Plant Pathology and Microbiology Publications Plant Pathology and Microbiology

11-2019

Protein expression and gene editing in monocots using foxtail mosaic virus vectors

Yu Mei Iowa State University

Bliss Beernink Iowa State University

Evan E. Ellison University of Minnesota - Twin Cities

Eva Konečná University of Minnesota - Twin Cities

Anjanasree K. Neelakandan Iowa State University, [email protected]

See next page for additional authors

Follow this and additional works at: https://lib.dr.iastate.edu/plantpath_pubs

Part of the Genetics Commons, and the Plant Pathology Commons

The complete bibliographic information for this item can be found at https://lib.dr.iastate.edu/ plantpath_pubs/283. For information on how to cite this item, please visit http://lib.dr.iastate.edu/howtocite.html.

This Article is brought to you for free and open access by the Plant Pathology and Microbiology at Iowa State University Digital Repository. It has been accepted for inclusion in Plant Pathology and Microbiology Publications by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Protein expression and gene editing in monocots using foxtail mosaic virus vectors

Abstract Plant viruses can be engineered to carry sequences that direct silencing of target host genes, expression of heterologous proteins, or editing of host genes. A set of foxtail mosaic virus (FoMV) vectors was developed that can be used for transient gene expression and single guide RNA delivery for Cas9‐mediated gene editing in maize, Setaria viridis, and Nicotiana benthamiana. This was accomplished by duplicating the FoMV capsid protein subgenomic promoter, abolishing the unnecessary open reading frame 5A, and inserting a cloning site containing unique restriction endonuclease cleavage sites immediately after the duplicated promoter. The modified oMVF vectors transiently expressed green fluorescent protein (GFP) and bialaphos resistance (BAR) protein in leaves of systemically infected maize seedlings. GFP was detected in epidermal and mesophyll cells by epifluorescence microscopy, and expression was confirmed yb Western blot analyses. Plants infected with FoMV carrying the bar gene were temporarily protected from a glufosinate herbicide, and expression was confirmed using a apidr antibody‐based BAR strip test. Expression of these proteins was stabilized by nucleotide substitutions in the sequence of the duplicated promoter region. Single guide RNAs expressed from the duplicated promoter mediated edits in the N. benthamiana Phytoene desaturase gene, the S. viridis Carbonic anhydrase 2 gene, and the maize HKT1 gene encoding a potassium transporter. The efficiency of editing was enhanced in the presence of synergistic viruses and a viral silencing suppressor. This work expands the utility of FoMV for virus‐induced gene silencing (VIGS), virus‐mediated overexpression (VOX), and virus‐enabled gene editing (VEdGE) in monocots.

Keywords Cas9, CRISPR, gene editing, gene expression, guide RNA, virus

Disciplines Genetics | Plant Pathology | Plant Sciences

Comments This article is published as Mei, Yu, Bliss M. Beernink, Evan E. Ellison, Eva Konečná, Anjanasree K. Neelakandan, Daniel F. Voytas, and Steven A. Whitham. "Protein expression and gene editing in monocots using foxtail mosaic virus vectors." Plant direct 3 (2019): e00181. doi: 10.1002/pld3.181.

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 License.

Authors Yu Mei, Bliss Beernink, Evan E. Ellison, Eva Konečná, Anjanasree K. Neelakandan, Daniel F. Voytas, and Steven A. Whitham

This article is available at Iowa State University Digital Repository: https://lib.dr.iastate.edu/plantpath_pubs/283

Received: 8 October 2019 | Revised: 8 October 2019 | Accepted: 24 October 2019 DOI: 10.1002/pld3.181

ORIGINAL RESEARCH

Protein expression and gene editing in monocots using foxtail mosaic virus vectors

Yu Mei1 | Bliss M. Beernink1 | Evan E. Ellison2 | Eva Konečná2 | Anjanasree K. Neelakandan3 | Daniel F. Voytas2 | Steven A. Whitham1

1Department of Plant Pathology and Microbiology, Iowa State University, Ames, Abstract IA, USA Plant viruses can be engineered to carry sequences that direct silencing of target host 2 Department of Genetics, Cell Biology genes, expression of heterologous proteins, or editing of host genes. A set of fox- and Development, Center for Genome Engineering, Center for Precision Plant tail mosaic virus (FoMV) vectors was developed that can be used for transient gene Genomics, University of Minnesota, St. Paul, expression and single guide RNA delivery for Cas9-mediated gene editing in maize, MN, USA 3Department of Genetics, Development and Setaria viridis, and Nicotiana benthamiana. This was accomplished by duplicating the Cell Biology, Iowa State University, Ames, FoMV capsid protein subgenomic promoter, abolishing the unnecessary open read- IA, USA ing frame 5A, and inserting a cloning site containing unique restriction endonuclease Correspondence cleavage sites immediately after the duplicated promoter. The modified FoMV vec- Steven A. Whitham, Department of Plant Pathology and Microbiology, Iowa State tors transiently expressed green fluorescent protein (GFP) and bialaphos resistance University, 1344 Advanced Teaching and (BAR) protein in leaves of systemically infected maize seedlings. GFP was detected Research Building, 2213 Pammel Dr., Ames, IA 50011, USA. in epidermal and mesophyll cells by epifluorescence microscopy, and expression was Email: [email protected] confirmed by Western blot analyses. Plants infected with FoMV carrying the bar

Funding information gene were temporarily protected from a glufosinate herbicide, and expression was This work was supported by the Iowa confirmed using a rapid antibody-based BAR strip test. Expression of these proteins State University Plant Sciences Institute (SAW), the Iowa State University Crop was stabilized by nucleotide substitutions in the sequence of the duplicated pro- Bioengineering Center (SAW), USDA NIFA moter region. Single guide RNAs expressed from the duplicated promoter mediated Hatch project number 3808 (SAW), and Department of Energy grant DE-SC0018277 edits in the N. benthamiana Phytoene desaturase gene, the S. viridis Carbonic anhydrase (DFV). Iowa State University and University 2 gene, and the maize HKT1 gene encoding a potassium transporter. The efficiency of Minnesota are part of a team supporting DARPA's Insect Allies program. of editing was enhanced in the presence of synergistic viruses and a viral silencing suppressor. This work expands the utility of FoMV for virus-induced gene silenc- ing (VIGS), virus-mediated overexpression (VOX), and virus-enabled gene editing (VEdGE) in monocots.

KEYWORDS Cas9, CRISPR, gene editing, gene expression, guide RNA, virus

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2019 The Authors. Plant Direct published by American Society of Plant Biologists and the Society for Experimental Biology and John Wiley & Sons Ltd

 wileyonlinelibrary.com/journal/pld3 | 1 Plant Direct. 2019;3:1–16. 2 | MEI et al.

