- Open Access
TALEN-mediated genome-editing approaches in the liverwort Marchantia polymorpha yield high efficiencies for targeted mutagenesis
Plant Methods volume 13, Article number: 20 (2017)
The liverwort Marchantia polymorpha occupies a crucial position in land plant evolution and provides the opportunity to investigate adaptations to a terrestrial plant life style. Marchantia reverse genetic analyses have thus far been conducted by employing a homologous recombination approach, which yields an efficiency of around 3%. Availability of the characterized and suitable endogenous MpEF1α promoter prompted us to establish the TALEN gene targeting technique for Marchantia.
Here, two different TALEN techniques, using custom and self-assembled TALEN constructs, were applied and compared. The MpNOP1 gene was selected as a candidate gene, as the respective knockout mutant has been shown to lack air chamber formation, representing an easily traceable phenotype. We demonstrate that both TALEN approaches are successful in Marchantia yielding high gene targeting efficiencies of over 20%. Investigation of selected G1 up to G4 generations proved the stability of the knockout mutants. In 392 analyzed T1 plants, no additional phenotypes were observed and only one chimeric knockout plant was detected after an extended cultivation period. Interestingly, two out of the 24 sequenced mutants harbored indels causing in-frame mutations and revealed novel Mpnop1-related phenotypes. This demonstrates the potential to detect crucial amino acids and motives of targeted proteins, which is of special interest for essential genes where full knockouts are lethal. The FastTALE™ TALEN assembly kit enables the rapid assembly and ligation of the TALEN arms within half a day. For transformations, custom and assembled constructs were subcloned into Marchantia binary vectors possessing the MpEF1α promoter.
Considering time, costs and practicability, the assembly TALEN approach represents a rapid and highly efficient gene targeting system to generate Marchantia knockout mutants, which can be further adapted for future advanced genome-editing applications.
Marchantia polymorpha is a complex thalloid liverwort with a haploid gametophyte-dominant life cycle, where the sporophyte is largely dependent upon the gametophyte . Liverworts, together with mosses and hornworts, belong to the group of bryophytes which diverged approximately 470 Mya from the earliest terrestrial plants . Representing one of the earliest land plant lineages, liverworts are therefore a key group to investigate the genetic basis of evolutionary effects associated with the transition from water to land and the origin of land plant innovations. Marchantia polymorpha has a long research history with histological studies dating back into the last centuries . Pioneering work has been conducted on the gene content and organization of plant mitochondria, chloroplasts and the Y sex chromosome [4,5,6,7]. Whereas ancient polyploidization events increased the genome content in the moss model organism Physcomitrella patens , evidence for increased chromosome numbers is lacking for liverworts such as Marchantia that have likely experienced a low rate of chromosomal evolution. These findings, together with the relatively small genome size of about 280 Mb  and an easy cultivation in the laboratory spurred the development of molecular techniques to establish Marchantia as a basal land plant model organism. Robust and rapid Agrobacterium mediated transformation techniques enable small and large screens of transgenic T1 populations [10,11,12,13]. Analysis of nuclear and organellar M. polymorpha genomes is conducted under the Community Sequencing Program at the Joint Genome Institute (DOE-JGI: http://jgi.doe.gov/whysequence-a-liverwort/), further enhancing the attractiveness of this bryophyte model system.
For reverse genetic approaches, two gene-targeting techniques have been established to generate knockout mutants. Targeted genome modification by homologous recombination (HR) is highly effective in P. patens, but more difficult in other plant species . To enhance the knockout-rates in Marchantia, a positive/negative selection system has been established, which employs the hygromycin resistance gene along with the diphtheria toxin A fragment . MpNOP1 was chosen as a candidate gene as the knockout mutant lacks air chamber formation and can therefore be easily identified by phenotypic screening . The HR approach yielded an efficiency of about 3%  and has been successfully applied for the generation of several non-essential Marchantia knockout mutants, comprising MpGI, MpFKF , MpPHOT  and MpTAA .
In the past few years, targeted genome editing by employing artificial nucleases paved the ground for rapidly modifying genomes with high efficiencies in a large variety of organisms [20,21,22,23]. Most recently, mainly the TALEN (transcription activator-like effector nucleases) and CRISPR/Cas9 (clustered regularly interspaced short palindromic repeat/CRISPR-associated) genome-editing techniques have been successfully applied to edit plant genomes [24,25,26]. Both systems use endonucleases to create double-strand breaks (DSBs) at a selected target site. Subsequent screenings aim to identify transgenic plants in which imprecise non-homologous end-joining (NHEJ) events introduced deletions, insertions or larger chromosomal replacements at the target site enabling the selection of mutants, which can no longer synthesize functional proteins.
The CRISPR/Cas9 system, originally providing an adaptive immune apparatus for prokaryotes, has been applied for reverse genetic studies in different monocot and eudicot species [24, 27, 28]. In 2014, Sugano and coworkers adopted the CRISPR/Cas9 system for Marchantia . CRISPR/Cas9 sequence specificity is conferred by a short 20 nt sequence that mediates via Watson–Crick base pairing binding to the target site. This sequence is part of a chimeric single guide RNA (sgRNA) that directs the Cas9 endonuclease protein to cleave the targeted DNA thereby generating blunt ends .
