MicroRNA Functions in Stress Responses
Anthony K.L. Leung1and Phillip A. Sharp1,2,*
1Koch Institute for Integrative Cancer Research
2Department of Biology
Massachusetts Institute of Technology, Cambridge, MA 02139, USA
MicroRNAs (miRNAs) are a class of ?22 nucleotide short noncoding RNAs that play key roles in fundamental
cellular processes, including how cells respond to changes in environment or, broadly defined, stresses.
Responding to stresses, cells either choose to restore or reprogram their gene expression patterns. This
decision is partly mediated by miRNA functions, in particular by modulating the amount of miRNAs, the
amount of mRNA targets, or the activity/mode of action of miRNA-protein complexes. In turn, these changes
determine the specificity, timing, and concentration of gene products expressed upon stresses. Dysregula-
tion of these processes contributes to chronic diseases, including cancers.
Stress, broadly defined, is the state when cells deviate from the
status quo due to sudden environmental changes or frequent
fluctuations in environmental factors. Such changes can
damage existing macromolecules, including proteins, mRNAs,
DNA, and lipids, that need to be replenished; if the damage
is not dealt with, metabolic imbalance can be incurred and
redox potentials altered (Kultz, 2005). Depending on the
severity and duration of stress encountered, cells either re-
establish cellular homeostasis to the former state or adopt an
altered state in the new environment. Such stress responses
are mediated via multiple mechanisms (Kultz, 2005), such as
induction of molecular chaperones (Buchberger et al., 2010;
Richter et al., 2010), rapid clearance of damaged macromole-
cules (Kroemer et al., 2010), growth arrest, activation of certain
gene expression programs (Spriggs et al., 2010), and, when
cells can no longer cope with excessive damage, cell death.
In this review, we will examine how microRNAs (miRNAs) are
well positioned to help restore homeostasis upon sudden envi-
ronmental changes, or, in the long run, enforce a new gene
expression program so cells can tolerate the new environment.
We will explore how stress alters the transcription, processing,
and turnover of miRNAs; the expression of mRNA targets; and
the activities of miRNA-protein complexes. To conclude, we
will consider the possible roles of miRNA-mediated stress
responses in the context of diseases and potential therapeutic
What are MicroRNAs?
As the name implies, miRNAs areshort RNAs of ?22 nucleotides
that modulate the stability and/or translational potential of their
mRNA targets (Figure 1A) (Bushati and Cohen, 2007; Carthew
and Sontheimer, 2009; Fabian et al., 2010). Over 60% of all
mammalian mRNAs are predicted targets of miRNAs (Friedman
et al., 2009), and miRNAs constitute a sizable class of regulators
(e.g., R950 different miRNAs in humans according to the
miRBase release 15 in April 2010; http://www.mirbase.org/),
even outnumbering kinases and phosphatases, indicating their
pervasive roles in regulation of cellular processes.
In particular, multiple lines of genetic evidence indicate that
miRNAs play key roles in mediating stress responses (reviewed
in Leung and Sharp, 2007). Perhaps somewhat surprisingly,
systematic inactivation of individual miRNA in flies and worms
has revealed that many miRNAs are dispensable for develop-
ment or viability under standard laboratory conditions (Bushati
and Cohen, 2007; Leaman et al., 2005; Miska et al., 2007).
However, this is not the case when some of these animals are
subjected to changing environments. For example, miR-7
knockout flies no longer develop their eyes properly when sub-
jected to alternating temperatures (Li et al., 2009), miR-14
mutant flies are sensitive to high salinity (Xu et al., 2003), mice
deficient in miR-208 cannot cope with cardiac overload (van
Rooij et al., 2007), and inactivation of miR-8 renders zebrafish
incapable of responding to osmotic stress (Flynt et al., 2009).
Thus, miRNA mutants could appear apparently normal but
exhibit phenotypic crisis in stress conditions.
MicroRNAs and Stress
Emerging data suggest that stress conditions can alter the
biogenesis of miRNAs, the expression of mRNA targets, and
the activities of miRNA-protein complexes. In the sections that
follow, we first explore eight specific examples in the literature
regarding the multiple roles of miRNAs in stress responses.
Then we discuss how stress regulates miRNA activities and
how cells mediate stress responses through miRNAs.
Genome Guardian p53 Regulates miRNA Biogenesis
upon DNA Damage
As numerous stepsare required to produce mature miRNAs, this
sequence of events offers various ‘‘tuning knobs’’ to precisely
adjust the level of mature miRNAs (Figure 1A). One player that
senses environmental stress and modulates the level of miRNAs
is p53, a tumor suppressor that is commonly activated upon
p53 regulates the expression of specific miRNAs at two
different levels—transcription and processing (Figure 1B).
Upon DNA damage, p53 induces the transcription of the primary
transcripts that encompass miR-34a, miR-34b, and miR-34c
from distinct genomic loci. In turn, these miRNAs together
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
repress a number of shared target genes to promote growth
arrest and apoptosis (reviewed in Hermeking, 2007). In addition,
p53 enhances the processing of a restricted population of pri-
miRNAs in cancer cells by associating with DDX5, a cofactor
of Drosha (Suzuki et al., 2009). Overexpression of each of these
miRNAs (miR-16, miR-103, miR-143, miR-145, miR-26a, and
miR-206) decreases the rate of cell proliferation. Interestingly,
most p53 mutations found in cancers are located in a domain
that isrequired forboth the miRNAprocessing function andtran-
scriptional activity (Junttila and Evan, 2009; Suzuki et al., 2009).
Thus, a loss of p53 function in transcription and processing of
specific miRNAs might contribute to tumor progression.
On the other hand, to ensure a robust DNA damage response,
the expression level of p53 must be tightly regulated, and this is
partly mediated by the ubiquitin-mediated degradation of p53
(Junttila and Evan, 2009). To further keep its default expression
and zebrafish (Le et al., 2009). Such repression is relieved upon
DNA damage by a decrease in miR-125b levels through an
unknown mechanism. Therefore, during DNA damage, a relief in
eventually resulting in upregulation of genes including miRNAs
the default repression of p53 by miR-125b raises the threshold
required for p53 activation, providing another layer of safeguard
before the DNA damage response can be triggered.
Stress-Induced Target Expression Overcomes
Apart from the cellular concentration of miRNAs, the level of
target gene repression is also dependent on the concentration
of mRNA targets relative to the miRNA (e.g., Doench et al.,
2003). This is illustrated in the case of two extracellular ligands,
MICA and MICB, which are induced by stresses like heat shock,
oxidative stress, viral infection, and DNA damage (Stern-Ginos-
sar et al., 2008). These ligands are recognized by an immune
activating receptor, NKG2D, which is expressed on natural killer
virus-infected cells. In normal cells, the translation of mRNAs
encoding these ligands is inhibited by miR-20a, miR-93, or
Figure 1. Biogenesis of Mature miRNAs Can Be Regulated at Multiple Levels
(A) miRNAs are first transcribed as primary transcripts (pri-miRNA), which fold into hairpin structures and are subsequently processed—first by Drosha in the
nucleus and then by Dicer in the cytoplasm. These processing steps result in an ?22 nucleotide duplex in which one strand, the mature miRNA, binds to the
core protein Argonaute and associated protein factors to form the final effector complex. The activity of the effector complexes is determined by the half-lives
of the miRNAs, the protein composition of the effector complexes, and the posttranslational modifications of the individual components. TF, transcription factor;
Pol II, RNA polymerase II.
