Link to Publisher's site
Plant Physiol. 2016 Jul; 171(3): 1794–1800.
Published online 2016 May 15. doi: 10.1104/pp.16.00663
PMCID: PMC4936589
PMID: 27208253

An Effective Strategy for Reliably Isolating Heritable and Cas9-Free Arabidopsis Mutants Generated by CRISPR/Cas9-Mediated Genome Editing1,[OPEN]

Abstract

Mutations generated by CRISPR/Cas9 in Arabidopsis (Arabidopsis thaliana) are often somatic and are rarely heritable. Isolation of mutations in Cas9-free Arabidopsis plants can ensure the stable transmission of the identified mutations to next generations, but the process is laborious and inefficient. Here, we present a simple visual screen for Cas9-free T2 seeds, allowing us to quickly obtain Cas9-free Arabidopsis mutants in the T2 generation. To demonstrate this in principle, we targeted two sites in the AUXIN-BINDING PROTEIN1 (ABP1) gene, whose function as a membrane-associated auxin receptor has been challenged recently. We obtained many T1 plants with detectable mutations near the target sites, but only a small fraction of T1 plants yielded Cas9-free abp1 mutations in the T2 generation. Moreover, the mutations did not segregate in Mendelian fashion in the T2 generation. However, mutations identified in the Cas9-free T2 plants were stably transmitted to the T3 generation following Mendelian genetics. To further simplify the screening procedure, we simultaneously targeted two sites in ABP1 to generate large deletions, which can be easily identified by PCR. We successfully generated two abp1 alleles that contained 1,141- and 711-bp deletions in the ABP1 gene. All of the Cas9-free abp1 alleles we generated were stable and heritable. The method described here allows for effectively isolating Cas9-free heritable CRISPR mutants in Arabidopsis.

The advancement of CRISPR/Cas9 genome-editing technology offers unprecedented tools to precisely edit DNA sequences in Arabidopsis (Arabidopsis thaliana) and other organisms (Cong et al., 2013; Feng et al., 2013, 2014; Mali et al., 2013; Gao and Zhao, 2014a). Genome editing by CRISPR/Cas9 has only three requirements: expression of the Cas9 protein, production of a guide RNA (gRNA) that complements the DNA sequences of the target gene, and the existence of an NGG protospacer adjacent motif (PAM) site in the target sequence (Cong et al., 2013; Mali et al., 2013). Cas9 is recruited to the target DNA by the gRNA molecule, which targets a specific DNA sequence by base pairing. Once at the target site, the nuclease activities of Cas9 generate a double-strand break a few base pairs upstream of the PAM site. Small deletions or insertions in the target site are generated when the double-strand break is repaired by error-prone nonhomologous end-joining DNA repair. Because of its simplicity, CRISPR/Cas9 has been widely adopted by many laboratories. Several groups have developed CRISPR vectors for editing genes in Arabidopsis (Feng et al., 2013, 2014; Mao et al., 2013; Fauser et al., 2014; Gao and Zhao, 2014b; Jiang et al., 2014; Li et al., 2014; Xing et al., 2014; Lowder et al., 2015; Ma et al., 2015; Zhang et al., 2015). Successful editing events in Arabidopsis have been widely reported. It is evident that CRISPR/Cas9-mediated gene-editing technology can successfully produce various heritable mutations in Arabidopsis. However, the majority of the reported analyses of the heredity of mutations generated by CRISPR/Cas9 did not segregate out the CRISPR/Cas9 construct. There are two major concerns about the existence of the Cas9/gRNA DNA in CRISPR alleles of Arabidopsis mutants. First, it is difficult to determine whether the mutation in the T2 generation in a putative Arabidopsis mutant is actually inherited from the T1 generation or is newly produced by the Cas9/gRNA construct in the T2 generation. It is essentially impossible to distinguish the two possibilities if the mutation is heterozygous. This point is extremely important, because the newly produced mutation in the T2 generation is likely somatic and not heritable. Second, the prolonged existence of the CRISPR/Cas9 construct in the mutants greatly increases the risk of producing off-target mutations. Despite the many reports of successful gene-editing events in Arabidopsis, we believe that it is still an open question how efficient CRISPR/Cas9 is in generating stably heritable mutations in Arabidopsis because removal of the CRISPR/Cas9 construct was not emphasized in previous studies. It is imperative to segregate out the CRISPR/Cas9 construct before a mutation can be claimed heritable in Arabidopsis.

Effective isolation of targeted mutations generated by CRISPR/Cas9 requires not only reasonable editing efficiency but also an easy method to screen for the mutations. Editing events generated by CRISPR/Cas9 are normally identified by restriction enzyme digestion of PCR fragments or by in vitro digestion using purified Cas9 protein. Both methods are time consuming and laborious. Simplified screening methods are urgently needed.

