Purchase this article with an account.
Michael H Farkas, Tadeusz J Kaczynski, Elizabeth D. Au, Lara A. Skelton, Maria E Sousa, Steven J Fliesler; High-throughput localization of lncRNAs guides functional studies in the RPE. Invest. Ophthalmol. Vis. Sci. 2019;60(9):2357.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Long non-coding RNAs (lncRNAs) are rapidly being identified as important mediators of cellular function and disease, but remain vastly understudied, particularly in the eye. A major reason is the difficulty in functionally studying lncRNAs. Here, we present a detailed characterization of the sub-cellular localization of RPE-expressed lncRNAs and provide an example of the utility of this data in guiding functional studies.
Two distinct lines of human iPSCs were differentiated to RPE and matured to 90 days. The nucleus and cytoplasm were fractionated and total RNA isolated from each. RNA-Seq libraries were prepared and sequenced on a NextSeq. The reads were aligned to the hg38 human genome build using STAR, and gene quantification was performed using RSubread and the Gencode v24 transcriptome database, and validated using RNA fluorescence in situ hybridization. To functionally study lncRNAs, we used iCLiP to study RNA interactions with DNA methyltransferase 1 (DNMT1) in iPSC-RPE. RNA-Seq libraries were prepared from RNA isolated following the immunoprecipitation, sequenced, aligned, and counted as described above. Downstream processing was performed using custom scripts.
Roughly 45,000 RNA isoforms were found to be expressed in both iPSC-RPE lines. We determined that 833-1,067 lncRNAs were localized to the nucleus, 779-1,183 lncRNAs were roughly equally distributed between the nucleus and cytoplasm, and 366-419 were localized to the cytoplasm. Concordance between the two iPSC-RPE lines was between 0.58-0.85. The DNMT1 immunoprecipitation analysis identified 22 interacting lncRNAs. Using the RNA localization data, all of these lncRNAs were found to be localized to the nucleus, adding a layer of validation to the analysis. MALAT1, a particularly noteworthy lncRNA, was found to be the predominant lncRNA bound to DNMT1, based on RNA-Seq count data.
lncRNAs lack many of the characteristics that aid the study of mRNAs, making them inherently more difficult to study from a functional standpoint. Since lncRNA function is partly driven by localization, we have undertaken a high-throughput approach to define their localization. Using this data to supplement other lncRNA studies, we have identified MALAT1, an important lncRNA with regard to development, cellular function, and disease, to bind to DNMT1, a maintenance DNA methyltransferase, suggesting a possible role in iPSC-RPE differentiation.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
This PDF is available to Subscribers Only