NEW YORK – Broken String Biosciences, a Cardiff University spinout that offers a sequencing-based assay for the effects of genome editing, is looking to use new investment to help commercialize its technology and branch out into new markets.
Last month, the Cambridge, UK-based firm, announced it had raised $15 million in Series A financing. It will use the funds to build a service offering for its customers in the genome editing market that use its assay called Induce-seq.
"It's not just an off-target discovery tool," said Simon Reed, cofounder and CSO of Broken String. "That's the key question most collaborators want to know about, but in addition, they want to understand what else they can extract to improve the editing process. … What we're talking about is providing information that expands the editable genome."
Already, the firm has collaborated with AstraZeneca, Novartis, and TwinStrand Biosciences on a study of off-target editing in cell therapies.
Broken String is also exploring other applications for its tech. "At its core, it's really a technology that allows precise measurement of DNA breaks," said Cofounder and CEO Felix Dobbs. "The gene editing space has real need in the near term, but we're now making good progress in establishing other applications."
That includes risk assessment in the environmental and toxicology spaces, analysis of DNA damage response in oncology, and even prognostic and diagnostic applications in cancer.
Broken String was founded in 2020 by Reed — also a professor at Cardiff University — Dobbs, and Patrick van Eijk, both former lab members, as well as Simon Kerr, chair of the firm's board. Dobbs had come to the lab on a doctorate project sponsored by AstraZeneca to address the challenge of measuring the off-target effects of genome editing.
Their solution, Induce-seq, is a PCR-free library preparation method that labels DNA breaks with special sequencing adapters, producing an unbiased map of breaks. The method essentially uses an Illumina flow cell as a hybridization-based enrichment panel. "Every sequencing read corresponds to a single labeled break," Dobbs said.
"When we view those breaks in a genome browser, you can see immediately that there's a diagnostic pattern that exists in a gene-edited sample," which is different from endogenous or on-target breaks, Reed added.
The team formed Broken String in 2020 and were selected for the Illumina Accelerator program in 2021. "It really did accelerate the company," Dobbs said, bringing connections to investors and allowing the firm to get commercial traction.
A $4.3 million seed financing round in September 2021 helped move the firm into the Wellcome Genome Campus. Now, with 14 employees, the firm also has a second facility nearby. It plans to roughly double by June of next year. "We want to get to 30 soon," Dobbs said.
In July 2022, the firm's trio of scientific cofounders published a paper in Nature Communications describing Induce-seq. Their workflow introduces full-length Illumina P5 adapters into break sites in situ, followed by fragmentation and then library preparation with "half-functional" P7 adapters, where one side of the double-stranded oligo is clipped. This enables the method to use the Illumina flow cell as a hybridization-based capture device, enriching only for fragments that come from either side of the break, thereby reducing the need for amplification during library prep, which introduces noise and misrepresents the frequency of breaks.
"This means that whenever multiple reads are detected at the same position, they are derived from different cells," the authors wrote. "For this reason, it is now possible to normalize and quantify double-stranded breaks based on cell number."
So far, the AstraZeneca and Novartis work is the most public use of Induce-seq; however, those results have not yet been published. According to Reed, Broken String received gene-edited samples from both pharmaceutical firms, featuring about half a dozen CRISPR-guided edits in each of two different cell types. Broken String reported off-target and on-target edits, then TwinStrand's duplex sequencing method was used to look closer at the mutational outcomes.
Some customers are using Induce-seq to help determine which editing system to use. "If you have multiple guides targeting the same gene, it will reveal the specific profile of each and allow [customers] to make decisions on which to take forward," Dobbs said. A company may then choose between several effective editing systems to choose the one that appears least risky. "Our tech allows those key decisions to be taken," he said.
This includes novel gene-editing systems that expand the ways in which genomes can be edited. "At the moment, we're correcting single-gene mutations" in certain therapies, Reed said. "The move to more complex genetic disorders will require a whole suite of tools that will allow for more complex editing, which is also a part of what Broken String is seeking to achieve."
In the paper, the Broken String team noted that their method "can be adapted for the detection of a range of other genomic features that can be end-labeled in this way," including genome-wide mutations, single-strand breaks and gaps, and "other types of DNA damage that can be converted into breaks and subsequently ligated using this combination of sequencing adapters."
Another application that Broken String now plans to explore is analyzing the DNA damage response in the context of cancer therapy, following the path laid out by the use of PARP inhibitors. "The idea is there are interactions between genetics and exposure of certain chemicals to enhance killing," Reed said. "How can we identify molecules that, in the right context, the right cancer and treatment, change the tumor so that it responds?"
Genotoxicity testing is adjacent to the use of Induce-seq for characterizing cell therapies and could also eventually be a market for Broken String. Historically, this has been done with cell-based assays, Reed said, but new materials including nanomaterials "have very different features from small molecules. They really need a different way of testing."
"There is no question that [Induce-seq] is applicable in this context," he said. "We're able to detect, very sensitively, exposures of individuals to certain chemicals, which interestingly have their own signatures."