Impact of oligonucleotide chemistry and silencing mechanism on applications in gene silencing and genome editing

Abstract

Endogenous gene regulation is an essential tool for the survival of all organisms. In recent years, scientists have gained valuable mechanistic insight into these endogenous gene regulation pathways. Using synthetic oligonucleotides (chemically modified DNA or RNA), we have been able to tap into these pathways and manipulate gene expression.

The development of techniques such as solid phase synthesis have enabled researchers to make custom-made oligonucleotides quickly and efficiently. Although solid phase synthesis is a routine method of obtaining oligonucleotides, more complicated oligonucleotides, e.g. chemically modified or longer RNAs, require optimization of the synthesis cycle and deprotection methodology. In order to synthesize complex oligonucleotides in high yield and purity, we tested several synthesis reagents, cycles, deprotection conditions, and purification methods. Using these optimized conditions, we have successfully synthesized a variety of oligonucleotides for gene silencing and genome editing applications. For synthesis of locked nucleic acid (LNA)-containing phosphorothioate oligonucleotides, tetraethylthiuram disulphide (TETD) is insufficiently reactive but 3-Ethoyx-1,2,4-dithiazoline-5-one (EDITH) gives excellent results.

ADAM33 is a susceptibility gene for asthma and bronchial hyperresponsiveness and is implicated in airway remodeling, but its function is only partially understood. Oligonucleotide-mediated gene silencing of ADAM33 could provide valuable insight to its biology and allow us to observe airway development under low ADAM33 expression levels. We show potent silencing of ADAM33 in MRC-5 lung fibroblasts using four different classes of oligonucleotides: siRNAs, single-stranded siRNAs, LNA gapmers, and novel conjugates of antisense oligonucleotides. We observed that several LNA gapmers showed subnanomolar potency when transfected with a cationic lipid, and low micromolar potency when delivered gymnotically. Also, we observed that RNase H-dependent antisense oligonucleotides greatly outperformed RISC-dependent oligonucleotides for silencing ADAM33. As ADAM33 mRNA is 95% retained in the nucleus, this work is consistent with recent findings that antisense oligonucleotides are often more potent against nuclear-localised transcripts.

Single-stranded siRNAs (ss-siRNAs) are chemically modified single stranded oligonucleotides that engage the RISC complex, they have the potential to combine the advantages of both duplex siRNAs and antisense oligonucleotides. One disadvantage of the published single-stranded siRNA chemical modification scheme is the use of 2’-O-methoxyethyl-RNA at the 3’ terminus. This modification is not available to most researchers so the use of the ss-siRNA technology has been limited up to the present. During our ADAM33 work, we observed that making small changes to the 3’ terminus of our single-stranded siRNAs could greatly improve the potency of the oligonucleotide. We found that replacing the 3’ terminal 2’-O-methoxyethyl residues with the commercially available 2’-O-methyl or LNA modifications actually improved the potency of the single-stranded siRNA against ADAM33. We developed and optimized single-standed siRNAs based on four additional active siRNA duplex sequences targeting different genes within mammalian cells. In one additional gene, PR, we were able to support our ADAM33 findings that single-stranded siRNAs show improved potency with 2’-O-methyl or LNA modification at the 3’ terminus. However, the single-stranded siRNAs against three target genes in two HEK-293 cell lines failed to show any gene silencing activity with any 3’ terminus modification. The failure of the oligonucleotides could be related to the specific cell lines used for the experiments as we also observed increased toxicity with our single-stranded siRNAs in these cells compared to their parent siRNA duplex.

Additionally, as chemically modified oligonucleotides can be used to provide greater affinity and specificity to their target sequence, we wanted to explore whether chemical modifications can improve DNA cleavage activity and specificity for genome editing applications. For this work, we used the clustered regularly interspaced short palindromic repeats/crispr associated (CRISPR/Cas9) type II system, which is commonly used in genome editing applications as it requires only two guide RNAs (crRNA, tracrRNA) and a single protein (Cas9 nuclease) for cleavage of a target dsDNA. We wanted to explore whether the use of chemically modified crRNA could improve the specificity or efficiency of cleavage of target DNA when compared to an unmodified crRNA. However, as no work has been published on chemically modified crRNAs, it was unknown whether the CRISPR/Cas system could tolerate chemical modifications and retain cleavage ability. Using a variety of chemically modified crRNAs, we were successfully able to obtain cleavage of our target DNA sequence, although none of our chemically modified crRNAs were able to match the cleavage efficiency of our unmodified crRNA.

More information

Identifiers

Catalogue record

Export record