51
Chapter 4
Transcriptome Profiling Using Single-Molecule
Direct RNA Sequencing
Fatih Ozsolak and Patrice M. Milos
Abstract
Methods for in-depth characterization of transcriptomes and quantification of transcript levels have
emerged as valuable tools for understanding cellular physiology and human disease biology, and have
begun to be utilized in various clinical diagnostic applications. Today, current methods utilized by
the scientific community typically require RNA to be converted to cDNA prior to comprehensive mea-
surements. However, this cDNA conversion process has been shown to introduce many biases and arti-
facts that interfere with the proper characterization and quantitation of transcripts. We have developed a
direct RNA sequencing (DRS) approach, in which, unlike other technologies, RNA is sequenced directly
without prior conversion to cDNA. The benefits of DRS include the ability to use minute quantities (e.g.
on the order of several femtomoles) of RNA with minimal sample preparation, the ability to analyze short
RNAs which pose unique challenges for analysis using cDNA-based approaches, and the ability to per-
form these analyses in a low-cost and high-throughput manner. Here, we describe the strategies and
procedures we employ to prepare various RNA species for analysis with DRS.
Key words: RNA sequencing, Single-molecule sequencing, Transcriptome profiling, Polyadenylation
site mapping
The emergence of microarray (1–4) and high-throughput DNA/
cDNA sequencing technologies (5–10) and their application to
understanding biological processes and human disease initially
provided a relatively simplistic view of transcriptomes which has
since been replaced with a larger, more complicated view of
genome-wide transcription. We now have a much more compre-
hensive view of the genome in which a large fraction of transcripts
emanate from unannotated parts of the genome (reviewed in
(11)), and has highlighted our limited, yet rapidly emerging,
1. Introduction
Young Min Kwon and Steven C. Ricke (eds.), High-Throughput Next Generation Sequencing: Methods and Applications,
Methods in Molecular Biology, vol. 733, DOI 10.1007/978-1-61779-089-8_4, © Springer Science+Business Media, LLC 2011

52 Ozsolak and Milos
knowledge of the transcriptome and the intimate role RNA plays
in health and disease (12–17). New technologies and methods,
which offer unique approaches to transcriptome characterization
and quantitation, with particular emphasis on minimizing the
inherent biases seen with existing methods and the ability to work
with minute quantities of cellular RNA are critical to fully explore
transcriptome biology.
DNA/cDNA sequencing has eliminated some of the techni-
cal challenges posed by earlier hybridization-based microarray
strategies, including limited dynamic range of detection and the
relatively high background due to cross-hybridization, but several
fundamental shortcomings still remain which prevent us from
understanding the “true” nature of transcriptomes. One limita-
tion of cDNA-based approaches is the tendency of various reverse
transcriptases (RT) to generate spurious second-strand cDNA
due to their DNA-dependent DNA polymerase activities (18–20).
This is thought to occur through either a hairpin loop at the 3¢
end of the first-strand cDNA or by specific or nonspecific
re-priming, involving either RNA fragments or primers used for
the first-strand synthesis. This effect confounds analyses aimed at
identifying the strand of the genomic DNA which gives rise to the
RNA (e.g. detecting sense vs. antisense transcripts) (21). Another
cDNA limitation, known as template switching (22–25), occurs
during the process of reverse transcription in which the nascent
cDNA being synthesized can sometimes dissociate from the tem-
plate RNA and re-anneal to a different stretch of RNA with a
sequence similar to the initial template. This event creates an arti-
factual cDNA that comprises the 5¢ region of the initial template
attached to the 3¢ region of the second template. In addition to
causing difficulties in RNA quantification, template switching
causes problems in the identification of exon–intron boundaries
and true chimeric transcripts. RTs are also known to synthesize
cDNAs in a primer independent manner, thought to be caused by
self-priming due to RNA secondary structure, resulting in the
generation of random cDNAs (26, 27). While this self-priming
may only occur at a frequency between 2 and 10% of cDNA mol-
ecules, these self-primed products are thought to be a major
source of error in achieving accurate detection and quantification
of RNA expression. Furthermore, RTs are error-prone due to
their lack of proofreading mechanisms (28, 29) and yield low
quantities of cDNA, necessitating the use of large quantities of
input RNA and relatively high levels of amplification. Finally,
most commercial technologies for gene expression/whole tran-
scriptome analyses add further artifacts to their measurements by
requiring double-stranded cDNA for subsequent amplification by
PCR, which can eliminate information regarding which DNA
strand is encoding the RNA. While strand-specific libraries can be
prepared, they are laborious with many steps (30) or involve
RNA–RNA ligation, which is highly inefficient (31).