1
Title: The yeast mitochondrial degradosome: its composition, interplay between RNA helicase and RNase activities and the role in mitochondrial RNA metabolism.
Running title: Functional and biochemical analysis of the yeast mitochondrial
degradosome.
Andrzej Dziembowski, Department of Genetics, Warsaw University and Institute of Biochemistry
and Biophysics, Polish Academy of Sciences Pawinskiego 5A, 02-106 Warsaw, Poland; E-mail:
Jan Piwowarski, Department of Genetics, Warsaw University and Institute of Biochemistry and
Biophysics, Polish Academy of Sciences, Pawinskiego 5A, 02-106 Warsaw, Poland; E-mail:
5DIDá +RVHU, Department of Genetics, Warsaw University, Pawinskiego 5A, 02-106 Warsaw,
Poland;
0LFKDá 0L�F]XN, Department of Genetics, Warsaw University and Institute of Biochemistry and
Biophysics, Polish Academy of Sciences, Pawinskiego 5A, 02-106 Warsaw, Poland; E-mail:
Aleksandra Dmochowska, Department of Genetics, Warsaw University and Institute of
Biochemistry and Biophysics, Polish Academy of Sciences, Pawinskiego 5A, 02-106 Warsaw,
Poland;E-mail: [email protected]
Michel Siep, current address: Department of Endocrinology and Reproduction, Faculty of
Medicine and Health Sciences, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands; E-mail:
Hans van der Spek, Section for Molecular Biology, Swammerdam Institute for Life Sciences,
University of Amsterdam, Kruislaan 318, 1098 SM, Amsterdam, The Netherlands; E-mail:
Les Grivell, current address: EMBO, Meyerhofstrasse 1, 69117 Heidelberg, Germany; E-mail:
3LRWU 3� 6WSLH� , Department of Genetics, Warsaw University and Institute of Biochemistry and
Biophysics, Polish Academy of Sciences, Pawinskiego 5A, 02-106 Warsaw, Poland. Tel.: 48-22-
659-70-72 (ext. 22 40); Fax: 48-22-658 47 54; E-mail: [email protected].
* To whom correspondence should be addressed,
9 figures
Copyright 2002 by The American Society for Biochemistry and Molecular Biology, Inc.
JBC Papers in Press. Published on November 7, 2002 as Manuscript M208287200 by guest on January 7, 2020
http://ww
w.jbc.org/
Dow
nloaded from
2
Summary
The yeast mitochondrial degradosome (mtEXO) is an NTP-dependent exoribonuclease
involved in mitochondrial RNA metabolism. Previous purifications suggested that it was
composed of three subunits. Our results suggest that the degradosome is composed of only
two large subunits: an RNase and a RNA helicase encoded by nuclear genes DSS1 and
SUV3, respectively, and that it co-purifies with mitochondrial ribosomes. We have found
that the purified degradosome has RNA helicase activity which precedes and is essential
for exoribonuclease activity of this complex. The degradosome RNase activity is necessary
for mitochondrial biogenesis but in vitro the degradosome without RNAse activity is still
able to unwind RNA. In yeast strains lacking degradosome components there is a strong
accumulation of mitochondrial mRNA and rRNA precursors not processed at 3’- and 5’-
ends. The observed accumulation of precursors is probably the result of lack of degradation
rather than direct inhibition of processing. We suggest that the degradosome is a central
part of a mitochondrial RNA surveillance system responsible for degradation of aberrant
and unprocessed RNAs.
Keywords:
yeast/mitochondria/RNA turnover/mtEXO/mitochondrial degradosome
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
3
Introduction
RNA turnover is a very important step in regulation of gene expression. RNA
degradation is mediated mostly by large multiprotein complexes like the degradosome in
bacteria (1) or the exosome in the cytoplasm of eukaryotes (2). The function and
composition of these complexes has been the subject of intensive investigations in recent
years. Much less is known about enzymes involved in RNA turnover in mitochondria (3)
(4) (5).