(B) Upon DNA damage, transcription, processing, and stability of specific miRNAs are modulated, resulting in a p53-mediated gene expression program that
promotes cell-growth arrest and apoptosis.
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
miR-106b. However, upon stress, while the levels of these
miRNAs remain unchanged, the transcription of MICA and
levels are now detected. This result suggests that the increase in
mRNA level presumably exceeds the quantity that can be
‘‘titrated’’ by the cellular miRNAs, resulting in the expression of
MICA and MICB proteins (Stern-Ginossar et al., 2008). In effect,
the relative levels of miRNAs and mRNA targets determine how
much target protein is produced under normal conditions, thus
providing a threshold. During stress, the high level of mRNAs en-
coding the ligands exceeds the threshold and contributes to an
innate immune state (Figure 2A).
Relief of miRNA Repression by Target Mimicry
The level of miRNA-mediated repression depends not only on
the ratio of a particular mRNA target relative to miRNA, but
also on the amount of other mRNAs present in the transcriptome
that are targeted by the same miRNA. Plant cells exploit this
property to adapt to the amount of inorganic phosphate present
inthe environment (Franco-Zorrilla et al.,2007).Upon phosphate
deprivation, plants remobilize phosphate from old leaves to new
leaves by repressing a negative regulator of this process, PHO2,
through transcriptional induction of miR-399. As expected, the
level of PHO2 expression decreases as miR-399 accumulates
and binds to the PHO2 mRNA; however, such reduction halts
after 6 days, even though the level of miR-399 continues to
rise. Franco-Zorrilla and colleagues discovered that the halt in
miR-399-mediated repression of PHO2 is due to the induction,
starting on day 6, of a conserved family of noncoding RNA
species known as Mt4-TPS1 that harbors a miR-399 binding
site. This binding site competes with PHO2 transcripts for the
binding to miR-399. Therefore, overexpression of a target mimic
is another means that cells can exploit to modulate the degree of
miRNA-mediated repression during stress (Figure 2B).
Similar phenomenon of target mimicry was also observed in
animal cells (reviewed in Ebert and Sharp, 2010). Overexpres-
‘‘sponges’’) has been shown to successfully ‘‘soak up’’ specific
miRNAs in mammalian cell culture and in animals, resulting in
derepression of endogenous targets (Ebert et al., 2007; Krol
et al., 2010). As illustrated earlier, heat shock or viral infection
induces the expression of MICA/MICB transcripts to such an
extent that the existing level of miR-20a, miR-93, or miR-106b
can no longer inhibit their translation (Stern-Ginossar et al.,
2008), suggesting that these miRNAs are fully occupied. It will,
therefore, be of interest to examine whether there is an increase
in expression of other mRNAs that are targeted by the same
miRNAs under these conditions.
RNA-Mediated Association between miRNA-Protein
Complexes and Specific RNA-Binding Proteins
Determines miRNA Activity during Stress
Besides the relative concentration of miRNAs and mRNA
targets, the functional outcome of a miRNA depends on the
properties of other RNA-binding proteins that bind nearby to
miRNA-protein complexes on the same mRNA target (reviewed
in Leung and Sharp, 2007). For example, a cationic amino acid
transporter (CAT-1) mRNA is normally translationally repressed
by miR-122, but the repression is relieved during amino acid
starvation (Bhattacharyya et al., 2006). Such relief of repression
requires an AU-rich element-binding protein (ARE-BP), HuR.
Similarly, another ARE-BP, FXR1, mediates the switch for
miR-369-3 from being a translational repressor to an activator
for the expression of TNFa transcripts in quiescent cells upon
serum starvation (Vasudevan et al., 2007). However, the molec-
ular mechanisms by which these ARE-BPs alter the miRNA
activity are unclear. Given that the local AU content around
miRNA binding site is relatively high (Bartel, 2009) and that
AU-rich elements areenriched near miRNA binding sites(Jacob-
sen et al.,2010), there might be more examples to be discovered
for ARE-BPs in modulating miRNA activities.
Apart from these ARE-BPs, other RNA-binding proteins have
whereas nhl-2/CGH-1 complexes enhance miRNA-mediated
protein can have contrasting effects on miRNA activities depend-
ing on the sequence context. For example, binding of HuR adja-
of one mRNA target (CAT-1) (Bhattacharyya et al., 2006) but
reduce the mRNA and translation of another (c-myc) (Kim et al.,
2009). It remains to be determined whether this discrepancy is
mation in the miRNA recognition site, or a change in activities of
these RNA binding proteins during different cellular states.
Some Components of the miRNA-Protein Complex
Relocalize to Stress Granules upon Multiple Stress
Another property of miRNA that changes upon stress is its
subcellular location (Leung et al., 2006). The majority of miRNAs,
Target exceeds thresholdTarget mimicry
Figure 2. Stress-Induced Expression of
mRNA Targets or Target Mimics Affects
The protein output of an mRNA target is depen-
dent on (A) the level of mRNA targets relative to
the amount of miRNAs as well as (B) the level of
mRNA targets relative to other mRNAs targeted
by the same miRNA.
(A) Translation of mRNA targets is generally sup-
pressed by a constant level of miRNAs. However,
when a particular mRNA target expression level
exceeds the quantity of mRNA targets that can
be titrated by the amount of miRNAs, translation
of mRNA targets accelerates.
(B) Translation of mRNA targets decreases along with the increase in miRNAs over time. Such decrease in translation halts as a result of the expression of target
mimics, which compete with mRNA targets for the same miRNA.
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
mRNA targets, and Argonaute proteins are found diffuse in the
cytoplasm, with ?1% of Argonaute localized in cytoplasmic
structures called P bodies that are enriched with RNA degrada-
tion factors such as the decapping enzyme and other proteins
related to the miRNA pathway, including GW182 and p54/rck/
CGH-1. Upon multiple types of stresses, a subpopulation of
miRNAs, mRNA targets, Argonaute and miRNA-protein complex
components p54/rck/CGH-1, but not GW182, become enriched
in another cytoplasmic structure called stress granules (SGs)
(Leung et al., 2006). The formation of these granules can be trig-
gered by the phosphorylation of eIF2a, which is governed by five
stress-sensing kinases (Anderson and Kedersha, 2008): (1) PKR
responds to viral infection, heat, and UV irradiation; (2) PERK
senses ER stress; (3) GCN2 is activated by amino acid depriva-
tion; (4) HRI senses oxidative stress; and (5) Z-DNA kinases are
involved in antiviral responses. SGs are enriched with many
RNA regulators, such as FXR1, HuR, and p54/rck/CGH-1
(Anderson and Kedersha, 2008), which, as described earlier,
can modulate miRNA activity. It remains to be determined
whether some of the changes in miRNA-mediated repression
upon stress could be due to the local concentration of these
RNA-binding proteins or other factors within these granules,
and/or dueto theexclusion ofsomecomponents thatareusually
required for miRNA-mediated repression, such as GW182.
