ArticlePDF AvailableLiterature Review

Mechanisms of Post-Transcriptional Regulation by microRNAs: Are the Answers in Sight?

March 2008
Nature Reviews Genetics 9(2):102-14

March 2008
9(2):102-14

DOI:10.1038/nrg2290

Source
PubMed

Authors:

Witold Filipowicz

Friedrich Miescher Institute for Biomedical Research

Suvendra N Bhattacharyya

Indian Institute of Chemical Biology

Nahum Sonenberg

McGill University

MicroRNAs constitute a large family of small, approximately 21-nucleotide-long, non-coding RNAs that have emerged as key post-transcriptional regulators of gene expression in metazoans and plants. In mammals, microRNAs are predicted to control the activity of approximately 30% of all protein-coding genes, and have been shown to participate in the regulation of almost every cellular process investigated so far. By base pairing to mRNAs, microRNAs mediate translational repression or mRNA degradation. This Review summarizes the current understanding of the mechanistic aspects of microRNA-induced repression of translation and discusses some of the controversies regarding different modes of microRNA function.

Possible interplay between RNA binding proteins and micro-ribonucleoproteins interacting with the mRNAs 3' UTR.A single mRNA can have several cis-acting motifs interacting with different RNA binding proteins (RBPs) and micro-ribonucleoproteins (miRNPs), which together will determine mRNA translatability or stability. The suppressive effect of the 3' UTR-binding protein ELAV1 on the miRNA-mediated repression (not shown) has recently been documented45. However, it is possible that RBPs will also interact with miRNPs to augment their repressive function, and that miRNPs will have a positive or negative effect on the activity of RBPs bound at the 3' UTR. The 7-methylguanosine cap is represented by a red circle in the figure.

…

Figures - uploaded by Nahum Sonenberg

Content may be subject to copyright.

Content uploaded by Nahum Sonenberg

Content may be subject to copyright.

MicroRNAs (miRNAs), which are approximately 21-

nucleotide-long RNA regulators of gene expression

1–3

have become a major focus of research in molecular

biology. Although for a long time they were considered

to be exclusive to multicellular organisms and possibly

essential for the transition to a more complex organ-

ism design, the recent identification of miRNAs in the

unicellular algae Chlamydomonas reinhardtii

4,5

indi-

cates that miRNAs are probably evolutionarily older

than originally thought. One to two hundred miRNAs

are expressed in lower metazoans and plants, but at

least a thousand are predicted to operate in humans.

Functional studies indicate that miRNAs participate

in the regulation of almost every cellular process

investigated and that changes in their expression are

observed in — and might underlie — human patholo-

gies, including cancer

1,2,6–8

. These findings are perhaps

not so surprising as bioinformatic predictions indicate

that mammalian miRNAs can regulate ~30% of all

protein-coding genes.

With just one possible exception noted so far

miRNAs control gene expression post-transcription-

ally by regulating mRNA translation or stability in the

cytoplasm

10–14

. However, further functions of miRNAs

seem likely. For example, by virtue of base pairing to

RNA, miRNAs could regulate pre-mRNA processing

in the nucleus or act as chaperones that modify mRNA

structure or modulate mRNA–protein interactions.

Indications that mammalian miRNAs can be imported

into the nucleus

or even secreted from the cell

will

motivate searches for currently unidentified functions

for this class of molecule. What is already certain is that

the discovery of miRNAs has revealed an important new

dimension in the complexity of post-transcriptional

regulation of eukaryotic gene expression. We are begin-

ning to understand why the 3′ UTRs of mRNA, with

which the miRNAs and other factors interact, are often

so long and so important for gene function.

The mechanistic details of the function of miRNAs in

repressing protein synthesis are still poorly understood.

miRNAs can affect both the translation and stability

of mRNAs, but the results from studies conducted in dif-

ferent systems and different laboratories have often been

contradictory: a comprehensive and lucid picture of the

mechanism of miRNA-mediated repression is difficult

to elaborate. In this Review, after briefly introducing

miRNAs and their biogenesis, we summarize what is

currently known about the mechanistic aspects of their

function in controlling mRNA stability and translation,

focusing primarily on animal cells. We also discuss the

cellular localization and reversibility of miRNA-mediated

repression. For further recent Reviews covering these

topics, see REFS 11–14,17,18, and more general infor-

mation about the biogenesis, diversity and function of

miRNAs can be found in REFS 1–3,19–23.

miRNA and micro-ribonucleoprotein biogenesis

miRNA precursors fold into imperfect dsRNA-like

hairpins, from which miRNAs are excised in two steps,

both of which are catalyzed by Drosha (also known as

*Friedrich Miescher Institute

for Biomedical Research,

4002 Basel, Switzerland.

‡

Department of Biochemistry

and McGill Cancer Center,

McGill University, Montreal,

Quebec, Canada, H3G 1Y6.

Correspondence to W.F. or N.S.

e-mails:

witold.filipowicz@fmi.ch;

nahum.sonenberg@mcgill.ca

doi:10.1038/nrg2290

Published online

16 January 2008

Mechanisms of post-transcriptional

regulation by microRNAs: are the

answers in sight?

Witold Filipowicz*, Suvendra N. Bhattacharyya* and Nahum Sonenberg

‡

Abstract | MicroRNAs constitute a large family of small, approximately 21-nucleotide-long,

non-coding RNAs that have emerged as key post-transcriptional regulators of gene

expression in metazoans and plants. In mammals, microRNAs are predicted to control the

activity of approximately 30% of all protein-coding genes, and have been shown to

participate in the regulation of almost every cellular process investigated so far. By base

pairing to mRNAs, microRNAs mediate translational repression or mRNA degradation.

This Review summarizes the current understanding of the mechanistic aspects of

microRNA-induced repression of translation and discusses some of the controversies

regarding different modes of microRNA function.

REVIEWS

102

FEBRUARY 2008

VOLUME 9 www.nature.com/reviews/genetics

Dicer

An RNase III family

endonuclease that processes

dsRNA and pre-miRNAs into

small interfering RNAs and

microRNAs, respectively.

Small interfering RNAs

(siRNAs). Small RNAs that are

similar in size to microRNAs

but are derived from the

progressive cleavage of long

dsRNA by Dicer. Upon

incorporation into an

RISC, siRNAs guide the

endonucleolytic cleavage

of the target mRNA.

RN3) and the endoribonuclease Dicer — enzymes of

the RNase III family (BOX 1). The final processing of

the ~70-nucleotide pre-miRNA hairpin by Dicer yields

~21-bp miRNA duplexes with protruding 2-nucleotide

3′ ends, similar to small interfering RNAs (siRNAs)

operating in RNA interference (RNAi). Generally, the

strand with the 5′ terminus located at the thermody-

namically less-stable end of the duplex is selected to

function as a mature miRNA, and the other strand is

degraded

3,19,21–23

miRNAs function as components of ribonucleopro-

tein (RNP) complexes or RNA-induced silencing complexes

(RISCs), referred to as either micro-ribonucleoproteins

(miRNPs) or miRNA-induced silencing complexes

(miRISCs) (BOX 1). The most important and best-

characterized components of miRNPs are proteins

of the Argonaute family

24,25

. Mammals contain four

Argonaute (AGO) proteins, AGO1 to AGO4. Their

function in miRNA repression is demonstrated by their

association with similar sets of miRNAs and their ability

to repress protein synthesis when artificially tethered to

the mRNA 3′ UTR

26–28

(FIG. 1). AGO2 is the only AGO

that functions in RNAi because its RNaseH-like

P-element induced wimpy testis (PIWI) domain, but not

those of the other AGOs, can cleave mRNA at the cen-

tre of the siRNA–mRNA duplex (BOX 1). In Drosophila

melanogaster, Argonaute1 is dedicated to the miRNA

pathway, and Argonaute2 mainly functions in RNAi

24,25

Apart from the AGO proteins, miRNPs often include

other proteins, which probably function as miRNP

assembly or regulatory factors, or as effectors mediating

the repressive miRNP functions

Box 1 | Biogenesis of miroRNAs and their assembly into microribonucleoproteins

microRNAs (miRNAs) are processed from precursor molecules (pri-

miRNAs), which are either transcribed from independent miRNA

genes or are portions of introns of protein-coding RNA polymerase

II transcripts. A single pri-miRNA often contains sequences for

several different miRNAs. Pri-miRNAs fold into hairpin structures

containing imperfectly base-paired stems and are processed in two

steps, catalysed by the RNase III type endonucleases Drosha (also

known as RN3) and Dicer. Both Drosha and Dicer function in

complexes with proteins containing dsRNA-binding domains

(dsRBDs). The Drosha partners are the pasha protein in Drosophila

melanogaster or DiGeorge syndrome critical region gene 8 (DGCR8)

in mammals. The Drosha–DGCR8 complex processes pri-miRNAs to

~70-nucleotide hairpins known as pre-miRNAs

1,3,21,24

. Some spliced-out

introns in Caenorhabditis elegans, D. melanogaster and mammals

correspond precisely to pre-miRNAs (mirtrons), thus circumventing the

requirement for Drosha–DGCR8 (REFS 125–127). Plant genomes do not

encode Drosha homologues, and all miRNA biogenesis steps in Arabidopsis

thaliana are carried out by one of four Dicer-like proteins

. In animals, pre-

miRNAs are transported to the cytoplasm by exportin5, where they are cleaved

by Dicer (complexed with TAR RNA binding protein (TRBP) in mammals and the

loquacious gene product in D. melanogaster) to yield ~20-bp miRNA duplexes.

One strand is then selected to function as a mature miRNA, while the other strand

is degraded. Occasionally, both arms of the pre-miRNA hairpin give rise to mature

miRNAs

1,3,21,24

. Vertebrates and C. elegans contain single dicer genes, but some

other organisms like D. melanogaster and plants express two or more Dicer

proteins that function as heterodimers with different dsRBD proteins and have

specialized functions

1,3,21,24

Following their processing, miRNAs are assembled into ribonucleoprotein (RNP)

complexes called micro-RNPs (miRNPs) or miRNA-induced silencing complexes (miRISCs).

The assembly is a dynamic process, usually coupled with pre-miRNA processing by Dicer,

but its details are not well understood

1,3,21,24

. The key components of miRNPs are proteins

of the Argonaute (AGO) family. Of the many paralogues encoded in plant and metazoan

genomes, usually only some — known as AGO proteins — function in miRNA or

both miRNA and small interfering RNA (siRNA) pathways. In mammals, four AGO

proteins (AGO1 to AGO4) function in the miRNA repression but only AGO2

functions in RNAi. In C. elegans, which expresses 27 Argonaute proteins, RDE1

is involved in RNAi and ALG1 and ALG2 function in the miRNA pathway

24,25

Apart from AGOs, miRNPs can contain further proteins that function as

regulatory factors or effectors mediating inhibitory function of miRNPs

19,23,24

Examples are the fragile X mental retardation protein, FMRP, and its

D. melanogaster orthologue, dFXR, which are RNA-binding

proteins known to act as modulators of translation, particularly

in neurons (reviewed in REF. 128). Some P-body components

such as GW182 and RCK/p54 (see BOX 4) interact with miRNP

AGO proteins and are essential for inducing repression

78,92,104

Nature Reviews | Genetics

I I : I I I I I

Transcription

(A)

Pri-miRNA

Processing

Pre-miRNA

Exportin5

Maturation

Drosha

(A)

‘Mirtron’

Splicing

TRBP

Dicer

AGO1–4

AAAAA

I I I I I I I I I

AAAA

AGO2

AGO1–4

DGCR8

CCR4–

NOT

Strand selection;

miRNP assembly

Nucleus

Cytoplasm

Endonucleolytic cleavage Translational repression or deadenylation

REVIEWS

NATURE REVIEWS

GENETICS VOLUME 9

FEBRUARY 2008

103

RNA interference

The dsRNA-induced, sequence-

homology dependent gene-

silencing mechanism. The

dsRNA is processed to siRNAs,

which, upon incorporation

into an RISC, guide the

endonucleolytic cleavage

of the target mRNA.

RNA-induced silencing

complex

(RISC). The ribonucleoprotein

complex, consisting of small

interfering RNA and an AGO

protein, that harbours the

‘slicer’ activity, which cleaves

an mRNA target in the

middle of siRNA–mRNA

complementarity.

micro-ribonucleoprotein

(miRNP). A ribonucleoprotein

complex containing a miRNA

and one of the AGO proteins.

Depending on the identity of

the associated AGO, it might

harbour a ‘slicer’ activity,

characteristic of an RISC.

G cap

The 7-methylguanosine (m

that is linked by a 5–5′

triphosphate bridge to the first

transcribed nucleoside at the

5′ end of eukaryotic mRNAs.