We use the yeast Saccharomyces cerevisiae as a model to study mitochondrial
RNA metabolism. There are only 3 known ribonucleases which are involved in RNA
turnover in yeast mitochondria: Ynt20 (6), Nuc1 (7) and the multiprotein complex known
as the mitochondrial degradosome or MtEXO (4). Since YNT20 and NUC1 are not
essential for mitochondrial gene expression, their function is redundant (8); (6). In contrast
to this, the mitochondrial degradosome is necessary for mitochondrial biogenesis and
mutations in its subunits lead to respiratory incompetence (9,10).
The yeast mitochondrial degradosome was initially identified as a hydrolytic NTP-
dependent 3’:�¶ H[RULERQXFOHDVH DQG LQ DGGLWLRQ ZDV VKRZQ WR KDYH 51$-dependent
NTPase activity (11). SDS-PAGE analysis of the purified degradosome has shown 3
protein bands migrating at 110, 90 and 75 kDa. However, in contrast to this the native
molecular weight of the complex was estimated by size exclusion chromatography as 160
kDa (11). So far only 2 genes encoding degradosome subunits have been identified: SUV3
encoding an 84 kDa putative RNA helicase which was shown to be a bona fide
degradosome component (9,12) and DSS1 encoding a 105 kDa putative hydrolytic
exoribonuclease homologous to bacterial RNaseII (10). In the case of Dss1p there was no
direct proof that it is indeed part of the degradosome, but inactivation of the DSS1 gene
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
4
resulted in a complete loss of degradosome activity in mitochondria. (13). The putative
third (75 kDa) component remained to be identified.
Inactivation of either the SUV3 or DSS1 gene gave similar phenotypes: respiratory
incompetence, very strong inhibition of mitochondrial translation, a variety of disturbances
of RNA processing and stability, which finally lead to the loss of mitochondrial genomes
(10,13-15). Initially the research on SUV3 mutations was concentrated on their effect on
intron metabolism. It has been shown that the omega intron from the 21S rRNA transcript
was accumulated up to 90-fold, other group I introns accumulated as well, but to a smaller
extent (3,14-16). On the basis of these results it has been suggested that the physiological
function of the degradosome is to protect mitochondria from a toxic effect of undegraded
introns (12). The function of the mitochondrial degradosome is, however, not exclusively
intron-related: we have shown that introduction of intronless mitochondrial genomes to
∆SUV3 and ∆DSS1 strains does not restore respiratory competence (13,14). In such strains
15S rRNA and COB mRNA were found to be unstable, moreover precursors of 21S rRNA
and VAR1 mRNA were present (13). The above results indicated that more detailed
analysis of the involvement of the degradosome in mitochondrial RNA metabolism is
required.
In the present study we have purified the degradosome using tandem affinity
chromatography (17) and shown that in contrast to previously published data it contains
only 2 large subunits (Suv3p and Dss1p) The complex co-purifies with mitochondrial
ribosomes. We have shown that the degradosome has RNA helicase activity, which
precedes and is essential for RNA hydrolysis by this complex. We present data that the
Dss1 protein is responsible for exoribonuclease activity of the degradosome and that this
activity is essential for mitochondrial biogenesis. Finally, we have analyzed the influence
of inactivation of degradosome components on mitochondrial RNA processing and have
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
5
shown that the lack of degradosome activity causes aberrations of all processing steps of
rRNA and mRNA, but not tRNA.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
6
Materials and methods
Strains used and constructed. For an analysis of phenotypes of SUV3 and DSS1
disruptions we used the previously described strains: BWG ∆i (wt) [MATa, his1, ade1,
leu2, ura3] , ∆SUV3 ∆i [MATa, his1, ade1, leu2, Suv3∆::ura3], ∆DSS∆i [MATa, his1,
ade1, leu2, Dss1 ∆::ura3], (10,14)
Strains used for degradosome purifications were derived from W303. SUVTAP and
DSSTAP were constructed by in vivo recombination as described in (17) using plasmid
pBS1539 as template. Strain DSSSER was constructed by in vivo recombination. Strain
W303 was transformed by the PCR product obtained on template of total DNA from
DSSTAP strain and primers: SER (TTA TAC AGG TCG ACC TTT TAG ACA TGA
AAT GAT TGG AGC TAA ACA ATC TTT GAC AGT AAC) CON (AAC ATG AAT
TAA CAC CAT CGC AGC AAC GAG). Primer SER contains the DSS1 gene coding
sequence with a mutation changing tyrosine 814 to serine. The proper integration was
confirmed by western blotting and by sequencing of PCR products.
