Ambryon Biosciences

Engineered CRISPR-OsCas12f1 and RhCas12f1 with robust activities and expanded target range for genome editing

Kong et al. (2023). Nat Commun·Citations: 104CorrectedResearch

Corrected. A correction to this paper has been published. Read the notice.

Cas12f's small size suits single-AAV delivery, but known Cas12f1 nucleases were weak and PAM-restricted; arginine substitutions plus C-G sgRNA stem swaps yielded enOsCas12f1 and enRhCas12f1. One AAV9 of enOsCas12f1 restored dystrophin in a DMD mouse.

Key findings

  • Six uncharacterized Cas12f1 proteins are active in mammalian cells; four (OsCas12f1, HsCas12f1, Cb1Cas12f1, RhCas12f1) activate an EGFP reporter robustly but edit genomic loci modestly (1% to 20% indels). OsCas12f1 (433 aa) and RhCas12f1 (415 aa) are most active.

  • enOsCas12f1 and enRhCas12f1 repair outcomes are dominated by deletions that alter the protospacer rather than insertions, and the centre of the deletion sits in the PAM-distal region outside the protospacer — the same pattern seen with Un1Cas12f1 and AsCas12f1.

  • enOsCas12f1 tolerates no single mismatch at position 3, 5 or 11 and is impaired by adjacent double mismatches in positions 1 to 16, whereas enRhCas12f1 partly tolerates PAM-distal mismatches (positions 19 and 20) but PAM-proximal mismatches sharply reduce activity.

  • enOsCas12f1 showed on-target efficiency comparable to LbCas12a and low off-target editing, except a low signal at CLIC4 OT7; PEM-seq translocation at target 36 was 7.03% for enOsCas12f1 versus 8.44% Un1Cas12f1_ge4.1, 9.22% LbCas12a, 8.19% SpCas9.

  • DD fusion makes enOsCas12f1 trimethoprim-regulated; a single AAV9 in humanized DMD mouse tibialis anterior deletes exon 51, giving 22.7 ± 9.2% exon-51-deleted transcripts and dystrophin in 11.6 ± 4.0% of myofibers (DD version: 15.0 ± 7.0%, 7.6 ± 2.4%).

  • Catalytically dead enOsCas12f1 carries epigenetic effectors compactly: miniCRISPRoff (1444 aa versus CRISPRoff's 2361 aa) silences GFP in three of four versions tested and heavily methylates the Snrp promoter, and a VPR fusion robustly activates GFP in TRE3G-GFP reporter cells.

  • OsCas12f1 and RhCas12f1 recognize complementary PAMs, 5′-YTTH (Y = C or T, H = not G) and 5′-NCCD (D = not C) respectively, in vitro with a 7-bp random library; a 20-nt spacer is optimal for both.

  • Cleavage by OsCas12f1 requires the conserved RuvC active-site aspartates: either D228A or D406A abolishes its activity. The equivalent substitutions in RhCas12f1, D210A or D388A, likewise inactivate that nuclease.

  • Linear-plasmid assays suggest both nucleases cleave double-stranded DNA across a wide temperature range preferring 37 °C to 50 °C, and size-exclusion chromatography suggests both could form dimers with sgRNA, at least in the tested condition.

  • Arginine substitutions raise the activity of a subset of OsCas12f1 variants, presumably via stronger contacts: D52R was best of over 100 singles (1.31-fold), arginine at position 52 appeared better or slightly better, and T132R/D52R was the most active double mutant.

  • Truncating OsCas12f1 sgRNA stem base pairs impaired cleavage, while A-U to C-G substitutions in tracrRNA stem 1, which may increase sgRNA stability, gave 1.56-fold over wild type with Os-sg1.1, and a second round found Os-sg2.6 best.

  • Protein and sgRNA engineering of Cas12f1 are additive: T132R/D52R protein with Os-sg2.6 sgRNA gives enOsCas12f1, exceeding either component alone and reaching 9.4-fold wild-type OsCas12f1 activity at DMD; enRhCas12f1 combines RhCas12f1 variants with Rh-sg1.1.

  • enOsCas12f1 is active at 5′-NTTN sites (over 10% indels at 12 of 12 TTC, 7 of 9 TTA, 9 of 11 TTT, 4 of 11 TTG sites) and enRhCas12f1 at 5′-CCD sites, whereas Un1Cas12f1_ge4.1 stays largely confined to 5′-TTTR.

  • enOsCas12f1 outperforms Un1Cas12f1_ge4.1 at endogenous loci in PCSK9, TTR and VEGFA: over 1% indels at all 30 sites (maximum 96.2%) versus 22 of 27 (maximum 60.6%), with 54.7 ± 29.8% versus 14.0 ± 18.1% average efficiency, 3.9-fold higher.

Author abstract

The type V-F CRISPR-Cas12f system is a strong candidate for therapeutic applications due to the compact size of the Cas12f proteins. In this work, we identify six uncharacterized Cas12f1 proteins with nuclease activity in mammalian cells from assembled bacterial genomes. Among them, OsCas12f1 (433 aa) from Oscillibacter sp. and RhCas12f1 (415 aa) from Ruminiclostridium herbifermentans, which respectively target 5' T-rich Protospacer Adjacent Motifs (PAMs) and 5' C-rich PAMs, show the highest editing activity. Through protein and sgRNA engineering, we generate enhanced OsCas12f1 (enOsCas12f1) and enRhCas12f1 variants, with 5'-TTN and 5'-CCD (D = not C) PAMs respectively, exhibiting much higher editing efficiency and broader PAMs, compared with the engineered variant Un1Cas12f1 (Un1Cas12f1_ge4.1). Furthermore, by fusing the destabilized domain with enOsCas12f1, we generate inducible-enOsCas12f1 and demonstate its activity in vivo by single adeno-associated virus delivery. Finally, dead enOsCas12f1-based epigenetic editing and gene activation can also be achieved in mammalian cells. This study thus provides compact gene editing tools for basic research with remarkable promise for therapeutic applications.

Identifiers & revision