|Home | About | Journals | Submit | Contact Us | Français|
Recombineering technology in E. coli enables targeting of linear donor DNA to circular recipient DNA using short shared homology sequences. In this work, we demonstrate that recombineering is also able to support recombination between a pair of linear DNA substrates (linear/linear recombineering) in vivo in E. coli. Linear DNA up to 100 kb is accurately modified and remains intact without undergoing rearrangements after recombination. This system will be valuable for direct in vivo manipulation of large linear DNA including the N15 and PY54 prophages and linear animal viruses, and for assembly of linear constructs as artificial chromosome vectors.
Recombineering technology enables facile modification of large circular DNA using homologous recombination in E. coli without dependence on suitably placed restriction sites or in vitro enzymatic manipulations [1–3]. In the presence of short 40–60 nucleotide homology sequences, expression of the E. coli recE and recT genes and/or the bacteriophage λ red genes facilitates targeting of linear donor DNA (e.g., linear PCR products) to a specific site on circular recipient DNA in vivo [1,2].
To date, recombineering has been demonstrated for the following substrates: 1) linear DNA and a circular replicon (including plasmids [2,4], PACs and BACs [1,2,5], and the E. coli chromosome [6,7]), and 2) linear DNA and a linearized plasmid vector backbone that becomes recircularized after gap-repair following in vivo recombination . However, so far, recombineering between a pair of linear DNA substrates has not been tested in vivo because of the lack of a system that could serve as linear recipient plasmids in E. coli.
We recently developed a system to assemble BACs up to 100 kb as linear plasmids capped with telomeres derived from the bacteriophage N15 . This linear BAC was resistant to RecBCD, which degrades linear DNA in E. coli, and was functional after transfer into human cells and produced correctly spliced β-globin transcript . In this work, using this linear BAC DNA system, we demonstrate recombineering between a pair of linear DNA substrates (linear/linear recombineering) in vivo in E. coli and discuss its application.
To investigate recombineering between a pair of linear DNA substrates, a linear PCR product (donor DNA) was electroporated into electrocompetent E. coli DH10B containing a resident linear 100 kb human β-globin BAC and plasmid pGETrec [strain telN+ DH10B (pGETrec, linear BAC)]  (Figure 1a). The linear BAC serves as the recipient DNA (Figure 1a), while plasmid pGETrec provides the recombineering enzymes, Gam, RecE, and RecT. To initiate linear/linear recombineering the linear donor DNA (Kan60) is electroporated into these cells as described in Figure 1a.
First, to provide transient expression of the recombineering enzymes in this strain, the cellswere grown at 150 rpm at 30°C and plasmid pGETrec was induced for 10 mins with arabinose when the OD600 reached 0.55  (Figure 1a). The linear donor DNA (Kan60) contains the kanamycin resistance gene (Kmr) flanked by 60 bp of homology that directs this 1162 bp DNA to recombine to a sequence centered around the Bst 1107I site on the vector backbone of the linear BAC recipient DNA (Figure 1a). Kan60 was PCR amplified using primers kan60F (5′-TTC CGG TCA CAC CAC ATA CGT TCC GCC ATT CTT ATG CGA TGC ACA TGC TGT ATG CCG GTA caa gaa atc aca gcc gaa gc-3′) and kan60R (5′-AGA CTT CCG TTG AAC TGA TGG ACT TAT GTC CCA TCA GGC TTT GCA GAA CTT TCA GCG gta gcg tga tct gat cct tca act-3′) from pCyPAC7 (a gift from the late Dr. Panos Ioannou) according to our standard protocol  and electroporated into the host cells (Figure 1a).
After recombineering between the linear DNA substrate pair (linear/linear substrates), 58 Kmr recombinants were obtained (Figure 1b). In comparison, a parallel set of linear/circular substrates were recombined by electroporation of the same linear Kan60 donor fragment into a strain containing a circular β-globin BAC recipient [DH10B (pGETrec, circular BAC)] . This recombination event produced 211 Kmr recombinants (Figure 1b). Although linear/circular recombinantion produced about 3.6 times more colonies than linear/linear recombination, it is statistically not significant to suggest that recombination favors linear/circular substrates (Student’s t-test: t = 2.39, df = 3, p > 0.10). While linear/circular recombineering generated more clones, it is important to note that the linear/linear substrates produced an average of more than 50 recombinants per electroporation, far more colonies than needed for analysis. Thus, the linear/linear technique works as efficiently as standard linear/circular recombination and can be applied using existing recombineering protocols.
To verify that linear/linear recombineering precisely targeted the Kan60 fragment to the Bst1107I site on the vector backbone of the linear BAC, a PCR screening assay was employed with primers KF and KR  (black arrows in Figure 1a) to amplify across this site. In total, at least 20 independent Kmr colonies were screened using this assay and all clones were positive for a 1.4 kb product that is expected when the Kan60 PCR fragment recombined correctly into the Bst1107I site of the linear BAC (Figure 1a). Figure 2a shows five of the Kmr recombinants (lanes 1–5) analyzed that produced this 1.4 kb PCR product. As control for this assay, the non-recombinant parent linear BAC produced a 288 bp fragment (Figure 2a, lane 6), indicating absence of Kan60 insertion. This PCR assay confirmed that the linear/linear recombination between Kan60 was precise to the targeted Bst1107I site on the linear BAC DNA to produce the linear BAC::Kan60 (Figure 1a).
Next, to examine the integrity of the modified linear BAC::Kan60 from the telN+ DH10B (pGETrec, linear BAC::Kan60) strain after recombineering, this DNA was analyzed by pulsed field gel electrophoresis (PFGE). In order to purify the linear BAC::Kan60 for analysis without plasmid pGETrec contamination, plasmid pGETrec was first eliminated from the telN+ DH10B (pGETrec, linear BAC::Kan60) strain by addition of arabinose in the medium and omitting ampicillin selection for pGETrec .
Arabinose induction triggers elimination of pGETrec due to the toxicity of the recombineering proteins when over-expressed from this plasmid , leaving only the linear BAC::Kan60 DNA in the host cells. Pure linear BAC::Kan60 DNA was then purified from the telN+ DH10B (linear BAC::kan60) cells using the NucleoBond Plasmid Midi Kit (BD Biosciences, Clontech) and analyzed by PFGE after Nar I restriction digestion.
In the control digestion (Figure 2b, lane P) Nar I cuts the unmodified linear BAC DNA once to produce an 8024 bp vector fragment (see schematic below Figure 2b) and an intact ~ 95 kb fragment containing the β-globin insert. The recombinant linear BAC::Kan60 was also cut once with Nar I but produced a larger 9.2 kb vector band (Figure 2b, lanes 1–3), in addition to the 95 kb β-globin insert fragment, indicating insertion of the 1162 bp Kan60 PCR fragment into the vector backbone as intended (refer to schematic below Figure 2b). The Nar I PFGE (Figure 2b) demonstrated: i) that linear/linear recombineering between the linear Kan60 fragment and the 100 kb linear BAC DNA was precise, and ii) the stability of the linear BAC::Kan60 DNA was retained after linear/linear recombineering without undergoing gross rearrangements.
Recombineering technology has emerged as an important tool for manipulation of genomic loci for functional studies in in vitro and in vivo models [11–13]. The linear/linear recombineering mechanism described in this work functions as effectively as standard recombineering and should specifically appeal to researchers who are interested in applying this technique to manipulate large linear substrates such as the phages N15 and PY54 , animal viruses such as vaccinia virus , and for assembly of large linear constructs for use as artificial chromosome vectors  in mammalian cells.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.