two amplification profiles were tested for the RT-LDA reactions: one at 53⁰C with a total reaction time of 30 minutes and at 53⁰C for 1 hour. Both amplification profiles were performed using a range of ES RNA concentrations: of 4x10³, 4x10⁴, 4x10⁵, 4x10⁶, 4x10⁷ and 4x10⁸ copies/ml. The RT-LDA product was detected using FRET probes in all samples (excluding a blank sample) amplified for 30 minutes and 1 hour, respectively. Three RT-LDA samples were used for PCR to provide an additional confirmation of successful RT-LDA using new time and temperature profiles. In particular, the RT-LDA samples, containing ES RNA at a starting concentration of 4x10⁷ and 4x10⁶ copies/ml that were amplified for 1 hour and a sample containing 4x10³ copies/ml of ES RNA amplified for 30 minutes, were diluted 1 :10 and further amplified by PCR using primers Fw4A/Rev#1. A PCR product for the RT-LDA sample of 4x10⁷ copies/ml was clearly detected as a bright band on the 1% MetaPhor^® gel; the other two samples revealed faint bands (data not shown). In order to confirm the presence of the specific product in these faint bands, 18 μl of both (4x10³ and 4x10⁶ copies/ml) PCR reactions and 18 μl of the PCR product of the ES DNA control were used for FRET detection on the LightCycler^®. The specific PCR product was detected using the melting curve analysis in all samples except the blank sample (Figure 10). RT-LDA product obtained with high ES RNA concentrations (4x10⁷ and 4x10^a copies/ml) and primers Fw4A/Rev#1 was amplified again using conventional PCR and primers Fw3A/Rev#1. Primer Fw#3A has the same HIV-1 sequence specific region as the primer Fw#4A and was used due to the limited amounts of the primer Fw#4A available. PCR product (-150 bp) was gel purified and used for sequencing (Figure 11 ). The sequence obtained using primers Fw#4A and Rev#1 was aligned with the ES DNA (Figure 6).

Since successful RT-LDA was demonstrated using 1 hour and even 30 minutes amplification time at 53⁰C, initial evaluation of the method was performed to assess reproducibility of partially cycling RT-LDA under the new conditions. Reproducibility of RT-LDA was studied in five different experiments performed on five different days. Each experiment was performed using serial dilutions of a template ES RNA covering a range of concentrations: 4x10³, 4x10⁴, 4x10⁵, 4x10⁶, 4x10⁷ and 4x10⁸ copies/ml. RT-LDA reactions were performed at 53⁰C for 1 hour. Post-amplification detection of amplicons was performed using melting curve analysis and FRET hybridization probes. RT-LDA product was detected using FRET probes in all samples in each of the five repeated runs. Thus, no failed amplification was observed in any of the RT-LDA reactions performed on five different days using a concentration range of ES RNA covering six orders of magnitude. Figure 12 (A; B) shows the melting curve analysis for two out of five repeated RT-LDA experiments.

Preliminary evaluation of partially cycling RT-LDA also involved detection of the amplicons using anti-DNP (dinitrophenyl) striped dipsticks and NALF. RT-LDA reactions were performed using the short amplification profile described above and ES RNA template at concentrations of 4x10⁸ and 4x10⁷ copies/ml. Different dilutions (from 1 :10 to 1:30) and input concentrations (from 1 μl to 30 μl) of the diluted RT-LDA product were used for NALF detection. The samples that gave positive detection in a visual detection range of NALF are shown in figure 13. The most optimal detection was found in samples diluted 1 :10 and 10-30 μl of that dilution. Reproducibility of NALF detection of the RT-LDA amplicons was confirmed in a number of lateral flow experiments.

Further experiments, to detect the RT-LDA amplicon following amplification, involved autoradiography. In order to be able to visualise the amplicon, the RT-LDA reaction was cleaned up to remove any excess salts and proteins using a standard DNA phenol/chloroform ethanol extraction. The resulting extracted sample was run on a PAGE gel along with various controls; a non-extracted RT-LDA sample, an extracted RT-LDA water blank and a standard RT-PCR control amplicon (to use as a size reference). The amplicon of the correct size (-173 bp: 128bp gag sequence plus the primer overhang) could be visualised in the extracted sample but was not detectable in the non-extracted sample control due to smear formation (Figure 14). Bands of various sizes were also visible in the water template control sample. However, the correct sized amplicon was absent (figure 14).

