NEW YORK (GenomeWeb) – Broad Institute researchers David Liu and Feng Zhang have both developed new CRISPR-based systems, one for editing point mutations in the genome and the other for editing RNA, they revealed today in separate studies.
Liu's work, published in Nature, uses a guide RNA and catalytically impaired CRISPR-Cas9 to convert A-T base pairs to G-C base pairs in the genome, enabling the editing of single point mutations without the induction of double-stranded DNA breaks (DSBs).
The study builds on previous work done in Liu's lab in which researchers developed third-generation base editors called BE3s that consist of a catalytically impaired CRISPR-Cas9 mutant that cannot make DSBs, a single-strand-specific cytidine deaminase that converts C to uracil in the single-stranded DNA bubble created by Cas9, a uracil glycosylase inhibitor that impedes uracil excision and downstream processes that decrease base editing efficiency and product purity, and nickase activity to nick the non-edited DNA strand, directing cellular mismatch repair to replace the G-containing DNA strand.
In a press conference call about the study, Liu said this kind of editing enables the efficient and permanent conversion of C-G base pairs to T-A base pairs, and addresses point mutations found in roughly 15 percent of genetic diseases.
However, the spontaneous deamination of cytosine and 5-methylcytosine in DNA is a major source of de novo mutations, and about half of known pathogenic SNPs are C-G to T-A transitions, Liu and his coauthors at the Broad and Harvard University wrote in their paper. Therefore, converting A-T base pairs to G-C bases could help treat many more genetic diseases.
The deamination of adenine results in inosine, which is treated as guanine by DNA and RNA polymerases. However, there are no naturally occurring enzymes that deaminate adenine in DNA, so the researchers' first task was to directly engineer such an enzyme. They engineered E. coli TadA tRNA adenosine deaminase (TadA) through seven rounds of bacterial evolution to accept DNA as a substrate when fused to a catalytically impaired CRISPR-Cas9.
What resulted was an Adenine Base Editor (ABE). It transforms A to inosine, which is read as G. Similar to the BE3 base editor, the ABE also contains a nickase which causes the non-edited DNA strand to replace the T with a C. The CRISPR-Cas9 contained in the system binds to the guide RNA, and guides the ABE to the editing site — however, it has been impaired so that it can no longer induce DSBs.
"Extensive evolution and engineering to maximize ABE efficiency and sequence generality resulted in seventh-generation adenine base editors, such as ABE7.10, that convert target A-T to G-C base pairs efficiently (averaging 53 percent across 17 genomic sites in human cells) with very high product purity (typically more than 99 percent) and very low rates of indels comparable to those of untreated cells," the authors wrote. "We show that ABE7 variants introduce point mutations much more efficiently and cleanly than a current Cas9 nuclease-mediated [homology-directed repair] method, induce less off-target genome modification than Cas9 nuclease, and can be used both to correct disease-associated SNPs, and to introduce disease-suppressing SNPs in cultured human cells."
And while the average editing efficiency at the 17 sites was about 53 percent, it ranged from 34 percent to 68 percent across all sites and exceeded 50 percent at 11 sites, the team added.
Importantly, the action of the ABEs resulted in very clean edits, with an average of less than 0.1 percent indels, which is similar to untreated control cells. The authors also did not observe any A to non-G editing above that of untreated cells among the 17 genomic sites tested. "In retrospect, it's tempting to speculate that because nature is not known to have an enzyme that can convert A to inosine in DNA, perhaps the cell did not evolve effective ways to remove inosine from DNA once ABE makes the conversion of A to inosine, thereby resulting in a minimum of these undesired by-products," Liu said during the press conference. "In other words, the cleanliness of ABE might be an unexpected upside to the challenge of not having a natural enzyme that converts A to inosine in DNA."
They also tested the off-target activity of the ABEs and observed off-target base editing at only four of 12 known Cas9 off-target sites (33 percent) as compared to detectable modifications by Cas9 nuclease at nine of the 12 known off-target loci (75 percent).
"Moreover, the nine confirmed Cas9 off-target loci were modified with an average efficiency of 14 percent indels, while the four confirmed ABE off-target loci were modified with an average of only 1.3 percent A-T to G-C mutation," the authors added. "We note that although seven of the nine confirmed Cas9 off-target loci contained at least one A within the ABE activity window, three of these seven off-target loci were not detectably edited by ABE7.8, 7.9, or 7.10. Together, these data suggest that ABE7 variants may be less prone to off-target genome modification than Cas9 nuclease, even for off-target sites containing editable As."