miRNAs Upregulated by NF-kB during Inflammation
Downregulate the Proinflammatory Signaling Cascade
In some cases, a miRNA can act as a timer of the stress
response. Timing becomes particularly important in the case of
acute stress responses such as those during inflammation. In
response to inflammation, NF-kB upregulates the transcription
of miR-9, miR-155, and miR-146 along with other inflamma-
tory-responsive genes through a signaling cascade in macro-
phages (O’Connell et al., 2010). Like other immediate early-
response genes, the expression of these primary miRNA
transcripts peaks within ?2 hr. However, the mature miRNAs
take time to accumulate and peak at a much later time (e.g.,
miR-155 at?24hr) (O’Connell etal.,2007). Giventhat repression
mediated by miRNAs would not be effective until the concentra-
tion of mature miRNAs reach a certain level in cells (Calabrese,
2008), this would create a delay between the emerging presence
of mature miRNAs and the beginning of target repression. Inter-
estingly, some of the mRNA targets of miR-9, miR-155, and
miR-146 are themselves proinflammatory signaling molecules
(e.g., NF-kB) that upregulate these miRNAs. Thus, these sig-
naling molecules would only be expressed in a defined period
of time before the increasing level of mature miRNAs switches
off their expression. In effect, miRNAs activated by NF-kB reset
the proinflammatory signaling pathway after activation (Fig-
ure 3A). Such temporal-controlled expression allows macro-
phages to elicit a strong inflammatory attack to pathogens while
minimizing the duration to inflict unwanted damages to the host.
miRNAs Enforce a New Gene Expression Program upon
Even though the effect of miRNA-mediated repression is, in
general, quantitatively modest (?2-fold) (Baek et al., 2008; Lim
et al., 2005; Selbach et al., 2008), the signal can be amplified by
strated in a study by Iliopoulos and colleagues, in which they
show that interleukin-6 (IL-6), commonly involved in inflamma-
tion, can mediate anepigenetic switch through let-7 to transform
feedback loop (Iliopoulos et al., 2009). This feedback loop links
LIN-28, a negative regulator of let-7 through the transcription
sure for only 1 hr results in long-lasting cellular responses for
many generations (>10 doublings) without requiring additional
feedback loop and spreads to other cells through paracrine and
autocrine signaling by the secreted IL6.
miRNAs Ensure Robust Eye Development in Flies upon
miR-7 mutant flies are apparently normal in standard laboratory
conditions, but their eyes cannot develop properly when grown
in an environment that constantly fluctuates in temperature (Li
et al., 2009). This is partly because miR-7 normally stabilizes
the developmental transition thanks to its strategic location
within both a reciprocal negative feedback loop and a coherent
feed-forward loop (Figure 3C). The reciprocal negative feedback
loop involves miR-7 and the transcription factor Yan, where Yan
represses the transcription of miR-7 in progenitor cells and
miR-7 represses Yan posttranscriptionally in terminally differen-
tiated cells. Since a negative feedback loop has the tendency to
toggle between two states, the robustness of the switch from
progenitor to terminally differentiated states can be instilled by
coupling with an upstream feed-forward loop (Figure 3C). During
development, the homeostasis of progenitor cells is broken by
a transient induction of the epidermal growth factor (EGF)
signaling cascade. EGF signaling upregulates the transcription
factor PNTP1, which, in turn, induces the transcription of
miR-7. In this case, both miR-7 posttranscriptionally and
PNTP1 transcriptionally repress a common target, Yan. This
resultant coherent feed-forward loop, in which PNTP1 regulates
miR-7 and both negatively regulate Yan, creates stability in Yan
repression against fluctuations in the level of PNTP1 and miR-7.
As the signaling persists and photoreceptor cells terminally
differentiate, miR-7 levels accumulate above a certain threshold.
The accumulated high level of miR-7 and the maintenance of the
feed-forward loop in differentiated cells therefore persistently
repress Yan expression, ensuring that the change in cell fate
will not be spontaneously reverted.
Stress Responses: Restore or Reprogram Gene
Expression by miRNAs
miRNAs and their targets are often connected within a gene
expression network: (1) multiple targets are linked together
through common regulation by individual miRNAs; (2) transcrip-
tion factors regulate the expression of miRNAs; and (3) many
transcription factors themselves are subject to extensive miRNA
regulation (Shalgi et al., 2007). Depending on where a miRNA is
embedded within the gene circuitry, it can either act asa restorer
to homeostasis or as an enforcer of a new gene expression
program during stress. If the miRNA is embedded in a negative
feedback loop, it allows the cells to resume their original target
expression patterns despite external triggers (Figure 3A). On
the other hand, if the miRNA is embedded in a positive feedback
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
loop (Figure 3B) or double-negative feedback loop (Figure 3C),
it becomes part of a bistable switch to enforce a new gene
expression pattern. Therefore, the involvement of miRNAs in
negative or positive feedback loops commits cells either to
stay on course or to change for good, respectively, in adapting
to new environments.
Moreover, miRNAs are ideal for buffering transcript surges
during stress. At the single-cell level, transcription in eukaryotic
cells often occurs in a burst-like manner (reviewed in Raj and
van Oudenaarden, 2008), causing the number of mRNAs per
cell to fluctuate significantly over time. With this constant fluctu-
ation, cells need to decide whether new transcripts should be
committed for translation. Such decision making becomes
Figure 3. miRNAs Are Strategically Located
in Negative, Positive, and Double-Negative
Feedback Loops to Mediate Stress
(A) A negative feedback loop: A (red) activates B
(blue), but B represses A. This circuitry results in
the restoration of the original level of A and B after
the triggering stimulus is removed. For example,
inflammation triggers a signaling cascade that
results in a NF-kB-dependent transcription of
a set of miRNAs. These miRNAs in turn target
the components of the proinflammatory pathway,
hence resetting the activation status of the inflam-
(B) A positive feedback loop: in this circuit, A acti-
vates B and B activates A. Therefore, there could
be a stable state with both A and B on or both A
and B off. Upon inflammation, an increase in IL-6
triggers activation of NF-kB and expression of
LIN-28, which results in the reduction of let-7
expression. As let-7 normally represses the
expression of IL-6, this reduction in let-7 results
in an increase in expression of IL-6, thus further
propagating the cycle of events. Unlike a negative
feedback loop, this circuitry exhibits a persistent,
self-perpetuating response long after the trig-
gering stimulus is removed.
(C) A double-negative feedback loop: A represses
state with A on and B off, or vice versa, but not
both on or off. This is found in the case of miR-7
and Yan in the differentiation of Drosophila eyes.
Similar to a positive feedback loop, this results in
a self-perpetuating response long after the trig-
gering stimulus is removed.Examplesoffeedback
loops are highlighted in yellow on the left side of
the panel, and hypothetical timings of the trigger
and resultant expression/activity of protein A and
B are indicated on the right. A feed-forward loop
is circled by dotted lines in (C). Panel designs
were adapted from Ferrell (2002).
particularly essential in times of duress
become more frequent (e.g., Cai et al.,
2008). miRNAs can ensure a steady level
of gene expression unless the stress
signal is sustained long enough to
increase the amount of transcripts over
a certain threshold, as in the case of
MICA/MICB described earlier. In addi-
tion, the induction of miRNA expression
upon stress might partly explain the phenomenon of ‘‘stress
hardening,’’ in which the tolerance of a particular stress
increases after preconditioning with low doses, or ‘‘cross-toler-
ance,’’ in which preconditioning of one stress increases toler-
ance of another stress (Kultz, 2005). Due to the long half-lives
of miRNAs, their sustained presence from the previous round
of stress potentially allows certain miRNA-mediated gene regu-
lation to be already in place to tackle future assaults.
