NEW YORK (GenomeWeb News) — A transcriptome-sequencing study using a combination of long- and short-read sequencing technologies made by Roche 454 and Illumina has identified new gene fusions involved in prostate cancer.
In the study, which appears in yesterday's online edition of Nature, a team of University of Michigan researchers searched for new gene fusions in prostate cancer by sequencing the transcriptomes of patient cell lines and tumor samples.
Among the newly identified fusions, the researchers found a recurrent transcript read-through called SLC45A3-ELK4 and several fusions that seem to represent private mutations.
"We defined a new class of mutations in prostate cancer," senior author Arul Chinnaiyan, a pathology and urology researcher at the University of Michigan, said in a statement. "The recurrent fusions are thought to be the driving mechanism of cancer. But we found other fusions as well, some of which were unique to individual patients."
Chromosomal rearrangements in cancer can create situations in which two previously separate genes become fused. In some instances, two gene-coding sequences may be thrown together, leading to the expression of a fusion protein. In other cases, genetic rearrangements may put one gene under the control of regulatory elements for another gene causing abnormal gene expression.
Gene fusions have traditionally been linked to blood, bone, and soft-tissue cancers such as leukemia, lymphoma, and Ewing sarcoma. But they are also starting to turn up in solid tumors as well.
For example, recurrent gene fusions have been detected in some lung cancers. And in 2005, Chinnaiyan led a team of researchers who identified the first prostate cancer-associated gene fusion: between the 5' untranslated region of the serine protease gene TMPRSS2 and the transcription factors ERG and ETV1. The TMPRSS2-ERG fusion has since been tied to particularly aggressive prostate cancers.
In an effort to identify new prostate cancer-related fusions, Chinnaiyan and his colleagues used a combination of long-read sequencing with a Roche 454 FLX and short-read sequencing with an Illumina Genome Analyzer to decode prostate cancer transcriptomes.
But first the researchers validated their approach by doing a set of proof-of-concept experiments to see if they could detect a known gene fusion — BCR-ABL1 — in cells from a chronic myelogenous leukemia cell line.
After creating cDNA libraries, the researchers generated millions of 36-nucleotide sequence reads on the Genome Analyzer. They screened these reads to find sequences aligning to more than one gene. By integrating reference sequence information from longer sequences generated by 454 sequencing, the team came up with an approach for finding new and known gene fusions.
They were also able to successfully rediscover the TMPRSS2-ERG fusion in prostate cancer tumor samples and cell lines. In the process, the researchers found new fusions and read-through events in the prostate cancer cell lines. For instance, in the VCaP line, they detected a fusion between two chromosome 16 genes, USP10 and ZDHHC7, as well as fusions involving a chromosome 2 gene called HJURP.
In the LNCaP line, meanwhile, they found fusions between a gene called MIPOL1 on chromosome 14 and a gene called DGKB on chromosome 7.
Next, the team turned from cell lines to tumor samples, using their transcriptome sequencing approach to assess two metastatic prostate cancer samples that were known to carry the TMPRSS2-ERG fusion. In these lines, they turned up the TMPRSS2-ERG fusion, as expected, and several new gene fusions and read-through transcripts.
One of these, SLC45A3-ELK4, appears to occur regularly in prostate cancer — seven of the 20 metastatic tissues tested expressed the fusion. And, the authors noted, SLC45A3-ELK4 fusion appears to arise without detectable DNA changes.
"Overall SLC45A3-ELK4 appears to be the only recurrent chimeric transcript identified in our transcriptome sequencing study as other gene fusions tested in a panel of prostate cancer samples appeared to be restricted to the sample in which they were identified … and thus may represent rare or private mutations," they wrote in the Nature paper.
Chinnaiyan and his colleagues are reportedly applying a similar approach to search for gene fusions in breast and lung cancers and melanoma.
"Taken together, this study establishes a robust pipeline for the discovery of novel gene chimaeras using high-throughput sequencing, opening up an important class of cancer-related mutations for comprehensive characterization," the authors wrote.