Plant Biotechnology Journal (2016) 14, pp. 533–542
doi: 10.1111/pbi.12403
Multiplexed, targeted gene editing in Nicotiana benthamiana for glyco-engineering and monoclonal antibody production Jin Li1, Thomas J. Stoddard1, Zachary L. Demorest1, Pierre-Olivier Lavoie2, Song Luo1, Benjamin M. Clasen1, Frederic Cedrone3, Erin E. Ray1, Andrew P. Coffman1, Aurelie Daulhac1, Ann Yabandith1, Adam J. Retterath1, Luc Mathis1, Daniel F. Voytas1, Marc-Andr e D’Aoust2,* and Feng Zhang1,* 1 2 3
Cellectis Plant Sciences Inc., New Brighton, MN, USA Medicago Inc., Quebec, QC, Canada Cellectis SA, Paris, France
Received 17 December 2014; revised 20 April 2015; accepted 21 April 2015. *Correspondence (Tel +1(651)683-2807; fax +1(651)683-2805; email [email protected]) and (Tel +1(418)658-9393; fax +1(418)658-6699; email [email protected])
Keywords: Gene editing, transcription activator-like effector nucleases, glyco-engineering, plant-derived pharmaceuticals.
Summary Biopharmaceutical glycoproteins produced in plants carry N-glycans with plant-specific residues core a(1,3)-fucose and b(1,2)-xylose, which can significantly impact the activity, stability and immunogenicity of biopharmaceuticals. In this study, we have employed sequence-specific transcription activator-like effector nucleases (TALENs) to knock out two a(1,3)-fucosyltransferase (FucT) and the two b(1,2)-xylosyltransferase (XylT) genes within Nicotiana benthamiana to generate plants with improved capacity to produce glycoproteins devoid of plant-specific residues. Among plants regenerated from N. benthamiana protoplasts transformed with TALENs targeting either the FucT or XylT genes, 50% (80 of 160) and 73% (94 of 129) had mutations in at least one FucT or XylT allele, respectively. Among plants regenerated from protoplasts transformed with both TALEN pairs, 17% (18 of 105) had mutations in all four gene targets, and 3% (3 of 105) plants had mutations in all eight alleles comprising both gene families; these mutations were transmitted to the next generation. Endogenous proteins expressed in the complete knockout line had N-glycans that lacked b(1,2)-xylose and had a significant reduction in core a(1,3)-fucose levels (40% of wild type). A similar phenotype was observed in the N-glycans of a recombinant rituximab antibody transiently expressed in the homozygous mutant plants. More importantly, the most desirable glycoform, one lacking both core a(1,3)-fucose and b(1,2)-xylose residues, increased in the antibody from 2% when produced in the wild-type line to 55% in the mutant line. These results demonstrate the power of TALENs for multiplexed gene editing. Furthermore, the mutant N. benthamiana lines provide a valuable platform for producing highly potent biopharmaceutical products.
Introduction Biopharmaceuticals are typically protein-based medicines produced by genetically engineering living cells or organisms, such as microbes, insect or mammalian cell lines (e.g. Chinese hamster ovary cells or human fibroblast cells) (Yin et al., 2007). The lengthy development times, scalability, high costs and safety associated with these systems, however, have stimulated research into developing alternative production platforms. Recent successes with highly efficient plant expression systems offer a viable alternative to produce large quantities of biopharmaceuticals at reduced costs (Hiatt et al., 1989; Mett et al., 2008; Stoger et al., 2002; Twyman et al., 2003; Yusibov and Mamedov, 2010). The advantages of plant-based expression systems include the inability of plant cells to propagate mammalian viruses and pathogens, greater scalability, lower production and bioprocessing costs (Stoger et al., 2014), and the capacity to engineer posttranslational modifications, such as N-glycosylation, to maximize the potency of therapeutic glycoproteins (Loos and Steinkellner, 2014). N-glycosylation is the covalent linkage of an oligosaccharide side chain to a protein through the amide nitrogen of an asparagine (Asn) residue. As an important post-translational
modification, N-glycosylation affects the folding, stability and biological activity of biopharmaceuticals produced in heterologous expression systems (Gomord and Faye, 2004; Gomord et al., 2010). Whereas plants are able to synthesize an N-glycan core structure identical to mammalian cells, plant-derived recombinant glycoproteins can also have N-glycans with core a (1,3)-fucose and b(1,2)-xylose, which are absent in mammalianproduced proteins (Cabanes-Macheteau et al., 1999). The presence of plant-specific residues may negatively impact the quality of therapeutic proteins. For example, it has been shown that the presence of core a(1,3)-fucose on the N-glycan of the Fc region of monoclonal antibodies significantly reduces antibody-dependent cell-mediated cytotoxicity (ADCC) activity of the antibody (Cox et al., 2006; Schuster et al., 2007). Also, although it has been shown recently that administration of the plant-produced taliglucerase alfa and influenza virus-like particles carrying plant plant-specific N-glycans did not elicit adverse immune responses (Grabowski et al., 2014; Ward et al., 2014), the possibility remains that, in particular cases, the presence of plant-specific N-glycan residues on therapeutic proteins leads to undesirable immune responses. In order for plants to become a more competitive bioreactor for human-consumed biopharmaceuticals, plant expression systems would greatly benefit from the avail-
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd
533
534 Jin Li et al. ability of engineered plant lines with the capacity to produce optimal forms of glycosylated therapeutic proteins on the basis of modified N-glycosylation pathway. In recent years, extensive research efforts have been undertaken to modulate plant-specific N-glycosylation machinery, resulting in recombinant proteins with mammalian-like modifications. The enzymes responsible for the attachment of core a(1,3)-fucose and b(1,2)-xylose residues in plants are a(1,3)fucosyltransferase (FucT) and b(1,2)-xylosyltransferase (XylT) (Strasser et al., 2004) (Figure 1a). Interruption of FucT and XylT gene functions has been achieved via RNA interference (RNAi) technology and random mutagenesis methods in a number of plant species including, Arabidopsis, Lemna minor, Nicotiana benthamiana, alfalfa and rice (Castilho and Steinkellner, 2012).
These resulting mutant lines exhibit the desired phenotypes, namely significantly reduced levels of plant-specific a(1,3)-fucose and b(1,2)-xylose residues in recombinant proteins (Cox et al., 2006; Strasser et al., 2008; Yin et al., 2011). Unfortunately, these approaches suffer from several major drawbacks. For example, RNAi technology is often unable to completely deactivate the functions of the target genes, especially when more than one gene or gene family is targeted. Although traditional mutagenesis approaches, such as T-DNA, EMS, gamma rays and fast neutrons, can produce full knockout mutations, these mutations are randomly introduced into the genome with no control over location and the types of mutations created (Ostergaard and Yanofsky, 2004; Wang et al., 2012). This lack of specificity requires large-scale screens of mutagenized populations to
(a)
(b)
(c)
(d)
Figure 1 Targeting mutations in the FucT1/FucT2 and XylT1/XylT2 genes. (a) The FucT and XylT genes are responsible for the attachment of core a(1,3)fucose and b(1,2)-xylose residues to N-glycans in plants. (b) Schematics of the FucT1/FucT2 and XylT1/XylT2 genes. The FucT1/FucT2 genes contain 7 exons and 6 introns, and the XylT1/XylT2 genes contain 3 exons and 2 introns, represented by the open boxes and angled lines, respectively. TALEN target sites are indicated by arrowheads in the first exon of each gene. (c) DNA sequences of TALEN target sites in the FucT1/FucT2 and (d) the XylT1/XylT2 genes. Each TALEN target site consists of two TALEN recognition sequences, denoted in upper-case letters; the spacer sequences are shown in lower case. DNA sequence polymorphisms between FucT1 and FucT2 are indicated. ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
Glyco-engineering through genome editing 535 identify candidate plants with knockout mutations in the genes of interest. Moreover, the recovered plants often need to go through a multigeneration backcrossing regime to remove unwanted mutations randomly introduced at other loci. The shortcomings of the RNAi knock-down and random mutagenesis knockout approaches make it desirable to develop efficient methods to generate specific and targeted mutations in the genome of plants chosen for biopharmaceutical production. Recent advances in plant genome engineering have allowed for the generation of targeted mutations in a highly specific and efficient manner. Current genome engineering technology is based on the ability to engineer sequence-specific nucleases, such as meganucleases, zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs) or CRISPR/Cas nucleases (Puchta and Fauser, 2014; Voytas, 2013). These nucleases are engineered to recognize a genomic target sequence and subsequently cleave the DNA, resulting in double-stranded breaks (DSBs). The DSBs can then be repaired by either the nonhomologous end joining (NHEJ) or homologous recombination (HR) pathways, resulting in knockout mutations, precise site-directed point mutations or targeted gene insertions. In recent years, the efficiency of genome engineering technology has been dramatically improved, particularly with the development of TALEN and CRISPR/Cas nucleases. These two newest members in the designer nuclease family are easier to engineer for a desired target sequence and often have high cleavage activity. These benefits have been demonstrated in a wide range of organism at both single and multiple loci making them the reagents of choice for genome modification in many species (Baltes and Voytas, 2015). In this study, TALEN-mediated gene editing technology was employed to manipulate N-glycosylation pathways in N. benthamiana for the production of biopharmaceuticals with mammalian-like N-glycan structures. The highly efficient TALENs designed in this study enabled multiplexed gene knockouts in four genes, two XylT and two FucT genes, creating mutants with the desired glycosylation profiles in one generation. The resulting mutants were used to produce rituximab antibodies with an absence or reduction in plant-specific N-glycans. The high efficacy of TALEN technology for multiplexed gene editing of plants, as demonstrated in this study, will open up opportunities to accelerate research and development in plant metabolic engineering and synthetic biology.