Principles of miRNA–mRNA interactions

In plants, miRNAs generally base pair to mRNAs

with nearly perfect complementarity and trigger

endonucleolytic mRNA cleavage by an RNAi-like

mechanism

. In rare instances, a similar mechanism

is used by vertebrate and viral miRNAs (see REF. 21 for

examples). However, in most cases, metazoan miRNAs

pair imperfectly with their targets, following a set of

rules determined by experimental and bioinformatic

analyses

30–34

(BOX 2). The most stringent requirement is

a contiguous and perfect base pairing of the miRNA

nucleotides 2–8, representing the ‘seed’ region, which

nucleates the interaction. With few exceptions, miRNA-

binding sites in metazoan mRNAs lie in the 3′ UTR and

are usually present in multiple copies — this is required

for effective repression of translation

30–34

. miRNAs also

exert their repressive function when their binding sites

are artificially placed in 5′ UTRs or coding regions

35,36

although the physiological effects of the coding-region

sites might be only marginal

Modes of translational repression

mRNA translation can be divided into three steps: initi-

ation, elongation and termination. Initiation starts with

the recognition of the mRNA 5′-terminal cap structure

GpppN (in which N is any nucleotide) by the eIF4E

subunit of the eukaryotic translation initiation factor

(eIF) eIF4F, which also contains eIF4G, an important

scaffold for the assembly of the ribosome initiation

complex (BOX 3). Interaction of eIF4G with another

initiation factor, eIF3, facilitates the recruitment of the

40S ribosomal subunit

38,39

. eIF4G also interacts with

the polyadenylate-binding protein 1 (PABP1). The

ability of eIF4G to interact simultaneously with eIF4E

and PABP1 brings the two ends of the mRNA in close

proximity

40,41

. This ‘circularization’ stimulates transla-

tion initiation by increasing the affinity of eIF4E for

GpppN, and might facilitate ribosome recycling

Some cellular and viral mRNAs initiate translation

independently of the m

G cap and eIF4E; in this case,

40S ribosomes are recruited to the mRNA through

Nature Reviews | Genetics

AAAAA

a Cap-dependent reporter mRNAs

ORF

AAAAA

b IRES-containing mRNA reporters

GpppN

ApppN

ORF

AAAAA

N-AGO2 or N-GW182

c Tethering reporters

Box-B hairpins

IRES

ORF

ORF 1 ORF 2

ApppN

GpppN

Figure 1 | Examples of reporters used in studies of microRNA function. a | Capped reporters containing multiple

mRNA binding sites. mRNAs containing a non-functional ApppN cap (instead of the 7-methylguanosine (m

G) cap)

can be prepared by in vitro transcription with T7 phage RNA polymerase and either introduced into cells by

transfection or used in studies in cell-free extracts. b | Mono-cistronic and bi-cistronic reporters containing a viral

internal ribosomal entry sites (IRES). Reporters containing ApppN or pppN at the 5′ end can be prepared by in vitro

transcription and then transfected into cells. c | Reporters used to study the effects of tethering to mRNA of

Argonaute (AGO) proteins or GW182 on protein synthesis. The investigated proteins are expressed as fusions

with a phage λN-peptide, which can bind the short Box-B hairpins that are inserted to the mRNA 3′ UTR

27,106

. The

λN-peptide–Box-B system can also be used to tether initiation factors eIF4E or eIF4G to the intercistronic region of

the bi-cistronic reporter

. Reporters that are generated in vitro and used for either RNA transfection experiments

or studies in cell-free extracts can be prepared with or without the poly (A) tail

43,44,52,55

. Reporters can also differ in

the number of miRNA binding sites that are present in the 3′ UTR. Reporters that are devoid of microRNA binding

sites or that contain mutated sites are used as controls.

REVIEWS

104

FEBRUARY 2008

VOLUME 9 www.nature.com/reviews/genetics

Internal ribosomal entry site

(IRES). An RNA element,

usually present in the 5′ UTR,

that allows m

G-cap-

independent association of

ribosome with mRNA.

ApppN cap

An unmethylated cap analogue

that is not bound by eIF4E.

The mRNAs with an artificially

introduced ApppN cap instead

of a physiological m

GpppN

cap are translated inefficiently.

interaction with an internal ribosome entry site (IRES)

Joining of the 60S subunit at the AUG codon precedes

the elongation phase of translation.

Although it is now clear that the effects of miRNAs

on protein synthesis can result from mRNA destabili-

zation or translational repression, whether the latter

occurs at the initiation or post-initiation step (or both)

remains a matter of debate. Several recently published

papers provide important mechanistic insights into the

repression-at-the-initiation step, giving extra credence

to this model.

Repression at the initiation step. Investigations were

carried out using HeLa cells and reporter mRNAs that

had multiple binding sites for either natural or synthetic

miRNAs in their 3′ UTR. The investigations revealed

that the translation of m

G-capped mRNAs, but not of

mRNAs containing an IRES or a non-functional ApppN

cap, is repressed by miRNAs

43,44

. As in numerous subse-

quent studies, the specificity of repression was assessed

using reporters containing mutated miRNA sites or by

antisense oligonucleotides that specifically block the

targeting miRNA. The conclusion that the m

G cap is

essential for translational repression was corroborated by

experiments with bi-cistronic mRNAs. In these experi-

ments, the activity of the first cap-dependent cistron,

but not the second cistron, placed under the control

of eIF4E or eIF4G artificially tethered to the mRNA,

was repressed by the endogenous let‑7 miRNA

(FIG. 1).

Polysome gradient analysis independently supports an

effect on the initiation step: reporter mRNAs that either

contained functional let‑7-binding sites or that were

repressed by AGO2 (artificially tethered to the 3′ UTR)

showed a marked shift in sedimentation toward the top

of the gradient, indicating reduced ribosome loading

on the repressed mRNA

. Likewise, the amino-acid-

starvation-induced release of endogenous cationic amino

acid transporter 1 (CAT1) mRNA from repression that

was mediated by the miRNA miR‑122 was accompa-

nied by a more effective recruitment of CAT1 mRNA to

polysomes in human hepatoma cells

There is substantial evidence that factors bound at

the 3′ UTR exert their inhibitory effect on translational

initiation by recruiting proteins that either interfere with

the eIF4E–eIF4G interaction or bind directly to the cap

but, unlike eIF4E, are unable to associate with eIF4G

and promote assembly of the 40S initiation complex

46–48

Could miRNPs or tethered AGO proteins function in a

similar manner? Kiriakidou et al.

recently reported that

the central domain of AGO proteins contains limited

sequence homology to the cap-binding region of eIF4E.

Importantly, the similarity includes two aromatic resi-

dues (FIG. 2), which are crucial for cap binding in eIF4E

and other cap-binding proteins

49,50

. Mutations of one or

both aromatic amino acids in AGO2 to valine but, signif-

icantly, not to other aromatic amino acids, prevented the

interaction with m

GTP–Sepharose and abolished

the ability of AGO2 to repress translation when tethered

to the mRNA 3′ UTR. These data indicate that AGO2

and related proteins can compete with eIF4E for m

binding and thus prevent translation of capped, but

not IRES-containing, mRNAs

. The data also provide

a plausible explanation for the requirement of multiple

miRNPs or tethered AGO molecules for robust repres-

sion

27,30,43,51

. Multiple copies of AGO, with an apparently

lower affinity for m

G than eIF4E

, would increase the

likelihood of AGO association with the cap. It will be

important to determine whether the AGO aromatic

residues are essential for miRNA-mediated repression in

a physiological assay. Additional evidence, for example

from cross-linking experiments, should be obtained in

support of direct interaction of AGO with the mRNA

G cap structure.

Lessons from in vitro studies. Four different cell-free

extracts that recapitulate many features of the miRNA-

mediated in vivo effects have recently been described.

In all of them, the presence of the m

G cap was required

for translational repression

52–55

; the mRNAs containing

Box 2 | Principles of microRNA–mRNA interactions

MicroRNAs (miRNAs) interact with their mRNA targets by base pairing. In plants, most

miRNAs base pair to mRNAs with nearly perfect complementarity and induce mRNA

degradation by an RNAi-like mechanism — the mRNA is cleaved endonucleolytically

in the middle of the miRNA–mRNA duplex

. By contrast, with few exceptions,

metazoan miRNAs base pair with their targets imperfectly, following a set of rules

that have been identified by experimental and bioinformatics analyses

30–34

• One rule for miRNA–target base paring is perfect and contiguous base pairing of

miRNA nucleotides 2 to 8, representing the ‘seed’ region (shown in dark red and

green), which nucleates the miRNA–mRNA association. GU pairs or mismatches

and bulges in the seed region greatly affect repression. However, an A residue across

position 1 of the miRNA, and an A or U across position 9 (shown in yellow), improve

the site efficiency, although they do not need to base pair with miRNA nucleotides.

• Another rule is that bulges or mismatches must be present in the central region of the

miRNA–mRNA duplex, precluding the Argonaute (AGO)-mediated endonucleolytic

cleavage of mRNA.

• The third rule is that there must be reasonable complementarity to the miRNA 3′ half to

stabilize the interaction. Mismatches and bulges are generally tolerated in this region,

although good base pairing, particularly to residues 13–16 of the miRNA (shown in

orange), becomes important when matching in the seed region is suboptimal

31,33

Other factors that can improve site efficacy include an AU-rich neighbourhood

and, for long 3′ UTRs, a position that is not too far away from the poly(A) tail or the

termination codon; these factors can make the 3′ UTR regions less structured and

hence more accessible to miRNP recognition

33,34,129

. Indeed, accessibility of binding

sites might have an important effect on miRNA-mediated repression

130

. Some

experimentally characterized sites deviate significantly from these rules and can, for

example, even require a bulged nucleotide in the seed region pairing

131,132

. In addition,

combinations of sites can require a specific configuration (for example, separation by

a stretch of nucleotides of specific sequence and length) for efficient repression

131

Usually, miRNA-binding sites in metazoan mRNAs lie in the 3′ UTR and are present in

multiple copies. Importantly, multiple sites for the same or different miRNAs are

generally required for effective repression

30–34

. When they are present close to each

other (10–40 nucleotides apart) they tend to act cooperatively, that is, their effect

exceeds that expected from the independent contributions of two single sites

30,33

NNNNNNN

Nature Reviews | Genetics

Bulge

ORF AAAAAA

>15 nucleotides

‘Seed’

region

Bulge

3′ complementarity

816 13 1

NNNNNNNNNN

NNNNNNNNN miRNA

REVIEWS

NATURE REVIEWS

GENETICS VOLUME 9

FEBRUARY 2008

105

an IRES or an ApppN cap were not inhibited

53,54

. Extracts

derived from D. melanogaster embryos and mouse

Krebs2 ascites cells were used to define the repression

step more precisely. In both systems, miRNAs inhibited

the association of mRNA with either the 40S or the 80S

ribosome, consistent with miRNAs targeting translation

initiation, probably at the 40S–mRNA complex assembly

step

53,54

. In agreement with the model that AGO proteins

compete with eIF4E for cap binding

, the addition of

purified initiation factor eIF4F to the ascites extract res-

cued mRNA from the miRNA-mediated inhibition

Extracts that were prepared from rabbit reticulocytes

and from human HEK293 cells were also tested for the

poly(A)-tail requirement

— translational repression

occurred only when target mRNAs contained both

an m

G cap and a poly(A) tail. In the reticulocyte lysate,

G cap dependence could be partially relieved by the

addition of poly(A) tails of non-physiological length —

0.8 kb or longer — implying that polyadenylation might

have a role in miRNA-mediated repression

. Studies in

HEK293 cell extracts showed that mRNAs containing

miRNA-binding sites underwent deadenylation irre-

spective of whether they contained an m

G cap (trans-

lationally repressed mRNAs) or an ApppN cap or IRES

(non-repressed mRNAs). Thus, although the miRNA-

mediated deadenylation had no apparent effect on the

translation of IRES-containing or ApppN-containing

mRNAs, it might have contributed to the repression of

G-capped mRNAs by disrupting the eIF4G-mediated

mRNA circularization

Taken together, the data support the notion that by

targeting one of the two terminal mRNA structures,

miRNAs prevent the synergy between the 5′ cap and

3′ poly(A) tail. Notably, HEK293 cell lysate was supple-

mented with an extract of cells overexpressing GW182

(REF. 55), a protein that recruits the CCR4–NOT dead-

enylation complex to the miRNA-bound mRNA (dis-

cussed below). Hence, in this system, miRNA-mediated

repression might be biased towards deadenylation. Of

the remaining three systems, the D. melanogaster and

mouse ascites extracts originated from non-modified

cells and responded to endogenous miRNPs

53,54

. By con-

trast, the repression in the reticulocyte lysate required

pre-annealing of the synthetic miRNA to the template

mRNA

. It is not known how effectively such a pre-

formed miRNA–mRNA duplex associates with AGO

proteins and thus to what extent this system recapitulates

the physiological miRNA response. Interestingly, bind-

ing of miRNPs to the 3′ UTR in D. melanogaster extracts

resulted in the formation of heavy aggregates, termed

pseudo-polysomes, even in the absence of translation

Whether they are related to P-bodies (discussed below)

remains unknown.

Data on the requirement of a poly(A) tail for repres-

sion in vitro differ from some findings obtained in intact

cells. Reporter RNA transcripts that were directly trans-

fected to HeLa cells were repressed even in the absence

of a poly(A) tail. In one study

, its presence resulted

in stronger repression, but this effect was not seen in a

different study

. Although it is unlikely, the possibility

that the poly(A)-free RNA becomes polyadenylated in

Box 3 | Steps in eukaryotic translation

Translation of mRNA consists of three steps: initiation, elongation and termination.

Initiation is the most complex step and is subject to a large number of

interventions, with the phosphorylation of initiation factors being the key

regulator

133

. Translation requires the participation of at least 10 initiation factors,

many of them multisubunit complexes

38,39

. Initiation of translation of most cellular

mRNAs starts with the recognition of the mRNA 5′-terminal 7-methylguanosine

G) cap (represented by the red circle in the figure) by the eukaryotic translation

initiation factor (eIF) 4E subunit of the initiation factor eIF4F, which also contains

eIF4A (an RNA helicase) and eIF4G (a large multidomain protein that functions as a

scaffold for the assembly of the translation initiation complex). Interaction of

eIF4G with another multi-subunit initiation factor, eIF3, facilitates the recruitment

of the 40S subunit, which then begins scanning the mRNA 5′ UTR in search of the

AUG (or in rare cases its cognate) initiation codon. Following the joining of the 60S

ribosomal subunit the elongation phase ensues. The elongation step can also be

regulated by phosphorylation of the elongation factor eEF2 (REF. 134). When the

ribosome encounters a termination codon, translation release factors mediate

the termination process, in which the ribosomal subunits dissociate from both the

mRNA and from each other. An important function of eIF4G is its interaction with

the poly(A)-binding protein 1, PABP1, which is associated with the poly(A) tail. This

interaction brings about the circularization of the mRNA, which stimulates

translation initiation and possibly recycling of ribosomes

40,41

. eIF6 is required for

60S subunit biogenesis, and might also act as an initiation factor that regulates

subunit joining

58–61

. Some cellular and many viral mRNAs initiate translation

independently of the m

G cap and eIF4E, and sometimes also independently of

other initiation factors. During this mode of translation, ribosomes are recruited to

the mRNA through interaction with internal ribosome entry sites (IRES), which are

usually highly structured regions in the 5′-UTR. The best-studied IRES are those of

the encephalomyocarditis and polio viruses, hepatitis C virus and the insect cricket

paralysis virus. Bi-cistronic constructs, in which translation of the upstream cistron

requires the presence of the cap and eIF4E and that of the downstream

cistron requires internal initiation, are widely used as one of the means to identify

a putative IRES

38,42

(FIG. 1).