Here, we report an effective strategy to reliably isolate Cas9-free T2 plants that contain stably heritable mutations in Arabidopsis. We added a cassette that enables the expression of the mCherry gene under the control of a strong promoter to the CRISPR/Cas9 vector. The mCherry cassette allowed us to visually select Cas9-free plants in the T2 generation. We focused on the Cas9-free T2 plants because we hypothesized that once a mutation is identified in a Cas9-free T2 plant, the mutation must have been inherited from the previous generation and it will be stably transmitted to next generations. As a proof of concept, we targeted two sites in the AUXIN-BINDING PROTEIN1 (ABP1) gene. We found that less than 30% of T1 plants contained detectable mutations. About 50% of the positive T1 plants were able to produce Cas9-free plants that harbor a mutation near the target sites. The success rates for identifying a mutation in Cas9-free T2 plants varied among T2 populations, but most were in the single digits. Surprisingly, the ratio between homozygous and heterozygous mutations in the T2 generation apparently failed to match the expected Mendelian segregation. We also show that screening for mutations could be greatly simplified if two gRNAs are expressed simultaneously to generate a large deletion. Our strategy of using mCherry and dual gRNAs led us to effectively isolate Cas9-free plants with the desired mutations.

RESULTS

Development of a Visual Screen for Cas9-Free Plants

In order to obtain stably transmissible mutations in Arabidopsis generated by CRISPR/Cas9-mediated genome-editing technology, it is imperative to segregate out the CRISPR/Cas9 construct. Otherwise, it is very difficult to distinguish between a mutation transmitted from the previous generation and a newly generated mutation by Cas9. Traditionally, Cas9-free plants are identified by PCR using Cas9-specific primers. However, the PCR method is laborious and inefficient. In order to quickly identify Cas9-free plants, we inserted an mCherry-expressing cassette into the CRISPR/Cas9 vector so that Cas9-free plants can be visually identified using a microscope (Fig. 1A). We placed the mCherry gene under the control of the strong promoter At2S3 (Kroj et al., 2003). As shown in Figure 1B, seeds harvested from the T1 plants that contained the mCherry-expressing cassette segregate into two groups: one group displayed strong red fluorescence and the other group had no fluorescence. Because the mCherry cassette and the CRISPR/Cas9 unit are located on the same plasmid, a lack of red fluorescence is indicative of a Cas9-free state. Therefore, the mCherry cassette makes it very easy to visually differentiate the seeds with the Cas9 transgene from those without the Cas9 transgene (Fig. 1B).

Design of a CRISPR/Cas9 vector to facilitate a visual screen for Cas9-free Arabidopsis seeds in the T2 generation. A, Schematic representation of the new CRISPR/Cas9 vector that contains a Cas9 expression cassette driven by the cauliflower mosaic virus (CaMV) 35S promoter and a U6 promoter-controlled gRNA production unit. More importantly, it also expresses mCherry from the strong promoter At2S3 in seeds. NLS, Nuclear localization signal. B, Visual screen for T2 seeds that no longer harbor the CRISPR/Cas9 construct. The Cas9-free seeds do not produce the red fluorescence.

Generation of Mutations in the ABP1 Locus by CRISPR/Cas9

We previously obtained an abp1 mutant that contains a 5-bp deletion in the first exon of ABP1 using our ribozyme-based CRISPR technology (Gao et al., 2015). The mutation in the first exon was suggested not to be optimal because of the potential production of truncated proteins (Chen et al., 2015; Dai et al., 2015; Habets and Offringa, 2015; Pan et al., 2015). Here, we designed two new gRNAs to target two discrete sites in the ABP1 gene (Fig. 2A) to test our new vector and to generate additional abp1 alleles. The new target sites (named CRP2 and CRP3; Fig. 2A) were selected in an attempt to disrupt the auxin-binding pocket in ABP1. The CRP2 target has a BsaJI restriction site near the PAM motif, and the CRP3 target contains a TaqI site (Fig. 2A). The two restriction enzymes can be used to screen editing events at the targets (Fig. 2, B and C). We used the cauliflower mosaic virus 35S promoter to drive the expression of Cas9 and used a U6 promoter to express the specific gRNAs (Fig. 1A). We transformed the CRISPR/Cas9 constructs into wild-type Arabidopsis Columbia-0 and screened for potential gene-editing events in T1 plants. As shown in Table I, we were able to identify T1 plants that had undergone successful editing at the two ABP1 target sites. Interestingly, the editing efficiencies for the two target sites differed significantly. For the CRP3 target, only 3.5% of the T1 plants (three of 86) had detectable mutations at the target site. In contrast, the mutation rate was much higher at the CRP2 site: about 21% (seven of 33) of T1 plants had detectable mutations at the CRP2 site. We noticed that the mutation rate at the CRP2 site was significantly underestimated, because there are two overlapping BsaJI sites at the target. In addition, Cas9 usually cuts DNA 3 bp upstream of the PAM motif, and mutations there would not disrupt the BsaJI restriction site at the CRP2 target. The exact reasons for why editing efficiencies varied greatly between the two targets are not fully understood. Recent studies have clearly shown that certain features in gRNAs greatly affected editing efficiency, and some guidelines for designing better gRNAs have been proposed (Chari et al., 2015; Liang et al., 2016). We did not obtain any apparent homozygous abp1 T1 plants, even though plant 75 appeared to contain very little wild-type ABP1 DNA at the CRP3 target site (Fig. 2C).