Results TALEN design and activity assessment at the FucT and XylT target sites In the genome of N. benthamiana, five FucT and two XylT genes have been identified and characterized (Weterings and Van Eldik, 2010, 2013). Because of the high copy number of the FucT gene family, phylogenetic analysis of the FucT genes was first conducted to understand their relationship. The result indicated the formation of two clusters, one containing FucT1 and FucT2 and the other containing FucT3, FucT4 and FucT5 (Figure S1). In previous research, relative expression and activity of the FucT genes have been characterized (Weterings and Van Eldik, 2013). As a result, FucT1 and FucT2 showed the highest expression of all five genes. Thus, knockout of both FucT1 and FucT2 may result in the largest reduction in a(1,3)-fucose residues. Furthermore, the similarity in sequence between FucT1 and FucT2 allowed us to design TALEN pairs that target both genes simultaneously.
Sequence-specific TALENs were designed to knock out two FucT genes (FucT1 and FucT2) and two XylT genes (XylT1 and XylT2) in Nicotiana benthamiana. Conserved sequences in the first exon of the FucT1 and FucT2 genes were analysed by TALENTM hit software, and four potential target sites for each gene were identified (Figure 1b). Between FucT1 and FucT2, single nucleotide polymorphisms are present in the TALEN recognition sequences of target sites 1 and 2 and the spacer sequences of target sites 3 and 4 (Figure 1c). Eleven TALEN monomers were designed to recognize all the FucT1 and FucT2 target sites and were constructed using methods previously described (Beurdeley et al., 2013). Similarly, four TALEN target sites were identified in the first exon of the XylT1 and XylT2 genes (Figure 1b, d). As each target site is conserved in both XylT genes, only four TALEN pairs were required to target all sites in both XylT genes. To assess the activity of the TALENs targeting the FucT and XylT genes, a yeast-based assay was first performed using a method described previously (Haun et al., 2014; Christian et al., 2012). As indicated in Table S1, five of six FucT TALENs and three of four XylT TALENs showed high cleavage activities at both 37°C and 30°C, as compared to our positive control, I-SceI. TALENs with the highest activity were chosen for further testing at endogenous target sites in protoplasts isolated from N. benthamiana leaf tissue. TALEN expression cassettes were transformed into protoplasts and then the samples were sacrificed for preparation of genomic DNA after a 48-h incubation period. The genomic DNA was subjected to PCR to produce amplicons of approximately 300 bp that encompass the TALEN target sites. The amplicons were then subjected to 454 pyro-sequencing, and the resulting sequencing reads were evaluated for NHEJ-induced mutations. The frequency of mutated sequences in each sample was used to estimate the activity of the corresponding TALENs. The two FucT TALEN pairs, FucT1_T02 and FucT2_T02, displayed high frequencies of NHEJ-induced mutagenesis, 33.3% and 39.1%, respectively, at their corresponding target sites (Table 1). As the FucT1_T02 and FucT2_T02 target sites differ by only 1 bp (A/G) in the left TALEN recognition site, activity of each TALEN pair was examined at both FucT genes. Interestingly, the results indicate that both FucT1_T02 and FucT2_T02 recognize and cleave both genes with similar efficiencies; therefore, we predicted that it should be possible to use a single TALEN pair to disrupt both genes simultaneously. The XylT TALENs, XylT_03 and XylT_04, were also highly mutagenic, and 22.8% and 61.9% of the respective sequencing reads had mutations at the target sites (Table 1). Examples of TALENinduced mutations in both FucT and XylT genes are shown in Figure S3.
Generation of FucT and XylT mutants via multiplexed, targeted gene editing The two TALENs, FucT2_T02 and XylT_T04, which showed the highest activity in protoplasts, were chosen to create lines with mutations in the FucT and XylT genes. Leaf protoplasts were transformed with either the FucT2_T02 TALEN only (group 1), the XylT_T04 TALEN only (group 2), or both the FucT2_T02 and XylT_T04 TALENs (group 3). The transformed protoplasts were then regenerated to individual plantlets on nonselective media. Leaf tissues from regenerated plantlets were genotyped by PCR and DNA sequencing. From transformation group 1, 160 regenerated plantlets were screened, revealing 80 plantlets (50%) with mutations in at least
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
536 Jin Li et al. Table 1 TALEN activity assessment in N. benthamiana protoplasts
TALEN name FucT1_T02 FucT2_T02 XylT_T03 XylT_T04
Gene targets
NHEJ-induced
Protoplast
mutagenesis
transformation
frequency*
efficiency†
FucT1
33.3% (27 470)
87%
FucT2
21.4% (7259)
87%
FucT1
39.7% (12 295)
87%
FucT2
39.1% (6457)
87%
XylT1
22.8% (3169)
81%
XylT2
27.5% (420)
81%
XylT1
61.9% (1615)
84%
XylT2
53.3% (464)
84%
*Mutagenesis frequency is the percentage of 454 sequencing reads with NHEJinduced mutations. The total number of the sequencing reads for each sample is indicated in parentheses. The NHEJ mutagenesis frequency was zero for the negative control, which was only transformed with the YFP construct. †
Protoplast efficiency is determined by dividing the number of YFP-positive
protoplasts by the number of total protoplasts (plasmid encoding YFP was cotransformed with plasmids encoding TALENs).
one allele of either the FucT1 or FucT2 genes and 42 plantlets (26%) with mutations in both FucT1 and FucT2 genes (Table 2). Two plantlets, designated NB13-105a and NB12-213a, had mutations in all alleles of both FucT1 and FucT2. In the plant line NB13-105a, a 13-bp deletion was detected in both alleles of the FucT1 gene and a 14-bp deletion was found in both alleles of the FucT2 gene (Figure 2a). No wild-type sequences were identified for either FucT1 or FucT2, suggesting this plantlet had knockout mutations in all alleles of both FucT genes. In the plant line NB13-213a, 20-bp and 12-bp deletions occurred in the FucT1 alleles, and 13-bp and 2-bp deletions occurred in the FucT2 alleles (Figure 2a). No wild-type sequence was recovered for either gene; therefore, NB12-213a likely contains distinct mutations in all alleles of the FucT1 and FucT2 genes. In transformation group 2, 129 regenerated plantlets were screened with 94 plantlets (73%) having mutations in at least one allele of either XylT1 or XylT2, and 53 plantlets (41%) had mutations in both genes. Two plantlets, NB15-11d and NB12-113c, had knockout mutations in all alleles of both XylT genes. Plant line NB15-11d had two distinct 7-bp deletions in the XylT1 gene and an identical 8-bp deletion in both alleles of XylT2
(Figure 2a). Plant line NB12-113c had identical 7-bp deletions in XylT1 as well as 35-bp and 5-bp deletions in XylT2 (Figure 2a). No wild-type sequences for the target sites of either XylT1 or XylT2 were recovered from these two plant lines. With transformation group 3, the goal was to perform multiplexed, targeted gene editing and create mutant lines with knockout mutations in all alleles of all four XylT and FucT genes. We cotransformed protoplasts with four plasmids encoding the FucT2_T02 and XylT_T04 TALEN pairs and regenerated a total of 105 plantlets. Of these plantlets, 54 (51%) had mutations in at least one allele of either the XylT or FucT genes, and 18 (17%) had mutations in all four genes. Among the 18 mutants, three had bi-allelic mutations in all four genes. One of these mutants, NB14-29aT0, had identical 44-bp deletions in FucT1, a 2-bp and a 40-bp deletion in FucT2, identical 5-bp deletions in XylT1 and a 6-bp and 549-bp deletion in XylT2 (Figure 2b). The completely mutant plant, NB14-29aT0, derived from transformation group 3 was allowed to set seed in order to assess transmission of the mutations to the next generation. T1 seeds were collected, and 60 T1 plants were genotyped to detect mutations in the XylT and FucT genes. The genotyping results showed that all the mutations present in the T0 plants were transmitted to the T1 generation (data not shown). No wild-type alleles were found in any of the genotyped T1 plants, further confirming that all the mutant plants had all of the targeted genes completely knocked out.
Reduction of a(1,3)-fucose and b(1,2)-xylose levels in endogenous proteins from N. benthamiana mutants Plants with confirmed mutant genotypes were subjected to N-glycosylation profile analysis in which endogenous proteins were assayed for levels of a(1,3)-fucose and b(1,2)-xylose via methods previously described (North et al., 2009; Strasser et al., 2008). Both FucT and XylT mutant plants had significant changes in their N-glycan profiles. FucT mutant plants NB13-105a and NB13-213a had overall a(1,3)-fucose levels reduced by 61% (Figure 3a). XylT mutant plants, NB15-11d and NB12-113c, had b (1,2)-xylose levels that were 23 (14%§
2
XylT1/XylT2
129
94 (73%)
53 (41%)
>22 (17%)§
3
FucT1/FucT2
105
54 (51%)
18 (17%)
3 (3%)
XylT1/XylT2 *The number of plants that contain mutations in at least one targeted gene. †
The number of plants that contain mutations in all targeted genes. Mutations could occur in one or both alleles of a given target.