Diploid strains W303/DSSTAP and W303/DSSSER were constructed by genetic
crosses.
Isolation of mitochondria Mitochondria were isolated as described previously (18). For
mitochondrial RNA analysis yeast were grown in YPD medium (2% bactopeptone, 1%
yeast extract Difco) with 2% glucose, and for degradosome purification in YPD medium
containing 2% glycerol and 1% ethanol.
Western blotting. Proteins were resolved on 12 % SDS-PAGE and then transferred to a
Hybond-C membrane. Proteins were detected with the following antibodies: TAP-tag -
PAP (peroxidase-anti-peroxidase, Sigma), NAM9 (19) and visualized using ECL from
Amersham.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
7
Extract preparation and purification of TAP-tagged complexes. Total native extracts
were prepared as described previously (4). For preparation of native mitochondrial extracts
mitochondria isolated from 2.5 l yeast culture were resuspended in 5 ml of lysis buffer (20
mM TRIS pH 8.0, 150 mM KCl, 2% Triton X-100, 0.5 mM EDTA, 2X complete protease
inhibitor (Boehringer ), 1 mM PMSF) incubated with rotation for 15 min in 40C,
centrifuged 20,000 g at 4 0C and the supernatant was collected. TAP purification
procedure was performed exactly as described (17) with the exception that 5 ml of
mitochondrial extract was mixed with 5ml of IPP150 buffer (10mM Tris pH8, 150mM
NaCl, 0.1% NP-40) and directly loaded on a IgG affinity column without dialysis. For
SDS-PAGE proteins were precipitated by TCA. For biochemical analysis protein
complexes were dialyzed to buffer containing 50% glycerol 10mM Tris pH 8.0 50 mM
KCl, 10 mM MgCl2 and 1mM DTT.
Protein identification. Proteins were identified by Mass Spectrometry Laboratory at
Institute of Biochemistry and Biophysics, Polish Academy of Sciences. In-gel digestion of
the protein was performed by the protocol of Shevchenko, et al. (20), using bovine trypsin
(Promega). Q-Tof mass spectrometer (Micromass) was used for peptide mass
fingerprinting and peptide sequencing by tandem mass spectrometry. The Suv3 protein was
identified by peptide mass fingerprinting with peptide coverage of protein 18%. Dss1p and
ribosomal proteins were identified by peptide sequencing. Sequenced peptides were for:
Dss1p (VELDHTR, QYLTVTSPLR, LINSDFQLITK, NSNAVIFGEGFNK,
DISALYPSVIQLLK, ELDNDQATETVVDR, LYDLTNIEELQWK);
Mrpl3p (FLPESELAK, STVNEIPESVASK, LQLPNELTYSTLSR, MEPFEFTLGR,
FFNNSLNSK, SIIAAIWAVTEQK SPVFIVHV FSGEETLGEG YGSSLK);
Mrpl40p (GQPDL IIPWPKPDPI DVQTNLATDP, VIAREQTFWV DSVVR, VFEFLEK)
Mrp1p(GLFSIEGLQK, EVSYIPL LAIDASPK);
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
8
Mrpl35p (YSPPEHIDEIFRMSYDFLEQR, DIIDYDVPVYR, LETLAAIPDTLPTLVPR
FVVWVFR)
Isolation of ribosomes. Mitochondrial ribosomes were isolated by centrifugation through
sucrose cushion as described previously (19).
Analysis of degradosome enzymatic activities. NTP–dependent exoribonuclease activity
was analyzed as described previously (4).
RNA helicase assays. RNA helicase activity was assayed by the strand displacement
method. The partially double-stranded RNA substrate was prepared as described in (21)
with the exception that substrates after annealing were purified by 15% native acrylamide
gel electrophoresis. The RNA helicase activity assay was performed in 20 µl reaction
volume containing 10 mM Tris-Cl pH 8.0, 25 mM KCl, 10 mM MgCl2, 1 mM DTT, 0.1
mg/ml BSA, 1mM CTP , and 100 fM 32P-labelled partial duplex RNA substrate at 30°C.
