The Point of Base Editors: Correcting Point Mutations

Specific and precise base changes in genomic DNA and in RNA offer hope to the millions of people suffering genetic diseases associated with point mutations

Some genome editing systems are highly conspicuous. They introduce double-strand breaks to DNA that attract the attention of cellular mechanisms such as nonhomologous end joining and homology-directed repair. If a genome editing system is so brash as to attempt a sizable insertion of new DNA, homology-directed repair must ensure that the insertion succeeds.

Not all genome editing systems make waves. Base editing systems, for example, scarcely make ripples. These systems gently swap a single base for another in just one strand of DNA, leaving a mere nick in the other strand, a nick that is easily filled in via mismatch repair.

As specific as they are tidy, base editing systems spurn excess and embrace adequacy. They do just enough to correct harmful single base mutations. If that sounds modest, keep in mind that single-base mutations account for hundreds of life-threatening disorders. One of these disorders is sickle-cell disease, which affects approximately 100,000 individuals in the United States alone. Altogether, the disorders associated with single-base mutations affect millions.

The potential for base editing systems to treat genetic disorders was discussed at a recent event, the 2nd International Conference on Base Editing—Enzymes and Applications. As this article relates, the event’s speakers, gifted scientists all, aren’t limiting their work to the correction of DNA mutations. Some scientists are using base editors to revise RNA, reducing the risk of irreversible genomic alterations.

Base editor basics

Base editors were first developed in 2016 and 2017 at the laboratory led by David Liu, PhD, at Harvard University. They directly convert one DNA base pair at a specified position in the genome of a living cell into a different base pair, without cutting through the DNA double helix. Earlier gene editing methods—such as those relying on zinc finger nucleases, transcription activator–like effector-based nucleases, and CRISPR-Cas9—primarily disrupt, rather than precisely edit, genes in most cell types.

Each base editing target requires a thoughtful analysis of the target site against the available base editors. Currently, there are over 100. A publicly available machine learning model, BE-Hive, developed by the Liu group, helps predict which base editor will give the best results for a particular edit of interest.

Encouraging preclinical results

Progeria is an infamous, progressive, and fatal genetic disease characterized by rapid aging. Children with progeria quickly show characteristics of elderly people and typically die around age 14. No treatment has yet proven to substantially extend their lifespan, although a recently approved small-molecule drug, lonafarnib, can increase lifespan on average by approximately 2.5 years.

In 2017, Luke W. Koblan, then a Liu graduate student, realized that the mutation that causes progeria in 95% of cases was one that in theory could be corrected with the adenine base editor that was being developed in the Liu laboratory. The most common cause of progeria is a single C-to-T mutation in one position of LMNA, the gene that encodes the protein lamin A.

“In our recent paper (Koblan et al. Nature 2021; 589: 608–614), we used CRISPR base editing to profoundly rescue several of the symptoms of the disease in an animal model of human progeria—a mouse that carries the mutated human LMNA gene,” said Liu, a core faculty member of the Broad Institute, a professor of chemistry and chemical biology at Harvard University, and an investigator at the Howard Hughes Medical Institute. “These progeria mice show many of the hallmark symptoms of progeria in children, including cardiovascular problems, other signs of rapid aging, and early death.”

A base editor was programmed to directly convert the mutation back to the normal DNA sequence by converting the disease-causing T-A base pair back to a C-G base pair. When this base editor was tested in cultured cells derived from children with progeria, very efficient (~90%) conversion of the mutation was observed, with no off-target editing detected.

Packaged into an adeno-associated virus (AAV) delivery vehicle, the base editor was administered in a single injection to treat progeria mice. “The result surpassed our expectations,” Liu recalled. “We saw substantial correction of the mutation at the DNA, RNA, and protein levels in many organs; dramatic improvement in the health of heart tissue; and an unprecedented increase in lifespan following this one-time treatment.”

Although additional work lies ahead, this study raises the possibility of treating progeria in a patient by directly fixing the root cause of the disease with a one-time base editing therapy, rather than by treating the disease symptoms.

