| May
1, 2005 |
| By:
Caifu
Chen, Kelly
McDonald, Ada
H. Wong, Olga
V. Petrauskene, Manohar
R. Furtado |
| Pharmaceutical
Discovery |
|
Figure 1. MicroRNA cleavage.
Active mature miRNAs are cleaved from longer inactive
precursors. The inherent homology between the active and
precursor molecules can add complexity to miRNA analysis.
|
MicroRNAs The endogenous small RNAs
known as microRNAs (miRNAs) found in the genomes of both animals and
plants are highly conserved by evolution, suggesting a fundamental
biological function. MicroRNAs are cleaved from larger hairpin
precursors (Figure 1), and it is the mature, cleaved, approximately
22-nucleotide miRNA that currently is the focus of most research. The
inherent homology between the longer pre-miRNA and the mature, active
miRNA makes it difficult to preferentially detect the active molecule
using traditional hybridization-based methods. This has contributed to
the challenges of accurately quantifying and elucidating the role of
miRNAs in different tissues, developmental stages and disease states.
A total of approximately 850 unique miRNAs have been discovered so
far, including 222 human miRNAs (about 250 miRNAs for each species have
been predicted). Although miRNAs were only recently discovered, they are
relatively abundant, accounting for approximately 1% of the predicted
genes in animals and plants (1). Despite their general abundance, the
level of individual miRNAs varies dramatically between cell type and
developmental stage, ranging from a few to as many as 50,000 molecules
per cell (2). As with mRNA gene transcripts, these differences in miRNA
levels are believed to be a key indicator of miRNA activity. This large
variability in the amount of miRNAs in different tissues makes it
imperative that methods for identifying and quantifying miRNAs are
accurate over an unusually large dynamic range.
MicroRNAs are not translated into proteins; instead, they regulate
the expression of other genes by either cleaving or repressing the
translation of their messenger RNA (mRNA) targets (translational
repression, not cleavage, is the mechanism of miRNAs that usually occurs
in animals, although the mechanism of action appears to be driven by the
level of base complementarity between the miRNA and its target). The
discovery that miRNAs play an important role in the post-transcriptional
regulatory process has added a new layer to our understanding of the
complexity of gene regulation.
Recent studies have implicated miRNAs in such fundamental processes
as cell development, differentiation, communication and death.
Therefore, it is not surprising that miRNAs have been found to be
involved in such disparate areas as hematopoiesis, insulin secretion,
nervous system patterning and human cancer development. Current evidence
suggests that thorough studies of gene regulation will increasingly
include miRNA analysis, and thus the development of more efficient
laboratory tools to quantify microRNAs is becoming increasingly
important.
Current Methods for Assaying MicroRNAs
Today, many researchers are using hybridization-based methods, such as
Northern blots and microarrays, to identify and quantitate miRNAs in
tissues and cells. Yet, in the face of the special challenges presented
by miRNAs, these methods are not ideal.
Northern analysis. Northern analysis was one of the first
methods commonly used for miRNA analysis because it generally is a
homegrown technology that most labs can readily access without an
investment in new capital equipment and without learning new protocols.
Because it is such a labor-intensive, manual process, Northern analysis
is not amenable to large-scale screening experiments. However, it was
adequate for the first phase of miRNA biology, in which researchers
focused on discovering the general mechanisms of miRNA function. For
example, researchers have used Northern analysis to elucidate details
about miRNA maturation (3) and to determine that complementarity between
a miRNA and its RNA target directs the miRNA's cellular function (4).
The dynamic range of Northern blot
analysis typically only reaches two orders of magnitude — a
significant problem given the wide range in the number of miRNA
molecules in cells. For example, blot conditions that can measure 40,000
miRNA molecules per cell will not accurately detect 10 miRNA molecules
per cell. Further, because Northern blots are based on hybridization,
they often cannot distinguish well between miRNAs that only have small
sequence differences — a common situation among closely related miRNA
family members. For example, the miRNA let-7c differs from family
members let-7a and let-7b by only one base. In addition, the
reproducibility of Northern blots is known to be relatively low.
