Targeting Molecular Chimeras with RNA Sequencing

“She was…in the fore part a lion, in the hinder a serpent, and in the midst a goat breathing forth in terrible wise the might of blazing fire.” – Homer, Iliad

To ancient Greek storytellers like Homer, the Chimera was an effective way to both captivate an audience and evoke horror at the same time. The monster’s terrible form foreshadowed its terrible power. And while the Chimera that Homer describes was no more than a myth, the concept lives on in a very real beast that vexes molecular biologists to this day: fusion transcripts.

As its name suggests, a fusion transcript is formed when exons from multiple distinct genes are inappropriately combined, leading some to refer to these genes and their products as chimeras¹. Sometimes, these fusion events give rise to proteins that resemble—in both form and function—an unnatural combination of protein parts that drive the development of various diseases, most notably cancer. Accordingly, fusion transcripts have become increasingly important, not only for understanding disease pathology, but as potential biomarkers and therapeutic targets².

Unfortunately, studying chimeras is far from trivial, in part because they’re difficult to anticipate. Fusions events can occur in either DNA or RNA, and precisely which segments of genetic material are fused can vary widely. This variability makes it difficult for any one assay to suffice for all applications. Often researchers must choose between assays that either enhance discovery potential while sacrificing sensitivity, or else increase sensitivity at the expense of decreased breadth. Identifying rare or unexpected fusion events has thus been a considerable challenge.

To overcome this challenge, Twist Bioscience has partnered with the Center for Genomics and Transcriptomics (CeGaT)—a global provider of sequencing services for research, clinical studies, and human genetic diagnostics—to develop a sensitive approach to fusion transcript detection using targeted RNA sequencing. The unique design of this assay gives researchers the freedom to discover the unexpected, while preserving the confidence and sensitivity that comes with targeted sequencing. Below, we dive in on molecular chimeras and the tools we use to hunt them.

Chimeras In Disease

Gene fusions are important biological anomalies that can drive disease pathology, with some of the most well documented examples coming from the field of oncology. For example, Ewing’s Sarcoma—a type of cancer that originates in connective tissue proximal to bones—is driven by a chimeric transcription factor that is most often the EWS-Fli1 protein³. Formed from a reciprocal chromosomal translocation, the EWS-FLI1 protein is an amalgamation of two proteins: the RNA-binding EWSR1 protein and the FLI-1 transcriptional regulator. When the DNA encoding these genes becomes fused, it produces a fusion transcript and the resulting transcription factor wreaks havoc on cells, driving malignant transformation and disease progression.³

Evidence suggests that gene fusions like EWS-FLI1 play an important role in cancer development and may present unique therapeutic targets ^4,5. Discovery of the BCR-ABL1 chimeric protein in chronic myeloid leukemia, for example, led to the development of multiple first line treatments and diagnostics ⁶. In a study of 4,366 primary tumor samples, researchers reported finding 7,887 fusion transcripts across 13 different cancer types. Among these transcripts, at least 7% involved a protein kinase and stoke interest as potential drug targets ⁵.

And, fusions are not uncommon. CeGaT has reported that, in a retrospective analysis of 1,369 clinical cases spanning 89 different tumor types, clinically relevant fusions were found in 127 samples (9%).

Clearly, studying fusion transcripts is a valuable line of research with clinical implications. Yet, identifying these molecular chimeras is a considerable challenge.

Detecting RNA Fusion Transcripts

DNA sequencing is an invaluable tool for studying the genetic underpinnings of health and disease, yet it can also be limited in some critical ways that reduce its value in detecting chimeras. One substantial hurdle is the relative rarity of a fusion sequence within a cell, where the fusion may only exist on one or two DNA molecules. Additionally, the cells containing the fusion gene may be intermixed with several other normal cell lineages. Detecting such few molecules with confidence may require expansive (and costly) rounds of sequencing.

And, even with expansive sequencing, short-read DNA technology may still struggle to identify certain fusion events, such as those that originate at intronic break points (those that occur between exons) or between multiple distinct RNA transcripts. Therefore to detect fusion transcripts, many researchers are turning to RNA sequencing.

Unlike DNA sequencing, RNA sequencing allows the detection of fusion events originating in either DNA or RNA while also providing quantitative readouts of the chimeras’ expression levels. And, while fusions may only be present on a few DNA molecules, the resulting fusion transcripts can swell in number to hundreds or thousands of molecules. Detecting these abundant molecules with RNA sequencing is much more practical relative to DNA sequencing.

Generally, RNA sequencing approaches fall into one of two designs: those sequencing the whole transcriptome; or those focusing on predefined targets.

Whole transcriptome sequencing represents a relatively unbiased approach to fusion transcript detection, enabling researchers to cast a wide net and be surprised by what they catch. However, this comes at a cost: whole transcriptome sequencing is not only expensive, but it is often lower in sensitivity when compared to targeted sequencing approaches⁴.

Rather than spreading sequencing resources across the entire transcriptome, targeted RNA sequencing enables researchers to focus on deeply sequencing specific transcripts. When well-designed, a targeted approach can greatly improve the sensitivity of an assay and enable the reliable detection of rare fusion transcripts.

