Geneart strings

The sequences coding for the protein portions selected to be included in the fusion antigen were joined contiguously, respecting the open reading frames of interest. Sequences coding for a Flag-Tag and a His-Tag were added to the 3′ end, followed by the introduction of a stop codon. Correct translation was confirmed using the ExPASy translate tool (https://web.expasy.org/translate/), followed by codon optimization to a mammalian codon usage bias. The originated sequence was synthetized (GeneArt Strings—Thermo Fischer Scientific, MA, USA) and cloned into the VR2001-TOPO vector as previously described^{29 (link)}. The VR-2001-SfSPChimera plasmid was then purified as previously reported^{29 (link)}, and used either as a DNA vaccine or as a vector in the attempt of protein expression using a mammalian system, described elsewhere^{25 (link)}.
Additional in-silico determinations were performed now using the multiepitope antigen protein sequence as template. The prediction of antigenicity probability (ANTIGENpro), solubility upon overexpression (SOLpro), formation of disulphide bonds (DIpro) and presence of transmembrane domains (ABTMpro) were performed using the SCRATCH protein predictor online tool (https://scratch.proteomics.ics.uci.edu/). The prediction of allergenic potential was performed using the AlgPred online tool (https://crdd.osdd.net/raghava/algpred/)^{90 (link)}.

Pseudotyped lentiviruses displaying influenza HAs were produced by transfection of HEK 293 T/17 cells (ECCAC, Public Health England, UK) with 1.0 μg of gag/pol construct, p8.91, 1.5 μg of a luciferase reporter carrying construct, pCSFLW, 250 μg of TMPRSS4-expressing construct and 1.0 μg of HA glycoprotein-expressing construct^{12 (link)}. Transfections were performed in 10 ml of media DMEM 10% FCS, 1% penicillin–streptomycin, 20% L-glutamate and left for 8 hours. One unit of endogenous NA (Sigma, USA) was added to 10 ml of new media to induce virus budding. Media was removed 48 hours post the induction of budding and filtered with a 0.45-μm syringe. The pseudotyped influenza viruses were stored at −80 °C.
The strain accession numbers used for production of the pseudotyped lentiviruses are A/California/04/2009 (AEE69009), A/Solomon Islands/3/2006 (ABU99109), A/USSR/90/1977 (AAA43240), A/Denver/1957 (ABD15258), A/Iowa/1943 (ABO38373), A/PR/8/1934 (CCH23213), A/WSN/1933 (ACF54598) and A/South Carolina/1/1918 (AAC57065).
The p8.91, pCSFLW and TMPRSS4-expressing construct were gifts from Dr Nigel Temperton, whilst the HA-expressing plasmids were either gifts from Dr. Temperton or produced through the cloning of GeneArt Strings (Thermo Fischer Scientific, USA) into the pI.18 expression vector (also a gift from Dr. Temperton). All plasmids are available on request.

cDNA encoding 11S^{17 (link)}, DARPin 9.29^{36 (link)}, DARPin G3^{36 (link)}, VH and VL of trastuzumab^{35 (link)}, VH and VL of pertuzumab^{34 (link)}, and 73JFab^{43 (link)} were purchased as GeneArt Strings from ThermoFisher Scientific. The VH and VL of trastuzumab and pertuzumab were then appended onto CL and CH1 using the primers indicated in the primer table included in the supplemental material (Supplemental Table S4). The linker, His tag, and β9/β10 peptide were added to the sequences through PCR using primers listed in Supplemental Table S4. Each cDNA was inserted into pF1K T7 (Promega) at the PvuI and XbaI restriction enzyme sites. Δ11S was created by PCR amplifying all except the β9 sequence of 11S and inserting the resulting amplicon into pF1K T7 with PvuI/XbaI. β10Δ11S was created by addition of the β10 sequence onto the forward primer used to amplify Δ11S and inserting the amplicon into pF1K T7 with PvuI/XbaI. Standard genetic engineering protocols were used for all cloning, and each clone was sequence verified.

To obtain recombinant RNF213, we synthesized codon-optimized DNA fragments for insect cell expression. We focused our analysis on the physiologically most relevant mouse RNF213 isoform (RefSeq #NP_001035094.2), adding a Gly₃-His₁₀ tag to the C-terminus. The 7 cDNA fragments obtained as GeneArt Strings (Thermo Fisher Scientific) were assembled via Gibson Assembly into the pFastBAC1 vector. The final expression vector was confirmed to contain the target sequence by Sanger sequencing. Mutants were generated by splitting the full-length DNA into the separate cloning vectors, performing site-directed mutagenesis, and reassembling the expression vector by Golden Gate Assembly. The mutated cloning sites of the final vectors were validated by Sanger sequencing.

Isogenic mutants of P. gingivalis were generated, using a DNA construct obtained either through overlap extension PCR or synthesized commercially through gene synthesis (GeneArt^® Strings; ThermoFisher Scientific). Overlap extension PCR products were created through PCR amplification of ~500 bp genomic fragments upstream and downstream of the gene to be deleted and fused to the ermF marker through PCR, as previously detailed by (Kuwayama et al., 2002) and using primers described in Table 1 where the first codon of ermF replaces the native codon, thus ensuring expression of the antibiotic cassette and reducing chances of any polar effects on downstream gene expression. DNA constructs that were synthesised were designed in the same fashion, with the ermF marker flanked by the 500 bp upstream and downstream regions. Both synthetic constructs and PCR products were blunt‐end cloned into pJET1.2 (ThermoFisher Scientific) according to manufacturer's instructions. DNA constructs were introduced into P. gingivalis through the natural competence method as described by Tribble et al., (2012), and successful transformants selected on erythromycin (10 μg ml⁻¹) containing BA plates. Mutants were confirmed by PCR of extracted genomic DNA (Promega Wizard Genomic DNA), with PCR products sequenced at GATC Biotech to establish insertion of ermF at the expected position.