When miRNAs are rare, successful
detection of the precursor and mature miRNA signals can require exposure
times as long as three days, which not only makes experimentation slow
but also increases the risk of losing the signal in the background on
the blot. Another major disadvantage of Northern blots is the enormous
amount of starting sample required. Researchers typically only hybridize
one miRNA probe to a Northern blot at a time, and it is not uncommon to
need 5 to 10 micrograms of total RNA per lane to successfully detect
miRNAs — a situation that makes experiments impossible or prohibitive
with scarce samples such as stem cells or primary human tumor cells, and
that makes profiling multiple miRNAs impractical.
Microarrays. Microarrays
provide advantages over Northern blots in that they tend to have a
slightly larger dynamic range and greatly improved throughput.
Therefore, they can enable researchers to conduct cost-effective
large-scale screening experiments. For example, researchers recently
used high-density oligonucleotide microarrays to conduct a global
analysis of miRNA expression levels in multiple eukaryotes, including
human, mouse and rat (5).
Yet, microarrays still require
enough RNA (approximately 5 µg per array) that their usefulness is
limited to experiments where the sample is relatively abundant. This
limitation is compounded by the fact that several array replicates
usually are required for statistical validity. Further, because
microarrays do not inherently include size-separation like Northern
blots, the identical sequence shared between precursor and mature,
active miRNAs can make it difficult or impossible to distinguish between
the two; when the assessment of biological activity is the goal, this
limitation can confound interpretation. Similarly, as with Northern
blots, the hybridization basis of microarrays also means that they do
not distinguish well between miRNAs that have only small differences in
sequence.
Quantitative Real-time PCR
Because of the limitations associated with both microarray and Northern
experiments, researchers typically validate results obtained by these
methods using quantitative real-time polymerase chain reaction (PCR),
which is considered to be the gold standard for quantifying both DNA and
RNA levels. Therefore, it is reasonable to expect that researchers will
increasingly use quantitative real-time PCR as their primary assay,
rather than using it as a secondary validation step, especially when
ultra high throughput is not required.
Table I. Quantitative real-time
PCR outperforms Northern and microarray analysis on most
parameters for measuring miRNAs.
|
Quantitative, real-time PCR is based
on the quantitative relationship between the amount of starting target
sample and the amount of PCR product at any given cycle number. The
higher the starting copy number of the nucleic acid target, the sooner a
significant increase in fluorescence is observed (6, 7). The specificity
of TaqMan® (Roche Molecular Systems Inc., Basel,
Switzerland) assays (quantitative real-time PCR) relies on the combined
action of two PCR primers and a sequence-specific TaqMan probe (8).
These fluorogenic-labeled probes take advantage of the 5' nuclease
activity of Taq DNA polymerase (TaqMan reagent-based chemistry), greatly
improving real-time PCR detection and making it possible to eliminate
post-PCR processing for the analysis of probe degradation (e.g., signal
generation). This enables a simple, easily performed, homogeneous assay
for gene expression profiling (8).
Figure 2. Assay scheme. Step 1 —
Looped-primer RT: Looped primers are annealed to miRNA targets
and extended by reverse transcriptase. Step 2 — Real-time PCR:
miRNA-specific forward primer, a TaqMan probe and universal
reverse primer are used for PCR reactions. Quantitation of
miRNAs is estimated based on measured CT values.
|
Despite the fact that quantitative
real-time PCR requires dramatically less sample than Northern analysis
or microarrays and has significantly improved dynamic range and
sensitivity, this technology also has encountered challenges when faced
with miRNA detection. For example, the short length of mature miRNAs has
made it difficult to design effective primer and probe sets that are
specific for mature miRNAs. Because it is possible, with careful design,
to create quantitative real-time PCR assays for the longer precursor
molecules, some researchers have used the level of miRNA precursors as
surrogate markers for the mature, active miRNA (9). Yet, the level of
precursor miRNA present in cells is not necessarily a valid indication
of the level of the corresponding mature miRNA, making this method less
than ideal.
Figure 3. The dynamic range of a
real-time PCR amplification plot of synthetic miRNA lin-4 by
stem-loop RT-PCR. Estimated synthetic miRNA input is based on
OD: 70, 700, 7000, 70,000, 700,000, 7M and 70M copies in PCR.
Significant correlation is seen between the copy number and CT
value up to seven orders of magnitude. A) The amplification
plot. B) The amplification plot converted to log scale.
|
Recently, however, a new miRNA
quantitation method has been developed that overcomes these challenges
(Table I). The method uses stem-looped primers for reverse transcription
(RT) of the miRNA, followed by quantitative real-time PCR (Figure 2).