In a time course aliquots (5 µl) were analyzed on a native 8% polyacrylamide gel
containing 0.5× Tris-borate-EDTA and 0.1% SDS. Strand separation was visualized by
autoradiography.
RNA isolation. RNA was isolated from mitochondria by the guanidinium thiocyanate-acid
phenol procedure using TriReagent (MRC, USA). RNA was normalized with respect to the
amount of COX1 mRNA assayed by Northern blot.
Northern blotting. For Northern blot analysis RNA was resolved in 1 % agarose gels
containing 0.925 % formaldehyde, and 1X NBC buffer (0.5 M boric acid, 10 mM sodium
citrate, 50mM NaOH), transferred to nylon membranes by passive diffusion or resolved in
6% denaturing acrylamide gels and transferred to nylon membranes by electrophoretic
transfer in 0.5 TBE buffer in trans-Blot Cell apparatus (Bio-Rad). Blots were washed 2
times with 2X SSC. RNA was immobilized by UV crosslinking and incubation at 800C in
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
9
vacuum, hybridized at 550C for oligonuclotide probes or at 65 0C for DNA fragment probes
in : 7% SDS, 0.5M Na3P04 pH 7.4, 1mM EDTA, 1% BSA buffer.
Oligonucleotides used for hybridization were as follows: COB mRNA(TAT CTA TGT
ATT AAT TTA ATT ATA TAT TAT TTA TTA ACT CTA CCG AT) COB 3’- precursor
(TAT TAT TAT TAT TTA ATT TTT ATA AAT TAG AGA TAT), VAR1 mRNA( AAA
TAT AAT AGA AAA AAG AGT ATT ATA TAT TAA TAT AAA ATA TAT TAA
TAT) VAR1 precursor (GAA GGA GTT TGG TTA AAG AAG ATA AAG AAT AAA
A),15S rRNA (TAT AAG CCC ACC GCA GGT TCC CCT ACG GTA ACT GTA) 15S
rRNA 3’- precursor (ATA TAA TAT TAA TAT TTA TAT TAA TAT TTA GAT TAA
TAT TTA)
Probes for tRNA detection were the products of PCR reaction using radiolabelled [P32]
dATP. One of the primers was biotinylated and strands were resolved on streptavidin
paramagnetic beads (Dynal) using NaOH as for solid phase nuclease S1 mapping. For
tRNA Val and Thr: primers were Valth P (AAA TAA ATA TTA TTA TAT TAA GAT
AG) and Valth L + 5’-biotin (AAT AAT AT TAT AAT AAA TTTA TAA ATG) and both
strands were used as separate probes. For tRNA Ala I Ile primers were tRANL1 (TAT
AAT TTT ATA TTT TAA TAT AGG) and TRNAP+5’-biotin (GAA ACT AAC AGG
GAT TGA ACC) and as a probe was used the strand containing biotin.
S1 nuclease mapping. 3’- ends of COX1 and COB mRNAs were mapped by classical S1
mapping method using single stranded probes (22) 3’- ends for 21S rRNA and 15S rRNA
were mapped by solid phase S1 nuclease mapping (23). The probes:
For COX1 mRNA, the DNA fragment excised by EcoR1 and BstY1 restriction
enzymes from pUCCOX1 plasmid containing 3’- end of COX1 gene labeled at the 3’- end
with [P32] dATP using the Klenow fragment (Fermentas) were used as a probe. Single
stranded RNA was purified by denaturing acrylamide gel electrophoresis. Plasmid
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
10
pUCCOX1 was constructed by cloning of the PCR product obtained using primers:
MWG1 (TAA AAT GGA ACT AAT) MWG2 (TGT GGC TTC CCA ATG CAT TTC
TTA GG) and mitochondrial DNA as a template, digested with NsiI and BstY1 into
pUC18 vector digested with NsiI and BamH1.
For COB mRNA, the DNA fragment excised by Csp61 and BamHI restriction
enzymes from pUCCOB plasmid containing the 3’- end of COB, labeled at the 3’- end
with [P32] dATP using Klenow fragment (Fermentas) was used as a probe. Single stranded
DNA was purified by denaturing acrylamide gel electrophoresis. Plasmid pUCCOB was
constructed by cloning of the PCR product obtained using the primers: MWG23 (CGG
GAT CCT AGG ACA AAT TGG AGG TGCC) MWG 24 (CGG AAT CAA ATC TCC
TTG CGG GGT CC) and the template of mitochondrial DNA and digested with EcoRI and
BstY1 into pUC18 vector digested with Eco RI and BamH1.