Small Effect of miRNAs Writ Large during Stress?
A common conundrum in the miRNA field is how such subtle
effects (?2-fold) mediated by miRNAs can be important for
physiology. Yet, inactivation of the key processing enzyme Dicer
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
clearly indicates that miRNAs are important for a wide array of
biological processes (Bushati and Cohen, 2007), and apparently
normal single miRNA mutants exhibit phenotypic crisis in stress
conditions (Li et al., 2009; van Rooij et al., 2007; Xu et al., 2003).
The manifestation of phenotypes upon subtle changes in gene
dosage caused by the loss of particular miRNAs upon stress,
but not in normal conditions, is reminiscent of the phenomenon
of haploinsufficiency observed in yeast upon stress. Only 3%
of heterozygous yeast mutants exhibit growth phenotypes in
nutrient-rich media, but 66% of them display a growth or fitness
defect in at least one condition of chemical or environmental
stress (Hillenmeyer et al., 2008). Clinically, inactivation of one
allele of certain genes in humans and mice can lead to cancers
or other pathological conditions (Seidman and Seidman, 2002).
Thus, in an analogous fashion, a 2-fold change in the mRNA
target expression due to a loss of a particular miRNA could also
be important for physiology in animals. The question is which
mRNA targets are sensitive to such changes in concentration.
Based on observations in the literature, two particular classes
of proteins are particularly sensitive to changes in concentration,
where such changes can result in profound physiological effects
during stress—transcription factors and signaling molecules. A
change in the expression of transcription factors defines a new
pattern of miRNA and mRNA expressions during stress. p53
defines anew gene circuitry with new miRNA and mRNAexpres-
sion patterns upon DNA damage to promote growth arrest and
apoptosis. On the other hand, overexpression of mRNA target
mimics (Mt4-TPS1) or mRNA targets (MICA/MICB) results in an
escape from miRNA suppression (Franco-Zorrilla et al., 2007;
Stern-Ginossar et al., 2008). Since many transcription factors
regulate the expression of both miRNAs and their mRNA targets
(Tsang et al., 2007), we anticipate that the interplay among tran-
scription factors, miRNAs, and their mRNA targets can poten-
tially generate different types of expression timing patterns
(Figure 4). For example, a transcription factor can activate the
transcription of both miRNAs and their mRNA targets. Initially,
the level of miRNAs is not high enough to exert its suppressive
Figure 4. Timing of mRNA Target
Expression Can Be Modulated by the
Interplay among Transcription Factors,
miRNAs, and mRNA Targets
Depending on whether a transcription factor (TF)
activates or represses the level of miRNAs and/or
mRNA targets, different timing of mRNA target
expression results. The assumption is that it takes
time for mature miRNAs to accumulate (deplete)
upon transcriptional activation (repression) in sup-
pressing (derepressing) their mRNA targets. On
the other hand, the change in the expression of
mRNA targets is immediate.
increase in target mRNA expression
results. As miRNAs accumulate over
time, miRNAs exert their suppressive
effect on the mRNA targets, causing
a decrease in mRNA target expression.
As a result, a pulse of mRNA target
pulses, lags, and varied forms of decreases are indeed
commonly observed in gene expression responses of human
The other ideal mRNA targets are signaling complexes. They
are highly dynamic and nonstoichiometric in nature, thus making
them particularly sensitive to changes in concentration of their
components (reviewed in Inui et al., 2010). In fact, miRNAs play
instrumental roles in mediating stress responses through
components at all tiers of signaling pathways, from sensors to
transducers to effectors. For example, in zebrafish, miR-8
represses the expression of a membrane-bound component
NHERF1, which links extracellular sensing to actin cytoskeleton
rearrangement, and, as a result, regulates Na+/H+exchange
activity upon osmotic stress (Flynt et al., 2009). Upon hypoxia,
miR-210 is upregulated and targets iron-sulfur cluster assembly
proteins ISCU1/2 that are critical for electron transport and mito-
chondrial oxidation/reduction, thereby shutting down the
aerobic mitochondrial functions and switching the hypoxic cells
to anaerobic metabolism (Chan et al., 2009). Thus, the subtle
changes in the protein concentrations at these key nodes of
signaling cascades allow miRNAs to transmit information in
a timely manner and reconfigure cellular decisions.
Other Potential Modes of Modulation of miRNA
Activities upon Stress
Summarizing observations highlighted in previous sections, the
functions of miRNAs during stress can be modulated by four
factors: (1) the level of miRNAs, (2) the level of mRNA targets,
(3) the activity, or (4) the mode of action of miRNA-protein
complexes (Figure 5).The changes in thesefactors are regulated
by various signaling pathways. The level of miRNAs can be
modulated at the transcription and/or processing level by
stress-induced factors like p53 or NF-kB. In addition, the level
of miRNAs is also determined by their half-lives. In most cases
examined, miRNAs are relatively stable (e.g., half-life of
12 days in vivo [van Rooij et al., 2007]). However, increasing
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
evidence suggests that the stabilities of subsets of miRNAs are
subjected to regulation. For example, the decay of host
miR-27 is triggered by herpesvirus infection (Cazalla et al.,
2010), the turnover of miR-183/96/182 clusters, miR-204 and
miR-201 is accelerated in dark-adapted retina (Krol et al.,
2010), and the level of miR-125b decreases upon DNA damage
(Le et al., 2009). Though the molecular mechanisms involved
remain to be dissected, emerging data suggest that the speci-
ficity might lie at the 30end of the miRNAs (Ameres et al., 2010;
Cazalla et al., 2010), and several enzymes/cofactors have been
implicated (Kai and Pasquinelli, 2010). Therefore, the long half-
lives of miRNAs and thus their suppressive activity could provide
a stable memory of a cell state or, when certain miRNAs are
degraded, the ability to ‘‘forget’’ the cell state. Changes in the
expression or stability of the cellular miRNAs upon stress could
thus alter the homeostasis to a new cell state.
The level and species of miRNAs expressed during stress in
turn determine the specificity of mRNA targets. Friedman and
binding sites for different miRNAs and on average approximately
four different conserved sites per target (Friedman et al., 2009).
Such binding site architecture implies that expression of the
mRNA target is likely to be under combinatorial control of
multiple miRNAs (Krek et al., 2005). The repression effect of indi-
vidual binding sites is additive and can be cooperative when
sites are nearby (Doench et al., 2003; Krek et al., 2005). Poten-
tially, some of these sites could be considered as ‘‘stress’’
control elements that would only affect the target expression
under stress conditions when specific miRNAs are expressed
Apart from changes in the absolute amounts of miRNAs and
mRNA targets, the availability and nature of a particular miRNA
binding site can also be altered by RNA editing, alternative
splicing, or the use of alternative polyadenylation sites—
processes that are known to be regulated by changes in cellular
status or environment (Biamonti and Caceres, 2009; Farajollahi
and Maas, 2010; Neilson and Sandberg, 2010) (Figure 5B). In
addition, it has become increasingly evident that miRNA binding
sites, although commonly discovered and examined in the 30
untranslated regions, can also be present in the coding
sequences (Chi et al., 2009; Hafner et al., 2010; Zisoulis et al.,
2010). In particular, these miRNA binding sites can be located
miRNA accessibility to those binding sites might be sensitive to
the level of tRNA charging, which is commonly altered during
stress (e.g., Zaborske et al., 2009).