Nature Reviews | Genetics

AUG

60S

40S

eIF6

eIF4E

PABP1

eIF4G

40S

eIF3

eIF4A

PABP1

REVIEWS

106

FEBRUARY 2008

VOLUME 9 www.nature.com/reviews/genetics

Polysome gradient analysis

A technique that involves the

sedimentation of cell extracts

through a gradient of sucrose

or glycerol, thereby allowing

the determination of the

number of ribosomes that

are associated with a specific

mRNA. Repression of

translational initiation, which

results in the less efficient

loading of ribosomes onto

mRNA, is usually associated

with a shift of mRNA towards

the top of the gradient.

transfected cells was not excluded by these studies. With

this caveat in mind, the data suggest that a poly(A) tail

per se is not absolutely required for the repression, a

conclusion supported by the observation that mRNA

containing a 3′ histone stem-loop in place of a poly(A)

tail also undergoes translational repression in HEK293

cells

Repression by preventing 60S subunit joining. An alt-

ernative mechanism of miRNA action was recently

proposed by Chendrimada et al.

The authors reported

that eIF6 and 60S ribosomal subunit proteins co-

immunoprecipitate with the AGO2–Dicer–TRBP com-

plex. eIF6 was first described as a protein that binds the

60S subunit to prevent its precocious interaction with

the 40S subunit

, and was thought to act as an initia-

tion factor. However, it was shown later that eIF6 is not

involved in translation in yeast, but rather has a crucial

function (both in yeast and mammals) in the biogenesis

of the 60S subunit in the nucleolus, and accompanies the

60S subunit to the cytoplasm

59–61

. Chendrimada et al.

showed that partial depletion of eIF6 in either human

cells or Caenorhabditis elegans rescues mRNA tar-

gets from miRNA inhibition, possibly by reducing

eIF6-mediated impediment of 60S joining.

The involvement of eIF6 in ribosome biogenesis

complicates the interpretation of the data that support its

role in miRNA repression, and invites another possible

scenario. Sachs and Davis

62,63

demonstrated that muta-

tions in a ribosomal protein and a helicase involved in

yeast 60S biogenesis could act as bypass suppressors of

complete deletion of the gene encoding poly(A) bind-

ing protein (Pab1). As in metazoans, the yeast Pab1 is

an essential protein, contributing to translation initia-

tion through its function in mRNA circularization. The

bypass suppressor mutations, which all resulted in a 60S

ribosomal subunit deficit, allowed growth, albeit reduced,

in the absence of Pab1. This rescue can be explained

by an increase in the free 40S subunit pool (resulting

from a partial depletion of 60S ribosomes), leading

to an enhanced rate of their recruitment to mRNA.

This would partially compensate for the lack of Pab1,

which stimulates 40S recruitment, and would switch

the rate-limiting step from the 40S subunit-loading

step to the 60S joining step.

A similar switch, negating the advantages of the

circularization of bulk mRNAs, could be caused by

the knockdown of eIF6. The resulting limited 60S defi-

cit

would bring some relief of the miRNA-mediated

repression, because the target mRNAs could now com-

pete with the bulk of mRNAs on a more equal footing.

If this explanation is correct, the work of Chendrimada

et al.

would be consistent with the idea that miRNA-

mediated repression affects the initiation of translation

by targeting the 5′ cap and poly(A) tail, although perhaps

not because of a direct involvement of eIF6 in repres-

sion. Admittedly, this model does not explain why eIF6

co-purifies with the RISC.

Repression at post-initiation steps. Despite compelling

in vitro and in vivo evidence, targeting of translation

initiation is unlikely to be the only mechanism by which

miRNAs bring about mRNA repression. Early studies in

C. elegans showed that lin‑14 and lin‑28 mRNAs, which

are targets of lin‑4 miRNA, remain associated with

polysomes despite a strong reduction in their protein

products at a specific stage of larval development

64,65

Similar results, which are incompatible with the initia-

tion model, were seen in mammalian cells. In two studies

that used reporter mRNAs targeted by either synthetic

Nature Reviews | Genetics

DUF

1875

PAZ

C-GWM-GW

N-GW Q-richUBA

RBD

505

470

Cap binding

RNase H-like fold

AGO binding

P-body targeting

PIWI

Human AGO2

Drosophila

melanogaster

GW182

Figure 2 | Domain organization of Argonaute and GW182 proteins. The schemes represent human Argonaute2

(AGO2) and Drosophila melanogaster GW182, two proteins extensively characterized in mediating the microRNA

(miRNA)-mediated repression. AGO2 is the only mammalian AGO protein that, in addition to miRNA repression, also

functions in RNAi. Its RNAseH-like P-element induced wimpy testis (PIWI) domain is competent in endonucleolytically

cleaving the mRNA. The region separating the PIWI Argonaute Zwille (PAZ) and PIWI domains of AGO2 contains two

aromatic amino acids (phenylalanines F

470

and F

505

), mutation of which was reported to prevent both the repression of

translation in the AGO2 tethering assay and the binding of AGO2 to 7-methylguanosine triphosphate (m

GTP)-

Sepharose

. D. melanogaster contains only one GW182 protein

78,136

but there are three GW182 paralogues in mammals

(known as TNRC6A–C). One related protein, AIN1, is expressed in Caenorhabditis elegans

. N-GW, M-GW and C-GW are

regions enriched in glycine (G)–tryptophan (W) dipeptides. The N-GW domain or shorter GW-repeats-containing

peptides were shown to mediate interaction of GW182 proteins with the PIWI domain of AGO proteins

78,94

. The region

extending from the N terminus to the glutamine-rich domain is responsible for targeting GW182 to P-bodies

. DUF,

domain of unknown function; RBD, RNA binding domain; UBA, ubiquitin associated domain; Q; glutamine.

REVIEWS

NATURE REVIEWS

GENETICS VOLUME 9

FEBRUARY 2008

107

or endogenous miRNAs

66,67

, the repressed mRNAs

associated with active polysomes — as demonstrated by

sensitivity of the polysomes to different conditions that

inhibit translation. Moreover, Peterson et al.

found that,

like the cap-dependent upstream ORF, IRES-mediated

translation of the downstream ORF in the bi-cistronic

reporter is sensitive to miRNAs. Drawing on additional

data, the authors proposed a drop-off model, in which

miRNAs render ribosomes prone to premature termina-

tion of translation. Lytle et al. also reported repression of

IRES-containing reporters

The observation that three endogenous miRNAs and

KRAS mRNA, a known target of let‑7 miRNA, cosedi-

ment with polysomes led Maroney et al.

to conclude

that repression occurs at a post-initiation step. Because

puromycin or hypertonic conditions — factors causing

general inhibition of translation — shifted polysome-

associated miRNAs towards the top of the gradient

during polysome gradient analyses, whereas the shift

of KRAS mRNA was only partial, the authors pro-

posed that miRNAs decelerate translation elongation.

Cosedimentation of a significant fraction of cellular

miRNAs or AGO proteins with polysomes has also

been reported in other studies

69–71

and is often quoted

in support of the post-initiation mechanism. However,

it should be stressed that repression of mRNA targets by

miRNAs is generally only partial, and binding of a single

miRNP to mRNA frequently has no significant effect

(see REFS 43,51 for examples). Hence, cosedimentation of

miRNPs with polysomes is not necessarily diagnostic

of post-initiation repression, but might simply reflect

the association of miRNPs with mRNAs undergoing

productive translation.

How miRNAs could modulate the elongation or

termination process remains unclear. Apart from the

proposed miRNA-mediated control, few other exam-

ples of regulation targeting post-initiation steps have

been reported. Repression of pal1 mRNA by GLD1 in

C. elegans seems to involve the stalling or slowing down

of elongating ribosomes

, as does translational repres-

sion of unspliced HAC1 mRNA in yeast

. Other exam-

ples include nanos and oskar mRNAs in D. melanogaster

embryos

74,75

, although the proposed mode of their

regulation has recently been either reinterpreted or

questioned

46,76

. Despite undeniable evidence that

translational repression by miRNAs can occur by post-

initiation mechanisms, the findings do not demonstrate

unequivocally that the initiation and post-initiation

mechanisms are mutually exclusive. It is possible that ini-

tiation is always inhibited, but when the elongation step

is also repressed, ribosomes would queue on the mRNA,

thereby masking the effect of an initiation block.

The association of repressed mRNAs with transla-

tionally competent polysomes has also fuelled specula-

tions that proteins are continually synthesized from these

mRNAs but do not accumulate because they are rapidly

degraded by proteases recruited by miRNPs

(FIG. 3).

This possibility has been experimentally addressed; in

immunoprecipitation experiments, nascent polypep-

tides produced from the repressed reporter could not

be detected

. Likewise, in pulse-labelling experiments,

neither full-length nor nascent polypeptides could be

identified when the reporter mRNA was repressed

On the other hand, repression was not prevented when

reporter proteins were targeted to the endoplasmic

reticulum (ER). This excludes the possibility that nascent

proteins are degraded in the cytosol

. In conclusion,

the proteolysis proposal is at present based on negative

rather than positive data. Proteasome inhibitors had no

effect on miRNA-mediated repression

43,66,67

and other

proteases have not been identified.

mRNA deadenylation and decay

Although initial studies suggested that the levels of

miRNA-inhibited mRNAs remain mostly unchanged,

more recent work has demonstrated that the repres-

sion of many miRNA targets is frequently associated

with their destabilization

56,77–80

(FIG. 3). Likewise, micro-

array studies of transcript levels in cells and tissues

in which the miRNA pathway was inhibited

78,79,81–83

, or in

which miRNA levels were experimentally altered

84–87

revealed marked changes in the abundance of dozens of

validated or predicted miRNA targets, consistent with a

role for miRNAs in mRNA destabilization.

In eukaryotes, mRNA degradation can follow two

pathways, each of which is initiated by a gradual short-

ening of the mRNA poly(A) tail. The mRNA body can

then be degraded by progressive 3′→5′ decay, which is

catalysed by the exosome, or by the removal of the cap

followed by 5′→3′ degradation, which is catalysed by

the exonuclease XRN1

(REF. 88). Levels of mRNA are

controlled by mRNPs through the recruitment of decay

machinery components, leading to mRNA deadenyla-

tion and decapping. The degradation, or at least its final

steps, is thought to occur in P-bodies — cellular struc-

tures that are enriched in mRNA-catabolizing enzymes

and translational repressors

17,89

(BOX 4).

The mechanism of miRNA-mediated mRNA desta-

bilization is best understood in D. melanogaster. Studies

in D. melanogaster S2 cells demonstrated that the P-body

protein GW182 (product of the gawky gene), which

interacts with the miRNP Argonaute1 (the interaction

also occurs between mammalian and worm ortho-

logues

90–93

), is a key factor that marks mRNAs for decay

The AGO PIWI domain and glycine–tryptophan (GW)

dipeptide-containing domains or peptides of GW182

family proteins are important for this interaction

78,94

(FIG. 2). Consistent with its role in mediating mRNA

degradation, GW182 depletion leads to an upregulation

of many mRNA targets that are also upregulated in cells

that are depleted of Argonaute1. Tethering of GW182

to the mRNA bypasses the Argonaute1 requirement for

repression, further demonstrating that GW182 func-

tions in the same pathway downstream of Argonaute1.

Depletion of the components of the CCR4–NOT

deadenylating complex prevents the decay-promoting

activity of GW182, suggesting that it plays a part in

recruiting CCR4–NOT to repressed mRNAs. Likewise,

the knockdown of the decapping-complex proteins,

DCP1 and DCP2, or different combinations of decapping

activators, prevents miRNA-mediated degradation but

leads to an accumulation of deadenylated mRNAs

78,83

REVIEWS

108

FEBRUARY 2008

VOLUME 9 www.nature.com/reviews/genetics

Accelerated deadenylation also results in a reduced

abundance of miRNA-repressed mRNAs in mammalian

cells

. Moreover, knockdown experiments in C. elegans

and analysis of the decay intermediates originating from

repressed mRNAs in worms

and mammalian cells

56,82

support the role of decapping and 5′→3′ exonucleolytic

activities in these systems.

Widespread miRNA-mediated deadenylation

of mRNAs occurs during zebrafish embryogenesis.

The miRNA miR‑430 facilitates the removal of hundreds

of maternal mRNAs by inducing their deadenylation and

subsequent decay at the onset of zygotic transcription

Interestingly, some miR‑430 targets, such as nanos1 and

tudor-like tdrd7 mRNAs, are repressed by miR‑430 in

somatic but not germ cells, indicating that target destabi-

lization and/or repression can be tissue or cell specific

Likewise, mRNA reporters targeted by let‑7 miRNA are

destabilized to different degrees in different mammalian

cell lines

Although many of the mRNAs that are targeted by

miRNAs undergo substantial destabilization, there are

also numerous examples of repression at the transla-

tional level, with no or only a minimal effect on mRNA

decay (Supplementary information S1 (table)). Studies

using D. melanogaster S2 cells identified some endog-

enous or reporter miRNA targets, for which repression

could be entirely accounted for by either mRNA

degradation or translational repression, or by a com-

bination of both processes

78,83

. It is not known what

determines whether an mRNA follows the degradation or

translational-repression pathway. Accessory proteins

bound to the 3′ UTR might be involved, or structural sub-

tleties of imperfect miRNA–mRNA duplexes, particularly

of their central regions, could be important

82,96

Whether the deadenylation and the ensuing decay

are primary or secondary to the translational repres-

sion remains unknown. Clearly, the association of

AGO instead of eIF4E with the m

G cap would not

only prevent effective recruitment of ribosomes, but

would also disrupt the circularization of the mRNA,

probably rendering the poly(A) tail more vulnerable

to exonucleolytic degradation. Experiments that have

been carried out to explore whether deadenylation is

a primary or secondary event have not proved to be

conclusive. Reporter mRNAs that are repressed by

either oligonucleotides that are complementary to the

AUG codon or the 5′ UTR hairpins do not undergo

deadenylation unless they contain miRNA sites

79,80

However, it is unlikely that mRNA circularization is

disrupted by the oligonucleotide or the hairpin, both

of which act at some distance from the cap. By contrast,

the disruption could be effected by the miRNP AGO

Nature Reviews | Genetics

AAAAA

miRNPs

(mRNA storage

or degradation)

Proteolysis

(degradation of nascent peptide)

AAAAA

Elongation block

(slowed elongation or ribosome ‘drop-off’)

P-body

miRNPs

eIF4E

miRNPs

ORF

Initiation block

(repressed cap recognition or 60S joining)

Deadenylation

(followed by decapping and degradation)

CCR4–

NOT

ORF AAAAA

miRNP binding

miRNPs

Figure 3 | Possible mechanisms of the microRNA-mediated post-transcriptional gene repression in animal cells.

