CRISPR/Cas9-targeted removal of unwanted sequences from small-RNA sequencing libraries

• Published in: Nucleic acids research Vol. 47; no. 14; p. e84
• Main Authors: Hardigan, Andrew A, Roberts, Brian S, Moore, Dianna E, Ramaker, Ryne C, Jones, Angela L, Myers, Richard M
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 22.08.2019
• Subjects: Base Sequence; CRISPR-Cas Systems; Gene Expression Regulation; Gene Library; High-Throughput Nucleotide Sequencing - methods; Humans; Methods Online; MicroRNAs - genetics; Models, Genetic; Nucleic Acid Hybridization - methods; RNA, Ribosomal - genetics; RNA, Small Untranslated - genetics; Sequence Analysis, RNA - methods
• ISSN: 0305-1048; 1362-4962; 1362-4962
• DOI: 10.1093/nar/gkz425

Abstract In small RNA (smRNA) sequencing studies, highly abundant molecules such as adapter dimer products and tissue-specific microRNAs (miRNAs) inhibit accurate quantification of lowly expressed species. We previously developed a method to selectively deplete highly abundant miRNAs. However, this method does not deplete adapter dimer ligation products that, unless removed by gel-separation, comprise most of the library. Here, we have adapted and modified recently described methods for CRISPR/Cas9–based Depletion of Abundant Species by Hybridization (‘DASH’) to smRNA-seq, which we have termed miRNA and Adapter Dimer—DASH (MAD-DASH). In MAD-DASH, Cas9 is complexed with single guide RNAs (sgRNAs) targeting adapter dimer ligation products, alongside highly expressed tissue-specific smRNAs, for cleavage in vitro. This process dramatically reduces adapter dimer and targeted smRNA sequences, can be multiplexed, shows minimal off-target effects, improves the quantification of lowly expressed miRNAs from human plasma and tissue derived RNA, and obviates the need for gel-separation, greatly increasing sample throughput. Additionally, the method is fully customizable to other smRNA-seq preparation methods. Like depletion of ribosomal RNA for mRNA-seq and mitochondrial DNA for ATAC-seq, our method allows for greater proportional read-depth of non-targeted sequences.