Generation of abp1 mutants using the mCherry-containing CRISPR/Cas9 editing vector. A, Schematic representation of the ABP1 gene and the sequences of the selected target sites for editing ABP1. PAM sites (NGG or CCN) are highlighted in dark blue. CRP2 and CRP3 target opposite strands of the ABP1 genomic DNA. The restriction enzyme sites used for genotyping and screening for mutations are underlined. BsaJI recognizes CCNNGG, while TaqI cuts TCGA. Note that Cas9 usually cuts 3 bp upstream of the PAM site. Therefore, screening with BsaJI is not optimal. B, Restriction digestion screen of T1 plants transformed with CRP2/CRISPR vector using the enzyme BsaJI. Plants with mutations generate PCR bands resistant to BsaJI digestion (arrow). Among the 15 samples shown, four potentially have been edited at the ABP1 locus (3, 5, 11, and 14). C, Restriction digestion of PCR products from T1 plants that have disrupted the TaqI site at the CRP3 target site. Note that sample 75 has very little wild-type (WT) DNA. The arrow points to a TaqI-resistant PCR band. D, Three abp1 mutants with deletions/insertions at the CRP2 target site. abp1-c4d has a 4-bp deletion, and abp1-c3d has a 3-bp deletion; abp1-c8i has a very complex mutation. E, Two editing events at the CRP3 site that resulted in two stable Cas9-free abp1 alleles. One has a 12-bp deletion and the other deletes 42 bp near the target site. Note that the 42-bp deletion is not shown in full.

Table I.

Editing efficiencies by CRISPR/Cas9 for different target sites

The target sequences for CRP2 and CRP3 are described in Figure 2A. The RGR target site was described previously (Gao et al., 2015). The RGR sequence and design also are shown in Supplemental Figure S2. CRP2/RGR refers to targeting both CRP2 and RGR sites simultaneously. The ratios represent the editing efficiency. For example, 7:33 refers to seven positive plants out of 33 total plants. All of the T2 plants analyzed were Cas9 free based on a lack of red fluorescence (Fig. 1B). For the T2 plants in boldface, the number of mutant plants (both heterozygous and homozygous) from each T1 plant is shown (boldfaced indicates non-bi-allelic mutations). abp1-c12d and abp1-c42d are from the same T1 plant 75; abp1-c2 shows an unusual segregation pattern.

Target CRP2 CRP3 CRP2/RGR CRP3/RGR
T1 7:33 3:86 5:61 0:92
T2 T1 plant 3, died T1 plant 34, 0:72 T1 plant 14, 0:48 Not analyzed
T1 plant 5, 0:72 T1 plant 40, 0:72 T1 plant 29, 26:52 (abp1-c2) Not analyzed
T1 plant 11, 1:95 abp1-c2+/−, 13:52 Not analyzed
T1 plant 14, 3:94 T1 plant 75, 8:196 abp1-c2−/−, 13:52 Not analyzed
T1 plant 25, 1:95 abp1-c12d, 7:196 T1 plant 38, 0:72 Not analyzed
T1 plant 30, 0:72 abp1-c42d, 1:196 T1 plant 56, 2:96 (abp1-c3) Not analyzed
T1 plant 33, 0:96 T1 plant 65, 0:72 Not analyzed

