ARAUD Tanguy Hors Thèse
33 Pages
Downloading requires you to have access to the YouScribe library
Learn all about the services we offer

ARAUD Tanguy Hors Thèse

Downloading requires you to have access to the YouScribe library
Learn all about the services we offer
33 Pages


Département de Microbiologie et Médecine MoléculaireCentre Médicale UniversitaireOverview of the miRNA pathwaysby Tanguy ARAUDHors Thèse ManuscriptCo-directeurs de thèse: Dr J. CURRAN (CMU) Pr J.D. ROCHAIX (Science III)Examinateur: Pr D. BELIN (CMU) Pr W. REITH (CMU)Novembre 2008Overview of the miRNA pathwaysIndexOnce upon a time… 3What’s a microRNA? 4Universal nomenclature 5Genetic organisation and transcription 7miRNA Biogenesis 8Nuclear processing: Cleavage by the Drosha-DGCR8 Complex 8Nuclear export 10Cytoplasmic procy DICER 10Incorporation into RISC and activation 11Biological activity 12Translational inhibition at the elongation step. 13Translational inhibition at the initiation step 15Indirect degradation 15Translational activation 16Regulation of microRNA activity 16Disease 17miRNAs and cancer 17Viral infection and host defence 18Heart disease 20Conclusion 20Acknowledgements 21Bibliographie 21Hors-Thèse Tanguy ARAUD Page 2Overview of the miRNA pathwaysOnce upon a time…The micro RNA (miRNA) story started in the mid 1970’s with the isolation of a mutation (e912) in Caenorhabditis elegans (C. elegans) (Hodgkin, 1974). This mutation affected the lin-4 gene. It blocked ventral hypodermal cell division during development and generated a “Vulvaless” phenotype (Horvitz and Sulston, 1980) (figure1). It appeared that the e912 mutation was causing a failure of temporal development, indicating that lin-4 might encode a ...



Published by
Reads 17
Language English


Département de Microbiologie et Médecine Moléculaire
Centre Médicale Universitaire
Overview of the miRNA
by Tanguy ARAUD
Hors Thèse Manuscript
Co-directeurs de thèse: Dr J. CURRAN (CMU)
Pr J.D. ROCHAIX (Science III)
Examinateur: Pr D. BELIN (CMU)
Novembre 2008Overview of the miRNA pathways
Once upon a time… 3
What’s a microRNA? 4
Universal nomenclature 5
Genetic organisation and transcription 7
miRNA Biogenesis 8
Nuclear processing: Cleavage by the Drosha-DGCR8 Complex 8
Nuclear export 10
Cytoplasmic procy DICER 10
Incorporation into RISC and activation 11
Biological activity 12
Translational inhibition at the elongation step. 13
Translational inhibition at the initiation step 15
Indirect degradation 15
Translational activation 16
Regulation of microRNA activity 16
Disease 17
miRNAs and cancer 17
Viral infection and host defence 18
Heart disease 20
Conclusion 20
Acknowledgements 21
Bibliographie 21
Hors-Thèse Tanguy ARAUD Page 2Overview of the miRNA pathways
Once upon a time…
The micro RNA (miRNA) story started in the mid 1970’s with the isolation of a
mutation (e912) in Caenorhabditis elegans (C. elegans) (Hodgkin, 1974). This mutation
affected the lin-4 gene. It blocked ventral hypodermal cell division during development
and generated a “Vulvaless” phenotype (Horvitz and Sulston, 1980) (figure1). It appeared
that the e912 mutation was causing a failure of temporal development, indicating that
lin-4 might encode a master regulator of developmental timing (Chalfie et al., 1981).
Figure 1: Caenorhabditis elegans Vulvaless phenotype.
Those images are extracted from the online review of C.elegans biology WormBooK (http://www. A- WT phenotype. The vulva is present in the adult female worm. It is necessary for egg-
laying and for copulation with males. B- Vulvaless phenotype. Mutation e912 induces the loss of the lin-4
gene, and loss of the vulva in the female worm.