For 15S rRNA, the PCR product obtained using primers 15SL (ACA GGC GTT
ACA TTG TTG TC) 15SP( CCC TTA TTT ATT TAA AGA AGA TG) digested with
Sau3A restricton enzyme was used as a probe.
For 21S rRNA, the PCR product obtained using the primers 21SNL2 (AGT ACG
CAA GGA CCA TAA TG) 21SR2 (TTT CGA TTA CAA AAC GAA TGC) digested with
NcoI restriction enzyme was used as a probe.
Primer extension. 5 µg of mitochondrial RNA were hybridized with 2pM of γ[32P]ATP
labeled primer in 200 mM KCl, 10 mM Tris pH 8.3. The reaction mixture was heated at
850C for 5 min and slowly cooled to 420C and then 10 µl of elongation mix was added (2x
buffer for AMV, dNTP 2mM, 7u. AMV Boehringer) Reactions were stopped by addition
of 12 µl formamide and products were resolved in 6% acrylamide 7M urea gels.
Accompanying sequencing reactions were used as molecular weight markers. Primers were
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
11
used as follows: COX1(TAT ATT TAA TGA TAT TAA TAC TCTC); VAR1 (GAA ATA
TAT ATA TAT ATA TAA TAT GCA TCC) COB (CAA TTA TTA TTA TTA TTA TTA
TAC ATA AA); 15S rRNA (CGT ATG ACT CGT ATG CGT CAT GTC C); 21S rRNA
(TTT AAT TAT TAT ACT CCA TGT TAT CT); COX3 (ATT AAT ATA AAT CAT
TGA TAA TAT CTT); ATP6/8 ( ATA TTA GTA TTTA TTT ATA TAG TTC CC)
Mapping the 3’- end of COB mRNA by ribonuclease T1 digestion. The
procedure was very similar to that described previously (24). 5 µg of mitochondrial RNA
were hybridized to 1 pM oligonucleotide CobT1 (TAT CTA TGT ATT AAT TTA ATT
ATA TAT TAT TTA TTA ACT CTA CCG AT) in 20 µl by heating at 650C for 5 min, and
slow renaturation to 370C (90-120 min). Heteroduplexes were ethanol precipitated,
resuspended in 5µl of digestion buffer (20mM sodium citrate pH 5.0, 10mM EDTA, 5 u
RNase T1 (Gibco-BRL)) and incubated at 37 0C for 30 min. 5µl of formamide were added
to the samples. The samples were heated at 750C for 3 min and resolved in 8% acrylamide
7M urea gel, transferred to nylon membrane by electrophoresis transfer in 0.5 TBE buffer
in trans-Blot Cell (Bio-Rad) and hybridized using [32P] end labeled CobT1
oligonucleotide. The accompanying DNA sequencing reactions were used as molecular
weight markers.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
12
Results
The mitochondrial degradosome contains only 2 large subunits Dss1p and Suv3p.
To determine the structure and in vitro activity of the degradosome we decided to
purify this protein complex by the tandem affinity purification method (TAP) (17). In our
previous work (4) we constructed a yeast strain with C-terminal Suv3 TAP-tag fusion
(named SUVTAP). In addition, for the purpose of this study we constructed a C-terminal
Dss1p Tap-tag fusion strain (DSSTAP). Western blot analysis of total native yeast extracts
using anti-TAP antibodies revealed that in such conditions Dss1p always partially degrades
(Fig 1). Therefore we developed a method of degradosome isolation from purified
mitochondria which minimizes protein degradation. In such conditions, using either Suv3-
TAP fusion or Dss1-TAP fusion the purified degradosomes were active as a NTP-
dependent exoribonuclease and contained only 2 large proteins migrating as 75kDA and
105 kDa which we assumed to be the Suv3p and Dss1p proteins (Fig 2B). The two
proteins were excised from the gel and subjected to mass spectrometric analysis. As
expected they represented SUV3 and DSS1 gene products. This result correlates with the
predicted molecular masses of the Suv3p (81 kDa) and Dss1p (108 kDa). The differences
in gel migration of Suv3p and Dss1p in purifications from strains DSSTAP and SUVTAP
were due to the rest of the TAP-tag in Dss1p and Suv3p proteins, respectively. There was
no 90 kDa protein, which was previously believed to correspond to Suv3p (12). SDS-page
analysis of protein samples from mock TAP purification shows no proteins, confirming
specificity of purification (Fig 2B). On the basis of these results we suggest that the
degradosome contains only two large subunits: Suv3p and Dss1p.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
13
The mitochondrial degradosome complexes co-purify with mitochondrial ribosomes.
