3.3.1 LAMP and SmartAmp

This group of isothermal amplification depends on DNA polymerase and complicated primer sets (four to six primers) to generate intermediate products (stem-loop DNA structures), which serve as templates for the following exponential amplification. The formation of the stem-loop DNA structure by the primer sets is the key point of the reaction. Once the intermediate products are formed, the autocyclic strand displacement reaction can progress rapidly at a constant temperature.

Loop-mediated isothermal amplification (LAMP) and smart amplification process (SmartAmp) belong to this group, utilizing the most complicated primer design strategy in all the isothermal technologies (Notomi et al., 2000; Mitani et al., 2007). To make the following discussion and comparison clearly understandable, the terminology is defined, as shown in Table 3.3. The terminology difference between LAMP and SmartAmp is also shown in this table. Figure 3.1 shows the design scheme of the technologies and the relative positions of the primers. Other isothermal amplification technologies use fewer primer pairs. However, the inner primer pairs are designed as complicated structures that are crucial for most isothermal amplification technologies. Generally, the inner primers can be designed in two to three parts. The 3′ end is always the hybridizationregion, while the 5′ end and the middle region have different design purposes. The detailed design structures of inner primer pairs are listed and compared separately in Fig. 3.2.

The amplification principles and primer design of LAMP and SmartAmp are similar. They both require a set of at least four specific primers which hybridize to six different specific regions on the nucleic acid template. The two outer primers have a role in strand displacement during the non-cyclic step only (initialamplification stage). The inner primers (also called turn-back primers or folding primers) have both sense and antisense sequence, in such a way as to help in the formation of stem-loop structure, a key aspect of the technology. In order to speed up the reaction, two additional primers (loop primers or stem primers), hybridizing another two different specific regions, can be designed and added to the reaction mixture (Nagamine et al., 2002; Tisi et al., 2010; Gandelman et al., 2011). Loop primers are designed in the region between the inner primer and stem-loop forming region (shown in Fig. 3.1). Stem primers are designed in the region between the two stem-loop forming regions with flexible orientation. LAMP uses a symmetric primer design. The two inner primers are like the turn-back primers in SmartAmp, which uses an asymmetric primer design with a turn-back primer and a folding primer. Compared with the LAMP technology, SmartAmp also employs additional proteins (MutS) to increase the amplification and detection discrimination for SNP (single-nucleotide polymorphism) identification. Some animations on the websites are helpful to understand the complicated principles, e.g. http://loopamp.eiken.co.jp/e/lamp/anim.html and http://www.dnaform.jp/smartamp/smartamp/movie/amp_only_e.html.

Primer selection and design are very challenging for this group of technologies. The following factors need to be considered: specificity of six to eight primer regions, GC content, secondary structure, primer Tm, primer length, primer distance and stability of end base pairs of primer (Parida et al., 2008). The recommended primer Tm from the outermost region (outer primer) to the innermost region (stem-loop forming region) spans from 57 °C to 67 °C. Obviously, this is a difficult task for human efforts alone, even with an advanced educational degree. Therefore, dedicated research work has been done to make both technologies feasible for researchers. The Eiken Chemical Co. (Tokyo, Japan) has developed software to aid the primer selection. The current version is 4. Other choices of software have also been developed, which are available online. The selection rules are specified clearly for each software program and in the review by Parida et al. (2008). The currently available tools are listed as follows.