Isolation of Cas9-Free and Stably Transmissible abp1 Mutations

We harvested seeds from individual T1 plants and used a fluorescence-based visual screen to identify T2 seeds that did not contain the CRISPR/Cas9 construct (Fig. 1B). We then germinated the Cas9-free T2 seeds and transplanted the seedlings to soil. We genotyped at least 48 Cas9-free T2 plants harvested from each T1 plant. Less than 50% (three of seven) of the CRP2 T1 plants produced Cas9-free T2 plants with a mutation at the CRP2 target site (Table I). We genotyped 95 Cas9-free T2 plants from the CRP2 T1 plant 11 and obtained one plant in this population that had a mutation (Table I), which was a 4-bp deletion (abp1-c4d; Fig. 2D). The mutation was heterozygous. The 4-bp deletion resulted in a frame shift. In theory, abp1-c4d would produce a truncated ABP1 protein with the first 117 amino acid residues identical to those of wild-type ABP1. abp1-c4d is likely a loss-of-function mutant because it lacked the P(X)4H(X)3N fingerprint that is important for zinc and auxin binding (Woo et al., 2002). Among the 94 Cas9-free T2 plants from CRP2 T1 plant 14, three plants contained a 3-bp deletion, which were also heterozygous (abp1-c3d; Fig. 2D). We further identified one heterozygous plant out of 95 Cas9-free T2 plants from the CRP2 T1 plant 25 with a complex mutation pattern (abp1-c8i; Fig. 2D). abp1-c8i harbored a 9-bp insertion, a 1-bp deletion, and a point mutation (Fig. 2D). This allele also will be useful in future studies because of a lack of the key C-terminal region of ABP1.

To determine whether the mutations identified in the Cas9-free T2 plants could be stably transmitted to the next generations, we genotyped 28 T3 plants generated from selfing a T2 abp1-c3d plant. We found that 13 plants were heterozygous, eight homozygous, and seven without the mutation, indicating that the mutation identified in a Cas9-free plant at the T2 stage was stably transmitted to the T3 generation in a Mendelian fashion (Supplemental Table S1). Genotyping results of the T3 plants from the other Cas9-free T2 mutants at the CRP2 targets also were consistent with the expected Mendelian ratios, suggesting that once a mutation is confirmed in a Cas9-free T2 plant, the mutation would be stable and could be transmitted to next generations following Mendelian genetics (Supplemental Table S1).

We also analyzed the mutations generated at the CRP3 site. Among the three T1 plants that contained mutations at the CRP3 target, only one T1 plant produced Cas9-free offspring with mutations at the intended target site (Table I). Among the 196 Cas9-free T2 plants for the CRP3 target we genotyped, eight plants contained mutations. Moreover, seven out of the eight plants had a 12-bp deletion at the CRP3 site and one contained a 42-bp deletion (Fig. 2E). Interestingly, among the plants with the 12-bp deletion (abp1-c12d), two were homozygous and five were heterozygous. The 42-bp deletion (abp1-c42d) was homozygous. We crossed the abp1-c12d homozygous T2 plants to wild-type plants and found that all of the F1 plants were heterozygous for the mutation (Supplemental Table S1). We also genotyped some T3 plants from heterozygous T2 abp1-12d plants. It was very clear that the mutation segregated in a Mendelian fashion (Supplemental Table S1).

Generation of Large Deletions Using Two gRNAs

We tested whether we could delete a large fragment by simultaneously expressing two gRNAs that target the sites flanking the intended deletion. If successful, such a strategy will greatly simplify the screening process for gene-editing events because mutants will yield a much smaller PCR fragment than the wild type. Another advantage of a large deletion is that such a mutation would be an unambiguous knockout. We have shown previously that a ribozyme-flanked gRNA unit (RGR) that targeted the first exon of ABP1 successfully produced the abp1-c1 mutant with a 5-bp deletion (Fig. 3A; Gao et al., 2015). Here, we placed the same RGR unit under the control of a UBIQUITIN10 promoter (UBQ10) to produce a gRNA targeting the first exon of ABP1 (Fig. 3A; for vector map and RGR sequences, see Supplemental Figs. S1 and S2). We made two dual gRNA constructs to delete most of the genomic DNA of ABP1. The first construct combined the UBQ10:RGR unit with U6:CRP2, and the other combined UBQ10:RGR with U6:CRP3 (Fig. 3A). We transformed the two constructs into wild-type Arabidopsis and isolated T1 plants. For the RGR/CRP3 construct, we genotyped 92 T1 plants, but none of them produced the expected small PCR fragment. Given the lower editing efficiency that we had observed at the CRP3 site (Table I), the failure to generate a deletion from this construct was not a surprise. For the RGR/CRP2 construct, we obtained five T1 plants out of 61 that produced a smaller PCR fragment than the wild type, suggesting that these two gRNAs together were able to cause the deletion of part of the ABP1 gene.