‡
The number of plants that contain mutations in every allele of all targeted genes. Mutations occur in both alleles of a given target.
§
Plants containing in-frame mutations were not included in this number, even if the plant contained a mutation in every allele of all the targets.
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
Glyco-engineering through genome editing 537 (a)
(b)
Figure 2 Genetically transmissible mutations induced by TALENs in N. benthamiana plants. (a) TALEN-induced mutations in FucT1/FucT2 or XylT1/XylT2 knockouts. The plants NB13-105a and NB13-213a contain mutations in both alleles of FucT1 and FucT2; the plants NB15-11d and NB12-113c have mutations in both alleles of XylT1 and XylT2. TALEN recognition sequences are indicated in upper case; the spacer sequences are in lower case. Deletions are indicated by dashes, and the size of each deletion in base pairs is shown on the right. (b) TALEN-induced mutations in the plant NB14-29a that has mutations in all alleles of FucT1, FucT2, XylT1 and XylT2.
however, b(1,2)-xylose glycosylation was only reduced by 36%, as compared to the XylT-only mutant lines that had b(1,2)-xylose glycosylation 2 bp in the spacer
region were considered to be products of imprecise repair by NHEJ of a cleaved TALEN recognition site. Mutagenesis frequency is reported as the percentage of sequencing reads with NHEJ-induced mutations.
Plant regeneration from N. benthamiana protoplasts Immediately after transformation, protoplasts were suspended in liquid K3G1 medium at a cell density of 1 9 105/mL in a small petri dish and stored at 25°C in the dark. Four days after transformation, when the majority of the protoplasts had begun the first cell divisions, the culture was diluted fourfold with Media C (van den Elzen et al., 1985). At day 7 and day 10, the protoplast cultures were diluted twofold with MS media. At day 18, calli were observed under the light microscope, 1 month after transformation calli were visible to the eye. The individual calli were transferred to shoot-inducing MS medium (van den Elzen et al., 1985). After shoots emerged, the plantlets were cut at the base and transferred to root-inducing medium (van den Elzen et al., 1985). When roots formed, the plantlets were transferred to soil and grown to maturity.
Genotyping FucT and XylT mutant lines Genomic DNA isolated from regenerated plantlets was PCRamplified using gene-specific primer pairs for the FucT1, FucT2, XylT1 and XylT2 loci (primers are listed in Table S2). The PCR products were subjected to Sanger sequencing, and the sequencing trace files were analysed to identify candidate mutants. If a trace file had double peaks around the TALEN target sites, the plant was considered a candidate mutant for further analysis. Samples with clean traces were aligned with the wild-type target sequences to determine whether the same mutations were present at both alleles. Sequences of the mutant candidates were further analysed by cloning the PCR products. For each sample, 16–24 colonies were PCR-amplified and sequenced, and sequences from individual clones were aligned with wild-type target sequences to identify mutations. All primers used for genotyping are summarized in Table S2.
Glycosylation profiles of endogenous N. benthamiana leaf proteins One-month-old leaves from the N. benthamiana lines were harvested for protein isolation. Two grams of fresh leaves was ground into fine powder in liquid nitrogen and transferred to a 15-mL tube. Three volumes of extraction buffer (50 mM Tris, 150 mM NaCl, pH 8) were added along with 60 lL PMSF (100 mM in 100% isopropanol). The samples were then vortexed vigorously for 3 min and centrifuged at 5000 g for 15 min at 4°C. Supernatants were collected and mixed gradually with ammonium sulphate to a concentration of 10.6 g/ 100 mL to increase solubility of proteins via salting-in. The protein extracts were then centrifuged at 18 000 g for 20 min at 4°C. Supernatants were transferred to a fresh tube and mixed gradually with ammonium sulphate to a concentration of 38.7 g/100 mL to precipitate proteins via salting-out. Protein pellets were obtained by centrifuging at 18 000 g for 20 min at 4°C. Salts were removed from protein extracts using G2 dialysis cassettes (Thermo Fisher Scientific, Waltham, MA). N-glycans were released from the protein extracts by digestion with peptide-N-glycosidase A and F. Released N-glycans were purified using Ultra Clean SPE Carbograph columns (Thermo Fisher Scientific, Waltham, MA) and methylated in vitro. Permethylated N-glycans were analysed by ProteoDynamics Inc. (Clermont-
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
Glyco-engineering through genome editing 541 Ferrand, France) via MALDI TOF MS (Karas and Hillenkamp, 1988).
Rituximab antibody production in N. benthamiana Thirty-eight-day-old plants were transformed using Agrobacterium with T-DNA encoding the rituximab antibody light and heavy chains. The transient co-expression approach was similar to that described previously (Vezina et al., 2009) except that the rituximab light and heavy chain coding regions (GenBank Accession nos. AX709550 and AX709548, respectively) were placed under the control of a proprietary CPMV (cow pea mosaic virus)-derived expression system and the TBSV P19 was used as suppressor of silencing (Voinnet et al., 2003). Nine days post-agroinfiltration, the plant leaves were harvested and the rituximab antibody was purified. Total soluble proteins were extracted with 2 volumes of extraction buffer (50 mM Tris, pH 8.0, 150 mM NaCl) and the extract was centrifuged for 2 min at 20 000 g; the supernatant was filtered through Miracloth to remove insoluble material. The clarified protein extract was loaded onto a protein A affinity chromatography column (MabSure Select LX; GE Healthcare Life Sciences, Baie d’Urfe, QC, Canada). The column was washed with extraction buffer, and the antibody was eluted using 100 mM glycine pH 3.0. Elution fractions were immediately neutralized with 2M Tris pH 7.4 buffer, and the collected fractions were analysed by Coomassie-stained SDS-PAGE to assess antibody quantity and purity. Selected fractions were pooled and analysed for protein content.
LC-ESI MS/MS analysis of N-glycosylation in the rituximab antibody N-glycosylation of the rituximab antibody was analysed by LC-ESI MS/MS using a NanoAcquity UPLC coupled to a QTOF micro (Waters, Milford, MA). Reduction (DTT) and alkylation (iodoacetamide) of cysteine residues were performed prior to overnight digestion with trypsin (Promega, Madison, WI, Sequencing Grade) at 37°C. Half of the sample was subjected to deglycosylation with peptide-N(4)-(acetyl-b-glucosaminyl) asparagine amidase A (PNGase A; Roche, Ontario, Canada) at 37°C. Tryptic peptides and glycopeptides were chromatographically separated on a ProteoPep III C18 Integrafrit column (New Objective, Woburn, MA) using a water–acetonitrile gradient and analysed by MS/MS, using data-dependant acquisition. Peptides are identified using the Mascot search engine (Matrix Science, Boston, MA) and UniProt databases. Samples subjected to deglycosylation are used to identify the site of glycosylation, as PNGase A induces a substitution of Asn by Asp, which is detected by Mascot. The site of glycosylation in rituximab at position 301 was covered with two peptides: EEQYnSTYR and TKPREEQYnSTYR (n = site of N-glycosylation). Starting from these two identified peptides and the list of N-glycans typically found in plants, a list of putative glycopeptides, with their theoretical m/z, was created. The detection of the m/z in the LC-MS/MS analysis of samples without deglycosylation indicates which N-glycans are attached at site 301. Glycopeptide assignments were also confirmed by the presence of reporter ions in the MS/MS spectra: m/z 204, 366 and 512 indicating N-acetylglucosamine (GlcNAc), GlcNAc-hexose and GlcNAc-hexose-fucose, respectively.
Acknowledgements The authors would like to thank Dr. Nicholas Baltes for critical reading.
Conflict of interest JL, TJS, ZLD, SL, BMC, EER, AC, AD, AY, AR, LM, DFV and FZ are employees of Cellectis Plant Sciences Inc., a subsidiary of Cellectis SA. FC is an employee of Cellectis SA. POL and MAD are employees of Medicago R&D, a subsidiary of Medicago Inc.
