The yeast nucleolar protein Nop53p has previously been identified in the pre-60S complex and shown to affect pre-rRNA processing by directly binding to 5.8S rRNA, and to interact with Nop
Trang 1Synthesis of mature ribosomal subunits in yeast involves
many steps of rRNA processing, directed by at least 180
factors that include proteins and snoRNP complexes
The protein factors include rRNA-modifying enzymes,
endonucleases, exonucleases, RNA helicases, GTPases
and snoRNA-associated proteins [1,2] Three of the
rRNAs (18S, 5.8S and 25S) are transcribed as a 35S
pre-cursor, which undergoes a series of processing reactions,
including endo- and exonucleolytic cleavage and
nucleo-tide modifications Some of the processing factors and
ribosomal proteins assemble into the complex early
during transcription [3–6], leading to formation of
vari-ous pre-ribosomal particles, the first of which is the 90S
complex [7,8] Most of the factors forming the 90S
com-plex are involved in processing of 18S rRNA, or are part
of the 40S ribosome subunits [7,8]
Co-purification of proteins and mass spectrometry studies have identified many of the factors involved in rRNA processing, such as the small ribosomal subunit (SSU) complex processome and Dim2p [9,10] The pro-cessing factors of the large ribosomal subunit bind later during transcription of the 35S pre-rRNA, or after the early cleavages at sites A0, A1 and A2that separate the pre-40S and pre-60S complexes [8,11,12], and include some of the large ribosomal subunit proteins, as well as 27S processing factors [11] As some ribosomal proteins bind early during rRNA transcription, they also play
an important role in rRNA processing Rpl3p and the IPI complex have recently been shown to be involved in cleavages at ITS2, and their depletion leads to accumu-lation of the pre-rRNAs 35S and 27S, and a decrease in mature 25S levels [2,9]
exosome activation; pre-60S; pre-rRNA
processing; protein–RNA interaction;
ribosome biogenesis
Correspondence
C C Oliveira, Department of Biochemistry,
Institute of Chemistry, University of Sa˜o
Paulo, Av Prof Lineu Prestes 748, Sa˜o
Paulo, CEP 05508-900, Brazil
Fax: +55 11 38155579
Tel: +55 11 30913810 (ext 208)
E-mail: ccoliv@iq.usp.br
(Received 2 April 2008, revised 22 May
2008, accepted 20 June 2008)
doi:10.1111/j.1742-4658.2008.06565.x
osomal trans-acting factors that form intriguingly organized complexes One of the early stages of pre-rRNA processing includes formation of the two intermediate complexes pre-40S and pre-60S, which then form the mature ribosome subunits Each of these complexes contains specific pre-rRNAs, ribosomal proteins and processing factors The yeast nucleolar protein Nop53p has previously been identified in the pre-60S complex and shown to affect pre-rRNA processing by directly binding to 5.8S rRNA, and to interact with Nop17p and Nip7p, which are also involved in this process Here we show that Nop53p binds 5.8S rRNA co-transcriptionally through its N-terminal region, and that this protein portion can also par-tially complement growth of the conditional mutant strain Dnop53⁄ GAL::-NOP53 Nop53p interacts with Rrp6p and activates the exosome in vitro These results indicate that Nop53p may recruit the exosome to 7S pre-rRNA for processing Consistent with this observation and similar to the observed in exosome mutants, depletion of Nop53p leads to accumula-tion of polyadenylated pre-rRNAs
Abbreviations
ETS, external transcribed spacer; IPI, involved in processing of ITS2; ITS2, internal transcribed spacer 2; LSU, large ribosomal subunit; snoRNP, small nucleolar ribonucleoprotein; SSU, small ribosomal subunit; TAP, tandem affinity purification; TEV, tobacco etch virus protein; YNB, yeast minimal synthetic medium.