Multiple detection methods have been developed for the amplified products of LAMP, listed in Table 3.4. The amplification products are stem-loop DNA structures with several inverted repeats of the target, also described as cauliflowerlike structures with multiple loops (Gill et al., 2011). Some of the detection methods are based on the high yield of final products, a unique feature of the technology. A large amount of DNA (10–30 μg/25 μl) can be synthesized in a short time (15–60 min) with high specificity and sensitivity in a typical LAMPreaction (Mori et al., 2001). The yield of final products is usually at least 20-fold that of PCR or other isothermal amplification technologies. Pyrophosphate ions are produced in large amounts to form a white precipitate of magnesium pyrophosphate (insoluble by-product). The turbidity of insoluble complex formed from the high-yield products is a unique feature of LAMP technology. This characteristic greatly facilitates design of the detection platform. The detection can be performed instrument-free (naked eye) or with portable devices, or with the same instruments as PCR. Some detection chemistries, like intercalating dyes or cationic polymers, have certain (minor to strong) inhibitions to LAMP. In the early research stage, addition of the detection reagent after amplification was suggested and applied for the assays. However, detection systems with open tubes always increase the chance of cross-contamination. Tao et al. have described a method of embedding the detection reagent in a wax capsule (Tao et al., 2011). This is a promising technology to solve common problems. In this method, a visualized LAMP method is established by the addition of a microcrystalline wax-dye capsule containing the detection dye. The wax capsule remains intact during amplification, and releases the dye to the reaction mixture only when the temperature is raised to the melting point (95 °C) following amplification. Thecolor difference between positive samples and negative samples is visualized by the naked eye, while the cooling wax turns into a solid barrier on top of the mixture to minimize the risk of aerosol contamination.

LAMP depends on a single protein, DNA polymerase, which is another important feature of this technology. Proteins are the components most sensitive to the inhibitors existing in clinical specimens. The lower complexity of proteins makes it possible to tolerate inhibition by clinical samples. The rate-limiting step is the formation of the intermediate stem-loop structure. Once it is formed, the remaining amplification is less sensitive to the variations of clinical sample input. This might be the reason for the reported robustness of the LAMP assay for clinical diagnostics (Francois et al., 2011; Kaneko et al., 2007).

Since Notomi and his colleagues developed LAMP in 2000, the technology has gained popularity in both the research field and the diagnostic field because of the user-friendly software, available reagents (enzymes, buffers, etc.), simple operation, and multiple choices of detection methods. Many researchers from different countries have worked on the basic research for different applications. As a result, there are more than 500 publications in PubMed so far (at the end of 2011). And the technology has been put into practice in detection of various pathogens, SNP typing and many other diagnostics. It has been demonstrated that, even though the primer design of LAMP is the most complicated system compared with PCR and the other isothermal amplification methods, it is feasible and applicable to both research and diagnostic use.

3.3.2 SDA, NEAR, NEMA, CPA

This group of technologies requires an additional enzyme (a restriction endonuclease or a nicking enzyme) or uses other strategies to simulate the effects of the enzyme.

Strand displacement amplification (SDA) was developed by Walker et al. in 1992 (Walker et al., 1992a, 1992b). This method uses two sets of primers, a DNA polymerase, a restriction endonuclease, and modified nucleotides. The restriction enzyme nicks one strand of double-stranded DNA (dsDNA), and the DNA polymerase extends the 3′ end from the nick. New strands extending from the 3′ ends will displace the downstream strands, which are dispatched from the dsDNA as amplified products. The displaced strands can also serve as new templates. However, because the restriction enzymes typically cut both strands of non-denatured and unmodified DNA, they are not suitable candidates for use in the SDA methods. To create a nick site in a single strand, non-standard nucleotides, such as α-thio-dNTP (dNTP[αS]), must be added to the reaction mixture to alter the enzymes’ action. Thiol-modified nucleotides are incorporated into the synthesized products to inhibit cleavage of the synthesized strand. This modification creates a nick site on the primer side of the strand, which can be extended by the polymerase. SDA can operate at 37–41 °C with Klenowexo-DNA polymerase and a restriction enzyme, such as HincII or Aval. SDA can also perform at around 52 °C with Bst DNA polymerase and a restriction enzyme, such as BsoBI (Little et al., 1999).