Base editing of RNA

The ShapeTX RNAfix platform illustration
The ShapeTX RNAfix platform is a new technology for gene therapy that employs guide RNAs to recruit endogenous ADAR (adenosine deaminase acting on RNA) for site-directed RNA editing. ADAR catalyzes the deamination of adenosine into inosine, which is in turn read like guanosine by the translation and splicing machineries.

The ShapeTX RNAfix platform, a technology developed by Shape Therapeutics for gene therapy, employs guide RNAs (gRNAs) to recruit endogenous ADAR (adenosine deaminase acting on RNA) for site-directed RNA editing. ADAR catalyzes the deamination of adenosine into inosine, which is in turn read like guanosine by the translation and splicing machineries. The potential applications of ADAR-based editing are vast, from point mutation correction to tunable gene knockdown and upregulation, exon skipping, and even modulation of protein-protein interactions.

RNAfix opens the door to a wide range of therapeutic applications with potential advantages over DNA-based editing platforms:

  • First, since RNA is continuously expressed and replaced, RNA edits are transient, reducing the risk of irreversible genomic alterations.
  • Second, since RNA editing acts on many transcripts within a cell, it can have effects in almost every cell reached, whereas DNA editing can affect only those cells in which the desired genome editing event occurs, which can be a small proportion of all targeted cells.
  • Third, no foreign enzymes are delivered. Instead, the gRNA redirects an endogenous RNA-editing enzyme that is already present in the cells, thus improving the safety profile.

For targeted delivery of the gRNAs, Shape applies its AAVid platform to engineer novel AAV capsids with enhanced tissue-specific tropism, theoretically enabling one-time treatment for patients living with devastating diseases.

“We are also developing additional gene therapy approaches,” said Adrian W. Briggs, PhD, vice president, head of platform technologies, Shape. “Ultimately, our goal is to develop technologies to address as many
diseases as possible.”

For diseases caused by premature termination codon (PTC) mutations, Shape employs RNAskip, a proprietary suppressor tRNA technology that allows a PTC to be overcome through translational readthrough to produce a fully corrected wild-type protein. “There are only three possible PTC sequences,” Briggs noted. “This could enable RNAskip to be re-deployable across 20% of all genetic disorders without the need for new discovery campaigns. Rett syndrome is our current lead indication using the RNAskip technology.”

For other diseases where tailored expression of a transgene is required, the RNAswap platform applies smart-sensing vector designs to enable fine-tuned transgene expression in what is essentially “gene replacement 2.0,” according to Briggs. This is done by regulating interactions of host cell factors with host-responsive elements in promoters, untranslated regions, and coding domains.

The RNAswap discovery process employs artificial intelligence to identify critical natural and synthetic transgene elements to tailor gene expression for each cell gene and target cell type. Briggs stated that RNAswap may enable new treatment possibilities beyond the first-generation gene replacement approaches currently used by most gene therapy companies.

RNA sequencing–based assays

In the past decade, progress in sequencing technologies has facilitated the development of new types of RNA therapeutics. As translation of RNA biology begins, quantitative comparisons between disease and normal states or between drug-treated and control groups become increasingly important.

Inherently quantitative sequencing-based assays are available from Eclipse Bioinnovations. The company’s eCLIP technology produces transcriptome-wide protein-RNA binding maps with high accuracy and sensitivity. It is based on UV crosslinking of protein-RNA interactions followed by immunoprecipitation of a protein of interest.

“UV crosslinks are very stable, allowing for high-stringency washes that ensure high specificity,” explained Sergei Manakov, DPhil, associate director of bioinformatics at Eclipse. “An optimized sequencing adaptor ligation strategy explains the higher sensitivity of eCLIP compared with related CLIP technologies.”

Eclipse Bioinnovations has built a portfolio of assays based on eCLIP technology, including miR-eCLIP, an assay that can read out miRNA targets from AGO2 interactions, and m6A-eCLIP, an assay that can call m6A modifications with a single-nucleotide resolution by mapping m6A-antibody binding to RNA transcripts.