1 | INTRODUCTION (Draghici & Varrelmann, 2009). The overlapping ORFs 2, 3, and 4 are known as the triple gene block (TGB) with functions in virus move- Plant viruses provide surprisingly versatile technology platforms en- ment and suppression of host defense (Verchot-Lubicz, 2005). ORF5 abling the expression of a wide array of coding and non-coding se- encodes the coat protein (CP), which is indispensable for virus as- quences in plants (Pasin, Menzel, & Daros, 2019). Viruses engineered sembly and cell-to-cell movement (Cruz, Roberts, Prior, Chapman, & to carry heterologous open reading frames (ORFs) can express the Oparka, 1998). In addition to the five ORFs, the FoMV genome has a encoded proteins (VOX). Viruses that carry fragments of plant genes unique ORF5A that initiates 144 nucleotides upstream of the CP, but in sense and antisense orientation cause virus-induced gene silenc- it is not required for replication or for systemic infection of N. benth- ing (VIGS) of the targeted sequence, or microRNA inserts can initiate amiana or barley (Robertson et al., 2000). silencing with high specificity. Most recently, it has been demon- Recently, the FoMV-based viral vectors have been developed for strated that plant viruses or their derivatives can be used to deliver both VIGS and gene expression (Bouton et al., 2018; Liu et al., 2016; CRISPR-Cas reagents consisting of single guide RNAs (gRNAs), DNA Liu & Kearney, 2010; Mei et al., 2016). Two versions of FoMV-based repair templates, and site-specific nucleases such as Cas9 (Ali et VIGS vectors and their applications in maize and other monocots al., 2015; Ali, Eid, Ali, & Mahfouz, 2018; Butler, Baltes, Voytas, & have been reported. Our original FoMV-VIGS vector carried the Douches, 2016; Cody, Scholthof, & Mirkov, 2017; Dahan-Meir et al., insertion site for target gene fragments immediately after the stop 2018; Gao et al., 2019; Gil-Humanes et al., 2017; Jiang et al., 2019; codon of ORF5, which worked well for VIGS, but it cannot be used Mahas, Ali, Tashkandi, & Mahfouz, 2019; Wang et al., 2017). These for gene expression (Mei et al., 2016). The FoMV-VIGS vector re- capabilities show that plant viruses can be useful biotechnological ported by Liu et al. (2016) is designed to carry target sequences for tools for gene function studies in plants, and they can have practical silencing under the control of a duplicated FoMV CP subgenomic applications as well (Pasin et al., 2019; Zaidi & Mansoor, 2017). promoter. Inverted-repeats carried at this position were most effi- The development of viral vectors for monocot plants has been cient at inducing VIGS (Liu et al., 2016). Bouton et al. (2018) also emerging rapidly in recent years (Lee, Hammond-Kosack, & Kanyuka, duplicated the CP promoter and demonstrated that FoMV could be 2015). To date, at least eight different viruses have been developed used to transiently express marker and fungal effector proteins from into viral vectors for monocots, including barley stripe mosaic virus this position in N. benthamiana, wheat, and maize. Disarmed FoMV (BSMV) (Lee, Hammond-Kosack, & Kanyuka, 2012; Scofield, Huang, vectors for transient gene expression have also been developed by Brandt, & Gill, 2005), brome mosaic virus (BMV) (Ding, Schneider, substitution of the TGB or CP with genes of interest. In the presence Chaluvadi, Mian, & Nelson, 2006), cymbidium mosaic virus (CymMV) of a viral RNA silencing suppressor, this set of vectors achieves high (Hsieh et al., 2013), rice tungro bacilliform virus (RTBV) (Purkayastha, protein expression at the site of inoculation, but the recombinant Mathur, Verma, Sharma, & Dasgupta, 2010), wheat streak mosaic virus virus cannot spread systemically (Liu & Kearney, 2010). (WSMV) (Choi, Stenger, Morris, & French, 2000; Tatineni, McMechan, Here, we further investigated strategies to systemically express Hein, & French, 2011), bamboo mosaic virus (BaMV) together with its proteins in maize from FoMV using a duplicated CP strategy similar associated satellite RNA (Liou, Huang, Hu, Lin, & Hsu, 2014), cucum- to Bouton et al. (2018). Protein expression was demonstrated using ber mosaic virus (CMV) (Wang et al., 2016), barley yellow striate mo- the genes encoding bialaphos resistance (bar) and green fluorescent saic virus (BYSMV) (Gao et al., 2019), and foxtail mosaic virus (FoMV) protein (GFP), and time courses over plant development were used to (Bouton et al., 2018; Liu et al., 2016; Mei & Whitham, 2018; Mei, understand insert stability and limitations of the vector for protein ex- Zhang, Kernodle, Hill, & Whitham, 2016). Seven of the virus vectors are pression. We mapped subgenomic mRNA transcription start sites and designed for VIGS applications (BSMV, BMV, CymMV, RTBV, BaMV, modified duplicated CP promoter sequences to explore avenues for CMV, and FoMV), and four can be used for systemic gene expression better understanding FoMV gene expression and increasing stability of (BSMV, BYSMV, FoMV, and WSMV). BYSMV was also shown to deliver heterologous sequences. To demonstrate applications of FoMV in ed- both Cas9 and a single guide RNA to N. benthamiana where it induced iting plant genes, we expressed single gRNAs from the duplicated CP site-specific gene edits at the infiltrated site (Gao et al., 2019). promoter sequence in N. benthamiana, Setaria viridis, and maize plants Previous work by us and others demonstrated that FoMV is capa- carrying Cas9 transgenes. Systemic gene editing was observed in leaves ble of systemic infection and inducing VIGS in maize (Liu et al., 2016; of all three species, and it was enhanced in the presence of an RNA Mei & Whitham, 2018; Mei et al., 2016) or expressing proteins (Bouton silencing suppressor or a synergistic virus, demonstrating that FoMV et al., 2018). FoMV is a member of the genus Potexvirus, which has a can enable gene editing through the expression of functional gRNAs. single-stranded, positive-sense genomic RNA. The genome structure of FoMV is similar to potato virus X (PVX), which is the type mem- ber of the potexviruses. These viruses typically contain five major 2 | MATERIALS AND METHODS open reading frames (ORF) (Bruun-Rasmussen, Madsen, Johansen, & Albrechtsen, 2008; Huisman, Linthorst, Bol, & Cornelissen, 1988; 2.1 | Plant material and virus inoculation Robertson, French, & Morris, 2000). ORF1 encodes the RNA- dependent RNA polymerase (RdRp), which is necessary for viral The parent pFoMV-V infectious clone from which all FoMV clones RNA replication and subgenomic messenger RNA (sgRNA) synthesis were derived was previously described (Mei et al., 2016). The SCMV MEI et al. | 3 virus isolate was originally named MDMV-B and designated Iowa 5280–5333) and Bsu36I recognition site at 3’ end. The annealed 66-188 [ATCC-PV53]) (Ford, Bucholtz, & Lambe, 1967; Hill, Ford, & product was digested with Bsu36I and ligated into pFoMV-V-Δ5A- Benner, 1973). The viruses were propagated in sweet corn (Zea mays MSC that had been digested with PmlI and Bsu36I to produce the L. ‘Golden x Bantam’; American Meadows). Virus-infected leaf sap pFoMV-DP. The pFoMV-DC was constructed similarly using synthe- was prepared by grinding infected leaves in 50 mM potassium phos- sized oligonucleotides DCS and DCA. phate buffer, pH 7.0. Sweet corn plants at the two-leaf stage were mechanically inoculated by rubbing leaf sap on new leaves dusted with 600-mesh Carborundum. To inoculate plants with pFoMV infec- 2.3 | Generation of FoMV constructs for tious DNA clones, leaves of one-week-old seedlings were inoculated protein and RNA expression by particle bombardment using a Biolistic PDS-1000/He system (Bio- Rad Laboratories), 1.0-µm gold particles coated with 1 µg of pFoMV Bar was amplified by PCR using pBPMV-IA-GFP-BAR (Zhang, DNA, and 1,100-psi rupture disks at a distance of 6 cm as previously Bradshaw, Whitham, & Hill, 2010) as a template with primer pair described (Mei & Whitham, 2018). Plants were placed in the dark for BAR Bsu36I and BAR HpaI. The product was cloned into pGEM-T 12 hr before and after inoculation and then maintained in a green- Easy (Promega) and sequenced for verification. The bar gene was house room with a thermostat set to 20–22°C with a 16-hr photoper- released by Bsu36I and PspOMI double digestion and ligated into iod. N. benthamiana plants were grown in a growth room at 22°C with similarly digested pFoMV-DP or pFoMV-DC to generate the con- a 16-hr photoperiod, and the N. benthamiana Cas9 line was previously struct pFoMV-DP-BAR or pFoMV-DC-BAR. GFP was amplified by described (Baltes et al., 2015). The S. viridis Cas9 line was generated PCR using pSITE 2CA (Chakrabarty et al., 2007) as a template with by transforming strain A10 with a wheat codon-optimized Cas9 gene primer pairs GFP Bsu36I and GFP PspOMI. pGFP was generated by expressed from the maize ubiquitin promoter (ZmUbi) (Cermak et al., cloning the product into pGEM-T Easy, and it was verified by se- 2017; Van Eck, Swartwood, Pidgeon, & Maxson-Stein, 2017). The quencing. pGFP was digested with Bsu36I and PspOMI and ligated maize Cas9 line was described by Char et al. (2017). into similarly digested pFoMV-DC to generate pFoMV-DC-GFP.

2.2 | Construction of FoMV-derived vectors 2.4 | RT-PCR analysis

A three-step strategy was used to make pFoMV-V-derived expres- Non-infected wild-type leaves or leaves of plants infected by sion vectors. All oligonucleotide primers are listed in Table S1. In step pFoMV-V, pFoMV-V-Δ5A, pFoMV-DP, pFoMV-DC, pFoMV-DP-BAR, 1, pFoMV-V was modified to disrupt the predicted start codon of pFoMV-DC-BAR or pFoMV-DC-GFP were harvested for total RNA ORF5A. In PCR reaction A, primer pairs 5AmuS1 and 5AmuA1 were extraction using the RNeasy Plant Mini Kit (Qiagen). The leaves that used to amplify a product from pFoMV-V and the product was gel were sampled are indicated in the figure legends. After first-strand extracted. In PCR reaction B, primer pair 5AmuS2 and 5AmuA2 was cDNA synthesis, primer pair 5AmuS2 and 5AmuA2 was used to test used with pFoMV-V as template and the product was gel extracted. for FoMV infection and for the presence of insert. Z. mays actin was In overlapping PCR reaction C, primer pair 5AmuS1 and 5AmuA2 used as an internal control with primer pair ZmActS and ZmActR. was used with PCR products A and B as templates. PCR product C was digested with restriction enzymes SacII and SalI, gel extracted, and ligated into pFoMV-V that had been digested with the same en- 2.5 | Herbicide treatment and bar strip test zymes to produce the pFoMV-V-Δ5A. In step 2, a multiple cloning site was added into pFoMV-V-Δ5A. FoMV-DP/DC-BAR-infected sweet corn plants were sprayed with In PCR reaction D, primer pairs 5AmuS1 and 201DPA1 were used a dilution of Finale® herbicide that contained 0.05% glufosinate- to amplify a product from pFoMV-V-Δ5A and the product was gel ammonium w/v) (BASF) in deionized water. Maize plants were pho- extracted. In PCR reaction E, primer pairs 201DPS1 and 5AmuA2 tographed at 10 days after herbicide treatment. Leaf samples of were used with pFoMV-V-Δ5A as template and the product was non-inoculated, pFoMV-DC- or pFoMV-DC-BAR-infected plants gel extracted. In overlap PCR reaction F, primer pair 5AmuS1 and were used to test the presence of BAR protein using the QuickStix™ 5AmuA2 was used with PCR products D and E as templates. PCR Kit for LibertyLink® (bar) Cotton Leaf & Seed (EnviroLogix) according product F was digested with restriction enzymes SacII and SalI, gel to the manufacturer's instructions. extracted, and ligated into pFoMV-V that had been digested with the same enzymes to produce the pFoMV-V-Δ5A-MSC, which contains PmlI, Bsu36I, HpaI and PspOMI. 2.6 | Microscopy, protein extraction and western In step 3, the duplicated CP subgenomic promoters were added blot assays to pFoMV-V-Δ5A-MSC. Oligonucleotides DPS and DPA were syn- thesized and annealed to form double-strand DNA fragment that Leaf samples of non-inoculated, pFoMV-DC-, pFoMV-DC-GFP- contained the putative subgenomic promoter for CP (nucleotides infected plants were cut into small squares (1 cm × 0.5 cm) 4 | MEI et al.

pCAMBIA380-FoMV-infected plants. In each sample, 8 leaf disks were collected using a 0.5 cm diameter borer and homogenized in 100 µl of extraction buffer (50 mM NaCl, 20 mM Tris pH 7.5, 1 mM EDTA, 0.1% Triton X-100, 10% glycerol, 5 mM DTT, 2 mM NaF, 1 mM PMSF, and 1x protease inhibitor cocktail). After centrifugation, the supernatant was mixed with SDS-PAGE loading buffer and boiled for 5 min, and then, 10 µl of each boiled sample was used in SDS-PAGE followed by Western blotting using anti-GFP monoclonal antibody (Genscript).

