Abstract
N6-Methyladenosine (m6A) accounts for approximately 0.2% to 0.6% of all adenosine in mammalian mRNA, representing the most abundant internal mRNA modifications. m6A RNA immunoprecipitation followed by high-throughput sequencing (MeRIP-seq) is a powerful technique to map the m6A location transcriptome-wide. However, this method typically requires 300 μg of total RNA, which limits its application to patient tumors. In this study, we present a refined m6A MeRIP-seq protocol and analysis pipeline that can be applied to profile low-input RNA samples from patient tumors. We optimized the key parameters of m6A MeRIP-seq, including the starting amount of RNA, RNA fragmentation, antibody selection, MeRIP washing/elution conditions, methods for RNA library construction, and the bioinformatics analysis pipeline. With the optimized immunoprecipitation (IP) conditions and a postamplification rRNA depletion strategy, we were able to profile the m6A epitranscriptome using 500 ng of total RNA. We identified approximately 12,000 m6A peaks with a high signal-to-noise (S/N) ratio from 2 lung adenocarcinoma (ADC) patient tumors. Through integrative analysis of the transcriptome, m6A epitranscriptome, and proteome data in the same patient tumors, we identified dynamics at the m6A level that account for the discordance between mRNA and protein levels in these tumors. The refined m6A MeRIP-seq method is suitable for m6A epitranscriptome profiling in a limited amount of patient tumors, setting the ground for unraveling the dynamics of the m6A epitranscriptome and the underlying mechanisms in clinical settings.