Binding of micro-ribonucleoproteins (miRNPs), possibly complexed with accessory factors, to mRNA 3′ UTR can induce

deadenylation and decay of target mRNAs

56,78,79,83

(upper left). Alternatively, miRNPs can repress translation initiation

at either the cap-recognition stage

43,44,53–55

or the 60S subunit joining stage

(bottom left). mRNAs repressed by

deadenylation or at the translation-initiation stage are moved to P-bodies for either degradation or storage. The

repression can also occur at post-initiation phases of translation

66–68

, owing to either slowed elongation or ribosome

‘drop-off’ (bottom right). Proteolytic cleavage of nascent polypeptides was also proposed as a mechanism of the

miRNA-induced repression of protein production

(upper-right). A protease (X) that might be involved in the process

has not been identified. The 7-methylguanosine cap is represented by a red circle. eIF4E, eukaryotic initiation factor 4E.

REVIEWS

NATURE REVIEWS

GENETICS VOLUME 9

FEBRUARY 2008

109

interacting with the cap. Perhaps the strongest support

for deadenylation as a primary event comes from the

finding that the translationally inactive ApppN-capped

mRNA (which does not interact with eIF4E and hence is

unable to circularize) is deadenylated when injected into

zebrafish embryos only when it contains miR‑430 sites

in its 3′ UTR

. This, and other experiments

56,79,83

, indi-

cate that miRNA-dependent mRNA deadenylation and

decay is not dependent on active translation, although

examples of mRNA targets, decay of which requires

ongoing translation, have also been reported

Compartmentalization of miRNA repression

Translationally inactive eukaryotic mRNAs gener-

ally assemble into repressive mRNPs that accumulate

in discrete cytoplasmic foci known as P-bodies or

GW-bodies

17,89

. Another type of aggregate that contains

repressed mRNAs are stress granules (SGs), which accu-

mulate in response to various stress conditions

(BOX 4).

Originally considered as being primarily involved in

mRNA degradation

17,89

, P-bodies are now known to

also be temporary sites of storage for repressed mRNAs

in yeast and mammals

45,98,99

. The demonstration that

AGO proteins, miRNAs and mRNAs repressed by

miRNAs are all enriched in P-bodies

43,45,78,92,93,100,101

impli-

cated P-bodies in miRNA repression and in the fate of

repressed mRNAs. Relevant data are emerging, although

their interpretation is sometimes difficult owing to the

lack of a precise definition of P-bodies (microscopically-

visible versus submicroscopic) and limited information

on the distribution of miRNP components and other

factors between P-bodies and the cytosol.

There is a good correlation between miRNA-

mediated translational repression and accumulation

of mRNAs in visible P-bodies

43,45,100,102

. Moreover, there

is an inverse relationship between P-body localization

and polysome association of target mRNAs in mam-

malian cells

43,45,102

. The endogenous CAT1 mRNA, a

target of miR‑122, localizes to P-bodies when transla-

tion is repressed but not when it is reversed by stress.

In addition, transfection of miR‑122 into cells that

normally do not express it is sufficient to concentrate

CAT1 mRNA in P-bodies

. So far, quantitative data

on the cytosolic distribution of P-bodies are available

only for let‑7 miRNA and its reporter mRNA target,

both of which are ectopically expressed in HeLa cells.

Approximately 20% of each RNA was localized to visible

P-bodies

, indicating that the repression either involves

submicroscopic P-bodies or occurs outside them. Note

also that the knockdown of some P-body components

(such as LSM1 or LSM3), which results in dispersion

of microscopic P-bodies, has no effect on miRNA func-

tion

103,104

. Hence, the microscopically visible P-bodies

are not essential for repression, and their formation is

a consequence rather than the cause of silencing

43,103

These data are consistent with the recent analysis of yeast

cells that demonstrate that submicroscopic mRNPs, con-

taining a set of core P-body components, are sufficient

for basic control of translation repression and mRNA

decay

105

. In contrast to knockdowns of LSM1 and LSM3,In contrast to knockdowns of LSM1 and LSM3,

depletion of other P-body components such as DCP1 or of other P-body components such as DCP1 or

DCP2, GW182, and various decapping activators, either

individually (for example, RCK/p54) or in combinations,

prevents efficient inhibition of target mRNAs in cultured

cells

78,83,91–93,104,106,107

Notwithstanding the above findings, a functional

miRNA pathway is clearly essential for the formation

of large P-body aggregates. Global inhibition of miRNA

biogenesis or depletion of the proteins that are involved

in miRNA repression, such as GW182 or Argonaute1,

results in strong dispersal of visible P-bodies in mam-

malian and D. melanogaster S2 cells

103,108

. Interestingly,

depletion of Dicer2 or Argonaute2, which are involvedArgonaute2, which are involved2, which are involved

Box 4 | P-bodies and stress granules

P-bodies (also known as

GW-bodies) are discrete

granules that are

localized in the

cytoplasm of eukaryotic

cells. They are enriched

in proteins that are

involved in mRNA

catabolism

(deadenylation,

decapping and mRNA

degradation) and

translational

repression

17,89,97

. The core

P-body components,

conserved from budding

yeast (which are devoid

of RNA-silencing

pathways) to mammals,

include the decapping

enzyme complex DCP1–DCP2, the decapping activators RCK/p54 (Dhh1 in yeast),

Pat1 (or the Drosophila melanogaster orthologue CG5208, also known as HPat),

RAP55 (Scd6 in yeast), and EDC3 (Edc3 in yeast) and the heptameric LSm1–7

complex. Metazoa contain yet another decapping activator, Ge-1 or Hedls. P-bodies

also contain other mRNA decay enzymes: the deadenylase complex CAF1–CCR4–

NOT and the 5′→3′ exonuclease XRN1 (REFS 17,89). Some proteins involved in

nonsense-mediated mRNA decay (NMD) and other mRNA degradation pathways are

also enriched in P-bodies. P-bodies lack ribosomes and all translation initiation

factors with the exception of eukaryotic initiation factor (eIF) 4E. However, eIF4G

and Pab1 accumulate in P-bodies under specific repressive conditions in yeast

In metazoa, P-bodies are enriched in proteins participating in miRNA repression —

Argonaute (AGO) proteins and GW182 — and miRNAs themselves. Consistent with

their localization, AGO and GW182 proteins and miRNAs interact, directly or

indirectly, with different P-body components

43,78,90,92,93,100,104,108

(see figure). The

decapping activators RCK/p54 and Pat1, and another P-body-resident protein 4E-T,

have the ability to repress translation, with some affecting the initiation step. These

proteins can contribute to the repressive function of miRNAs

78,83,91–93,104,106,107,109

P-bodies are highly dynamic structures, fluctuating in size and number during the

cell cycle and in response to changes in the translational status of the cell. They

require a continuous supply of repressed mRNAs, and a global translation-initiation

block leads to an increase in P-body size in yeast and metazoa; inhibition of

elongation by cycloheximide, which retains mRNAs on polysomes, results in their

dispersion

17,89

. Likewise, depletion of some P-body components has a strong effect on

their integrity, at least as visualized by microscopy. The mRNAs targeted to P-bodies

either undergo degradation or are stored there for future use.

Stress granules (SGs)

are another type of mRNA-containing cytoplasmic

aggregates, formed in response to global repression of translation initiation or to

various stress conditions. Many proteins found in P-bodies are absent from SGs and

vice versa. However, they share some proteins, and P-bodies and SGs are frequently

located adjacent to each other, possibly exchanging their cargo material

115,135

Nature Reviews | Genetics

10 µm

GW182

RCK/p54

Co-localization

REVIEWS

110

FEBRUARY 2008

VOLUME 9 www.nature.com/reviews/genetics

in RNAi, also results in dispersion of large P-bodies

in D. melanogaster cells, arguing for a role of both

RNA-silencing pathways in P-body formation

103

Most P-body components, including AGO proteins,

are also found throughout the cytosol

. Hence, it is prob-

able that repression by miRNPs is initiated in the cytosol

(or at least outside P-bodies) and that the repressed

mRNAs form P-body aggregates, either small or large,

upon run-off from the ribosomes. P-body-residentP-body-resident-body-residentbody-resident

proteins such as RCK/p54 (and the yeast orthologue

Dhh1)

104,109–111

, 4E-T

112,113

, Pat1 (and the D. melanogaster

orthologue HPat1)

103,109

and RAP55 (REF. 114) have anhave an

established inhibitory activity on translation, some at

the initiation step. These proteins, as well as GW182

(GW182 functions as a translational repressor in additionGW182 functions as a translational repressor in addition in addition

to recruiting the CAF1–CCR4–NOT deadenylaseCAF1–CCR4–NOT deadenylase

),,

can assist miRNPs in initiating the repression. Whereas

RCK/p54 and GW182 can be enrolled directly through

their interaction with AGO proteins

78,94,104

, recruitment

of others might occur through RCK/p54 or GW182.

Surprisingly, only ~1.3% of enhanced GFP (EGFP)-

tagged AGO2 localized to P-bodies in HeLa cells

101

Moreover, the P-body-associated EGFP–AGO2

exchanged with the cytoplasm at a much slower rate

than DCP1–DCP2 or LSM6, the P-body components

involved in decapping

101,113

; GW182 also exchanges

slowly at P-bodies

115

. Rationalization of these observa-

tions is difficult at present. P-bodies could consist of

compartments with differing component dynamics

43,116

Alternatively, miRNPs and associated proteins, such

as GW182, could be ‘anchored’ to some cytoplasmic

structures and not be readily available for diffusion

into the pre-existing photo-bleached P-bodies. In

support of this model, most cellular AGO proteins

fractionate with the ER or Golgi

117,118

. Moreover, fol-

lowing permeabilization of the plasma membrane, only

a small fraction of AGO2 is readily extractable and is

probably cytosolic

. However, the observation that the

EGFP–AGO2 that accumulates in SGs following stress,

or treatment with initiation inhibitors, exchanges rap-

idly with the cytosolic AGO2

pool

101

is at odds with the

anchoring model. Association of AGO proteins with

mRNAs stored in P-bodies, but not those undergoing

degradation, could be another explanation for their low

enrichment in these structures.

Leung et al.

101

found that, in addition to AGO pro-

teins, miRNA mimics and the repressed reporter mRNA

accumulate in SGs. Moreover, the localization of AGO

proteins to SGs but not P-bodies was miRNA-dependant.

Because SGs are now known to form not only in

response to stress but also following general inhibi-

tion of translational initiation

18,119

, SGs (like P-bodies)

might have a role in the miRNA-mediated regula-

tion of translation

18,101

. Alternatively, localization of

miRNP components to SGs might reflect dragging

of the mRNA-associated, but not necessarily inhibitory,

miRNPs to SGs that are formed in response to general

translational inhibition. This scenario could also explain

why the localization of AGO to SGs, but not to P-bodies,

is miRNA dependent: AGO proteins directly interact

with other P-body components

78,94,104

but their localiza-

tion to SGs might require assembly into miRNP to allow

association with mRNA by base pairing.

Reversibility of miRNA-mediated repression

Recent findings indicate that under certain conditions,

or in specific cells, miRNA-mediated repression can be

effectively reversed or prevented

45,95,120,121

, and miRNPs

or their components can even act as translational acti-

vators

. The ability to disengage miRNPs from the

repressed mRNA, or render them stimulatory, makes

miRNA regulation much more wide-ranging and

dynamic.

In human hepatoma cells, CAT1 mRNA is trans-

lationally repressed by the liver-specific miR‑122 and

accumulates in P-bodies. Following amino-acid starva-

tion or other types of stress, CAT1 mRNA is released

from P-bodies and recruited to polysomes, in a proc-

ess that depends on binding of ELAVL1 (also known

as HuR), a member of the embryonic lethal abnormal

vision (ELAV) protein family, to the CAT1 3′ UTR.

APOBEC3G (apolipoprotein B mRNA-editing enzyme

catalytic polypeptide-like 3G), also interferes with the

miRNA action, possibly by altering the distribution of

target messages between P-bodies and polysomes

102

Other examples of the reversible action of miRNAs

have been reported in neuronal cells. In neurons, some

mRNAs are transported along the dendrites as repressed

mRNPs to become translated at dendritic spines upon

synaptic activation

122

. miR‑134 is implicated in the regu-

lation of LIMK1, a protein kinase that is important for

the development of the spine. In response to extracel-

lular stimuli, miR‑134-mediated repression of Limk1

mRNA is partially relieved at dendritic spines of

rat neurons

120

. In D. melanogaster, stimulation of olfac-

tory neurons is associated with proteolysis of the armitage

(armi) protein, which is essential for the assembly of

miRNPs. Following armi degradation, mRNAs that are

normally repressed by miRNAs become translated at

the synapse

121

. Given that many miRNAs are specifically

expressed in the brain

123

, and that three of the four mam-

malian ELAV proteins — ELAV2, ELAV3 and ELAV4

Nature Reviews | Genetics

ORF

AAAAAA

RBP X

RBP Y

miRNP

CCR4–

NOT

Figure 4 | Possible interplay between RNA binding proteins and micro-

ribonucleoproteins interacting with the mRNAs 3′ UTR. A single mRNA can

have several cis-acting motifs interacting with different RNA binding proteins (RBPs)

and micro-ribonucleoproteins (miRNPs), which together will determine mRNA

translatability or stability. The suppressive effect of the 3′ UTR-binding protein

ELAV1 on the miRNA-mediated repression (not shown) has recently been

documented

. However, it is possible that RBPs will also interact with miRNPs to

augment their repressive function, and that miRNPs will have a positive or negative

effect on the activity of RBPs bound at the 3′ UTR. The 7-methylguanosine cap is

represented by a red circle in the figure.