2.7 | Insertion of FoMV into a T-DNA plasmid for agroinoculation

First, the XhoI restriction site in pFoMV-DC was replaced by PacI. To achieve this, a PCR reaction was performed using the primer pair FM-PacIFor and NosRev with pFoMV-DC as template. The PCR product was digested with restriction enzymes XbaI and ClaI, gel ex- tracted, and ligated into pFoMV-DC that also had been digested with XbaI and ClaI to produce the pFoMV-DC-PacI. Next, the cauliflower mosaic virus 35S promoter to the nopaline synthase terminator was amplified from the pFoMV-DC-PacI using primer pair DCPacI1380F and DCPacI1380R. The pCAMBIA1380 backbone was amplified with primer pair 1380F and 1380R. The two PCR products were purified and assembled using the Gibson Assembly® Cloning Kit according to the manufacturer's instructions (New England Biolabs). The fidelity of the resulting pCAMBIA380-FoMV-DC* (pFoMV-DC*) was further confirmed by sequencing. To generate constructs for gRNA delivery, oligonucleotides NbPDSg1 and NbPDSg2 (Table S1) were synthesized and annealed, and then ligated into Bsu36I- and PspOMI-digested FIGURE 1 Diagrams of the genomes of FoMV clones and the pFoMV-DC* to generate the construct pCAMBIA1380-FoMV-gN- nucleotide modifications to produce the FoMV-DP and FoMV-DC bPDS. The construct pCAMBIA1380-FoMV-gZmHKT1 was gener- expression vectors. (a) The FoMV-V silencing vector (pFoMV-V; Mei et al., 2016) is shown in the upper panel. The lower panel is annotated ated similarly using oligonucleotides ZmHKT1g1 and ZmHKT1g2 FoMV sequence beginning in ORF 4 and continuing to the ORF5 (Table S1). The construct pCAMBIA1380-FoMV-gSvCA2 was gen- start codon. ORF 4 coding sequence is highlighted in blue, ORF 5A is erated by PCR amplifying the sgRNA scaffold with oligonucleotides in green text, the predicted core of the ORF 5 promoter is in purple SvCA2g1 and sgRNA:7xT:PspOMI. This PCR product was purified text with a dashed underline, the transcription start site for the ORF and assembled into Bsu361- and PspOMI-digested pFoMV-DC* using 5 subgenomic mRNA transcript is indicated by the black arrow, the ® beginning of ORF 5 is highlighted in purple, and start or stop codons the Gibson Assembly Cloning Kit according to the manufacturer's for all ORFs are indicated by underlining. Multiple cloning site 1 (MCS1) instructions (New England Biolabs). contains the XbaI and XhoI restriction enzyme sites. (b) FoMV-V Δ5A in which a point mutation was introduced into the ORF 5A start codon was mutated to render it non-functional (ATG -> ACG) as indicated 2.8 | Agroinfiltration of N. benthamiana by the orange text. (c) FoMV-DP and the sequence showing the duplication of the ORF5 promoter followed by a second multiple cloning site (MCS2). The bold black text indicates the duplicated Nicotiana benthamiana plants between 5 and 6 weeks old were used promoter sequence, and the lowercase gray letters indicate the in this study. Agrobacterium tumefaciens strain GV3101 harboring the MCS2 with the Bsu36I, HpaI, and PspOMI sites, respectively, in bold. pFoMV-DC* plasmids was cultured and re-suspended in infiltration (d) Schematic representation of the FoMV-DC genome in which the buffer (200 μM acetosyringone, 10 mM MES, pH 5.6, and 10 mM duplicated ORF5 promoter was modified by five point mutations that MgCl ) (OD = 1.0) and kept at room temperature for 3 hr before reduced redundancy of the duplicated sequence. Single nucleotide 2 600 changes in the duplicated promoter region are in bold red text infiltration into plant leaves with a needleless syringe. Three weeks later, the infected systemic leaves were harvested and used as inocu- and examined with a Zeiss Axioplan II microscope with FITC lum for rub-inoculation onto new sweet corn seedlings at the two-leaf cube. Total protein was extracted from leaf samples of non- stage. Three-leaf stage S. viridis seedlings were rub-inoculated using inoculated, pFoMV-DC- and pFoMV-DC-GFP-infected and systemically infected N. benthamiana leaves 10 days after infiltration. MEI et al. | 5

3 | RESULTS

3.1 | Construction of FoMV expression vectors

We previously developed the pFoMV-V vector for VIGS in maize (Mei et al., 2016), but it cannot be used to express proteins, because the cloning site (MCS1) for plant gene fragments was placed after the stop codon of the CP gene (ORF5) (Figure 1a). To engineer FoMV vectors for protein expression, we designed a CP promoter duplication strat- egy similar to what has been used in PVX vectors (Dickmeis, Fischer, & Commandeur, 2014; Lacomme & Chapman, 2008; Sablowski, Baulcombe, & Bevan, 1995; Wang et al., 2014) and recently in another FoMV expression vector (Bouton et al., 2018). The CP promoter du- plication strategy was potentially complicated by ORF5A, which over- laps ORF4 and ORF5 (Figure 1a). ORF5A was previously shown to be expressed but dispensable for systemic infection in barley (Robertson et al., 2000). To confirm that ORF5A is not required for infection of FIGURE 2 Infection of sweet corn (Golden × Bantam) by the maize, we made pFoMV-V Δ5A in which the predicted start codon was FoMV vectors. (a) Images of leaves from control and inoculated changed from ATG to ACG without altering the amino acid sequence plants. From left to right: non-inoculated (NI), FoMV silencing vector of ORF4 (Figure 1b). The infectivity of FoMV-V Δ5A was tested by (FoMV-V), FoMV-V with mutated ORF5A (FoMV-V- Δ5A), FoMV expression vectors FoMV-DP (DP) and FoMV-DC (DC). Bar = 1 cm. both biolistic inoculation and subsequent rub-inoculation. Plants sys- (b) RT-PCR assay detected FoMV in the systemic leaves shown in temically infected by FoMV-V Δ5A and FoMV-V developed mosaic panel A. Maize actin was included as internal control for RT-PCR symptoms that were indistinguishable (Figure 2a). The stability of the mutation was tested in 14 biolistically inoculated plants and 13 plants For NbPDS gene-editing experiments, N. benthamiana transgenic that were rub-inoculated with sap from a biolistically inoculated plant. plants expressing the SpCas9 were used. The infiltrated and sys- The mutation was detected in all the infected samples, and no wild- temic leaves were sampled at 7, 14, and 21 DPI. Flowers and cap- type sequence was observed. These results demonstrated that the sules of infiltrated plants were also sampled to test the NbPDS gene Δ5A mutation did not revert to wild type and the inability to produce editing. In the case of FoMV and TuMV co-infection, suspensions the putative 5A protein did not impair the infectivity of FoMV in maize. of GV3101 harboring the pFoMV-DC*-gNbPDS and A. tumefaciens Subsequently, pFoMV-V Δ5A was further modified by introducing strain GV2260 harboring the pCB-TuMV-GFP (Lellis, Kasschau, two duplicated promoter variants and multiple cloning site 2 (MCS2)

Whitham, & Carrington, 2002) were mixed (OD600 = 1.0 for each containing the Bsu36I, HpaI and PspOMI endonuclease cleavage sites. strain, mixed at 1:1) and infiltrated. The infiltrated and systemic pFoMV-DP contains a 54-nucleotide duplication of the CP subgenomic leaves were sampled at 7 DPI. promoter (nucleotides 5280–5333) (Figure 1c), and pFoMV-DC was made by changing five codons in the sequence that overlaps with ORF4 without altering the encoded amino acid sequence (Figure 1d). 2.9 | Verification of mutations in N. benthamiana, Sweet corn seedlings infected with pFoMV-DP or pFoMV-DC S. viridis, and maize developed mosaic symptoms indistinguishable from the parental pFoMV-V or pFoMV-V Δ5A clones (Figure 2a). RT-PCR was used to Genomic DNA was extracted from N. benthamiana or maize samples amplify a 396-nucleotide fragment containing the duplicated pro- using the DNeasy Plant Mini Kit (Qiagen) according to the manufac- moter and MCS2. The RT-PCR product was detected in symptomatic turer's instructions. The DNA fragments encompassing the target sweet corn plants that were biolistically inoculated with pFoMV-V, sites were amplified using Q5 High-Fidelity DNA Polymerase (New pFoMV-V Δ5A, pFoMV-DP, or pFoMV-DC but not in non-inocu- England Biolabs) with primer pair NbPDSs and NbPDSa for NbPDS, lated control plants (Figure 2b). The slightly larger band detected in primer pair oEE374 and oEE375 for SvCA2, or primer pair ZmIDTF0 FoMV-DP- or FoMV-DC-infected leaves suggested that the dupli- and ZmIDTR0 for ZmHKT1 (Table S1). The NbPDS fragment was di- cated promoter remained intact during maize infection (Figure 2b). gested with NcoI-HF and then separated on a 1.0% agarose gel. The We also confirmed the authenticity of subgenomic RNAs expressed intensity of the DNA bands was quantified using ImageJ software to by using 5’ RACE to map the transcription start sites for the dupli- estimate editing efficiency. The undigested bands were individually cated CP promoter and the authentic CP promoter. This analysis ver- gel purified and cloned into pGEM-T Easy (Promega, Madison, WI, ified that the duplicated CP promoter produced subgenomic mRNA USA) for sequence analysis. The ZmHKT1 and SvCA2 amplicons were transcripts that initiated 14 nucleotides downstream of the core analyzed in a similar way except XcmI and PvuII, respectively, were promoter sequence at a GAA sequence as does the authentic CP used for the digestion. promoter (Figure 1). 6 | MEI et al.

