Semi-Site-Specific Primer PCR: A Simple but Reliable Genome-Walking Tool

Abstract

Genome-walking has been frequently applied to molecular biology and related areas. Herein, a simple but reliable genome-walking technique, termed semi-site-specific primer PCR (3SP-PCR), is presented. The key to 3SP-PCR is the use of a semi-site-specific primer in secondary PCR that partially overlaps its corresponding primary site-specific primer. A 3SP-PCR set comprises two rounds of nested amplification reactions. In each round of reaction, any primer is allowed to partially anneal to the DNA template once only in the single relaxed-stringency cycle, creating a pool of single-stranded DNAs. The target single-stranded DNA can be converted into a double-stranded molecule directed by the site-specific primer, and thus can be exponentially amplified by the subsequent high-stringency cycles. The non-target one cannot be converted into a double-strand due to the lack of a perfect binding site to any primer, and thus fails to be amplified. We validated the 3SP-PCR method by using it to probe the unknown DNA regions of rice hygromycin genes and Levilactobacillus brevis CD0817 glutamic acid decarboxylase genes.

1. Introduction

Genome-walking is a series of technologies that are exploited to retrieve unknown DNA sequences flanking a known region. Genome-walking is one of the worthy tools in many fields of life science [1,2]. By identifying unknown flanking regions, genome-walking has successfully realized a variety of applications, such as obtaining the full length of a gene, accessing the regulatory sequence of a gene, and identifying transgenic DNA [3,4,5,6].

Numerous strategies have been proposed for genome-walking. Screening the genomic DNA library is the earliest genome-walking method. This method, however, is rather frustrating because it is laborious and time-consuming [7]. With the proceeding of biotechnology, many PCR-based genome-walking methods have been devised. These PCR-based methods vary substantially in their experimental operations, but can be divided into three categories according to the involved rationales: (i) inverse PCR [8,9]; (ii) digestion-ligation-mediated PCR [10,11]; and (iii) randomly primed PCR [12,13].

Inverse PCR features high specificity and low efficiency. In inverse PCR, genomic DNA is firstly digested by an endonuclease that has no recognition site on the known sequence; the digested DNA is then self-cyclized by a DNA ligase. The cyclized target product is subsequently amplified using two site-specific primers with an orientation of 5′-end-facing-5′-end (opposite to classic 3′-end-facing-3′-end). Due to the use of a site-specific primer pair, the amplification specificity of inverse PCR is high [8,9,14]. However, the length of the fragment captured by inverse PCR is unsatisfactory because a smaller fragment is always preferentially ligated and amplified [15].

Digestion-ligation-mediated PCR is compromised in specificity and efficiency, attributed mainly to the necessity of using a linker/adaptor/cassette primer. Like the inverse PCR, digestion-ligation-mediated PCR requires the pretreatment of genomic DNA prior to PCR. Genomic DNA is digested with an endonuclease, followed by ligation with a linker/adaptor/cassette DNA. The DNA of interest is enriched by 2–3 rounds of nested PCRs sequentially driven by the linker/adaptor/cassette primer pairing with nested site-specific primers [10,16]. Apparently, a non-negligible background will inevitably arise from the amplification of irrelevant DNAs by the linker/adaptor/cassette primer [13,17].

Randomly primed PCR is a completely PCR-based method [13,18]. In this PCR, a walking primer partially anneals to some place(s) on the unknown flank in the single non-stringency cycle and afterwards synthesizes a target single-stranded DNA (ssDNA). This target DNA is then amplified by 2–3 successive rounds of PCRs performed by the walking primer pairing with nested site-specific primers [13,18]. Thermal asymmetric interlaced PCR [12], partially overlap** primer-based PCR [7], and wristwatch PCR [13] are types of randomly primed PCR. Nevertheless, undesired DNAs resulting from the walking primer are a big issue. Moreover, these PCRs involve complicated experimental processes or need multiple walking primers [7,12,13].

In this work, a new genome-walking approach, termed semi-site-specific primer PCR (3SP-PCR), is proposed for the rapid amplification of unknown DNA sequences adjacent to a known region. The 3SP-PCR efficiently enriches target DNA while it suppresses non-target ones through the partial overlap between the semi-site-specific primer used in secondary PCR and the corresponding site-specific primer in primary PCR. We confirmed the feasibility of this method by capturing the flanks of the rice hygromycin (hyg) gene [7] and Levilactobacillus brevis CD0817 glutamic acid decarboxylase genes (gadA/R) [19].

2. Materials and Methods

2.1. Genomic DNA

L. brevis CD0817 was cultivated as previously described [20,21]. The genomic DNA of L. brevis CD0817 was extracted using the TIANamp Bacteria DNA Kit (Tiangen Biotech Co., Ltd., Bei**g, China) according to the manufacturer’s guidance. Rice genomic DNA was kindly provided by the lab of Prof. ** primer-based PCR [7], and stepwise partially overlap** primer-based PCR [15], have features of rapidity and simplicity. However, these methods are unsatisfactory owing to their limited level of efficiency, high background, or complex operations [13]. Therefore, it is attractive to develop a randomly primed PCR-based genome-walking method of real practicability. The proposed 3SP-PCR is such a genome-walking protocol.