TALENs are a combination of the FokI endonuclease and the TAL effector (TALE) DNA-binding domain derived from the transcription activator-like effectors from Xanthomonas spec. [20, 22]. Binding specificity is conveyed by multiple, nearly identical repeats comprising 33–35 aa sequences which differ only at the positions 12 and 13, called repeat-variable di-residue (RVDs). Each repeat binds to a certain single nucleotide (HD = C, NG = T, NI = A, NS = A, NN = G/A, NK = G). Differing from the Cas9 endonuclease of the CRISPR/Cas9 system, the FokI endonuclease acts as a dimer. The catalytic FokI domain is fused C-terminally to the right and left TALEN arms. Two binding sites of 16–20 bp are involved, separated by a spacer of around 15 bp lengths, which significantly enhances the sequence specificity. The right and left TALEN arms bind in a tail-to-tail orientation enabling the dimerization of the FokI domains such that the dimer is able to cut the DNA in the spacer region and generates sticky ends. In the CRISPR/Cas9 system, a shorter target DNA sequence together with a mismatch tolerance at the 5′ end of the sgRNA increases the risk of generating off-targets . Restrictions for the target site design differ as the TALEN arms require only a 5′ thymine  whereas CRISPR/Cas9 target sites need the presence of a 3′ NGG (or NAG) PAM sequence . CRISPR/Cas9 offers the advantage that cloning can be rapidly conducted by a PCR-approach. Meanwhile, generation of the TALEN arms by ligation of repetitive modules has been simplified by the availability of efficient novel assembly techniques [31,32,33].
Over the last few years, the accumulated experience with different genome-editing techniques revealed that each technique has its specific, inherent advantages as well as shortcomings [24,25,26, 34]. The availability of a comprehensive toolbox for genome editing of the new basal land plant model system Marchantia would therefore be ideal and facilitate the selection of the best-suited approach for a specific genome-editing application. In this study, we compared the efficiencies of custom and assembled TALENs, which disrupt the MpNOP1 function. We demonstrate that Marchantia knockout mutants can be generated with an efficiency of over 20% and anticipate that this technique will further advance the use of this novel model organism to investigate the evolution of land plant gene functions.
Construction of custom TALEN vectors
For the custom TALEN approach, TALEN coding sequences were purchased from Cellectis Bioresearch (SAS Paris, France, Fig. 1a). The MpNOP1 (AB830886) target sites (TS) TS1 and TS2 (Fig. 1b, c) were selected in collaboration with Cellectis Bioresearch considering the requirement of a thymine preceding the 5′ end of the target sites. The custom TALEN arms belong to the second-generation design, for which the N-terminus has been shortened and a nuclear localization signal (NLS) has been added. Left arms contain the HA and right arms the S-peptide epitope. The TALEN coding sequences were delivered and cloned into a vector based on the pTAL.pLESS backbone, which lacks a suitable promoter. We used the Gateway® Cloning Technology to insert the TALEN arms in the binary expression vectors pMpGWB103 and pMpGWB403 allowing a double selection for hygromycin (Hyrr) and geneticin (G418r) (Fig. 2a; ). These vectors comprise the efficient MpEF1α promoter and have been successfully used in generating transgenic M. polymorpha plants [35, 36].
For subcloning, each TALEN arm was amplified with primers containing attB sites (5′-GGGGACAAGTTTGTACAAAAAAGCAGGCTTCATGGGCGATCCTAAAAAGAAACG-3′, 5′-GGGGACCACTTTGTACAAGAAAGCTGGGTCTTATCAGTCGGCCGCGAAGTTGATC-3′) and cloned into the entry vector pDONR207 using the Gateway® Technology (Invitrogen, USA) following the manufacturer’s instructions. The constructs were verified by sequencing and the TALEN arms then cloned into the binary expression vectors via the Gateway® LR recombination reaction.
Construction of assembled FastTALE™ vectors
For assembly of the TALEN arms, the FastTALE™ TALEN assembly kit from Sidansai Biotechnology was employed according to the manufacturer’s instructions. To compare the custom-designed and assembled TALEN approaches, the same TS1 and TS2 binding sequences were used (Fig. 1). The TALEN assembly kit provides all required materials and reagents, except for the PCR sequencing primers and restriction enzymes for control digestions. The available plant destination vectors contain the CaMV35S promoter, which requires a subcloning step into the suitable Marchantia binary vectors pMpGWB103 and pMpGWB403 possessing the MpEF1α promoter. The kit includes 172 modules of ready-to-use DNA fragments that code for different TALE repeats and the vectors pL23–pL26 for the left arm cloning, which comprise sequences coding for the last 3′ half-repeat binding A, T, C or G, respectively. Likewise, vectors pR15–pR18 contain the last 3′ half-repeat binding A, T, C or G for the right arm. The procedure is summarized exemplarily for the left TALEN arm binding to the MpNOP1 TS1 (5′-TTGGGGCTCAGGTTAAC-3′, Fig. 1c). The first 5′ T as well as the last 3′ base of the target site are present in the destination vector and are thus not required as modules for the assembly procedure. The modules T1, G2, G3, GG4, CT5, CA6, GG7, TT8 and AA9 for the remaining binding sequences were selected according to the suppliers’ instructions. As the last required 3′ base of the left arm binding-site is a cytosine, vector pL25 was included in the ligation reaction. Likewise, target sequences for the other arms were divided into modules and inserted into vectors pR15 for the right arm of TS1 and into pL26 and pR17 for TS2, respectively. All selected modules were simultaneously assembled and inserted in the provided Ptalen vectors which harbor the CaMV35S promoter, requiring about 20 min bench time and in total about 5 h. Reactions were transformed into the supplied D18 E. coli cells and positive colonies were identified by PCR. Plasmids from positive clones were analyzed by using the unique restriction sites SacI/SpeI together with SphI to generate recognizable, diagnostic fragments and further verified by sequencing (forward primer: 5′-CTCCCCTTCAGCTGGACAC-3′, reverse primer: 5′-AGCTGGGCCACGATTGAC-3′). To subclone the TALEN arms by conventional cloning, pMpGWB103 and pMpGWB403 vectors were modified such that an additional SpeI restriction site was introduced 5′ of a SacI restriction site. The assembled TALEN constructs were digested with SpeI/SacI/SphI and ligated into the modified binary vectors (Fig. 2b). The resulting expression vectors were verified by restriction digests and by sequencing (forward primer: 5′-GGGGACAAGTTTGTACAAAAAAGCAGGCT-3′, reverse primer: 5′-CTCATAAATAACGTCATGCATTAC-3′).