the herbicide Finale® (BASF) (Whitham, Yamamoto, & Carrington, 1999), which has glufosinate-ammonium as the active compound. Within ten days after biolistic inoculation with pFoMV-DP-BAR and pFoMV-DC-BAR, 50%–90% of inoculated plants developed typical mosaic symptoms starting from the 3rd leaf (e.g., Figure 3a). To con- firm infection and test stability of the insertion, RNA was extracted from the 4th and 6th leaves of 10 FoMV-DP-BAR and 18 FoMV- DC-BAR plants for RT-PCR analysis using FoMV primers that flanked the cloning site. A 475-bp PCR product was expected in the plants infected by the empty vectors, and a 1,027-bp product in the plants infected by FoMV-DP-BAR or FoMV-DC-BAR. In the leaf 4 samples, the 1,027-bp PCR product was detected in 9 out of 10 FoMV-DP- BAR plants, and all of them also contained smaller deletion deriva- tives (Figure 3b). The 1,027-bp product was detected in 17 out of 18 FoMV-DC-BAR plants, and 6 of them also contained smaller deletion derivatives (Figure 3c). The deletion of BAR was more extensive in the 6th leaf samples in which all the FoMV-DP-BAR and FoMV-DC- BAR plants carried partial to total deletions (Figure 3d,e). Overall, the bar insert appeared to be less stable in FoMV-DP than in the FoMV-DC context, which was expected since the FoMV-DC vector was designed to reduce redundancy with the native CP promoter. Expression of the BAR protein was confirmed by an immunoassay in leaf four of two plants infected by FoMV-DC-BAR (Figure 3f). To test whether the expression of BAR could provide herbicide resistance, plants were sprayed with 0.05% glufosinate-ammonium twice at a 3-day interval and then scored for response at 10 days after the first application. The herbicide treatment killed the non-infected, FoMV-DP, and FoMV-DC plants, but plants infected by FoMV-DP- BAR and FoMV-DC-BAR were partially or totally protected (Figure 4a). The protective effect varied among plants, but the 4th leaf remained FIGURE 3 FoMV-mediated BAR expression in sweet corn green on most of the plants (Figure 4b,c). When the herbicide was (Golden × Bantam). (a) Sweet corn inoculated with pFoMV-DP-BAR applied at a later stage beginning at 23 dpi with pFoMV-DP-BAR and (DP-BAR, left) and pFoMV-DC-BAR (DC-BAR, right). (b) RT-PCR pFoMV-DC-BAR, all plants were killed (Figure S1), which is consistent analysis of bar insert stability in the 4th leaves of FoMV-DP-BAR- infected plants. (c) RT-PCR analysis of bar insert stability in the with the extensive deletion of the bar gene in leaf 6 (Figure 3d,e). 4th leaves of FoMV-DC-BAR-infected plants. (d) RT-PCR analysis To test whether the expression of BAR could be passaged, the of bar insert stability in the 6th leaves of FoMV-DP-BAR-infected sap from leaves of biolistically inoculated FoMV-DC-BAR plants plants. (e) RT-PCR analysis of bar insert stability in the 6th leaves of with confirmed BAR expression (Figure 3f) was used to rub-inoc- FoMV-DC-BAR-infected plants. The upper gel images in (b, c, d, and ulate the first two leaves of healthy sweet corn seedlings. Mosaic e) are the RT-PCR control showing amplification of a maize actin symptoms were observed in 25 of 30 plants approximately one week mRNA fragment in all samples. The lower gel images are RT-PCR amplification across the FoMV cloning site (MCS2). EV indicates after rub-inoculation. These plants and corresponding controls were the FoMV-DP or DC empty vector that carries no insert. (F) Strip treated with 0.05% glufosinate-ammonium at 13 DPI. At 10 days test for the expression of BAR protein. Positive signals indicated by after herbicide treatment, 16 FoMV-DC-BAR plants were protected the arrow are detected in FoMV-DC-BAR-infected leaf tissue, but with at least one leaf remaining green, and all leaves remained green not in leaves of non-infected (NI) or FoMV-DC empty vector (EV) on 8 of the plants (Figure S2). These data showed that functional control plants. The red stars in panel C indicate the plants infected with FoMV-DC-BAR that were used in the strip test BAR protein was expressed when leaf sap from pFoMV-DC-BAR-in- oculated plants was rub-inoculated onto new sweet corn plants, and the protective effect is similar in biolistic and rub-inoculated plants. 3.2 | Expression of the Bialaphos Resistance (BAR) protein from FoMV 3.3 | Expression of GFP from FoMV To test protein expression, the bar gene was inserted into pFoMV-DP and pFoMV-DC to produce pFoMV-DP-BAR and pFoMV-DC-BAR, The GFP sequence was inserted at the Bsu36I and PspOMI cloning respectively. Expression of BAR was expected to protect plants from sites in pFoMV-DC to produce pFoMV-DC-GFP. GFP expression MEI et al. | 7

FIGURE 4 Expression of BAR from FoMV transiently partially protects sweet corn plants from herbicide treatment. (a) Sweet corn plants before (upper panels) and after (lower panels) treatment with glufosinate herbicide. Plants were photographed at 18 dpi, just prior to herbicide treatment, and at 10 days after the first herbicide application. From left to right: non-inoculated (NI), plants infected with pFoMV-DC (DC), pFoMV-DP (DP), pFoMV-DC-BAR (DC-BAR), and pFoMV- DP-BAR (DP-BAR). (b) Individual DC-BAR plants after herbicide treatment. (c) A representative image of a DP-BAR plant after herbicide treatment. Red stars in b and c indicate the 4th leaves

was not investigated using pFoMV-DP, because bar was more stable Expression of GFP protein was analyzed by Western blot using in pFoMV-DC. Sweet corn seedlings were inoculated with pFoMV- an anti-GFP antibody. Consistent with the RT-PCR results, GFP was DC-GFP using biolistic bombardment, and mosaic symptoms devel- detected in all the leaf 4 samples and most of the leaf 6 samples but oped that were indistinguishable from pFoMV-DC-inoculated plants not in any of the leaf 9 samples (Figure 5c). In addition, bands from (Figure 5a). The presence and stability of the GFP insert was tested the leaf 4 samples were the most intense, indicating that GFP accu- by RT-PCR analysis using primers that flanked the cloning site. A mulated to the highest levels in leaf 4 relative to leaf 6. The leaves 475-bp PCR product was expected in plants infected by FoMV-DC, were examined using a fluorescence microscope to detect GFP flu- and an 1,192-bp PCR product was expected in plants infected by orescence. Green fluorescence was observed in epidermal and me- FoMV-DC-GFP. No deletion of GFP was detected in leaf 4 samples, sophyll cells of leaves infected by FoMV-DC-GFP but not in control because we only observed the 1,192-bp band, and we saw evi- leaves infected by FoMV-DC (Figure 5d). The green fluorescence dence for a deletion only in the leaf 6 sample of plant 5 (Figure 5b; occurred in a patchy manner, reflecting the symptoms of FoMV in- P5). However, the 1,192-bp band was not detected in any of the fection (Figure 5d). The GFP signals were relatively strong in leaf 4 leaf 9 samples, indicating that no intact FoMV-DC-GFP remained and became weaker as the virus moved to upper leaves and were not (Figure 5b). detectable in the 9th leaf (Figure S3). 8 | MEI et al.

FIGURE 5 Expression of GFP from FoMV in sweet corn. (a) Sweet corn inoculated with pFoMV-DC (DC, left) and by pFoMV-DC-GFP (DC-GFP, right). (b) RT-PCR analyses for the GFP insert stability in FoMV-DC-GFP-infected plants. The upper gel image is the RT-PCR control showing amplification of a maize actin mRNA in all samples. The lower gel image shows RT-PCR amplification across the FoMV cloning site MSC2. EV indicates the FoMV-DC empty vector that carries no insert. L4, L6, and L9 indicate the leaf number that was sampled. (c) Immunoblot assay showing GFP expression in FoMV-DC-GFP-infected leaf tissues that are used in panel B. The upper images are immunoblots using anti-GFP antibody, and the lower images show the protein loading control. (d) Green fluorescence in leaf 4 of plants infected with FoMV-DC or FoMV-DC-GFP. The area in the red box is enlarged in the lower panels, which are in two different focal planes to illustrate GFP fluorescence in epidermal (left) and mesophyll cells (right)

FIGURE 6 FoMV agroinoculation and expression of guide RNAs. (a) Schematic representation of FoMV vector in the pCAMBIA1380 binary vector. MCS1* indicates that the MCS1 site was modified to XbaI and PacI. (b) The yellow box indicates the position of insertion of a single 20-nucleotide guide RNA sequence fused to the 83-nucleotide scaffold (gRNA) at MCS2 of pFoMV-DC*. (c) gRNA target sites for N. benthamiana Pds1 S. viridis CA2, and Z. mays HKT1. Red text indicates the 20-nucleotide gRNA sequence, the gray boxes indicate the recognition sites for NcoI, PvuII and XcmI restriction endonucleases, the yellow arrows indicate the cleavage sites for the restriction endonucleases, and the protospacer adjacent motifs are indicated by the underlined text. (d) Strategy to express gRNA from the pFoMV-DC* vector in plants expressing the Cas9 protein. The red cross-hatch indicates that gene edits were expected to be observed in inoculated and systemic leaves in which FoMV-DC*-gRNA is accumulating MEI et al. | 9

FIGURE 7 FoMV-mediated editing of NbPDS in agroinfiltrated and systemic leaves of N. benthamiana plants expressing Cas9. (a) PCR- based assay to detect edits in NbPDS at 7 days post-inoculation (DPI) in agroinfiltrated (INF) and systemic (SYS) leaves. Oligonucleotide primers were used to generate 797-bp amplicons flanking the NbPDS guide RNA target site, and the amplicons were incubated with NcoI. The wild-type amplicons are cleaved into two bands of 541 bp and 256 bp, amplicons carrying edits that disrupt the NcoI site are not cleaved. F-g, FoMV-DC* expressing the NbPDS guide RNA; HC, co-infiltration with a construct expressing the TuMV helper component- proteinase; P19, co-infiltration with a construct expressing the TBSV 19 kDa protein; M, DNA size marker; P#, plant number; % Indels, percent of each PCR amplicon that was not digested; --, indicates that a non-digested product was not detected and therefore was not quantified; +, wild-type amplicon cleaved by NcoI; -, wild-type amplicon that was not incubated with NcoI. (b) PCR-based assay to detect edits in NbPDS at 14 DPI in agroinfiltrated (INF) and systemic (SYS) leaves. The systemic leaves were subdivided into two categories: middle systemic (MID-SYS) and top systemic (TOP-SYS). The “nd” and the red triangle indicate the sample for which indels were not quantified and sequence not determined due to inefficient PCR. (c) PCR-based assay to detect edits in NbPDS at 7 DPI in agroinfiltrated (INF) and systemic (SYS) leaves during a co-infection of FoMV-DC*-gNbPDS and turnip mosaic virus (TuMV). F + T, FoMV-DC* empty vector + TuMV; I, agroinfiltrated; S, systemic

3.4 | Development of a FoMV vector compatible 3.5 | FoMV-mediated Gene Editing in with agroinoculation N. benthamiana

