UC Santa Cruz

Most of the research on cancer has focused on mutations in protein-coding genes. However, less than 2% of the genome codes for proteins and most of the transcriptome is composed of noncoding RNAs. Long noncoding RNAs (lncRNAs) have been found to modulate each one of the hallmarks of cancer. Pancreatic cancer has one of the lowest survival rates of all cancers and is difficult to detect until it has metastasized. Therefore, there is an urgent need to find biomarkers for the detection of this cancer. Exosomes are a potential source for biomarkers since they are released by cells for cell-to-cell communication, they contain different RNAs, and are involved in cancer. The research presented here answers the question of which pancreatic-cancer-associated RNAs are detectable in the exosomes released from pancreatic cancer cells. To this end, we first performed RNA-seq from control PANC1 cells compared to KRAS knockdown PANC1 cells. KRAS is one of the main driver mutations of pancreatic cancer. We then took the list of downregulated lncRNAs and protein coding RNAs and filtered it using TCGA data to obtain a list of validated pancreatic-cancer-associated RNAs. From this validated list, we found that two lncRNAs, AL121772.1 and LINC00294, were detectable in the exosomes. In addition, we found 30 protein coding RNAs that were detectable in the exosomes and two of these, AURKA and PAK2, were strongly correlated to poor prognosis when expressed at elevated levels. These findings represent new set of candidates for further studies of biomarkers for pancreatic cancer.