References Arnould, S., Chames, P., Perez, C., Lacroix, E., Duclert, A., Epinat, J.C., Stricher, F., Petit, A.S., Patin, A., Guillier, S., Rolland, S., Prieto, J., Blanco, F.J., Bravo, J., Montoya, G., Serrano, L., Duchateau, P. and Paques, F. (2006) Engineering of large numbers of highly specific homing endonucleases that induce recombination on novel DNA targets. J. Mol. Biol. 355, 443–458. Baltes, N.J. and Voytas, D.F. (2015) Enabling plant synthetic biology through genome engineering. Trends Biotechnol. 33, 120–131. Beurdeley, M., Bietz, F., Li, J., Thomas, S., Stoddard, T., Juillerat, A., Zhang, F., Voytas, D.F., Duchateau, P. and Silva, G.H. (2013) Compact designer TALENs for efficient genome engineering. Nat. Commun. 4, 1762. Bombarely, A., Rosli, H.G., Vrebalov, J., Moffett, P., Mueller, L.A. and Martin, G.B. (2012) A draft genome sequence of Nicotiana benthamiana to enhance molecular plant-microbe biology research. Mol. Plant-Microbe Interact. 25, 1523–1530. Cabanes-Macheteau, M., Fitchette-Laine, A.C., Loutelier-Bourhis, C., Lange, C., Vine, N.D., Ma, J.K., Lerouge, P. and Faye, L. (1999) N-Glycosylation of a mouse IgG expressed in transgenic tobacco plants. Glycobiology, 9, 365–372. Castilho, A. and Steinkellner, H. (2012) Glyco-engineering in plants to produce human-like N-glycan structures. Biotechnol. J. 7, 1088–1098. Christian, M.L., Demorest, Z.L., Starker, C.G., Osborn, M.J., Nyquist, M.D., Zhang, Y., Carlson, D.F., Bradley, P., Bogdanove, A.J. and Voytas, D.F. (2012) Targeting G with TAL effectors: a comparison of activities of TALENs constructed with NN and NK repeat variable di-residues. PLoS ONE, 7, e45383. Cox, K.M., Sterling, J.D., Regan, J.T., Gasdaska, J.R., Frantz, K.K., Peele, C.G., Black, A., Passmore, D., Moldovan-Loomis, C., Srinivasan, M., Cuison, S., Cardarelli, P.M. and Dickey, L.F. (2006) Glycan optimization of a human monoclonal antibody in the aquatic plant Lemna minor. Nat. Biotechnol. 24, 1591–1597. Davis, A.P. and Symington, L.S. (2004) RAD51-dependent break-induced replication in yeast. Mol. Cell. Biol. 24, 2344–2351. Gomord, V. and Faye, L. (2004) Posttranslational modification of therapeutic proteins in plants. Curr. Opin. Plant Biol. 7, 171–181. Gomord, V., Fitchette, A.C., Menu-Bouaouiche, L., Saint-Jore-Dupas, C., Plasson, C., Michaud, D. and Faye, L. (2010) Plant-specific glycosylation patterns in the context of therapeutic protein production. Plant Biotechnol. J. 8, 564–587. Grabowski, G.A., Golembo, M. and Shaaltiel, Y. (2014) Taliglucerase alfa: an enzyme replacement therapy using plant cell expression technology. Mol. Genet. Metab. 112, 1–8. Haun, W., Coffman, A., Clasen, B.M., Demorest, Z.L., Lowy, A., Ray, E., Retterath, A., Stoddard, T., Juillerat, A., Cedrone, F., Mathis, L., Voytas, D.F. and Zhang, F. (2014) Improved soybean oil quality by targeted mutagenesis of the fatty acid desaturase 2 gene family. Plant Biotechnol J. 12, 934–940. Hiatt, A., Cafferkey, R. and Bowdish, K. (1989) Production of antibodies in transgenic plants. Nature, 342, 76–78. Karas, M. and Hillenkamp, F. (1988) Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal. Chem. 60, 2299–2301. Kraus, E., Leung, W.Y. and Haber, J.E. (2001) Break-induced replication: a review and an example in budding yeast. Proc. Natl Acad. Sci. USA, 98, 8255–8262. Loos, A. and Steinkellner, H. (2014) Plant glyco-biotechnology on the way to synthetic biology. Front. Plant Sci. 5, 523. Mashiko, D., Fujihara, Y., Satouh, Y., Miyata, H., Isotani, A. and Ikawa, M. (2013) Generation of mutant mice by pronuclear injection of circular plasmid expressing Cas9 and single guided RNA. Sci. Rep. 3, 3355. Mett, V., Farrance, C.E., Green, B.J. and Yusibov, V. (2008) Plants as biofactories. Biologicals, 36, 354–358.
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
542 Jin Li et al. Moscou, M.J. and Bogdanove, A.J. (2009) A simple cipher governs DNA recognition by TAL effectors. Science, 326, 1501. North, S.J., Hitchen, P.G., Haslam, S.M. and Dell, A. (2009) Mass spectrometry in the analysis of N-linked and O-linked glycans. Curr Opin Struct Biol 19, 498–506. Ostergaard, L. and Yanofsky, M.F. (2004) Establishing gene function by mutagenesis in Arabidopsis thaliana. Plant J. 39, 682–696. Puchta, H. and Fauser, F. (2013) Gene targeting in plants: 25 years later. Int. J. Dev. Biol. 57, 629–637. Puchta, H. and Fauser, F. (2014) Synthetic nucleases for genome engineering in plants: prospects for a bright future. Plant J., 78, 727–741. Schuster, M., Jost, W., Mudde, G.C., Wiederkum, S., Schwager, C., Janzek, E., Altmann, F., Stadlmann, J., Stemmer, C. and Gorr, G. (2007) In vivo glycoengineered antibody with improved lytic potential produced by an innovative non-mammalian expression system. Biotechnol. J. 2, 700–708. Stoger, E., Fischer, R., Moloney, M. and Ma, J.K. (2014) Plant molecular pharming for the treatment of chronic and infectious diseases. Annu. Rev. Plant Biol. 65, 743–768. Stoger, E., Sack, M., Fischer, R. and Christou, P. (2002) Plantibodies: applications, advantages and bottlenecks. Curr. Opin. Biotechnol. 13, 161–166. Strasser, R., Altmann, F., Mach, L., Glossl, J. and Steinkellner, H. (2004) Generation of Arabidopsis thaliana plants with complex N-glycans lacking beta1,2-linked xylose and core alpha1,3-linked fucose. FEBS Lett. 561, 132–136. Strasser, R., Stadlmann, J., Schahs, M., Stiegler, G., Quendler, H., Mach, L., Glossl, J., Weterings, K., Pabst, M. and Steinkellner, H. (2008) Generation of glyco-engineered Nicotiana benthamiana for the production of monoclonal antibodies with a homogeneous human-like N-glycan structure. Plant Biotechnol. J. 6, 392–402. Twyman, R.M., Stoger, E., Schillberg, S., Christou, P. and Fischer, R. (2003) Molecular farming in plants: host systems and expression technology. Trends Biotechnol. 21, 570–578. van den Elzen, P.J.M., Townsend, J., Lee, K.Y. and Bedbrook, J.R. (1985) A chimaeric hygromycin resistance gene as a selectable marker in plant cells. Plant Mol. Biol. 5, 299–302. V ezina, L.P., Faye, L., Lerouge, P., D’Aoust, M.A., Marquet-Blouin, E., Burel, C., Lavoie, P.O., Bardor, M. and Gomord, V. (2009) Transient co-expression for fast and high-yield production of antibodies with human-like N-glycans in plants. Plant Biotechnol. J. 7, 442–455. Voinnet, O., Rivas, S., Mestre, P. and Baulcombe, D. (2003) An enhanced transient expression system in plants based on suppression of gene silencing by the p19 protein of tomato bushy stunt virus. Plant J. 33, 949–956. Voytas, D.F. (2013) Plant genome engineering with sequence-specific nucleases. Annu. Rev. Plant Biol. 64, 327–350. Wang, H., Yang, H., Shivalila, C.S., Dawlaty, M.M., Cheng, A.W., Zhang, F. and Jaenisch, R. (2013) One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering. Cell, 153, 910–918.
Wang, T.L., Uauy, C., Robson, F. and Till, B. (2012) TILLING in extremis. Plant Biotechnol. J. 10, 761–772. Ward, B.J., Landry, N., Trepanier, S., Mercier, G., Dargis, M., Couture, M., D’Aoust, M.A. and Vezina, L.P. (2014) Human antibody response to N-glycans present on plant-made influenza virus-like particle (VLP) vaccines. Vaccine, 32, 6098–6106. Weterings, K. and Van Eldik, G. (2013) Nicotiana benthamiana plants deficient in fucosyltransferase activity. International patent application WO 2013/ 050155. Weterings, K. and Van Eldik, G. (2010) Nicotiana benthamiana plants deficient in xylosyltransferase activity. International patent application WO 2010/ 145846. Wright, D.A., Townsend, J.A., Winfrey, R.J. Jr, Irwin, P.A., Rajagopal, J., Lonosky, P.M., Hall, B.D., Jondle, M.D. and Voytas, D.F. (2005) Highfrequency homologous recombination in plants mediated by zinc-finger nucleases. Plant J. 44, 693–705. Yin, B.J., Gao, T., Zheng, N.Y., Li, Y., Tang, S.Y., Liang, L.M. and Xie, Q. (2011) Generation of glyco-engineered BY2 cell lines with decreased expression of plant-specific glycoepitopes. Protein Cell, 2, 41–47. Yin, J., Li, G., Ren, X. and Herrler, G. (2007) Select what you need: a comparative evaluation of the advantages and limitations of frequently used expression systems for foreign genes. J. Biotechnol. 127, 335–347. Yoo, S.D., Cho, Y.H. and Sheen, J. (2007) Arabidopsis mesophyll protoplasts: a versatile cell system for transient gene expression analysis. Nat. Protoc. 2, 1565–1572. Yusibov, V.M. and Mamedov, T.G. (2010) Plants as an alternative system for expression of vaccine antigens. Proc. ANAS (Biol. Sci.) 65, 195–200. Zhang, Y., Zhang, F., Li, X., Baller, J.A., Qi, Y., Starker, C.G., Bogdanove, A.J. and Voytas, D.F. (2013) Transcription activator-like effector nucleases enable efficient plant genome engineering. Plant Physiol. 161, 20–27.
Supporting information Additional Supporting information may be found in the online version of this article: Figure S1 Phylogenetic analysis of the FucT genomic sequences. Figure S2 Illustration of a representative TALEN-expression plasmid. Figure S3 Examples of TALEN-induced mutations in the FucT and XylT gene target sequences. Table S1 TALEN activity assessment in a yeast single-strand annealing assay. Table S2 Primers used for 454 deep sequencing and genotyping analysis.