Figure 5. Different Ways to Modulate miRNA Activities upon Stress
The miRNA function can be modulated at multiple levels by changing (A) the level of mature miRNAs, (B) the level of mRNA targets, (C) the activity of miRNA-
protein complex, and (D) the mode of action of miRNA-protein complex.
(A) Shown is an mRNA target that has threebinding sites for three different miRNAs. Innormal condition, the target is repressed by miRNAs A and B. Upon stress,
expression of mRNA target decreases if the level of miRNA C increases (hence three sites are bound by miRNAs); alternatively, the expression increases if the
level of miRNA B decreases (only one site is bound).
(B) The expression of mRNA targets (black) increases if the expression of mRNA target mimics (green) increases upon stress. These mimics compete with the
same miRNAs as mRNA targets, thereby causing a relief in the repression of mRNA targets (‘‘Target mimicry’’). Alternatively, upon stress, cells could express
different isoforms of the mRNA targets where miRNA binding sites could be created or deleted.
(C)A change intheactivity of miRNA-protein complex upon stress couldbearesult ofposttranslationalmodifications of components inthecomplex,directasso-
ciation with other stress-specific cofactors, activation/repression from adjacent RNA-binding proteins bound on the same mRNA targets, or sequestration to
specific subcellular locales, such as stress granules or P bodies.
(D) Stress might alter the balance between the two major modes of action of miRNA-protein complexes: accelerating mRNA decay or inhibiting translation. Note
that mRNA degradation is irreversible by nature and, hence, tipping toward this mode of action upon stress could potentially alter the composition of the
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
The activity of miRNA-protein complexes is another determi-
nant of miRNA function that is regulated by signaling pathways
in response to environmental cues (Figure 5C). The activity of
the complexes can be modulated by its direct association with
a stress-sensing chaperone Hsp90 (Johnston et al., 2010; Pare
et al., 2009), or by posttranslational modifications of individual
components (Heo and Kim, 2009). For example, the core miRNA
binding protein Argonaute has been shown to be modified by
different posttranslational modifications, including phosphoryla-
tion by p38 MAP kinase pathway (Zeng et al., 2008) and proline
hydroxylation (Qi et al., 2008), whose enzymes are upregulated
upon hypoxia. Alternatively, the miRNA processing complex or
the miRNA-protein complex itself can contain components that
sense the environment. For example, Drosha cofactor DGCR8
is a heme-binding protein, and heme binding stimulates Dro-
sha-DGCR8 activity in vitro, suggesting a potential connection
between miRNA processing and a heme-regulated, oxygen-
sensing signaling cascade in vivo (Faller et al., 2007).
Finally, miRNAs have two major modes of action: accelerating
and Cohen, 2007; Carthew and Sontheimer, 2009; Fabian et al.,
2010). Yet, the consequences for mRNA targets are very
different: mRNA degradation results in an irreversible removal
from the transcriptome, whereas the level of mRNA targets can
remain constant upon inhibition of translation. Depending on
which mode of action is taken, miRNAs can effectively mold
the composition of the transcriptome upon stress. It has recently
been shown in worms that the repression of lin-14 by miRNA
lin-4 at mRNA level is alleviated upon nutrient deprivation while
its suppressive effect at protein level remains unchanged (Holtz
and Pasquinelli, 2009). Therefore, it will be of interest to deter-
mine whether other stresses or miRNA/mRNA target pairs favor
one mode of miRNA action over the other (Figure 5D).
Diseases and Potential Therapeutics
Organisms face constantly changing environments in the wild,
including fluctuations in temperature and limited availability of
food and water; on the other hand, modern humans in the devel-
oped world face the other extreme, such as excess nutrients,
new dietary components, and a lack of physical activity. In either
case, homeostasis is achieved through biological processes
associated with stress responses. When the system cannot
cope with these challenges, homeostasis breaks down, which
can result in diseases such as cancers, diabetes, neurodegener-
ative disorders, cardiovascular diseases, viral infections, and
many others. In some of these cases, homeostasis can be medi-
ated viamiRNAs. Forexample, herpes simplexvirus 1expresses
its own set of miRNAs that downregulate viral immediate-early
proteins to lie dormant in the trigeminal nerve of the face
(Umbach et al., 2008). However, upon excessive sunlight, fever,
or other stress, such miRNA-mediated homeostasis is broken
down and results in cold sores. In this section, we will further
describe how dysregulation of miRNA expression and activity
can contribute to stress-related chronic diseases, using cancer
as a model for discussion.
microenvironment, avoiding host immune system detection,
proliferating, surviving, and, in some cases, metastasizing
(Hanahan and Weinberg, 2000; Kroemer and Pouyssegur,
2008; Luo et al., 2009). To do so, cancer cells usually subvert
apoptosis, metabolic pathways, and stress responses, many of
which are regulated by miRNAs. One obvious way to take
away such controls in cancer cells is to reduce the total amount
of miRNAs and/or mutate key components of the miRNA path-
ways, including Dicer and Argonaute family members. Indeed,
such reductions and mutations are commonly observed in
tumors (Esquela-Kerscher and Slack, 2006; Kumar et al., 2007;
Lu et al., 2005). Take the positive feedback loop involving let-7
and IL-6 as an example (Figure 3B): a reduction of let-7 level,
which is commonly observed in tumors (Esquela-Kerscher and
Slack, 2006; Kumar et al., 2008; Lu et al., 2005), lowers the
threshold of extracellular IL-6 that is required to trigger cancer
cell transformation (Iliopoulos et al., 2009). On the other hand,
certain miRNAs are commonly upregulated in cancer cells (Calin
and Croce, 2006). For example, metastatic tumor cells can
potentially avoid immune recognition by upregulating miRNAs
(including miR-373 and miR-502c) that repress the stress-
induced ligands MICA/MICB, which serve as extracellular
signals for elimination by natural killer cells or T lympocytes
(Huang et al., 2008; Stern-Ginossar et al., 2008).
One other recently established hallmark of cancer is the pres-
ence of stress phenotypes (Luo et al., 2009). These stress
phenotypes, including those resulting from DNA damage/repli-
cation stress, proteotoxic stress, mitotic stress, metabolic
stress, and oxidative stress, may not be responsible for initiating
tumorigenesis. Instead, cancer cells develop tolerance to these
stresses, and, in turn, become dependent on stress response
pathways to maintain their growth and survival. Dependency
on these stress response pathways regulated by miRNAs thus
miRNA levels or activities.
One advantage of using miRNAs as therapeutic targets is that
they often regulate multiple mRNA targets that belong to the
same signaling pathway or protein complexes at the same time
(Tsang et al., 2010). The key is to identify which miRNAs and
which targets are involved in each particular disease. High-
throughput biochemical techniques have been developed to
identify endogenous mRNA targets (Chi et al., 2009; Hafner
et al., 2010; Zisoulis et al., 2010). The identification of certain
miRNAs that act as enforcers of a new gene expression program
to trigger an amplification cycle within a specific positive feed-
back loop. There are ongoing efforts to deliver miRNA mimics
or perfectly complementary ‘‘antagomir’’ inhibitors to increase
or decrease, respectively, the levels of specific miRNAs in vivo.
to lower hepatitis C virus replication levels and cholesterol levels
(Elmen et al., 2008; Krutzfeldt et al., 2005; Lanford et al., 2010).