FoMV can infect N. benthamiana, which is inoculated easily by To test whether single gRNA produced from FoMV-DC* can medi- Agrobacterium infiltration (agroinoculation), a much simpler method ate gene editing, we inserted a gRNA targeting the N. benthamiana than biolistic inoculation. To facilitate agroinoculation, the CaMV 35S Phytoene desaturase gene (NbPDS) at MCS2 to produce the pFoMV- promoter, genome of FoMV-DC, and nopaline synthase terminator DC*-gNbPDS construct (Figure 6b,c). This gRNA was previously re- were amplified as a single fragment from pFoMV-DC and inserted ported to mediate gene editing when expressed from tobacco rattle into the binary T-DNA vector pCAMBIA1380 (Figure 6a). The XhoI virus in N. benthamiana expressing Cas9 (Ali et al., 2015). The gRNA restriction endonuclease site at MCS1 was not unique in the pCAM- target site overlaps a recognition site for the NcoI restriction endonu- BIA1380 context, so it was replaced with PacI to create MCS1*. The clease (Figure 6c). N. benthamiana plants expressing Cas9 (Baltes et infectivity of pCAMBIA1380-FoMV-DC* was then tested by agroin- al., 2015) were agroinoculated with pFoMV-DC*-gNbPDS (Figure 6d). oculation of N. benthamiana plants. Mild mosaic symptoms were ob- Genomic DNA was extracted from infiltrated and systemic leaves served on systemic N. benthamiana leaves at approximately 3 weeks sampled at 7, 14, and 21 DPI to test modification of NbPDS. PCR was after inoculation (Figure S4a). Sweet corn seedlings were subse- used to produce 797-bp amplicons containing the target sequence quently rub-inoculated with sap from FoMV-infected N. benthami- that were then digested with NcoI (Figure 7). After NcoI digestion, ana leaves, and typical mosaic symptoms were observed within a the wild-type amplicon yielded two bands of 541 bp and 256 bp, week (Figure S4b). FoMV infection in the agroinoculated N. bentha- but many of the amplicons from FoMV-DC*-gNbPDS leaves yielded miana plants and rub-inoculated sweet corn plants was confirmed an additional band of approximately 797 bp that was resistant to by Western blot using anti-FoMV-CP antibody (Figure S4c). These NcoI digestion (Figure 7). The presence of the non-digested band results demonstrated that the pCAMBIA1380-FoMV-DC* (pFoMV- suggested that the target sequence was altered. At 7 DPI, the per- DC*) was infectious through agroinoculation and could be passaged centage of the modified DNA was estimated in the range of 73%- by rub-inoculation to maize. 91% in the infiltrated leaves and from 0% to 8% in the top systemic 10 | MEI et al.

FIGURE 8 Guide RNAs (gRNAs) are stably maintained at MCS2 in the FoMV-DC* vector. A single guide RNA (gRNA) targeting S. viridis Carbonic anhydrase 2 (SvCA2, Sevir.5G247900) fused to the 83-nucleotide tracrRNA scaffold was inserted into MCS2. (a) SvCA2 gRNA stability was tested in S. viridis plants. The FoMV-DC*-gSvCA2 construct was infiltrated into N. benthamiana and rub-inoculated onto young S. viridis leaves. The agarose gel shows the presence of stable inserts in S viridis leaves. + indicates a plasmid control, I indicates the inoculated leaf, and S indicates an upper systemic leaf. (b) Sanger sequencing of the RT-PCR bands from S. viridis inoculated with FoMV-DC*- gSvCA2. The insert is maintained after passage through N. benthamiana into S. viridis leaves without acquiring any deletions or mutations leaves (Figure 7a). To confirm that the NbPDS gene was modified at only sampled at 7 DPI. The co-infection elevated the estimated per- the gRNA target site, the NcoI-resistant bands were cloned and se- centage of indels in systemic leaves from 4%–6% to >70%, which was quenced. The sequences confirmed that the NcoI site was disrupted comparable to the infiltrated leaves in the FoMV-DC*-gNbPDS single mostly by small deletions (1 to 23 bp) that initiated 3 bp upstream infection (Figure 7c; Figure S8). As observed for the single FoMV- of the PAM (Figure S5). At 14 DPI, we collected samples from the DC*-gNbPDS experiments, the gene edits were characterized by infiltrated leaves, top systemic leaves that were asymptomatic at the small deletions ranging in size from 1 to 11 bp (Figure S9). time, and the middle systemic leaves that displayed viral symptoms. To further test the effect of TuMV HcPro, the Cas9 N. benthami- Similar to the 7 DPI samples, there was much more non-digested am- ana plants were crossed with TuMV HcPro N. benthamiana plants. plicon in the infiltrated leaves compared to the top leaves (Figure 7b). F1 plants were genotyped for presence of Cas9 and HcPro and then Interestingly, the amount of non-digested amplicon was similar in the inoculated with pFoMV-DC*-gNbPDS (Figure S10). The percent of middle systemic leaves and the infiltrated leaves. Sequence analysis indels in systemic leaves was 2%–4% in the absence of HcPro and of the undigested amplicons showed a similar spectrum of small in- 6%–37% in the presence of HcPro. This result indicates that the dels occurring within 3bp of the PAM (Figure S6). TuMV HcPro silencing suppressor promotes gene editing, although To test whether editing could be observed in flowers, a set of the effect is variable. The plants were grown to maturity, and all plants was inoculated and flowers were collected beginning at ap- seeds from the first two pods produced on each plant were sown proximately 2 months after inoculation. NbPDS gene editing was in soil. We expected to observe albino seedlings if any carried her- detected in the flower tissues of FoMV-DC*-gNbPDS-infected itable NbPDS loss-of-function mutations (Ali et al., 2015), but there Cas9-overexpressing plants with the percent of indels based on were none, which indicates that FoMV-DC*-gNbPDS did not induce NcoI digestion ranging from 6.9% to 14.8% (Figure S7). These data heritable edits in N. benthamiana. demonstrated that FoMV continued to express functional guide RNAs over the course of growth and development in N. benthami- ana. Seed was collected from capsules derived from the two flowers 3.6 | FoMV-mediated gene editing in S. viridis above those that were collected for DNA extraction and sown in soil. We did not observe any photobleached (albino) seedlings, which was To test FoMV-mediated gRNA delivery in monocots, we used a the expected phenotype if NbPds loss-of-function mutations were gRNA that targets the first exon of Carbonic anhydrase 2 (SvCA2, inherited (Ali et al., 2015). Sevir.5G247900) of S. viridis (Figure 6b). The FoMV-DC*-gSvCA2 To test whether the efficiency of FoMV-mediated NbPDS gene clone was first agroinoculated onto N. benthamiana to create inocu- editing could be elevated in the presence of viral silencing suppres- lum that was rub-inoculated onto a S. viridis line expressing Cas9. sors, we first co-infiltrated the Cas9 N. benthamiana with a mixture of To test gRNA stability, RT-PCR analysis was performed on inocu- Agrobacterium strains carrying pFoMV-DC*-gNbPDS and p35S-Hc- lated and systemic leaf tissues. The gRNA insertion was detected Pro or pCB-P19. We observed similar amounts of the non-digested in both the inoculated and systemic leaves, and sequence analysis band in infiltrated and systemic leaves at 7 and 14 DPI when com- confirmed that the gRNA remained intact following passage from pared to pFoMV-DC*-gNbPDS alone (Figure 7a,b), suggesting that N. benthamiana to S. viridis (Figure 8a,b). To determine whether gene the localized expression of the silencing suppressors did not enhance edits occurred in the inoculated or systemic S. viridis leaves, DNA editing. Next, we co-infiltrated pFoMV-DC*-gNbPDS with an in- was extracted and used as a template to PCR amplify the target re- fectious clone of turnip mosaic virus (TuMV) expressing GFP (pCB- gion in non-inoculated plants, inoculated leaves and systemic leaves. TuMV-GFP) (Lellis et al., 2002). TuMV is a potyvirus that systemically A 360-bp amplicon that is cleaved into bands of 228 bp and 132 bp infects N. benthamiana, and because it carries HcPro, we expected it by PvuII is expected in the wild type, but if the PvuII site is disrupted to interact synergistically with FoMV and increase the frequency of due to gene editing, then a mixture of cleaved and uncleaved am- editing in systemic tissues. This TuMV-GFP clone is highly virulent plicons is expected. Infection by FoMV-DC*-gSvCA2 resulted in on N. benthamiana plants and, in our conditions, causes plants to die about 45% indels in the inoculated leaf and 60% indels in the sys- at about 10 DPI, so infiltrated leaves and top systemic leaves were temic leaf (Figure 9a). The non-digested amplicons were cloned and MEI et al. | 11

sequenced, and we observed a series of small deletions (1–6 bp) that had occurred within 2–4 bp of the PAM sequence, disrupting the PvuII cleavage site (Figure 9b). These data demonstrate that FoMV delivered a functional guide RNA that could mediate Cas9-directed cleavage of SvCA2. Progeny of the plants infected with FoMV-DC*- gSvCA2 did not carry edits in SvCA2 indicating that FoMV did not induce heritable edits in S. viridis.