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
doi: 10.1111/pbi.12403
Multiplexed, targeted gene editing in Nicotiana benthamiana for glyco-engineering and monoclonal antibody production Jin Li1, Thomas J. Stoddard1, Zachary L. Demorest1, Pierre-Olivier Lavoie2, Song Luo1, Benjamin M. Clasen1, Frederic Cedrone3, Erin E. Ray1, Andrew P. Coffman1, Aurelie Daulhac1, Ann Yabandith1, Adam J. Retterath1, Luc Mathis1, Daniel F. Voytas1, Marc-Andr e D’Aoust2,* and Feng Zhang1,* 1 2 3
Cellectis Plant Sciences Inc., New Brighton, MN, USA Medicago Inc., Quebec, QC, Canada Cellectis SA, Paris, France
Received 17 December 2014; revised 20 April 2015; accepted 21 April 2015. *Correspondence (Tel +1(651)683-2807; fax +1(651)683-2805; email [email protected]) and (Tel +1(418)658-9393; fax +1(418)658-6699; email [email protected])
Keywords: Gene editing, transcription activator-like effector nucleases, glyco-engineering, plant-derived pharmaceuticals.
Summary Biopharmaceutical glycoproteins produced in plants carry N-glycans with plant-specific residues core a(1,3)-fucose and b(1,2)-xylose, which can significantly impact the activity, stability and immunogenicity of biopharmaceuticals. In this study, we have employed sequence-specific transcription activator-like effector nucleases (TALENs) to knock out two a(1,3)-fucosyltransferase (FucT) and the two b(1,2)-xylosyltransferase (XylT) genes within Nicotiana benthamiana to generate plants with improved capacity to produce glycoproteins devoid of plant-specific residues. Among plants regenerated from N. benthamiana protoplasts transformed with TALENs targeting either the FucT or XylT genes, 50% (80 of 160) and 73% (94 of 129) had mutations in at least one FucT or XylT allele, respectively. Among plants regenerated from protoplasts transformed with both TALEN pairs, 17% (18 of 105) had mutations in all four gene targets, and 3% (3 of 105) plants had mutations in all eight alleles comprising both gene families; these mutations were transmitted to the next generation. Endogenous proteins expressed in the complete knockout line had N-glycans that lacked b(1,2)-xylose and had a significant reduction in core a(1,3)-fucose levels (40% of wild type). A similar phenotype was observed in the N-glycans of a recombinant rituximab antibody transiently expressed in the homozygous mutant plants. More importantly, the most desirable glycoform, one lacking both core a(1,3)-fucose and b(1,2)-xylose residues, increased in the antibody from 2% when produced in the wild-type line to 55% in the mutant line. These results demonstrate the power of TALENs for multiplexed gene editing. Furthermore, the mutant N. benthamiana lines provide a valuable platform for producing highly potent biopharmaceutical products.
Introduction Biopharmaceuticals are typically protein-based medicines produced by genetically engineering living cells or organisms, such as microbes, insect or mammalian cell lines (e.g. Chinese hamster ovary cells or human fibroblast cells) (Yin et al., 2007). The lengthy development times, scalability, high costs and safety associated with these systems, however, have stimulated research into developing alternative production platforms. Recent successes with highly efficient plant expression systems offer a viable alternative to produce large quantities of biopharmaceuticals at reduced costs (Hiatt et al., 1989; Mett et al., 2008; Stoger et al., 2002; Twyman et al., 2003; Yusibov and Mamedov, 2010). The advantages of plant-based expression systems include the inability of plant cells to propagate mammalian viruses and pathogens, greater scalability, lower production and bioprocessing costs (Stoger et al., 2014), and the capacity to engineer posttranslational modifications, such as N-glycosylation, to maximize the potency of therapeutic glycoproteins (Loos and Steinkellner, 2014). N-glycosylation is the covalent linkage of an oligosaccharide side chain to a protein through the amide nitrogen of an asparagine (Asn) residue. As an important post-translational
modification, N-glycosylation affects the folding, stability and biological activity of biopharmaceuticals produced in heterologous expression systems (Gomord and Faye, 2004; Gomord et al., 2010). Whereas plants are able to synthesize an N-glycan core structure identical to mammalian cells, plant-derived recombinant glycoproteins can also have N-glycans with core a (1,3)-fucose and b(1,2)-xylose, which are absent in mammalianproduced proteins (Cabanes-Macheteau et al., 1999). The presence of plant-specific residues may negatively impact the quality of therapeutic proteins. For example, it has been shown that the presence of core a(1,3)-fucose on the N-glycan of the Fc region of monoclonal antibodies significantly reduces antibody-dependent cell-mediated cytotoxicity (ADCC) activity of the antibody (Cox et al., 2006; Schuster et al., 2007). Also, although it has been shown recently that administration of the plant-produced taliglucerase alfa and influenza virus-like particles carrying plant plant-specific N-glycans did not elicit adverse immune responses (Grabowski et al., 2014; Ward et al., 2014), the possibility remains that, in particular cases, the presence of plant-specific N-glycan residues on therapeutic proteins leads to undesirable immune responses. In order for plants to become a more competitive bioreactor for human-consumed biopharmaceuticals, plant expression systems would greatly benefit from the avail-
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd
533
534 Jin Li et al. ability of engineered plant lines with the capacity to produce optimal forms of glycosylated therapeutic proteins on the basis of modified N-glycosylation pathway. In recent years, extensive research efforts have been undertaken to modulate plant-specific N-glycosylation machinery, resulting in recombinant proteins with mammalian-like modifications. The enzymes responsible for the attachment of core a(1,3)-fucose and b(1,2)-xylose residues in plants are a(1,3)fucosyltransferase (FucT) and b(1,2)-xylosyltransferase (XylT) (Strasser et al., 2004) (Figure 1a). Interruption of FucT and XylT gene functions has been achieved via RNA interference (RNAi) technology and random mutagenesis methods in a number of plant species including, Arabidopsis, Lemna minor, Nicotiana benthamiana, alfalfa and rice (Castilho and Steinkellner, 2012).
These resulting mutant lines exhibit the desired phenotypes, namely significantly reduced levels of plant-specific a(1,3)-fucose and b(1,2)-xylose residues in recombinant proteins (Cox et al., 2006; Strasser et al., 2008; Yin et al., 2011). Unfortunately, these approaches suffer from several major drawbacks. For example, RNAi technology is often unable to completely deactivate the functions of the target genes, especially when more than one gene or gene family is targeted. Although traditional mutagenesis approaches, such as T-DNA, EMS, gamma rays and fast neutrons, can produce full knockout mutations, these mutations are randomly introduced into the genome with no control over location and the types of mutations created (Ostergaard and Yanofsky, 2004; Wang et al., 2012). This lack of specificity requires large-scale screens of mutagenized populations to
(a)
(b)
(c)
(d)
Figure 1 Targeting mutations in the FucT1/FucT2 and XylT1/XylT2 genes. (a) The FucT and XylT genes are responsible for the attachment of core a(1,3)fucose and b(1,2)-xylose residues to N-glycans in plants. (b) Schematics of the FucT1/FucT2 and XylT1/XylT2 genes. The FucT1/FucT2 genes contain 7 exons and 6 introns, and the XylT1/XylT2 genes contain 3 exons and 2 introns, represented by the open boxes and angled lines, respectively. TALEN target sites are indicated by arrowheads in the first exon of each gene. (c) DNA sequences of TALEN target sites in the FucT1/FucT2 and (d) the XylT1/XylT2 genes. Each TALEN target site consists of two TALEN recognition sequences, denoted in upper-case letters; the spacer sequences are shown in lower case. DNA sequence polymorphisms between FucT1 and FucT2 are indicated. ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
Glyco-engineering through genome editing 535 identify candidate plants with knockout mutations in the genes of interest. Moreover, the recovered plants often need to go through a multigeneration backcrossing regime to remove unwanted mutations randomly introduced at other loci. The shortcomings of the RNAi knock-down and random mutagenesis knockout approaches make it desirable to develop efficient methods to generate specific and targeted mutations in the genome of plants chosen for biopharmaceutical production. Recent advances in plant genome engineering have allowed for the generation of targeted mutations in a highly specific and efficient manner. Current genome engineering technology is based on the ability to engineer sequence-specific nucleases, such as meganucleases, zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs) or CRISPR/Cas nucleases (Puchta and Fauser, 2014; Voytas, 2013). These nucleases are engineered to recognize a genomic target sequence and subsequently cleave the DNA, resulting in double-stranded breaks (DSBs). The DSBs can then be repaired by either the nonhomologous end joining (NHEJ) or homologous recombination (HR) pathways, resulting in knockout mutations, precise site-directed point mutations or targeted gene insertions. In recent years, the efficiency of genome engineering technology has been dramatically improved, particularly with the development of TALEN and CRISPR/Cas nucleases. These two newest members in the designer nuclease family are easier to engineer for a desired target sequence and often have high cleavage activity. These benefits have been demonstrated in a wide range of organism at both single and multiple loci making them the reagents of choice for genome modification in many species (Baltes and Voytas, 2015). In this study, TALEN-mediated gene editing technology was employed to manipulate N-glycosylation pathways in N. benthamiana for the production of biopharmaceuticals with mammalian-like N-glycan structures. The highly efficient TALENs designed in this study enabled multiplexed gene knockouts in four genes, two XylT and two FucT genes, creating mutants with the desired glycosylation profiles in one generation. The resulting mutants were used to produce rituximab antibodies with an absence or reduction in plant-specific N-glycans. The high efficacy of TALEN technology for multiplexed gene editing of plants, as demonstrated in this study, will open up opportunities to accelerate research and development in plant metabolic engineering and synthetic biology.