In addition to Suv3p and Dss1p, the TAP purified degradosome preparations
contained smaller proteins. Their concentrations changed slightly from purification to
purification (data not shown). On the basis of mass spectrometry analysis we identified
three proteins from the large ribosomal subunit (Mrpl3p, Mrpl40p and Mrpl35p) and one
protein from the small subunit (Mrp1p). This suggested the association of the degradosome
complexes with ribosomes. Therefore, we purified mitochondrial ribosomes by
centrifugation through a sucrose cushion and analyzed the presence of the degradosome
subunit Dss1p and ribosomal proteins by immunobloting analysis using antibodies specific
for TAP-tag and the ribosomal protein Nam9p (Fig 3). Our analyses indicated that the
Dss1p protein was present exclusively in the ribosomal fraction, suggesting that in vivo all
degradosome particles are associated with ribosomes.
The mitochondrial degradosome has RNA helicase activity followed by
exoribonuclease activity.
RNA-dependent NTP-ase activity of the degradosome and homology of Suv3
protein to known RNA helicases suggested that the degradosome has RNA helicase
activity, but so far no biochemical data had been reported. Our attempts to isolate Suv3p
and Dss1p in a variety of heterologous expression systems failed, therefore we performed
an RNA helicase assay by a strand displacement method using the degradosome purified
from the DSSTAP strain (Fig 4). As can be seen in Fig 4 the degradosome has the ability
to unwind double-stranded RNA regions; this results in formation of single stranded RNA
which was subsequently degraded exoribonucleolytically. There was no partial degradation
of the substrate before unwinding, so RNA helicase activity precedes exoribonuclease
activity of this complex. This result is in agreement with the previously reported absolute
requirement for nucleotide triphosphates for RNA degradation by MtEXO (11).
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
14
Dss1p is responsible for RNase activity of the degradosome, which is essential for
respiratory competence of mitochondria.
RNase activity of the degradosome seems to be associated with the Dss1 protein
which has regions of homology to bacterial RNase II. To understand the role of the Dss1p
in degradosome functions we constructed the yeast strain DSSSER containing a point
mutation within the DSS1 coding sequence and a TAP-tag at the C-terminus of protein.
The tyrosine 814 residue was changed to serine. This tyrosine is conserved in the
exoribonuclease family to which Dss1p belongs and it may be involved in the catalytic
mechanism (25). The constructed strain appeared to be respiratory incompetent with very
unstable mitochondrial genomes (data not shown). This indicates that this amino acid is
essential for Dss1p function and mitochondrial biogenesis.
We have tried to purify the mitochondrial degradosome from the mutant strain and
from a non-respiring strain DSSTAP rho0 as a control, but western blot analysis revealed
that Dss1p is unstable in mitochondrial extracts from respiratory incompetent strains (data
not shown). To avoid this problem we constructed respiring diploid strains W303/DSSSER
and W303/DSSTAP (wild type Dss1p) as a control. TAP purified degradosomes were
assayed for NTP-dependent exoribonuclease and RNA helicase activity. Results presented
in Figure 5 show that the mutation Tyr814Ser of the Dss1p causes complete loss of the
exoribonuclease activity of the degradosome but does not abolish RNA helicase activity.
These results show that Dss1p is responsible for exoribonuclease activity of the
degradosome and that it is essential for proper mitochondrial biogenesis.
Lack of degradosome components causes accumulation of RNA precursors not
processed at the 3’- and 5’-ends.