In cases where a specific miRNA-mRNA target interaction
should be modulated, short oligos termed ‘‘target protectors’’
(e.g., Choi et al., 2007). The idea of a target protector is that
a single-strand oligonucleotide could specifically interfere with
the interaction of a miRNA with a single target while leaving the
regulation of other targets of the same miRNA unaffected.
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
Alternatively, multiple signaling molecules in stress responses
modulate the transcription, processing, and stability of miRNAs.
These provide further avenues for modulating the miRNA
pathway. For example, a small molecule has been identified
from a pilot screen for positive modulators of miRNA processing
(Shan et al., 2008). This raises the possibility of restoring the
counterparts. As our understanding of these fundamental
processes deepens, therapeutic possibilities will continue to
Conclusions and Perspectives
Our current understanding of miRNA pathways is mostly based
on experiments performed under standardized laboratory
‘‘normal’’ conditions. However, by doing so, we might be over-
looking the ‘‘normal’’ function of miRNAs in stressful conditions
that cells commonly face. For example, endothelial cells are
constantly under exposure to shear stress, epidermal cells to
varying light and temperature levels, and transitional epithelial
cells to fluctuations in pH and osmotic pressures. When these
cells fail to adapt, pathological conditions arise. Yet, miRNA
functions in stress responses are not just restricted to patholog-
ical conditions; instead, these stress-mediated functions are
actuallyparts of the natural physiological processes taking place
in multicellular organisms every day.
Currently, there are only a handful of observations in the liter-
ature, which are likely to be the first of many examples to be
discovered, concerning the roles of miRNAs in stress responses
and how stress modulates miRNA activities. Multiple studies
have shown that miRNA mutant animals appear normal and
viable until subjected to stress. miRNAs can behave as temporal
agents to control a process over time or to set a threshold in
determining when target gene expression becomes sufficiently
strong to signal a transition. miRNA functions can be modulated
at multiple, distinct steps by various signaling pathways.
Depending on where miRNAs are embedded in the gene regula-
tory network, they can either function to restore homeostasis or
to enforce a new program, allowing cells to adapt to the new
environment until conditions change yet again. Together, this
emerging evidence clearly indicates that miRNAs play indis-
pensable roles in stress responses. Yet, in many of these cases,
the molecular mechanisms remain unclear and should be inves-
tigated further to elucidate these fundamental roles of miRNAs in
controlling mRNA regulation during stress.
We thank M. Lindstrom for figure illustrations, and A. Ravi, M. Ebert, and J.
Wilusz for insightful comments. This work was supported by RO1-CA133404
from the National Institutes of Health (NIH), PO1-CA42063 from the National
Cancer Institute (NCI) to P.A.S., and partially by Cancer Center Support
(core) grant P30-CA14051 from NCI. A.K.L.L. was a special fellow of the
Leukemia and Lymphoma Society.
Ameres, S.L., Horwich, M.D., Hung, J.H., Xu, J., Ghildiyal, M., Weng, Z., and
Zamore, P.D. (2010). Target RNA-directed trimming and tailing of small
silencing RNAs. Science 328, 1534–1539.
Anderson, P., and Kedersha, N. (2008). Stress granules: the Tao of RNA triage.
Trends Biochem. Sci. 33, 141–150.
Baek, D., Villen, J., Shin, C., Camargo,F.D., Gygi, S.P., and Bartel, D.P. (2008).
The impact of microRNAs on protein output. Nature 455, 64–71.
Bartel, D.P. (2009). MicroRNAs: target recognition and regulatory functions.
Cell 136, 215–233.
Bhattacharyya, S.N., Habermacher, R., Martine, U., Closs, E.I., and Filipowicz,
W. (2006). Relief of microRNA-mediated translational repression in human
cells subjected to stress. Cell 125, 1111–1124.
Biamonti, G., and Caceres, J.F. (2009). Cellular stress and RNA splicing.
Trends Biochem. Sci. 34, 146–153.
Buchberger, A., Bukau, B., and Sommer, T. (2010). Protein quality control in
the cytosol and the endoplasmic reticulum: brothers in arms. Mol. Cell 40,
this issue, 238–252.
Biol. 23, 175–205.
Cai, L., Dalal, C.K., and Elowitz, M.B. (2008). Frequency-modulated nuclear
localization bursts coordinate gene regulation. Nature 455, 485–490.
Calabrese, J.M. (2008). Dicer deletion and short RNA expression analysis in
mouse embryonic stem cells. PhD thesis. Massachusetts Institute of Tech-
nology, Cambridge, Massachusetts.
Calin, G.A., and Croce, C.M. (2006). MicroRNA signatures in human cancers.
Nat. Rev. Cancer 6, 857–866.
Carthew, R.W., and Sontheimer, E.J. (2009). Origins and mechanisms of
miRNAs and siRNAs. Cell 136, 642–655.
Cazalla, D., Yario, T., and Steitz, J. (2010). Down-regulation of a host
microRNA by a Herpesvirus saimiri noncoding RNA. Science 328, 1563–1566.
Chan, S.Y., Zhang, Y.Y., Hemann, C., Mahoney, C.E., Zweier, J.L., and
Loscalzo, J. (2009). MicroRNA-210 controls mitochondrial metabolism during
hypoxia by repressing the iron-sulfur cluster assembly proteins ISCU1/2. Cell
Metab. 10, 273–284.
Chi, S.W., Zang, J.B., Mele, A., and Darnell, R.B. (2009). Argonaute HITS-CLIP
decodes microRNA-mRNA interaction maps. Nature 460, 479–486.
Choi, W.Y., Giraldez, A.J., and Schier, A.F. (2007). Target protectors reveal
dampening and balancing of Nodal agonist and antagonist by miR-430.
Science 318, 271–274.
Doench, J.G., Petersen, C.P., and Sharp, P.A. (2003). siRNAs can function as
miRNAs. Genes Dev. 17, 438–442.
Ebert, M.S., and Sharp, P.A. (2010). Emerging roles for natural microRNA
sponges. Curr. Biol. 20, R858–R861.
Ebert, M.S., Neilson, J.R., and Sharp, P.A. (2007). MicroRNA sponges:
competitive inhibitors of small RNAs in mammalian cells. Nat. Methods 4,
Elmen, J., Lindow, M., Schutz, S., Lawrence, M., Petri, A., Obad, S., Lindholm,
M., Hedtjarn, M., Hansen, H.F., Berger, U., et al. (2008). LNA-mediated micro-
RNA silencing in non-human primates. Nature 452, 896–899.
Esquela-Kerscher, A., and Slack, F.J. (2006). Oncomirs—microRNAs with
a role in cancer. Nat. Rev. Cancer 6, 259–269.
Fabian, M.R., Sonenberg, N., and Filipowicz, W. (2010). Regulation of mRNA
translation and stability by microRNAs. Annu. Rev. Biochem. 79, 351–379.
Faller, M., Matsunaga, M., Yin, S., Loo, J.A., and Guo, F. (2007). Heme is
involved in microRNA processing. Nat. Struct. Mol. Biol. 14, 23–29.