Trang 2The exosome is a complex of exoribonucleases that
is involved in the late steps of pre-rRNA processing,
and is directly responsible for the 3¢ fi 5¢
exonucleo-lytic digestion of the 3¢ extension of the 7S pre-rRNA
and formation of the mature 5.8S rRNA [13]
Inter-estingly, despite being directly involved in the late
steps of processing, depletion of essential subunits of
the exosome leads to accumulation of the pre-rRNAs
35S, 27S and 7S [13–16] The exosome is also
involved in processing of snoRNAs and degradation
of defective rRNAs and cytoplasmic mRNAs [17,18]
These results indicate that the exosome has two types
of substrates, one type that requires maturation
through removal of 3¢ extensions, and another type
that has not been correctly processed and is going to
be subjected to rapid and complete degradation For
the exosome to differentiate between these two kinds
of substrates, it requires either RNA signals or
associ-ation with other proteins [19] One of the
exosome-interacting proteins is Rrp47p, which also participates
in 3¢ fi 5¢ processing of nuclear stable RNAs [20] The
exosome also associates with the TRAMP complex
(composed of the factors Trf4p⁄ Trf5p–Air1p ⁄ Air2p–
Mtr4p) that is responsible for the polyadenylation of
aberrant RNAs, thereby stimulating exosome activity
in vitroand in vivo [21–24]
Many other exosome-interacting proteins have been
identified in yeast Rrp43p has been reported to
inter-act with Nip7p and Nop17p [25,26] The nuclear Lsm
complex has been shown to be a necessary cofactor for
5¢ and 3¢ exoribonucleases involved in the processing
of 7S pre-rRNA [27] The Rex complex, formed by the
RNase D class of RNases, is also required for 5.8S
rRNA maturation [28] In addition, the Ski complex,
formed by proteins Ski2p, Skip3p and Ski8p, is an
exosome cofactor involved in 3¢ fi 5¢ cytoplasmic
mRNA degradation [29,30]
Nop53p has been shown to bind 5.8S rRNA, and its
depletion leads to accumulation of 7S, a phenotype
similar to that caused by the depletion of core
exo-some subunits [31] Nop53p interacts with the
nucleo-lar proteins Nop17p and Nip7p [31], both of which
interact with the exosome and are involved in
pre-rRNA processing [25,26] In this study, we
demon-strate that Nop53p binds 5.8S rRNA through its
N-terminal region, and that Nop53p is recruited to
pre-rRNA early during transcription We also show
that Nop53p interacts directly with the exosome
sub-unit Rrp6p and with the TRAMP subsub-unit Trf4p, and
demonstrate that Nop53p activates the exosome in
in vitroRNA degradation assays These results indicate
that Nop53p is an exosome regulatory factor
Results
Nop53p is recruited co-transcriptionally to pre-rRNA
Nop53p is a nucleolar protein that has previously been shown to be involved in pre-rRNA processing and to co-immunoprecipitate the 27S and 7S pre-rRNAs and the mature 5.8S rRNA [31–33], and to bind 5.8S rRNA in vitro [31] In order to determine whether Nop53p interacts with the pre-rRNA early during transcription, chromatin immunoprecipitation (ChIP) experiments were performed, using the fusion protein protein A–Nop53p, and protein A as a negative con-trol Immunoprecipitated chromatin was analyzed by PCR reactions using primers complementary to vari-ous regions of the rDNA, or to the snR37 (box
H⁄ ACA) and snR74 (box C ⁄ D) snoRNA genes, with the latter being used as controls The results show that protein A–Nop53p immunoprecipitates 5.8S and 25S chromatin and, to a lesser extent, 18S chromatin (Fig 1) In order to verify whether protein A–Nop53p chromatin binding was dependent on active transcrip-tion, ChIP was also performed in the presence of RNases A⁄ T1 The results show that, in the presence
of RNases A⁄ T1, protein A–Nop53p chromatin immu-noprecipitation is reduced to the same levels as the control protein A (Fig 1) Further evidence for the Nop53p co-transcriptional interaction with pre-rRNA was obtained by observation of direct interaction between Nop53p and RNA polymerase I transcription factor Rrn3p [34] by protein pull-down (Fig 1E) In these experiments, recombinant GST–Rrn3p pulled down His–Nop53p, whereas GST did not (Fig 1E) These results indicate that Nop53p binds 5.8S rRNA co-transcriptionally, which is in accordance with its nucleolar localization
Analysis of Nop53p regions involved in RNA interaction
Although Nop53p binds RNA [31], no RNA recog-nition motif could be identified in its sequence In order to determine the region of Nop53p that is responsible for the interaction with RNA in the pre-60S complex, truncated Nop53p mutants were obtained, which correspond to the breakdown frag-ments of Nop53p visualized on SDS–PAGE gels (Fig 1) [31], and may contain stable structural domains of the protein Co-immunoprecipitation experiments were then performed using the truncated mutants fused to protein A (A–N-Nop53p and
Trang 3Nop53p) In these experiments, various salt
concentra-tions were used to analyze the strength of the
interaction between truncated Nop53p mutants and
the pre-60S complex The protein A–Nop53p fusion
efficiently precipitated 27S, 7S and 5.8S rRNAs, even
in the presence of 500 mm potassium acetate (Fig 2A),
indicating that Nop53p binds stably to the pre-60S
complex Although Nop53p precipitates 25S, lower
relative amounts of this rRNA were co-precipitated at
higher salt concentrations, indicating that Nop53p
binds less efficiently to mature 60S subunits, which is also consistent with its nucleolar localization
The truncated Nop53p mutant fusion A–N-Nop53p (N-terminal portion of Nop53p) also precipitates pre-60S rRNAs, but much less efficiently than the full-length protein, and only in the presence of up to
300 mm potassium acetate (Fig 2A) Interestingly, the A–C-Nop53p fusion (C-terminal portion of Nop53p) co-purifies 27S, 25S, 7S and 5.8S rRNAs more effi-ciently than the N-terminal portion of Nop53p
Fig 1 Nop53p immunoprecipitates 5.8S chromatin and interacts with RNA polymerase I A ChIP assay with A–Nop53p or protein was per-formed, followed by PCR reactions with primers for amplification of various regions of the rDNA and snoRNAs (A) PCR for amplification of 18S and 25S chromatin regions (B) Amplification of 5.8S region using samples from ChIP in the absence (upper panel) or presence (lower panel) of RNases A ⁄ T 1 (C) Amplification of snoRNA chromatin For (A)–(C), ‘Int’ represents the intergenic region of chromosome V, used as
an internal control; I, input; S, sheared; E, eluted (D) Quantification of the bands obtained in the PCR reactions Values represent the ratio of the rDNA bound to column to the input Bars represent standard deviation (E) Western blot for detection of proteins after pull-down assay Total extracts from cells expressing either GST or GST–Rrn3p (TE 1 ) were incubated with glutathione–Sepharose, the flow-through fraction was collected (FT1), and, after washing, total extracts of cells expressing His–Nop53p (TE2) were loaded The flow-through fraction was collected again (FT2), the resin was washed, and the bound fraction was obtained (B) His–Nop53p is pulled down by GST–Rrn3p, but not by GST His–Nop53p was detected using monoclonal antibody against His GST and GST–Rrn3p were detected using anti-GST serum Bands corresponding to full-length and breakdown products of His–Nop53p are indicated on the right The asterisks indicate a protein present in Escherichia coli extract that runs close to His–Nop53p.
Trang 4(Fig 2A) A–C-Nop53p co-precipitates 27S pre-rRNA
even in the presence of 500 mm potassium acetate,
indicating that the C-terminal portion of Nop53p is
stably bound to pre-rRNP complexes A western blot
of bound fractions from the same experiments showed
that protein A fusions bound efficiently to the columns
under all conditions used (Fig 2B), showing that the
differences in efficiency of rRNA precipitation between
truncated Nop53p mutants are due to different
stabili-ties of interaction with the pre-60S complex and not
inefficient binding to the column
In order to determine whether the Nop53p
trun-cated mutants also bind RNA directly, in vitro RNA
binding assays were performed In these experiments,
full-length Nop53p bound RNAs corresponding to
various fragments of pre-rRNA (Fig 3A) Although
Nop53p specifically co-immunprecipitates pre-60S
chromatin (Fig 1) and rRNAs (Fig 2) [31], Nop53p
did not show a clear sequence specificity for binding
in these in vitro RNA binding assays Interestingly,
however, all the rRNAs regions tested were AU-rich
and predicted to form secondary structures
N-Nop53p also bound RNA, although not as
effi-ciently as the full-length protein (Fig 3A) C-Nop53p,
on the other hand, did not bind RNA in vitro, show-ing the same result as the negative control GST (Fig 3A) We therefore conclude that Nop53p binds RNA through its N-terminal region and has affinity for AU-rich and structured RNAs In the pre-60S complex, Nop53p binding to rRNA might be more specific and stabilized by protein–protein interactions with its C-terminal portion
In order to analyze the affinity of Nop53p for AU-rich RNA sequences in more detail, in vitro RNA binding assays were performed using RNA oligonucleo-tides Full-length Nop53p bound poly-rU and poly-rAU oligomers, but not poly-rC (Fig 3B), corrob-orating the results described above In these exp-eriments, 5.8S rRNA was used as a positive control for Nop53p interaction In summary, although no sequence specificity was detected in these in vitro assays, Nop53p showed higher affinity for U- and AU-rich sequences
Truncated Nop53p mutants still localize to the nucleolus
Although Nop53p is a nucleolar protein, no nuclear localization signal could be detected in its sequence In
Fig 2 Truncated mutants of Nop53p still
associate with pre-60S (A)
Co-immunopre-cipitation of RNA with full-length or
trun-cated Nop53 Northern blot hybridization of
RNA co-immunoprecipitated with
pro-tein A–Nop53p, propro-tein A–N-Nop53p or
protein A–C-Nop53p Probes used were
specific against rRNAs or scR1 (internal
control) (B) Western blot of the proteins
obtained from the same experiments,
detected using anti-mouse IgG.
Trang 5order to identify the portion of Nop53p that is
respon-sible for its subcellular localization, the truncated
mutants of the protein were fused to a GFP tag
(GFP–N-Nop53p and GFP–C-Nop53p) Confocal
images of split fluorescence channels showed the same
pattern of localization for RFP–Nop1p and
GFP–N-Nop53p and GFP–C-GFP–N-Nop53p Interestingly, although
GFP–C-Nop53p is concentrated in the nucleolus, it
can also be visualized throughout the nucleus The
co-localization of GFP–Nop53p truncated mutants
and RFP–Nop1p was confirmed by fluorescence
profiles in several cell images (Fig 4A,B) These results
indicate that protein interactions might be responsible
for directing Nop53p to the nucleus and for its
concen-tration in the nucleolus
The N-terminal half of Nop53p complements
a conditional mutant strain
As shown above, the N-terminal portion of Nop53p binds 5.8S rRNA directly and is concentrated in the nucleolus, whereas the C-terminal portion of Nop53p might interact with the proteins in the pre-60S complex, but is less concentrated in the nucleolus These results raised the question of whether any of the truncated mutants of Nop53p, when under control of a constitu-tive promoter, could complement the growth of the con-ditional strain Dnop53⁄ GAL::A-NOP53 in glucose medium Interestingly, N-Nop53p partially comple-ments growth of the conditional strain (Fig 5A) When pre-rRNA processing was analyzed in these
Fig 3 RNA binding assay with truncated Nop53p mutants (A) Radioactively labeled
in vitro transcribed fragments of rRNA were incubated with 10 pmol of full-length Nop53p, or the truncated forms GST–N-Nop53p or GST–C-GST–N-Nop53p, or with GST RNA–protein complexes were analyzed by native gel electrophoresis and visualized by phosphorimaging Lanes 1, 6, 11 and 16, RNAs incubated with full-length Nop53p; lanes 2, 7, 12 and 17, RNAs incubated with GST–C-Nop53p; lanes 3, 8, 13 and 18, RNAs incubated with GST–N-Nop53p; lanes 4, 9,
14 and 19, RNAs incubated with GST; lanes
5, 10, 15 and 20, free RNA (B) Nop53p shows a preference for U-rich sequences Increasing amounts of Nop53p were incu-bated with various RNA oligonucleotides Free RNA and RNA–protein complexes (RNP) are indicated on the right No protein was added in lanes 1, 5, 9 and 13.
Trang 6transformants, it was possible to see that, although
27S pre-rRNA and 25S rRNA levels in the strains
Dnop53⁄ GAL::A-NOP53 expressing either N-Nop53p
or C-Nop53p were very similar to those of the control
strain Dnop53⁄ GAL::A-NOP53 ⁄ pGAD, expression of
N-Nop53p led to lower accumulation of the 7S
pre-rRNA intermediate (Fig 5B) The strain expressing
C-Nop53p showed higher levels of 7S pre-rRNA and
lower levels of the mature 5.8S rRNA (Fig 5B)
Quanti-fication of 7S⁄ 5.8S ratio in these strains showed that
N-Nop53p partially complements the function of the
Dnop53⁄ GAL::A-NOP53 strain (Fig 5C) These results
indicate that direct binding to 5.8S rRNA is important
for Nop53p function
Dnop53::GAL-NOP53 accumulates polyadenylated
forms of pre-rRNA
Nop53p affects pre-rRNA processing, and its depletion
leads to the accumulation of 27S and 7S pre-rRNAs,
which are degraded from the 5¢ end [31] Therefore, cells depleted of Nop53p show similar phenotypes to exosome mutants, indicating that unprocessed rRNA intermediates may accumulate in a polyadenylated form in the Dnop53::GAL-NOP53 strain, as demon-strated for Drrp6 mutants [35,36] In order to analyze the polyadenylation of rRNA processing intermediates
in the Dnop53::GAL-NOP53 strain, total RNA was extracted after 12 h of Nop53p depletion, and poly-A RNA was isolated using oligo(dT) cellulose columns Analysis of the purified poly-A RNA demonstrated that 27S and 7S pre-rRNAs accumulated in the poly-adenylated form in Dnop53::GAL-NOP53 (Fig 6A), indicating that these RNAs are not efficiently processed or degraded by the exosome in the absence
of Nop53p
In order to determine whether the effect of Nop53p depletion on 5.8S processing by the exosome was indi-rect, or whether it involved direct exosome binding, protein interaction experiments were performed We
Fig 4 Subcellular localization and protein interaction of the Nop53p N- and C-terminal fragments Yeast strain NOP53 expressing GFP–N-Nop53p and RFP–Nop1p (A) or GFP–C-GFP–N-Nop53p and RFP–Nop1p (B) was analyzed by laser scanning confocal microscopy Each channel labeling is shown separately and merged in the lower right panels The upper panels show representative profiles of green and red fluores-cence, indicating RFP–Nop1p (red line) and GFP–N-Nop53p (green line) or GFP–C-Nop53p (green line) co-localization.
Trang 7have previously tried to identify interactions between
Nop53p and the exosome subunits using the
two-hybrid system, but no positive interaction was detected
[31] Therefore, we tested the interaction between
Nop53p and some of the exosome subunits using GST
pull-down assays In these experiments, we detected a
specific interaction between the recombinant proteins
His–Nop53p and GST–Rrp6p (Fig 6B) Control
experiments with GST–Mtr3p showed that His–
Nop53p does not interact with this other exosome
sub-unit, nor does it interact with GST, which was used as
a negative control (Fig 6B) Despite the higher level of
GST expression compared to GST–Rrp6p or to GST–
Mtr3p, His–Nop53p was only pulled down by GST–
Rrp6p, confirming the specificity of this interaction
These results led to the conclusion that, by binding to
the 5.8S rRNA and through its interaction with
Rrp6p, Nop53p may direct the exosome to the 7S
intermediate for processing
The TRAMP complex has been shown to be
respon-sible for polyadenylation of the RNAs that are
sub-strates for degradation by the exosome [22] As
depletion of Nop53p leads to the accumulation of
polyadenylated pre-rRNAs, this raises the question of
whether Nop53p also interacts with TRAMP subunits
The TRAMP subunit Trf4p was therefore fused to GST, expressed in Escherichia coli, and its interaction with Nop53p tested through GST pull-down The results show that GST–Trf4p pulls down His–Nop53p, but the negative control GST does not (Fig 6C) These results indicate that Nop53p not only interacts with the exosome, but also with the TRAMP complex, corroborating the view that it is a regulatory factor for processing of 7S pre-rRNA
Nop53p activates the exosome RNase activity
in vitro
In order to test whether the Nop53p–Rrp6p interaction
is important for control of exosome function, in vitro RNA degradation assays were performed Yeast exosome was isolated by tandem affinity purification (TAP)–Rrp43p purification and was incubated with a substrate RNA for in vitro RNA degradation, in the presence of Nop53p or BSA, the latter being used as a negative control (Fig 7A) The results show that TAP–Rrp43p exosome degrades an in vitro transcribed RNA corresponding to a region of ITS2, a natural exosome substrate during rRNA maturation (Fig 7A; lane 2) Upon incubation of the substrate RNA with
Fig 5 Analysis of complementation of Dnop53 ⁄ GAL::A-NOP53 by truncated mutants of Nop53p (A) Analysis of comple-mentation of strain Dnop53 ⁄ GAL::A-NOP53
by truncated Nop53p mutants, under the control of a constitutive promoter, by growth in glucose medium N-Nop53p par-tially complements growth on glucose plates (B) rRNA processing was analyzed in yeast strains Dnop53 ⁄ GAL::A-NOP53 expressing either Nop53p, N-Nop53p or C-Nop53p by Northern blot hybridization with probes against rRNAs, indicated on the right (C) Quantification of the 7S ⁄ 5.8S rRNA ratio, showing the efficiency of 5.8S rRNA maturation in cells expressing the truncated forms of Nop53p.
Trang 8the TAP–Rrp43p complex, there is a 20% decrease in
the intensity of the substrate band and a corresponding
increase in the intensity of faster-migrating bands that
correspond to degradation products (Fig 7A; lane 2)
Although Nop53p does not degrade the RNA by itself
(Fig 7A; lane 10), addition of 10 pmol Nop53p to the
reaction containing the TAP–Rrp43p complex
increases the RNase activity of the exosome by 16%
(Fig 7A; lane 7) Addition of 20 and 30 pmol Nop53p further increased the RNase activity of the exosome by 27% and 42%, respectively (Fig 7A; lanes 8 and 9) Addition of BSA to the reaction had no effect (Fig 7A; lanes 2–6) Control experiments with TAP– Nop58p-purified box C⁄ D snoRNP complex showed
no degradation of the RNA, as expected (Fig 7A; lanes 11–16)
Fig 6 Analysis of rRNA polyadenylation in
the strain Dnop53 ⁄ GAL::A-NOP53 and
inter-action with the exosome (A) Total RNA
was isolated from strains NOP53 and
Dnop53 ⁄ GAL::A-NOP53, and run through
oli-go(dT)–Sepharose columns Polyadenylated
RNAs were analyzed by Northern blot
hybridization against probes specific for
rRNAs I, input; FT, flow-through; EL, eluted
polyadenylated RNA (B, C) Western blot for
detection of proteins after pull-down assay.
(B) Total extract from Escherichia coli cells
expressing either GST, GST–Rrp6p or GST–
Mtr3p (TE 1 ) was incubated with
glutathi-one–Sepharose resin, the flow-through
frac-tion was collected (FT1), and, after washing,
the total extract of cells expressing His–
Nop53p (TE 2 ) was loaded The flow-through
fraction was collected again (FT2), the resin
was washed (not shown), and bound
frac-tion was obtained (B) His–Nop53p is pulled
down by GST–Rrp6p His–Nop53p was
detected using antibody against His GST,
GST–Rrp6p and GST–Mtr3p were detected
using anti-GST serum Bands corresponding
to full-length and breakdown products of
fusion proteins are indicated on the right.
(C) Same procedure as in (B), but with total
extract from E coli cells expressing either
GST or GST–Trf4p (TE 1 ), or expressing His–
Nop53p (TE 2 ) His–Nop53p is pulled down
by GST–Trf4p.
Trang 9The recovery of TAP-purified complexes was
analyzed by the detection of other subunits of the
exosome and box C⁄ D snoRNP by western blotting
(Fig 7B) TAP–Nop58p co-purified endogenous
Nop1p and TAP–Rrp43p co-purified Mtr3p, indicating
that the box C⁄ D snoRNP and exosome complexes,
respectively, were recovered In addition, incubation of
TAP complexes with His–Nop53p from E coli extracts
showed that His–Nop53p is recovered with TAP–
Rrp43p, further confirming the interaction of Nop53p
the role that Nop53p plays in rRNA processing in yeast It had previously been demonstrated that Nop53p binds 5.8S rRNA and participates in the late steps of maturation of the large ribosomal subunit RNAs [31–33], and here we show that the role played
by Nop53p involves protein–protein and protein–RNA interactions Nop53p co-precipitates 5.8S and 25S chromatin and, to a lower extent, 18S chromatin, which indicates that it binds pre-rRNA co-transcrip-tionally Nop53p recruitment to rDNA chromatin is dependent on active transcription, as no precipitation
of chromatin above background level was obtained with A-Nop53p after treatment with RNases We also show here that Nop53p interacts directly with RNA polymerase I transcription factor Rrn3p [34] These data indicate that, although Nop53p is present in the pre-60S complex [1,11] and affects 7S pre-rRNA processing by the exosome [31], it binds 5.8S rRNA co-transcriptionally Other protein complexes have been shown to interact with transcription factors and also influence pre-rRNA processing, including the CURI complex formed by CK2, Utp21, Rrp7p and Ifh1p, which is proposed to couple rRNA and ribo-somal protein transcription [37] Some of the U3 snoRNP protein subunits (Utp) have also been shown
to bind rRNA early during transcription and to partic-ipate in rRNA processing [4,5,7] SSU processome factors, mainly involved in processing of the 18S rRNA, bind the precursor rRNA co-transcriptionally [4] Later during processing, factors involved in the maturation of 27S pre-rRNA assemble onto the RNA, forming the large ribosomal subunit (LSU) complex [8,38] Nop53p may participate in formation of the LSU knob, and as it is present in the gradient frac-tions that contain LSU pre-rRNAs, it may remain bound to the 5.8S rRNA during its processing [32] The nucleolar localization of Nop53p seems to be the result of protein–protein interactions, as no nuclear localization signal could be identified in the Nop53p sequence and truncated versions of this protein still localize to the nucleolus We have shown that Nop53p interacts with various nuclear proteins – Nop17p,
Fig 7 Effect of Nop53p on RNA degradation by the exosome.
In vitro RNA degradation assay to test the effect of Nop53p on
exosome RNase activity (A) A radioactively labeled RNA oligo
cor-responding to the 5¢ region of the rRNA spacer ITS2 was incubated
with 100 ng of the exosome complex isolated using TAP–Rrp43p,
or with 100 ng of box C ⁄ D snoRNP isolated using TAP–Nop58p,
and 10, 20 or 30 pmol of His–Nop53p or BSA Reaction mixtures
were incubated for 1 h at 30 C, and the products were analyzed
by denaturing acrylamide gel electrophoresis The main degradation
products generated by the exosome complex are indicated (B)
Analysis of protein complexes recovered through TAP purification.
TAP–Nop58p co-purified Nop1p and TAP–Rrp43p co-purified Mtr3p,
indicating that the box C ⁄ D snoRNP and exosome complexes,
respectively, were intact Both complexes co-purified His–Nop53p
in the pull-down assay, although Nop53p interaction with the
exosome was much stronger.
Trang 10Nip7p, Rrn3p, Rrp6p and Trf4p (Figs 1 and 6) [31].
Identification of these protein interactions indicates
that one of these factors, or the whole complex, might
be responsible for directing Nop53p to the nucleolus
A recent example of an rRNA processing factor that is
dependent on protein interaction for its subcellular
localization is human hRrp47p, an exosome cofactor,
which was shown to depend on its interaction with the
exosome subunit PM⁄ Scl_100 (an Rrp6p ortholog) for
direction to the nucleus [39] Interestingly, the Nop53p
C-terminal region co-immunoprecipitates the pre-60S
complex more efficiently than the N-terminal portion
of the protein The Nop53p N-terminal region, on the
other hand, is involved in RNA interaction and can
partially complement the conditional strain Dnop53⁄
GAL::NOP53 in glucose medium These results
indi-cate that interaction with RNA is responsible for
Nop53p molecular function in 27S and 7S pre-rRNA
processing, and that this interaction may be stabilized
in the pre-60S complex through protein interaction
with the C-terminal portion of Nop53p Similarly, the
ribosomal protein Rpl25p affects processing of 27S
pre-rRNA and has three functional domains for
nuclear import, RNA binding and 60S subunit
assem-bly [40] Mutations of each of these domains result in
defective ITS2 processing and accumulation of
pre-rRNA 27S, indicating that assembly of Rpl25p is
necessary but not sufficient for processing [40] Despite
not having a canonical RNA-binding motif, Nop53p
binds RNA, but does not show strict RNA sequence
specificity in in vitro RNA binding experiments
Simi-larly, Nop9p, another example of an RNA-binding
protein involved in pre-rRNA processing, associates
with 20S pre-rRNA but does not show sequence
speci-ficity for in vitro binding [41]
Depletion of Nop53p leads to accumulation of the
7S pre-rRNA and polyadenylated RNAs, a
pheno-type very similar to that resulting from depletion of
exosome subunits The results shown here indicate
that Nop53p function is directly related to the
inter-action with 5.8S rRNA and the exosome in the
pre-60S complex In this context, Nop53p could
be responsible for directing the exosome to 7S
pre-rRNA, thereby regulating the function of the
complex In the absence of Nop53p, the exosome is
not efficiently directed to the 7S pre-rRNA for
pro-cessing, leading to the accumulation of its
polyadeny-lated form As RNAs polyadenypolyadeny-lated by the TRAMP
complex are targeted for degradation by the exosome
[22], polyadenylated 7S was expected to be degraded
in strain Dnop53 ⁄ GAL::A-NOP53 However,
poly-adenylated 7S pre-RNA accumulates in this strain
and appears to be degraded in the 5¢ fi 3¢ direction
[31], leading to the conclusion that Nop53p is an exosome cofactor
Accordingly, in vitro RNA degradation assays with the exosome complex isolated using TAP–Rrp43p showed that, although Nop53p does not degrade RNA
by itself, its presence stimulates the RNase activity of the exosome It is possible that the stimulation of the exosome activity is due to recruitment of the complex
to the substrate via Nop53p–Rrp6p interaction, or through TRAMP recruitment Interestingly, Nop53p has also been identified as interacting with components
of the TRAMP complex [42], corroborating the results shown here A similar role is seen for another RNA-binding protein, of the Nrd1 complex, which can direct the exosome to specific RNA substrates and stimulate exosome degradation of substrates [43] We can conclude that Nop53p must play an important role in exosome activity
In summary, we show here that Nop53p binds 5.8S rRNA co-transcriptionally through its N-terminal por-tion and may interact with other pre-60S processing factors through its C-terminal portion As depletion of Nop53p leads to accumulation of polyadenylated 7S pre-rRNA, and as Nop53p interacts with the exosome subunit Rrp6p and activates the RNase activity of the complex in vitro, Nop53p may be involved in recruit-ment of the exosome to the 7S pre-rRNA for proce-ssing and formation of the mature 5.8S rRNA
Experimental procedures
Plasmid construction
The plasmids used in this study are listed in Table 1 and cloning procedures are summarized below DNA fragments
of NOP53 coding for the N-terminal (amino acids 1–210) and C-terminal (amino acids 210–456) portions of the pro-tein were PCR-amplified from Saccharomyces cerevisiae genomic DNA and cloned into vectors pBTM or pGADC2 for two-hybrid analyses and into YCplac33GAL-A, fused
to the protein A tag, under the control of the GAL1 promoter [31] Subsequently, the fragments coding for the N- and C-terminal regions of NOP53 were subcloned into pGEX (GE Healthcare, Piscataway, NJ, USA), using the restriction sites BamHI⁄ SalI and EcoRI ⁄ PstI, respectively, generating vectors pGEX-N-NOP53 and pGEX-C-NOP53 Plasmids pGAD-N-NOP53 and pGAD-C-NOP53, contain-ing NOP53 truncation mutants under the control of the constitutive ADH1 promoter, were also used for comple-mentation analysis of conditional strain Dnop53⁄ GAL:: NOP53 The plasmids pYCplac33GAL-A-NOP53, pET28-NOP53, pGADC2-NOP53 and pGEM-5.8S have been described previously [31]