Two sets of forward and reverse primers (outer and inner primer pairs) are designed to achieve exponential amplification. The inner primer pairs (F_inner and R_inner) are called amplification primers in SDA. The 3′ end of an inner primer, also called the target binding sequence, hybridizes at the target sequence (Fig. 3.2). It is about 10–25 nucleotides in length and confers hybridization specificity on the amplification primer. The inner primer also comprises a recognition site for a restriction endonuclease, 5′ to the target binding sequence. The recognition site is for a restriction endonuclease which will nick one strand of a DNA duplex when the recognition site is semi-modified. The 5′ tail of the inner primer is about 10–25 nucleotides, located at 5′ to the restriction endonuclease recognition site. It serves as a polymerase repriming site when the remainder of the inner primer is nicked and displaced during SDA. The repriming role of the 5′ tail nucleotides sustains the SDA reaction and allows the synthesis of multiple amplicons from a single target molecule. The sequence of the 5′ tail is not critical; it can be routinely selected and modified to obtain the desired Tm for hybridization (Walker et al., 1992b). The outer primer pairs (F_outer and R_outer) are also called bumper primers in SDA. The bumper primer anneals to a target sequence upstream of the amplification primer such that extension of the bumper primer displaces the downstream amplification primer and its extension product.

Little et al. developed a real-time detection method for SDA based on a dualdye labeled hairpin probe (Little et al., 1999). The region between these labels includes a stem-loop structure. The loop comprises a recognition sequence for the same restriction enzyme used in the assay. A target-specific sequence is at the 3′ end of the probe. Before SDA amplification, the two labels are proximal to each other, such that any excitation of the fluorescein (one label) leads to transfer of the emitted energy to the rhodamine label (the other label). The net effect is that very little emission from the excited fluorescein is detected. After SDA, the probes hybridize to the specific amplified sequences, and are cleaved by the restriction enzyme. This cleavage causes the physical separation of the two labels, such that no energy transfer from the excited fluorescein to rhodamine can occur. Therefore, the excited fluorescent signal is detectable, which is indicative of specific amplification of the target sequence.

However, several drawbacks have hindered the general application of SDA. First, SDA requires a heat denaturation step prior to isothermal amplification. SDA enzymes must be added stepwise to the reaction after heat denaturation of target DNA, so the workflow is complicated. The multiple-step workflow necessitates opening the reaction vessel and exposing the sample to potential contamination. Second, the use of modified nucleotides will not only increase the manufactory cost, but also make the reaction mixture complicated. The added restriction enzymes are much less efficient in nicking the modified substrates, thereby leading to a slower amplification rate and lower product yield, while requiring a much higher amount of the requisite enzyme.

With recent research and discovery of nicking enzymes, the technologies of nicking enzyme amplification reaction (NEAR) and nicking enzyme mediated amplification (NEMA) have been developed (http://bioustar.com/newSite/platforms/jspt_3.jsp; Maples et al., 2009). Both technologies use nicking enzymes instead of restriction enzymes. Therefore, modified nucleotides are no longer required. NEAR only uses the inner primer pairs, while NEMA uses two sets of primers similarly to SDA. The recognition site for a restriction endonuclease of the inner primers is replaced by the corresponding site for a nicking enzyme. The reaction is performed at 54–60 °C with Bst DNA polymerase (Maples et al., 2009).

However, there are some risks of template-independent nucleic acid amplification (background noise) when employing DNA polymerase together with restriction endonucleases or nicking enzymes for in vitro nucleic acid amplification (Liang et al., 2004; Zyrina et al., 2007; Liang et al., 2006). Sometimes the non-specific amplification can affect the sensitivity of the assay.

