An Improved Recombineering Toolset for Plants

Abstract

Gene functional studies often rely on the expression of a gene of interest as transcriptional and translational fusions with specialized tags. Ideally, this is done in the native chromosomal contexts to avoid potential misexpression artifacts. Although recent improvements in genome editing have made it possible to directly modify the target genes in their native chromosomal locations, classical transgenesis is still the preferred experimental approach chosen in most gene tagging studies because of its time efficiency and accessibility. We have developed a recombineering-based tagging system that brings together the convenience of the classical transgenic approaches and the high degree of confidence in the results obtained by direct chromosomal tagging using genome-editing strategies. These simple, scalable, customizable recombineering toolsets and protocols allow a variety of genetic modifications to be generated. In addition, we developed a highly efficient recombinase-mediated cassette exchange system to facilitate the transfer of the desired sequences from a bacterial artificial chromosome clone to a transformation-compatible binary vector, expanding the use of the recombineering approaches beyond Arabidopsis (Arabidopsis thaliana). We demonstrated the utility of this system by generating more than 250 whole-gene translational fusions and 123 Arabidopsis transgenic lines corresponding to 62 auxin-related genes and characterizing the translational reporter expression patterns for 14 auxin biosynthesis genes.

Figure 1.

Recombineering Process Comparing Classical and New Accelerated Strategies. (A) Schematic representation of the basic concept of recombineering where the lambda red proteins Exo, Beta, and Gam mediate the integration of a linear fragment of DNA electroporated into a recombineering E. coli strain carrying a GOI in a BAC. During this process, the 5′-3′ exonuclease Exo produces 3′-protruding ends in the linear DNA that, upon binding to Beta, find homology regions of as little as 40 nucleotides in the GOI and mediate the integration of the linear DNA molecule. Gam inhibits several E. coli nucleases, preventing the degradation of the linear DNA. After cells with the modified DNA are selected, the whole BAC is transferred to Agrobacterium and used for plant transformation. (B) The first step in any recombineering experiment is the identification of a genomic clone (typically a TAC or a BAC) containing the gene or sequences of interest. (C) In the classical galK-based system, the galK positive/negative selectable marker is amplified using a pair of primers that contain at least 40 nucleotides of sequence corresponding to the sequence flanking the desired insertion site in the target genomic DNA clone. In this example, the amplification of the galK cassette with the GS1 and GS2 primers will result in the production of an amplicon (GS1-galK-GS2). The GS1 and GS2 sequences in the amplicon will target the galK selectable marker to the desired location in the gene, in this particular example, the 3′ of the gene just before the stop codon. (D) Electroporation of this amplicon in a recombineering competent E. coli strain such as SW105 and the selection of the galK-positive colonies in minimal medium supplemented with Gal as the only carbon source will result in a clone containing the galK marker at the desired location in the GOI, in this particular example, immediately upstream of the stop codon. Because of the slow growth of bacteria in minimal medium, this process takes up to 7 d. (E) Using the same set of primers used to amplify the galK cassette, a TAG/AnyDNA cassette (such as GFP) is amplified to generate the amplicon GS1-TAG/AnyDNA-GS2. (F) As before, the GS1 and GS2 sequences will target the amplicon to the desired location, replacing the galK by the TAG/AnyDNA sequence. This sequence replacement can be accomplished by electroporating the GS1-TAG/AnyDNA-GS2 amplicon into the recombineering cells carrying the GOI tagged with galK and selecting for clones that lost galK in minimum medium supplemented with 2-deoxygalactose. Again, this may take up to 7 d due to the slow growth of bacteria in minimal medium. Only galK-negative colonies will survive in the presence of this chemical. (G) The faster and user-friendly bifunctional cassette system combines the selectable marker (such as an antibiotic resistance gene) and the tag of interest in a single cassette. By flanking the sequences of the selectable marker with the FRTs, the selectable marker sequence can be readily removed post-insertion by a highly efficient in vivo FLP reaction. UA, universal adaptor. (H) Similar to the classical approach, the bifunctional large cassette, GS1-5′UA-TAG/AnyDNA-FRT-galK/AmpR-FRT-5′UA-GS2 is first amplified with a pair of primers, GS1 and GS2, to add the gene-specific sequences that will target the recombineering cassette to the desired location in the gene. UA, universal adaptor. (I) By electroporating this cassette into the recombineering E. coli strain SW105 containing the GOI and selecting for, in this example, ampicillin-resistant clones, the bacteria with the desired construct can be efficiently and rapidly identified. The use of regular LB and antibiotic selection allows for the identification of the bacteria with the desired construct in as little as 2 d. UA, universal adaptor. (J) Finally, the induction of FLP recombinase already engineered in the SW105 strain would result in the removal of the sequences corresponding to the selectable marker (bottom), leading to the tag containing the reporter or epitope of interest followed by a 36-nucleotide (nt)-long FRT-containing scar that encodes 12 extra amino acids. The approximate time period required for each step is indicated. The GS1 primer should have the following structure: 5′-40 nt just upstream of the nucleotide, after which you want to insert your tag, followed by the 5′UA sequence -GGA​GGT​GGA​GGT​GGA​GCT-3′. Similarly, the GS2 primer should have the structure: 5′-40 nt corresponding to the reverse complement of the sequence just downstream of the nucleotide, in front of which you want to insert your tag, followed by the 3′UA sequence GGC​CCC​AGC​GGC​CGC​AGC​AGC​ACC-3′. UA, universal adaptor.