Discussion

The development of a novel method for isothermal amplification of RNA, termed reverse transcription loop dependant amplification (RT-LDA) has been described. RT-LDA represents front-end amplification for a proposed/designed POC RNA based HIV-1 diagnostic assay and it is designed to fulfil the criteria required for POC diagnostics: rapid and efficient amplification of single stranded DNA amplicons using viral RNA template. The RT-LDA technique combines a known concept of strand displacement amplification using restriction enzyme BsoB\ and Bst DNA polymerase and a novel primer design. RT-LDA, like most amplification techniques, uses a reverse and a forward primer. The RT-LDA reverse primer, similarly to primers used for strand displacement amplification, possesses a BsoB\ recognition site that gets nicked, and this ssDNA nick initiates a series of strand displacement/polymerisation steps driven by Bst DNA polymerase. A novel primer design is applied to the RT-LDA forward primer - it possesses a 5' overhang sequence, which is complementary to the sequence of a sense strand extended from this primer. During the RT-LDA reaction, dsDNA targets (T1 , T2 and T3) are formed (Figure 4), and from each template, displacement of multiple anti-sense strands/amplicons begins. During strand polymerisation these strands/amplicons copy an overhang from the forward primer. Thus, the displaced anti-sense amplicons have two regions at their 3' end with complementary sequences, which hybridize to each other causing a partially "looped" structure of the ssDNA amplicons (Figures 2, 3 and 4). Accumulation of partially "looped" ssDNA amplicons represents a key feature of RT-LDA. The "loop" structure closes the region complementary to the forward primer and thus, prevents conversion of anti-sense ssDNA amplicons into dsDNA amplicons. Another novel aspect of the RT-LDA design is the absence of bumping primers that are normally used in the SDA technique. This is done to avoid the production of multiple dsDNA species in SDA (Figure 1) and to make RT-LDA less complex. Instead of bumping primers, RT-LDA makes use of AMV reverse transcriptase that has combined RNase H activity, which digests viral RNA in the RNA/cDNA hybrid, leaving cDNA in a single stranded form. Functionality of the novel modifications applied to the RT-LDA design has been proven experimentally.

Two variations of the RT-LDA technique were designed and evaluated namely: linear and semi-cycling (Figures 3 and 4). Originally the RT-LDA method with a linear type of amplification was designed. Partially cycling RT-LDA was later designed after preliminary evaluation of linear RT-LDA. Initial experiments were performed to set up a novel technique and find the most optimal design for a RT-LDA forward primer containing the "loop" overhangs (LMP design). Theoretically, the "loop" overhang has to be complementary to a short sequence at the 3' end of the forward primer and to the longer sequence of the extended sense strand. The length of a complementary "loop" sequence has to be long enough to preserve the single stranded structure of the DNA amplicons, but not too long to interfere with the extension of the forward primer. The exact dimensions of the "loop" overhang in terms of the number of nucleotides and GC content were established experimentally. This was achieved by performing RT-LDA reactions using 7 forward primers with 5'- "loop" overhangs of different lengths and sequence composition. According to the experimental data obtained for linear RT-LDA reactions, the most optimal design for RT-LDA forward primer suggests short sequence (9 bases) of a "loop" forming overhang with a high (~50%) GC% content. However, the amplicons produced with the linear RT-LDA method could not be sufficiently detected using FRET hybridization probes and the LightCycler^® platform. RT-LDA product from linear amplification could only be seen by further amplification using conventional PCR and was confirmed additionally using DNA sequencing. This issue was attributed to the insufficient power of amplification provided by the linear RT-LDA. However, linear RT-LDA can still be used in combination with more sensitive detection formats, and for applications where the amplification methods have a long incubation time.