Cross priming amplification (CPA) can perform strand displacement amplification without nicking activity from additional enzymes (restriction or nicking enzymes) (Fang et al., 2009). It uses two sets of primers to simulate the effects of the additional enzyme. The outer primer pairs are similar to those of SDA. The inner primer pairs contain no recognition sites for any enzymes (Fig. 3.2). Instead, the inner primers, also called cross primers, are designed as follows. The 3′ end of the F_inner sequence hybridizes to the amplification target. The 5′ end of F_inner is identical to the 3′ end sequence of R_inner, and vice versa (Fang et al., 2009). The displaced strand contains newly introduced priming sites on both ends. Thus it can serve as a template with priming sites for both cross primers on its 3′ end. A new priming site is introduced after each round of extension/displacement, resulting in multiple primer binding sites which accelerate the amplification process. The cross primer can also be used as a displacement primer in the following amplification. Overall, the primary purpose of the 5′ end sequence of the inner primer pairs in CPA is to introduce additional priming sites at both ends of the target. The DNA extension mainly relies on the annealing of multiple primers to multiple priming sites of both strands to drive the synthesis of new DNA (You, 2011). Additional primers may be designed for detection purposes. The final products are a mixture with different lengths and multiple forms of secondary structures. Although this technology employs multiple primer binding sites to increase the efficiency, the yield of final product is still far away from that of the LAMP method (similar to other isothermal amplification technologies). The amplification time is also in the range of 1 h.

Selection and design of the inner primer pairs are essential to the speed, specificity and sensitivity of the above technologies. Generally, the concentration of outer primers is lower than that of inner primers, which is similar to LAMP technology. However, there is no available software for primer design for the above technologies.

3.3.3 ICAN, ICA

This group of isothermal amplification technologies is highly dependent on an additional enzyme, RNAse H (Ribonuclease H). The enzyme can specifically degrade the RNA portion of DNA/RNA hybrids (Tadokoro and Kanaya, 2009). Prokaryotic RNase H, which is involved in DNA replication, repair and transcription, has been classified into RNases HI, HII and HIII based on the difference in their amino acid sequences. RNase HI represents type 1 RNase H, and RNases HII and HIII represent type 2 RNase H. Both types of enzymes are useful for nucleic acid amplification and detection, with more recent reports on Type 2 enzymes (Gašparič et al., 2008; Dobosy et al., 2011). Type 1 RNase H requires multiple RNA bases in the substrate for full activity. A DNA/RNA/DNA oligonucleotide with only one or two RNA bases is not cleaved by this type of enzyme when hybridized to a DNA oligonucleotide, while Type 2 RNase H can cleave a single ribonucleotide embedded within a DNA sequence when hybridized to a DNA oligonucleotide (Eder et al., 1993). The cleavage by both types of enzymes occurs on the 5′ side of the RNA residue, leaving a DNA oligonucleotide with a 3′-hydroxyl that is competent to serve as a primer. This unique feature of cleavage is crucial for this group of isothermal amplification technologies. The inner primer pairs are designed with one or more RNA bases, which can be cleaved by the RNase H enzymes to generate nick sites for isothermal amplification. Isothermal and chimeric primer-initiated amplification of nucleic acid (ICAN) and isothermal chain amplification (ICA) belong to this group.

ICAN was proposed in 2002 (Shimada et al., 2002; Mukai et al., 2007). It uses only one pair of primers (inner primer pairs) together with RNase H and DNA polymerase. In Mukai’s paper, Thermococcus litoralis RNaseH and BcaBEST DNA polymerase (Takara Bio Inc., Japan) were used for amplification at 55 °C. The inner primers of ICAN, also described as chimeric primers, are designed as 5′-DNA-RNA-3′, with a few RNA residues (one to six) at the 3′ end (Fig. 3.2). RNase H can cleave the RNA portion of the extended strand and introduce a nick site at the 5′-RNA/DNA-3′ junction of an extended strand synthesized from the chimeric primer. The yield of final products can be increased by adding more primers. Three amplification products are formed, which can be differentiated by gel electrophoresis. The long one contains both primer sequences (forward and reverse), the middle one contains only one primer sequence (either forward or reverse), and the short one contains no primer sequence. Two unique mechanisms, multi-priming and template-switching, have been proposed to explain the phenomenon (Uemori et al., 2007).