Figure 2.

Schematic Representation of Two Applications for the tag-generator Cassette. (A) to (H) A tag-generator cassette consisting of the negative selectable marker gene RPSL and the positive selectable marker Amp^R conferring ampicillin resistance can be used for the easy generation of new bifunctional recombineering cassettes containing any desired tag (see [A] to [C]), or to perform precise gene editing (such as introducing point mutations, deletions, or insertions) in the GOI (see [D] to [H]). To facilitate the use of this tag-generator cassette, in addition to the negative (RPSL) and positive (Amp^R) selectable markers, the construct contains the 5′ and 3′ universal adaptors (UAs) that allow for the amplification of any recombineering cassette in our collection and the TGR sequence that allows for the in-frame insertion of any tag, making it possible to use the resulting cassettes in tagging experiments at any position in the GOI (N-terminal, C-terminal, or internal). Finally, this cassette also includes FRT sites flanking the sequences conferring ampicillin resistance (Amp^R), allowing for the precise and efficient elimination of the selectable marker gene post-insertion (A). The tag-generator cassette can be used to construct new recombineering cassettes (see [B] and [C]). A ready-to-use SW105 E. coli strain containing a TAC clone that harbors the tag-generator cassette has been constructed (A). Using the primers TGF (5′-TAA​AAA​GGG​TTC​TCG​TTG​CTA​AGG​AGG​TGG​AGG​TGG​AGC​T-3′ in-frame with 20 nucleotides of the 5′ end of the new tag) and TGR (5′-GAA​AGT​ATA​GGA​ACT​TCC​CAC​CTG​CAG​CTC​CAC​CTG​CAG​C-3′ in-frame with 20 nucleotides that anneal to the 3′ end of the tag of interest), the tag of interest (TAG/AnyDNA) can be amplified, generating the 5′UA-TAG/AnyDNA-TGR amplicon (B). By electroporating this amplicon in the SW105 recombineering strain carrying the tag-generator cassette and selecting for the absence of RPSL (streptomycin-resistant colonies), a new bifunctional recombineering cassette for the tag of interest will be obtained (C). The tag-generator cassette can also be used in a two-step recombination procedure similar to the classical galK approach to generate any type of sequence modification, such as seamless insertion of a tag, introduction of point mutations, and so on. In this case, the process starts with the identification of the genomic clone containing the GOI (D). Using GS1 and GS2 primers (Figure 1) to PCR-amplify the tag-generator cassette, an amplicon containing the sequences flanking the point where the gene editing will take place is obtained (E). By electroporating this amplicon in SW105 recombineering cells carrying the BAC or TAC clone with the desired gene and selecting for ampicillin-resistant colonies, the GOI is tagged with the tag-generator cassette (F). Next, a replacement DNA construct containing the edited sequence (point mutations, deletions, insertions, and so on; depicted as a red box in [H]) flanked by long regions of homology to the GOI (100 to 200 bp on each side of the region to be edited are recommended) is produced, typically by commercial DNA synthesis (G). When designing these constructs, it is important to consider that recombination can take place at any point within the regions of homology between the replacement sequence and the GOI tagged with the tag-generator cassette (bottom). By electroporating the replacement DNA and selecting for colonies resistant to streptomycin, the desired final product is obtained (H).