CRISPR/Cas9-mediated deletions of a large DNA fragment between two gRNA target sites in Arabidopsis. A, We produced CRISPR plasmids that target three sites of the ABP1 gene. We combined the RGR and CRP2 modules to delete the first three exons. We also combined the RGR and CRP3 modules in another plasmid. RGR is controlled by the UBQ10 promoter. Green boxes refer to ABP1 exons. Vertical arrows point to gRNA target sites. ABP1-U409 and ABP1-CRP2-GT2 are the primer pair used in the PCR screening. The RGR sequence and design are shown in Supplemental Figure S2. B, PCR amplification using ABP1-U409 and ABP1-CRP2-GT2 primers and the genomic DNA from Cas9-free T2 plants generated from a single T1 plant transformed with the RGR-CRP2 dual gRNA vector. About half of the plants contained a deletion. Note that this primer pair preferentially amplifies the small fragment and cannot differentiate homozygous from heterozygous plants. C, Schematic representation of the abp1-c2 mutation, which is a deletion of 1,141 bp including the first three exons and 304 bp of the ABP1 promoter. The dashed line represents the deleted region. D, Identification of a second abp1 allele that has a large deletion. Only two plants (105 and 115) out of 96 Cas9-free T2 plants from a single RGR-CRP2 T1 plant contained a deletion (arrows). E, Further sequencing analysis shows that the deletion is 711 bp, which is the exact expected size generated by gRNAs targeting RGR and CRP2 sites.

We then screened Cas9-free T2 plants from the T1 plants that were positive for deletions to identify stably heritable abp1 deletion mutations. From the five positive T1 plants generated with RGR/CRP2, two T1 plants produced Cas9-free T2 offspring that contained a large deletion in the ABP1 gene. We genotyped 52 Cas9-free T2 plants generated from the single T1 plant 29 (Fig. 3B). We found that 26 of the 52 T2 plants contained a large deletion at the ABP1 locus and 13 of the mutants were homozygous for the mutation (Table I). The results apparently did not match the expected results from Mendelian genetics (χ2 = 19.5). We further sequenced the small PCR fragment and found that the deletion was 1,141 bp (abp1-c2; Fig. 3C), which included part of the ABP1 promoter and the first three exons (Fig. 3C; for the deleted sequences, see Supplemental Document S1). Interestingly, the designed deletion between the two gRNAs was only 711 bp.

We also identified two plants in the T2 generation that had a 711-bp deletion in the ABP1 locus (abp1-c3) after screening 96 Cas9-free progeny from the T1 plant 56 (Fig. 3, D and E; Table I). One plant was homozygous and the other was heterozygous. We further tested whether the deletion mutations identified in the T2 plants could be stably transmitted to next generations by genotyping T3 plants generated from selfing and by genotyping F1 plants that resulted from a cross between the T2 mutants and the wild type. Our results demonstrated that the two deletion mutants were stable and segregated into T3 plants following the rules of Mendelian genetics (Supplemental Table S1).

DISCUSSION

We designed a new CRISPR/Cas9 vector to generate Cas9-free T2 plants with targeted mutations in Arabidopsis. We successfully isolated at least two different mutations at each intended target site. The mutations in the Cas9-free plants are stable and are transmitted to next generations in a Mendelian fashion (Supplemental Table S1). The method described in this article (Fig. 4A) is reliable and effective.

Reliably isolating stable and heritable targeted mutants using CRISPR/Cas9 genome-editing technology in Arabidopsis. A, Flow chart for isolating CRISPR alleles of Arabidopsis mutants. The key is to use the visual screen to quickly identify Cas9-free T2 seeds. Mutations in Cas9-free T2 plants are stably transmitted to next generations following Mendelian genetics (Supplemental Table S1). B, Schematic representation of the mosaic nature of mutations generated by CRISPR/Cas9 in T1 plants. If a founder cell for a flower is mutated, the seeds generated from that particular flower will contain heritable mutations (blue or purple). However, seeds in the majority of the siliques do not contain heritable mutations. Red refers to seeds with the mCherry-CRISPR/Cas9 construct.