ICA is a new technology published in 2010 (Jung et al., 2010). It uses two pairs of primers (outer and inner primer pairs) together with RNase H and DNA polymerase. Hybridase thermostable RNase H (Epicentre Biotechnologies, Madison, WI, USA) and Bst DNA polymerase were used for the amplification at 60 °C. The outer primer pairs are similar to those of SDA. The inner primer pairsare also similar to those of SDA, as shown in Fig. 3.2. The only difference is that a stretch of RNA residues is designed in the nicking region for RNase H cleavage. The major products are double-stranded nucleic acids, including either the forward inner primer or the reverse inner primer. Besides the main amplification products, two other double-stranded nucleic acids are also expected to form, one that includes both the forward and the reverse inner primers and another that is composed of the region inside both the forward and reverse inner primers. However, this technology easily forms non-specific products (primer dimers) from negative samples and low copy number of template samples, based on the published information (Jung et al., 2010).

Cycling probe technology (CPT, dual-labeled chimeric DNA-RNA-DNA probe) is a sequence-specific real-time detection method (Duck et al., 1990). It can be applied for both ICAN and ICA technologies. In the real-time ICAN/ICA, RNase H has two roles. It introduces a nick to chimeric primers to drive the amplification reaction, as well as cleaving a hybridized probe to separate the fluorescence quencher from the reporter, resulting in increased fluorescence.

3.3.4 RCA and related technologies

Rolling circle amplification (RCA) of circulable oligonucleotides (‘padlock’ probes) is a method for the detection and amplification of short DNA sequences. The padlock probe consists of a single-stranded oligonucleotide whose 5′ and 3′ ends hybridize to a target of interest and are then ligated to create a single-stranded DNA (ssDNA) circle, which is then a substrate for RCA (Fire and Xu, 1995). However, RCA is a linear amplification process with only one primer, whose sequence is complementary to the ssDNA. It can take several hours to obtain a detectable signal. In order to improve the amplification efficiency, different formats of exponential RCA have been proposed, such as ramification amplification (Zhang et al., 1998), hyperbranched RCA (Lizardi et al., 1998), cascade RCA (Thomas et al., 1999) and exponential RCA (Alsmadi et al., 2003; Beals et al., 2010). All of the above exponential RCA technologies use one additional primer (second primer), with a sequence identical to a part of the ssDNA, to generate tandem repeat products. The second primer binds at every tandem sequence synthesized by the first primer. The second primers are extended by the polymerase to displace downstream growing strands, which in turn contain a binding site for the first primer. Therefore, exponential amplification is achieved by the two primers, continually initiating new rounds of strand displacement synthesis to provide new binding sites for the primers.

The amplified products are different from linear RCA and exponential RCA. Linear RCA generates a population of long single-stranded products, also called linear concatenated DNA molecules. Each product contains repeats complementary to the individual padlock template. It looks like a broad smear of high molecular weight DNA filaments by gel electrophoresis. The tandem repeatsgenerated by exponential RCA are shown as a double-stranded DNA ladder by gel electrophoresis.

Primer generation RCA (PG-RCA) is a special kind of exponential RCA (Murakami et al., 2009). It uses only ssDNA, nicking enzyme and DNA polymerase. No exogenous primers are required. The circular probe carries not only the hybridization sequence to the target but also the complementary sequence of the nicking enzyme. The reaction starts from ssDNA hybridization to a template DNA, and is followed by a cascade reaction of linear RCA and nicking reaction. The nicking reaction creates multiple primers for the circular probe from the linear RCA product. The smear and ladder DNA are observed on gel electrophoresis. However, background noise (template-independent DNA ladder products) is a crucial concern for this technology to be further applied. This is similar to some other isothermal technologies using a nicking/restriction enzyme together with a DNA polymerase.