3.7 | FoMV-mediated gene editing in maize

To test gene editing in maize, a gRNA targeting ZmHKT1 was in- serted into pFoMV-DC* to create pFoMV-DC*-gZmHKT1. The ZmHKT1 gRNA targets a sequence overlapping a recognition site for the XcmI endonuclease (Figure 6c) (Xing et al., 2014). XcmI diges- tion of the wild-type 732-bp PCR amplicon containing the target sequence yielded two bands of 504 bp and 228 bp. Seedlings from a maize line segregating for Cas9 (Char et al., 2017) were co-inoc- ulated with FoMV-DC*-gZmHKT1 and the potyvirus SCMV, which expresses an HcPro (Zhang et al., 2008). Co-infection of FoMV and SCMV enhances FoMV accumulation and results in more severe disease symptoms (Figure S11), which is typical of potyvirus syn- ergisms (Pruss, Ge, Shi, Carrington, & Bowman Vance, 1997; Shi, Miller, Verchot, Carrington, & Vance, 1997). Presence of the Cas9 transgene mRNA was detected using RT-PCR, and the viruses were detected by Western blot using anti-FoMV-CP and anti-SCMV- CP antibodies (Figure 10). Genomic DNA was extracted from leaf 9 of the plants, and amplicons containing the ZmHKT1 target site were produced by PCR and then incubated with XcmI. Only ampli- FIGURE 9 FoMV-mediated editing of SvCA2 in systemic cons produced from plants that were Cas9 positive and infected by leaves of S. viridis plants expressing Cas9. (a) PCR-based assay FoMV-DC*-gZmHKT1 or FoMV-DC*-gZmHKT1 + SCMV were par- to detect edits in SvCA2 at 14 days post-inoculation in an upper tially resistant to XcmI digestion. We estimated that 3%–6% of the systemic leaf. FoMV-gRNA vectors were propagated in Nicotiana amplicons from Cas9 + FoMV-DC*-gZmHKT1 contained indels that benthamiana, and at 10 days after infiltration, a systemic leaf was used to rub-inoculate Cas9 S. viridis plants. Oligonucleotide primers disrupted the XcmI site and 7%–38% of the amplicons from Cas9 + were used to generate 360-bp amplicons flanking the SvCA2 FoMV-DC*-gZmHKT1 + SCMV contained indels that disrupted the guide RNA target site for multiple infected plants. The amplicons Xcm1 site (Figure 10a). However, not all plants that were positive for were incubated with PvuII restriction enzyme to create wild-type Cas9 and FoMV-DC*-gZmHKT1 alone or in combination with SCMV cleaved bands of 228 bp and 132 bp. Amplicons carrying edits produced amplicons resistant to XcmI (see plants 14, 13, and 16; that disrupt the PvuII site are not cleaved. The gel shown here represents a plant with successful disruption of the PvuII target Figure 10a). As expected, no non-digested amplicons were detected site by the FoMV-gRNA vector. + indicates a plasmid control, I in either Cas9-negative or FoMV-DC*-gZmHKT1-negative plants indicates the inoculated leaf, and S indicates an upper systemic (see plants 1, 15, 5, 9, 11, 12, 17; Figure 10a). Similar to N. bentha- leaf. % Indels, percent of each PCR amplicon that was not digested miana and S. viridis, small deletions within two to three bp of the as measured by ImageJ software; --, indicates that a non-digested PAM were most common in maize (Figure 10b). We were not able to product was not detected and therefore was not quantified. (b) obtain seeds from these plants due to growth chamber conditions Sequence analysis of the non-cleaved amplicons that were gel purified and cloned. The PvuII recognition site is indicated by and the severe symptoms of co-infection with SCMV. Therefore, the the gray box in the wild-type sequence, and the cleavage site is possibility of heritable mutations in ZmHKT1 could not be tested. indicated by the yellow arrow. The protospacer adjacent motif (PAM) is indicated by the underlined sequence, and the guide RNA sequence is indicated by the red text. The blue dashes represent 4 | DISCUSSION nucleotide deletions. Each sample is from an individual clone, negative numbers to the right of a sequence show the number of nucleotides deleted, and (#x) shows the number of times that Here, we describe a set of recombinant FoMV vectors that have ap- sequence occurred plications for gene expression and single gRNA delivery, and they 12 | MEI et al.

