Fig. 4: Model-based design of shorter 5’UTRs for gene editing mRNA therapeutics.

(A) Top: schematic of mRNA vector, with a 25nt-long variable 5’UTR segment as in Fig. 3. Bottom: Absolute editing efficiencies for mRNAs with a megaTAL targeting the TGFBR2 gene, for 21 different 5’UTR including designs and controls. Each group of four bars represents one 5’UTR sequence transfected at four mRNA doses (0.25, 0.5, 1, or 2 pmol mRNA). Two biological replicates were performed per 5’UTR and mRNA dosage, and are represented by individual markers. Colors represent the source in the case of controls or the design method. Editing efficiencies for the first No VAE Fast SeqProp design and the second +VAE DEN design were close to zero only at a dosage of 0.25 pmol mRNA, and were deemed to be the result of experimental error and excluded from subsequent analysis. (B) Editing efficiencies normalized to the Strong Kozak control at the corresponding mRNA dosage. (C) Absolute editing efficiency as a function of mRNA dosage for a few selected 5’UTRs indicated with a vertical arrow in (B). (D) Comparison of Kozak-normalized editing efficiencies when using a megaTAL targeting the TGFBR2 gene vs. the PDCD1 gene. (E) Comparison of Kozak-normalized editing efficiencies for the TGFBR2 megaTAL in K562 cells or HepG2. In (D) and (E), each marker and error bar represent the mean and standard deviation of the Kozak-normalized editing efficiencies ((B) and the bottom panels of Supplementary Fig. 23 and Supplementary Fig. 24) across all mRNA dosages for a particular 5’UTR (n = 8 for most 5’UTRs: 2 biological replicates and 4 dosages per replicate. The 0.25 pmol dosage was excluded for two sequences as described above, therefore n = 6 for these 5’UTRs only). Data from designs with the fixed-end 50nt architecture (Fig. 2 and relevant Supplementary Figs.) are also included in (D) and (E). Source data are provided as a Source Data file.