We show that it is extremely important to focus on Cas9-free T2 plants in order to unambiguously identify heritable mutations generated by CRISPR/Cas9 in Arabidopsis. Identification of a targeted mutation generated from CRISPR/Cas9 is based mainly on analyses of PCR fragments digested with enzymes. The PCR usually uses genomic DNA isolated from a piece of leaf tissue as a template. Results from such assays often cannot reveal the mosaic nature of the mutations if the majority of the cells contain the mutation (Fig. 2C), thus often yielding false positives. We previously identified putative homozygous T2 plants based on restriction digestion (Supplemental Fig. S3), but in the T3 generation we found that the mutation was not heritable, because none of the Cas9-free T3 plants contained the mutation. However, mutations observed in Cas9-free T2 plants must have been transmitted from the previous generation. Because less than 25% of the T2 plants are Cas9 free, it would be 75% more genotyping workload if we did not preselect the Cas9-free plants. In addition, we found that the chance of identifying a mutation in Cas9-free T2 plants is usually low (Table I). For example, we only identified one plant that contained a heterozygous abp1 mutation (abp1-c8i) after we genotyped 95 Cas9-free T2 plants generated from the CRP2 T1 plant 25 (Table I; Fig. 2D). In order to identify one mutant plant in this case, we would have to genotype at least 380 T2 plants if we did not preselect the Cas9-free plants. Given that less than 50% of the positive T1 plants produced Cas9-free plants with a mutation (Table I), the workload would be so heavy that identification of a heritable mutation in a Cas9-free plant becomes prohibitive if we do not preselect the Cas9-free T2 plants. Expression of the mCherry gene in seeds makes the selection of Cas9-free T2 plants very convenient and efficient (Fig. 1B).

We were puzzled by why the ratio between heterozygous and homozygous mutants in some cases was not 2:1. For example, the ratio was 1:1 (13:13) for abp1-c2 in the T2 generation (Fig. 3C; Table I). Moreover, 50% of the Cas9-free T2 plants contained the abp1-c2 mutation. The 1:1 ratio and the 50% of mutants observed apparently cannot be explained by Mendelian genetics, which would give 75% of plants with the mutation (homozygous and heterozygous) among the T2 plants and a 2:1 ratio between heterozygous and homozygous mutants. We realized that mutations in T1 plants are probably mosaic. If a given floral founder cell contains a mutation in the target site, seeds in the silique developed from that particular flower will have the mutation (Fig. 4B). If the mutation in the founder cell is heterozygous, seeds in the silique will segregate according to Mendelian genetics. However, if the mutation in the founder cell is homozygous, every seed in the silique will contain the homozygous mutation (Fig. 4B). When we harvest seeds from a T1 plant, seeds from all of the siliques are mixed. Because the founder cells for the majority of flowers/siliques do not contain a mutation, mutated seeds from a few siliques will be diluted, greatly decreasing the chance for identifying Cas9-free mutants at the T2 stage. That could explain the single-digit rate observed for identifying Cas9-free mutants (Table I). For the abnormal heterozygous-to-homozygous ratio, we believe that it is determined by the ratio between homozygous siliques and heterozygous siliques (Fig. 4B). Another example of the mosaic nature in T1 plants was observed in the CRP3 T1 plant 75 (Table I). This T1 plant produced two different mutations (Table I). We think that plant 75 was not biallelic, because one mutation appeared predominantly (Table I).

Recently, it was reported that the expression Cas9 under the control of some specialty promoters could greatly increase gene-editing efficiency in Arabidopsis (Wang et al., 2015; Yan et al., 2015; Mao et al., 2016). It was reported that homozygous mutants could be obtained in the T1 generation (Wang et al., 2015). However, the studies did not determine what percentage of Cas9-free T2 plants contained the edited mutations. Given the dramatically increased editing efficiency at the T1 stage when these specialty promoters were used instead of the 35S promoter, it is worth combining our mCherry cassette with the specialty promoter-driven Cas9 unit to increase the efficiency for isolating Cas9-free heritable Arabidopsis mutations.

The generation of a large deletion by employing two gRNAs greatly simplifies the screening process (Fig. 3). We did not observe a dramatic decrease in editing efficiency when dual gRNAs were used (Table I). Another advantage is that large-deletion mutations are more likely null compared with small-deletion mutations.

In summary, CRISPR/Cas9 gene-editing technology is a powerful tool for creating targeted mutations in Arabidopsis, but it is important to identify mutations in Cas9-free T2 plants to ensure that the mutations observed can be stably transmitted to future generations. Our fluorescence-based visual screen facilitates the isolation of Cas9-free T2 seeds easily and quickly. Combined with the use of dual gRNAs, our method reliably identifies useful targeted mutations in Arabidopsis.

MATERIALS AND METHODS

Plasmids and Constructs

Two CRISPR/Cas9 vectors were used in this study: pHDE-35SCas9-mCherry and pHDE-35SCas9-UBQ10-mCherry. The complete sequence information of the two vectors is shown in Supplemental Document S2. The maps and annotated vector sequences are shown in Supplemental Figure S1. The U6-gRNA unit was cloned into the PmeI site in both vectors by Gibson assembly (Gibson et al., 2008). The RGR unit was cloned into the MfeI site in pHDE-35SCas9-UBQ10-mCherry by Gibson assembly. The RGR design and sequences are shown in Supplemental Figure S2.