Farajollahi, S., and Maas, S. (2010). Molecular diversity through RNA editing:
a balancing act. Trends Genet. 26, 221–230.
Ferrell, J.E., Jr. (2002). Self-perpetuating states in signal transduction: positive
feedback, double-negative feedback and bistability. Curr. Opin. Cell Biol. 14,
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
Flynt, A.S., Thatcher, E.J., Burkewitz, K., Li, N., Liu, Y., and Patton, J.G. (2009).
miR-8 microRNAs regulate the response to osmotic stress in zebrafish
embryos. J. Cell Biol. 185, 115–127.
Franco-Zorrilla, J.M., Valli, A., Todesco, M., Mateos, I., Puga, M.I., Rubio-
Somoza, I., Leyva, A., Weigel, D., Garcia, J.A., and Paz-Ares, J. (2007). Target
mimicry provides a new mechanism for regulation of microRNA activity. Nat.
Genet. 39, 1033–1037.
lian mRNAs are conserved targets of microRNAs. Genome Res. 19, 92–105.
Rothballer, A., Ascano, M., Jr., Jungkamp, A.-C., Munschauer, M., et al.
(2010). Transcriptome-wide identification of RNA-binding protein and micro-
RNA target sites by PAR-CLIP. Cell 141, 129–141.
Hammell, C.M., Lubin, I., Boag, P.R., Blackwell, T.K., and Ambros, V. (2009).
nhl-2 Modulates microRNA activity in Caenorhabditis elegans. Cell 136,
Hanahan, D., and Weinberg, R.A. (2000). The hallmarks of cancer. Cell 100,
Heo, I., and Kim, V.N. (2009). Regulating the regulators: posttranslational
modifications of RNA silencing factors. Cell 139, 28–31.
Hermeking, H. (2007). p53 enters the microRNA world. Cancer Cell 12,
Hillenmeyer, M.E., Fung, E., Wildenhain, J., Pierce, S.E., Hoon, S., Lee, W.,
Proctor, M., St Onge, R.P., Tyers, M., Koller, D., et al. (2008). The chemical
genomic portrait of yeast: uncovering a phenotype for all genes. Science
Holtz, J., and Pasquinelli, A.E. (2009). Uncoupling of lin-14 mRNA and protein
repression by nutrient deprivation in Caenorhabditis elegans. RNA 15,
Huang, Q., Gumireddy, K., Schrier, M., le Sage, C., Nagel, R., Nair, S., Egan,
D.A., Li, A., Huang, G., Klein-Szanto, A.J., et al. (2008). The microRNAs miR-
373 and miR-520c promote tumour invasion and metastasis. Nat. Cell Biol.
Iliopoulos, D., Hirsch, H.A., and Struhl, K. (2009). An epigenetic switch
involving NF-kappaB, Lin28, Let-7 microRNA, and IL6 links inflammation to
cell transformation. Cell 139, 693–706.
Inui, M., Martello, G., and Piccolo, S. (2010). MicroRNA control of signal trans-
duction. Nat. Rev. Mol. Cell Biol. 11, 252–263.
Jacobsen, A., Wen, J., Marks, D.S., and Krogh, A. (2010). Signatures of RNA
binding proteins globally coupled to effective microRNA target sites. Genome
Res. 20, 1010–1019.
Johnston, M., Geoffroy, M.C., Sobala, A., Hay, R., and Hutvagner, G. (2010).
HSP90 protein stabilizes unloaded argonaute complexes and microscopic
P-bodies in human cells. Mol. Biol. Cell 21, 1462–1469.
Junttila, M.R., and Evan, G.I. (2009). p53—a Jack of all trades but master of
none. Nat. Rev. Cancer 9, 821–829.
Kai, Z.S., and Pasquinelli, A.E. (2010). MicroRNA assassins: factors that regu-
late the disappearance of miRNAs. Nat. Struct. Mol. Biol. 17, 5–10.
Kedde, M., Strasser, M.J., Boldajipour, B., Oude Vrielink, J.A., Slanchev, K., le
Sage, C., Nagel, R., Voorhoeve, P.M., van Duijse, J., Orom, U.A., et al. (2007).
RNA-binding protein Dnd1 inhibits microRNA access to target mRNA. Cell
Kim, H.H., Kuwano, Y., Srikantan, S., Lee, E.K., Martindale, J.L., and Gorospe,
M. (2009). HuR recruits let-7/RISC to repress c-Myc expression. Genes Dev.
Krek, A., Grun, D., Poy, M.N., Wolf, R., Rosenberg, L., Epstein, E.J., MacMe-
namin, P., da Piedade, I., Gunsalus, K.C., Stoffel, M., and Rajewsky, N. (2005).
Combinatorial microRNA target predictions. Nat. Genet. 37, 495–500.
Kroemer, G., and Pouyssegur, J. (2008). Tumor cell metabolism: cancer’s
Achilles’ heel. Cancer Cell 13, 472–482.
Kroemer, G., Marino, G., and Levine, B. (2010). Autophagy and the integrated
stress response. Mol. Cell 40, this issue, 280–293.
Krol, J., Busskamp, V., Markiewicz, I., Stadler, M.B., Ribi, S., Richter, J.,
Duebel, J., Bicker, S., Fehling, H.J., Schubeler, D., et al. (2010). Characterizing
light-regulated retinal microRNAs reveals rapid turnover as a common prop-
erty of neuronal microRNAs. Cell 141, 618–631.
Krutzfeldt, J., Rajewsky, N., Braich, R., Rajeev, K.G., Tuschl, T., Manoharan,
M., and Stoffel, M. (2005). Silencing of microRNAs in vivo with ‘antagomirs’.
Nature 438, 685–689.
Kultz, D. (2005). Molecular and evolutionary basis of the cellular stress
response. Annu. Rev. Physiol. 67, 225–257.
Kumar, M.S., Lu, J., Mercer, K.L., Golub, T.R., and Jacks, T. (2007). Impaired
microRNA processing enhances cellular transformation and tumorigenesis.
Nat. Genet. 39, 673–677.
Kumar, M.S.,Erkeland,S.J., Pester, R.E.,Chen,C.Y., Ebert, M.S.,Sharp, P.A.,
and Jacks, T. (2008). Suppression of non-small cell lung tumor development
by the let-7 microRNA family. Proc. Natl. Acad. Sci. USA 105, 3903–3908.
Lanford, R.E., Hildebrandt-Eriksen, E.S., Petri, A., Persson, R., Lindow, M.,
Munk, M.E., Kauppinen, S., and Orum, H. (2010). Therapeutic silencing of
microRNA-122 in primates with chronic hepatitis C virus infection. Science
Le, M.T., Teh, C., Shyh-Chang, N., Xie, H., Zhou, B., Korzh, V., Lodish, H.F.,
andLim,B.(2009). MicroRNA-125b isanovelnegativeregulatorofp53. Genes
Dev. 23, 862–876.
Leaman, D., Chen, P.Y., Fak, J., Yalcin, A., Pearce, M., Unnerstall, U., Marks,
D.S., Sander, C., Tuschl, T., and Gaul, U. (2005). Antisense-mediated deple-
tion reveals essential and specific functions of microRNAs in Drosophila
development. Cell 121, 1097–1108.
