Stimulation of homologous recombination in plants expressing heterologous recombinases

Abstract

Background: Current excitement about the opportunities for gene editing in plants have been prompted by advances in CRISPR/Cas and TALEN technologies. CRISPR/Cas is widely used to knock-out or modify genes by inducing targeted double-strand breaks (DSBs) which are repaired predominantly by error-prone non-homologous end-joining or microhomology-mediated end joining resulting in mutations that may alter or abolish gene function. Although such mutations are random, they occur at sufficient frequency to allow useful mutations to be routinely identified by screening. By contrast, gene knock-ins to replace entire genes with alternative alleles or copies with specific characterised modifications, is not yet routinely possible. Gene replacement (or gene targeting) by homology directed repair occurs at extremely low frequency in higher plants making screening for useful events unfeasible. Homology directed repair might be increased by inhibiting non-homologous end-joining and/or stimulating homologous recombination (HR). Here we pave the way to increasing gene replacement efficiency by evaluating the effect of expression of multiple heterologous recombinases on intrachromosomal homologous recombination (ICR) in Nicotiana tabacum plants.

Results: We expressed several bacterial and human recombinases in different combinations in a tobacco transgenic line containing a highly sensitive β-glucuronidase (GUS)-based ICR substrate. Coordinated simultaneous expression of multiple recombinases was achieved using the viral 2A translational recoding system. We found that most recombinases increased ICR dramatically in pollen, where HR will be facilitated by the programmed DSBs that occur during meiosis. DMC1 expression produced the greatest stimulation of ICR in primary transformants, with one plant showing a 1000-fold increase in ICR frequency. Evaluation of ICR in homozygous T2 plant lines revealed increases in ICR of between 2-fold and 380-fold depending on recombinase(s) expressed. By comparison, ICR was only moderately increased in vegetative tissues and constitutive expression of heterologous recombinases also reduced plant fertility.

Conclusion: Expression of heterologous recombinases can greatly increase the frequency of HR in plant reproductive tissues. Combining such recombinase expression with the use of CRISPR/Cas9 to induce DSBs could be a route to radically improving gene replacement efficiency in plants.

Keywords: 2A; Gene editing; Gene targeting; Intrachromosomal homologous recombination; Pollen; Translational-recoding.

Fig. 1

Recombinases constructs and intrachromosomal recombination (ICR) assay. a The coding sequences of bacterial (RecA, RecG and RuvC) and human (Rad51, Rad52 and DMC1) recombinases (white boxes) were amplified by PCR. Bacterial recombinases were tagged at their N-terminus with SV40 nuclear localisation signal (hatched box). The multigene constructs were made by inserting the 2A sequence from foot and mouth disease virus (black box) between different recombinases in a single open reading frame. These fragments (single and multiple genes) were inserted between the CaMV 35S promoter (35Sp) and nopaline synthase terminator (NosT) of pGSC plasmid containing left (LB) and right (RB) T-DNA borders and the sulphonamide resistant gene (Sul1) for plants selection. Different elements of these constructs schematics are not drawn to scale. b The transgene used as ICR substrate in the tobacco transgenic line N1DC4 is formed of two defective overlapping fragments of β-glucuronidase (GUS) separated by hygromycin resistance gene (hpt). ICR restore a functional GUS gene that can be detected by histochemical staining as blue spots on seedlings (left) and blue pollen (right)

Fig. 2

Test of plasmid constructs using in vitro transcription and translation. a An example of a construct containing multiple coding sequences (RecA, RecG and RuvC) in a single open reading frame (arrow) cloned in pGEM®-T Easy vector (Promega) is shown. During translation, the 2A sequence of foot and mouth disease virus (black box) allows the production of the individual proteins including their nuclear localisation signal (hatched box). b TNT® wheat germ lysate was performed with plasmid DNA of pRecA-2A-RuvC (A-C), pRecA-2A-RecG-2A-RuvC (A-G-C), pRad52-2A-Rad51 (R52-R51) and pRad52-2A-DMC1-2A-Rad51 (R52-D-R51). Arrows indicate individual products

Fig. 3

Growth and fertility of transgenic lines expressing various recombinases. a) Growth in the glasshouse of homozygous lines expressing recombinases and the control N1DC4 (left, scale bar = 10 cm) and flowers of some lines showing longer pistil (right, scale bar = 5 mm). b Intrachromosomal recombination (ICR) frequency in pollen. Pollen of 3 flowers of the control N1DC4 and T0 transgenic lines expressing recombinases was stained and scored for GUS activity. The value of ICR frequency in N1DC4 control was 0.013 × 10^− 4. c The fertility in different homozygous lines compared to the control N1DC4. Data correspond to an average of 10 pods and error bars indicate standard errors. The transgenes are A-C, RecA-2A-RuvC; A-G-C, RecA-2A-RecG-2A-RuvC and R-D-R, Rad52-2A-DMC1-2A-Rad51

Fig. 4

Intrachromosomal recombination (ICR) frequency in homozygous lines. a Six-week-old seedlings of the control N1DC4 and homozygous lines expressing recombinases were stained for GUS activity and the number of blue spots per seedling was scored. b Pollen of the controls wild type (WT) and N1DC4 and two homozygous lines expressing DMC1 in N1DC4 background were stained for GUS activity to detect blue pollen (arrow, scale bar = 100 μm). c ICR events (blue pollen) were scored for the control N1DC4 and different homozygous lines expressing recombinases. The average value of ICR frequency in N1DC4 control was (0.134 ± 0.04)× 10^− 4. A-C, RecA-2A-RuvC construct. Data correspond to an average of 3–12 plants (3 flowers/plant) and error bars indicate standard errors