Results TALEN design and activity assessment at the FucT and XylT target sites In the genome of N. benthamiana, five FucT and two XylT genes have been identified and characterized (Weterings and Van Eldik, 2010, 2013). Because of the high copy number of the FucT gene family, phylogenetic analysis of the FucT genes was first conducted to understand their relationship. The result indicated the formation of two clusters, one containing FucT1 and FucT2 and the other containing FucT3, FucT4 and FucT5 (Figure S1). In previous research, relative expression and activity of the FucT genes have been characterized (Weterings and Van Eldik, 2013). As a result, FucT1 and FucT2 showed the highest expression of all five genes. Thus, knockout of both FucT1 and FucT2 may result in the largest reduction in a(1,3)-fucose residues. Furthermore, the similarity in sequence between FucT1 and FucT2 allowed us to design TALEN pairs that target both genes simultaneously.
Sequence-specific TALENs were designed to knock out two FucT genes (FucT1 and FucT2) and two XylT genes (XylT1 and XylT2) in Nicotiana benthamiana. Conserved sequences in the first exon of the FucT1 and FucT2 genes were analysed by TALENTM hit software, and four potential target sites for each gene were identified (Figure 1b). Between FucT1 and FucT2, single nucleotide polymorphisms are present in the TALEN recognition sequences of target sites 1 and 2 and the spacer sequences of target sites 3 and 4 (Figure 1c). Eleven TALEN monomers were designed to recognize all the FucT1 and FucT2 target sites and were constructed using methods previously described (Beurdeley et al., 2013). Similarly, four TALEN target sites were identified in the first exon of the XylT1 and XylT2 genes (Figure 1b, d). As each target site is conserved in both XylT genes, only four TALEN pairs were required to target all sites in both XylT genes. To assess the activity of the TALENs targeting the FucT and XylT genes, a yeast-based assay was first performed using a method described previously (Haun et al., 2014; Christian et al., 2012). As indicated in Table S1, five of six FucT TALENs and three of four XylT TALENs showed high cleavage activities at both 37°C and 30°C, as compared to our positive control, I-SceI. TALENs with the highest activity were chosen for further testing at endogenous target sites in protoplasts isolated from N. benthamiana leaf tissue. TALEN expression cassettes were transformed into protoplasts and then the samples were sacrificed for preparation of genomic DNA after a 48-h incubation period. The genomic DNA was subjected to PCR to produce amplicons of approximately 300 bp that encompass the TALEN target sites. The amplicons were then subjected to 454 pyro-sequencing, and the resulting sequencing reads were evaluated for NHEJ-induced mutations. The frequency of mutated sequences in each sample was used to estimate the activity of the corresponding TALENs. The two FucT TALEN pairs, FucT1_T02 and FucT2_T02, displayed high frequencies of NHEJ-induced mutagenesis, 33.3% and 39.1%, respectively, at their corresponding target sites (Table 1). As the FucT1_T02 and FucT2_T02 target sites differ by only 1 bp (A/G) in the left TALEN recognition site, activity of each TALEN pair was examined at both FucT genes. Interestingly, the results indicate that both FucT1_T02 and FucT2_T02 recognize and cleave both genes with similar efficiencies; therefore, we predicted that it should be possible to use a single TALEN pair to disrupt both genes simultaneously. The XylT TALENs, XylT_03 and XylT_04, were also highly mutagenic, and 22.8% and 61.9% of the respective sequencing reads had mutations at the target sites (Table 1). Examples of TALENinduced mutations in both FucT and XylT genes are shown in Figure S3.
Generation of FucT and XylT mutants via multiplexed, targeted gene editing The two TALENs, FucT2_T02 and XylT_T04, which showed the highest activity in protoplasts, were chosen to create lines with mutations in the FucT and XylT genes. Leaf protoplasts were transformed with either the FucT2_T02 TALEN only (group 1), the XylT_T04 TALEN only (group 2), or both the FucT2_T02 and XylT_T04 TALENs (group 3). The transformed protoplasts were then regenerated to individual plantlets on nonselective media. Leaf tissues from regenerated plantlets were genotyped by PCR and DNA sequencing. From transformation group 1, 160 regenerated plantlets were screened, revealing 80 plantlets (50%) with mutations in at least
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
536 Jin Li et al. Table 1 TALEN activity assessment in N. benthamiana protoplasts
TALEN name FucT1_T02 FucT2_T02 XylT_T03 XylT_T04
Gene targets
NHEJ-induced
Protoplast
mutagenesis
transformation
frequency*
efficiency†
FucT1
33.3% (27 470)
87%
FucT2
21.4% (7259)
87%
FucT1
39.7% (12 295)
87%
FucT2
39.1% (6457)
87%
XylT1
22.8% (3169)
81%
XylT2
27.5% (420)
81%
XylT1
61.9% (1615)
84%
XylT2
53.3% (464)
84%
*Mutagenesis frequency is the percentage of 454 sequencing reads with NHEJinduced mutations. The total number of the sequencing reads for each sample is indicated in parentheses. The NHEJ mutagenesis frequency was zero for the negative control, which was only transformed with the YFP construct. †
Protoplast efficiency is determined by dividing the number of YFP-positive
protoplasts by the number of total protoplasts (plasmid encoding YFP was cotransformed with plasmids encoding TALENs).
one allele of either the FucT1 or FucT2 genes and 42 plantlets (26%) with mutations in both FucT1 and FucT2 genes (Table 2). Two plantlets, designated NB13-105a and NB12-213a, had mutations in all alleles of both FucT1 and FucT2. In the plant line NB13-105a, a 13-bp deletion was detected in both alleles of the FucT1 gene and a 14-bp deletion was found in both alleles of the FucT2 gene (Figure 2a). No wild-type sequences were identified for either FucT1 or FucT2, suggesting this plantlet had knockout mutations in all alleles of both FucT genes. In the plant line NB13-213a, 20-bp and 12-bp deletions occurred in the FucT1 alleles, and 13-bp and 2-bp deletions occurred in the FucT2 alleles (Figure 2a). No wild-type sequence was recovered for either gene; therefore, NB12-213a likely contains distinct mutations in all alleles of the FucT1 and FucT2 genes. In transformation group 2, 129 regenerated plantlets were screened with 94 plantlets (73%) having mutations in at least one allele of either XylT1 or XylT2, and 53 plantlets (41%) had mutations in both genes. Two plantlets, NB15-11d and NB12-113c, had knockout mutations in all alleles of both XylT genes. Plant line NB15-11d had two distinct 7-bp deletions in the XylT1 gene and an identical 8-bp deletion in both alleles of XylT2
(Figure 2a). Plant line NB12-113c had identical 7-bp deletions in XylT1 as well as 35-bp and 5-bp deletions in XylT2 (Figure 2a). No wild-type sequences for the target sites of either XylT1 or XylT2 were recovered from these two plant lines. With transformation group 3, the goal was to perform multiplexed, targeted gene editing and create mutant lines with knockout mutations in all alleles of all four XylT and FucT genes. We cotransformed protoplasts with four plasmids encoding the FucT2_T02 and XylT_T04 TALEN pairs and regenerated a total of 105 plantlets. Of these plantlets, 54 (51%) had mutations in at least one allele of either the XylT or FucT genes, and 18 (17%) had mutations in all four genes. Among the 18 mutants, three had bi-allelic mutations in all four genes. One of these mutants, NB14-29aT0, had identical 44-bp deletions in FucT1, a 2-bp and a 40-bp deletion in FucT2, identical 5-bp deletions in XylT1 and a 6-bp and 549-bp deletion in XylT2 (Figure 2b). The completely mutant plant, NB14-29aT0, derived from transformation group 3 was allowed to set seed in order to assess transmission of the mutations to the next generation. T1 seeds were collected, and 60 T1 plants were genotyped to detect mutations in the XylT and FucT genes. The genotyping results showed that all the mutations present in the T0 plants were transmitted to the T1 generation (data not shown). No wild-type alleles were found in any of the genotyped T1 plants, further confirming that all the mutant plants had all of the targeted genes completely knocked out.
Reduction of a(1,3)-fucose and b(1,2)-xylose levels in endogenous proteins from N. benthamiana mutants Plants with confirmed mutant genotypes were subjected to N-glycosylation profile analysis in which endogenous proteins were assayed for levels of a(1,3)-fucose and b(1,2)-xylose via methods previously described (North et al., 2009; Strasser et al., 2008). Both FucT and XylT mutant plants had significant changes in their N-glycan profiles. FucT mutant plants NB13-105a and NB13-213a had overall a(1,3)-fucose levels reduced by 61% (Figure 3a). XylT mutant plants, NB15-11d and NB12-113c, had b (1,2)-xylose levels that were 23 (14%§
2
XylT1/XylT2
129
94 (73%)
53 (41%)
>22 (17%)§
3
FucT1/FucT2
105
54 (51%)
18 (17%)
3 (3%)
XylT1/XylT2 *The number of plants that contain mutations in at least one targeted gene. †
The number of plants that contain mutations in all targeted genes. Mutations could occur in one or both alleles of a given target.
‡
The number of plants that contain mutations in every allele of all targeted genes. Mutations occur in both alleles of a given target.
