Best Basic Part
BBa_K4830015: PFV RT
Among the eight alternative RTs, PFV demonstrated the highest level of performance, achieving 28% (Figure 1), 52.5% (Figure 2), and 8% (Figure 3) on the RNF2, HEK3, and EMX1 genomic loci, respectively. Another noteworthy performer was MMTV RT, which exhibited a significant editing efficiency of 67% on HEK3, and 7% and 4% on RNF2 and EMX1, respectively.
Figure 1.Screening of RT variants for prime editing activity at RNF2 target site. Bar graphs show the mean value and error bars indicate S.D. of n = 3, independent biological replicates;
Figure 2.Screening of RT variants for prime editing activity at HEK3 target site. Bar graphs show the mean value and error bars indicate S.D. of n = 3, independent biological replicates.
Figure 3.Screening of RT variants for prime editing activity at EMX1 target site.
To further assess the performance of alternative RTs, we conducted in vitro prime editing using a TLR reporter plasmid, an advanced TLR system designed in our laboratory. Cell analysis was conducted at both 24 hours and 72 hours after transfection, with 24 hours providing a more stringent selection criterion. Figures 4 and 5 illustrate that PFV and MMTV outperformed other RTs in correlation with previous results.
Figure 4. Prime editing efficiency on TLR reporter plasmid 24hr after transfection. Bar graphs show the mean value and error bars indicate S.D. of n = 3, independent biological replicates.
Figure 5. Prime editing efficiency on TLR reporter plasmid 72hr after transfection. Bar graphs show the mean value and error bars indicate S.D. of n = 3, independent biological replicates.
The consistent and robust performance of PFV RT across three distinct endogenous sites and in the TLR reporter system positions it as an excellent candidate for serving as an alternative RT in prime editing.
In a study involving the Split-PE system, various RTs sourced from both bacterial and viral origins were screened, with HFV RT (1827 bp) being one of the options considered [8]. As depicted in Figure 6 below, HFV RT exhibited notably lower (or even negligible) editing frequency on RNF2, RUNX1, and HEK sites, respectively.
Figure 6. (c) PPE frequencies of seven non-MMLV RTs tested : Human codon-optimized non-MMLV RTs tested were from human foamy virus (HFV), human endogenous retrovirus K (HERV-Kcon), lactococcal group II intron Ll.ltrB (LtrA), Thermosynechococcus elongatus group II intron (TeI4c), Methanosarcina aromaticovorans intron 5 (Ma-Int5), Geobacillus stearothermophilus GsI-IIC intron (GsI-IIC) and E. rectale (Eu.re.I2) group II intron (Marathon). n = 3, technical replicates. (d) Schematic showing the lengths of all non-MMLV RTs tested in (c) in comparison to MMLV-RT (without counting start codons) [8].
To assess the similarity between the HFV RT examined in the study and the PFV RT from our lab, we conducted a protein sequence alignment using Serial Cloner Software. The alignment results, illustrated in Figure 7, reveal a 97.44% similarity between the two sequences, with the main disparity being a truncation of 120 amino acids at the front end of the HFV. This incomplete sequence of HFV likely contributes to its lower prime editing efficiency.
Figure 7. PFV and HFV RT protein sequence alignment. Sequence one represents PFV while sequence two represents HFV.
In an attempt to replicate the study's results, we deleted 96 amino acids from the front end of PFV. As depicted in Figures 8 and 9, this led to a significant drop in efficiency, from 32.5% and 35.5% of the original full-length PFV RT to 13.5% and 5.5% with the 1-96 deletion on RNF2 and HEK3 target sites, respectively. This suggests that the deleted region plays a crucial role in the function of the RT.
Given our understanding that the RT structure contains an RNase H domain, we hypothesized that the removal of the RNase H domain in PFV will not significantly affect the gene editing efficiencies. RNase H is only for RNA template degradation and this domain is also inactive in the mutant MMLV RT of PE2. We opted to take a similar approach with the truncated MMLV RT and removed the RNase H domain of PFV (deletion 583-748). The results, as seen in Figures 8 and 9, indicated a comparable to full-length PFV level of editing efficiency.
Figure 8. Prime editing efficiency at RNF2. Bar graphs show the mean value and error bars indicate S.D. of n = 3, independent biological replicates.
Figure 9. Prime editing efficiency at HEK3. Bar graphs show the mean value and error bars indicate S.D. of n = 3, independent biological replicates.
Finally, to validate the performance of the newly identified PFV RT, we conducted a test at a therapeutic site, specifically targeting the single nucleotide polymorphism (SNP) responsible for Primary Open-Angle Glaucoma (POAG). This site was recommended for testing by a clinician after it was discovered by Gharahkhani, et al. [11], and pegRNA and ngRNA for prime editing were custom-designed in our laboratory. Figure 10 demonstrates the presence of editing at the target site, albeit at a lower efficiency.
Figure 10. Prime editing efficiency at HEK3. Bar graphs show the mean value and error bars indicate S.D. of n = 3, independent biological replicates.
Conclusion
In conclusion, we assessed eight RTs and identified PFV RT as a promising candidate, displaying noteworthy editing efficiency across three genomic loci, the TLR reporter system, and a therapeutic site. PFV RT can be developed similarly to MMLV RT by incorporating mutations to enhance processivity, structural stability, fidelity, or other pertinent characteristics relevant to prime editing. We anticipate that our findings will provide researchers with an alternative option and contribute to the continued improvement of PFV RT as has been seen with MMLV.