Xbai

pGRS-88: The gene fragment was amplified by PCR with the primers O-GRS-112 (aattattTCTAGAGCCACCATGTCGCAAGGCCTCCAGCTCCTGTTTC) and O-GRS-113 (GCTCCTGTACTTCCTGAAAGTTGACTCTGTAG) with the Phusion polymerase (NEB, Ipswich, MA, USA) from RNA isolated from stimulated lymphocytes. The PCR product was digested with XbaI and phosphorylated with T4 polynucleotide kinase 3′ phosphatase minus (NEB, Ipswich, MA, USA) and ligated to the vector pGRS-12 (59 (link)) digested with XbaI (NEB, Ipswich, MA, USA) and AfeI (NEB, Ipswich, MA, USA).

Regions flanking the PSPTO_1404 (hrpL) gene were amplified by primers oSWC04110/oSWC04112 and oSWC04114/oSWC04116 (see Table S1 for all primer sequences) from DC3000 genomic DNA, purified by gel electrophoresis and gel extraction (Qiagen), and joined by SOEing PCR [39] (link). The joined fragment was then digested with XbaI (all enzymes were obtained from New England Biolabs unless otherwise noted), ligated with XbaI digested pK18mobsacB, and then transformed into E. coli TOP10. To ensure that the resulting construct was free from unwanted mutations, it was sequenced using primers M13F, M13R, oSWC05110, oSWC05111, oSWC05112, and oSWC05113. The FLAG-tagged construct was introduced into DC3000 by electroporation to generate HLN090. Merodiploid intermediates were selected for growth on medium containing kanamycin. Recombinants that had eliminated pK18mobsacB plasmid sequences were identified by sucrose counter-selection. Sucrose-resistant, Kan-sensitive colonies were analyzed by Sanger sequencing (with primers oSWC04110, oSWC04112, oSWC04114, oSWC04116, oSWC05110, oSWC05111, oSWC05112, and oSWC05113) to confirm successful tagging.

pLVTHM vectors containing shRNA targeting SULF2 (GGAGTGGTGGTGTCAATA) or a non-targeted scrambled control (AACAGTCGCGTTTGCGACTGG) were used as previously described^{13 (link)}. The pBARLHyg lentiviral reporter plasmid was a gift from Dr. Randall Moon (University of Washington)^{74 (link)}. pGL3 BRE luciferase was a gift from Martine Roussel and Peter ten Dijke (Addgene, plasmid # 45126)^{75 (link)}. The lentiviral pTRIP-EF1a backbone was a gift from Abdel Benraiss (University of Rochester, Rochester, NY)^{76 (link)}. To generate L-BRE-Luc, pGL3 BRE Luciferase was digested with KpnI (NEB), blunted using a commercial blunting kit (NEB), and digested with XbaI (NEB). The pTRIP-Ef1α plasmid was linearized by digestion with MluI (NEB), blunted, and digested with XbaI to remove Ef1α-mCherry-WPRE. The BRE-Luc fragment and the pTRIP-EF1a backbone were purified by gel extraction (Qiagen) and ligated with T4 DNA ligase (Invitrogen) to generate L-BRE-Luc. pBARLHyg, L-BRE-Luc and pLVTHM were packaged in lentivirus as previously described^{52 (link)}. Briefly, HEK 293T cells were triple transfected with viral vector and packaging plasmids pLP/VSVG (Invitrogen) and psPAX2 (AddGene), and viral supernatant was collected 42 h later. pLVTHM viral titer was determined by quantification of pLVTHM-driven GFP expression.

Plasmid pHellsgate 8 (Helliwell and Waterhouse, 2003 (link)) was the backbone vector for cloning the virus segments in direct and reverse directions separated by a hairpin. The plasmid was digested with XhoI and XbaI (New England Biolabs, UK). Next, amplicons were obtained using the cDNA from CGMMV and ToLCNDV and the primers described in Supplementary Table 2. The amplicons, derived from the cp and mp genes of CGMMV and the AV1 and BC1 genes from ToLCNDV were used for Gibson assembly to pHellsgate 8. For each construction, four segments were assembled following the manufacturer's recommendations (New England Biolabs, UK), including the virus gene(s) with direct and reverse positions and the corresponding segments for the XbaI-XhoI digestions of pHellsgate8 vector that included the hairpin. The resulting plasmids, pGHE-CP, pGHE-MP, pGHE-AV1, and pGHE-BC1 were used to transform Top10 E. coli cells that were grown in LB plates supplemented with streptomycin (100 μg/ml). Plasmids were extracted from the cells and checked by restriction fragment analysis and Sanger sequencing. The plasmids were then transferred to Agrobacterium tumefaciens LBA4404 by electroporation and selected with the same antibiotic.

pLVX-Puro (Cat. 632164, TakaraBio, San Jose, CA, USA) was digested using XbaI (Cat. R0145, New England Biolabs NEB, Ipswich, MA, USA), and empty ligation was minimized using Phosphatase (NEB, Cat. M0290). The open vector was then purified using the MinElute PCR Purification Kit (Cat. 28004, Qiagen, Germantown, MD, USA), blunted at room temperature using the Quick Blunting Kit from NEB (Cat. E1201), and heat inactivated for 10 min at 70 °C. The ZsGreen1 gene from pRetroX-IRES-ZsGreen1 (TakaraBio, Cat. 632520) and the c-MYC gene from a 293T cell-derived in-house cDNA library were cloned by PCR using Phusion High-Fidelity PCR Master Mix with HF Buffer (Cat. F531L, Thermo Scientific, Waltham, MA, USA). The PCR product was phosphorylated using T4 Polynucleotide Kinase (NEB, Cat. M0201). Blunt ligation was performed with the PCR-amplified and phosphorylated ZsGreen1-c-Myc fusion construct and XbaI-cut and blunted recipient vector using Quick Ligase (NEB, Cat. M2200S). The ligation mixture was transformed into Oneshot Stbl3 competent cells (Invitrogen, Cat. C7373-03) according to the recommended protocol and plated on LB + Ampicillin plates (100 μg/mL). The final construct was confirmed by 5′ sequencing at the BCM DNA Sequencing Core.

To construct riboswitch and CDS plasmids for this study, we started with the pFTV1 vector backbone, which contains mRFP1 modified to contain an N-terminal SacI restriction site^{41 (link)}. We used the Riboswitch Calculator to design candidate riboswitch sequences and the Operon Calculator to codon-optimize the MS2 coat protein CDS and design an optimal RBS sequence (Source Data)^{39 (link),64 (link)}. We designed and ordered gBlocks, containing primer binding sites and additional restriction sites, and PCR primers for both the riboswitches and MS2 coat protein CDS (Integrated DNA Technologies). We PCR amplified the gBlocks using Phusion or Q5 DNA polymerase (New England Biolabs). For the riboswitches, we digested the riboswitch amplicons and pFTV1 vector backbone with XbaI and SacI-HF (New England Biolabs). For the MS2 coat protein CDS, we digested the CDS amplicon and pFTV1 vector backbone with XbaI and NotI-HF (New England Biolabs). For both the riboswitches and CDS, we ligated the digested inserts with digested backbone using T4 DNA ligase (New England Biolabs), and heat-shock transformed the ligation product into chemically competent DH10B. We then performed Sanger sequencing to verify that the insert had been cloned correctly.