§
Plants containing in-frame mutations were not included in this number, even if the plant contained a mutation in every allele of all the targets.
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
Glyco-engineering through genome editing 537 (a)
(b)
Figure 2 Genetically transmissible mutations induced by TALENs in N. benthamiana plants. (a) TALEN-induced mutations in FucT1/FucT2 or XylT1/XylT2 knockouts. The plants NB13-105a and NB13-213a contain mutations in both alleles of FucT1 and FucT2; the plants NB15-11d and NB12-113c have mutations in both alleles of XylT1 and XylT2. TALEN recognition sequences are indicated in upper case; the spacer sequences are in lower case. Deletions are indicated by dashes, and the size of each deletion in base pairs is shown on the right. (b) TALEN-induced mutations in the plant NB14-29a that has mutations in all alleles of FucT1, FucT2, XylT1 and XylT2.
however, b(1,2)-xylose glycosylation was only reduced by 36%, as compared to the XylT-only mutant lines that had b(1,2)-xylose glycosylation 2 bp in the spacer
region were considered to be products of imprecise repair by NHEJ of a cleaved TALEN recognition site. Mutagenesis frequency is reported as the percentage of sequencing reads with NHEJ-induced mutations.
Plant regeneration from N. benthamiana protoplasts Immediately after transformation, protoplasts were suspended in liquid K3G1 medium at a cell density of 1 9 105/mL in a small petri dish and stored at 25°C in the dark. Four days after transformation, when the majority of the protoplasts had begun the first cell divisions, the culture was diluted fourfold with Media C (van den Elzen et al., 1985). At day 7 and day 10, the protoplast cultures were diluted twofold with MS media. At day 18, calli were observed under the light microscope, 1 month after transformation calli were visible to the eye. The individual calli were transferred to shoot-inducing MS medium (van den Elzen et al., 1985). After shoots emerged, the plantlets were cut at the base and transferred to root-inducing medium (van den Elzen et al., 1985). When roots formed, the plantlets were transferred to soil and grown to maturity.
Genotyping FucT and XylT mutant lines Genomic DNA isolated from regenerated plantlets was PCRamplified using gene-specific primer pairs for the FucT1, FucT2, XylT1 and XylT2 loci (primers are listed in Table S2). The PCR products were subjected to Sanger sequencing, and the sequencing trace files were analysed to identify candidate mutants. If a trace file had double peaks around the TALEN target sites, the plant was considered a candidate mutant for further analysis. Samples with clean traces were aligned with the wild-type target sequences to determine whether the same mutations were present at both alleles. Sequences of the mutant candidates were further analysed by cloning the PCR products. For each sample, 16–24 colonies were PCR-amplified and sequenced, and sequences from individual clones were aligned with wild-type target sequences to identify mutations. All primers used for genotyping are summarized in Table S2.
Glycosylation profiles of endogenous N. benthamiana leaf proteins One-month-old leaves from the N. benthamiana lines were harvested for protein isolation. Two grams of fresh leaves was ground into fine powder in liquid nitrogen and transferred to a 15-mL tube. Three volumes of extraction buffer (50 mM Tris, 150 mM NaCl, pH 8) were added along with 60 lL PMSF (100 mM in 100% isopropanol). The samples were then vortexed vigorously for 3 min and centrifuged at 5000 g for 15 min at 4°C. Supernatants were collected and mixed gradually with ammonium sulphate to a concentration of 10.6 g/ 100 mL to increase solubility of proteins via salting-in. The protein extracts were then centrifuged at 18 000 g for 20 min at 4°C. Supernatants were transferred to a fresh tube and mixed gradually with ammonium sulphate to a concentration of 38.7 g/100 mL to precipitate proteins via salting-out. Protein pellets were obtained by centrifuging at 18 000 g for 20 min at 4°C. Salts were removed from protein extracts using G2 dialysis cassettes (Thermo Fisher Scientific, Waltham, MA). N-glycans were released from the protein extracts by digestion with peptide-N-glycosidase A and F. Released N-glycans were purified using Ultra Clean SPE Carbograph columns (Thermo Fisher Scientific, Waltham, MA) and methylated in vitro. Permethylated N-glycans were analysed by ProteoDynamics Inc. (Clermont-
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
Glyco-engineering through genome editing 541 Ferrand, France) via MALDI TOF MS (Karas and Hillenkamp, 1988).
Rituximab antibody production in N. benthamiana Thirty-eight-day-old plants were transformed using Agrobacterium with T-DNA encoding the rituximab antibody light and heavy chains. The transient co-expression approach was similar to that described previously (Vezina et al., 2009) except that the rituximab light and heavy chain coding regions (GenBank Accession nos. AX709550 and AX709548, respectively) were placed under the control of a proprietary CPMV (cow pea mosaic virus)-derived expression system and the TBSV P19 was used as suppressor of silencing (Voinnet et al., 2003). Nine days post-agroinfiltration, the plant leaves were harvested and the rituximab antibody was purified. Total soluble proteins were extracted with 2 volumes of extraction buffer (50 mM Tris, pH 8.0, 150 mM NaCl) and the extract was centrifuged for 2 min at 20 000 g; the supernatant was filtered through Miracloth to remove insoluble material. The clarified protein extract was loaded onto a protein A affinity chromatography column (MabSure Select LX; GE Healthcare Life Sciences, Baie d’Urfe, QC, Canada). The column was washed with extraction buffer, and the antibody was eluted using 100 mM glycine pH 3.0. Elution fractions were immediately neutralized with 2M Tris pH 7.4 buffer, and the collected fractions were analysed by Coomassie-stained SDS-PAGE to assess antibody quantity and purity. Selected fractions were pooled and analysed for protein content.
LC-ESI MS/MS analysis of N-glycosylation in the rituximab antibody N-glycosylation of the rituximab antibody was analysed by LC-ESI MS/MS using a NanoAcquity UPLC coupled to a QTOF micro (Waters, Milford, MA). Reduction (DTT) and alkylation (iodoacetamide) of cysteine residues were performed prior to overnight digestion with trypsin (Promega, Madison, WI, Sequencing Grade) at 37°C. Half of the sample was subjected to deglycosylation with peptide-N(4)-(acetyl-b-glucosaminyl) asparagine amidase A (PNGase A; Roche, Ontario, Canada) at 37°C. Tryptic peptides and glycopeptides were chromatographically separated on a ProteoPep III C18 Integrafrit column (New Objective, Woburn, MA) using a water–acetonitrile gradient and analysed by MS/MS, using data-dependant acquisition. Peptides are identified using the Mascot search engine (Matrix Science, Boston, MA) and UniProt databases. Samples subjected to deglycosylation are used to identify the site of glycosylation, as PNGase A induces a substitution of Asn by Asp, which is detected by Mascot. The site of glycosylation in rituximab at position 301 was covered with two peptides: EEQYnSTYR and TKPREEQYnSTYR (n = site of N-glycosylation). Starting from these two identified peptides and the list of N-glycans typically found in plants, a list of putative glycopeptides, with their theoretical m/z, was created. The detection of the m/z in the LC-MS/MS analysis of samples without deglycosylation indicates which N-glycans are attached at site 301. Glycopeptide assignments were also confirmed by the presence of reporter ions in the MS/MS spectra: m/z 204, 366 and 512 indicating N-acetylglucosamine (GlcNAc), GlcNAc-hexose and GlcNAc-hexose-fucose, respectively.
Acknowledgements The authors would like to thank Dr. Nicholas Baltes for critical reading.
Conflict of interest JL, TJS, ZLD, SL, BMC, EER, AC, AD, AY, AR, LM, DFV and FZ are employees of Cellectis Plant Sciences Inc., a subsidiary of Cellectis SA. FC is an employee of Cellectis SA. POL and MAD are employees of Medicago R&D, a subsidiary of Medicago Inc.