An advanced RT-LDA design is achieved by the addition of the second BsoB\ nickable site into the sequence of the forward primer (Figure 4). This modification of a forward primer does not interfere with displacement of ssDNA partially "looped" anti-sense amplicons. The modified forward primer enhances amplification due to formation of three different types of dsDNA targets (T1 , T2 and T3) as apposed to only one such target in a linear type of RT-LDA design (Figures 2 and 3). Thus, another novel primer design converted linear RT-LDA into a partially cycling amplification technique. In particular, modification was applied to a forward primer Fw#3A, which was shown previously to produce the most optimal linear amplification. For ease of differentiating between primers, a modified Fw#3A containing a BsoB\ restriction site and a random sequence for "an anchoring artificial" primer was termed Fw#4A. RT-LDA experiments performed with new primer Fw#4A and Rev#1 immediately showed an improved detection using melting curve analysis with the FRET hybridization probes. Optimisation of reaction conditions allowed further improvement of RT-LDA. Reduction in amplification time of RT-LDA that was performed using one constant temperature (53⁰C) revealed good sensitivity of this method. In particular, 4000 copies/ml of ES RNA template could be detected using only 30 minutes amplification. The PCR product amplified from RT-LDA using the primer set Fw#4A/Rev#1 confirmed the presence of a specific product. For the fourth time during the development of the RT-LDA technique, the PCR product was sequenced and the sequence of the amplicons aligned with HIV-1 subtype C sequence (Figure 6). Preliminary evaluation of semi-cycling RT-LDA was performed using a wide range of ES RNA concentrations from 4x10³ to 4x10⁸ copies/ml, which were tested in five replicate RT-LDA experiments. Each RT-LDA run was performed at 53⁰C for 1 hour. Repeated experiments showed that the newly designed RT-LDA technique is reproducible over six orders of magnitude. Preliminarily sensitivity of a one hour RT-LDA reaction combined with the FRET detection format can be set at 4000 copies/ml. Preliminary evaluation of RT-LDA also included NALF experiments in order to demonstrate compatibility of a new amplification technique with NALF detection using nitrocellulose dipsticks as one of the most widely used in POC tests detection formats. It was important to exclude possible interference between NALF reaction components and RT-LDA reactions mix. Most importantly, the detection of RT-LDA product using NALF dipsticks, without a prior denaturation step, provides experimental proof of the single stranded structure of the partially looped RT-LDA amplicons. Positive detection observed for RT-LDA reactions (Figure 13) showed that the newly developed amplification technique is fully compatible with NALF detection using anti-DNP striped dipsticks and anti-superparamagnetic conjugates. In conclusion, an innovative loop mediating primer (LMP) design applied to a novel amplification technique, termed RT-LDA, has been shown that a partially cycling RT-LDA is reproducible over a wide range of template concentrations (4x10³ - 4x10⁸ copies/ml) and provides good sensitivity using only one hour of amplification. RT-LDA is fully compatible with NALF detection using dipsticks, and thus the new method can be used as an amplification front end for POC NAT based diagnostic tests (e.g. near-patient diagnostic assay for detection of HIV-1 RNA in plasma). These results warrant further optimization and evaluation of RT-LDA using instrument based detection of ssDNA amplicons captured on the dipsticks and application of this method to different targets (e.g. viral and microbial pathogens) will be carried out.

References

Hellyer, T.J., DesJardin, L.E., Teixeira, L., Perkins, M. D., Cave, M. D. and Eisenach, K.D. (1999) Detection of viable Mycobacterium tuberculosis by reverse transcriptase-strand displacement amplification of mRNA. J Clin Microbiol 37(3), 518-523.

Nadeau, J.G., Pitner, B.J., Linn, C. P., Schram, J.L., Dean, CH. and Nycz, CM. (1999) Real-time, sequence-specific detection of nucleic acids during strand displacement amplification. Anal Biochem. 276, 177-187.

Nycz, CM., Dean, CH., Haaland, P.D., Spargo, CA. and Walker, T.G. (1998) Quantitative reverse transcription strand displacement amplification: quantitation of nucleic acids using an isothermal amplification technique. Anal Biochem. 259, 226-234.

Spargo, C.A., Fraiser, M.S., Van Cleve, M., Wright, D.J., Nycz, CM., Spears, P.A. and Walker, T.G. (1996) Detection of M. tuberculosis DNA using thermophilic strand displacement amplification. MoI Cell Probes 10(4), 247-56. Walker, GT. (1993) Empirical aspects of strand displacement amplification. PCR

Methods Appl. 3(1 ), 1-6.

Walker, G.T., Nadeau, J.G. and Linn, CP. (1995) A DNA probe assay using strand displacement amplification (SDA) and filtration to separate reacted and unreacted detector probes. MoI Cell Probes 9(6), 399-403. Walker, T.G., Fraiser, M.S., Schram, J.L., Little, M.C, Nadeau, J.G. and Malinowski,

D.P. (1992) Strand displacement amplification - an isothermal, in vitro DNA amplification technique. Nucleic Acids Res 20(7), 1691-1696.

Walker, T.G., Linn, CP. and Nadeau, J.G. (1996) DNA detection by strand displacement amplification and fluorescent polarization with signal enhancement using a DNA binding protein. Nucleic Acids Res 24(2), 348-353.

Walker, T.G., Nadeau, J.G., Spears, P.A., Schram, J.L., Nycz, CM. and Shank, D. D. (1994) Multiplex strand displacement amplification (SDA) and detection of DNA sequences from Mycobacterium tuberculosis and other mycobacteria. Nucleic Acids Res 22(13), 2670- 2677.

Claims

CLAIMS:

1. A nucleic acid molecule for amplifying a target region of a target nucleic acid molecule, the nucleic acid molecule comprising:

2. A nucleic acid molecule according to claim 1 , wherein the strand to which the primer binds is the sense strand of the target molecule.

3. A nucleic acid molecule according to claim 1 , wherein the strand to which the primer binds is the antisense strand of the target molecule.

