Abstract P4-07-09: Breast cancer specific expression of miRNA deciphered using next generation sequencing of LCM procured cells

Abstract The objective of this study is to decipher miRNA expression profiles of laser capture microdissection (LCM)-procured carcinoma cells compared to those of intact serial sections of a breast cancer biopsy. Our hypothesis is that miRNA signatures discerned from specific carcinoma cell populations more precisely correlate with clinical behavior than that provided by conventional biomarkers of intact tissue biopsies. De-identified frozen biopsies of invasive ductal carcinomas of known grade and biomarker status containing 35-70% cancer were selected from an IRB-approved Biorepository (JLW). Serial tissue sections were stained with H & E and 12-15,000 carcinoma cells were collected from an adjacent section. RNA was extracted using PureLink RNA Mini Kit™ (Invitrogen), evaluated (Agilent Bioanalyzer) and sequenced for miRNA expression using the Ion Torrent™ System (Thermo Fisher). Total RNAs were enriched for small RNA species using mirVana miRNA™ kits (Thermo Fisher) and RNA libraries were constructed from 5 ng of enriched RNA using Ion Total RNA-Seq Kit v2. Barcodes were utilized to multiplex libraries for template preparation and sequencing on Proton PI™ chips as twelve-plex library pools. Each library was sequenced to an average of 30M reads on the Ion Proton™sequencer with the Ion PI™ chip. Sequence reads were aligned to miRNA precursor hairpins available from the miRBase (miRBase.org) miRNA repository. Aligned reads to each miRBase reference miRNA were then reported. Using the R statistical software package, DESeq (Bioconductor), counts for all libraries were normalized and relative expression was calculated. Mapping statistics (e.g., aligned reads (range 59-79%) and miRBase matches (range 69-85%)) were assessed for each library. Comparison of expressed miRNAs from intact tissue sections with those of cognate carcinoma cells procured by LCM revealed, in general, that smaller defined miRNA gene sets were expressed in isolated populations of carcinoma cells. miRNA expression patterns of experimental pairs (intact vs LCM-procured) using MA-plots were highly variable in carcinomas with different grades, suggesting a relationship to disease status. Gene frequency plots, comparing expression from intact tissue sections to that of LCM-procured cell population, revealed subsets of differently expressed miRNAs. To increase statistical power, a follow-up experiment was performed with triplicate libraries from 4 different representative carcinoma samples. In addition to miRNA sequencing, targeted RNA sequencing with an Ion AmpliSeq™ RNA panel was used to capture gene expression information from the12 additional samples. From these replicated libraries, we are able to combine mRNA and miRNA expression information to create an expected profile from these breast carcinoma tissue samples. Application of Next Generation Sequencing of miRNAs and Ion AmpliSeq™ RNA panels using LCM-procured cells and intact tissue provides an innovative approach for assessing differential expression of miRNA and mRNA levels involved in breast cancer behavior. Supported in part by a grant from the Phi Beta Psi Charity Trust (JLW & SAA) and a CTSP Award from the Commonwealth of Kentucky (JLW). For research use only. Citation Format: Kelli S Bramlett, James L Wittliff, Jose G Cienfuegos, Sarah A Andres, Jeoffrey J Schageman. Breast cancer specific expression of miRNA deciphered using next generation sequencing of LCM procured cells [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P4-07-09.