A modular XNAzyme cleaves long, structured RNAs under physiological conditions and enables allele-specific gene silencing

Abstract

Nucleic acid catalysts (ribozymes, DNA- and XNAzymes) cleave target (m)RNAs with high specificity but have shown limited efficacy in clinical application. Here we report on the in vitro evolution and engineering of a highly specific modular RNA endonuclease XNAzyme, FR6_1, composed of 2’-deoxy-2’-fluoro-β-D-arabino nucleic acid (FANA). FR6_1 overcomes activity limitations of previous DNA- and XNAzymes and can be retargeted to cleave highly structured full-length (>5 kb) BRAF and KRAS mRNAs at physiological Mg 2+ concentrations with allelic selectivity for tumour-associated ( BRAF V600E and KRAS G12D) mutations. Phosphorothioate-FANA modification enhances FR6_1 biostability and enables rapid KRAS mRNA knockdown in cultured human adenocarcinoma cells with a G12D-allele-specific component provided by in vivo XNAzyme cleavage activity. These results provide a starting point for the development of improved gene silencing agents based on FANA or other XNA chemistries.

XNAzymes targeting the SARS-CoV-2 genome inhibit viral infection

“…Next, we sought to combine SARS-CoV-2 XNAzymes in order to target the genome synergistically and to stabilise catalysts prior to examining activity in vivo. Previously, we found that a cocktail of three FR6_1derived XNAzymes was able to cleave a long, structured RNA target at expected sites without mutual interference 31 . However, as FANA displays only modest resistance to serum exonucleases, further modifications (FANA phosphorothioate linkages) had to be introduced to improve biostability.…”

“…The modular RNA endonuclease XNAzyme FR6_1 was initially selected to cleave an RNA sequence in the genome of Zaire Ebolavirus ('Sub_-Ebo') and is predicted to have an architecture comprised of a catalytic core flanked by two guide strands or substrate-binding 'arms' 31 . Substrate specificity arises from a combination of the complementarity of the binding arms for target RNA, as well as putatively unpaired substrate residues positioned in a three-residue bulge opposite the catalytic core-although FR6_1 tolerates almost any dinucleotide pair (except GG) at the cleavage site in this bulge, there is an absolute requirement for an A one nucleotide downstream 31 . Initially, we searched the RNA genome of SARS-CoV-2 (NCBI NC_045512.2) for sequences with similarity to the original target sequence of FR6_1, but with no similar (i.e.…”

“…1a), a screen of mutations in the XNAzyme core (Supplementary Fig. 1b, c) previously found to be beneficial for re-targeting FR6_1 to KRAS mRNA 31 , enabled the discovery of an improved ORF1b-targeting XNAzyme, 'Fz_CoV2_1b [A28G]' (Supplementary Fig. 1d).…”

“…5b). We have previously shown that it is critical to remove XNAzymes from RNA preparations when assessing RNA cleavage by RT-PCR as catalytic activity can occur during typical workups, as well as inhibition of target site RNA reverse transcription, yielding two sources of 'false positive' knockdown when measuring in vivo activity 31 . We, therefore, verified that our protocol for removal of FANAzymes also removed the TFz 3 nanostructure by performing control assays after the addition of TFz 3 and indeed observed no knockdown when the initial incubation step was omitted (Supplementary Fig.…”

“…'X10-23' 27 , which has also been used to target SARS-CoV-2 RNA as the basis of a viral biosensor 28 ) or the de novo evolution of fullymodified XNAzymes 29,30 . Recently, we described a modular XNAzyme, 'FR6_1', composed of 2′-deoxy-2′-fluoro-β-D-arabino nucleic acid (FANA), originally selected to cleave a sequence in the RNA genome of the Zaire Ebolavirus, which has the ability to cleave long (>5 kb), structured RNA under physiological conditions in vitro and in vivo 31 .…”