Leung, A.K., and Sharp, P.A. (2007). microRNAs: a safeguard against turmoil?
Cell 130, 581–585.
Leung, A.K., Calabrese, J.M., and Sharp, P.A. (2006). Quantitative analysis of
Argonaute protein reveals microRNA-dependent localization to stress gran-
ules. Proc. Natl. Acad. Sci. USA 103, 18125–18130.
Li, X., Cassidy, J.J., Reinke, C.A., Fischboeck, S., and Carthew, R.W. (2009). A
microRNA imparts robustness against environmental fluctuation during devel-
opment. Cell 137, 273–282.
Lim, L.P., Lau, N.C., Garrett-Engele, P., Grimson, A., Schelter, J.M., Castle, J.,
Bartel, D.P., Linsley, P.S., and Johnson, J.M. (2005). Microarray analysis
shows that some microRNAs downregulate large numbers of target mRNAs.
Nature 433, 769–773.
Lu, J., Getz, G., Miska, E.A., Alvarez-Saavedra, E., Lamb, J., Peck, D., Sweet-
Cordero, A., Ebert, B.L., Mak, R.H., Ferrando, A.A., et al. (2005). MicroRNA
expression profiles classify human cancers. Nature 435, 834–838.
Luo, J., Solimini, N.L., and Elledge, S.J. (2009). Principles of cancer therapy:
oncogene and non-oncogene addiction. Cell 136, 823–837.
Miska, E.A., Alvarez-Saavedra, E., Abbott, A.L., Lau, N.C., Hellman, A.B.,
McGonagle, S.M., Bartel, D.P., Ambros, V.R., and Horvitz, H.R. (2007). Most
Caenorhabditis elegans microRNAs are individually not essential for develop-
ment or viability. PLoS Genet. 3, e215. 10.1371/journal.pgen.0030215.
Murray, J.I., Whitfield, M.L., Trinklein, N.D., Myers, R.M., Brown, P.O., and
Botstein, D. (2004). Diverse and specific gene expression responses to
stresses in cultured human cells. Mol. Biol. Cell 15, 2361–2374.
Neilson, J.R., and Sandberg, R. (2010). Heterogeneity in mammalian RNA 30
end formation. Exp. Cell Res. 316, 1357–1364.
O’Connell, R.M., Taganov, K.D., Boldin, M.P., Cheng, G., and Baltimore, D.
(2007). MicroRNA-155 is induced during the macrophage inflammatory
response. Proc. Natl. Acad. Sci. USA 104, 1604–1609.
O’Connell, R.M., Rao, D.S., Chaudhuri, A.A., and Baltimore, D. (2010). Physi-
ologicalandpathologicalroles formicroRNAs intheimmune system.Nat.Rev.
Immunol. 10, 111–122.
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.
Pare, J.M., Tahbaz, N., Lopez-Orozco, J., LaPointe, P., Lasko, P., and Download full-text
Hobman, T.C. (2009). Hsp90 regulates the function of argonaute 2 and its
recruitment to stress granules and P-bodies. Mol. Biol. Cell 20, 3273–3284.
Qi, H.H., Ongusaha, P.P., Myllyharju, J., Cheng, D., Pakkanen, O., Shi, Y., Lee,
S.W., and Peng, J. (2008). Prolyl 4-hydroxylation regulates Argonaute 2
stability. Nature 455, 421–424.
Raj, A., and van Oudenaarden, A. (2008). Nature, nurture, or chance:
stochastic gene expression and its consequences. Cell 135, 216–226.
Richter, K., Haslbeck, M., and Buchner, J. (2010). Life on the verge of death:
the heat shock response revisited. Mol. Cell 40, this issue, 253–266.
Seidman, J.G., and Seidman, C. (2002). Transcription factor haploinsuffi-
ciency: when half a loaf is not enough. J. Clin. Invest. 109, 451–455.
Selbach, M., Schwanhausser, B., Thierfelder, N., Fang, Z., Khanin, R., and
Rajewsky, N. (2008). Widespread changes in protein synthesis induced by
microRNAs. Nature 455, 58–63.
Shalgi, R., Lieber, D., Oren, M., and Pilpel, Y. (2007). Global and local architec-
ture of the mammalian microRNA-transcription factor regulatory network.
PLoS Comput. Biol. 3, e131. 10.1371/journal.pcbi.0030131.
Shan, G., Li, Y., Zhang, J., Li, W., Szulwach, K.E., Duan, R., Faghihi, M.A.,
Khalil, A.M., Lu, L., Paroo, Z., et al. (2008). A small molecule enhances RNA
interference and promotes microRNA processing. Nat. Biotechnol. 26,
Spriggs, K.A., Bushell, M., and Willis, A.E. (2010). Translational regulation of
gene expression during conditions of cell stress. Mol. Cell 40, this issue,
Stern-Ginossar, N., Gur, C., Biton, M., Horwitz, E., Elboim, M., Stanietsky, N.,
Mandelboim, M., and Mandelboim, O. (2008). Human microRNAs regulate
stress-induced immune responses mediated by the receptor NKG2D. Nat.
Immunol. 9, 1065–1073.
Suzuki, H.I., Yamagata,K.,Sugimoto,K.,Iwamoto,T.,Kato,S.,and Miyazono,
K. (2009). Modulation of microRNA processing by p53. Nature 460, 529–533.
back and feedforward loops are recurrent network motifs in mammals. Mol.
Cell 26, 753–767.
Tsang, J.S., Ebert, M.S., and van Oudenaarden, A. (2010). Genome-wide
dissection of microRNA functions and cotargeting networks using gene set
signatures. Mol. Cell 38, 140–153.
Umbach, J.L., Kramer, M.F., Jurak, I., Karnowski, H.W., Coen, D.M., and
Cullen, B.R. (2008). MicroRNAs expressed by herpes simplex virus 1 during
latent infection regulate viral mRNAs. Nature 454, 780–783.
van Rooij, E., Sutherland, L.B., Qi, X., Richardson, J.A., Hill, J., and Olson, E.N.
(2007). Control of stress-dependent cardiac growth and gene expression by
a microRNA. Science 316, 575–579.
Vasudevan, S., Tong, Y., and Steitz, J.A. (2007). Switching from repression to
activation: microRNAs can up-regulate translation. Science 318, 1931–1934.
Mir-14 suppresses cell death and is required for normal fat metabolism. Curr.
Biol. 13, 790–795.
Zaborske, J.M., Narasimhan, J., Jiang, L., Wek, S.A., Dittmar, K.A., Freimoser,
F.,Pan,T.,andWek, R.C.(2009). Genome-wide analysisoftRNAcharging and
activation of the eIF2 kinase Gcn2p. J. Biol. Chem. 284, 25254–25267.
Zeng, Y., Sankala, H., Zhang, X., and Graves, P.R. (2008). Phosphorylation of
Argonaute 2 at serine-387 facilitates its localization to processing bodies.
Biochem. J. 413, 429–436.
Zisoulis, D.G., Lovci, M.T., Wilbert, M.L., Hutt, K.R., Liang, T.Y., Pasquinelli,
A.E., and Yeo, G.W. (2010). Comprehensive discovery of endogenous Argo-
naute binding sites in Caenorhabditis elegans. Nat. Struct. Mol. Biol. 17,
Molecular Cell 40, October 22, 2010 ª2010 Elsevier Inc.