A single Argonaute protein
mediates both transcriptional
and posttranscriptional silencing
in Schizosaccharomyces pombe
Alla Sigova, Nicholas Rhind,1and Phillip D. Zamore2
Department of Biochemistry and Molecular Pharmacology, University of Massachusetts Medical School,
Worcester, Massachusetts 01605, USA
The Schizosaccharomyces pombe genome encodes only one of each of the three major classes of proteins
implicated in RNA silencing: Dicer (Dcr1), RNA-dependent RNA polymerase (RdRP; Rdp1), and Argonaute
(Ago1). These three proteins are required for silencing at centromeres and for the initiation of transcriptionally
silent heterochromatin at the mating-type locus. Here, we show that the introduction of a double-stranded
RNA (dsRNA) hairpin corresponding to a green fluorescent protein (GFP) transgene triggers classical RNA
interference (RNAi) in S. pombe. That is, GFP silencing triggered by dsRNA reflects a change in the
steady-state concentration of GFP mRNA, but not in the rate of GFP transcription. RNAi in S. pombe requires
dcr1, rdp1, and ago1, but does not require chp1, tas3, or swi6, genes required for transcriptional silencing.
Thus, the RNAi machinery in S. pombe can direct both transcriptional and posttranscriptional silencing using
a single Dicer, RdRP, and Argonaute protein. Our findings suggest that these three proteins fulfill a common
biochemical function in distinct siRNA-directed silencing pathways.
[Keywords: RNAi; PTGS; TGS; siRNA; Argonaute; Dicer]
Supplemental material is available at http://www.genesdev.org.
Received May 4, 2004; revised version accepted August 6, 2004.
In many eukaryotic cells, exogenous long double-
stranded RNA (dsRNA) triggers the specific degradation
of cellular mRNAs with corresponding sequences, a phe-
nomenon termed RNA interference (RNAi) (Fire et al.
1998). The dsRNA is cleaved by the multidomain ribo-
nuclease III enzyme, Dicer, into a population of 21–27-nt
dsRNAs termed small interfering RNAs (siRNAs) (Bern-
stein et al. 2001). siRNAs are the specificity determi-
nants of the RNAi pathway (Hamilton and Baulcombe
1999; Hammond et al. 2000; Zamore et al. 2000). siRNAs
are assembled into a protein–RNA complex, the RNA-
induced silencing complex (RISC), which directs cleav-
age of the target RNA (Hammond et al. 2000; Zamore et
al. 2000; Nykänen et al. 2001). Among the various pro-
tein factors required for RNAi, members of the Argo-
naute family of proteins are universally associated with
siRNA-directed gene silencing (Cogoni and Macino
1997; Tabara et al. 1999, 2002; Cerutti et al. 2000; Fagard
et al. 2000; Catalanotto et al. 2002; Kennerdell et al.
2002; Mochizuki et al. 2002; Pal-Bhadra et al. 2002;
Tijsterman et al. 2002; Williams and Rubin 2002; Shi et
al. 2004). Plants and animals contain multiple Argonaute
paralogs; the Drosophila genome encodes at least five
distinct Argonaute proteins; humans have eight; and
worms, 27! The remarkable diversity of Argonaute pro-
teins suggests that each Argonaute paralog plays a spe-
cific role in RNA silencing. The Ago2 protein is a core-
component of biochemically purified RISC from both
Drosophila (Hammond et al. 2001) and mammals
(Hutvágner and Zamore 2002; Martinez et al. 2002). In
fact, among the human Argonaute proteins, only Ago2
can mediate microRNA (miRNA)- or siRNA-directed en-
donucleolytic cleavage of target RNA (Liu et al. 2004;
Meister et al. 2004). The recent three-dimensional struc-
ture of an archeal Argonaute protein, together with ex-
periments evaluating the importance of predicted cata-
lytic residues in human Ago2, suggests that human Ago2
itself is the small RNA-guided endoribonuclease that
cleaves target RNA (Liu et al. 2004; Song et al. 2004).
Other human Argonaute proteins may be specialized to
direct translational repression of mRNA or transcrip-
tional silencing of DNA sequences by the small RNA-
1E-MAIL email@example.com; FAX (508) 856-6231.
2E-MAIL firstname.lastname@example.org; FAX (508) 856-2003.
Article published online ahead of print. Article and publication date are
GENES & DEVELOPMENT 18:2359–2367 © 2004 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/04; www.genesdev.org 2359
directed production of silent heterochromatin. Such
functional specialization might also extend to the ani-
mal miRNA pathway, in which small RNAs typically
direct translational repression, but not destruction, of
their target mRNAs (Lagos-Quintana et al. 2001; Lau et
al. 2001; Lee and Ambros 2001). In Caenorhabditis el-
egans, the Argonaute protein Rde-1 is required for RNAi
but not for miRNA function, whereas the Argonaute pro-
teins Alg-1 and Alg-2 are required for miRNA function
but not RNAi (Tabara et al. 1999; Grishok et al. 2001). In
Drosophila, Ago2 function is restricted to the siRNA-
directed posttranscriptional gene silencing (PTGS) path-
way, whereas miRNA function is associated with Ago1
(Okamura et al. 2004). In Arabidopsis thaliana, AGO1 is
required for miRNA function and PTGS, whereas AGO4
acts in a silencing pathway that targets chromatin,
rather than mRNA (Fagard et al. 2000; Morel et al. 2002;
Zilberman et al. 2003; Vaucheret et al. 2004).
In contrast to the multiplicity of Argonaute proteins in
higher organisms, the sequenced fungal genomes appear
to encode one (Neurospora crassa and Schizosaccharo-
myces pombe) or no Argonaute proteins (Saccharomyces
cerevisiae). The Neurospora Argonaute protein QDE-2 is
required for quelling, an RNAi-like phenomenon (Co-
goni and Macino 1997) and is a component of an siRNA-
containing complex (Catalanotto et al. 2002). In contrast,
Ago1 is required for the silencing of transcription at cen-
tromeres in the fission yeast S. pombe (Volpe et al. 2002)
and initiation of silent heterochromatin at the mating
type locus (Hall et al. 2002). S. pombe Ago1 is a compo-
nent of the RNA-induced initiation of transcriptional si-
lencing (RITS) complex (Verdel et al. 2004). In addition to
Ago1, the RITS contains siRNAs derived from centro-
meric sequences, as well as the chromodomain protein
Chp1 and the protein Tas3, whose function is unknown.
Current evidence suggests that the RITS uses the se-
quence information in its siRNA component to direct
the methylation of Lys 9 (K9) of histone H3 bound to
centromeric DNA (Verdel et al. 2004). Histone H3 K9
methylation, in turn, triggers the formation of transcrip-
tionally silent heterochromatin, a process dependent on
the histone methyltransferase Clr4 and a second chro-
modomain protein, Swi6 (Elgin and Grewal 2003). The
initiation of silent heterochromatin requires not only
the RITS complex, but also Dcr1, the S. pombe homolog
of Dicer, and a putative RNA-dependent RNA polymer-
ase, Rdp1 (Hall et al. 2002). By analogy to silencing in
higher organisms, Dcr1 is presumed to generate siRNA
from dsRNA derived from centromeric DNA. Rdp1 has
been postulated either to generate this dsRNA itself or to
amplify the production of siRNA, but the precise func-
tion of RNA-dependent RNA polymerases in siRNA-di-
rected silencing is not yet understood.
In S. pombe, as in other eukaryotes, silencing can be
triggered by the introduction of exogenous long dsRNA
transcribed from transgenic DNA. Such dsRNA can si-
lence both endogenous (Schramke and Allshire 2003) and
exogenous genes (Raponi and Arndt 2003), although it
has not been determined whether this silencing is tran-
scriptional—like the endogenous silencing of centro-
meric DNA—or posttranscriptional, as is observed in the
RNAi pathway in animals and plants. Grewal, Martiens-
sen, and colleagues (Volpe et al. 2002) have shown that
the production of heterochromatin at the S. pombe cen-
tromere contains features characteristic of both tran-
scriptional and posttranscriptional silencing. Forward
transcription at this locus appears to be repressed by the
formation of heterochromatin, yet reverse transcription
through the same locus is not, suggesting that reverse
transcripts are degraded posttranscriptionally. To ad-
dress the question of whether PTGS exists in S. pombe
in the absence of transcriptional gene silencing (TGS),
we developed a simplified silencing system based on
green fluorescent protein (GFP) expression. Here, we
show that the introduction of a dsRNA hairpin corre-
sponding to a GFP transgene triggers classical RNAi in S.
pombe: introduction of GFP dsRNA causes a change in
the steady-state concentration of GFP mRNA, but not
the rate of GFP transcription. RNAi in S. pombe requires
Ago1, Dcr1, and Rdp1, but does not require Chp1, Tas3,
or Swi6, which are required for transcriptional silencing.
Our data suggest that Argonaute, Dicer, and RdRP play
common biochemical roles in functionally distinct si-
To test whether dsRNA triggers RNAi in S. pombe, we
engineered a strain containing the enhanced GFP (gfp)
protein coding sequence fused in-frame to the endog-
enous adh1 gene. PCR analysis (data not shown) verified
that the strain contained the adh1:gfp fusion gene inte-
grated at the adh1 locus on chromosome 3. The fusion
mRNA encoded by this locus was the only source of
adh1 mRNA in the strain. The strain also contained the
kanamycin resistance gene (aph, encoding the enzyme
aminoglycoside-3?-phosphotransferase) downstream from
adh1:gfp, with 628 bp separating the start of aph open
reading frame (ORF) from the end of the gfp ORF.
To trigger silencing of adh1:gfp, we engineered a plas-
mid-borne GFP hairpin (Fig. 1). The 760-bp gfp ORF was
cloned as an inverted repeat, with the sense and anti-
sense arms of the repeat separated by a 67-bp spacer con-
taining the first intron of the rad9 gene. The intron was
included, because intron-containing hairpin RNAs in-
duce PTGS in plants with nearly 100% efficiency (Smith
et al. 2000). The construct used here, when spliced, is
presumed to leave a loop of 14 unpaired nucleotides (nt).
Transcription of the GFP hairpin was under the control
of the thiamine-repressible nmt1 promoter; in the ab-
sence of thiamine, nmt1 is among the strongest promot-
ers in S. pombe (Moreno et al. 1991). The plasmid also
contained the ura4 gene, to permit selection for reten-
tion of the plasmid in the absence of uracil. (The reporter
strain contains a ura4 mutation and thus cannot synthe-
size uracil.) Expression of the GFP hairpin in the ura4-
strain had no effect on cell morphology or growth rate
(data not shown).
When the plasmid encoding the GFP hairpin was in-
troduced into the adh1:gfp target strain, the GFP fluo-
Sigova et al.
2360GENES & DEVELOPMENT
rescence intensity was reduced by more than twofold
(adh1:gfp + hairpin; Figs. 2, 5 [below]). Silencing was ob-
served only in the presence of the GFP hairpin plasmid.
No silencing was observed in a adh1:gfp strain trans-
formed with the same plasmid lacking the hairpin (Fig.
2A, empty vector), nor was silencing triggered by the
same plasmid expressing only sense or antisense GFP
transcript (Fig. 2B). Furthermore, when the silenced
strain was grown in the presence of uracil, the plasmid
expressing the GFP hairpin was lost (adh:gfp-hairpin), as
expected, and GFP fluorescence was restored to that of
the original adh1:gfp strain (Fig. 2A). The loss of silenc-
ing in nonselective conditions argues against epigenetic
(i.e., heritable) GFP silencing. Thus, silencing in our
GFP/GFP hairpin system is distinct from that observed
previously at both the centromeres and the mating-type
locus, where dsRNA is proposed to trigger assembly of
epigenetically heritable, repressive chromatin structures
(Volpe et al. 2002).
We could imagine two possible mechanisms for the
silencing of adh1:gfp by the GFP hairpin. Silencing
might reflect either unstable transcriptional repression
of the locus or bona fide posttranscriptional gene silenc-
ing, that is, RNAi. To distinguish between these possi-
bilities, we measured both the steady-state level (Fig. 3)
and the rate of nuclear transcription (Fig. 4) of the
adh1:gfp mRNA. To measure steady-state mRNA abun-
dance, we performed quantitative RT–PCR using prim-
ers that spanned the fusion site in the adh1:gfp transgene
(Fig. 1); actin (act1) mRNAs levels were measured as a
control. In the presence of the GFP hairpin, the level of
gfp mRNA was reduced more than twofold. The steady-
state level of aph mRNA was unchanged, demonstrating
the specificity of the silencing triggered by the GFP hair-
In contrast to its effect on steady-state mRNA levels,
the GFP hairpin caused no observable change in the tran-
scription of the adh1:gfp mRNA. Transcription was as-
sessed by the nuclear run-on method. Briefly, yeast cells
were permeabilized with detergent to permit the intro-
duction of ?-32P-UTP to label transcripts from elongat-
ing RNA polymerase II (RNA Pol II) (Volpe et al. 2002).
This technique measures the density of RNA Pol II on a
gene at the time of lysis, a reflection of the transcrip-
tional rate of the gene in vivo. To distinguish transcrip-
tion of the adh1:gfp mRNA from that of the GFP hairpin,
we measured the rate of transcription for the adh1:gfp
fusion mRNA using a probe for adh1. The rate of tran-
scription of the act1 locus provided a normalization con-
trol. The GFP hairpin did not detectably decrease the
rate of transcription of adh1:gfp (Fig. 4B). Control experi-
ments demonstrate that the nuclear run-on assay can
detect changes in the rate of transcription at least as
small as 25% (Supplementary Fig. 1). Thus, the nuclear
run-on data suggest that the reduction of steady-state
resentative FACS data demonstrating that GFP silencing occurs
only in the presence of the GFP hairpin. Silencing of GFP ex-
pression, measured by fluorescence intensity, was not observed
for the empty vector alone, and silencing was lost when the
silenced strain was grown in the presence of uracil to cause loss
of the hairpin-encoding plasmid. Ten-thousand cells were ana-
lyzed for each genotype. See also Figure 5. (B) Representative
FACS data demonstrating that GFP silencing is triggered by the
GFP hairpin but not by either sense or antisense GFP tran-
The GFP hairpin triggers adh1:gfp silencing. (A) Rep-
was fused in frame with the adh1 locus on chromosome 3. The
kanamycin resistance gene (aph) was inserted adjacent to the
adh1:gfp transgene as an insertion marker under the control of
the Ashbya gossypii translation elongation factor 1? gene pro-
moter. Red arrows indicate the position of adh1:gfp-specific
primers used in the quantitative RT–PCR assays in Figure 3.
The silencing trigger was expressed from an episomal plasmid
encoding a hairpin transcript corresponding to GFP expressed
from the nmt1 promoter. The GFP dsRNA hairpin contained
within the loop sequences the 67-bp intron 1 from rad9 to fa-
cilitate hairpin expression.
Experimental strategy. The coding sequence of GFP
Classical RNAi in S. pombe
GENES & DEVELOPMENT 2361
level of gfp mRNA triggered by the hairpin was posttran-
To test our hypothesis by an independent method, we
performed quantitative chromatin immunoprecipitation
(ChIP) using an antibody to RNA Pol II. Like the nuclear
run-on method, this method measures the relative den-
sity of RNA Pol II on a DNA sequence, a reflection of its
transcriptional rate. Again, we could detect no decrease
in the association of RNA Pol II with adh1:gfp in the
presence of the GFP hairpin, relative to its association in
the absence of the hairpin or in the presence of the plas-
mid lacking the hairpin sequences (empty vector) (Fig.
4C). We conclude that silencing of the adh1:gfp locus by
the GFP hairpin occurs posttranscriptionally.
Silencing at the centromere locus can be accompanied
by the spread of silencing across at least 3 kb of adjacent
noncentromeric sequence; the spread of silencing corre-
lates with the coating of the adjacent DNA with the
Swi6 protein (Partridge et al. 2000). In contrast, silencing
of adh1:gfp by the GFP hairpin was not accompanied by
silencing of the adjacent aph locus (Fig. 4), whose pro-
moter is less than 600 bp downstream from the end of
the gfp ORF (Fig. 1). Thus, in every respect, silencing of
adh1:gfp by the GFP hairpin appears to be entirely post-
transcriptional. We conclude that in S. pombe, as in ani-
mals, plants, protozoa, and basal eukaryotes, dsRNA
Dicer orthologs are required for posttranscriptional
gene silencing in Arabidopsis thaliana, Drosophila, C.
elegans, Dictyostelium discoideum, and Neurospora
(Bernstein et al. 2001; Knight and Bass 2001; Martens et
al. 2002; Catalanotto et al. 2004). Similarly, genes encod-
ing putative RdRP enzymes are required for PTGS in
plants, worms, Neurospora, and Dictyostelium (Cogoni
and Macino 1999; Dalmay et al. 2000; Smardon et al.
2000; Martens et al. 2002). The S. pombe dcr1 and rdp1
genes are required for the initiation of silent heterochro-
matin at centromeres and the mating-type locus (Hall et
al. 2002; Volpe et al. 2002). In fact, the Rdp1 protein is
physically associated with centromeric repeats, suggest-
ing that it plays a direct role in generating the dsRNA
silencing trigger (Volpe et al. 2002).
transcriptional. (A) Representative nuclear run-on experiment.
(B) The average value ± standard deviation for three indepen-
dent run-on experiments is presented. For each strain, the rates
of transcription for both aph and adh1 were measured and nor-
malized to the rate of transcription of act1. Firefly luciferase (Pp
luc) served as a negative control. No decrease in transcription of
adh1:gfp was observed when GFP was silenced by introduction
of the GFP hairpin-expressing plasmid. The rate of aph tran-
scription was essentially constant in all aph-containing geno-
types. (C) RNA Pol II chromatin immunoprecipitation (ChIP).
The density of RNA Pol II at the adh1:gfp locus was not de-
creased by the introduction of the GFP hairpin, reinforcing the
view that silencing triggered by this hairpin is posttranscrip-
tional. Data are the average of two independent trials. The data
for adh1:gfp and those for aph were separately normalized to
their respective values in the absence of the empty or hairpin
GFP silencing triggered by the GFP hairpin is post-
level of adh1:gfp mRNA. The steady-state level of mRNA was
measured for gfp and aph, relative to the expression level of the
act1 mRNA. For each strain three independent cultures were
assayed; each assay was analyzed by quantitative RT–PCR in
triplicate. The data are presented as the average of the three
independent trials (using the average value for the triplicate
quantitative RT–PCR reactions) ± the standard deviation of the
three independent trials.
GFP silencing reflects a reduction in the steady-state
Sigova et al.
2362 GENES & DEVELOPMENT
Small RNAs corresponding to GFP sequences accumu-
late in strains containing the GFP hairpin, but not the
empty vector (Supplementary Fig. 2). No small RNAs
were detected in either the dcr1? strain (Supplementary
Fig. 2) or the rdp1? strain (data not shown). Small RNA
production required only the GFP hairpin, not the
adh1:gfp target RNA, suggesting that the RdRP plays a
direct role in the production of the small RNAs from the
hairpin trigger itself or their subsequent stability, rather
than in the generation of secondary siRNAs templated
from the target RNA. Small RNAs could be detected in
the ago1? strain, consistent with a role for Ago1 in the
function, but not the production, of small RNAs (data
Next, we measured GFP fluorescence in the dcr1?,
rdp1?, and ago1? strains containing both the adh1:gfp
locus and the GFP hairpin. GFP was not silenced in the
absence of Dcr1 or Rdp1 (Fig. 5), consistent with the
requirement for these enzymes in small RNA produc-
tion. Thus, both Dicer and RdRP are required for post-
transcriptional silencing of GFP in S. pombe, just as they
are in C. elegans, the classical model for RNAi. Unlike
plants and Drosophila, S. pombe encodes a single dicer
gene. Our data, together with previous studies of silent
heterochromatin in S. pombe, demonstrate that a single
Dicer protein can support both TGS and PTGS. The
functional specializations of Dicer proteins in plants (Xie
et al. 2004) and flies (Lee et al. 2004) is unlikely to reflect
an inherent biochemical limitation of the ancestral
Dicer protein. Unlike C. elegans, which contains four
RdRP genes, S. pombe encodes a single RdRP homolog,
demonstrating that a single RdRP protein can mediate
both transcriptional and posttranscriptional gene silenc-
ing. The simplest explanation is that the S. pombe RdRP
protein supplies a common biochemical function to both
pathways. In C. elegans, both somatic and germ-line
RNAi triggered by dsRNA requires an RdRP (Smardon et
al. 2000). The finding that in S. pombe RNAi triggered by
dsRNA requires a functional RdRP suggests that the
mechanism of S. pombe silencing is more closely related
to RNAi in worms than PTGS in plants, where the RdRP
homolog SDE1 is required only for silencing triggered by
sense RNA-producing transgenes, not dsRNA (Dalmay
et al. 2000). We note that it remains possible that one of
the RdRP genes distinct from SDE1 could be required for
dsRNA-triggered silencing in plants.
In contrast to the Dicer and RdRP genes, Argonaute
proteins form a highly ramified family in all higher eu-
karyotes whose genomes have been fully sequenced. For
example, the C. elegans genome encodes 27 Argonaute
proteins. Current evidence suggests that Argonaute pro-
teins are functionally specialized in higher organisms.
For example, the C. elegans Argonaute protein Rde-1 is
required for RNAi, but not for miRNA function. Con-
versely, the Argonaute proteins Alg-1/Alg-2 are required
for miRNA function in worms. In contrast, miRNAs as-
sociate with at least four distinct Argonaute proteins in
humans, only one of which, Ago2, can direct target
mRNA cleavage (Liu et al. 2004; Meister et al. 2004). The
Drosophila genome encodes at least five Argonaute pro-
teins. In Drosophila, miRNAs associate with the Argo-
naute protein Ago1, whereas siRNAs associate with
Ago2; both Ago1 and Ago2 can direct small RNA-guided
target RNA cleavage (Okamura et al. 2004). Both of the
Drosophila Argonaute protein-encoding genes, piwi and
aubergine (aub) are required for H3 K9 methylation and
correct localization of two heterochromatic proteins,
HP1 and HP2, all processes associated with TGS (Pal-
Bhadra et al. 2004). Yet, mutations in aub abrogate the
RNAi-based silencing of the Stellate locus (Aravin et al.
2001), RNAi triggered by exogenous dsRNA (Kennerdell
et al. 2002), and RISC assembly (Tomari et al. 2004).
Furthermore, mutations in piwi partially block PTGS of
Adh transgenes (Pal-Bhadra et al. 2002). Because Argo-
naute proteins have been implicated in both the produc-
tion of the siRNA silencing trigger (Tabara et al. 2002)
and the execution of target RNA (Hammond et al. 2001;
Hutvágner and Zamore 2002; Martinez et al. 2002), a
partial resolution to this apparent paradox would be if
Argonaute proteins were required at multiple steps in
both the TGS and RNAi pathways. In this view, discrete
Argonaute proteins would be biochemically specialized
for functions common to TGS and PTGS, and others
dedicated to functions unique to each process. Alterna-
tively, the diversity of Argonaute proteins might simply
reflect their specialized patterns of spatial or temporal
expression or their intracellular localization.
The genome of S. pombe encodes a single Argonaute
protein, Ago1, which is required for transcriptional si-
lencing (Hall et al. 2002; Volpe et al. 2002). Is it also
required for RNAi in S. pombe? We introduced the GFP
hairpin into an ago1? strain bearing the adh1:gfp trans-
gene. Posttranscriptional silencing of adh1:gfp by the
GFP hairpin required ago1 (Fig. 5). Thus, a single Argo-
naute protein mediates both RNAi and TGS in S. pombe.
Ago1, Rdp1, and Dcr1 are all required for both TGS
and PTGS in S. pombe. Are these two distinct silencing
pathways mediated by a common complex? Ago1 is a
not the transcriptional silencing proteins Chp1, Tas3, or Swi6.
The geometric mean of fluorescence intensity was determined
for each strain from experiments like that in Figure 2, and nor-
malized to the geometric mean for the wild-type adh1:gfp
strain. The data are the average ± standard deviation for three
GFP silencing depends on the RNAi machinery but
Classical RNAi in S. pombe
GENES & DEVELOPMENT2363
component of the RITS complex, which also contains
Chp1, Tas3, and siRNA (Verdel et al. 2004). To test
whether the RITS complex mediates RNAi in S. pombe,
we asked whether either Chp1 or Tas3 is required for
silencing of the adh1:gfp transgene by the GFP hairpin.
We also asked whether the Swi6 or Clr4 proteins, which
are not components of the RITS, but are required for
RITS-initiated production of silent heterochromatin, are
required for posttranscriptional silencing of the adh1:gfp
transgene by the GFP hairpin. Deletion of chp1, tas3, or
swi6 had no effect on GFP silencing by the GFP hairpin
(Fig. 5). Thus, the requirements for TGS and PTGS in S.
pombe are genetically distinct. In contrast, deletion of
clr4 had an effect on posttranscriptional silencing of the
adh:gfp locus. This finding is consistent with a previous
report that siRNAs do not accumulate to normal levels
in the clr4 mutant strain (Schramke and Allshire 2003);
how Clr4 functions in siRNA biogenesis is not under-
stood. That RNAi in S. pombe does not require Chp1,
Tas3, or Swi6 suggests that a complex distinct from RITS
mediates siRNA-directed target mRNA degradation. Al-
ternatively, once siRNAs are produced by the action of
Dcr1 and Rdp1, only Ago1 itself may be required for
PTGS in S. pombe.
In this study, we have demonstrated that a dsRNA de-
rived from a hairpin transcript can trigger posttranscrip-
tional silencing of a corresponding mRNA in S. pombe.
Schramke and Allshire (2003) demonstrated that a simi-
lar hairpin transcript, corresponding to the ura4 locus,
could trigger transcriptional silencing. In both studies,
silencing triggered by a hairpin transcript required the
RNAi machinery—Dcr1, Rdp1, and Ago1. Transcrip-
tional silencing, unlike posttranscriptional silencing, re-
quired components of the transcriptional silencing appa-
ratus—Chp1, Tas3, or Swi6. Robust silencing by both
pathways requires the chromodomain protein Clr4,
which appears to play a role in siRNA biogenesis or sta-
bility. Why does the GFP hairpin construct presented
here trigger exclusively posttranscriptional silencing,
whereas the previously studied ura4 hairpin triggered
transcriptional silencing? One possible explanation is
that the GFP hairpin used here included an efficiently
spliced intron between the two arms of the hairpin. We
presume that splicing of the intron promotes the accu-
mulation of GFP dsRNA in the cytoplasm. In contrast,
the ura4 hairpin construct of Schramke and Allshire
(2003) contained an unspliced spacer sequence between
the hairpin arms. Thus, the ura4 hairpin may be local-
ized largely to the nucleus. A difference in subcellular
localization might explain the different results obtained
by the two studies. Alternatively, silencing of ura4 by
the ura4-specific hairpin might comprise a mixture of
transcriptional and posttranscriptional silencing. In this
case, transcriptional silencing might not occur at the
adh1 locus, even if the GFP hairpin-derived siRNAs trig-
ger histone modification, perhaps because the gene is
strongly expressed or is in a region of the genome other-
wise refractory to heterochromatin formation. Nonethe-
less, our data, together with those of Schramke and
Allshire (2003), clearly show that at least two distinct
silencing responses can be initiated by a common RNAi
machinery, without resorting to specialized forms of
Dicer, RdRP, or Argonaute proteins. The demonstration
that fission yeast contain a functional RNAi pathway
now provides a simplified, genetically tractable model in
which to study how the nature of the silencing trigger or
of the silencing target determines the silencing pathway
evoked—posttranscriptional or transcriptional.
Materials and methods
S. pombe strain construction
Fission yeast were grown and manipulated as described (Moreno
et al. 1991). Unless otherwise stated, all strains (Table 1) were
grown at 30°C in EMM2 media supplemented with histidine,
leucine, adenine, and uracil as appropriate. Deletion and fusion
strains were constructed by PCR-based cassette mutagenesis as
described (Bahler et al. 1998) using the following oligonucleo-
CTTTCCAAGCGGATCCCCGGGTTAATTAA-3? and 5?-AAA
GGAATTCGAGCTCGTTTAAAC-3? for adh1:gfp; 5?-GTTTGG
CGCCACTTCTAAATAAGC-3? and 5?-TAAGGAAGTAAAAG
Strains used in this study
leu1-32 h+ura4-294 pAS1
h−leu1-32 ura4-D18 his7-366 ade6-210 adh1:gfp
h−leu1-32 ura4-D18 his7-366 ade6-210 adh1:gfp
h+leu1-32 ura4-? ade6-210 adh1:gfp pRIP2
h+leu1-32 ura4-? ade6-210 ago1?kanMX adh1:gfp
h−leu1-32 ura4-? ade6-210 rdp1?kanMX adh1:gfp
h−leu1-32 ura4-? dcr1?kanMX adh1:gfp pAS1
h+leu1-32 ura4-D18 his7-366 ade6-210 chp1?his7
h−leu1-32 ura4-D18 ade6-210 tas3?kanMX
h+leu1-32 ura4-? his7-366 ade6-210 arg3D
swi6?arg3 adh1:gfp pAS1
h+leu1-32 ura4-? his7-366 ade6-210 clr4?LEU2
A question mark indicates it is not known if the strain contains
either the ura4-294 or ura4-D18 allele of ura4.
Sigova et al.
2364GENES & DEVELOPMENT
for rdp1?kanMX; and 5?-TAGCTTAGGATTCATTATTTTTT
AAGTCGAATTCGAGCTCGTTTAAAC-3? for dcr1?kanMX.
Construction of the silencing trigger
Full-length eGFP sequence was amplified using the following
ATGTG-3?. PCR fragments were directly subcloned into pCR4-
TOPO (Invitrogen) to produce pCR4-TOPO-B and pCR4-
TOPO-N constructs. Both constructs were digested with NheI
and NotI restriction enzymes; restriction fragments were gel-
purified and ligated to produce pCR4-TOPO-B+N. This con-
struct was digested with NdeI and BamHI, and the insert was
subcloned into the corresponding restriction sites of pRIP2
(Maundrell 1993) to produce pAS1. A second plasmid bearing an
ars1 sequence was also prepared by subcloning the EcoRI frag-
ment of pREP2 into the corresponding site in pAS1 to produce
pAS2. When transformed, both pAS1 and pAS2 produced epi-
somal transformants that gave similar results in our experi-
ments. Results for the pAS1 construct only are presented here.
Cells were harvested from liquid cultures at mid-exponential
phase (O.D.600= 0.1–0.4). Total RNA was extracted from cells
using the hot phenol method (Lyne et al. 2003). Purified RNA
was treated with DNase (RQ1, Promega) and analyzed by Quan-
titative RT—PCR in a DNA Engine OPTICON2 (MJ Research)
using the QuantiTect SYBR Green PCR Kit (QIAGEN) accord-
ing to the manufacturer’s instruction Analysis was performed
using Opticon Monitor (MJ Research), Excel (Microsoft), and
IgorPro 5.0 (Wavemetrics) software. PCR primers were 5?-GAT
TGCCGGCCGTATCGTCTT-3? and 5?-GCCCATTAACATC
ACCATCTA-3? for adh1:gfp; 5?-CTGAAACATGGCAAAGG
TAGC-3? and 5?-GGGATCGCAGTGGTGAGTAAC-3? for aph;
5?-ACTTTGCTACGTCGCTTTGGAC-3? and 5?-CGTTTCCG
mRNA levels were determined from the threshold cycle for
amplification using the 2−??CTmethod (Livak and Schmittgen
2001). Control experiments measuring the change in ?CTwith
template dilution demonstrated that the efficiency of amplifi-
cation of the target gene (adh1:gfp or aph) and the control (act1)
was approximately equal.
Nuclear run-on analysis
RNA transcriptional elongation was analyzed by nuclear run-on
essentially as described (Volpe et al. 2002). Labeled RNA from
permeabilized cells was purified with hot phenol as described
above and hybridized to nylon membranes containing strand-
specific RNA probes for firefly luciferase Photinus pyralis (Pp-
luc), aph, adh1, and act1. Full-length RNA probes were tran-
scribed with T7 RNA polymerase from PCR templates prepared
with the following oligonucleotides: 5?-gcgtaatacgactcactatag
GAAAAACTCATCGAGCATCAAATG-3? and 5?-GGGTAAG
GAAAAGACTCACG-3? for aph; 5?-gcgtaatacgactcactatagggC
TTGGAAAGGTCCAAGACGATAC-3? and 5?-GACTATTCC
TGACAAGCAGTTG-3? for adh1; 5?-gcgtaatacgactcactatagGA
AGCACTTACGGTAAACGATAC-3? and 5?-GGAAGAAGAA
ATCGCAGCGTTG-3? for act1, where the T7 promoter se-
quence is in lowercase. Pp luc RNA was transcribed from a
PCR-generated DNA template amplified from the pGL-2 Con-
trol vector (Promega) using 5?-gcgtaatacgactcactatagGAGAG
Log-phase cells were harvested, washed in PBS (10 mM
Na2HPO4, 1.7 mM KH2PO4 at pH 7.4, 137 mM NaCl, and 2.7
mM KCl) and resuspended in 500 µL PBS. Fluorescence was
analyzed using a FACScan Flow Cytometer (Becton-Dickinson).
Constant settings were maintained for all experiments. Data
were acquired from 10,000 cells in all experiments and analyzed
with Cell Quest (Becton-Dickinson) and FACSPress 1.3 (Ray
Chromatin immunoprecipitation (ChIP)
The density of RNA polymerase II on the adh1:gfp and aph
genes was measured essentially as described (Takahashi et al.
2000). Briefly, 50-mL cultures were grown until O.D.600= 0.8.
Protein was cross-linked to DNA by adding formaldehyde to a
final concentration of 1% and incubating the cultures for 10–15
min at room temperature with gentle shaking. Cross-linking
was quenched by adding glycine to a final concentration of
0.125 M. Cells were harvested by centrifugation, washed three
times with ice-cold PBS, frozen in liquid nitrogen, and stored at
−80°C until use. Cell pellets were resuspended in 200 µL of lysis
buffer (50 mM HEPES-KOH at pH 7.5, 150 mM NaCl, 1 mM
EDTA, 0.1% (w/v) sodium deoxycholate, 1% [w/v] Triton
X-100, 0.1% [w/v] SDS) containing for each 50 mL one tablet of
Complete, EDTA-free Protease Inhibitor Cocktail (Roche) and
500 µL of Protease Inhibitor Cocktail For Fungal and Yeast Cells
(Sigma), then vortexed with an equal volume of silica beads for
20 min at 4°C. The volume of the cell suspension was then
adjusted by adding lysis buffer to a final volume of 650 µL. After
sonicating to shear the chromosomal DNA to 250–500 bp, cell
suspensions were centrifuged at 16,000 × g for 10 min at 4°C,
and the supernatants were divided into three equal portions.
The volume of each portion was adjusted to 500 µL, anti-RNA
polymerase II (Pol II) antibody (Covance), anti-HA antibody (Co-
vance), or no antibody was added, and the samples were incu-
bated on a rotating wheel for 4 h at 4°C. Fifty microliters salmon
sperm DNA/protein A agarose beads (Upstate) was added and
the incubation continued for 1.5–2 h at 4°C. Beads were pelleted
by centrifugation at 500 × g for 2 min, supernatants discarded,
then the beads were washed once for 5 min with 1 mL of each
of the following buffers: buffer 1 (lysis buffer with 1% sodium
deoxycholate), buffer 2 (lysis buffer with 1% sodium deoxycho-
late and 1M NaCl), buffer 3 (50 mM Tris-HCl at pH 8.0, 0.25 M
LiCl, 1 mM EDTA, 0.5% NP-40, 0.5% sodium deoxycholate),
and twice with 10 mM Tris-HCl (pH 7.6) containing 10 mM
EDTA. Samples were treated with proteinase K for 4 h at 42°C
and cross-linking reversed by overnight incubation at 65°C. Af-
ter phenol-chloroform extraction and ethanol precipitation, the
DNA was resuspended in 20 µL of water. Five percent of the
total DNA was used in quantitative PCR analysis as described
above (“RNA analysis”). The difference in threshold cycles ob-
tained when amplifying Pol II immunoprecipitated DNA and
control DNA precipitated using nonspecific (HA) or no antibody
Classical RNAi in S. pombe
GENES & DEVELOPMENT2365
was at least 10 cycles. Control experiments demonstrated that
less than twofold differences in the density of RNA Pol II on a
gene were readily detectable with our protocol.
We thank members of the Zamore lab for support and for com-
ments on the manuscript, Tamal Raha and Michael Green for
help with Pol II ChIP, Janet Partridge and Robin Allshire for the
chp1?, clr4?, and swi6? strains, and Shiv Grewal and Danesh
Moazed for the tas3? strain.
Aravin, A.A., Naumova, N.M., Tulin, A.V., Vagin, V.V., Ro-
zovsky, Y.M., and Gvozdev, V.A. 2001. Double-stranded
RNA-mediated silencing of genomic tandem repeats and
transposable elements in the D. melanogaster germline.
Curr. Biol. 11: 1017–1027.
Bahler, J., Wu, J.Q., Longtine, M.S., Shah, N.G., McKenzie III,
A., Steever, A.B., Wach, A., Philippsen, P., and Pringle, J.R.
1998. Heterologous modules for efficient and versatile PCR-
based gene targeting in Schizosaccharomyces pombe. Yeast
Bernstein, E., Caudy, A.A., Hammond, S.M., and Hannon, G.J.
2001. Role for a bidentate ribonuclease in the initiation step
of RNA interference. Nature 409: 363–366.
Catalanotto, C., Azzalin, G., Macino, G., and Cogoni, C. 2002.
Involvement of small RNAs and role of the qde genes in the
gene silencing pathway in Neurospora. Genes & Dev.
Catalanotto, C., Pallotta, M., ReFalo, P., Sachs, M.S., Vayssie,
L., Macino, G., and Cogoni, C. 2004. Redundancy of the two
dicer genes in transgene-induced posttranscriptional gene si-
lencing in Neurospora crassa. Mol. Cell Biol. 24: 2536–2545.
Cerutti, L., Mian, N., and Bateman, A. 2000. Domains in gene
silencing and cell differentiation proteins: The novel PAZ
domain and redefinition of the Piwi domain. Trends Bio-
chem. Sci. 25: 481–482.
Cogoni, C. and Macino, G. 1997. Isolation of quelling-defective
(qde) mutants impaired in posttranscriptional transgene-in-
duced gene silencing in Neurospora crassa. Proc. Natl. Acad.
Sci. 94: 10233–10238.
———. 1999. Gene silencing in Neurospora crassa requires a
protein homologous to RNA-dependent RNA polymerase.
Nature 399: 166–169.
Dalmay, T., Hamilton, A., Rudd, S., Angell, S., and Baulcombe,
D.C. 2000. An RNA-dependent RNA polymerase gene in
Arabidopsis is required for posttranscriptional gene silencing
mediated by a transgene but not by a virus. Cell 101: 543–
Elgin, S.C. and Grewal, S.I. 2003. Heterochromatin: Silence is
golden. Curr. Biol. 13: R895–R898.
Fagard, M., Boutet, S., Morel, J.-B., Bellini, C., and Vaucheret, H.
2000. AGO1, QDE-2, and RDE-1 are related proteins re-
quired for post-transcriptional gene silencing in plants,
quelling in fungi, and RNA interference in animals. Proc.
Natl. Acad. Sci. 97: 11650–11654.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E.,
and Mello, C.C. 1998. Potent and specific genetic interfer-
ence by double-stranded RNA in Caenorhabditis elegans.
Nature 391: 806–811.
Grishok, A., Pasquinelli, A.E., Conte, D., Li, N., Parrish, S., Ha,
I., Baillie, D.L., Fire, A., Ruvkun, G., and Mello, C.C. 2001.
Genes and mechanisms related to RNA interference regulate
expression of the small temporal RNAs that control C. el-
egans developmental timing. Cell 106: 23–34.
Hall, I.M., Shankaranarayana, G.D., Noma, K.-I., Ayoub, N.,
Cohen, A., and Grewal, S.I.S. 2002. Establishment and main-
tenance of a heterochromatin domain. Science 297: 2232–
Hamilton, A.J. and Baulcombe, D.C. 1999. A species of small
antisense RNA in posttranscriptional gene silencing in
plants. Science 286: 950–952.
Hammond, S.M., Bernstein, E., Beach, D., and Hannon, G.J.
2000. An RNA-directed nuclease mediates post-transcrip-
tional gene silencing in Drosophila cells. Nature 404: 293–
Hammond, S.M., Boettcher, S., Caudy, A.A., Kobayashi, R., and
Hannon, G.J. 2001. Argonaute2, a link between genetic and
biochemical analyses of RNAi. Science 293: 1146–1150.
Hutvágner, G. and Zamore, P.D. 2002. A microRNA in a mul-
tiple-turnover RNAi enzyme complex. Science 297: 2056–
Kennerdell, J.R., Yamaguchi, S., and Carthew, R.W. 2002. RNAi
is activated during Drosophila oocyte maturation in a man-
ner dependent on aubergine and spindle-E. Genes & Dev.
Knight, S.W. and Bass, B.L. 2001. A role for the RNase III en-
zyme DCR-1 in RNA interference and germ line develop-
ment in Caenorhabditis elegans. Science 293: 2269–2271.
Lagos-Quintana, M., Rauhut, R., Lendeckel, W., and Tuschl, T.
2001. Identification of novel genes coding for small ex-
pressed RNAs. Science 294: 853–858.
Lau, N.C., Lim, L.P., Weinstein, E.G., and Bartel, D.P. 2001. An
abundant class of tiny RNAs with probable regulatory roles
in Caenorhabditis elegans. Science 294: 858–862.
Lee, R.C. and Ambros, V. 2001. An extensive class of small
RNAs in Caenorhabditis elegans. Science 294: 862–864.
Lee, Y.S., Nakahara, K., Pham, J.W., Kim, K., He, Z., Sonthei-
mer, E.J., and Carthew, R.W. 2004. Distinct roles for Dro-
sophila Dicer-1 and Dicer-2 in the siRNA/miRNA silencing
pathways. Cell 117: 69–81.
Liu, J., Carmell, M.A., Rivas, F.V., Marsden, C.G., Thomson,
J.M., Song, J.-J., Hammond, S.M., Joshua-Tor, L., and Han-
non, G.J. 2004. Argonaute2 is the catalytic engine of mam-
malian RNAi. Science (in press). [Epub ahead of print July 29,
Livak, K.J. and Schmittgen, T.D. 2001. Analysis of relative gene
expression data using real-time quantitative PCR and the
2−??CTmethod. Methods 25: 402–408.
Lyne, R., Burns, G., Mata, J., Penkett, C.J., Rustici, G., Chen, D.,
Langford, C., Vetrie, D., and Bahler, J. 2003. Whole-genome
microarrays of fission yeast: Characteristics, accuracy, re-
producibility, and processing of array data. BMC Genomics
Postlethwait, P., and Nellen, W. 2002. RNAi in Dictyos-
telium: The role of RNA-directed RNA polymerases and
double-stranded RNase. Mol. Biol. Cell 13: 445–453.
Martinez, J., Patkaniowska, A., Urlaub, H., Lührmann, R., and
Tuschl, T. 2002. Single stranded antisense siRNA guide tar-
get RNA cleavage in RNAi. Cell 110: 563–574.
Maundrell, K. 1993. Thiamine-repressible expression vectors
pREP and pRIP for fission yeast. Gene 123: 127–130.
Meister, G., Landthaler, M., Patkaniowska, A., Dorsett, Y.,
Teng, G., and Tuschl, T. 2004. Human Argonaute2 mediates
RNA cleavage targeted by miRNAs and siRNAs. Mol. Cell
Mochizuki, K., Fine, N.A., Fujisawa, T., and Gorovsky, M.A.
Sigova et al.
2366 GENES & DEVELOPMENT
2002. Analysis of a piwi-related gene implicates small RNAs Download full-text
in genome rearrangement in Tetrahymena. Cell 110: 689–
Morel, J.B., Godon, C., Mourrain, P., Beclin, C., Boutet, S.,
Feuerbach, F., Proux, F., and Vaucheret, H. 2002. Fertile hy-
pomorphic ARGONAUTE (ago1) mutants impaired in post-
transcriptional gene silencing and virus resistance. Plant
Cell 14: 629–639.
Moreno, S., Klar, A., and Nurse, P. 1991. Molecular genetic
analysis of fission yeast Schizosaccharomyces pombe. Meth-
ods Enzymol. 194: 795–823.
Nykänen, A., Haley, B., and Zamore, P.D. 2001. ATP Require-
ments and small interfering RNA structure in the RNA in-
terference pathway. Cell 107: 309–321.
Okamura, K., Ishizuka, A., Siomi, H., and Siomi, M.C. 2004.
Distinct roles for Argonaute proteins in small RNA-directed
RNA cleavage pathways. Genes & Dev. 18: 1655–1666.
Pal-Bhadra, M., Bhadra, U., and Birchler, J.A. 2002. RNAi related
mechanisms affect both transcriptional and posttranscrip-
tional transgene silencing in Drosophila. Mol. Cell 9: 315–
Pal-Bhadra, M., Leibovitch, B.A., Gandhi, S.G., Rao, M., Bhadra,
U., Birchler, J.A., and Elgin, S.C. 2004. Heterochromatic si-
lencing and HP1 localization in Drosophila are dependent on
the RNAi machinery. Science 303: 669–672.
Partridge, J.F., Borgstrom, B., and Allshire, R.C. 2000. Distinct
protein interaction domains and protein spreading in a com-
plex centromere. Genes & Dev. 14: 783–791.
Raponi, M. and Arndt, G.M. 2003. Double-stranded RNA-me-
diated gene silencing in fission yeast. Nucleic Acids Res.
Schramke, V. and Allshire, R. 2003. Hairpin RNAs and retro-
transposon LTRs effect RNAi and chromatin-based gene si-
lencing. Science 301: 1069–1074.
Shi, H., Djikeng, A., Tschudi, C., and Ullu, E. 2004. Argonaute
protein in the early divergent eukaryote Trypanosoma bru-
cei: Control of small interfering RNA accumulation and ret-
roposon transcript abundance. Mol. Cell Biol. 24: 420–427.
Smardon, A., Spoerke, J., Stacey, S., Klein, M., Mackin, N., and
Maine, E. 2000. EGO-1 is related to RNA-directed RNA
polymerase and functions in germ-line development and
RNA interference in C. elegans. Curr. Biol. 10: 169–178.
Smith, N.A., Singh, S.P., Wang, M.B., Stoutjesdijk, P.A., Green,
A.G., and Waterhouse, P.M. 2000. Total silencing by intron-
spliced hairpin RNAs. Nature 407: 319–320.
Song, J.-J., Smith, S.K., Hannon, G.J., and Joshua-Tor, L. 2004.
Crystal structure of Argonaute and its implications for RISC
slicer activity. Science (in press). [Epub ahead of print July
29, 2004; 10.1126/science.1102514.]
Tabara, H., Sarkissian, M., Kelly, W.G., Fleenor, J., Grishok, A.,
Timmons, L., Fire, A., and Mello, C.C. 1999. The rde-1 gene,
RNA interference, and transposon silencing in C. elegans.
Cell 99: 123–132.
Tabara, H., Yigit, E., Siomi, H., and Mello, C.C. 2002. The
dsRNA binding protein RDE-4 interacts with RDE-1, DCR-
1, and a DexH-box helicase to direct RNAi in C. elegans.
Cell 109: 861–871.
Takahashi, K., Saitoh, S., and Yanagida, M. 2000. Application of
the chromatin immunoprecipitation method to identify in
vivo protein–DNA associations in fission yeast. Science
STKE 2000: PL1.
Tijsterman, M., Okihara, K.L., Thijssen, K., and Plasterk, R.H.
2002. PPW-1, a PAZ/PIWI protein required for efficient ger-
mline RNAi, is defective in a natural isolate of C. elegans.
Curr. Biol. 12: 1535–1540.
Tomari, Y., Du, T., Haley, B., Schwarz, D.S., Bennett, R., Cook,
H.A., Koppetsch, B.S., Theurkauf, W.E., and Zamore, P.D.
2004. RISC assembly defects in the Drosophila RNAi mu-
tant armitage. Cell 116: 831–841.
Vaucheret, H., Vazquez, F., Crete, P., and Bartel, D.P. 2004. The
action of ARGONAUTE1 in the miRNA pathway and its
regulation by the miRNA pathway are crucial for plant de-
velopment. Genes & Dev. 18: 1187–1197.
Verdel, A., Jia, S., Gerber, S., Sugiyama, T., Gygi, S., Grewal, S.I.,
and Moazed, D. 2004. RNAi-mediated targeting of hetero-
chromatin by the RITS complex. Science 303: 672–676.
Volpe, T.A., Kidner, C., Hall, I.M., Teng, G., Grewal, S.I.S., and
Martienssen, R.A. 2002. Regulation of heterchromatic si-
lencing and Histone H3 Lysine-9 methylation by RNAi. Sci-
ence 297: 1833–1837.
Williams, R.W. and Rubin, G.M. 2002. Argonaute1 is required
for efficient RNA interference in Drosophila embryos. Proc.
Natl. Acad. Sci. 99: 6889–6894.
Xie, Z., Johansen, L.K., Gustafson, A.M., Kasschau, K.D., Lellis,
A.D., Zilberman, D., Jacobsen, S.E., and Carrington, J.C.
2004. Genetic and functional diversification of small RNA
pathways in plants. PLoS Biol. 2: E104.
Zamore, P.D., Tuschl, T., Sharp, P.A., and Bartel, D.P. 2000.
RNAi: Double-stranded RNA directs the ATP-dependent
cleavage of mRNA at 21 to 23 nucleotide intervals. Cell
Zilberman, D., Cao, X., and Jacobsen, S.E. 2003. Argonaute4
control of locus-specific siRNA accumulation and DNA and
histone methylation. Science 299: 716–719.
Classical RNAi in S. pombe
GENES & DEVELOPMENT2367