REVIEWS

NATURE REVIEWS

GENETICS VOLUME 9

FEBRUARY 2008

111

(also known as HuB, HuC and HuD) — are restricted to

neurons

124

, reversible miRNA regulation might have a

general role in brain development and function.

It is likely that RNA-binding proteins (RBPs) other

than ELAV proteins act as modifiers of miRNA-mediated

repression. miR‑430 repression of nanos1 and tdrd7

mRNAs in somatic but not germline cells can be attrib-

uted to a specific 3′ UTR-binding protein that prevents

miR‑430 function in germline cells

. Intriguingly,

together with the FMRP-related protein FXR1, AGO2

(possibly as a part of an miRNP) acts as an activator of

translation when bound to the 3′ UTR of tumour necro-

sis factor-α mRNA in serum-starved human cells

. This

finding possibly reveals one of many potential combina-

tions of the interplay between the miRNPs and RBPs that

are interacting with mRNA 3′ UTRs. Because RBPs such

as ELAV1 can act as translational activators by interfer-

ing with the miRNP-mediated repression of translation,

it is also possible that miRNPs might act as translational

activators by either displacing or modulating inhibitory

RBPs bound at the 3′ UTR. Likewise, in other circum-

stances, miRNPs and RBPs might act synergistically to

either repress or activate mRNA translation (FIG. 4).

Conclusions and prospects

Perhaps the paramount open question is whether

miRNAs inhibit protein synthesis by a primary single

mechanism or by different mechanisms. In other words, is

it possible that miRNAs trigger an initial event that

is then amplified by different mechanisms? On the

basis of the many lines of evidence, it is widely believed

that miRNAs suppress protein synthesis by a bevy of

mechanisms. Although this could be the case, it is too

early to draw this conclusion with certainty. A simple,

alternative mechanistic model posits that the earliest

event in protein-synthesis repression is the inhibition of

cap-dependent translation through the binding of AGO

to the cap structure. Secondary effects of this inhibition

could then be manifested at other steps, such as mRNA

degradation or proteolysis of the nascent polypeptide

chains. It is conceivable that the different outcomes

of the miRNA repression experiments occur partially

because of the different experimental systems and

methodologies. Although the use of in vitro systems

allows identification and biochemical characterization

of early events during repression, the reporter mRNAs

lack a nuclear history that could involve deposition of

RBPs that modify the mRNA properties and affect the

response to miRNAs. The same applies to the in vitro

transcribed mRNAs that were transfected into cultured

cells. Indeed, differences in the outcome of miRNA-

mediated repression have been reported, depending

on whether RNA or DNA was used for transfection, or

even on the method of transfection

. Finally, it should

be recognized that the steps that limit protein expression

can differ among different transfected reporter genes or

in vitro transcribed mRNAs

It will be crucial to understand the regulation of

miRNA function through modulation of the activity

of RISC components and associated factors, possibly

by phosphorylation and other protein modifications.

Thus, the involvement of different signalling pathways

in the control of miRNA function should be studied. It

is also highly likely that the mechanisms that control

translation initiation will have a significant impact on

miRNA-regulated gene expression. It will also be impor-

tant to determine the precise contributions of different

cellular structures, such as P-bodies and SGs, to miRNA-

mediated repression of translation. The fact that miRNA

function can be recapitulated in cell-free extracts argues

against a primary and essential role of P-bodies and SGs in

miRNA repression, inasmuch as these microscopic struc-

tures are unlikely to exist in cell-free extracts. However,

pseudo-polysomes that are formed in extracts from

D. melanogaster embryos

might contain constituents

of P-bodies, and it will be interesting to find out if this

is indeed the case. The availability of cell-free systems

to study miRNA function is a significant development.

It is hoped that these systems will generate a detailed

and precise mechanistic picture of the miRNA-mediated

inhibition of protein synthesis, as has been accomplished

for transcription, translation and splicing.

Finally, a complete and accurate understanding of the

mechanism of miRNA function will require elucidation

of three-dimensional structures of animal AGO proteins,

their complexes with the miRNA and the cap structure,

and ultimately the structure of miRNP bound to mRNA.

Structural information would help validate or refute the

current models for miRNA function.

Note added in proof

Two papers have recently appeared which add new

information about the miRNA-mediated repression.

Vasudevan et al.

137

showed that while miRNAs repress

translation in proliferating mammalian cells, they

induce translation upregulation of target mRNAs upon

cell-cycle arrest. Kedde et al.

138

identified dead end 1

(Dnd1) as a protein that, by binding to the target mRNA

3′UTR, counteracts the function of miRNAs in human

cells and zebrafish germ cells.

1. Bushati, N. & Cohen, S. M. microRNA functions.

Annu. Rev. Cell Dev. Biol. 23, 175–205 (2007).

2. Kloosterman, W. P. & Plasterk, R. H. The diverse

functions of microRNAs in animal development and

disease. Dev. Cell 11, 441–450 (2006).

3. Rana, T. M. Illuminating the silence: understanding the

structure and function of small RNAs. Nature Rev.

Mol. Cell Biol. 8, 23–36 (2007).

4. Molnar, A., Schwach, F., Studholme, D. J.,

Thuenemann, E. C. & Baulcombe, D. C. miRNAs

control gene expression in the single-cell alga

Chlamydomonas reinhardtii. Nature 447, 1126–1129

(2007).

5. Zhao, T. et al. A complex system of small RNAs in the

unicellular green alga Chlamydomonas reinhardtii.

Genes Dev. 21, 1190–1203 (2007).

6. Esquela-Kerscher, A. & Slack, F. J. Oncomirs —

microRNAs with a role in cancer. Nature Rev. Cancer

6, 259–269 (2006).

7. Chang, T. C. & Mendell, J. T. microRNAs in

vertebrate physiology and human disease.

Annu. Rev. Genomics Hum. Genet. 8, 215–239

(2007).

8. Krutzfeldt, J. & Stoffel, M. microRNAs: a new

class of regulatory genes affecting metabolism.

Cell Metab. 4, 9–12 (2006).

9. Bao, N., Lye, K. W. & Barton, M. K. MicroRNA

binding sites in Arabidopsis class III HD-ZIP

mRNAs are required for methylation of the

template chromosome. Dev. Cell 7, 653–662

(2004).

10. Valencia-Sanchez, M. A., Liu, J., Hannon, G. J. &

Parker, R. Control of translation and mRNA

degradation by miRNAs and siRNAs. Genes Dev. 20,

515–524 (2006).

11. Pillai, R. S., Bhattacharyya, S. N. & Filipowicz, W.

Repression of protein synthesis by miRNAs: how

many mechanisms? Trends Cell Biol. 17, 118–126

(2007).

REVIEWS

112

FEBRUARY 2008

VOLUME 9 www.nature.com/reviews/genetics

12. Standart, N. & Jackson, R. J. MicroRNAs repress

translation of m7Gppp-capped target mRNAs in vitro

by inhibiting initiation and promoting deadenylation.

Genes Dev. 21, 1975–1982 (2007).

13. Jackson, R. J. & Standart, N. How do microRNAs

regulate gene expression? Sci. STKE 2007, re1 (2007).

14. Nilsen, T. W. Mechanisms of microRNA-mediated gene

regulation in animal cells. Trends Genet. 23, 243–249

(2007).

15. Hwang, H. W., Wentzel, E. A. & Mendell, J. T. A

hexanucleotide element directs microRNA nuclear

import. Science 315, 97–100 (2007).

16. Valadi, H. et al. Exosome-mediated transfer of mRNAs

and microRNAs is a novel mechanism of genetic

exchange between cells. Nature Cell Biol. 9, 654–659

(2007).

17. Eulalio, A., Behm-Ansmant, I. & Izaurralde, E.

P-bodies: at the crossroads of post-transcriptional

pathways. Nature Rev. Mol. Cell Biol. 8, 9–22 (2007).

18. Leung, A. K. & Sharp, P. A. Function and localization

of microRNAs in mammalian cells. Cold Spring Harb.

Symp. Quant. Biol. 71, 29–38 (2006).

19. Sontheimer, E. J. Assembly and function of RNA

silencing complexes. Nature Rev. Mol. Cell Biol. 6,

127–138 (2005).

20. Bartel, D. P. MicroRNAs: genomics, biogenesis,

mechanism and function. Cell 116, 281–297 (2004).

21. Du, T. & Zamore, P. D. microPrimer: the biogenesis

and function of microRNA. Development 132,

4645–4652 (2005).

22. Kim, V. N. & Nam, J. W. Genomics of microRNA.

Trends Genet. 22, 165–173 (2006).

23. Filipowicz, W., Jaskiewicz, L., Kolb, F. A. & Pillai, R. S.

Post-transcriptional gene silencing by siRNAs and

miRNAs. Curr. Opin. Struct. Biol. 15, 331–341 (2005).

24. Peters, L. & Meister, G. Argonaute proteins: mediators

of RNA silencing. Mol. Cell 26, 611–623 (2007).

25. Tolia, N. H. & Joshua-Tor, L. Slicer and the argonautes.

Nature Chem. Biol. 3, 36–43 (2007).

26. Liu, J. et al. Argonaute2 is the catalytic engine of

mammalian RNAi. Science 305, 1437–1441 (2004).

27. Pillai, R. S., Artus, C. G. & Filipowicz, W. Tethering of

human Ago proteins to mRNA mimics the miRNA-

mediated repression of protein synthesis. RNA 10,

1518–1525 (2004).

28. Meister, G. et al. Human Argonaute2 mediates RNA

cleavage targeted by miRNAs and siRNAs. Mol. Cell

15, 185–197 (2004).

29. Jones-Rhoades, M. W., Bartel, D. P. & Bartel, B.

MicroRNAS and their regulatory roles in plants.

Annu. Rev. Plant Biol. 57, 19–53 (2006).

30. Doench, J. G. & Sharp, P. A. Specificity of microRNA

target selection in translational repression. Genes Dev.

18, 504–511 (2004).

31. Brennecke, J., Stark, A., Russell, R. B. & Cohen, S. M.

Principles of microRNA-target recognition. PLoS Biol.

3, 404–418 (2005).

32. Lewis, B. P., Burge, C. B. & Bartel, D. P. Conserved

seed pairing, often flanked by adenosines, indicates

that thousands of human genes are microRNA targets.

Cell 120, 15–20 (2005).

33. Grimson, A. et al. MicroRNA targeting specificity

in mammals: determinants beyond seed pairing.

Mol. Cell 27, 91–105 (2007).

34. Nielsen, C. B. et al. Determinants of targeting by

endogenous and exogenous microRNAs and siRNAs.

RNA 13, 1894–1910 (2007).

35. Lytle, J. R., Yario, T. A. & Steitz, J. A. Target mRNAs

are repressed as efficiently by microRNA-binding sites

in the 5′ UTR as in the 3′ UTR. Proc. Natl Acad. Sci.

USA 104, 9667–6972 (2007).

36. Kloosterman, W. P., Wienholds, E., Ketting, R. F. &

Plasterk, R. H. Substrate requirements for let-7

function in the developing zebrafish embryo. Nucleic

Acids Res. 32, 6284–6291 (2004).

37. Easow, G., Teleman, A. A. & Cohen, S. M. Isolation

of microRNA targets by miRNP immunopurification.

RNA 13, 1198–1204 (2007).

38. Merrick, W. C. Cap-dependent and cap-independent

translation in eukaryotic systems. Gene 332, 1–11

(2004).

39. Kapp, L. D. & Lorsch, J. R. The molecular mechanics

of eukaryotic translation. Annu. Rev. Biochem. 73,

657–704 (2004).

40. Wells, S. E., Hillner, P. E., Vale, R. D. & Sachs, A. B.

Circularization of mRNA by eukaryotic translation

initiation factors. Mol. Cell 2, 135–140 (1998).

41. Derry, M. C., Yanagiya, A., Martineau, Y. &

Sonenberg, N. Regulation of poly(A)-binding protein

through PABP-interacting proteins. Cold Spring Harb.

Symp. Quant. Biol. 71, 537–543 (2006).

42. Jackson, R. J. Alternative mechanisms of initiating

translation of mammalian mRNAs. Biochem. Soc.

Trans. 33, 1231–1241 (2005).

43. Pillai, R. S. et al. Inhibition of translational initiation

by let-7 microRNA in human cells. Science 309,

1573–1576 (2005).

44. Humphreys, D. T., Westman, B. J., Martin, D. I. &

Preiss, T. MicroRNAs control translation initiation

by inhibiting eukaryotic initiation factor 4E/cap and

poly(A) tail function. Proc. Natl Acad. Sci. USA 102,

16961–16966 (2005).

This paper and reference 43 provide the first

evidence that miRNAs repress translational

initiation, probably by interfering with the function

of the cap-binding factor eIF4E.

45. Bhattacharyya, S. N., Habermacher, R., Martine, U.,

Closs, E. I. & Filipowicz, W. Relief of microRNA-

mediated translational repression in human cells

subjected to stress. Cell 125, 1111–1124 (2006).

46. Chekulaeva, M., Hentze, M. W. & Ephrussi, A.

Bruno acts as a dual repressor of oskar translation,

promoting mRNA oligomerization and formation of

silencing particles. Cell 124, 521–533 (2006).