Eclipse Bioinnovations’ assay portfolio illustration
Eclipse Bioinnovations’ assay portfolio includes RBP-eCLIP (to identify RNA targets transcriptome wide); miR-eCLIP (to read out miRNA targets from AGO2 interactions); m6A-eCLIP (to call m6A modifications with single-nucleotide resolution); End-Seq (to map 5’ and 3’ transcript ends); Ribo-eCLIP (to profile mRNA translation rates transcriptome wide); and SHAPE (to achieve RNA structural probing).

The accuracy of eCLIP reveals differences in binding profiles and functions of specific isoforms of RNA-binding proteins such as the p110 and p150 isoforms of ADAR1. Manakov said that the nuclear-localized and constitutively expressed p110 isoform of ADAR1 binds and edits Alu-repeat-derived double-stranded RNA, presumably unwinding it and preventing autoimmune activation of innate double-stranded RNA sensors in the cytoplasm.

On the other hand, the p150 isoform of ADAR1 is known to be interferon inducible and to play a part in antiviral defense. In the absence of viral infection, the p150 isoform primarily binds to LINE1 retrotransposon RNA transcripts in HEK293xT and K562 cells. “This is interesting,” Manakov remarked. “LINE1 is sufficiently different from retroviruses and proposed to have specific functions related to memory and learning as well as oncogenic processes.”

The company has optimized RNA isolation and adaptor ligation, and it has developed expertise and know-how to apply to other useful assays. “This is where End-Seq and FLI-Seq fit in,” Manakov indicated. End-seq maps 5′ and 3′ transcript ends with high precision and sensitivity. The key is the ability to isolate intact RNA and to ligate sequencing adaptors to RNA ends with high efficiency. Based on a probe capture enrichment strategy, FLI-seq facilitates readouts of CRISPR knockout screens.

According to Manakov, the intent is to continue to streamline and simplify workflows, make bioinformatics analysis more accessible, and expand the assay portfolio. Examples include Ribo-eCLIP (to profile mRNA translation rates transcriptome wide) and SHAPE (for RNA structural probing).

Access to genome editing resources

As a nonprofit biological resource center, Addgene distributes plasmids that have been constructed in academic and industrial laboratories. The intent is to accelerate research and discovery by improving access to useful research materials and information.

Addgene offers the genome editing community an extensive catalog. For example, Addgene plasmids can enable genome editing with zinc finger nucleases, transcription activator–like effector-based nucleases, Cre-lox systems, and CRISPR systems. Addgene also offers help with viral vectors, gene expression tools, fluorescent and luminescent markers, and optical and chemical switches.

The repository established strong relationships with the laboratories that pioneered the plasmid-based tools for CRISPR systems, including the laboratories led by Jennifer Doudna, PhD, Feng Zhang, PhD, and David Liu, PhD. “We continue to work closely with these laboratories,” said Eric Perkins, PhD, senior director of customer experience, Addgene. “We are also now working closely with their alumni, who have gone on to lead their own academic groups and companies.”

Perkins stressed that Addgene is committed to maintaining a good relationship with this next generation of innovators, to staying active within their community, and to remaining an important resource by providing the newest plasmid tools quickly and inexpensively.

Since 2016, Addgene has served as a repository for plasmids for use in base editing, and since the first publications in 2020, the organization has distributed prime editing plasmids. Almost 700 of these base editing and prime editing plasmids are currently deposited with Addgene. “We have observed a significant demand for these plasmids since they were first developed,” Perkins noted.

Plasmids have been deposited from a number of different laboratories during the last year. Some examples include mitochondrial and CGBE (C:G to G:C) editors from the Liu laboratory; CGBE editors from the laboratory led by Wei Leong Chew, PhD, at the Genome Institute of Singapore; modified cytidine base editors from the laboratory of Dali Li, PhD, at the MD Anderson Cancer Center; and improved prime editors from the laboratory of Scot Wolfe, PhD, at the University of Massachusetts Medical School.

According to Perkins, Addgene not only provides useful tools to the scientific community, it also distributes instructional information for those tools. In addition to physical products, Addgene offers a variety of educational resources related to base editing. Blog posts are released on a regular basis, and a curated page exists on the Addgene website. The nonprofit also provides a CRISPR eBook.