Finally, the researchers tested the potential of ABEs in human cell lines to both correct a mutation that causes hereditary hemochromatosis and to introduce a disease-suppressing mutation that protects against the effects of sickle cell anemia. While further work is needed, they noted, the results from these experiments suggested that ABEs would be a viable editing tool for patients with these conditions.
During his press conference, Liu noted that his lab is collaborating with several other groups to use base editing to study or validate possible treatments for blood diseases, genetic deafness and blindness, and some neurological disorders.
He also said that the base editor can be interfaced with a variety of Cas9 homologs, both naturally occurring and engineered, in order to bind to different sets of DNA targets. "The more different versions of Cas9 you can integrate into a base editor, the broader and broader the potential reach of your base editor in terms of what fraction of the genome you can edit," he said.
However, he added, these would have to be applied separately, and not multiplexed, as the various Cas9 homologs each bind to different PAM sequences.
REPAIRing RNA nucleotides
Meanwhile in Science, Zhang and his coauthors at the Massachusetts Institute of Technology and Harvard Medical School described their new CRISPR-based RNA editing system, called RNA Editing for Programmable A to I Replacement (REPAIR), which allows for the temporary repair of single RNA nucleotides in mammalian cells without permanently altering the genome.
While DNA editing holds promise for treating genetic disease, temporary editing of RNA has potential for treating diseases caused by temporary changes in cell state. In their study, the researchers profiled Type VI CRISPR systems to engineer a Cas13 ortholog capable of robust knockdown, and used catalytically-inactive Cas13 (dCas13) to direct adenosine-to-inosine deaminase activity by ADAR2 to transcripts in mammalian cells.
To demonstrate the broad applicability of the REPAIR system for RNA editing in mammalian cells, the team designed guides against two disease-relevant mutations: 878G>A in X-linked nephrogenic diabetes insipidus and 1517G>A in Fanconi anemia. They transfected expression constructs for cDNA of genes carrying these mutations into HEK293FT cells and tested whether REPAIR could correct the mutations, and found that they were able to achieve 35 percent correction in the first and 23 percent correction in the second.
They then tested the ability of REPAIR to correct 34 different disease-relevant G-to-A mutations and found that they were able to achieve significant editing at 33 sites with up to 28 percent editing efficiency. "The mutations we chose are only a fraction of the pathogenic G-to-A mutations (5,739) in the ClinVar database, which also includes an additional 11,943 G-to-A variants," the authors wrote. "Because there are no sequence constraints, REPAIRv1 is capable of potentially editing all these disease-relevant mutations, especially given that we observed editing regardless of the target motif."
The team then sought to improve the specificity of REPAIR to reduce the number of off-target edits, so it employed structure-guided protein engineering of ADAR2DD. The resulting high-specificity variant system, called REPAIRv2, was 919 times more specific than REPAIRv1, the researchers said.
"We targeted REPAIRv2 to endogenous genes to test if the specificity-enhancing mutations reduced nearby edits in target transcripts while maintaining high-efficiency on-target editing. For guides targeting either KRAS or PPIB, we found that REPAIRv2 had no detectable off-target edits, unlike REPAIRv1, and could effectively edit the on-target adenosine at efficiencies of 27.1 percent (KRAS) or 13 percent (PPIB)," the authors wrote. "This specificity extended to additional target sites…. Overall, REPAIRv2 eliminated off-targets in duplexed regions around the edited adenosine and showed dramatically enhanced transcriptome-wide specificity."
For now, the researchers added, REPAIR can convert adenosine to inosine. However, they hypothesized that additional fusions of dCas13 with other catalytic RNA editing domains, such as APOBEC, could enable cytidine-to-uridine editing, and that mutagenesis of ADAR could relax the substrate preference to target cytidine, allowing for the enhanced specificity conferred by the duplexed RNA substrate requirement to be exploited by C-to-U editors.
In the press conference for his Nature paper, Liu called Zhang's work "an impressive and exciting development," and said that he's hopeful that DNA base editing and RNA base editing can be used together as complementary tools to address a broad range of research and therapeutic applications.