FIGURE 10 FoMV-mediated editing of ZmHKT1 in systemic leaves of maize plants expressing Cas9. (a) PCR-based assay to detect edits in ZmHKT1 systemic leaf 9. Oligonucleotide primers were used to generate 732-bp amplicons flanking the ZmHKT1 guide RNA target site, and the amplicons were incubated with XcmI. The wild-type amplicons are cleaved into two bands of 504 bp and 228 bp, amplicons carrying edits that disrupt the XcmI site are not cleaved. RT-PCR was used to determine the presence of the Cas9 transgene, FoMV-DC*-gZmHKT1 (F-g), and sugarcane mosaic virus (SCMV) in each leaf sample. The numbers above each lane indicate the plant from which the leaf sample was derived. M, DNA size marker; % Indels, percent of each PCR amplicon that was not digested; --, indicates that a non-digested product was not detected and therefore was not quantified; 1L9, a leaf 9 sample from plant 1; 1L10, a leaf 10 sample from plant 1; -, wild-type amplicon that was not digested. (b) Sequence analysis of amplicons that were gel purified and cloned. The Xcm1 recognition site is indicated by the gray box in the wild-type sequence, and the cleavage site is indicated by the yellow arrow. The protospacer adjacent motif (PAM) is indicated by the blue line, and the guide RNA sequence is indicated by the red text. The blue dashes represent nucleotide deletions, blue letters represent nucleotide insertions, and a red dash indicates a gap in the alignment to the guide RNA sequence due to a nucleotide insertion in one of the amplicon sequences. P# indicates the corresponding plant number, and the ratio to the right of the wild-type sequences indicate the number of clones that carried an indel out of the total number of clones sequenced for that plant sample. Negative numbers to the right of a sequence show the number of nucleotides deleted, positive numbers show the number of nucleotides inserted, and (x#) shows the number of times that sequence occurred were demonstrated to function in maize, S. viridis, and N. benthami- ORF5A starts 144 nucleotides upstream of and is in frame with ana. These vectors were derived from a previous FoMV vector de- the CP ORF, and it is predicted to produce a variant of CP with veloped by us for VIGS, which could not be used for gene expression a 48-amino acid N-terminal extension (Bruun-Rasmussen et al., because of the position of MCS1 after the stop codon of ORF5 (Mei 2008; Robertson et al., 2000). Disrupting ORF5A simplified the et al., 2016). Our new vectors can be inoculated into plants using duplicated CP cloning strategy for the expression of proteins and either DNA particle bombardment or agroinfiltration. Although allowed us to reduce the duplicated CP promoter region to 54 nu- VIGS was not tested using the new vector set, they are expected cleotides. We aimed to minimize the size of the duplication, while to be useful for this purpose, given a previous demonstration of retaining the promoter function, to reduce the risk of recombi- VIGS using a duplicated CP promoter strategy in FoMV by Liu et al. nation (Dickmeis et al., 2014). We confirmed that the expected (2016). In order to enable foreign gene expression, the subgenomic subgenomic mRNA transcripts initiating at 14 nucleotides down- promoter duplication strategy, which has been successfully demon- stream of the core promoter sequence at GAA were produced strated in PVX vectors (Dickmeis et al., 2014; Lacomme & Chapman, from the duplicated CP subgenomic promoter by using 5’ RACE. 2008; Sablowski et al., 1995; Wang et al., 2014), was adopted to gen- In the FoMV-DC vectors, the duplicated nucleotides overlapping erate pFoMV-DP/DC vectors. A similar strategy was used in another with ORF4 were further modified by changing the third base of FoMV expression vector developed by Bouton et al. (2018), although each codon, which yielded better stability of the bar insertion. It different approaches were taken to design and build the vectors. may be possible to further stabilize expression from FoMV using To engineer the pFoMV-DP/DC vectors, we first confirmed a heterologous promoter strategy as shown for PVX by Dickmeis that the ORF5A was dispensable for infection of our primary tar- et al. (2014) and for other viruses, such as TMV (Roy et al., 2011). get plant, maize. Robertson et al. (2000) previously showed that However, this approach may not be straightforward, because the ORF5A is not required for systemic infection of barley plants. 8 nucleotide CP subgenomic promoter core sequence differs at MEI et al. | 13 two positions relative to most other characterized potexviruses to plants expressing Cas9 alone. HcPro's ability to promote gene ed- (Kim & Hemenway, 1997), and the FoMV CP subgenomic promoter iting is consistent with results from Mao et al. (2018) who showed did not function in the context of the PVX genome (Dickmeis et that the tomato bushy stunt P19 silencing suppressor improved gene al., 2014). editing by the CRISPR/Cas9 system in Arabidopsis thaliana. To test gene expression, the coding sequences of the BAR (552 bp) To extend this idea further, we co-inoculated maize seedlings and GFP (720 bp) marker proteins were inserted into the FoMV-DC segregating for Cas9 with FoMV-DC*-gZmHKT1 and SCMV. The and FoMV-DP vectors. Although expression varied from plant to presence of SCMV promoted FoMV accumulation and resulted in plant, when inoculated at the two-leaf stage, protein expression was more severe disease than with either virus alone, which is consistent observed from L4 to L6. In the case of GFP expression, green fluo- with a synergistic interaction mediated by a potyvirus. Consistent rescence was detected through L8 and no fluorescence signal was with TuMV, co-infection of SCMV with FoMV-DC*-gZmHKT1 re- observed in L9 samples, which was consistent with the results of RT- sulted in the highest levels of gene edits in systemic leaves, although PCR showing total loss of the GFP insertion in all L9 samples. Our pre- the frequency was variable. Taken together, we conclude that syn- vious work found that when plants were infected by the FoMV vector ergistic viruses or their silencing suppressors can greatly enhance carrying 200–300 bp insertions at MCS1 after the ORF5 stop codon, FoMV-mediated gene editing. However, SCMV + FoMV co-infection plants began to show evidence for deletions of the inserts by L9 and results in severe symptoms that prevent proper maize development, partial or total deletion was observed in L top (leaf 12–14) (Mei et al., and so phenotypic effects of edits would be difficult to discern. An 2016). Considering that the insertion size of bar and GFP in FoMV-DC important line of investigation for the future will be to optimize si- and FoMV-DP is approximately twice as big, an earlier and more ex- lencing suppressor activity to enable more efficient gene editing tensive deletion was not surprising. In line with a relationship between without such severe symptoms. insert stability and size, we found that single gRNA, which are 103 The vector set and experiments presented here demonstrate that nucleotides in size, was very stable at MCS2 in both the FoMV-DC recombinant FoMV clones have a wide variety of potential uses for and FoMV-DP contexts, with no evidence for deletions in S. viridis or VIGS, VOX, and VEdGE (virus-enabled gene editing). The utility of maize plants. This influence of insert size is in agreement with obser- FoMV for these different applications can supplement conventional vations in PVX vectors (Avesani et al., 2007). genetics and transgenic plant approaches for identifying gene func- The most exciting aspect of this work is a demonstration that tions in maize, or to investigate the functions of novel proteins in FoMV can deliver functional gRNA in monocot plants, maize and maize. Our work demonstrates that there are limitations when per- S. viridis, expressing Cas9 and in N. benthamiana. We tested proof forming studies using VOX, because of the stability of inserts, but of concept in N. benthamiana by agroinoculation where FoMV-DC*- we expect that FoMV can be useful for expressing proteins that can gNbPDS induced indels at a high frequency in the agroinoculated leaf induce phenotypes in leaves 4–6 of young maize plants. The single patches. Indels were also observed in the systemic leaves at differing gRNA inserts appear to be stable, and so FoMV could be used to in- frequencies depending on the timing and location of the sampled duce gene edits in vegetative tissues of monocot plants expressing leaves. To enhance the frequency of editing in systemic tissues, viral Cas9. Based on the frequency of indels in maize plants expressing silencing suppressors were provided by transient agroinfiltrations, Cas9, we do not expect to be able to observe phenotypes due to gene co-infections, and stable transformation. Transient agroinfiltration editing. of HcPro and P19 did not enhance editing, because this localized Heritable gene edits were not observed in the progeny of method of delivery could not promote FoMV accumulation in sys- N. benthamiana or S. viridis plants. In N. benthamiana, edits were temic tissues. Co-inoculation of FoMV with the potyvirus, TuMV, observed in the flowers, but they must have occurred in mater- resulted in remarkably high levels of gene edits in systemic leaves nal tissues, such as sepals and petals. The lack of heritable edits at 7 dpi, but plants died within a few days, so further analyses were is consistent with the inability of FoMV to be seed transmitted not possible. The high frequency of editing was in stark contrast with in S. viridis (Paulsen & Niblett, 1977). Because single guide RNA the low frequency of editing in comparable systemic leaves of plants can vary in their efficiency and specificity (Doench et al., 2016), that had been inoculated with FoMV-DC*-gNbPDS alone. These the ability of FoMV to induce gene edits in monocot vegetative results indicate that TuMV and FoMV interact synergistically, likely tissues is expected to have useful applications for rapidly screen- due to the HcPro silencing suppressor carried by TuMV (Kasschau ing single gRNA candidates prior to producing gRNA-Cas9 trans- et al., 2003). The increased frequency of editing is expected to be genic lines carrying heritable edits. A very interesting possibility is due to increased accumulation of FoMV genomic and subgenomic that Cas9 and gRNA could be expressed simultaneously from the RNAs resulting in more production of guide RNAs. In addition, the FoMV genome as was recently shown for the negative strand cy- suppression of silencing may also enhance expression of the Cas9 torhabdovirus, BYSMV, which can tolerate insertion of the 4.4 kb transgene. Additional experiments will be required to distinguish Cas9 coding sequence and a single gRNA (Gao et al., 2019). The among these possibilities. Consistent with the idea that HcPro may recombinant BYSMV could induce gene edits in the agroinfiltrated have promoted these effects, inoculation of plants expressing both leaves of N. benthamiana plants, but systemic gene editing was not Cas9 and HcPro with FoMV-DC-gNbPDS also resulted in a higher, reported. We expect that additional work will be needed to stabi- but variable frequency of gene editing in systemic leaves compared lize an ORF as large as Cas9 in the FoMV genome, so that it could 14 | MEI et al. be used to express both Cas9 and gRNA for systemic gene editing Bruun-Rasmussen, M., Madsen, C. T., Johansen, E., & Albrechtsen, M. in maize and other monocots. (2008). Revised sequence of foxtail mosaic virus reveals a triple gene block structure similar to potato virus X. Archives of Virology, 153, 223–226. Butler, N. M., Baltes, N. J., Voytas, D. F., & Douches, D. S. (2016). CONFLICT OF INTEREST Geminivirus-mediated genome editing in potato (Solanum tuberosum L.) using sequence-specific nucleases. Frontiers in Plant Science, 7, 1045. The authors declare no conflict of interest associated with the work Cermak, T., Curtin, S. J., Gil-Humanes, J., Cegan, R., Kono, T. J. Y., Konecna, described in this manuscript. E., … Voytas, D. F. (2017). A multipurpose toolkit to enable advanced genome engineering in plants. The Plant Cell, 29, 1196–1217. https​:// doi.org/10.1105/tpc.16.00922​ Chakrabarty, R., Banerjee, R., Chung, S. M., Farman, M., Citovsky, V., AUTHOR CONTRIBUTIONS Hogenhout, S. A., … Goodin, M. (2007). PSITE vectors for stable inte- gration or transient expression of autofluorescent protein fusions in Y.M., E.K., D.F.V., and S.A.W. conceived the original research plans; plants: Probing Nicotiana benthamiana-virus interactions. Molecular D.F.V. and S.A.W. supervised the experiments; Y.M., B.M.B., E.E.E., Plant-Microbe Interactions, 20, 740–750. Char, S. N., Neelakandan, A. K., Nahampun, H., Frame, B., Main, M., and E.K. performed most of the experiments; A.K.N. provided Spalding, M. H., … Yang, B. (2017). An Agrobacterium-delivered critical materials and guidance; and Y.M., B.M.B., E.E.E., D.F.V., and CRISPR/Cas9 system for high-frequency targeted mutagenesis in S.A.W. wrote the article with contributions of all the authors. S.A.W. maize. Plant Biotechnology Journal, 15, 257–268. agrees to serve as the author responsible for contact and ensures Choi, I. R., Stenger, D. C., Morris, T. J., & French, R. (2000). A plant virus vector for systemic expression of foreign genes in cereals. The Plant Journal, communication. 23, 547–555. https​://doi.org/10.1046/j.1365-313x.2000.00820.x Cody, W. B., Scholthof, H. B., & Mirkov, T. E. (2017). Multiplexed gene editing and protein overexpression using a tobacco mosaic virus viral RESPONSIBILITIES OF THE AUTHOR FOR CONTACT vector. Plant Physiology, 175, 23–35. Cruz, S. S., Roberts, A. G., Prior, D. A., Chapman, S., & Oparka, K. J. (1998). Cell-to-cell and phloem-mediated transport of potato virus X: The It is the responsibility of the author for contact to ensure that all role of virions. The Plant Cell, 10, 495–510. https​://doi.org/10.1105/ scientists who have contributed substantially to the conception, tpc.10.4.495 design, or execution of the work described in the manuscript are Dahan-Meir, T., Filler-Hayut, S., Melamed-Bessudo, C., Bocobza, S., included as authors, in accordance with the guidelines from the Czosnek, H., Aharoni, A., & Levy, A. A. (2018). Efficient in planta gene targeting in tomato using geminiviral replicons and the CRISPR/Cas9 Committee on Publication Ethics (COPE) (http://publi​catio​nethi​ system. The Plant Journal, 95, 5–16. cs.org/resou​rces/guide​lines​). It is the responsibility of the author for Dickmeis, C., Fischer, R., & Commandeur, U. (2014). Potato virus X-based contact also to ensure that all authors agree to the list of authors and expression vectors are stabilized for long-term production of pro- the identified contributions of those authors. The author responsi- teins and larger inserts. Biotechnology Journal, 9, 1369–1379. Ding, X. S., Schneider, W. L., Chaluvadi, S. R., Mian, M. A., & Nelson, R. ble for contact and ensuring the distribution of materials integral to S. (2006). Characterization of a brome mosaic virus strain and its use the findings presented in this article in accordance with the Journal as a vector for gene silencing in monocotyledonous hosts. Molecular policy described in the Instructions for Authors (http://www.plant​ Plant-Microbe Interactions, 19, 1229–1239. physi​ol.org) is S. A. Whitham ([email protected]). Doench, J. G., Fusi, N., Sullender, M., Hegde, M., Vaimberg, E. W., Donovan, K. F., … Root, D. E. (2016). Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9. REFERENCES Nature Biotechnology, 34, 184–191. Ali, Z., Abul-faraj, A., Li, L., Ghosh, N., Piatek, M., Mahjoub, A., … Mahfouz, Draghici, H. K., & Varrelmann, M. (2009). Evidence that the linker be- M. M. (2015). Efficient virus-mediated genome editing in plants using tween the methyltransferase and helicase domains of potato virus the CRISPR/Cas9 system. Molecular Plant, 8, 1288–1291. X replicase is involved in homologous RNA recombination. Journal of Ali, Z., Eid, A., Ali, S., & Mahfouz, M. M. (2018). Pea early-browning Virology, 83, 7761–7769. https​://doi.org/10.1128/JVI.00179-08 virus-mediated genome editing via the CRISPR/Cas9 system in Ford, R. E., Bucholtz, W. F., & Lambe, R. C. (1967). Occurrence of maize Nicotiana benthamiana and Arabidopsis. Virus Research, 244, 333–337. dwarf mosaic virus in Iowa in 1966. Plant Disease Reporter, 51, https​://doi.org/10.1016/j.virus​res.2017.10.009 388–389. Avesani, L., Marconi, G., Morandini, F., Albertini, E., Bruschetta, M., Gao, Q., Xu, W. Y., Yan, T., Fang, X. D., Cao, Q., Zhang, Z. J., … Wang, Bortesi, L., … Porceddu, A. (2007). Stability of potato virus X expres- X. B. (2019) Rescue of a plant cytorhabdovirus as versatile ex- sion vectors is related to insert size: Implications for replication mod- pression platforms for planthopper and cereal genomic studies. els and risk assessment. Transgenic Research, 16, 587–597. https​://doi. New Phytologist, 223(4), 2120–2133. https​://doi.org/10.1111/ org/10.1007/s11248-006-9051-1 nph.15889​ Baltes, N. J., Hummel, A. W., Konecna, E., Cegan, R., Bruns, A. N., Bisaro, Gil-Humanes, J., Wang, Y., Liang, Z., Shan, Q., Ozuna, C. V., Sanchez- D. M., & Voytas, D. F. (2015). Conferring resistance to geminiviruses Leon, S., … Voytas, D. F. (2017). High-efficiency gene targeting in with the CRISPR-Cas prokaryotic immune system. Nature Plants, 1, hexaploid wheat using DNA replicons and CRISPR/Cas9. The Plant 15145. Journal, 89, 1251–1262. https​://doi.org/10.1111/tpj.13446​ Bouton, C., King, R. C., Chen, H., Azhakanandam, K., Bieri, S., Hammond- Hill, J. H., Ford, R. E., & Benner, H. I. (1973). Purification and partial char- Kosack, K. E., & Kanyuka, K. (2018). Foxtail mosaic virus: A viral vec- acterization of maize dwarf mosaic virus strain B (sugarcane mosaic tor for protein expression in cereals. Plant Physiology, 177, 1352–1367. virus). Journal of General Virology, 20, 327–339. MEI et al. | 15