47. Richter, J. D. & Sonenberg, N. Regulation of cap-

dependent translation by eIF4E inhibitory proteins.

Nature 433, 477–480 (2005).

48. Cho, P. F. et al. A new paradigm for translational

control: inhibition via 5′–3′ mRNA tethering by Bicoid

and the eIF4E cognate 4EHP. Cell 121, 411–423

(2005).

49. Kiriakidou, M. et al. An mRNA m

G cap binding-like

motif within human Ago2 represses translation.

Cell 129, 1141–1151 (2007).

This paper reports that human AGO2 has the

potential to directly interact with the m

G cap and

to repress translational initiation by competing

with eIF4E for cap binding.

50. Marcotrigiano, J., Gingras, A. C., Sonenberg, N. &

Burley, S. K. Cocrystal structure of the messenger RNA

5′ cap-binding protein (eIF4E) bound to 7-methyl-GDP.

Cell 89, 951–961 (1997).

51. Doench, J. G., Petersen, C. P. & Sharp, P. A. siRNAs

can function as miRNAs. Genes Dev. 17, 438–442

(2003).

52. Wang, B., Love, T. M., Call, M. E., Doench, J. G. &

Novina, C. D. Recapitulation of short RNA-directed

translational gene silencing in vitro. Mol. Cell 22,

553–560 (2006).

53. Thermann, R. & Hentze, M. W. Drosophila miR2

induces pseudo-polysomes and inhibits translation

initiation. Nature 447, 875–878 (2007).

54. Mathonnet, G. et al. microRNA inhibition of translation

initiation in vitro by targeting the cap-binding complex

eIF4F. Science 317, 1764–1767 (2007).

55. Wakiyama, M., Takimoto, K., Ohara, O. & Yokoyama, S.

Let-7 microRNA-mediated mRNA deadenylation and

translational repression in a mammalian cell-free

system. Genes Dev. 21, 1857–1862 (2007).

References 52–55 describe the characterization of

cell-free extracts, recapitulating many features

of the miRNA-mediated repression established in

mammalian and D. melanogaster cell lines.

56. Wu, L., Fan, J. & Belasco, J. G. microRNAs direct rapid

deadenylation of mRNA. Proc. Natl Acad. Sci. USA

103, 4034–4039 (2006).

57. Chendrimada, T. P. et al. microRNA silencing through

RISC recruitment of eIF6. Nature 447, 823–828

(2007).

This report identifies eIF6 as a potential target of

miRNA-meditated repression. The authors propose

that, by interacting with eIF6, AGO proteins

repress translation by preventing the 60S

ribosomal subunit joining to the 40S initiation

complex.

58. Russell, D. W. & Spremulli, L. L. Identification of

a wheat germ ribosome dissociation factor distinct

from initiation factor eIF-3. J. Biol. Chem. 253,

6647–6649 (1978).

59. Sanvito, F. et al. The β4 integrin interactor p27(BBP/

eIF6) is an essential nuclear matrix protein involved in

60S ribosomal subunit assembly. J. Cell Biol. 144,

823–837 (1999).

60. Si, K. & Maitra, U. The Saccharomyces cerevisiae

homologue of mammalian translation initiation factor

6 does not function as a translation initiation factor.

Mol. Cell Biol. 19, 1416–1426 (1999).

61. Basu, U., Si, K., Warner, J. R. & Maitra, U. The

Saccharomyces cerevisiae TIF6 gene encoding

translation initiation factor 6 is required for 60S

ribosomal subunit biogenesis. Mol. Cell Biol. 21,

1453–1462 (2001).

62. Sachs, A. B. & Davis, R. W. Translation initiation and

ribosomal biogenesis: involvement of a putative rRNA

helicase and RPL46. Science 247, 1077–1079

(1990).

63. Sachs, A. B. & Davis, R. W. The poly(A) binding protein

is required for poly(A) shortening and 60S ribosomal

subunit-dependent translation initiation. Cell 58,

857–867 (1989).

64. Olsen, P. H. & Ambros, V. The lin-4 regulatory RNA

controls developmental timing in Caenorhabditis

elegans by blocking LIN-14 protein synthesis after the

initiation of translation. Dev. Biol. 216, 671–680

(1999).

65. Seggerson, K., Tang, L. & Moss, E. G. Two genetic

circuits repress the Caenorhabditis elegans

heterochronic gene lin-28 after translation initiation.

Dev. Biol. 243, 215–225 (2002).

66. Petersen, C. P., Bordeleau, M. E., Pelletier, J. &

Sharp, P. A. Short RNAs repress translation after

initiation in mammalian cells. Mol. Cell 21, 533–542

(2006).

67. Nottrott, S., Simard, M. J. & Richter, J. D. Human

let-7a miRNA blocks protein production on actively

translating polyribosomes. Nature Struct. Mol. Biol.

13, 1108–1114 (2006).

68. Maroney, P. A., Yu, Y., Fisher, J. & Nilsen, T. W.

Evidence that microRNAs are associated with

translating messenger RNAs in human cells.

Nature Struct. Mol. Biol. 13, 1102–1107 (2006).

References 66 to 68 demonstrate that repressed

mRNAs are associated with actively translating

polyribosomes and argue that miRNAs block

protein synthesis at steps after initiation. The data

supporting this mechanism are also reported in

references 64 and 65.

69. Kim, J. et al. Identification of many microRNAs that

copurify with polyribosomes in mammalian neurons.

Proc. Natl Acad. Sci. USA 101, 360–365 (2004).

70. Nelson, P. T., Hatzigeorgiou, A. G. & Mourelatos, Z.

miRNP: mRNA association in polyribosomes in a

human neuronal cell line. RNA 10, 387–394 (2004).

71. Vasudevan, S. & Steitz, J. A. AU-rich-element-

mediated upregulation of translation by FXR1 and

Argonaute 2. Cell 128, 1105–1118 (2007).

This paper demonstrates that interaction of AGO2

(in a complex with FXR1) with the mRNA 3′ UTR

can, under specific cellular conditions, lead to

upregulation rather than downregulation of

translation.

72. Mootz, D., Ho, D. M. & Hunter, C. P. The STAR–Maxi-

KH domain protein GLD-1 mediates a developmental

switch in the translational control of C. elegans PAL-1.

Development 131, 3263–3272 (2004).

73. Ruegsegger, U., Leber, J. H. & Walter, P. Block of

HAC1 mRNA translation by long-range base pairing

is released by cytoplasmic splicing upon induction of

the unfolded protein response. Cell 107, 103–114

(2001).

74. Clark, I. E., Wyckoff, D. & Gavis, E. R. Synthesis of

the posterior determinant nanos is spatially restricted

by a novel cotranslational regulatory mechanism.

Curr. Biol. 10, 1311–1314 (2000).

75. Braat, A. K., Yan, N., Arn, E., Harrison, D. &

Macdonald, P. M. Localization-dependent oskar

protein accumulation; control after the initiation of

translation. Dev. Cell 7, 125–131 (2004).

76. Tomancak, P. et al. Global analysis of patterns of gene

expression during Drosophila embryogenesis. Genome

Biol. 8, R145 (2007).

77. Bagga, S. et al. Regulation by let-7 and lin-4 miRNAs

results in target mRNA degradation. Cell 122,

553–563 (2005).

78. Behm-Ansmant, I. et al. mRNA degradation by

miRNAs and GW182 requires both CCR4:NOT

deadenylase and DCP1:DCP2 decapping complexes.

Genes Dev. 20, 1885–1898 (2006).

79. Giraldez, A. J. et al. Zebrafish MiR-430 promotes

deadenylation and clearance of maternal mRNAs.

Science 312, 75–79 (2006).

This paper and references 56, 77 and 78 provide

compelling evidence that miRNA can induce

pronounced target mRNA degradation, which is

initiated by removal of the poly(A) tail.

80. Wu, L. & Belasco, J. G. Micro-RNA regulation

of the mammalian lin-28 gene during neuronal

differentiation of embryonal carcinoma cells.

Mol. Cell Biol. 25, 9198–9208 (2005).

81. Rehwinkel, J. et al. Genome-wide analysis of mRNAs

regulated by drosha and Argonaute proteins in

Drosophila melanogaster. Mol. Cell Biol. 26,

2965–2975 (2006).

REVIEWS

NATURE REVIEWS

GENETICS VOLUME 9

FEBRUARY 2008

113

82. Schmitter, D. et al. Effects of Dicer and Argonaute

down-regulation on mRNA levels in human HEK293

cells. Nucleic Acids Res. 34, 4801–4815 (2006).

83. Eulalio, A. et al. Target-specific requirements for

enhancers of decapping in miRNA-mediated gene

silencing. Genes Dev. 21, 2558–2570 (2007).

84. Lim, L. P. et al. Microarray analysis shows that some

microRNAs downregulate large numbers of target

mRNAs. Nature 433, 769–773 (2005).

85. Krutzfeldt, J. et al. Silencing of microRNAs in vivo

with ‘antagomirs’. Nature 438, 685–689 (2005).

86. Esau, C. et al. miR-122 regulation of lipid metabolism

revealed by in vivo antisense targeting. Cell. Metab. 3,

87–98 (2006).

87. Linsley, P. S. et al. Transcripts targeted by the

microRNA-16 family cooperatively regulate cell cycle

progression. Mol. Cell Biol. 27, 2240–2252 (2007).

88. Parker, R. & Song, H. The enzymes and control of

eukaryotic mRNA turnover. Nature Struct. Mol. Biol.

11, 121–127 (2004).

89. Parker, R. & Sheth, U. P bodies and the control of

mRNA translation and degradation. Mol. Cell 25,

635–646 (2007).

90. Ding, L., Spencer, A., Morita, K. & Han, M. The

developmental timing regulator AIN-1 interacts with

miRISCs and may target the argonaute protein ALG-1

to cytoplasmic P bodies in C. elegans. Mol. Cell 19,

437–447 (2005).

91. Liu, J. et al. A role for the P-body component

GW182 in microRNA function. Nature Cell Biol. 7,

1261–1266 (2005).

92. Jakymiw, A. et al. Disruption of GW bodies impairs

mammalian RNA interference. Nature Cell Biol. 7,

1267–1274 (2005).

93. Meister, G. et al. Identification of novel argonaute-

associated proteins. Curr. Biol. 15, 2149–2155 (2005).

94. Till, S. et al. A conserved motif in Argonaute-

interacting proteins mediates functional interactions

through the Argonaute PIWI domain. Nature Struct.

Mol. Biol. 14, 897–903 (2007).

95. Mishima, Y. et al. Differential regulation of germline

mRNAs in soma and germ cells by zebrafish miR-430.

Curr. Biol. 16, 2135–2142 (2006).

96. Aleman, L. M., Doench, J. & Sharp, P. A. Comparison

of siRNA-induced off-target RNA and protein effects.

RNA 13, 385–395 (2007).

97. Anderson, P. & Kedersha, N. RNA granules.

J. Cell Biol. 172, 803–808 (2006).

98. Brengues, M., Teixeira, D. & Parker, R. Movement of

eukaryotic mRNAs between polysomes and cytoplasmic

processing bodies. Science 310, 486–489 (2005).

99. Brengues, M. & Parker, R. Accumulation of

polyadenylated mRNA, Pab1p, eIF4E, and eIF4G with

P-bodies in Saccharomyces cerevisiae. Mol. Biol. Cell

18, 2592–2602 (2007).

100. Liu, J., Valencia-Sanchez, M. A., Hannon, G. J. &

Parker, R. MicroRNA-dependent localization

of targeted mRNAs to mammalian P-bodies.

Nature Cell Biol. 7, 719–723 (2005).

This paper and references 43, 45, 78, 91 and 92

establish the connection between miRNA-mediated

repression and P-bodies by demonstrating that

miRNP components and repressed mRNAs

accumulate in P-bodies and that many P-body

proteins are essential for the repression.

101. Leung, A. K., Calabrese, J. M. & Sharp, P. A.

Quantitative analysis of Argonaute protein reveals

microRNA-dependent localization to stress granules.

Proc. Natl Acad. Sci. USA 103, 18125–18130 (2006).

102. Huang, J. et al. Derepression of micro-RNA-mediated

protein translation inhibition by apolipoprotein B

mRNA-editing enzyme catalytic polypeptide-like 3G

(APOBEC3G) and its family members. J. Biol. Chem.

282, 33632–33640 (2007).

103. Eulalio, A., Behm-Ansmant, I., Schweizer, D. &

Izaurralde, E. P-body formation is a consequence,

not the cause, of RNA-mediated gene silencing.

Mol. Cell Biol. 27, 3970–3981 (2007).

104. Chu, C. Y. & Rana, T. M. Translation repression in

human cells by microRNA-induced gene silencing

requires RCK/p54. PLoS Biol. 4, 1122–1136(2006).

This paper and reference 83 identify the decapping

activators that are associated with P-bodies as

proteins that are essential for miRNA-mediated

repression.

105. Decker, C. J., Teixeira, D. & Parker, R. Edc3p and

a glutamine/asparagine-rich domain of Lsm4p

function in processing body assembly in

Saccharomyces cerevisiae. J. Cell Biol. 179,

437–449 (2007).

106. Rehwinkel, J., Behm-Ansmant, I., Gatfield, D. &

Izaurralde, E. A crucial role for GW182 and the

DCP1:DCP2 decapping complex in miRNA-

mediated gene silencing. RNA 11, 1640–1647

(2005).

107. Barbee, S. A. et al. Staufen- and FMRP-containing

neuronal RNPs are structurally and functionally

related to somatic P bodies. Neuron 52,

997–1009 (2006).

108. Pauley, K. M. et al. Formation of GW bodies is a

consequence of microRNA genesis. EMBO Rep. 7,

904–910 (2006).

109. Coller, J. & Parker, R. General translational

repression by activators of mRNA decapping.

Cell 122, 875–886 (2005).

110. Minshall, N. & Standart, N. The active form of

Xp54 RNA helicase in translational repression is

an RNA-mediated oligomer. Nucleic Acids Res. 32,

1325–1334 (2004).