Figure 3.

Schematic Representation of Two Applications for the Trimming Cassettes. (A) to (F) Two trimming cassettes, one conferring tetracycline resistance (FRT2-Tet-FRT2 trimming cassette) and another conferring ampicillin resistance (FRT5-Amp-FRT5 trimming cassette), have been generated to facilitate the elimination of undesired sequences in TAC or BAC clones ([A] to [D]), as well as for the efficient transfer of large fragments of DNA from BAC clones to binary vectors ( [E] and [F]). To make these actions possible, each antibiotic selectable marker in the trimming cassettes is flanked by a different pair of orthogonal FRT sequences, FRT2 or FRT5, which not only allow for the elimination of the antibiotic resistance sequences after the trimming process ([C] and [D]) but also for the efficient in vivo transfer of large fragments of DNA from a BAC or TAC clone to a modified binary vector ( [E] and [F]). The first step in the process of trimming a genomic sequence is to identify a BAC or TAC clone carrying the GOI (A). Using DNA for the ready-to-use FRT2-Tet-FRT2 and FRT5-Amp-FRT5 trimming cassettes as PCR templates and two pairs of primers, FRT2F/FRT2R, and FRT5F/FRT5R, two amplicons containing the sequences of the trimming cassettes flanked by 40 nucleotides (nt) homologous to the sequences flanking the region to be deleted in the target genomic DNA are produced by PCR (B). Electroporating these amplicons into electrocompetent SW105 cells carrying the TAC clone harboring the GOI and selecting for colonies resistant to both ampicillin and tetracycline results in the replacement of the undesired genomic DNA sequences by the trimming cassette sequences (C). Inducing the expression of the FLP recombinase present in the genome of the SW105 cells results in the elimination of the ampicillin and tetracycline selectable sequences, leaving behind a single FRT2 and FRT5 site at each flank, respectively (D). The trimming product obtained in (D) contains the desired genomic DNA fragment flanked by two orthogonal FRT sites, opening the possibility of using cassette-exchange strategies to move this potentially large (at least up to 78 kb) DNA from the original BAC/TAC to a binary vector. To generate binary vectors suitable for this cassette-exchange reaction, a derivative of the Gateway pDONR221 vector containing the negative selectable marker SacB flanked by the head-to-toe FRT2 and FRT5 sites was generated (E). Using this new vector, the FRT2-SacB-FRT5 cassette can be easily transferred to any attR1-attR2–containing destination vector such as pGWB1 (E). To transfer the genomic DNA fragment flanked by the FRT2 and FRT5 sites to the pGWB1-FRT2-SacB-FRT5 vector, SW105 cells carrying the trimmed BAC or TAC clone (D) can be electroporated with the pGWB1-FRT2-SacB-FRT5 vector. In the presence of Suc (negative selection for the SacB gene) and hygromycin (positive selection for the pGWB1 backbone), the product of a successful cassette-exchange reaction can be efficiently selected (F). Dark green arrows indicate resistant genes that work both in plants and bacteria. The primers used to amplify the trimming cassettes have the following structure: FRT2 F: 5′-40 nt corresponding to the sequence upstream of the nucleotide in front of which one wants to insert the FRT2 site followed by the sequence -TTC​AAA​TAT​GTA​TCC​GCT​CA-3′. FRT2 R: 5′-40 nt corresponding to the reverse complement sequence downstream of the nucleotide after which one wants to insert the FRT2 site followed by the sequence -TTA​CCA​ATG​CTT​AAT​CAG​TG-3′. FRT5 F: 5′-40 nt corresponding to the sequence upstream of the nucleotide in front of which one wants to insert the FRT5 site-AACGAATGCTAGTCTAGCTG-3′. FRT5 R: 5′-40 nt corresponding to the reverse complement sequence downstream of the nucleotide after which one wants to insert the FRT5 site-TTAGTTGACTGTCAGCTGTC-3′.