Hsieh, M. H., Lu, H. C., Pan, Z. J., Yeh, H. H., Wang, S. S., Chen, W. assembly and biotechnologies of plant viruses. Plant Biotechnology H., & Chen, H. H. (2013). Optimizing virus-induced gene silenc- Journal, 17(6), 1010–1026. https​://doi.org/10.1111/pbi.13084​. ing efficiency with cymbidium mosaic virus in Phalaenopsis Paulsen, A. Q., & Niblett, C. L. (1977). Purification and properties of fox- f l o w e­ r . Plant Science, 201–202, 25–41. https​://doi.org/10.1016/j. tail mosaic virus. Phytopathology, 67, 1346–1351. plant​sci.2012.11.003 Pruss, G., Ge, X., Shi, X. M., Carrington, J. C., & Bowman Vance, V. (1997). Huisman, M. J., Linthorst, H. J., Bol, J. F., & Cornelissen, J. C. (1988). Plant viral synergism: The potyviral genome encodes a broad-range The complete nucleotide sequence of potato virus X and its ho- pathogenicity enhancer that transactivates replication of heterologous mologies at the amino acid level with various plus-stranded RNA viruses. The Plant Cell, 9, 859–868. https​://doi.org/10.1105/tpc.9.6.859 viruses. Journal of General Virology, 69, 1789–1798. https​://doi. Purkayastha, A., Mathur, S., Verma, V., Sharma, S., & Dasgupta, I. (2010). org/10.1099/0022-1317-69-8-1789 Virus-induced gene silencing in rice using a vector derived from Jiang, N., Zhang, C., Liu, J. Y., Guo, Z. H., Zhang, Z. Y., Han, C. G., & a DNA virus. Planta, 232, 1531–1540. https​://doi.org/10.1007/ Wang, Y. (2019). Development of beet necrotic yellow vein vi- s00425-010-1273-z rus-based vectors for multiple-gene expression and guide RNA Robertson, N. L., French, R., & Morris, T. J. (2000). The open reading delivery in plant genome editing. Plant Biotechnology Journal, 17, frame 5A of foxtail mosaic virus is expressed in vivo and is dispens- 1302–1315. able for systemic infection. Archives of Virology, 145, 1685–1698. Kasschau, K. D., Xie, Z., Allen, E., Llave, C., Chapman, E. J., Krizan, K. A., & Roy, G., Weisburg, S., Foy, K., Rabindran, S., Mett, V., & Yusibov, V. (2011). Carrington, J. C. (2003). P1/HC-Pro, a viral suppressor of RNA silenc- Co-expression of multiple target proteins in plants from a tobacco ing, interferes with Arabidopsis development and miRNA function. mosaic virus vector using a combination of homologous and heterol- Developmental Cell, 4, 205–217. ogous subgenomic promoters. Archives of Virology, 156, 2057–2061. Kim, K. H., & Hemenway, C. (1997). Mutations that alter a conserved https​://doi.org/10.1007/s00705-011-1069-x element upstream of the potato virus X triple block and coat protein Sablowski, R. W., Baulcombe, D. C., & Bevan, M. (1995). Expression of a genes affect subgenomic RNA accumulation. Virology, 232, 187–197. flower-specific Myb protein in leaf cells using a viral vector causes https​://doi.org/10.1006/viro.1997.8565 ectopic activation of a target promoter. Proceedings of the National Lacomme, C., & Chapman, S. (2008). Use of potato virus X (PVX)-based Academy of Sciences of the United States of America, 92, 6901–6905. vectors for gene expression and virus-induced gene silencing (VIGS). https​://doi.org/10.1073/pnas.92.15.6901 Current Protocols in Microbiology, 8, 16I.1.1–16I.1.13. Scofield, S. R., Huang, L., Brandt, A. S., & Gill, B. S. (2005). Development Lee, W. S., Hammond-Kosack, K. E., & Kanyuka, K. (2012). Barley stripe of a virus-induced gene-silencing system for hexaploid wheat and its mosaic virus-mediated tools for investigating gene function in cereal use in functional analysis of the Lr21-mediated leaf rust resistance plants and their pathogens: Virus-induced gene silencing, host-medi- pathway. Plant Physiology, 138, 2165–2173. ated gene silencing, and virus-mediated overexpression of heterolo- Shi, X. M., Miller, H., Verchot, J., Carrington, J. C., & Vance, V. B. (1997). gous protein. Plant Physiology, 160, 582–590. Mutations in the region encoding the central domain of helper com- Lee, W.-S., Hammond-Kosack, K. E., & Kanyuka, K. (2015). In planta ponent-proteinase (HC-Pro) eliminate potato virus X/potyviral syner- transient expression systems for monocots. In K. Azhakanandam, A. gism. Virology, 231, 35–42. https​://doi.org/10.1006/viro.1997.8488 Silverstone, H. Daniell, & M. R. Davey (Eds.), Recent advancements in Tatineni, S., McMechan, A. J., Hein, G. L., & French, R. (2011). Efficient and gene expression and enabling technologies in crop plants (pp. 391–422). stable expression of GFP through wheat streak mosaic virus-based New York, NY: Springer. vectors in cereal hosts using a range of cleavage sites: Formation of Lellis, A. D., Kasschau, K. D., Whitham, S. A., & Carrington, J. C. (2002). dense fluorescent aggregates for sensitive virus tracking. Virology, Loss-of-susceptibility mutants of Arabidopsis thaliana reveal an 410, 268–281. https​://doi.org/10.1016/j.virol.2010.10.043 essential role for eIF(iso)4E during potyvirus infection. Current Van Eck, J., Swartwood, K., Pidgeon, K., & Maxson-Stein, K. (2017). Biology, 12, 1046–1051. Agrobacterium tumefaciens-mediated transformation of Setaria vir- Liou, M. R., Huang, Y. W., Hu, C. C., Lin, N. S., & Hsu, Y. H. (2014). A dual idis. In A. Doust, & X. Diao (Eds.), Genetics and genomics of Setaria gene-silencing vector system for monocot and dicot plants. Plant (pp. 343–356). Cham, Switzerland: Springer International Publishing. Biotechnology Journal, 12, 330–343. Verchot-Lubicz, J. (2005). A new cell-to-cell transport model for potexvi- Liu, N., Xie, K., Jia, Q., Zhao, J. P., Chen, T. Y., Li, H. G., … Liu, Y. L. (2016). ruses. Molecular Plant-Microbe Interactions, 18, 283–290. https​://doi. Foxtail mosaic virus-induced gene silencing in monocot plants. Plant org/10.1094/MPMI-18-0283 Physiology, 171, 1801–1807. Wang, M., Lu, Y., Botella, J. R., Mao, Y., Hua, K., & Zhu, J. K. (2017). Liu, Z., & Kearney, C. M. (2010). An efficient foxtail mosaic virus vector Gene targeting by homology-directed repair in rice using a gemini- system with reduced environmental risk. BMC Biotechnology, 10, 88. virus-based CRISPR/Cas9 system. Molecular Plant, 10, 1007–1010. https​://doi.org/10.1186/1472-6750-10-88 Wang, R., Yang, X., Wang, N., Liu, X., Nelson, R. S., Li, W., … Zhou, T. Mahas, A., Ali, Z., Tashkandi, M., & Mahfouz, M. M. (2019). Virus- (2016). An efficient virus-induced gene silencing vector for maize mediated genome editing in plants using the CRISPR/Cas9 system. functional genomics research. The Plant Journal, 86, 102–115. Methods in Molecular Biology, 1917, 311–326. https​://doi.org/10.1111/tpj.13142​ Mao, Y., Yang, X., Zhou, Y., Zhang, Z., Botella, J. R., & Zhu, J. K. (2018). Wang, Y., Cong, Q. Q., Lan, Y. F., Geng, C., Li, X. D., Liang, Y. C., … Li, X. D. Manipulating plant RNA-silencing pathways to improve the gene ed- (2014). Development of new potato virus X-based vectors for gene iting efficiency of CRISPR/Cas9 systems. Genome Biology, 19, 149. over-expression and gene silencing assay. Virus Research, 191, 62–69. https​://doi.org/10.1186/s13059-018-1529-7 https​://doi.org/10.1016/j.virus​res.2014.07.018 Mei, Y., & Whitham, S. A. (2018). Virus-induced gene silencing in maize Whitham, S. A., Yamamoto, M. L., & Carrington, J. C. (1999). Selectable with a foxtail mosaic virus vector. Methods in Molecular Biology, 1676, viruses and altered susceptibility mutants in Arabidopsis thaliana. 129–139. Proceedings of the National Academy of Sciences of the United States of Mei, Y., Zhang, C., Kernodle, B. M., Hill, J. H., & Whitham, S. A. (2016). A America, 96, 772–777. https​://doi.org/10.1073/pnas.96.2.772 foxtail mosaic virus vector for virus-induced gene silencing in maize. Xing, H. L., Dong, L., Wang, Z. P., Zhang, H. Y., Han, C. Y., Liu, B., … Plant Physiology, 171, 760–772. https​://doi.org/10.1104/pp.16.00172​ Chen, Q. J. (2014). A CRISPR/Cas9 toolkit for multiplex genome Pasin, F., Menzel, W., & Daros, J. A. (2019). Harnessed viruses in the age of editing in plants. BMC Plant Biology, 14, 327. https​://doi.org/10.1186/ metagenomics and synthetic biology: an update on infectious clone s12870-014-0327-y 16 | MEI et al.

Zaidi, S. S., & Mansoor, S. (2017). Viral vectors for plant genome engi- SUPPORTING INFORMATION neering. Frontiers in Plant Science, 8, 539. Additional supporting information may be found online in the Zhang, C., Bradshaw, J. D., Whitham, S. A., & Hill, J. H. (2010). The Supporting Information section. development of an efficient multipurpose bean pod mottle virus viral vector set for foreign gene expression and RNA silencing. Plant Physiology, 153, 52–65. https​://doi.org/10.1104/pp.109. 151639 How to cite this article: Mei Y, Beernink BM, Ellison EE, et al. Zhang, X., Du, P., Lu, L., Xiao, Q., Wang, W., Cao, X., … Li, Y. (2008). Protein expression and gene editing in monocots using foxtail Contrasting effects of HC-Pro and 2b viral suppressors from sugar- mosaic virus vectors. Plant Direct. 2019;3:1–16. h t t p s :​ / / d o i . cane mosaic virus and tomato aspermy cucumovirus on the accumu- lation of siRNAs. Virology, 374, 351–360. https​://doi.org/10.1016/j. org/10.1002/pld3.181 virol.2007.12.045