The stem-loop structure, which is specific to the 3´ end of the mature
miRNA, extends the very short mature miRNA molecule and adds a universal
3´ priming site for real-time PCR. The stem-loop also is believed to
create steric hindrance that prevents priming of the precursor miRNA.
Real-time PCR then provides a highly specific and quantitative method to
screen miRNA expression levels across various biological contexts using
a very small amount of starting material (1 to 10 ng of total RNA or
equivalent). For example, the stem-loop method has been found to have a
dynamic range of up to seven orders of magnitude (Figure 3), making it
ideal for detecting both high- and low-abundance miRNAs. In addition,
this method enables researchers to discriminate between miRNAs that
differ by only one base (Figure 4). Finally, as with traditional
quantitative real-time PCR assays, carrying out these new stem loop
assays in the lab is fast and simple.
Figure 4. Single-base
discrimination of TaqMan® MicroRNA Assays. Relative
detection (%) is calculated based on CT difference between
perfectly matched and mismatched assays.
|
RNA Interference
The discovery that cells possess a mechanism for gene regulation that
employs small RNAs led to the development of methods for introducing
synthetic small RNAs to harness the power of that mechanism. This
technique is called RNA interference (RNAi) and can be used to determine
gene function, study signaling pathways, identify and validate drug
targets and generate animal knockdown models. RNAi is carried out in the
lab by introducing small, double-stranded, non-coding RNAs to
"silence" the expression of targeted genes in a
sequence-specific manner. The short, non-coding RNAs used in RNAi
experiments include both short interfering RNAs (siRNAs) and short
hairpin RNAs (shRNAs).
Once the si/shRNAs are delivered to
the cells, researchers routinely measure the expression level of the
target mRNA to determine the extent of transcript silencing. Although
methods like Northern blots and ribonuclease protection assays are
available, quantitative real-time PCR with fluorescently labeled probes
provides the most sensitive and accurate method for detecting and
measuring mRNA (8, 10). It is thus ideal for correlating the extent of
gene knockdown with results from a follow-up assay or with an RNAi-induced
phenotype. Finally, as the design technology for quantitative real-time
stem-loop PCR assays advances, in the future these assays also should be
appllicable to direct quantitation of si/shRNA levels in cells. This
capability will enable researchers to directly measure the success of
their RNAi delivery efforts and correlate these levels to experimental
results.
References
1. L.P. Lim, M.E. Glasner, S. Yekta et al., Science 299, 1540.
(2003).
2. L.P. Lim, N.C. Lau, E.G.
Weinstein et al., Genes Dev. 17, 991 (2003).
3. Y. Lee, K. Jeon, J.T. Lee et
al., Embo. J. 21, 4663 (2002).
4. G. Hutvagner and P.D. Zamore, Science
297, 2056 (2002).
5. Y. Sun, S. Koo, N. White et al.,
Nucleic Acids Res. 32, e188. (2004).
6. K.J. Livak, S.J. Flood, J.
Marmaro et al., PCR Methods Appl. 4, 357 (1995).
7. S.A. Bustin, J. Mol.
Endocrinol. 29, 23 (2002).
8. C.A. Heid, J. Stevens, K.J.
Livak and P. M. Williams, Genome Res. 6, 986 (1996).
9. T.D. Schmittgen, J. Jiang, Q.
Liu and L. Yang, Nucleic Acids Res. 32, e43. (2004).
10. M.R. Furtado, O.V. Petrauskene
and K.J. Livak, Real-time Quantitative PCR in the Analysis of Gene
Expression. In: "DNA Amplification: Current Technologies and
Applications" (Norwich, UK, Horizon Biosciences, 2004), p. 131.
For Research Use Only. Not for use
in diagnostic procedures.
The PCR process and 5' nuclease
process are covered by patents owned by Roche Molecular Systems, Inc.
and F. Hoffmann-La Roche Ltd.
Caifu Chen and Kelly
McDonald are senior staff scientists in the advanced research and
technology group, Ada H. Wong is scientist 2 in applied markets
product research, Olga V. Petrauskene is senior staff scientist
in applied markets product research and Manohar R. Furtado is
director of scientific operations in applied markets product research at
Applied Biosystems in Foster City, California, USA. Caifu Chen can be
reached at Tel. 650-638-5245; e-mail chencx@appliedbiosystems.com
.
|