111. Smillie, D. A. & Sommerville, J. RNA helicase p54

(DDX6) is a shuttling protein involved in nuclear

assembly of stored mRNP particles. J. Cell Sci. 115,

395–407 (2002).

112. Ferraiuolo, M. A. et al. A role for the eIF4E-binding

protein 4E-T in P-body formation and mRNA decay.

J. Cell Biol. 170, 913–924 (2005).

113. Andrei, M. A. et al. A role for eIF4E and eIF4E-

transporter in targeting mRNPs to mammalian

processing bodies. RNA 11, 717–727 (2005).

114. Tanaka, K. J. et al. RAP55, a cytoplasmic

mRNP component, represses translation in

Xenopus oocytes. J. Biol. Chem. 281,

40096–40106 (2006).

115. Kedersha, N. et al. Stress granules and processing

bodies are dynamically linked sites of mRNP

remodeling. J. Cell Biol. 169, 871–884 (2005).

116. Durand, S. et al. Inhibition of nonsense-mediated

mRNA decay (NMD) by a new chemical molecule

reveals the dynamic of NMD factors in P-bodies.

J. Cell Biol. 178, 1145–1160 (2007).

117. Tahbaz, N., Carmichael, J. B. & Hobman, T. C.

GERp95 belongs to a family of signal-transducing

proteins and requires Hsp90 activity for stability

and Golgi localization. J. Biol. Chem. 276,

43294–43299 (2001).

118. Tahbaz, N. et al. Characterization of the interactions

between mammalian PAZ PIWI domain proteins and

Dicer. EMBO Rep. 5, 189–194 (2004).

119. Mazroui, R. et al. Inhibition of ribosome recruitment

induces stress granule formation independently of

eukaryotic initiation factor 2α phosphorylation.

Mol. Biol. Cell 17, 4212–4219 (2006).

120. Schratt, G. M. et al. A brain-specific microRNA

regulates dendritic spine development. Nature 439,

283–289 (2006).

This paper and reference 45 provide the first

evidence that, under specific cellular conditions,

mRNAs can be relieved from the miRNA-mediated

repression and relocate from P-bodies to enter

active translation.

121. Ashraf, S. I., McLoon, A. L., Sclarsic, S. M. & Kunes, S.

Synaptic protein synthesis associated with memory

is regulated by the RISC pathway in Drosophila.

Cell 124, 191–205 (2006).

122. Sutton, M. A. & Schuman, E. M. Dendritic protein

synthesis, synaptic plasticity, and memory. Cell 127,

49–58 (2006).

123. Kosik, K. S. The neuronal microRNA system.

Nature Rev. Neurosci. 7, 911–920 (2006).

124. Lu, J. Y. & Schneider, R. J. Tissue distribution

of AU-rich mRNA-binding proteins involved in

regulation of mRNA decay. J. Biol. Chem. 279,

12974–12979 (2004).

125. Ruby, J. G., Jan, C. H. & Bartel, D. P. Intronic

microRNA precursors that bypass drosha processing.

Nature 448, 83–86 (2007).

126. Okamura, K., Hagen, J. W., Duan, H., Tyler, D. M. &

Lai, E. C. The mirtron pathway generates

microRNA-class regulatory RNAs in Drosophila.

Cell 130, 89–100 (2007).

127. Berezikov, E., Chung, W. J., Willis, J., Cuppen, E. &

Lai, E. C. Mammalian mirtron genes. Mol. Cell 28,

328–336 (2007).

128. Jin, P., Alisch, R. S. & Warren, S. T. RNA and

microRNAs in fragile X mental retardation.

Nature Cell Biol. 6, 1048–1053 (2004).

129. Gaidatzis, D., van Nimwegen, E., Hausser, J. &

Zavolan, M. Inference of miRNA targets using

evolutionary conservation and pathway analysis.

BMC Bioinformatics 8, 69 (2007).

130. Kertesz, M., Iovino, N., Unnerstall, U., Gaul, U.

& Segal, E. The role of site accessibility in

microRNA target recognition. Nature Genet. 39,

1278–1284 (2007).

131. Vella, M. C., Choi, E. Y., Lin, S. Y., Reinert, K. &

Slack, F. J. The C. elegans microRNA let-7 binds

to imperfect let-7 complementary sites from

the lin-41 3′ UTR. Genes Dev. 18, 132–137

(2004).

132. Reinhart, B. J. et al. The 21-nucleotide let-7

RNA regulates developmental timing in

Caenorhabditis elegans. Nature 403, 901–906

(2000).

133. Raught, B. & Gingras, A.-C. in Translational Control

in Biology and Medicine (eds Mathews, M. B.,

Sonenberg, N. & Hershey, J. B.) 369–400 (Cold

Spring Harbor Laboratory Press, Cold Spring

Harbor, 2007).

134. Herbert, T. P. & Proud, C. G. in Translational

Control in Biology and Medicine (eds Mathews,

M. B., Sonenberg, N. & Hershey, J. B.) 601–624

(Cold Spring Harbor Laboratory Press, Cold Spring

Harbor, 2007).

135. Wilczynska, A., Aigueperse, C., Kress, M.,

Dautry, F. & Weil, D. The translational regulator

CPEB1 provides a link between dcp1 bodies

and stress granules. J. Cell Sci. 118, 981–992

(2005).

136. Schneider, M. D. et al. gawky is a component of

cytoplasmic mRNA processing bodies required for

early Drosophila development. J. Cell Biol. 174,

349–358 (2006).

137. Vasudevan S, Tong Y, Steitz J. A. Switching from

repression to activation: microRNAs can up-regulate

translation. Science 318, 1931–1934 (2007).

138. Kedde, M. et al. RNA-binding protein Dnd1

inhibits microRNA access to target mRNA. Cell

131, 1273–1286 (2007).

Acknowledgements

We thank R. Jackson for drawing our attention to the

bypass suppressors of the yeast Pab1 deletion. We thank

H. Grosshans, N. Standart, R. Jackson and members of the

Filipowicz and Sonenberg groups for their comments. S.N.B

is a rec ipien t o f H uman Frontier Sc ience Pro gra m

Organization (HFSPO) long-term fellowship. The Friedrich

Miescher Institute is supported by the Novartis Research

Foundation. N.S. was supported by grants from the HFSPO

and the Canadian Institute of Health Research and is a

Howard Hughes Medical Institute International Scholar.

Research by W.F. is also supported by the EC FP6 Program

‘Sirocco’.

DATABASES

Entrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query.

fcgi?db=gene

HAC1 | KRAS | let‑7 | lin‑4 | lin‑14 | lin‑28 | miR‑430 | pal1 |

tdrd7

FlyBase: http://flybase.bio.indiana.edu

loquacious | pasha

UniProtKB: http://ca.expasy.org/sprot

AGO1 | AGO2 | AGO4 | APOBEC3G | CAT1 | DCP1 | DCP2 |

FXR1 | GLD1 | GW182 | LIMK1 | LSM1 | LSM3 | PABP1 | RN3 |

TRBP | XRN1

FURTHER INFORMATION

Witold Filipowicz’s homepage: http://www.fmi.ch/

SUPPLEMENTARY INFORMATION

See online article: S1 (table)

ALL LINKS ARE ACTIVE IN THE ONLINE PDF

REVIEWS

114

FEBRUARY 2008

VOLUME 9 www.nature.com/reviews/genetics

cbm-34-cbm210456

Article

Jun 2024

Saeed M. El-Nahaas

Preanalytical considerations in quantifying circulating miRNAs that predict end-stage kidney disease in diabetes

Article

Full-text available

Jun 2024

Role of long non-coding RNA in inflammatory bowel disease

Article

Full-text available

Jun 2024

Inflammatory bowel disease (IBD) is a group of recurrent chronic inflammatory diseases, including Crohn’s disease (CD) and ulcerative colitis (UC). Although IBD has been extensively studied for decades, its cause and pathogenesis remain unclear. Existing research suggests that IBD may be the result of an interaction between genetic factors, environmental factors and the gut microbiome. IBD is closely related to non-coding RNAs (ncRNAs). NcRNAs are composed of microRNA(miRNA), long non-coding RNA(lnc RNA) and circular RNA(circ RNA). Compared with miRNA, the role of lnc RNA in IBD has been little studied. Lnc RNA is an RNA molecule that regulates gene expression and regulates a variety of molecular pathways involved in the pathbiology of IBD. Targeting IBD-associated lnc RNAs may promote personalized treatment of IBD and have therapeutic value for IBD patients. Therefore, this review summarized the effects of lnc RNA on the intestinal epithelial barrier, inflammatory response and immune homeostasis in IBD, and summarized the potential of lnc RNA as a biomarker of IBD and as a predictor of therapeutic response to IBD in the future.

Catechin-Induced changes in PODXL, DNMTs, and miRNA expression in Nalm6 cells: an integrated in silico and in vitro approach

Article

Full-text available

Jun 2024

Background This study explored the impact of predicted miRNAs on DNA methyltransferases (DNMTs) and the PODXL gene in Nalm6 cells, revealing the significance of these miRNAs in acute lymphocytic leukemia (ALL). Methods A comprehensive approach was adopted, integrating bioinformatic analyses encompassing protein structure prediction, molecular docking, dynamics, and ADMET profiling, in conjunction with evaluations of gene and miRNA expression patterns. This methodology was employed to elucidate the therapeutic potential of catechin compounds in modulating the activity of DNA methyltransferases (DNMTs) and the PODXL gene. Results The findings from our investigation indicate that catechins possess the capability to inhibit DNMT enzymes. This inhibitory effect is associated with the upregulation of microRNAs miR-200c and miR-548 and a concurrent downregulation of PODXL gene expression. These molecular interactions culminate in an augmented apoptotic response within ALL (Nalm6) cells. Conclusion The study posits that catechins may represent a viable therapeutic avenue for inducing apoptosis in ALL cells. This is achieved through the modulation of epigenetic mechanisms and alterations in gene expression profiles, highlighting the potential of catechins as agents for cancer therapy.

The miRNA Contribution in Adipocyte Maturation

Article

Full-text available

Jun 2024

Mesenchymal stem cells, due to their multipotent ability, are considered one of the best candidates to be used in regenerative medicine. To date, the most used source is represented by the bone marrow, despite the limited number of cells and the painful/invasive procedure for collection. Therefore, the scientific community has investigated many alternative sources for the collection of mesenchymal stem cells, with the adipose tissue representing the best option, given the abundance of mesenchymal stem cells and the easy access. Although adipose mesenchymal stem cells have recently been investigated for their multipotency, the molecular mechanisms underlying their adi-pogenic potential are still unclear. In this scenario, this communication is aimed at defining the role of miRNAs in adipogenic potential of adipose-derived mesenchymal stem cells via real-time PCR. Even if preliminary, our data show that cell culture conditions affect the expression of specific miRNA involved in the adipogenic potential of mesenchymal stem cells. The in vitro/in vivo validation of these results could pave the way for novel therapeutic strategies in the field of regenerative medicine. In conclusion, our research highlights how specific cell culture conditions can modulate the adipogenic potential of adipose mesenchymal stem cells through the regulation of specific miR-NAs.

A clathrin mediated endocytosis scaffolding protein, Intersectin 1, changes in an isoform, brain region, and sex specific manner in Alzheimer’s disease

Article

Full-text available

Jun 2024

Alzheimer’s disease (AD) is the most common form of dementia and is characterized by the accumulation of amyloid-beta (Aβ) plaques and neurofibrillary Tau tangles in the brain. We previously identified a set of candidate AD microRNAs (miRNAs) in human cerebrospinal fluid (CSF) and used a target prediction pipeline to identify mRNAs and pathways that could potentially be regulated by the miRNAs. Of these pathways, clathrin mediated endocytosis (CME) was selected for further investigation. CME is altered in multiple brain cell types in AD and is implicated in early cellular phenotypes such as enlarged early endosomes and pathogenic processing of Aβ. However, a comprehensive evaluation of major CME hub proteins in humans with AD across multiple brain regions is lacking. Thus, we used immunoblots to evaluate human post-mortem AD and control (CTL) frontal cortex (FC; AD n = 22, CTL n = 23) and hippocampus (HP; AD n = 34, CTL n = 22) for changes in Intersectin 1 (ITSN1), Phosphatidylinositol Binding Clathrin Assembly Protein gene (PICALM), Clathrin Light Chain (CLT), FCH and Mu Domain Containing Endocytic Adaptor 1 (FCHO1), Adaptor Related Protein Complex 2 (AP2) Subunit Alpha 1 (AP2A1), and Dynamin 2 (DNM2). Of these, we found that in AD, ITSN1-long (ITSN1-L) was decreased in the FC of males and HP of females, while ITSN1-short was increased in the HP of both males and females. We further evaluated ITSN1-L levels in cortex (CTX) and HP of the 5xFAD mouse model of Aβ pathology at different timepoints during aging and disease progression by immunoblot ( n = 5–8 per group). At 3 months, female 5xFAD exhibited an increase of ITSN1-L in CTX but a decrease at 6 and 9 months. Additionally, immunofluorescent staining of 5xFAD primary HP neurons showed an increase of ITSN1-L in matured 5xFAD neurons at 21 and 28 days in vitro . Together, our studies show that in AD, isoforms of ITSN1 change in a brain region-and sex-dependent manner. Further, changes in ITSN1-L are transient with levels increasing during early Aβ accumulation and decreasing during later progression. These findings suggest that ITSN1 expression, and consequently CME activity, may change depending on the stage of disease progression.

Elevation of endogenous RBM4 prevents Ang II-induced cardiomyocyte hypertrophy via down-regulating the expression of PTBP1

Article

Jun 2024

Evaluation of miRNA-146a, miRNA-34a, and Pro-inflammatory Cytokines as a Potential Early Indicators for Type 1 Diabetes Mellitus.