Screen for Editing Events in Arabidopsis

The CRISPR/Cas9 constructs were transformed into Arabidopsis (Arabidopsis thaliana) wild-type Columbia-0 through floral dipping. T1 plants were selected either by red fluorescence or on 16 µg L−1 hygromycin. Genomic DNA samples extracted from leaf tissues of 2-week-old T1 plants were used as templates for PCR. To screen mutations at the CRP2 target, we used the primer pair ABP1-U409 (5′-CCTCATCACACAACAAAGTCACTC-3′) and ABP1-CRP2-GT2 (5′-CATGAGGACCTGCAGGTGTTG-3′) to amplify the CRP2 target-containing fragment. The PCR product was digested using restriction enzyme BsaJI. Putative mutations should produce a BsaJI-resistant band. To genotype mutations at the CRP3 site, we used primers ABP1-2E (5′-TTGCCAATCGTGAGGAATATTAG-3′) and ABP1-CRP2-GT2 for PCR. Then, we digested the PCR product with TaqI. To screen for large deletions, we conducted PCR using ABP1-U409 and ABP1-CRP2-GT2 to screen for smaller fragments.

Cas9-free T2 seeds were isolated using a dissecting fluorescence microscope equipped with an mCherry filter. We focused our PCR screening for mutants on the identified Cas9-free plants.

Supplemental Data

The following supplemental materials are available.

Acknowledgments

We thank Drs. Rui Wu and Detlef Weigel at the Max Planck Institute for Developmental Biology for suggesting the use of mCherry-based screening, for providing the mCherry expression cassette, and for valuable discussions.

Notes

Glossary

gRNA guide RNA
PAM protospacer adjacent motif

Footnotes

1This work was supported by the National Institutes of Health (grant no. R01GM114660 to Y.Z.).

[OPEN]Articles can be viewed without a subscription.