Figure 4.

Schematic Representation of the 96-Well-Format Recombineering Pipeline. The process starts by growing 96 DH10B strains carrying the desired TAC clones (the best TAC clones from the two available Arabidopsis libraries for any given gene can be found in our genome browser at https://brcwebportal.cos.ncsu.edu/plant-riboprints/ArabidopsisJBrowser) in a 96-deep-well plate (1). The cells are pelleted by centrifugation and a 96-well-format alkaline-lysis DNA miniprep protocol is used to obtain DNA for the corresponding 96 TACs (2). Electrocompetent SW105 cells are prepared and aliquoted into a 96-well electroporator cuvette (3). DNA for each of the selected 96 TAC clones is added to the electroporation cuvette wells and electroporated into the SW105 competent cells (4). After the electroporation, cells are resuspended in LB and transferred to a 96-deep-well plate where they are allowed to recover before they are plated in selection medium. Individual clones grown in selection medium are tested by PCR and arranged back into a 96-well format (dashed arrow indicates that several steps are not shown; [5]). The SW105 strains carrying 96 TAC clones selected in step 5 are grown overnight in a 96-deep-well plate (6). Cells from the overnight culture are used to inoculate eight cultures corresponding to pools of 12 clones each (7). Electrocompetent cells from each of the eight pools of 12 clones are prepared (8). Aliquots of cells from each pool are placed into the wells of the corresponding rows of the 96-well electroporation cuvette. For example, from pool 1, 12 identical aliquots would be placed in each of the wells of the first row of the 96-well electroporator cuvette and so on (9). In parallel, a pair of 60-mers per gene are designed (primer sequences for generating N- and C-terminal amplicons for any gene and any of our ready-to-use recombineering cassettes can be obtained from our genome browser at https://brcwebportal.cos.ncsu.edu/plant-riboprints/ArabidopsisJBrowser; [10]) and used to generate the corresponding 96 amplicons using the DNA from one of our ready-to-use cassettes as a template (11). The amplicons are purified by simple chloroform extraction and ethanol precipitation in a 96-well plate (12). The corresponding 96 amplicons are added to the electrocompetent cells and electroporated in the 96-well electroporation cuvette (13). As before, the cells are resuspended in LB and transferred to a 96-deep-well plate to allow them to recover (14). The cells from each transformation are then streaked onto LB plates with the proper antibiotic (15). Individual colonies (one or two per construct) are examined by colony PCR using a combination of gene- and tag-specific primers, and the integrity and fidelity of the recombination is checked by PCR-fragment sequencing.

Figure 5.

GUS Staining Patterns of Translational Recombineering Fusions of Auxin Biosynthesis Genes and DR5:GUS in Roots. Seedlings were germinated for 3 d in the dark in control AT medium or in AT medium supplemented with 10 μM NPA, 10 μM ACC, 10 μM NPA + 10 μM ACC, or 50 nM NAA. Samples were optically cleared with ClearSee. Bar = 100 μm.

Figure 6.

GUS Staining Patterns of Translational Recombineering Fusions of Auxin Biosynthesis Genes and DR5:GUS in Shoots. Seedlings were germinated for 3 d in the dark in control AT medium or in AT medium supplemented with 10 μM NPA, 10 μM ACC, 10 μM NPA + 10 μM ACC, or 50 nM NAA. Samples were optically cleared with ClearSee. Bar = 200 μm.

Figure 7.

GUS Staining Patterns of Translational Recombineering Fusions of Auxin Biosynthesis Genes and DR5:GUS in Inflorescences and Flowers. Images of individual flowers represent the enlarged versions of the boxed areas of inflorescences. Red arrows mark the GUS activity domains of interest highlighted in the text. The samples of DR5:GUS and the TAA1 recombineering fusion with GUS were optically cleared with ClearSee to enable visualization of GUS activity in the ovules and developing seeds. Black bars in the inflorescence images = 2.5 mm. White bars in the flower pictures = 250 μm.