The 3SP-PCR is of high specificity because it owns the following two specificity safeguard mechanisms. First, the relatively short (10 nt) 3′ overlap, between the semi-specific primer and its corresponding outer site-specific primer is the key to guarantee the specificity. In each round of 3SP-PCR, the single relaxed-stringency cycle limits the partial annealing of any primer to once only, and produces a pool of ssDNAs. The non-target ssDNA cannot be amplified in the subsequent moderate high-stringency cycles, as it lacks an authentic binding site for any primer. The target one, however, can be amplified in the subsequent moderate high-stringency cycles, as its 3′-end has the binding site for the site-specific primer (Figure 1). In this regard, 3SP-PCR possesses a specificity safeguard mechanism similar to that of stepwise partially overlap** primer-based PCR [7,15] or wristwatch PCR [13]. Second, types of non-target DNAs are reduced in 3SP-PCR. The existing randomly primed PCRs have a common feature; that is, they all have to use a random walking primer. Clearly, three types of non-target products are thus produced in these methods: (i) produced from a random walking primer alone; (ii) produced from random walking primer and site-specific primer; and (iii) produced from site-specific primer alone [1,12,13]. Our 3SP-PCR does not involve a random walking primer given that the primary PCR is a single-primer PCR, in which the outer site-specific primer simultaneously acts as a walking primer. Hence, types i and ii non-target molecules are absent from the amplified product. This is an extra mechanism exclusive to 3SP-PCR for removing non-target background.

The 3SP-PCR is simpler than the available randomly primed PCRs. As discussed above, three types of non-target products are formed in routine randomly primed PCRs primary amplification [1,12,13]. Therefore, three rounds of amplification reactions are generally required in these methods, so as to dilute non-target background to an acceptable level [1,12]. For the proposed 3SP-PCR, two rounds of amplification reactions suffice to give a positive outcome (Figure 2), because its primary amplification generates only one type of non-target DNA (Figure 1).

The multi-band phenomenon predominated by short amplicons, common in routine randomly primed PCRs [1,13,15,23], is significantly ameliorated in 3SP-PCR. In general, a walking primer should identify numerous sites on the unknown flanking DNA under the low temperature (25 °C). As a result, the multi-band phenomenon occurs [7,15]. In addition, a smaller fragment is always preferentially amplified because it has a higher amplification efficiency than that of a longer one. Therefore, small amplicons seem to be predominant in traditional randomly primed PCRs [12,13]. In the primary 3SP-PCR, amplicons are encompassed by the site-specific primer and its inverted repeat. The amplification of a shorter fragment is inhibited because it is a preferentially formed hairpin via intra-strand annealing at the both termini. However, the amplification of a longer DNA is much less inhibited [24]. As shown in Figure 2, each secondary 3SP-PCRs exhibits only 1–2 distinct bands, demonstrating the multi-band phenomenon is improved in this method. Moreover, the largest fragment achieved in each walking ranged from nearly 1.5 to 8.0 kb, indicating a satisfactory walking efficiency of 3SP-PCR.

The success rate of 3SP-PCR is also guaranteed. The walking should succeed as long as the primer partially anneals to the unknown flank in the single low-stringency (25 °C) cycle of primary PCR [7,12,13,15]. We suspect that, at such a low temperature, the primer can find at least one site on the flank suitable for partial annealing, not to mention multiple primers can be individually used in parallel. In this work, one walking comprising three 3SP-PCR sets was performed for each of the selected genes. As shown in Figure 2, all 3SP-PCRs generated positive results, indicating a high success rate of the 3SP-PCR. A comparison of 3SP-PCR with the existing randomly primed PCRs is presented in

Table 3. Comparison of 3SP-PCR with the available PCR-based methods.

Approach	Rationale and Process	ESGM	Walk Range	Reference
Inverse PCR	Genomic DNA is digested with an endonuclease then self-circularized. The resultant DNA undergoes PCR performed by two SSPs with opposite extension orientations.	No	0.3–1.8	[1,8]
DL-PCR	DNA is digested with an endonuclease then ligated to a random oligo. The resultant DNA undergoes 2–3 consecutive PCRs by nested pairing the oligo primer with SSPs.	No	0.3–3.0	[1]
TAIL-PCR	A short random oligo is used as a walking primer. One low-stringency cycle has to be performed in every three cycles to aid walking primer annealing. The amplification of a target DNA is over a non-target one due to the differential amplification efficiency.	Yes	0.3–3.5	[1,12]
POP-PCR	A set of walking primers with 3′-overlaps are individually paired with nested SSPs. A walking primer can partially anneal to a DNA template only in the one relaxed-stringency cycle of each PCR, generating a pool of ssDNAs. A target ssDNA can be converted into dsDNA by the SSP in the next high-stringency cycle, but a non-target one cannot.	Yes	0.3–3.5	[7]
3SP-PCR	The rationale and process are shown in the section of “3.1 Principle of 3SP-PCR” of this study.	Yes	0.4–8.0	This study

Note: SSP denotes site-specific primer. DL-PCR, TAIL-PCR, POP-PCR, WW-PCR, and 3SP-PCR represent digestion-ligation-mediated PCR, thermal asymmetric interlaced PCR, partially overlap** primer-based PCR, wristwatch PCR, and semi-specific-primer PCR, respectively. ESGM: extra success-guaranteeing mechanism.

.

Reportedly, a primer can work even if the base mismatch rate is up to 60% [25]. In this study, the matched bases between the outer site-specific primer and their partial annealing loci were from 8 to 17, with at least two matched bases at the primers 3′ ends (Figure 3), which is accordant with the reported data [24,25,26]. At least two bases pairing at the 3′ end should be essential for the functional priming of a primer [24,25,26].

4. Conclusions

In this study, a simple, rapid but reliable 3SP-PCR has been devised for genome-walking. The feasibility of the proposed method was validated by retrieving the unknown flanking DNA sequences in rice and a microbe. The 3SP-PCR should be a promising alternative to the available genome-walking approaches.