References Arnould, S., Chames, P., Perez, C., Lacroix, E., Duclert, A., Epinat, J.C., Stricher, F., Petit, A.S., Patin, A., Guillier, S., Rolland, S., Prieto, J., Blanco, F.J., Bravo, J., Montoya, G., Serrano, L., Duchateau, P. and Paques, F. (2006) Engineering of large numbers of highly specific homing endonucleases that induce recombination on novel DNA targets. J. Mol. Biol. 355, 443–458. Baltes, N.J. and Voytas, D.F. (2015) Enabling plant synthetic biology through genome engineering. Trends Biotechnol. 33, 120–131. Beurdeley, M., Bietz, F., Li, J., Thomas, S., Stoddard, T., Juillerat, A., Zhang, F., Voytas, D.F., Duchateau, P. and Silva, G.H. (2013) Compact designer TALENs for efficient genome engineering. Nat. Commun. 4, 1762. Bombarely, A., Rosli, H.G., Vrebalov, J., Moffett, P., Mueller, L.A. and Martin, G.B. (2012) A draft genome sequence of Nicotiana benthamiana to enhance molecular plant-microbe biology research. Mol. Plant-Microbe Interact. 25, 1523–1530. Cabanes-Macheteau, M., Fitchette-Laine, A.C., Loutelier-Bourhis, C., Lange, C., Vine, N.D., Ma, J.K., Lerouge, P. and Faye, L. (1999) N-Glycosylation of a mouse IgG expressed in transgenic tobacco plants. Glycobiology, 9, 365–372. Castilho, A. and Steinkellner, H. (2012) Glyco-engineering in plants to produce human-like N-glycan structures. Biotechnol. J. 7, 1088–1098. Christian, M.L., Demorest, Z.L., Starker, C.G., Osborn, M.J., Nyquist, M.D., Zhang, Y., Carlson, D.F., Bradley, P., Bogdanove, A.J. and Voytas, D.F. (2012) Targeting G with TAL effectors: a comparison of activities of TALENs constructed with NN and NK repeat variable di-residues. PLoS ONE, 7, e45383. Cox, K.M., Sterling, J.D., Regan, J.T., Gasdaska, J.R., Frantz, K.K., Peele, C.G., Black, A., Passmore, D., Moldovan-Loomis, C., Srinivasan, M., Cuison, S., Cardarelli, P.M. and Dickey, L.F. (2006) Glycan optimization of a human monoclonal antibody in the aquatic plant Lemna minor. Nat. Biotechnol. 24, 1591–1597. Davis, A.P. and Symington, L.S. (2004) RAD51-dependent break-induced replication in yeast. Mol. Cell. Biol. 24, 2344–2351. Gomord, V. and Faye, L. (2004) Posttranslational modification of therapeutic proteins in plants. Curr. Opin. Plant Biol. 7, 171–181. Gomord, V., Fitchette, A.C., Menu-Bouaouiche, L., Saint-Jore-Dupas, C., Plasson, C., Michaud, D. and Faye, L. (2010) Plant-specific glycosylation patterns in the context of therapeutic protein production. Plant Biotechnol. J. 8, 564–587. Grabowski, G.A., Golembo, M. and Shaaltiel, Y. (2014) Taliglucerase alfa: an enzyme replacement therapy using plant cell expression technology. Mol. Genet. Metab. 112, 1–8. Haun, W., Coffman, A., Clasen, B.M., Demorest, Z.L., Lowy, A., Ray, E., Retterath, A., Stoddard, T., Juillerat, A., Cedrone, F., Mathis, L., Voytas, D.F. and Zhang, F. (2014) Improved soybean oil quality by targeted mutagenesis of the fatty acid desaturase 2 gene family. Plant Biotechnol J. 12, 934–940. Hiatt, A., Cafferkey, R. and Bowdish, K. (1989) Production of antibodies in transgenic plants. Nature, 342, 76–78. Karas, M. and Hillenkamp, F. (1988) Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal. Chem. 60, 2299–2301. Kraus, E., Leung, W.Y. and Haber, J.E. (2001) Break-induced replication: a review and an example in budding yeast. Proc. Natl Acad. Sci. USA, 98, 8255–8262. Loos, A. and Steinkellner, H. (2014) Plant glyco-biotechnology on the way to synthetic biology. Front. Plant Sci. 5, 523. Mashiko, D., Fujihara, Y., Satouh, Y., Miyata, H., Isotani, A. and Ikawa, M. (2013) Generation of mutant mice by pronuclear injection of circular plasmid expressing Cas9 and single guided RNA. Sci. Rep. 3, 3355. Mett, V., Farrance, C.E., Green, B.J. and Yusibov, V. (2008) Plants as biofactories. Biologicals, 36, 354–358.
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
542 Jin Li et al. Moscou, M.J. and Bogdanove, A.J. (2009) A simple cipher governs DNA recognition by TAL effectors. Science, 326, 1501. North, S.J., Hitchen, P.G., Haslam, S.M. and Dell, A. (2009) Mass spectrometry in the analysis of N-linked and O-linked glycans. Curr Opin Struct Biol 19, 498–506. Ostergaard, L. and Yanofsky, M.F. (2004) Establishing gene function by mutagenesis in Arabidopsis thaliana. Plant J. 39, 682–696. Puchta, H. and Fauser, F. (2013) Gene targeting in plants: 25 years later. Int. J. Dev. Biol. 57, 629–637. Puchta, H. and Fauser, F. (2014) Synthetic nucleases for genome engineering in plants: prospects for a bright future. Plant J., 78, 727–741. Schuster, M., Jost, W., Mudde, G.C., Wiederkum, S., Schwager, C., Janzek, E., Altmann, F., Stadlmann, J., Stemmer, C. and Gorr, G. (2007) In vivo glycoengineered antibody with improved lytic potential produced by an innovative non-mammalian expression system. Biotechnol. J. 2, 700–708. Stoger, E., Fischer, R., Moloney, M. and Ma, J.K. (2014) Plant molecular pharming for the treatment of chronic and infectious diseases. Annu. Rev. Plant Biol. 65, 743–768. Stoger, E., Sack, M., Fischer, R. and Christou, P. (2002) Plantibodies: applications, advantages and bottlenecks. Curr. Opin. Biotechnol. 13, 161–166. Strasser, R., Altmann, F., Mach, L., Glossl, J. and Steinkellner, H. (2004) Generation of Arabidopsis thaliana plants with complex N-glycans lacking beta1,2-linked xylose and core alpha1,3-linked fucose. FEBS Lett. 561, 132–136. Strasser, R., Stadlmann, J., Schahs, M., Stiegler, G., Quendler, H., Mach, L., Glossl, J., Weterings, K., Pabst, M. and Steinkellner, H. (2008) Generation of glyco-engineered Nicotiana benthamiana for the production of monoclonal antibodies with a homogeneous human-like N-glycan structure. Plant Biotechnol. J. 6, 392–402. Twyman, R.M., Stoger, E., Schillberg, S., Christou, P. and Fischer, R. (2003) Molecular farming in plants: host systems and expression technology. Trends Biotechnol. 21, 570–578. van den Elzen, P.J.M., Townsend, J., Lee, K.Y. and Bedbrook, J.R. (1985) A chimaeric hygromycin resistance gene as a selectable marker in plant cells. Plant Mol. Biol. 5, 299–302. V ezina, L.P., Faye, L., Lerouge, P., D’Aoust, M.A., Marquet-Blouin, E., Burel, C., Lavoie, P.O., Bardor, M. and Gomord, V. (2009) Transient co-expression for fast and high-yield production of antibodies with human-like N-glycans in plants. Plant Biotechnol. J. 7, 442–455. Voinnet, O., Rivas, S., Mestre, P. and Baulcombe, D. (2003) An enhanced transient expression system in plants based on suppression of gene silencing by the p19 protein of tomato bushy stunt virus. Plant J. 33, 949–956. Voytas, D.F. (2013) Plant genome engineering with sequence-specific nucleases. Annu. Rev. Plant Biol. 64, 327–350. Wang, H., Yang, H., Shivalila, C.S., Dawlaty, M.M., Cheng, A.W., Zhang, F. and Jaenisch, R. (2013) One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering. Cell, 153, 910–918.
Wang, T.L., Uauy, C., Robson, F. and Till, B. (2012) TILLING in extremis. Plant Biotechnol. J. 10, 761–772. Ward, B.J., Landry, N., Trepanier, S., Mercier, G., Dargis, M., Couture, M., D’Aoust, M.A. and Vezina, L.P. (2014) Human antibody response to N-glycans present on plant-made influenza virus-like particle (VLP) vaccines. Vaccine, 32, 6098–6106. Weterings, K. and Van Eldik, G. (2013) Nicotiana benthamiana plants deficient in fucosyltransferase activity. International patent application WO 2013/ 050155. Weterings, K. and Van Eldik, G. (2010) Nicotiana benthamiana plants deficient in xylosyltransferase activity. International patent application WO 2010/ 145846. Wright, D.A., Townsend, J.A., Winfrey, R.J. Jr, Irwin, P.A., Rajagopal, J., Lonosky, P.M., Hall, B.D., Jondle, M.D. and Voytas, D.F. (2005) Highfrequency homologous recombination in plants mediated by zinc-finger nucleases. Plant J. 44, 693–705. Yin, B.J., Gao, T., Zheng, N.Y., Li, Y., Tang, S.Y., Liang, L.M. and Xie, Q. (2011) Generation of glyco-engineered BY2 cell lines with decreased expression of plant-specific glycoepitopes. Protein Cell, 2, 41–47. Yin, J., Li, G., Ren, X. and Herrler, G. (2007) Select what you need: a comparative evaluation of the advantages and limitations of frequently used expression systems for foreign genes. J. Biotechnol. 127, 335–347. Yoo, S.D., Cho, Y.H. and Sheen, J. (2007) Arabidopsis mesophyll protoplasts: a versatile cell system for transient gene expression analysis. Nat. Protoc. 2, 1565–1572. Yusibov, V.M. and Mamedov, T.G. (2010) Plants as an alternative system for expression of vaccine antigens. Proc. ANAS (Biol. Sci.) 65, 195–200. Zhang, Y., Zhang, F., Li, X., Baller, J.A., Qi, Y., Starker, C.G., Bogdanove, A.J. and Voytas, D.F. (2013) Transcription activator-like effector nucleases enable efficient plant genome engineering. Plant Physiol. 161, 20–27.
Supporting information Additional Supporting information may be found in the online version of this article: Figure S1 Phylogenetic analysis of the FucT genomic sequences. Figure S2 Illustration of a representative TALEN-expression plasmid. Figure S3 Examples of TALEN-induced mutations in the FucT and XylT gene target sequences. Table S1 TALEN activity assessment in a yeast single-strand annealing assay. Table S2 Primers used for 454 deep sequencing and genotyping analysis.
ª 2015 Society for Experimental Biology, Association of Applied Biologists and John Wiley & Sons Ltd, Plant Biotechnology Journal, 14, 533–542