RCA can use several real-time detection methods for product detection, such as molecular beacon (Nilsson et al., 2002), PNA probe (Smolina et al., 2004; Kuhn et al., 2001) and molecular zipper (Yi et al., 2006). Padlock probes are powerful reagents capable of distinguishing single nucleotide changes in DNA samples. However, it is difficult to conduct circularization of padlock probes and amplification in one tube simultaneously. Therefore, multiple steps are required. This limits its application in the field of rapid diagnostics. So far there are no commercial kits available for in vitro diagnostics.

3.3.5 RPA and HDA

Recombinase polymerase amplification (PRA) (Piepenburg et al., 2006) and helicase-dependent amplification (HDA) (An et al., 2005) both depend on synchronization of multiple proteins to amplify nucleic acid in vitro. Compared with previously discussed isothermal technologies, these two methods utilize just one pair of primers, which is similar to PCR. The core idea of the technologies is to maximally mimic the natural process in vitro. RPA mimics the process of recombination, while HDA mimics the process of the mismatch repair pathway.

PRA is mediated by the coordinated activities of four proteins: T4 UvsX (recombinase, RecA homologue, ATP-dependent ssDNA binding protein), T4 UvsY (recombinase loading factor), T4 gp32 (ssDNA binding protein) and Bsu DNA polymerase (from Bacillus subtilis). The recombinase forms a complex with DNA primers (pre-synaptic filament) and scans the duplex template for homologous sequences. When the homologous sequence is found, primers hybridize to the target sequence through a strand exchange mechanism and displace the parent strand. Primers are extended by DNA polymerases in both directions. UvsY stimulates the reaction in two ways: displacing the gp32 protein from ssDNA and interacting with UvsX to stabilize its interaction with the primers. Gp32 can bind the parent strand when the strand is displaced duringsynthesis. A low reaction temperature (37–42 °C) and fast reaction speed (15–30 min) represent the main advantages of RPA.

However, the precise rules of primer design are not yet known. Trial-and-error method is applied for primer design and selection. Usually primers are selected as 30–35 nt in length, longer than those used in PCR. As a matter of fact, longer primer sequences increase the chances of non-specific amplification at low temperature. Primer dimers (non-specific amplification) are typically seen in non-template control and very low copy number of target amplification (manufacturer’s package insert, TwistDx Limited, Babraham, Cambridge, UK). In order to resolve the issue, target-specific probes, such as exo probe, fpg probe or LF probe, are applied in real-time and end-point detection (manufacturer’s package insert). However, it is not a fundamental solution. The sensitivity loss by dimer formation is not solved. Furthermore, there is still a certain distance from the PCR assay in terms of robustness and repeatability of the real-time data, based on the published information. All the above issues have limited its broad application in the research and diagnostic field.

HDA is mediated by the coordinated activities of three proteins: Tte UvrD (helicase, unwinding duplex), ET SSB (thermostable ssDNA binding protein) and Bst DNA polymerase. It employs a similar reaction mechanism to PCR, with the exception that HDA uses a helicase enzyme rather than heat to separate doublestranded DNA. Like PCR, the simple reaction scheme requires a pair of primers, a protein mix and buffer. This is the simplest platform in all the isothermal technologies. The similarity between HDA and PCR and the simplicity of the HDA scheme makes it easy to apply variant real-time PCR detection chemistries to HDA (Tong et al., 2008; Tong et al., 2012). The primer design and selection are aided by Primer 3 software, http://bioinfo.ut.ee/primer3-0.4.0/primer3/ (manufacturer’s package insert, BioHelix Corp., Beverly, MA, USA). However, the trial-and-error method is still important for assay optimization. This technology is more attractive with recent improvements in speed, sensitivity and robustness (Tong et al., 2011). Moreover, it can also be adapted to an instrument-free detection format, the BESt™ analyzer, for point-of-care diagnostics (Chow et al., 2008).