Marchantia growth and transformation
Experiments were performed using the M. polymorpha ecotype BoGa, which was obtained from the Botanical Garden of Osnabrück, Germany. Agrobacterium tumefaciens mediated transformation of sporelings with two constructs was carried out as previously described by Ishizaki et al.  with minor modifications. Differing from Ishizaki et al. , the separation of sporelings from A. tumefaciens after three days of co-cultivation was conducted by a 15 min sedimentation followed by decantation of the added washing solution (½ Gamborg B5 medium with vitamins, Duchefa). Sedimentation and decantation were repeated four times and the sporelings were then plated on agar plates with ½ Gamborg B5 medium with vitamins supplemented with 10 µg/ml hygromycin (Sigma-Aldrich), 5 µg/ml G418 (Sigma-Aldrich) and 100 µg/ml cefotaxime (Duchefa). Gemmae were investigated from the G1 up to the G4 generation. Plants were grown under sterile conditions on solid medium (1% agar medium and ½ Gamborg B5 medium with vitamins, Duchefa) in 10 × 10 × 2 cm petri dishes (Sarstedt, Nümbrecht, Germany) under a 16 h/8 h day/night regimen at 22 °C in climate cabinets (Sanyo MLR351).
Analyses of transgenic lines
Mpnop1 knockout mutants were identified by phenotypic screening of the T1 plants. T1 plants represent the first generation of thalli, which develop from the transformed sporelings. Transgenic lines were investigated under the stereomicroscope (Leica M165 FC) to confirm the lack of air chamber formation and photographed with an integrated Leica DFC490 camera. In total, 24 plants exhibiting the mutant phenotype after 5 weeks were selected for sequencing using the primer 5′-GACAGGGAAGACTTTGGAGAG-3′. Genomic DNA (gDNA) was extracted from each line according to Edwards et al. . The gDNA served as a template to amplify a 440 bp fragment of the MpNOP1 locus containing the TALEN target sites (5′-ATGGTCGATGTGATTGGTAATATTTTTG-3′, 5′-GATCTACACGCCCTTTGCTTGAG-3′). The stability of the genomic modifications was investigated in six selected Mpnop1 ge lines by continuous phenotype analysis and sequencing of the target sites from two lines up to the forth gemmae generation.
Targeted TALEN mutagenesis strategy
To establish an efficient TALEN technique in M. polymorpha, two different TALEN approaches were conducted. To reduce the labor-intensive and time-consuming multiple cloning steps of the different TALEN RVDs, we compared purchased custom TALENs from Cellectis with self-assembled TALENs. Several TALEN assembly kits are commercially available and two have been successfully applied for plant gene targeting. The Golden Gate method (Addgene) has been used for assembling multiple DNA fragments in an ordered fashion via custom repeat arrays in about three days and has enabled the targeting of genomic sites in rice . We applied the FastTALE™ TALEN assembly kit (Sidansai Biotechnology) for Marchantia, which has been employed to generate an efficient targeted mutagenesis in soybean . This kit offers the advantage that TALEN repeat arrays can be constructed by a one-step assembly and ligation procedure in only half a day.
To facilitate the detection of knockout mutants, we chose to target the MpNOP1 gene, as Mpnop1 mutants exhibit an easily identifiable phenotype . Wild-type Marchantia plants develop regularly distributed air chambers on their dorsal thallus surface and each air chamber forms a central air pore, a stomata-like structure (Fig. 3a, b). In Mpnop1 mutants, general thallus growth is not affected but air chamber formation is impaired and air pores do not develop. Instead, a single continuous epidermal cell layer comprising chloroplasts is formed. As Mpnop1 plants lack the numerous chlorophyll-rich filamentous cells formed within the air chambers, their thalli have a slightly more transparent appearance (Fig. 3c–f, ). Right and left TALEN arms were designed for two distinct MpNOP1 target sites in order to compare gene-editing efficiencies for different targeted nucleotide sequences. Furthermore, in case of off-target activities, these should differ for the two target sites and generate different phenotypes. MpNOP1 encodes a plant U-box E3 ubiquitin ligase that exhibits its ligase activity in a U-box dependent manner . NHEJ events should preferably occur before the U-box to avoid the generation of truncated MpNOP1 proteins that possess the crucial N-terminal U-box (Fig. 1b). Therefore, TS1 and TS2 were selected such that they start 36 and 88 bp downstream of the MpNOP1 start codon, respectively. For each target site, left and right TALEN arms bind to a 17 bp long target sequence and the respective spacer sequences comprise 15 bp for TS1 and 16 bp for TS2 (Fig. 1c).