Article

Jun 2024

Exploring the role of pericardial miRNAs and exosomes in modulating cardiac fibrosis

Article

Jun 2024
CARDIOVASC PATHOL

Epigenetic regulation of androgen dependent and independent prostate cancer

Chapter

Jan 2024

Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells

Article

Full-text available

Jan 2007

Exosomes are vesicles of endocytic origin released by many cells. These vesicles can mediate communication between cells, facilitating processes such as antigen presentation. Here, we show that exosomes from a mouse and a human mast cell line (MC/9 and HMC-1, respectively), as well as primary bone marrow-derived mouse mast cells, contain RNA. Microarray assessments revealed the presence of mRNA from approximately 1300 genes, many of which are not present in the cytoplasm of the donor cell. In vitro translation proved that the exosome mRNAs were functional. Quality control RNA analysis of total RNA derived from exosomes also revealed presence of small RNAs, including microRNAs. The RNA from mast cell exosomes is transferable to other mouse and human mast cells. After transfer of mouse exosomal RNA to human mast cells, new mouse proteins were found in the recipient cells, indicating that transferred exosomal mRNA can be translated after entering another cell. In summary, we show that exosomes contain both mRNA and microRNA, which can be delivered to another cell, and can be functional in this new location. We propose that this RNA is called " exosomal shuttle RNA " (esRNA). Exosomes are small (50–90 nm) membrane vesicles of endocytic origin that are released into the extracellular environment on fusion of multivesicular bodies (MVB) with the plasma membrane 1. Many cells have the capacity to release exosomes, including reticulo-cytes 2 , dendritic cells 3 , B cells 4 , T cells 5 , mast cells 6 , epithelial cells 7 and tumour cells 8. The functions of exosomes are not completely understood, although it has been shown that exosomes can participate in the signalling events contributing to antigen presentation to T cells 4 and the development of tolerance 9. Several mechanisms have been hypothesized describing the interactions of exosomes and recipient cells. Exosomes can bind to cells through recep-tor–ligand interactions, similar to cell–cell communication mediating , for example, antigen presentation 4. Alternatively, exosomes could putatively attach or fuse with the target-cell membrane, delivering exosomal surface proteins and perhaps cytoplasm to the recipient cell 10,11. Finally, exosomes may also be internalized by the recipient cells by mechanisms such as endocytosis 12. Exosomes were isolated from a mast-cell line (MC/9), primary bone marrow-derived mast cells (BMMC) and a human mast-cell line (HMC-1) through a series of microfiltration and ultracentrifugation steps modified from what has been previously described 4. To confirm that the structures studied indeed are exosomes, they were examined by electron microscopy (Fig. 1a), flow cytometric analysis (FACS; Fig. 1b), and proteomic analysis (see Supplementary Information, Table S1). The electron micrographs of the exosomes revealed rounded structures with a size of approximately 50–80 nm, similar to previously described exo-somes 4,13–15. The identity of the studied vesicles was further confirmed as exosomes by FACS analysis (Fig. 1b), which show the presence of the surface protein CD63 — a commonly used marker of exosomes. Finally, extensive protein analysis of the MC/9 exosomes was performed on multiple samples using LC-MS/MS technology. A total of 271 proteins were identified (see Supplementary Information, Table S1) from three preparations of the isolated vesicles, of which 47 proteins were present in all three samples. More importantly, a large number of the proteins found in the preparations were the same as proteins previously identified in exosomes produced by other cells (that is, exosomes derived from intestinal epithelial cells, urine, dendritic cells, microglia, melanoma, T-cells and B-cells). In particular, 60% of the 47 proteins found in all samples of mast-cell exosomes have been previously found in other types of exosomes. Moreover, 39% of the 271 total proteins found in the analysed exosome samples have also been previously found in other types of exosomes. Thus, the electron microscopy, the FACS, and the detailed protein analyses each provided significant evidence in favour of the identification of the isolated vesicles as exosomes. The presence of nucleic acids was examined in exosomes derived from MC/9, BMMC and HMC-1 cells to define a potential mechanism by which exosomes may mediate cell–cell communication. These assessments showed that isolated exosomes contain no DNA (see Supplementary Information, Fig. S1). However, substantial amounts of RNA were detected by agarose gel electrophoresis, spectrophotometry

Erratum: Inference of miRNA targets using evolutionary conservation and pathway analysis (BMC Bioinformatics)

Article

Full-text available

Jul 2007
BMC BIOINFORMATICS

Background: MicroRNAs have emerged as important regulatory genes in a variety of cellular processes and, in recent years, hundreds of such genes have been discovered in animals. In contrast, functional annotations are available only for a very small fraction of these miRNAs, and even in these cases only partially. Results: We developed a general Bayesian method for the inference of miRNA target sites, in which, for each miRNA, we explicitly model the evolution of orthologous target sites in a set of related species. Using this method we predict target sites for all known miRNAs in flies, worms, fish, and mammals. By comparing our predictions in fly with a reference set of experimentally tested miRNA-mRNA interactions we show that our general method performs at least as well as the most accurate methods available to date, including ones specifically tailored for target prediction in fly. An important novel feature of our model is that it explicitly infers the phylogenetic distribution of functional target sites, independently for each miRNA. This allows us to infer species-specific and clade-specific miRNA targeting. We also show that, in long human 3' UTRs, miRNA target sites occur preferentially near the start and near the end of the 3' UTR. To characterize miRNA function beyond the predicted lists of targets we further present a method to infer significant associations between the sets of targets predicted for individual miRNAs and specific biochemical pathways, in particular those of the KEGG pathway database. We show that this approach retrieves several known functional miRNA-mRNA associations, and predicts novel functions for known miRNAs in cell growth and in development. Conclusion: We have presented a Bayesian target prediction algorithm without any tunable parameters, that can be applied to sequences from any clade of species. The algorithm automatically infers the phylogenetic distribution of functional sites for each miRNA, and assigns a posterior probability to each putative target site. The results presented here indicate that our general method achieves very good performance in predicting miRNA target sites, providing at the same time insights into the evolution of target sites for individual miRNAs. Moreover, by combining our predictions with pathway analysis, we propose functions of specific miRNAs in nervous system development, inter-cellular communication and cell growth. The complete target site predictions as well as the miRNA/pathway associations are accessible on the ElMMo web server.

The beta 4 Integrin Interactor p27BBP/eIF6 Is an Essential Nuclear Matrix Protein Involved in 60S Ribosomal Subunit Assembly

Article

Full-text available

Mar 1999
J CELL BIOL

p27BBP/eIF6 is an evolutionarily conserved protein that was originally identified as p27BBP, an interactor of the cytoplasmic domain of integrin β4 and, independently, as the putative translation initiation factor eIF6. To establish the in vivo function of p27BBP/eIF6, its topographical distribution was investigated in mammalian cells and the effects of disrupting the corresponding gene was studied in the budding yeast, Saccharomyces cerevisiae. In epithelial cells containing β4 integrin, p27BBP/eIF6 is present in the cytoplasm and enriched at hemidesmosomes with a pattern similar to that of β4 integrin. Surprisingly, in the absence and in the presence of the β4 integrin subunit, p27BBP/eIF6 is in the nucleolus and associated with the nuclear matrix. Deletion of the IIH S. cerevisiae gene, encoding the yeast p27BBP/eIF6 homologue, is lethal, and depletion of the corresponding gene product is associated with a dramatic decrease of the level of free ribosomal 60S subunit. Furthermore, human p27BBP/eIF6 can rescue the lethal effect of the iihΔ yeast mutation. The data obtained in vivo suggest an evolutionarily conserved function of p27BBP/eIF6 in ribosome biogenesis or assembly rather than in translation. A further function related to the β4 integrin subunit may have evolved specifically in higher eukaryotic cells.

Principles of MicroRNA–Target Recognition

Article

Full-text available

Nov 2004
PLOS BIOL

MicroRNAs (miRNAs) are short non-coding RNAs that regulate gene expression in plants and animals. Although their biological importance has become clear, how they recognize and regulate target genes remains less well understood. Here, we systematically evaluate the minimal requirements for functional miRNA-target duplexes in vivo and distinguish classes of target sites with different functional properties. Target sites can be grouped into two broad categories. 5' dominant sites have sufficient complementarity to the miRNA 5' end to function with little or no support from pairing to the miRNA 3' end. Indeed, sites with 3' pairing below the random noise level are functional given a strong 5' end. In contrast, 3' compensatory sites have insufficient 5' pairing and require strong 3' pairing for function. We present examples and genome-wide statistical support to show that both classes of sites are used in biologically relevant genes. We provide evidence that an average miRNA has approximately 100 target sites, indicating that miRNAs regulate a large fraction of protein-coding genes and that miRNA 3' ends are key determinants of target specificity within miRNA families.

Oncomirs: MicroRNAs with a role in cancer

Article

Full-text available

Apr 2006
CANCER-AM CANCER SOC

MicroRNAs (miRNAs) are an abundant class of small non-protein-coding RNAs that function as negative gene regulators. They regulate diverse biological processes, and bioinformatic data indicates that each miRNA can control hundreds of gene targets, underscoring the potential influence of miRNAs on almost every genetic pathway. Recent evidence has shown that miRNA mutations or mis-expression correlate with various human cancers and indicates that miRNAs can function as tumour suppressors and oncogenes. miRNAs have been shown to repress the expression of important cancer-related genes and might prove useful in the diagnosis and treatment of cancer.

P-Body Formation Is a Consequence, Not the Cause, of RNA-Mediated Gene Silencing

Article

Full-text available

Jun 2007

P bodies are cytoplasmic domains that contain proteins involved in diverse posttranscriptional processes, such as mRNA degradation, nonsense-mediated mRNA decay (NMD), translational repression, and RNA-mediated gene silencing. The localization of these proteins and their targets in P bodies raises the question of whether their spatial concentration in discrete cytoplasmic domains is required for posttranscriptional gene regulation. We show that processes such as mRNA decay, NMD, and RNA-mediated gene silencing are functional in cells lacking detectable microscopic P bodies. Although P bodies are not required for silencing, blocking small interfering RNA or microRNA silencing pathways at any step prevents P-body formation, indicating that P bodies arise as a consequence of silencing. Consistently, we show that releasing mRNAs from polysomes is insufficient to trigger P-body assembly: polysome-free mRNAs must enter silencing and/or decapping pathways to nucleate P bodies. Thus, even though P-body components play crucial roles in mRNA silencing and decay, aggregation into P bodies is not required for function but is instead a consequence of their activity.

Synaptic Protein Synthesis Associated with Memory Is Regulated by the RISC Pathway in Drosophila

Article

Aug 2006
CELL

The Saccharomyces cerevisiae Homologue of Mammalian Translation Initiation Factor 6 Does Not Function as a Translation Initiation Factor

Article

Feb 1999

Eukaryotic translation initiation factor 6 (eIF6) binds to the 60S ribosomal subunit and prevents its association with the 40S ribosomal subunit. The Saccharomyces cerevisiae gene that encodes the 245-amino-acid eIF6 (calculated M r 25,550), designated TIF6 , has been cloned and expressed in Escherichia coli . The purified recombinant protein prevents association between 40S and 60S ribosomal subunits to form 80S ribosomes. TIF6 is a single-copy gene that maps on chromosome XVI and is essential for cell growth. eIF6 expressed in yeast cells associates with free 60S ribosomal subunits but not with 80S monosomes or polysomal ribosomes, indicating that it is not a ribosomal protein. Depletion of eIF6 from yeast cells resulted in a decrease in the rate of protein synthesis, accumulation of half-mer polyribosomes, reduced levels of 60S ribosomal subunits resulting in the stoichiometric imbalance in the 40S/60S subunit ratio, and ultimately cessation of cell growth. Furthermore, lysates of yeast cells depleted of eIF6 remained active in translation of mRNAs in vitro. These results indicate that eIF6 does not act as a true translation initiation factor. Rather, the protein may be involved in the biogenesis and/or stability of 60S ribosomal subunits.

Developmental Biology Takes on a Latin American Rhythm

Article

Aug 2005
CELL

Latin America, the vast geopolitical region from Mexico to Cape Horn at the tip of South America, evokes images of the rain forests of the Amazon and snow-capped peaks of the Andes, ancient Incan and Mayan civilizations, Spanish and Portuguese colonialism, and iconic cities like Rio de Janeiro. Biological research in Latin America has traditionally focused on disciplines such as physiology, pharmacology, biochemistry, genetics, immunology, and parasitology as well as on agricultural and ecological sciences. However, an energetic group of developmental biologists has recently emerged from Argentina, Brazil, Chile, Ecuador, Mexico, Peru, Uruguay, and Venezuela. In 2003, this group, composed of senior scientists and young Latin American developmental biologists who recently returned to their home countries after training in America and Europe, founded the Latin American Society of Developmental Biology (LASDB). Led by their current president Roberto Mayor (University of Chile, Santiago/University College London), the LASDB (http://www.lasdb.org) is promoting developmental biology research in Latin America.

Bruno Acts as a Dual Repressor of oskar Translation, Promoting mRNA Oligomerization and Formation of Silencing Particles

Article

Feb 2006
CELL

Prior to reaching the posterior pole of the Drosophila oocyte, oskar mRNA is translationally silenced by Bruno binding to BREs in the 3' untranslated region. The eIF4E binding protein Cup interacts with Bruno and inhibits oskar translation. Validating current models, we directly demonstrate the mechanism proposed for Cup-mediated repression: inhibition of small ribosomal subunit recruitment to oskar mRNA. However, 43S complex recruitment remains inhibited in the absence of functional Cup, uncovering a second Bruno-dependent silencing mechanism. This mechanism involves mRNA oligomerization and formation of large (50S-80S) silencing particles that cannot be accessed by ribosomes. Bruno-dependent mRNA oligomerization into silencing particles emerges as a mode of translational control that may be particularly suited to coupling with mRNA transport.

Mechanisms of Post-Transcriptional Regulation by microRNAs: Are the Answers in Sight?

Abstract and Figures

Recommended publications

Plasma MicroRNA as a novel diagnostic

RNA Interference

Low p21 level is necessary for the suppressive effects of micoRNA-31 on glioma cell migration and in...

A Least Angle Regression Model for the Prediction of Canonical and Non-Canonical miRNA-mRNA Interact...