Highly efficient homology-directed repair using transient CRISPR/Cpf1-geminiviral replicon in tomato

12 Genome editing via homology-directed repair (HDR) pathway in somatic plant cells was 13 very inefficient compared to illegitimate repair by non-homologous end joining (NHEJ). 14 Here, compared to a Cas9-based replicon system, we enhanced approximately 3-fold in 15 the HDR-based genome editing efficiency via transient geminiviral replicon system 16 equipping with CRISPR/LbCpf1 in tomato and obtained replicon-free, but with stable 17 HDR alleles. Efficiency of CRISPR/LbCpf1-based HDR was significantly modulated by 18 physical culture conditions such as temperature or light. A ten-day incubation at 31 o C 19 under light/dark cycles after Agrobacterium-mediated transformation performed the 20 best among conditions tested. Further, we developed multi-replicon system which is a 21 novel tool to introduce effector components required for the increase of HDR efficiency. 22 Even if it is still challenging, we also showed a feasibility of HDR-based genome editing 23 without genomic integration of antibiotic marker or any phenotypic selection. Our work 24 may pave a way for transgene-free rewriting of alleles of interest in asexually as well as 25 sexually reproducing plants. 26


57
CRISPR/LbCpf1-based geminiviral replicon system highly enhanced HDR in tomato 58 To test the hypothesis, we re-engineered a Bean Yellow Dwarf Virus (BeYDV) replicon to 59 supply high doses of homologous donor templates, and used a CRISPR/LbCpf1 system 60 (Zetsche et al., 2015) for DSB formation ( Figure 1A and 1B). Selection of HDR events was 61 supported by a double selection/screening system using kanamycin resistance and 62 anthocyanin overproduction ( Figure 1A).  Figure 1). The LbCpf1 system using two guide RNAs for targeting the ANT1 69 gene, a key transcription factor controlling anthocyanin pathway, showed the much higher 70 HDR efficiency, at 4.51±0.63 %, visible as purple calli and/or shoots ( Figure 1C and 1D), 71 compared to the other control constructs including a "minus Rep" (pRep -), "minus gRNA" 72 (pgRNA -), and comparable to a CRISPR/SpCas9-based construct (pTC217). The data 73 revealed that functional geminiviral replicons were crucial for the enhancement of HDR 74 efficiencies ( Figure 1C) as shown in other works (Čermák et al., 2015). This is the first report 75 showing highly efficient HDR in plants using Cas12a expressing from a geminiviral replicon.

76
Light conditions or photoperiods enhanced HDR efficiency of CRISPR/LbCpf1 system 77 Boyko and coworkers (2005) showed the strong impact of short-day conditions on 78 intrachromosomal recombination repair (ICR) in Arabidopsis. We tested if the same could be 79 true in tomato somatic cells. Using various lighting regimes, including complete darkness 80 (DD), short (8-h light/16-h dark; 8L/16D) and long (16L/8D) day conditions, we found that 81 HDR efficiencies achieved under short and long day conditions were higher than those in the 82 DD condition in the case of LbCpf1, but not SpCas9, and reached up to 6.62±1.29% (p<0.05, 83 Figure 1E). The advancement of LbCpf1-based HDR system might be explained by stress-84 responses of the host cells which rushed for maintenance of genome stability (Boyko et al.,85 2005) by any means of DNA repairs including HDR. 86 CRISPR/LbCpf1-based HDR was significantly higher compared to CRISPR/Cas9-based 87 system at high temperature 88 Temperature is an important factor controlling ICR (Boyko et al., 2005) and CRISPR/Cas9-89 based targeted mutagenesis in plants (LeBlanc et al., 2018) and CRISPR/Cpf1-based HDR To compete with the efficient NHEJ pathway, protein involving in the HDR pathway were 105 over-expressed, activated or enhanced leading to significant higher efficiencies (Ye et al., 106 2018;Pawelczak et al., 2018). For further improvement of our system, we used several 107 molecular approaches for HDR improvement in tomato. The first was to activate nine HDR 108 pathway genes (Supplemental Table 1) using the dCas9-sun tag/scFv-VP64 activation system 109 (Tanenbaum et al., 2014). A single construct system (pHR01-Activ, Supplemental Figure 2A) 110 showed negative effects on HDR (data not shown), which may be due to its large size (~32 kb 111 as T-DNA and ~27 kb as circularized replicon).