Generation of custom and assembled TALEN constructs
Both custom and assembled TALEN arms required subcloning into suitable binary vectors. The gateway compatible destination vectors pMpGWB103 and pMpGWB403 were used as final binary vectors for Marchantia transformation. Both vectors contain the MpEF1α promoter, which is strongly expressed in Marchantia tissues including the meristematic regions and has successfully been used for the generation of transgenic M. polymorpha plants [35, 36]. Double selection with hygromycin and G418 enabled the selection of transgenic T1 plants carrying right and left TALEN arm integrations. The custom TALENs from Cellectis (Fig. 2a) belong to the improved second-generation, which has been shown to enhance the gene targeting efficiency in Arabidopsis . To introduce the custom TALEN pairs into the binary Marchantia vectors, the right and left TALEN arms were first amplified from the supplied vectors using primers containing attB sites, subcloned into a Gateway-compatible entry vector and transferred into the pMpGWB103/403 vectors (Fig. 2a). The FastTALE™ TALEN arms were assembled and ligated into the provided expression vectors by a one-step-reaction within half a day. TALEN constructs were digested with SpeI/SacI/SphI and subcloned into modified pMpGWB103/403 destination vectors comprising an introduced SacI site to express the TALEN arms under control of the MpEF1α promoter (Fig. 2b).
Analysis of transgenic T1 populations
After Agrobacterium-mediated sporeling transformation, plants were grown for 5 weeks under double selection conditions. Altogether, 392 surviving T1 plants were analyzed to identify transgenic plants revealing the impaired air chamber formation phenotype characteristic for the Mpnop1 mutant (Table 1; ). 44% (20/45) of the TS1 custom TALEN plants displayed the Mpnop1 mutant phenotype, which is exemplarily shown for all confirmed mutant lines in Fig. 3c, d. For the TS2 site, 19% (38/199) of the T1 plants revealed an impaired air chamber formation. The custom TALEN pairs were transformed three times to analyze the variability between independent transformations using identical constructs. The efficiencies were 71, 38 and 20% for TS1 and 21, 22 and 9% for TS2, demonstrating that over threefold efficiency differences can occur between independently conducted transformation experiments. One transformation per target site was conducted with self-assembled TALEN vectors. For TS1, 6% (4/64) and for TS2, 42% (35/84) of the T1 plants exhibited the Mpnop1 knockout phenotype. Considering that both approaches used the same TS1 and TS2 sites, the overall TALEN mutagenesis showed a high efficiency of over 20%, reaching values of 22% for TS1 and 26% for TS2 (Table 1).
For further analysis, 22 T1 plants of either approach (10 for TS1, 12 for TS2) exhibiting the Mpnop1 mutant phenotype were selected and sequenced to confirm the genome-editing event of TS1 and TS2 and also to determine the genetic modifications that occurred by imprecise NHEJ repair of the cleaved genomic MpNOP1 locus. Corroborating the mutant phenotype, 21 of the investigated 22 lines were altered such that open reading frame shifts occurred, indicated by the suffix ge-o. The generation of premature stop codons likely led to the formation of truncated, non-functional MpNOP1 proteins (Fig. 4). 13 of the 22 T1 lines harbored deletions at the target sites varying between 4 nt up to 2000 nt. Most frequently, in four lines, 8 nt deletions were observed. Four insertion lines were obtained, where the most extreme case is represented by Mpnop1-10 ge-o which carries an integration of over 2000 nt. Five lines with a combination of deletions and insertions were detected. Noteworthy, Mpnop1-22 ge-i carried an in-frame deletion (ge-i) of 18 nt, suggesting that the absence of six amino acids directly upstream of the U-box severely affects the MpNOP1 protein activity in air chamber formation. These findings are consistent with previous reports, where TALEN approaches generated a variety of sequence modifications in plants [38,39,40].
Interestingly, two of the 392 T1 lines showed novel phenotypes. In contrast to a full Mpnop1 knockout lacking all air chambers, Mpnop1-23 ge-i formed only exceptionally single air chambers (Fig. 3g, h) and Mpnop1-24 ge-i developed a reduced number of air chambers with a more irregular shape (Fig. 3i, j). Sequencing analysis by testing different thallus areas from each plant revealed that these lines are not chimeras but that gene targeting generated in-frame changes at TS2. L37 and Y38 were replaced by S37 and N38 in Mpnop1-23 ge-i and L37 was deleted in Mpnop1-24 ge-i (Fig. 4), emphasizing the importance of the region upstream of the U-box for normal air chamber formation.
To detect eventual phenotypic changes occurring later in the T1 lines, we continued to observe the development of all mutant and wild-type-like transgenic T1 lines. One initially wild-type-like plant showed after cultivation for 7 weeks a sector of Mpnop1 mutant-like tissue on one site of one thallus arm (Fig. 5a, b). Distinct wild-type-like and Mpnop1-mutant-like tissues were harvested from this plant and sequenced. This revealed a 94 nt insertion in TS2 in the Mpnop1-like tissue, whereas the wild-type-like sector possessed the wild-type MpNOP1 sequence. The different tissue parts of the chimeric lobe from this late-formed Mpnop1-chimera line were transferred to plates with new medium. None of the newly formed thalli exhibited a mosaic phenotype and each tissue developed according to its origin either in a wild-type-like or Mpnop-like plant. As the T1 lines were cultivated continuously on the original selection media, the mosaic phenotype might have resulted from a late transformation event caused by residual A. tumefaciens, which can be impeded by transferring T1 plants to new plates supplemented with antibiotics to restrict A. tumefaciens growth.
Stability of TALEN-induced mutations throughout gemmae generations
Marchantia offers the advantage to propagate transgenic T1 plants asexually via clonal propagation of their gemmae, the G1 plants. These isogenic gemmae are produced from single initial cells in the T1 gemmae cups on the dorsal surface of the thallus . To investigate the stability of the mutations throughout successive G1, G2, G3 and G4 generations, five Mpnop ge-o lines were further analyzed. For each target site we selected from the custom and assembled TALEN approach at least one verified Mpnop1 knockout plant, namely Mpnop1-4 ge-o, -5 ge-o, -11 ge-o, -12 ge-o, -13 ge-o. Additionally, one line harboring an in-frame mutation, Mpnop1-22 ge-i, was also further analyzed. All six lines retained the knockout mutant phenotype and exhibited no reversions or additional phenotypes in the four analyzed gemmae generations (Fig. 3e, f). Consistency of the unaltered target sites was further confirmed by sequence analysis of two G4 lines.
Efficiency of the two Marchantia TALEN approaches
Genome editing using TALENs has been employed in several plant species and efficiencies of over 20% and occasionally up to 70% have been reported [31, 32, 39, 41, 42]. In this study, analysis of two TALEN approaches targeting the MpNOP1 gene revealed in 392 T1 plants an overall efficiency of 22% for TS1 and 26% for TS2. Our results demonstrate that the application of the MpEF1α promoter driving the expression of custom TALENs from Cellectis as well as assembled TALENs using the FastTALE™ TALEN assembly kit generated highly efficient promoter-TALEN combinations. The establishment of the first targeted Marchantia knockout technique by HR succeeded in the generation of approximately 3% mutants in T1 populations . For the recently established Marchantia CRISPR/Cas9 genome-editing technology, the sgRNA was expressed under the control of the endogenous Marchantia U6 polymerase III promoter to target MpARF1, coding for the AUXIN RESPONSE FACTOR1. This enabled a pre-selection step, as plants lacking MpARF1 activity fail to respond to synthetic auxin and thus do not form rhizoids. Analysis of a phenotypically pre-screened T1 population revealed that 11% of knockout mutants were obtained . The high TALEN efficiencies obtained in this study enable the identification of transgenic plants that do not display visible phenotypes by sequencing 50–100 T1 plants. Ideally, these numbers can be further decreased given the possibility to incorporate pre-selection steps.
TALEN construction involves re-engineering a new protein for each target. The FastTALE™ kit enables the easy and rapid assembly and ligation of the TALEN arms within 5 h. In comparison, generation of plasmids using the GoldenGate TALEN repeat modules is conducted in several steps and requires approximately three days (http://www.addgene.org/TALeffector/goldengateV2/k, ). Custom and self-assembled TALEN techniques require subcloning of the TALEN arms into suitable Marchantia vectors, which is also necessary for the CRISPR/Cas9 approach , making the practicability of these two genome-editing techniques comparable with respect to time requirements. Recent comparisons between TALENs and CRISPR/Cas9 showed that depending on the application, both techniques have their specific benefits [34, 44, 45]. CRISPR/Cas9 Arabidopsis lines can be rapidly established but were often found to be mosaic in the T1 generation, emphasizing the necessity to identify appropriate promoters driving Cas9 expression in the required tissues [46,47,48]. Furthermore, pol III promoters such as the U6 promoter are commonly used to express the sgRNAs, which require a precise transcription start and termination for successful recognition by the Cas9 protein. Deep sequencing revealed that the identification of a suitable U6 promoter is critical in order to avoid an inaccuracy of 5′ sgRNA ends, as this might lead to formation of transcripts with variable lengths ultimately reducing the CRISPR/Cas9 editing efficiencies . The TALEN approach does not rely on a correct sgRNA folding for endonuclease recognition but depends solely on the correct expression of the TALEN arm proteins. The MpEF1α promoter has been employed for the generation of several successfully applied binary Marchantia vectors [35, 36] and was therefore applied to establish the TALEN technique in Marchantia.
The TALEN and CRISPR/Cas9 approaches both enable precise gene targeting, but off-target formation is a possible side effect. CRISPR/Cas9 achieves specificity through the sgRNA, which can tolerate up to five mismatches within an up to 20 nt long target site . The TALEN strategy requires protein binding to two distinct sites, each approximately 16 bp in length, which are separated by an about 15 bp long spacer. Compared to the short CRISPR/Cas9 target site, such a sequence is less likely to occur elsewhere in the genome, which is supported by recent studies detecting extremely low TALEN off-target rates [39, 50]. Therefore, CRISPR/Cas9 is considered to be more prone to generate off-targets than TALENs . In addition, by screening the barley genome, Gurushidze et al.  did not detect potential TALEN off-targets. Next generation sequencing of an Arabidopsis T2 plant generated by a custom TALEN approach revealed only three additional deletions, which were unrelated to the TALEN binding site and seemed to have occurred spontaneously . TALEN off-target activities are, therefore considered to be extremely low and might not be distinguishable from the occurrence of unrelated, random mutations in the genome. We conducted two TALEN approaches for two distinct MpNOP1 target sites. In 392 analyzed T1 plants, no phenotypes besides Mpnop1-related ones were detectable. To exclude that TALEN off-target effects are hampering knockout mutant identification it is advisable to conduct the TALEN approach for at least two target sites.
Besides the different CRISPR/Cas9 and TALEN target site length requirements and coinciding specificity differences, the CRISPR sgRNA target site selection is constrained by the requirement of a 3′ NGG (PAM) site following the binding sequence, whereas TALEN pair design provides a higher target site flexibility as it requires only a 5′ thymine for efficient TALEN binding [20,21,22]. The shorter 20 nt long CRISPR binding sites are better suited for multiplex approaches to knock out several target sites of genes which share a high degree of sequence homology [52, 53]. This is particularly interesting for redundantly acting genes belonging to larger gene families. However, as the number of genes per family is exceptionally low in Marchantia compared to other land plants , the demand for such an application might be higher for investigations in angiosperm species.
TALENs generate out-of-frame and in-frame mutations
Small indels ranging from 3 to 20 bp occurred most frequently in the analyzed G1 lines, similar to the reports from other TALEN approaches [31, 38, 39, 41, 42]. Sequencing revealed that 21 out of 22 lines are out-of-frame mutants. The occurrence of premature stop codons might cause the formation of truncated, non-functional MpNOP1 proteins not comprising the crucial U-box of the E3 ubiquitin-protein ligase. Also, aberrant mRNAs might be eliminated via the non-sense mediated mRNA decay mechanism which prevents the synthesis of truncated proteins . Three lines carrying in-frame mutations localized before the predicted start of the MpNOP1 U-box (A40) showed interesting phenotypes. Although these lines likely still synthesize normal MpNOP1 protein levels, the mutations seem to interfere with the formation of a functional U-box. Mpnop1-22 ge-i lacks six amino acids and resembles full Mpnop1 knockout plants. The change of L37 and Y38 into S37 and N38 in Mpnop1-23 ge-i leads to a knockout-like phenotype with an extremely rare air chamber formation. The loss of L37 in Mpnop1-24 ge-i plants generates a less severe phenotype with reduced air chamber formation. The data emphasize the importance of the hydrophobic L37 for normal MpNOP1 activity, which might negatively affect folding of the downstream U-box. As CRISPR/Cas9 mutants predominately result from shorter deletions and insertions, larger deletions and in-frame mutations seem more likely to occur with the TALEN approach [46, 47]. This might be advantageous for the analysis of Marchantia genes, where full knockout mutants are lethal. Therefore, the TALEN approach is also suitable for generating gene modifications with reasonable frequencies such that an essential gene activity is not fully disrupted but allows insight into the function of crucial amino acids and/or motives.
The TALEN technique involves two DNA binding regions, which reduces off-target effects and a relatively unrestricted design enables the targeting of almost any sequence. Different strategies, such as custom TALENs supplied by a company or self-assembled TALENs generated by kits can be employed to generate the two TALEN arms. Both strategies require the same time for further subcloning into destination vectors harboring suitable promoters. Here, we show that TALEN constructs can be produced within only half a day using the FastTALE™ assembly kit, which provides a convenient and rapid alternative to commercial TALENs. In Marchantia, homologous recombination yielded 3% knockout mutants  and the CRISPR method generated 11% mutants in a T1 population that has been pre-screened by a positive selection step . Given the obtained targeting efficiencies of over 20% for the MpNOP1 gene, the TALEN approach is a highly efficient tool for Marchantia knockout mutant generation. Optimizing the efficiencies of different genome-editing techniques further will provide a toolbox for the selection of the best-suited approach for individual advanced applications. Besides knockout mutant generation, this will enable future genomic DNA replacements to be conducted and facilitates the introduction of protein tags for tracing endogenous gene expression to elucidate gene functions in the novel basal land plant model system Marchantia polymorpha.
Shimamura M. Marchantia polymorpha. Taxonomy, phylogeny and morphology of a model system. Plant Cell Physiol. 2016;57:230–56.
Wellman CH, Osterloff PL, Mohiuddin U. Fragments of the earliest land plants. Nature. 2003;425:282–5.
Bowman JL, Araki T, Kohchi T. Marchantia: past: Present and future. Plant Cell Physiol. 2016;57:205–9.
Oda K, Yamato K, Ohta E, Nakamura Y, Takemura M, Nozato N, Akashi K, Kanegae T, Ogura Y, Kohchi T, et al. Gene organization deduced from the complete sequence of liverwort Marchantia polymorpha mitochondrial-DNA—a primitive form of plant mitochondrial genome. J Mol Biol. 1992;223:1–7.
Ohyama K, Fukuzawa H, Kohchi T, Shirai H, Sano T, Sano S, Umesono K, Shiki Y, Takeuchi M, Chang Z, et al. Chloroplast gene organization deduced from complete sequence of liverwort Marrchantia polymorpha chlorplast DNA. Nature. 1986;322:572–4.
Okada S, Sone T, Fujisawa M, Nakayama S, Takenaka M, Ishizaki K, Kono K, Shimizu-Ueda Y, Hanajiri T, Yamato KT, et al. The Y chromosome in the liverwort Marchantia polymorpha has accumulated unique repeat sequences harboring a male-specific gene. Proc Natl Acad Sci USA. 2001;98:9454–9.
Yamato KT, Ishizaki K, Fujisawa M, Okada S, Nakayama S, Fujishita M, Bando H, Yodoya K, Hayashi K, Bando T, et al. Gene organization of the liverwort Y chromosome reveals distinct sex chromosome evolution in a haploid system. Proc Natl Acad Sci USA. 2007;104:6472–7.
Rensing SA, Lang D, Zimmer AD, Terry A, Salamov A, Shapiro H, Nishiyama T, Perroud P-F, Lindquist EA, Kamisugi Y, et al. The Physcomitrella genome reveals evolutionary insights into the conquest of land by plants. Science. 2008;319:64–9.
Ishizaki K, Nishihama R, Yamato KT, Kohchi T. Molecular genetic tools and techniques for Marchantia polymorpha research. Plant Cell Physiol. 2016;57:262–70.
Ishizaki K, Chiyoda S, Yamato KT, Kohchi T. Agrobacterium-mediated transformation of the haploid liverwort Marchantia polymorpha L., an emerging model for plant biology. Plant Cell Physiol. 2008;49:1084–91.
Kubota A, Ishizaki K, Hosaka M, Kohchi T. Efficient Agrobacterium-mediated transformation of the liverwort Marchantia polymorpha using regenerating thalli. Biosci Biotechnol Biochem. 2013;77:167–72.
Tsuboyama S, Kodama Y. AgarTrap: a simplified agrobacterium-mediated transformation method for sporelings of the liverwort Marchantia polymorpha L. Plant Cell Physiol. 2014;55:229–36.
Tsuboyama-Tanaka S, Kodama Y. AgarTrap-mediated genetic transformation using intact gemmae/gemmalings of the liverwort Marchantia polymorpha L. J Plant Res. 2015;128:337–44.
Puchta H. Gene replacement by homologous recombination in plants. Plant Mol Biol. 2002;48:173–82.
Ishizaki K, Johzuka-Hisatomi Y, Ishida S, Iida S, Kohchi T. Homologous recombination-mediated gene targeting in the liverwort Marchantia polymorpha L. Sci Rep. 2013;3:1532.
Ishizaki K, Mizutani M, Shimamura M, Masuda A, Nishihama R, Kohchi T. Essential role of the E3 ubiquitin ligase nopperabo1 in schizogenous intercellular space formation in the liverwort Marchantia polymorpha. Plant Cell. 2013;25:4075–84.
Kubota A, Kita S, Ishizaki K, Nishihama R, Yamato KT, Kohchi T. Co-option of a photoperiodic growth-phase transition system during land plant evolution. Nat Commun. 2014;5:3668.
Komatsu A, Terai M, Ishizaki K, Suetsugu N, Tsuboi H, Nishihama R, Yamato KT, Wada M, Kohchi T. Phototropin encoded by a single-copy gene mediates chloroplast photorelocation movements in the liverwort Marchantia polymorpha. Plant Physiol. 2014;166:U411–598.
Eklund DM, Ishizaki K, Flores-Sandoval E, Kikuchi S, Takebayashi Y, Tsukamoto S, Hirakawa Y, Nonomura M, Kato H, Kouno M, et al. Auxin produced by the indole-3-pyruvic acid pathway regulates development and gemmae dormancy in the liverwort Marchantia polymorpha. Plant Cell. 2015;27:1650–69.
Boch J, Scholze H, Schornack S, Landgraf A, Hahn S, Kay S, Lahaye T, Nickstadt A, Bonas U. Breaking the code of DNA binding specificity of TAL-Type III effectors. Science. 2009;326:1509–12.
Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA, Charpentier E. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science. 2012;337:816–21.
Moscou MJ, Bogdanove AJ. A simple cipher governs DNA recognition by TAL effectors. Science. 2009;326:1501.
Wiedenheft B, Sternberg SH, Doudna JA. RNA-guided genetic silencing systems in bacteria and archaea. Nature. 2012;482:331–8.
Bortesi L, Fischer R. The CRISPR/Cas9 system for plant genome editing and beyond. Biotechnol Adv. 2015;33:41–52.
Paul JW, Qi Y. CRISPR/Cas9 for plant genome editing: accomplishments, problems and prospects. Plant Cell Rep. 2016;35:1417–27.
Sprink T, Metje J, Hartung F. Plant genome editing by novel tools: TALEN and other sequence specific nucleases. Curr Opin Biotechnol. 2015;32:47–53.
Belhaj K, Chaparro-Garcia A, Kamoun S, Nekrasov V. Plant genome editing made easy: targeted mutagenesis in model and crop plants using the CRISPR/Cas system. Plant Methods. 2013;9:39.
Voytas DF, Gao C. Precision genome engineering and agriculture: opportunities and regulatory challenges. PLoS Biol. 2014;12:e1001877.
Sugano SS, Shirakawa M, Takagi J, Matsuda Y, Shimada T, Hara-Nishimura I, Kohchi T. CRISPR/Cas9-mediated targeted mutagenesis in the liverwort Marchantia polymorpha L. Plant Cell Physiol. 2014;55:475–81.
Doyle EL, Booher NJ, Standage DS, Voytas DF, Brendel VP, Vandyk JK, Bogdanove AJ. TAL, Effector-Nucleotide Targeter (TALE-NT) 2.0: tools for TAL effector design and target prediction. Nucleic Acids Res. 2012;40:W117–22.
Chen K, Shan Q, Gao C. An efficient TALEN mutagenesis system in rice. Methods. 2014;69:2–8.
Du H, Zeng X, Zhao M, Cui X, Wang Q, Yang H, Cheng H, Yu D. Efficient targeted mutagenesis in soybean by TALENs and CRISPR/Cas9. J Biotechnol. 2016;217:90–7.
Weber E, Gruetzner R, Werner S, Engler C, Marillonnet S. Assembly of designer TAL effectors by golden gate cloning. PLoS ONE. 2011;6:e19722.
He Z, Proudfoot C, Whitelaw CB, Lillico SG. Comparison of CRISPR/Cas9 and TALENs on editing an integrated EGFP gene in the genome of HEK293FT cells. Springerplus. 2016;5:814.
Ishizaki K, Nishihama R, Ueda M, Inoue K, Ishida S, Nishimura Y, Shikanai T, Kohchi T. Development of gateway binary vector series with four different selection markers for the liverwort Marchantia polymorpha. PLoS ONE. 2015;10:e0138876.
Althoff F, Kopischke S, Zobell O, Ide K, Ishizaki K, Kohchi T, Zachgo S. Comparison of the MpEF1alpha and CaMV35 promoters for application in Marchantia polymorpha overexpression studies. Transgenic Res. 2014;23:235–44.
Edwards K, Johnstone C, Thompson C. A simple and rapid method for the preparation of plant genomic DNA for PCR analysis. Nucleic Acids Res. 1991;19:1349.
Forner J, Pfeiffer A, Langenecker T, Manavella PA, Lohmann JU. Germline-transmitted genome editing in Arabidopsis thaliana Using TAL-effector-nucleases. PLoS ONE. 2015;10:e0121056.
Gurushidze M, Hensel G, Hiekel S, Schedel S, Valkov V, Kumlehn J. True-breeding targeted gene knock-out in barley using designer TALE-nuclease in haploid cells. PLoS ONE. 2014;9:e92046.
Zhang H, Gou F, Zhang J, Liu W, Li Q, Mao Y, Botella JR, Zhu JK. TALEN-mediated targeted mutagenesis produces a large variety of heritable mutations in rice. Plant Biotechnol J. 2016;14:186–94.
Shan QW, Wang YP, Chen KL, Liang Z, Li J, Zhang Y, Zhang K, Liu JX, Voytas DF, Zheng XL, et al. Rapid and efficient gene modification in rice and brachypodium using TALENs. Mol Plant. 2013;6:1365–8.
Wendt T, Holm PB, Starker CG, Christian M, Voytas DF, Brinch-Pedersen H, Holme IB. TAL effector nucleases induce mutations at a pre-selected location in the genome of primary barley transformants. Plant Mol Biol. 2013;83:279–85.
Cermak T, Doyle EL, Christian M, Wang L, Zhang Y, Schmidt C, Baller JA, Somia NV, Bogdanove AJ, Voytas DF. Efficient design and assembly of custom TALEN and other TAL effector-based constructs for DNA targeting. Nucleic Acids Res. 2011;39:e82.
Brandl C, Ortiz O, Rottig B, Wefers B, Wurst W, Kuhn R. Creation of targeted genomic deletions using TALEN or CRISPR/Cas nuclease pairs in one-cell mouse embryos. Febs Open Bio. 2015;5:26–35.
Yu ZS, Chen HQ, Liu JY, Zhang HT, Yan Y, Zhu NN, Guo YW, Yang B, Chang Y, Dai F, et al. Various applications of TALEN- and CRISPR/Cas9-mediated homologous recombination to modify the Drosophila genome. Biol Open. 2014;3:271–80.
Fauser F, Schiml S, Puchta H. Both CRISPR/Cas-based nucleases and nickases can be used efficiently for genome engineering in Arabidopsis thaliana. Plant J. 2014;79:348–59.
Feng Z, Mao Y, Xu N, Zhang B, Wei P, Yang DL, Wang Z, Zhang Z, Zheng R, Yang L, et al. Multigeneration analysis reveals the inheritance, specificity, and patterns of CRISPR/Cas-induced gene modifications in Arabidopsis. Proc Natl Acad Sci USA. 2014;111:4632–7.
Wang ZP, Xing HL, Dong L, Zhang HY, Han CY, Wang XC, Chen QJ. Egg cell-specific promoter-controlled CRISPR/Cas9 efficiently generates homozygous mutants for multiple target genes in Arabidopsis in a single generation. Genome Biol. 2015;16:144.
Ma H, Wu Y, Dang Y, Choi JG, Zhang J, Wu H. Pol III promoters to express small RNAs: delineation of transcription initiation. Mol Ther Nucleic Acids. 2014;3:e161.
Mussolino C, Morbitzer R, Lutge F, Dannemann N, Lahaye T, Cathomen T. A novel TALE nuclease scaffold enables high genome editing activity in combination with low toxicity. Nucleic Acids Res. 2011;39:9283–93.
Kuscu C, Arslan S, Singh R, Thorpe J, Adli M. Genome-wide analysis reveals characteristics of off-target sites bound by the Cas9 endonuclease. Nat Biotechnol. 2014;32:677–83.
Wang C, Shen L, Fu Y, Yan C, Wang K. A Simple CRISPR/Cas9 system for multiplex genome editing in rice. J Genet Genomics. 2015;42:703–6.
Zhang Z, Mao Y, Ha S, Liu W, Botella JR, Zhu JK. A multiplex CRISPR/Cas9 platform for fast and efficient editing of multiple genes in Arabidopsis. Plant Cell Rep. 2016;35:1519–33.
Chang YF, Imam JS, Wilkinson MF. The nonsense-mediated decay RNA surveillance pathway. Annu Rev Biochem. 2007;76:51–74.
SK, ES and SZ designed the experiments. SK developed the custom TALEN and FA the assembly TALEN technique. ES applied the techniques and acquired the data. SK, ES and SZ analyzed the data and wrote the manuscript. All authors read and approved the final manuscript.
The authors thank T. Kohchi, Kyoto, for MpNOP1 sequence information prior to publication and for providing Marchantia vectors. We appreciated comments on the manuscript from Henrik Buschmann and Tom Dierschke. We thank Claudia Gieshoidt for Marchantia transformation support and Lucille Schmieding for proofreading.
The authors declare that they have no competing interests.
This work was supported by the German Research Foundation (DFG) (Grant Nos. SFB944, ZA256/7-1).
Sarah Kopischke, Esther Schüßler and Felix Althoff contributed equally to this work
About this article
Cite this article
Kopischke, S., Schüßler, E., Althoff, F. et al. TALEN-mediated genome-editing approaches in the liverwort Marchantia polymorpha yield high efficiencies for targeted mutagenesis. Plant Methods 13, 20 (2017). https://doi.org/10.1186/s13007-017-0167-5
- Genome editing
- Marchantia polymorpha