References

  • Chari R, Mali P, Moosburner M, Church GM (2015) Unraveling CRISPR-Cas9 genome engineering parameters via a library-on-library approach. Nat Methods 12: 823–826 [PMC free article] [PubMed] [Google Scholar]
  • Chen J, Wang F, Zheng S, Xu T, Yang Z (2015) Pavement cells: a model system for non-transcriptional auxin signalling and crosstalks. J Exp Bot 66: 4957–4970 [PMC free article] [PubMed] [Google Scholar]
  • Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, Hsu PD, Wu X, Jiang W, Marraffini LA, et al. (2013) Multiplex genome engineering using CRISPR/Cas systems. Science 339: 819–823 [PMC free article] [PubMed] [Google Scholar]
  • Dai X, Zhang Y, Zhang D, Chen J, Gao X, Estelle M, Zhao Y (2015) Embryonic lethality of Arabidopsis abp1-1 is caused by deletion of the adjacent BSM gene. Nat Plants 1: 15183. [PMC free article] [PubMed] [Google Scholar]
  • Fauser F, Schiml S, Puchta H (2014) Both CRISPR/Cas-based nucleases and nickases can be used efficiently for genome engineering in Arabidopsis thaliana. Plant J 79: 348–359 [PubMed] [Google Scholar]
  • Feng Z, Mao Y, Xu N, Zhang B, Wei P, Yang DL, Wang Z, Zhang Z, Zheng R, Yang L, et al. (2014) Multigeneration analysis reveals the inheritance, specificity, and patterns of CRISPR/Cas-induced gene modifications in Arabidopsis. Proc Natl Acad Sci USA 111: 4632–4637 [PMC free article] [PubMed] [Google Scholar]
  • Feng Z, Zhang B, Ding W, Liu X, Yang DL, Wei P, Cao F, Zhu S, Zhang F, Mao Y, et al. (2013) Efficient genome editing in plants using a CRISPR/Cas system. Cell Res 23: 1229–1232 [PMC free article] [PubMed] [Google Scholar]
  • Gao Y, Zhang Y, Zhang D, Dai X, Estelle M, Zhao Y (2015) Auxin binding protein 1 (ABP1) is not required for either auxin signaling or Arabidopsis development. Proc Natl Acad Sci USA 112: 2275–2280 [PMC free article] [PubMed] [Google Scholar]
  • Gao Y, Zhao Y (2014a) Self-processing of ribozyme-flanked RNAs into guide RNAs in vitro and in vivo for CRISPR-mediated genome editing. J Integr Plant Biol 56: 343–349 [PubMed] [Google Scholar]
  • Gao Y, Zhao Y (2014b) Specific and heritable gene editing in Arabidopsis. Proc Natl Acad Sci USA 111: 4357–4358 [PMC free article] [PubMed] [Google Scholar]
  • Gibson DG, Benders GA, Andrews-Pfannkoch C, Denisova EA, Baden-Tillson H, Zaveri J, Stockwell TB, Brownley A, Thomas DW, Algire MA, et al. (2008) Complete chemical synthesis, assembly, and cloning of a Mycoplasma genitalium genome. Science 319: 1215–1220 [PubMed] [Google Scholar]
  • Habets ME, Offringa R (2015) Auxin Binding Protein 1: a red herring after all? Mol Plant 8: 1131–1134 [PubMed] [Google Scholar]
  • Jiang W, Yang B, Weeks DP (2014) Efficient CRISPR/Cas9-mediated gene editing in Arabidopsis thaliana and inheritance of modified genes in the T2 and T3 generations. PLoS ONE 9: e99225. [PMC free article] [PubMed] [Google Scholar]
  • Kroj T, Savino G, Valon C, Giraudat J, Parcy F (2003) Regulation of storage protein gene expression in Arabidopsis. Development 130: 6065–6073 [PubMed] [Google Scholar]
  • Li JF, Zhang D, Sheen J (2014) Cas9-based genome editing in Arabidopsis and tobacco. Methods Enzymol 546: 459–472 [PubMed] [Google Scholar]
  • Liang G, Zhang H, Lou D, Yu D (2016) Selection of highly efficient sgRNAs for CRISPR/Cas9-based plant genome editing. Sci Rep 6: 21451. [PMC free article] [PubMed] [Google Scholar]
  • Lowder LG, Zhang D, Baltes NJ, Paul JW III, Tang X, Zheng X, Voytas DF, Hsieh TF, Zhang Y, Qi Y (2015) A CRISPR/Cas9 toolbox for multiplexed plant genome editing and transcriptional regulation. Plant Physiol 169: 971–985 [PMC free article] [PubMed] [Google Scholar]
  • Ma X, Zhang Q, Zhu Q, Liu W, Chen Y, Qiu R, Wang B, Yang Z, Li H, Lin Y, et al. (2015) A robust CRISPR/Cas9 system for convenient, high-efficiency multiplex genome editing in monocot and dicot plants. Mol Plant 8: 1274–1284 [PubMed] [Google Scholar]
  • Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, Norville JE, Church GM (2013) RNA-guided human genome engineering via Cas9. Science 339: 823–826 [PMC free article] [PubMed] [Google Scholar]
  • Mao Y, Zhang H, Xu N, Zhang B, Gou F, Zhu JK (2013) Application of the CRISPR-Cas system for efficient genome engineering in plants. Mol Plant 6: 2008–2011 [PMC free article] [PubMed] [Google Scholar]
  • Mao Y, Zhang Z, Feng Z, Wei P, Zhang H, Botella JR, Zhu JK (2016) Development of germ-line-specific CRISPR-Cas9 systems to improve the production of heritable gene modifications in Arabidopsis. Plant Biotechnol J 14: 519–532 [PMC free article] [PubMed] [Google Scholar]
  • Pan X, Chen J, Yang Z (2015) Auxin regulation of cell polarity in plants. Curr Opin Plant Biol 28: 144–153 [PubMed] [Google Scholar]
  • Wang ZP, Xing HL, Dong L, Zhang HY, Han CY, Wang XC, Chen QJ (2015) Egg cell-specific promoter-controlled CRISPR/Cas9 efficiently generates homozygous mutants for multiple target genes in Arabidopsis in a single generation. Genome Biol 16: 144. [PMC free article] [PubMed] [Google Scholar]
  • Woo EJ, Marshall J, Bauly J, Chen JG, Venis M, Napier RM, Pickersgill RW (2002) Crystal structure of auxin-binding protein 1 in complex with auxin. EMBO J 21: 2877–2885 [PMC free article] [PubMed] [Google Scholar]
  • Xing HL, Dong L, Wang ZP, Zhang HY, Han CY, Liu B, Wang XC, Chen QJ (2014) A CRISPR/Cas9 toolkit for multiplex genome editing in plants. BMC Plant Biol 14: 327. [PMC free article] [PubMed] [Google Scholar]
  • Yan L, Wei S, Wu Y, Hu R, Li H, Yang W, Xie Q (2015) High-efficiency genome editing in Arabidopsis using YAO promoter-driven CRISPR/Cas9 system. Mol Plant 8: 1820–1823 [PubMed] [Google Scholar]
  • Zhang Z, Mao Y, Ha S, Liu W, Botella JR, Zhu JK (2015) A multiplex CRISPR/Cas9 platform for fast and efficient editing of multiple genes in Arabidopsis. Plant Cell Rep http://dx.doi.org/10.1007/s00299-015-1900-z [PMC free article] [PubMed] [Google Scholar]

Articles from Plant Physiology are provided here courtesy of American Society of Plant Biologists

Formats: