Proc. Natl. Acad. Sci. USA
Vol. 94, pp. 13536–13541, December 1997
Functional separation of pre-rRNA processing steps revealed by
truncation of the U3 small nucleolar ribonucleoprotein
SARAH J. LEE* AND SUSAN J. BASERGA†‡
Departments of *Molecular Biophysics and Biochemistry and†Therapeutic Radiology and Genetics, Yale School of Medicine, New Haven, CT 06520-8040
Communicated by Sidney Altman, Yale University, New Haven, CT, October 8, 1997 (received for review August 8, 1997)
(snoRNP) is required for three cleavage events that generate
the mature 18S rRNA from the pre-rRNA. In Saccharomyces
cerevisiae, depletion of Mpp10, a U3 snoRNP-specific protein,
halts 18S rRNA production and impairs cleavage at the three
U3 snoRNP-dependent sites: A0, A1, and A2. We have iden-
tified truncation mutations of Mpp10 that affect 18S rRNA
synthesis and confer cold-sensitivity and slow growth. How-
ever, distinct from yeast cells depleted of Mpp10, the mutants
carrying these truncated Mpp10 proteins accumulate a novel
precursor, resulting from cleavage at only A0. The Mpp10
truncations do not alter association of Mpp10 with the U3
snoRNA, nor do they affect snoRNA or protein stability. Thus,
the role in processing of the U3 snoRNP can be separated into
cleavage at the A0 site, which occurs in the presence of
truncated Mpp10, and cleavage at the A1?A2 sites, which
occurs only with intact Mpp10. These results strongly argue
for a role for Mpp10 in processing at the A1?A2 sites.
The U3 small nucleolar ribonucleoprotein
Four rRNAs and many proteins comprise the eukaryotic
ribosome: the 60S subunit contains 25S (Saccharomyces cer-
evisiae), 5.8S, and 5S rRNAs, and the 40S subunit houses 18S
rRNA. In the nucleolus RNA polymerase I transcribes long,
25S, 5.8S, and 18S rRNAs. A schematic of pre-rRNA process-
ing (Fig. 1) shows that the rRNA sequences are flanked by two
external transcribed spacers (5? and 3?) and separated by two
internal transcribed spacers (ITS1 and ITS2). Complexes of a
small nucleolar RNA and proteins, called snoRNPs, and other
nucleolar proteins process this original transcript into the
mature rRNAs. The depletion or mutation of many of these
processing molecules disrupts pre-rRNA processing (1, 2).
Many small nucleolar RNAs (snoRNAs) have been identi-
fied but U3 is the most abundant in vertebrate cells and exists
in a wide evolutionary range of species (from trypanosomes to
humans), making it an attractive molecule for studies aimed at
elucidation of pre-rRNA processing. In vivo and?or in vitro
studies in three different systems, mouse, Xenopus laevis
oocytes, and S. cerevisiae, have demonstrated that it is as an
essential component for production of 18S rRNA (3–9).
In yeast, processing of the pre-18S rRNA occurs before the
cleavage steps that release the mature 5.8S and 25S rRNAs
(Fig. 1). The U3 snoRNA has been shown to be necessary for
the early processing events that begin at the A0 site and then
proceed to the A1 and A2 sites (4, 6, 7, 9). However, other
small nucleolar ribonucleoprotein (snoRNPs), including U14
and snR30, are required for formation of 18S rRNA (10, 11).
Thus, pre-rRNA processing may be mediated by a multi-
snoRNP ‘‘processome,’’ similar to the spliceosome that carries
out splicing of pre-mRNAs (12).
Two U3 snoRNA-pre-rRNA base pairing interactions have
been identified in yeast (6, 7, 9, 13). One of these interactions,
between the U3 snoRNA and a site within the 5? external
transcribed spacers (?150 nucleotides upstream of the A0
cleavage site), is essential for pre-18S rRNA processing. Base
complementarity in this interaction is essential as mutations in
the 5? external transcribed spacer that disrupt the U3-pre-
rRNA association impair 18S rRNA production. Processing
can be restored with these by making the compensatory
mutations in the U3 snoRNA. The sizable distance of this
necessary, base paired region from any of the cleavage sites led
to a model characterizing the U3 snoRNA as an RNA chap-
erone. Instead of catalyzing the cleavage, U3 snoRNA may
help fold the pre-rRNA and expose cleavage sites to the
enzymatically active molecule. Indeed, Rnt1, the yeast ho-
molog of Escherichia coli RNase III (14), is sufficient for
cleavage at this site in vitro and cleaves in vivo at this site. The
components responsible for cleavage at A1 and A2 have not
yet been identified.
Although several U3-specific proteins have been identified
in vertebrate cells through immunoprecipitation experiments,
only one has been purified and none of their genes have been
cloned (15, 16). The genes for two proteins associated with the
yeast U3 snoRNP, however, have been cloned: NOP1 (17–19),
the yeast homolog of the vertebrate fibrillarin protein, and
SOF1 (20). Unlike Nop1, a vertebrate homolog for Sof1
remains unknown. Both proteins are required for yeast via-
bility and production of 18S rRNA (21).
Nop1 is a protein component common to a number of
snoRNAs (22), so the phenotypes resulting from mutations in
NOP1 may not reflect a function associated with U3. Sof1, on
the other hand, is a U3-specific protein that was identified in
yeast as an extragenic suppressor of a temperature-sensitive
fibrillarin mutant (20). Cells lacking Sof1 mimic the phenotype
observed in cells depleted of U3, with impaired cleavage of
pre-18S rRNA at the three U3-dependent sites but normal
production of 5.8S and 25S rRNAs. Sof1, it appears, plays a
necessary role in U3 snoRNP function. Without more detailed
studies, however, one cannot acquire further insight into
whether Sof1 affects U3 snoRNP assembly and?or the role of
the snoRNP in cleavage of the three sites.
Recent studies from our laboratory have characterized a
novel protein in yeast and vertebrates, Mpp10, that associates
specifically with the U3 snoRNA (J. M. Westendorf, K. N.
Konstantinov, S. Wormsley, M. D. Shu, N. Matsumoto-Taniura,
F. Pirollet, F. G. Klier, L. Gerace and S.J.B., unpublished
results) (23). MPP10 is essential for cell survival in yeast and
The publication costs of this article were defrayed in part by page charge
payment. This article must therefore be hereby marked ‘‘advertisement’’ in
accordance with 18 U.S.C. §1734 solely to indicate this fact.
© 1997 by The National Academy of Sciences 0027-8424?97?9413536-6$2.00?0
PNAS is available online at http:??www.pnas.org.
Abbreviations: snoRNA, small nucleolar RNA; snoRNP, small nucle-
‡To whom reprint requests should be addressed at: Therapeutic
Radiology and Genetics, Yale School of Medicine, HRT 317, P.O.
Box 208040, New Haven, CT 06520-8040. e-mail: susan.baserga@
anti-Mpp10 antibodies specifically immunoprecipitate the U3
snoRNA from yeast cells (23). Depletion of Mpp10 has no
effect on 25S rRNA production but halts 18S rRNA produc-
tion by inhibiting processing at the three U3-dependent sites:
A0, A1, and A2.
To better understand its role in pre-rRNA processing, we
have made selected mutations in the Mpp10 protein and
assayed their functional consequences in yeast. We report here
that truncation of only the C terminus of Mpp10 causes the
cells to become cold-sensitive, whereas truncation at both ends
imparts a slow growth phenotype at all temperatures. Surpris-
ingly, the slow growth results from impaired processing at only
the A1 and A2 sites in the pre-rRNA, distinct from the
deficiency in pre-rRNA processing that we observe upon
depletion of Mpp10. Neither of the truncations affect associ-
ation of Mpp10 with the U3 snoRNA, nor is U3 snoRNA or
protein stability affected. Thus, in characterizing the truncated
proteins, we have separated the roles of the U3 snoRNP into
cleavage at A0 and cleavage at A1?A2. These results allow
MATERIALS AND METHODS
Microbiological Medium and Yeast Manipulation. S. cer-
evisiae were grown and transformed as described in Dunbar et
al. (23). All media used for cell growth, once strains were
constructed, was yeast?peptone?dextrose (yeast extract 1%,
peptone 2%, glucose 2%).
Yeast Strains. All constructs were transformed into the
haploid yeast strain mpp10::HIS3 pGAL1::MPP10. The con-
struction of this strain and its genotype is described in Dunbar
et al. (23).
Cloning of the Truncation Mutations. Oligonucleotides
targeted to the desired ends for each of the truncation
mutations were used to perform PCR using a full-length
MPP10 gene in pET28a as a template. mpp10–1 was generated
using ympp10.3 (5?-CCGCGGATCCATGTCAGAACTCTT-
TGGAGTATTGAAATC-3?) and ympp10.9R (5?-CCCGGA-
GCTCTCAGACATTGTATATCTCTTGAGG-3?) for the 5?
end and the 3? end of the mutant gene, respectively. mpp10–2
was generated using ympp10.10 (5?-CCGCGGATCCATGG-
CAGAACTGGACGAAATC-3?) and ympp10.9R. mpp10–3
required use of ympp10.10 and ympp10.16 (5?-CCCGCTCG-
AGTCAAAGTTTTCTATTTGTGCT-3?). The fragment
corresponding to the terminus, necessary to construct
mpp10–4, was generated using ympp10.15 (5?-CCCGCTCG-
AGTCATGTCGAATGCCTCT-3?) and ympp10.16.
Reactions were carried out under standard conditions using
reagents supplied by Perkin–Elmer on a GeneAmp PCR 2400
machine for 20 rounds of amplification using Taq DNA
polymerase. Each cycle consisted of 30 sec at 94°C, 30 sec at
55°C, and 30 sec at 72°C.
All PCR fragments were purified using the QIAquick PCR
Purification Kit (Qiagen, Chatsworth, CA). The fragments for
all the mutants except mpp10–4 were digested with BamHI
and AvaI, resolved on a 0.8% gel, and purified using Gene-
Clean (Bio101, Vista, CA). Digested and purified fragments
were ligated into p415GPD (AmpR, LEU2, ARS?CEN) at the
BamHI and XhoI sites (24). The C-terminus fragment was
ligated into the XhoI site of mpp10–2 to make mpp10–4. The
fragment was prepared in a similar manner to the other
fragments but digested only with XhoI and purified on a 2.0%
Plasmids were extracted from E. coli using QIAprep Spin
Miniprep Kits (Qiagen) or Jetstar kits for large scale prepa-
rations. Construct identity was verified by restriction digest
(BamHI and XhoI) and sequencing of the 5? and 3? ends of
each insert (automated on an Applied Biosystems 373 Stretch
sequencer) using oligos ympp10.13 (5?-GCTACCATCAGA-
GAGATTGTGAGAAGGC-3?) and ympp10.14 (5?-GAGCT-
TCAAAAGGCACATTCCG-3?), respectively. Restriction
mapping using ClaI and BglII confirmed the correct orienta-
tion of the C terminus fragment in mpp10–4. All reagents for
digests were obtained from New England Biolabs and reac-
tions were carried out according to their specifications.
Growth Curves. Yeast were grown in yeast?peptone?
dextrose at 30°C until OD600 0.1–0.9. Cells were diluted to
OD6000.050 in yeast?peptone?dextrose and regrown at 30°C.
Optical density at 600 nm was measured on a Beckman DU-64
Spectrophotometer at the indicated time points.
Plasmid Retrieval from Yeast. Plasmids were harvested
from yeast using the method shown in ref. 25. To verify the
presence of the truncated Mpp10 genes in our mutant strains,
the plasmids were restriction mapped and the ends were
sequenced on an Applied Biosystems DNA Sequencer at the
William Keck facility at Yale.
Western Blot Analysis. Extracts were made from yeast
grown in yeast?peptone?dextrose at 30°C to an OD600 of
0.6–0.8. Cells were washed with water, resuspended in NET-2
(20 mM Tris?Cl, pH 7.5?150 mM NaCl?0.05% Nonidet P-40)
with protease inhibitors and lysed by vortexing (6 ? 45 sec)
with 0.45–0.5 mm glass beads. The lysate was cleared by
centrifugation for 5 min at 13,200 rpm at room temperature.
Total protein (2.5 mg) was separated on 10% SDS?PAGE and
transferred onto a nitrocellulose membrane. Expression of
rRNA transcript undergoes successive cleavages at the A0, A1, and A2
sites, dividing it into precursors to the small and large subunit rRNAs.
The 20S is processed into the small subunit 18S rRNA. The 27SA2
precursor gives rise to the large subunit rRNAs, 5.8S and 25S. The
large subunit RNAs can also be processed from the 27SB precursor,
produced by either exonucleolytic degradation to B1 or by direct
cleavage at B1. The 27SB precursor undergoes further processing steps
to form the mature large subunit rRNAs. (B) Pre-rRNA processing in
S. cerevisiae expressing truncated Mpp10 proteins. The levels of rRNA
precursors that are generated by cleaving at A1 and A2, the 32S, 20S,
and 27SA2 pre-rRNAs, are reduced, whereas the levels of 33S and 23S
precursors are similar. A new 22S precursor is visible, resulting from
competent cleavage at A0 and deficient cleavage at A1 and A2. It is
not known if the 22S precursor is degraded or is slowly processed to
form the 18S rRNA.
Pre-rRNA processing in S. cerevisiae. (A) The nascent 35S
Biochemistry: Lee and BasergaProc. Natl. Acad. Sci. USA 94 (1997)13537
Mpp10 in each of the strains was tested by Western blot
analysis (enhanced chemluminescence, Amersham) using the
anti-yeast Mpp10 antibody characterized in Dunbar et al. (23).
Immunoprecipitations. 2.5 mg of protein A-Sepharose
CL-4B (Pharmacia) beads were complexed with anti-yeast
Mpp10 (50 ?l of rabbit serum) in 0.5 ml of NET-2 (20 mM
Tris?Cl, pH 7.5?150 mM NaCl?0.05% Nonidet P-40) by nu-
tating overnight at 4°C. Protein A-Sepharose CL-4B was also
prepared without antibody for a control (mock). The beads
were washed three times with 1 ml NET-2.
Yeast extracts were prepared by the same methods as those
used for the Western blot assay except the cells were harvested
at OD600of 5. Extract for 20 OD600units of cells was complexed
with the antibody-bound beads for 1 hr at 4°C. The pellet was
washed eight times with 1 ml NET-2. RNA was recovered by
PCA acid extraction, ethanol precipitated, and resolved on an
8% denaturing polyacrylamide gel. RNA was transferred to a
Zeta-Probe membrane (Bio-Rad) and analyzed by Northern
blot with32P-?UTP labeled anti-sense U3 RNA according to
Dunbar et al. (23).
Analysis of Pre-rRNA Processing. Total RNA preparation,
gel conditions, and blotting procedure, were carried out as in
in the text. Equal amounts of total RNA were loaded in each
lane. Blots were probed with several oligonucleotides de-
scribed in Berges et al. (26). Blots were also probed with oligo
z (called oligo c in ref. 27).
Cloning of Mpp10 Truncation Mutations. Although there is
only 30% identity between the yeast and human MPP10 genes,
the distribution of charged residues is highly conserved, indi-
cating potential functional significance. We based our yeast
MPP10 truncations on these charged regions. Truncated gene
fragments of MPP10 were generated by PCR from an intact
gene with oligonucleotides directed at the designated trunca-
tion sites. All of the Mpp10 gene mutations and the full-length
Mpp10 were cloned into a yeast expression vector, p415GPD.
We constructed the following genes: mpp10–1, a C-terminal
truncation of 109 amino acids; mpp10–3, an N-terminal trun-
cation of 46 amino acids; and mpp10–2, carrying both trun-
cations. The identity of the mutants was verified by restriction
digests and DNA sequencing of the ends from plasmids
rescued from yeast. Sequence data revealed several identical
amino acid substitutions in the C terminus of mpp10–1 and
mpp10–2, both carried in the oligonucleotide used to make the
C-terminal truncation. The last five amino acids in these
sequences in the wild-type protein are Gln-Pro-Leu-Tyr-Met.
To test for the possibility that these substitutions contribute to
the observed phenotypes, we cloned a fourth construct,
mpp10–4. We constructed this gene by reintroducing the
truncated C-terminal sequence onto mpp10–2 rescued from
yeast. This mutant represents an N-terminal truncation inclu-
sive of these mutations.
Truncation of Mpp10 Affects Cell Growth. Strains were
constructed by shuffling the plasmids into the mpp10::HIS3
pGAL1::MPP10 strain (described in ref. 23). Following selec-
tion against the pGAL::MPP10 plasmid on 5-FOA, these new
strains, carrying only full-length or truncated MPP10 genes on
the p415GPD plasmid, were streaked onto rich media and
grown at 18°C, 22°C, 30°C, and 37°C to test for cold and
temperature-sensitivity. The results are shown in Fig. 2. Nei-
ther mpp10–3 nor mpp10–4 strains exhibited any cold-
sensitive phenotypes and none of the mutants were tempera-
ture sensitive (data not shown). Thus, the amino acid substi-
tutions introduced by PCR do not produce a detectable growth
phenotype. In contrast, the mpp10–1 strain grows normally at
30°C but shows inhibited growth at 22°C. The mpp10–2 strain,
the double truncation, confers slow growth even at 30°C and
37°C, and almost no growth at lower temperatures. Therefore
and truncation of both ends causes slow growth even at 30°C.
To quantify growth differences between MPP10, mpp10–1,
and mpp10–2, we recorded growth curves for each strain at
30°C (data not shown). As expected, the strains with MPP10
and mpp10–1 grew at similar rates and the mpp10–2 strain
grew at a retarded rate but eventually reached stationary
phase. The doubling times were 2, 2.2, and 4 hr for the MPP10,
mpp10–1, and mpp10–2 strains, respectively.
To investigate the nature of the growth defect conferred by
our truncated proteins, we examined protein levels, U3
snoRNP association, and snoRNA stability, and pre-rRNA
processing in our yeast strains.
Truncation of Mpp10 Does Not Affect Levels of Expression.
The expression and stability of the truncated proteins was
in lane 1, runs anomalously at 110 kDa, even though it has a
predicted molecular mass of 67 kDa. This has been observed
with other nucleolar proteins that, like Mpp10, also have
distinct regions of positive and negative charges (28, 29), and
is observed when the Mpp10 protein is expressed in E. coli
(23). A band of ?85 kDa, corresponding to the truncated
proteins, is present in both lanes 2 and 3. No size difference
between the two proteins is resolvable, even on a lower
percentage gel (data not shown). This is not unexpected since
even wild-type Mpp10 exhibits anomalous migration. Verifi-
cation that the yeast strains contain the correct Mpp10 trun-
and slow growth of yeast. The Mpp10 protein is represented by ?. All
four strains bearing Mpp10 truncations, indicated by I, were streaked
Mpp10 protein. Growth was compared at 30°C and 22°C. For the
N-terminal truncation, the first 46 amino acids were deleted, and an
initiating methionine was added at amino acid 47. For the C-terminal
truncation, the last 109 amino acids (amino acids 485–593) were
deleted. The asterisk (?) in mpp10–1, mpp10–2, and mpp10–4 indicate
the amino acid substitutions that arose from the oligo used for PCR
(refer to Materials and Methods).
Truncations of the Mpp10 protein result in cold-sensitivity
13538 Biochemistry: Lee and BasergaProc. Natl. Acad. Sci. USA 94 (1997)
cations was accomplished by plasmid rescue from the yeast
followed by restriction digests and DNA sequencing of the
ends of the plasmid inserts (data not shown).
Truncation of Mpp10 Does Not Affect U3 snoRNA Associ-
ation. To investigate if the truncated Mpp10 proteins were
deficient in association with the U3 snoRNA, immunoprecipi-
tations were performed on yeast cell extracts with anti-Mpp10
antibodies. RNA was recovered, from both pellets and total
extracts, and resolved on a polyacrylamide gel. For the total
extract lanes, we isolated RNA from one-tenth of the volume
of extract used for the immunoprecipitation to avoid smearing.
Isolated RNA was analyzed via Northern blot by probing with
proteins associate with the U3 snoRNA to the same extent as
wild-type (lanes 5–7). A single band is present in the total
extract lanes of all three strains, showing that U3 snoRNA
stability is unaffected by truncation of Mpp10 (lanes 1–3). No
U3 snoRNA is immunoprecipitable when beads alone are used
(lane 4). Therefore the growth defect that we observe in the
mpp10–2 strain is not due to a lack of association with the U3
snoRNA nor due to an unstable U3 snoRNA.
Truncation of Mpp10 Affects Processing of Pre-18S rRNA
Precursors. The effects of the truncated Mpp10 proteins on
pre-rRNA processing was investigated. Total RNA was iso-
? 1), resolved on an agarose-formaldehyde gel and analyzed
by Northern blot with oligonucleotides that differentiate
among mature and precursor rRNAs. The first panel of Fig. 5A
levels of the mature products remain consistent among all
three strains (lanes 1–3) when the yeast are grown this way
prior to RNA isolation. Refer to Fig. 1 for processing inter-
mediates. The 32S rRNA, resulting from cleavage at A1, is
detectable in the wild-type (lane 1), but longer exposures of
this panel show that in relation to wild type, levels of 32S are
slightly decreased in mpp10–1, and remarkably reduced in
mpp10–2 (data not shown). Equal levels of the 33S precursor,
across all three strains, is also observable at longer exposures
of this panel (data not shown).
Probing with an oligonucleotide that recognizes 18S rRNA
precursors demonstrates a deficiency in processing between
the two mutants and wild-type (lanes 4–6). Levels of 33S and
23S are consistent among the different strains, but we see a
gradual decrease in the accumulation of 20S, the immediate
precursor to 18S rRNA, in the mpp10–1 and mpp10–2 strains.
Interestingly, the mpp10–1 strain seems to accumulate an
intermediate level of 20S, between the levels of wild-type and
the mpp10–2 strain. Also significant is the over-accumulation
of 35S precursor in the mpp10–2 strain (lane 6) but not in the
mpp10–1 strain (lanes 5 and 8). Most importantly, a new 22S
in wild-type (lane 4). This precursor appears in yeast that are
deficient in processing at the A1 and A2 sites and is produced
by cleavages at the A0 and A3 sites (9, 27). To confirm its
identity in our experiments, we reprobed our blots with an
oligonucleotide that hybridizes to sequences between the A0
and A1 sites (lanes 7–9). These results clearly indicate that this
aberrant precursor is not present in the wild-type (lane 10), but
appears in the mutants (lanes 11 and 12). A third oligonucle-
otide was used to probe for the presence of a precursor to the
large subunit RNAs that normally results from cleavage at the
A2 site, called 27SA2. The results (lanes 10–12) indicate that
the 27SA2 precursor is present in the strain with wild-type
Mpp10, but decreased in the two mutant strains. This confirms
that cleavage at A2 is impaired in the two mutant strains.
Taken together, the results in Fig. 5 suggest that truncation of
Mpp10 causes a deficiency in processing primarily at sites A1
and A2 in the pre-rRNA.
We were surprised to detect equivalent levels of 18S and 25S
rRNA in our mutant strains when they also exhibit a clear
defect in processing of pre-18S rRNA precursors. However, we
learned that when we grow the yeast cells, seeded equally, for
the same amount of time (10 hr), we observe a graded decrease
in 18S levels from wild-type to the mpp10–1 and mpp10–2
mutant strains when compared with 25S rRNA levels (Fig. 5B,
lanes 1–3). The effects of the truncation mutations on the
pre-rRNA precursors that we observed in Fig. 5B are also
visible in Northern blots of yeast grown this way, though they
are not as pronounced (data not shown). Growth to equal
density prior to RNA analysis, then, aids in determining the
nature of the processing defect in cells expressing the Mpp10
We have made mutations in the Mpp10 protein, a U3 snoRNP
component, to better understand its role in pre-rRNA pro-
cold-sensitivity and slow growth. The slow growth is due to a
deficiency in processing at only 2 of the 3 U3 snoRNP-
dependent sites in the pre-rRNA, visualized by the presence of
a new 22S precursor. This is distinct from the deficiency in
pre-rRNA processing that we have observed in yeast cells
depleted of Mpp10 (23). This processing defect is neither due
to instability of the truncated Mpp10 or U3 snoRNA, nor to
expression levels similar to that of full-length Mpp10. Western blot
analysis of yeast extracts was performed with anti-Mpp10 rabbit serum
diluted 1:10,000. Proteins were resolved on 10% SDS?PAGE. The
lower, fainter bands in the mpp10–1 lane are degradation products
particular to this batch of extract.
Truncated Mpp10 proteins are stable in yeast and have
ation or U3 snoRNA stability. Lanes 1–3 represent RNA isolated from
total yeast extract. Lanes 4–7 represent RNA that was isolated from
the pellet after immunoprecipitation of yeast extracts with protein
A-Sepharose CL 4B beads alone (‘‘mock,’’ lane 4) or anti-Mpp10
rabbit serum (lanes 5–7). For lanes 1–3, RNA was isolated from 1?10
the volume of extract used for the immunoprecipitations. RNA was
resolved on an 8% denaturing polyacrylamide gel. Northern blots were
probed with an anti-sense U3 snoRNA.
Truncations of Mpp10 do not affect U3 snoRNA associ-
Biochemistry: Lee and BasergaProc. Natl. Acad. Sci. USA 94 (1997) 13539
a lack of association of the truncated proteins with the U3
snoRNA. Thus, our results suggests that an important function
of Mpp10 is to enable cleavage at the A1?A2 sites, and these
experiments demonstrate a novel separation of the functions
of the U3 snoRNP into cleavage at A0 and cleavages at A1 and
A2 (Fig. 1).
In yeast, depletion of Sof1, Mpp10, or the RNA component
of the U3 snoRNP causes a deficiency in pre-rRNA processing
at the A0, A1, and A2 sites in the pre-rRNA. This is indicated
by an increase in the 23S and 35S pre-rRNA precursors, and
a decrease in the 27SA2 precursor on Northern blots following
depletion of any of these components individually (4, 20, 23).
This is consistent with the role of the U3 snoRNP in pre-rRNA
processing at these three sites. Since the yeast homolog of the
E. coli Rnase III protein, Rnt1, cuts at the A0 site in vitro and
in vivo (14), the U3 snoRNP may be acting as a chaperone on
the pre-rRNA for this reaction.
Although U3 snoRNA depletion causes a deficiency in
cleavage at all three sites, specific mutations cause a deficiency
at the A1?A2, but not A0, cleavage sites. John Hughes
observed the 22S precursor (though referred to as 21S) in yeast
that are cold-sensitive and that harbor mutations in the
conserved box A sequence of the U3 snoRNA (9). At both
30°C and 16°C, these mutants demonstrate a processing pat-
tern similar to those of our Mpp10 truncation mutants; spe-
cifically, a reduction of 32S and 27SA2 and an accumulation of
22S (21S). We do not know if the 22S precursor is a dead-end
in pre-rRNA processing, or if it is merely seen when processing
at the A1?A2 sites is slowed. Pulse-chase analysis, which might
differentiate between these possibilities, has so far been un-
informative, since we have been unable to detect aberrant
precursors with this methodology even when Mpp10 is de-
pleted (data not shown and ref. 23).
Studies on the U3 snoRNA mutations that confer cold-
sensitivity and a deficiency in pre-rRNA processing at the A1
and A2 sites suggests that the U3 snoRNP may play role in the
formation of a predicted, conserved pseudoknot at the 5? end
of the 18S rRNA, adjacent to the A1 cleavage site (9, 30).
Furthermore, it has been proposed that this pseudoknot-
forming interaction is linked to U3-dependent cleavage at sites
A1 and A2, since sequences within the highly-conserved box
A region of the U3 snoRNA have the potential to base pair
with the pseudoknot-forming sequences in the rRNA. One
possibility is that the U3 snoRNP, by virtue of base pairing
between the snoRNA and the pre-rRNA, brings the
pseudoknot-forming sequences proximal to each other, prior
to nearby cleavage at A1. This is supported by experimental
evidence derived from studying the cold-sensitive U3 snoRNA
mutants, since they bear nucleotide changes that would disrupt
these interactions. Because our Mpp10 truncation mutants
also show deficient A1?A2 processing, perhaps these muta-
tions target the same mechanism. One possibility is that
Mpp10 is required for facilitating or maintaining snoRNA-
pre-rRNA base pairing, perhaps by contacting both the
snoRNP and the pre-rRNA at the same time.
Strikingly, both the Mpp10 truncations and these U3 mu-
involved in macromolecular assembly. Arguments based on
thermodynamic principles have asserted that macromolecular
assembly mutants are particularly sensitive to cold tempera-
tures. Because these processes are largely driven by hydropho-
bic interactions, the effects of assembly mutations can be
exaggerated at lower temperatures (31). These arguments are
also supported by former studies that found that cold-sensitive
mutants affecting ribosome assembly were easier to generate
than temperature-sensitive ones (32, 33). We do not yet know
if the Mpp10 truncations and the cold-sensitive U3 mutants,
when coexpressed, cause lethality. If so, this would suggest that
the two molecules are functioning in the same pathway.
Our results suggest that slowed growth in the truncation
mutant strains is the result of impaired processing of pre-
MPP10 or the truncations, mpp10–1 and mpp10–2,was resolved on a 1% formaldehyde-agarose gel. Blots were probed with the indicated32P-labeled
oligonucleotides. The relevant cleavage sites are indicated. (A) Northern analysis on RNA that was isolated from strains grown to the same density.
(B) Northern blot analysis on RNA isolated from strains grown the same amount of time (10 hr).
Truncation of the Mpp10 protein impairs processing at sites A1 and A2 in the pre-rRNA. Total RNA from strains bearing full-length
13540 Biochemistry: Lee and Baserga Proc. Natl. Acad. Sci. USA 94 (1997)
rRNA. Although both mpp10–1 and mpp10–2 strains show
inefficient processing at these sites at 30°C, only the double
mutant, the mpp10–2 strain, incurs a growth defect. Northern
analysis on steady-state levels of rRNA reveals that both the
mpp10–1 and mpp10–2 strains have decreased levels of 18S
rRNA, mpp10-2 more severely affected than mpp10–1. From
wild type to mpp10–1 to mpp10–2, a successive under-
accumulation of both 32S and 20S rRNA is observed, indic-
ative that processing efficiency also declines in this graded
fashion among the three strains. This suggests that there may
be a level of processing deficiency that is tolerable for the cells
and allows them to divide normally, as in the mpp10–1 strain.
Rates of rRNA production in mpp10–1 seem to fall above this
level and the strain grows as quickly as wild-type yeast.
However, if the processing efficiency falls below this permis-
sible level, the rate of rRNA production is no longer able to
match the demands of a cell dividing at wild-type rates.
Therefore, even though processing in the mpp10–2 strain is not
drastically different from the mpp10–1 strain, the decrease in
efficiency compared with wild-type must exceed the threshold,
causing the double mutant to show a sharp two-fold decrease
in cell division rate.
Of note is the accumulation of the primary transcript, the
35S rRNA, in the double truncation mutant strain (mpp10–2),
but not in the single truncation strain (mpp10–1). It is possible
that the appearance of an increase in the 35S precursor occurs
when pre-rRNA processing is only severely affected, as an
increase in the 35S precursor is also observed for yeast
individually depleted of three U3 snoRNP components (9, 20,
23). This is in contrast to results obtained after disrupting
pre-rRNA processing in Xenopus oocytes, where levels of the
primary transcript remain constant following depletion of the
U22 snoRNA, which is also required for pre-18S rRNA
processing (34, 35). This may reflect a difference between
Although our results indicate that part of the Mpp10 protein
is required for U3 snoRNA association, it is not known with
which part of the U3 snoRNA it is associated. Experiments in
vertebrates have indicated that only the 3? half of U3 is
required for fibrillarin binding, and this association requires an
intact box C (36). A 55 kDa U3 snoRNP protein identified in
hamster cells is also associated with the 3? half of the U3
snoRNA (16). Determination of the U3 nucleotides required
for binding of Mpp10 may strengthen our view that it is
involved in the formation of the pseudoknot at the 5? end of
the 18S rRNA, if similar sequences are required for U3
snoRNP association and pre-rRNA processing.
We wish to thank these members of the Baserga lab for their help
and advice: Steven Wormsley, David Dunbar, and Tina Agentis. We
thank Peter Glazer and Chris Yoo for critical reading of the manu-
script. S.J.B. is a member of the Yale Cancer Center.
1. Maxwell, E. S. & Fournier, M. J. (1995) Annu. Rev. Biochem. 64,
Venema, J. & Tollervey, D. (1995) Yeast 11, 1629–1650.
Kass, S., Tyc, K., Steitz, J. A. & Sollner-Webb, B. (1990) Cell 60,
Hughes, J. M. X. & Ares, M., Jr. (1991) EMBO J. 10, 4231–4239.
Mougey, E. B., Pape, L. K. & Sollner-Webb, B. (1993) Mol. Cell.
Biol. 13, 5990–5998.
Beltrame, M. & Tollervey, D. (1992) EMBO J. 11, 1531–1542.
Beltrame, M., Henry, Y. & Tollervey, D. (1994) Nucleic Acids
Res. 22, 4057–4065.
Savino, R. & Gerbi, S. A. (1990) EMBO J. 9, 2299–2308.
Hughes, J. M. (1996) J. Mol. Biol. 259, 645–654.
Li, H. V., Zagorski, J. & Fournier, M. J. (1990) Mol. Cell. Biol.
Morrissey, J. P. & Tollervey, D. (1993) Mol. Cell. Biol. 13,
Fournier, M. J. & Maxwell, E. S. (1993) Trends Biochem. Sci. 18,
Beltrame, M. & Tollervey, D. (1995) EMBO J. 14, 4350–4356.
Elela, S. A., Igel, H. & Ares, M., Jr. (1996) Cell 85, 115–124.
Parker, K. A. & Steitz, J. A. (1987) Mol. Cell. Biol. 7, 2899–2913.
Nucleic Acids Res. 21, 5377–5385.
Henriquez, R. G., Blobel, G. & Aris, J. P. (1990) J. Biol. Chem.
Jansen, R. P., Hurt, E. C., Kern, H., Lehtonen, H., Carmo-
Fonseca, M., Lapeyre, B. & Tollervey, D. (1991) J. Cell Biol. 113,
Schimmang, T., Tollervey, D., Kern, H., Frank, R. & Hurt, E. C.
(1989) EMBO J. 8, 4015–4024.
Jansen, R., Tollervey, D. & Hurt, E. C. (1993) EMBO J. 12,
Tollervey, D., Lehtonen, H., Carmo-Fonseca, M. & Hurt, E. C.
(1991) EMBO J. 10, 573–583.
Balakin, A. G., Smith, L. & Fournier, M. J. (1996) Cell 86,
Dunbar, D. A., Wormsley, S., Agentis, T. M. & Baserga, S. J.
(1997) Mol. Cell. Biol. 17, 5803–5812.
Mumberg, D., Muller, R. & Funk, M. (1995) Gene 156, 119–122.
Hoffman, C.-S. & Winston, F. (1987) Gene 57, 267–272.
Berges, T., Petfalski, E., Tollervey, D. & Hurt, E. C. (1994)
EMBO J. 13, 3136–3148.
Lafontaine, D., Vandenhaute, J. & Tollervey, D. (1995) Genes
Dev. 9, 2470–2481.
Benton, B. M., Zang, J. H. & Thorner, J. (1994) J. Cell Biol. 127,
Meier, U. T. (1996) J. Biol. Chem. 271, 19376–19384.
Dennis, P. P., Russell, A. G. & Moniz, D. S. M. (1997) RNA 3,
Strauss, E. J. & Guthrie, C. (1991) Genes Dev. 5, 629–641.
Guthrie, C., Nashimoto, H. & Nomura, M. (1969) Proc. Natl.
Acad. Sci. USA 63, 384–391.
Tai, P. C., Kessler, D. P. & Ingraham, J. (1969) J. Bacteriol. 97,
Tycowski, K. T., Shu, M. D. & Steitz, J. A. (1994) Science 266,
Dunbar, D. A., Ware, V. C. & Baserga, S. J. (1996) RNA 2,
Baserga, S. J., Yang, X. W. & Steitz, J. A. (1991) EMBO J. 10,
Biochemistry: Lee and BasergaProc. Natl. Acad. Sci. USA 94 (1997)13541