112
The size of viral replicons is inversely correlated with their copy numbers (Suarez-Lopez and 113 Gutierrez, 1997;Baltes et al., 2014). In this work we also tested a novel idea to use a T-DNA 114 producing multiple replicons (pHR01-MR, Figure 2A, and Supplemental Figure 2B).

115
Compared to pHR01, the construct showed HDR efficiencies with 39% increase. We also 116 confirmed the release of three replicons from a single vector (pHR01-MR) used in this work 117 ( Figure 2B). To our best knowledge, this is the first report that multiple replicons can be used 118 for efficient genome editing via HDR pathway. This multiple replicon system may also 119 provide more flexible choices for expressing multiple genes/genetic tools/DNA agents with 120 high copies in plant cells.

121
The true HDR events were obtained at high frequency 122 To verify the HDR repair events in the study, PCR analyses were conducted using primers 123 specific for the right (UPANT1-F1/NptII-R1) and left (ZY010F/TC140R) ( Figure 1A;  Figure 5B).

144
The HDR allele was stably inherited in offspring by self-pollination as well as backcrossing 145 To confirm stable heritable edits, we grew Genome Edited generation 1 (GE1) plants ( Figure   146 2E) obtained from self-pollination of LbCpf1-based HDR GE0 events, and found segregating 147 population in purple phenotype (Supplemental Table 4) similar to data shown by Čermák and    Table 2) or pTC217 (Supplemental Table 3).

251
HDR efficiencies were recorded in at least three replicates and statistically analyzed and plotted 252 using PRISM 7.01 software. In Figure 1C, multiple comparisons of the HDR efficiencies of the 253 other constructs with that of pRepwere done by one-way ANOVA analysis (Uncorrected 254 Fisher LSD test, n=3, df=2, t=4.4; 4.4 and 1.5 for pTC217; pHR01 and pgRNA -, respectively).

450
Step by step protocol is presented with each number in the circles indicates number of days after 451 seed sowing (upper panel) and treatments used in each steps are shown in below panel.   showing perfectly edited HKT1;2 N217 to D217 allele with WT allele as a reference. The nucleotides highlighted in the discontinuous red boxes denote intended modifications for N217D; PAM and core sequences (to avoid re-cutting).
(B) HDR construct layout for HKT1;2 editing. There is neither selection nor visible marker integrated into the donor sequence. The NptII marker was used for enrichment of transformed cells.
(C) Morphology of the HKT1;2 N217D edited event compared to its parental WT in greenhouse conditions. Scale bar = 1 cm.

Supplemental Figure 1
The de novo engineered geminiviral amplicon (named as pLSL.R.Ly) and its replication in tomato.
(A) Map of pLSL.R.Ly. The DNA amplicon is defined by its boundary sequences (Long Intergenic Region, LIR) and a terminated sequence (Short Intergenic Region, SIR). The replication associated protein (Rep/RepA) is expressed from the LIR promoter sequence. All of the expression cassettes of HDR tools were cloned into the vector by replacing the red marker (Lycopene) using a pair of type IIS restriction enzyme (BpiI, flanking ends are TGCC and GGGA). Left (LB) and right (RB) denote the borders of a T-DNA. (B) Circulated DNA detection in tomato leaves infiltrated with pLSL.R.Ly compared to that of pLSLR. Agrobacteria containing the plasmids were infiltrated into tomato leaves (Hongkwang cultivar) and infiltrated leaf were collected at 6, 8 and 11 dpi and used for detection of circulated DNAs. N: water; P1: positive control for pLSL.R.Ly; positive control for P2: pLSLR; Cx: Control samples collected at x dpi; Ixy: infiltrated sample number y collected at x dpi; I11V: sample collected from leaves infiltrated with pLSLR at 11 dpi. PCRs using primers specific to GAPDH were used as loading control.  Alignment of targeted regions isolated from the HKT12 events. 18/25 events (highlighted in yellow) showed strong double peaks indicating single/bi-allelic mutations. 6/25 events showed clearly bi-allelic mutations. C77 showed weak (30%) double peaks. C83 and C105 showed large truncations.

Supplemental Figure 10
Timeline and contents of Agro-mediated transformation protocol used in this work.
Step by step protocol is presented with each number in the circles indicates number of days after seed sowing (upper panel) and treatments used in each steps are shown in below panel.