4. A nucleic acid molecule according to any one of claims 1 to 3, wherein the portion of the target region to which the second nucleotide sequence is complementary is sufficiently close to the 5' end of the target region to cause the complementary sequences to hybridise to form the partial loop.

5. A nucleic acid molecule according to any one of claims 1 to 4, which further includes a third nucleotide sequence at the end of the second nucleotide sequence, the third nucleotide sequence comprising at least two nucleotides which are complementary to at least the last two nucleotides at the 3' end of the primer.

6. A nucleic acid molecule according to any one of claims 1 to 4, which includes a fourth nucleotide sequence linking the second nucleotide sequence and the primer.

7. A nucleic acid molecule according to claim 5, which includes a fourth nucleotide sequence linking the third nucleotide sequence and the primer.

8. A nucleic acid molecule according to any one of claims 1 to 7, which further comprises a restriction enzyme recognition site.

9. A nucleic acid molecule according to claim 8, wherein the restriction enzyme recognition site is a BsoB1 recognition site.

10. A nucleic acid molecule according to claim 7 or 8, wherein the restriction enzyme recognition site is located within the primer.

11. A nucleic acid molecule according to claim 7 or 8, wherein the restriction enzyme recognition site is located between the first and second nucleotide sequences.

12. A nucleic acid molecule according to any one of claims 1 to 11, wherein the target molecule is from a pathogen.

13. A nucleic acid molecule according to claim 12, wherein the pathogen is a virus or microbe.

14. A method according to claim 12 or 13, wherein the target molecule is HIV-1.

15. A nucleic acid molecule according to any one of claims 1 to 14, wherein the first nucleotide sequence is from about 3 to about 5 nucleotides in length.

16. A nucleic acid molecule according to any one of claims 1 to 14, wherein the first nucleotide sequence is from about 15 to about 18 nucleotides in length.

17. A nucleic acid molecule according to any one of claims 1 to 16, wherein the second nucleotide sequence is from about 7 to 12 nucleotides in length.

18. A nucleic acid molecule according to any one of claims 1 to 17, wherein the second nucleotide sequence has a GC content of about 50%.

19. A nucleic acid molecule according to any one of claims 1 to 18, wherein the target region is SEQ ID NO: 1 or a sequence which has at least 90% identity thereto.

20. A nucleic acid molecule according to any one of claims 1 to 19, which has at least 90% identity to any one of SEQ ID NOs: 3 to 10.

21. A primer set for amplifying a target region of a nucleic acid molecule, the primer set including a forward primer comprising a nucleic acid molecule as claimed in any one of claims 1 to 20 and a reverse primer.

22. A primer set according to claim 21 , wherein the reverse primer comprises:

(a) a sequence specific for the target region;

(b) a restriction enzyme recognition site; and

(c) a sequence which is non-specific to the nucleic acid molecule.

23. A kit for carrying out a point of care test, the kit comprising: (a) a nucleic acid molecule according to any one of claims 1 to 20 or a primer set according to either of claims 21 or 22;

(b) reagents for extracting nucleic acid from a patient sample;

(c) reagents for amplification of the nucleic acid; and/or

(d) means for detecting the presence of a target region in the nucleic acid.

24. A kit according to claim 23, wherein the detecting means is a dipstick or microfluidic device.

25. A method of amplifying a target region of a target nucleic acid molecule, the method including the steps of:

(a) mixing a nucleic acid sample with a nucleic acid molecule according to any one of claims 1 to 20;

(b) allowing the nucleic acid molecule of any one of claims 1 to 20 to bind to the target region of a strand of the target molecule if the target molecule is present in the sample;

(c) initiating strand polymerization to generate a double stranded DNA amplicon;

(e) displacing the two strands of the double stranded DNA amplicon; and

(f) allowing the complementary regions of the strand in step (d) to hybridize to each other and form a partially looped structure at its 3' end, thereby maintaining the strand in a single-stranded state.

26. A method according to claim 25, which further includes an initial reverse transcription step (RT) if the target molecule is RNA, so as to convert RNA to cDNA.

27. A method according to claim 25 or 26, wherein the strand of the target molecule to which the nucleic acid molecule binds is a sense strand.

28. A method according to claim 25 or 26, wherein the strand of the target molecule to which the nucleic acid molecule binds is an antisense strand.

29. A method according to any one of claims 25 to 28, which further includes the step of cleaving the strands of the DNA amplicon at a recognition site with a restriction enzyme after step (c).

30. A method according to any one of claims 25 to 29, wherein the target molecule is from a pathogen.

31. A method according to claim 30, wherein the pathogen is a virus or microbe.

32. A method according to claim 30 or 31 , wherein the target molecule is HIV-1.

33. A method according to any one of claims 25 to 32, wherein steps (c) and (e) are initiated using Bst DNA polymerase and BsoBI restriction enzymes.