NEW YORK – Researchers from Stanford University and their collaborators have devised a way to effectively isolate and study large human extrachromosomal DNA (ecDNA).
Describing their approach in a study published in Nature Genetics in October, the researchers showed that the amplification-free method — which was adapted from a previously developed method for isolating bacterial chromosomal segments named Cas9-assisted targeting of chromosome segments (CRISPR-CATCH) — enabled targeted profiling of megabase-sized ecDNA from human cancer cells, allowing researchers to gain insights into their molecular heterogeneity and epigenomic landscape.
"Extrachromosomal DNA represents a really important challenge for cancer patients," said Howard Chang, a physician-scientist at Stanford University and the senior author of the study. "A lot of the most important cancer-causing genes are transcribed from extrachromosomal DNAs."
Despite their importance, ecDNA has been difficult to study due to its large size and sequence complexity. According to the Stanford team, DNA fluorescence in situ hybridization (FISH), bulk whole-genome sequencing (WGS), and exonuclease digestion of linear DNA followed by DNA amplification are currently the three main tools for researchers to analyze human ecDNA.
However, these methods come with their limitations. For instance, while DNA FISH can help detect the presence of certain targeted genes on ecDNA, it has a low throughput. Meanwhile, with bulk WGS, ecDNA and chromosomal DNA are processed together. Thus, the ecDNA sequences need to be computationally inferred and are hard to validate. In the case of exonuclease digestion followed by rolling DNA amplification, the method is hindered by ecDNA size, since larger DNA molecules are prone to breakage. Plus, the amplification step of the approach prevents epigenetic analysis of the ecDNA.
To overcome these bottlenecks, the researchers sought to use CRISPR-CATCH, a CRISPR-Cas9-based method that was previously developed for isolating bacterial chromosomal segments.
Mechanistically, CRISPR-CATCH works by in vitro CRISPR-Cas9 treatment followed by pulsed-field gel electrophoresis (PFGE) of agarose-entrapped genomic DNA. According to Chang, the method starts by embedding the genomic sample in agarose plugs to prevent DNA shearing. Then the encapsulated sample is treated with CRISPR-Cas9 and a single guide RNA (sgRNA), which will make a precise cut to linearize the ecDNA.
"The key insight is that when you separate large pieces of DNA on a gel, called pulsed-field gel electrophoresis, large [DNA] circles get stuck in the well; they don't go into the gel. If you cut the circle just once, the circle turns into a line, and then that moves into their gel according to the right size," he explained. "We realize that we can use CRISPR-Cas9 to make precisely one cut in the ecDNA, and then enrich it."
Meanwhile, if the same cut happens on a chromosome, the two resulting DNA fragments will still be much larger and migrate at a vastly different speed than ecDNA, effectively separating ecDNA from its corresponding chromosome, Chang added.
Overall, the study showed that CRISPR-CATCH led to "massive enrichment" of the ecDNA, Chang said, providing researchers a window to study their structure, diversity, origin, and epigenomic signatures.
In particular, when the team tested CRISPR-CATCH targeting the EGFR locus on patient-derived glioblastoma neurosphere (GBM39) cells, the method enriched the ecDNA by 30-fold, resulting in ultrahigh sequencing coverage downstream. Similarly, the authors demonstrated that CRISPR-CATCH can also isolate targeted ecDNAs on flash-frozen patient tumor samples.
In addition, the study showed that CRISPR-CATCH enabled researchers to study ecDNA epigenomic profiles, such as DNA cytosine methylation (5mC), when paired with nanopore sequencing. It also enabled them to identify the chromosomal origins of ecDNA by phasing the oncogenic variants.
Chang's team also developed an analytical pipeline for de novo amplicon reconstruction using the sequencing data obtained from CRISPR-CATCH. This helped reveal heterogeneous structural variations and altered enhancer landscapes of ecDNA.
"CRISPR-CATCH enables [the] isolation and structural characterization of ecDNAs, this providing a 'best-of-both-worlds' between cytogenetic stains and whole-genome sequencing, while not being limited in circle size as is the case for existing circular DNA library enrichment methods," Roel Verhaak, a computational cancer biologist at the Jackson Laboratory who was not involved in this study, wrote in an email.
Verhaak, whose lab has also been studying ecDNA, said he is "very interested" in trying out CRISPR-CATCH in his own lab. The method "combines the specificity of cytogenetic methods with the comprehensive characterization benefits of whole-genome sequencing and appears to be relatively straightforward," he said, adding that he believes CRISPR-CATCH "will enable higher throughput in the number of samples that can be characterized for their presence of ecDNA, for example in xenograft studies of the same cancer cell line model."
Verhaak pointed out that one limitation of CRISPR-CATCH is that the ecDNA cargo oncogene needs to be known to design the appropriate ecDNA-targeting guide RNA. Also, he said future studies may further improve the method's sensitivity.
Echoing Verhaak's point, Chang said one possible future direction for CRISPR-CATCH is to devise a way to identify the targeted cut site de novo. "Right now, we rely on cancer whole-genome sequencing data, which obviously is routinely done for cancer patients, so we know what the focal amplification is and that guides the targeted cutting," he said. "There might be a way to … even bypass that step in the future."
Furthermore, because CRISPR-CATCH currently relies on gel electrophoresis, the throughput is still limited and has room for improvement, Chang noted.
Chang said that his lab is interested in using CRISPR-CATCH to help decipher the biogenesis and origin of ecDNA in cancer patients and also how these molecules evolve with cancer treatment.
"It is a very useful way of enriching our understanding of basically a very important cause of cancer," he said. "I think that following the same patients over time is going to teach us a lot about how ecDNA is driving cancer biology."