Such focused, sensitive sequencing can be achieved by enriching RNA samples for target transcripts using amplicons, anchor multiplexed PCR, or hybrid capture probes. While each approach has unique and non-overlapping advantages, hybrid capture has shown particular benefits for fusion transcript detection. Unlike amplicon sequencing, hybrid capture is less prone to amplification bias and is likely to have a lower dropout rate ^4,7,8. Additionally, hybrid capture can be designed to enable hypothesis-free detection of fusion transcripts involving specific genes, making it possible to discover both known and novel chimeras.

Because of these benefits, targeted RNA sequencing using hybrid capture represents a powerful tool for fusion transcript detection. As with any next-generation sequencing tool, though, the value of targeted RNA sequencing relies heavily on the quality of assay design.

Hunting Chimeras With The Twist Alliance CeGaT RNA Fusion Panel

Twist has partnered with CeGaT to develop a targeted RNA sequencing panel that captures fusion transcripts involving any of 160 different genes. Notably, fusions involving these genes have been linked to over 30 different types of cancer, making them attractive biomarkers (Table 1).

Beyond simply capturing expected fusion transcripts, the Twist Alliance CeGaT RNA Fusion Panel is able to detect unexpected fusion breakpoints and other coding variants in any of the CDS-targeted genes covered by this panel. In leaving room for the unexpected, this panel increases the researcher’s ability to detect fusions and other structural mutations that might affect target gene function and activity.

Like all Twist target enrichment panels, this panel is synthesized with an industry-low error rate and a high degree of uniformity. The very low error rate in Twist capture probe synthesis can significantly reduce missed or off-target binding, which increases the likelihood of novel fusion discovery. Similarly, Twist's industry-leading capture probe synthesis ensures unbiased enrichment of all transcripts and, subsequently, reduces the number of overall sequencing reads required to analyze under-represented transcripts.

In emphasizing this point, the Twist Alliance CeGaT RNA Fusion Panel was put to the test on publicly available reference materials that included roughly 15 known fusion variants. Measurement of fusion fragments per million (FFPM) showed high detection of all expected fusion variants across three replicate experiments, demonstrating a 100% sensitivity for this reference set.

Fusion panel validation results from the SeraCare FFPE NTRK Fusion RNA standard, sequenced in triplicates by researchers at CeGaT. Table shows FFPM (fusion fragments per million) values. All expected fusions are found with high support. Additional (putative false-positive) calls are present but with very low support.

Further comparison was done to test panel performance against total RNA sequencing in 91 samples collected from either fresh frozen or formalin fixed, paraffin embedded tissue. The panel identified fusions in 27 samples. Notably, one-third of fusion events were left undetected by total RNA sequencing. In a subset of samples (59) for which whole exome data was available, more than half of the identified fusion events were unrecognized in the DNA sequencing data.

Together, these results indicate that the Twist Alliance CeGaT RNA Fusion Panel is a powerful tool that can help researchers detect these elusive molecular chimeras.

References

1. Wu, Hao, et al. “Gene Fusions and Chimeric RNAs, and Their Implications in Cancer.” Genes & Diseases, vol. 6, no. 4, Dec. 2019, pp. 385–390, https://doi.org/10.1016/j.gendis.2019.08.002.
2. Powers, Martin J. The Ever-Changing World of Gene Fusions in Cancer: A Secondary Gene Fusion and Progression. Vol. 38, no. 47, 15 Oct. 2019, pp. 7197–7199, https://doi.org/10.1038/s41388-019-1057-2.
3. Grünewald, Thomas G. P., et al. “Ewing Sarcoma.” Nature Reviews. Disease Primers, vol. 4, no. 1, 5 July 2018, p. 5, pubmed.ncbi.nlm.nih.gov/29977059/, https://doi.org/10.1038/s41572-018-0003-x.
4. Mertens, Fredrik, et al. “The Emerging Complexity of Gene Fusions in Cancer.” Nature Reviews Cancer, vol. 15, no. 6, June 2015, pp. 371–381, https://doi.org/10.1038/nrc3947.
5. Yoshihara, K, et al. “The Landscape and Therapeutic Relevance of Cancer-Associated Transcript Fusions.” Oncogene, vol. 34, no. 37, 15 Dec. 2014, pp. 4845–4854, https://doi.org/10.1038/onc.2014.406.
6. Kantarjian, Hagop, et al. “Dasatinib versus Imatinib in Newly Diagnosed Chronic-Phase Chronic Myeloid Leukemia.” The New England Journal of Medicine, vol. 362, no. 24, 2010, pp. 2260–70, www.ncbi.nlm.nih.gov/pubmed/20525995/, https://doi.org/10.1056/NEJMoa1002315.
7. Samorodnitsky, Eric, et al. “Evaluation of Hybridization Capture versus Amplicon-Based Methods for Whole-Exome Sequencing.” Human Mutation, vol. 36, no. 9, 15 July 2015, pp. 903–914, https://doi.org/10.1002/humu.22825.
8. Heyer, Erin E., et al. “Diagnosis of Fusion Genes Using Targeted RNA Sequencing.” Nature Communications, vol. 10, no. 1, 27 Mar. 2019, https://doi.org/10.1038/s41467-019-09374-9.