Later, during amplification of these mutated nematodes the Horvitz group isolated
revertants showing normal development. lin-14 null alleles were identified in these
animals. Surprisingly, the developmental timing defects of lin-14 were opposite to those
of lin-4 (Ambros and Horvitz, 1984), suggesting that lin-4 might encode a trans-acting
negative regulator of lin-14. Fortunately, two new nematode mutants, showing the same
phenotype as e912 were isolated (n355 and n536). These gain-of-function mutations
affected the lin-14 gene (Ruvkun et al., 1989). Deletions in the 3′ untranslated region
(UTR) of the lin-14 mRNA were characterized in both n355 and n536. This was correlated
with an inappropriately high level of LIN-14 expression during development (Wightman
et al., 1991). Despite these observations the unusual characteristics of miRNA delayed
there discovery. A size <700bp and no well characterized ORF (Open Reading Frame) were
not compatible with the definition of a gene at that time. Furthermore, two transcripts
of about 60 bp and 20 bp were detected using the RNase Protection Assay (RPA), with the
shorter form being the most abundant. It was only with the appearance of bioinformatic
software, that the antisense theory was elaborated. The complementary sequence of the
shorter lin-4 transcript was found to be repeated seven times within the 3’UTR of lin-14.
Hors-Thèse Tanguy ARAUD Page 3Overview of the miRNA pathways
The formation of this RNA:RNA duplex with a temporal gradient, down regulated the
expression of lin-14 during development (Lee et al., 1993; Wightman et al., 1993).
However, the discovery of the world’s first miRNA did not trigger a gold rush until
the discovery of a second tiny RNA, let-7, in C. elegans. This 21 nt RNA was complementary
to regions within the 3’UTR of lin-14, lin-28, lin-41, lin-42 and daf-12, indicating that
expression of these genes might be directly controlled by let-7 (Reinhart et al., 2000).
Nonetheless, the term “microRNA” was only introduced in 2001 in a set of three articles
(Lau et al., 2001; Lee and Ambros, 2001; Ruvkun, 2001).
The generation of full genome databases for different species, demonstrated that
let-7 was conserved across most of the animal phylogeny, with a temporal regulation also
conserved in a wide range of species (Pasquinelli et al., 2000). Computational methods
predicted more than 6000 potential miRNA genes covering 58 species (Artzi et al., 2008;
Lim et al., 2003a). Finally, the number of repository miRNA databases and prediction
tools also increased (miBase (Griffiths-Jones et al., 2008), miRNApath (Chiromatzo et al.,
2007), miRGen (Megraw et al., 2007), miRNAMap (Hsu et al., 2006)), providing resources
for further studies (figure 2).
600 Figure 2: Increasing numbers of
publication in the miRNA field.500
Since the 70’s the number of
400 publications has increased with the
exponential growth in the last five 300
years, indicating that this field is
200 very active. Data used to create the
charts was extracted from the web 100
site “ISI web of Knowledge” (http://
Year of publication
What’s a microRNA?
MiRNAs are a class of non-coding RNA genes. They belong to the RNA interference
familly (RNAi family) along with small interfering RNA (siRNA) and PIWI interacting RNA
(piRNA). There final products are small single strand RNA molecules, about 19-21
nucleotides (nt) long with 3’ hydroxyl and 5’ phosphate ends. piRNAs are a little bit longer,
about 24-27 nt. It is largely accepted that these highly conserved RNAs generally regulate
the expression of genes. They acts as guide molecules by base-pairing with the 3’-UTR (or
5’UTR) of specific mRNAs leading to translational repression and/or mRNA cleavage (see
Hors-Thèse Tanguy ARAUD Page 4
Number of publicationsOverview of the miRNA pathways
Ambros and coworkers defined a small RNA molecule as a miRNA using five
criteria (Ambros et al., 2003; Griffiths-Jones et al., 2006):
• Detection of a distinct 22-nt RNA transcript.
• Identification of the 22-nt sequence ina library of cDNAs madefrom size-fractionated
RNA. Such sequences must precisely match the genomic sequence of the organism
from which they were cloned.
• Prediction of a hairpin precursor (usually about 60–80 nt) containing the 22-nt
miRNA sequence within one arm.
• Phylogenetic conservation of the 22-nt miRNA sequence and its predicted precursor
secondary structure.
• Detection of increased precursor accumulation in organisms with reduced Dicer
function (see later).
Those five criteria are necessary, but nowadays it seems that they are not sufficient
to unambiguously define a small RNA as a miRNA. Indeed siRNA are very similar to miRNA
and the distinction between them is increasingly becomingobscure. The most important
requirement for a miRNA is a contiguous and perfect base pairing with its target of 6
nucleotides (nucleotide 2 to 8 from the 5’ extremity). This is called the “seed” region
(figure 3). It nucleates the interaction of the miRNA with its mRNA target. Another
important feature is the presence of a bulge in the middle of the sequence. Both miRNA
and siRNA are produced via similar pathways and have similar mechanisms of action
(Ambros, 2004; Bartel, 2004). That’s why most of the discoveries about siRNA processing
were simply extended to miRNAs and vice versa. Recent observations have shown that
both can suppress translation of mRNAs (in the case of an imperfect match, miRNAs) or
can cleave target RNAs (in the case of a perfect match, siRNAs), and that they can be
interchangeable (Cullen, 2004; Yekta et al., 2004; Zeng et al., 2003). siRNAs were originally
proposed to act mainly as an antiviral defence and transposon repression system via the
phenomenon of RNAi (Hunter, 2000), but recent findings indicate that such RNAs may
play a much broader role in gene and genome regulation. Finally, we can ask if the same
small RNA can act as a miRNA on one mRNA target and as a siRNA on another?
Universal nomenclature
Because the number of miRNAs discovered is growing rapidly, 2909 in 36 genomes
in June 2005 to 5071 in 58 genomes in August 2007 (Griffiths-Jones et al., 2008), an
universal nomenclature became necessary.
The primary transcript, which is the result of transcription (see below), is named
pri-miR, while the product of Drosha is called pre-mir (figure 5). Finally, miRNAs are
named using the “mir” abbreviation surrounded by a three or four letter species prefix
Hors-Thèse Tanguy ARAUD Page 5Overview of the miRNA pathways
primary transcript of lin4A
U5’ U U C A U C A
UU C C C A G3’ C A
Guide strand lin-4 duplexC
5’ U U C A
3’ C A
passenger strand
exemple of a site of lin-4:lin-14 3’UTR interactionD
Figure 3: Schematic representation of lin-4 – lin-14 interaction.
The sequences were obtained from the miRBase website ( and folded with
the online program RNAfold ( The targeted site in the 3’UTR of lin-14 was
predicted by the online program TargetScanWorm (, and verified experimentally
by Lee et al., 1993. A: pri-cel-lin4, B: pre-cel-lin4, C: guide and passenger strand duplex of lin-4, D: Example
of lin4-lin14 interaction. There are 8 characterized sites of lin-4 interaction in the 3’UTR of lin-14 and the
one presented is the most characteristic. The green arrows indicate the site of Drosha cleavage, and the
blue arrows indicate the site of Dicer cleavage (see later).
and a unique identifying number suffix (e.g. hsa-mir-212). The identifying numbers are
assigned sequentially, with identical miRNAs having the same number, regardless of
organism. miRNAs with only 1 or 2 bp of differences are assigned suffixes (e.g. mmu-mir-
181a and mmu-mir-181b). If a miRNA is expressed from more than one hairpin precursor
locus, it is denoted with further numeric suffixe(e.g. dme-mir-6-1 and dme-mir-6-2). The
genes that encode the miRNA are also named using the same nomenclature, with
capitalization, hyphenation, and italics according to the conventions of the organism.
(e.g. hsa-mir-212, mmu-mir-181a, dme-mir-6-1), (Ambros et al., 2003; Griffiths-Jones et
al., 2006).
Hors-Thèse Tanguy ARAUD Page 6Overview of the miRNA pathways
Genetic organisation and transcription
The first annotation of miRNAs indicated that most were located in intergenic
regions (outside annotated genes). 50% of those known miRNAs are found in close
proximity to other miRNAs (Lagos-Quintana et al., 2001; Lau et al., 2001) which raised
the possibility that these clustered miRNAs might be transcribed from a single polycistronic
transcriptional unit (Figure 4) (Lee et al., 2002; Mourelatos et al., 2002). It was initially
thought that miRNAs are produced by RNA polymerase III, because it transcribes most of
the small RNAs, such as tRNAs and U6 snRNA.
Use of the Cap-Switching RACE protocol (which selects capped RNAs) suggested
that the miR-172 precursor, EAT, in Arabidopsis contains a cap structure (Aukerman and
Sakai, 2003). Moreover, analysis of the noncoding RNAs containing miRNA sequences
showed that they are polyadenylated and spliced (Aukerman and Sakai, 2003; Tam, 2001).
Transcription of some miRNAs is sensitive to -amanitin at a concentration specific with
the inhibition of RNA pol II, and chromatin immunoprecipitation analyses demonstrated
that RNA pol II is physically associated with different pri-miRNA promoters (Lee et al.,
2004). Full analysis of the human pri-miR-21 indicated that it is not only structurally
similar to mRNAs but can, in fact, function both as a pri-miRNA and a mRNA (Cai et al.,
2004). Nonetheless, the possibility that a small number of miRNA genes could be
transcribed by other RNA polymerases cannot be excluded. Ectopic expression of miR-
142 from a pol III promoter produces a miRNA that functions in vitro, indicating that
there is no obligate link between polymerase type, future miRNA processing and function
(Chen et al., 2004).
However, more careful analysis has revealed that ~ 70% of mammalian miRNA
genes (161 out of 232) are located in defined transcription units (Rodriguez et al., 2004).
The majority of the human miRNA loci are in fact located within the intronic regions of
either coding or noncoding transcription units (figure 4). So, miRNA genes can be grouped
on the basis of their genomic locations: exonic miRNA in non-coding transcription units,
intronic miRNA in non-coding transcription units and intronic miRNA in protein-coding
transcription units. The expression of the last group of miRNAs largely coincides with the
transcription of the hosting transcription unit, indicating that the intronic miRNAs and
their hosting genes may be coregulated and generated from a common precursor
transcript (Baskerville and Bartel, 2005).
Introns have been considered as evolutionary debris resulting from splicing
excision. Added to this, the increasing number of non-protein-coding transcripts being
detected in mammalian cells has been suggested to be “transcriptional noise” (Dennis,
Hors-Thèse Tanguy ARAUD Page 7Overview of the miRNA pathways
Pol II transcription
A= exonic miRNA in non-coding transcription units B= intronic miRNA in non-coding transcription units C= intronic miRNA in protein-coding transcription units
intronic pri-miRNA intronic pri-miRNA
Figure 4: Genetic organization and possible precursor of the primary miRNA.
Most, if not all, primary transcripts of miRNAs are transcribed by RNA pol II, capped and polyadenylated. A
= exonic miRNA in non-coding transcription units; B = intronic miRNA in non-coding transcription units; C
= intronic miRNA in protein-coding transcription units. Open Reading Frame is represented by green box.
The cleavage of intronic miRNAs by the Microprocessor complex is independent of the Spliceosome.
2002). For example, some excised introns have half-lives comparable with mRNA and are
even exported from the nucleus to the cytoplasm (Clement et al., 2001; Clement et al.,
This raises an interesting question. How do these intronic or non-protein-coding
transcript miRNAs get processed?
miRNA Biogenesis
Nuclear processing: Cleavage by the Drosha-DGCR8 Complex
Like intronic small nucleolar RNAs (snoRNAs) (Filipowicz and Pogacic, 2002), the
pri-miRNA is cleaved by the nuclear RNase III type enzyme Drosha in a process known as
“cropping” (Lee et al., 2003). Drosha functions in a complex with DGCR8 (also called
Pasha in flies and nematodes) (Denli et al., 2004; Gregory et al., 2004; Han et al., 2004).
This complex is called “Microprocessor”. It must recognize the junction of single-stranded
and double-stranded RNA and count ~ 11 bp, one helical RNA turn, to cut the
phosphodiester bond.
DGCR8 is the product of a gene deleted in DiGeorge syndrome (Landthaler et al.,
2004). Unlike Drosha, it can be crosslinked to pri-miRNA, suggesting that it acts as a
Hors-Thèse Tanguy ARAUD Page 8P68
Overview of the miRNA pathways
see FEBS letters %/) (2005) 5822-5829
dicing and slicing the core machinery of the RNA interfer-
ence pathway
miRNA duplex
TRBPP68passanger strand
translationnal repression
Figure 5: miRNA processing.
The pri-miRNA is cleaved in the nucleus with the help of DGRCR8 by Drosha. The resulting product, called
pre-miRNA, is exported to the cytoplasm in an RAN-GTP dependent manner by Exportin 5. Then pre-miRNA
is cleaved by Dicer, and a double strand miRNA duplex of 22nt is obtained. The passenger strand is removed
by TRBP and P68, and the guide strand is incorporated into the miRISC with an Ago protein to induce the
silencing of specific mRNA.
Hors-Thèse Tanguy ARAUD Page 9
TRBPOverview of the miRNA pathways
molecular ruler (Han et al., 2006). Drosha belongs to the class II of the RNase III family
characterized by a tandem of RNase III domains (RIIID) and a double stranded RNA binding
domain (dsRBD). The C-terminal RIIID, proximal to the dsRBD, cleaves the 5’-strand of the
hairpin, whereas the other RIIID (RIIIDa) cleaves the 3’-strand (Han et al., 2004). The
processing reaction releases a stem-loop of ~ 70nt with a 2-nt overhang at its 3’ end. This
intermediate product is called pre-miRNA (figure 5).
Intronic miRNAs can be processed from spliced or unspliced intronic regions. The
cleavage of an unspliced intron by Drosha does not significantly affect the production of
mature mRNA. The separation of the two machines, Spliceosome and Microprocessor,
suggests that a continuous intron may not be required for splicing (Kim and Kim, 2007).
Total, nuclear and cytoplasmic RNA fractions, were prepared from HeLa cells and
the subcellular localization of endogenous pri-miRNAs was then analysed by generating
cDNA using oligo-dT primed reverse transcription followed by PCR using specific primers.
This demonstrated that relatively little full-length pri-miRNA reaches the cytoplasm (Cai
et al., 2004) indicating that Drosha cleavage occurs mainly in the nucleus (figure 5).
Nuclear export
All RNAs appeared to use the same saturatable export pathway. Exporin 5 (Exp5)
has been shown to bind RNA with a high degree of double stranded character (Gwizdek
et al., 2003; Zeng and Cullen, 2004). Depletion of Exp5 decreases the levels of mature
miRNAs by 40 to 60% (Lund et al., 2004). This export is carrier-mediated in a Ran GTP
manner (Fried and Kutay, 2003).
The spatial separation and sequential action of Drosha and Dicer, which are
respectively localized in the nucleus and the cytoplasm, appears to promote correct and
efficient processing of precursors in the generation ofmature miRNAs (Lee et al., 2003).
Exp5, which functions in the middle of this pathway, may facilitate miRNA biogenesis by
monitoring the integrityof pre-miRNAs and by promoting efficient release of pre-miRNAs
from Drosha in the nucleus where the level of RanGTP is high. Conversely, in the cytoplasm,
where the level of RanGTP is low, Exp5 would release pre-miRNAs to Dicer for further
processing (figure 5).
Cytoplasmic processing: Cleavage by DICER
Following export, pre-miRNA is transfered to the cytoplasmic RNase III Dicer. This
was simultaneously confirmed by different groups using different approaches.
Immunoprecipitated Dicer (or the Drosophila homologue) generates ~ 22-nucleotide
Hors-Thèse Tanguy ARAUD Page 10