Our previous results have shown that a disruption of SUV3 or DSS1 genes results in
alterations in RNA stability and processing in mitochondria in a system where intronless
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
15
mitochondrial genomes were introducted into mutant strains (13). 15S rRNA and COB
mRNAs were unstable, in addition in the mutants we detected several precursors of 21S
rRNA and VAR1, which were not visible on Northern blots prepared from the wild-type
strains. We decided to look more carefully at mitochondrial RNA processing in ∆SUV3
and ∆DSS1 strains with intronless mtDNA.
We analyzed processing of 3’- ends of 21S rRNA, 15S rRNA and mRNA for COB
and COX1 by S1 nuclease mapping (Fig 6a). Our results show that the mature form of
RNA is produced for all RNA classes tested. Processing of 21S rRNA is strongly inhibited
and the precursor accumulates. In the case of 15S rRNA, and mRNA for COX1 and COB
precursor molecules were not seen because they were longer than the probes used for S1
nuclease mapping. Therefore we analyzed 3’- end processing of 15S rRNA, COB mRNA
and VAR1 mRNA by Northern blot hybridizations using two oligonucleotide probes: one
complementary to the 3’- end of mature RNA and the other complementary to the region
adjacent to the precursor (Fig 6b). For all RNAs tested the accumulation of precursors not
processed at the 3’- end was observed. On autoradiograms of Northern blots with precursor
specific probes it was impossible to detect any new RNA classes not seen on Northern
blots with mRNA specific probes. This suggests that all RNA detected with precursor
specific probes contains mature RNA and there is no accumulation of noncoding RNA
classes arising after 3’- end processing.
The mRNAs for Var1 and Cob are located at the end of polycistronic transcripts
and accumulation of very high molecular weight precursors may suggest problems with
transcription termination. This would, however, be very difficult to prove, as the sites of
transcription termination in yeast mitochondria are not known and noncoding RNA classes
arising after 3’- end processing are very unstable.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
16
To analyze the role of the degradosome in 5’-end RNA processing we used the
primer extension method (fig 7). We analyzed all transcripts known to be specifically
processed at the 5’-end: 15S rRNA and COB, VAR1 and ATP6/8 mRNA. As a control we
used 21S rRNA, COX1 mRNA not requiring 5’-end processing and COX3 mRNA with the
mature 5’-end generated by tRNA excision. The analysis was also done for Val tRNA. For
all specifically processed RNA there was strong accumulation of precursors which are not
processed at the 5’-end. Processing of COX3 mRNA and tRNAVal was not impaired.
The mitochondrial degradosome is not involved in tRNA processing.
The mature mitochondrial tRNA is generated in three steps: 5’-endonucleolytic
cleavage by RNase P, 3’-end processing by an unidentified endonuclease and CCA
synthesis by tRNA nucleotide transferase (26). In mammalian mitochondria
exoribonucleases are involved in tRNA repair processing after incorrect CCA addition
(27).
In order to check if the degradosome is involved in tRNA processing we analyzed
the maturation of tRNA in ∆DSS1 and ∆SUV3 strains by Northern blotting. Four tRNAs
were tested:
a) tRNAVal and tRNAThr which are encoded by complementary strands. Their transcripts
partially overlap. Single stranded DNA fragments complementary to tRNA and
overlapping regions were used as probes. As can be seen, there are no aberrancies in tRNA
processing, and there is no accumulation of transcribed intergenic regions (Fig 8).
b) tRNAAla and tRNAIle which are produced from one transcript (fig 8). The processing of
these tRNAs was also found to be correct and no accumulation of precursors or intergenic
regions was observed.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
17
In all cases the amount of mature tRNAs was 2-3 times lower in our mutant strain
in comparison to respiring wild type yeast, but no effect on processing was detected.
Inactivation of degradosome components does not cause aberrations at the 3’- end of
mature COB mRNA and there is no polyadenylation of this transcript.
Mature COB mRNA is unstable in ∆DSS and ∆SUV3 mutant stains (13). We asked
if this instability is caused by small changes at the 3’- end of mRNA or by possible
polyadenylation. So far there were no reports showing conclusively whether
polyadenylation occurs in yeast mitochondria (28,29). In plant mitochondria as in bacteria
polyadenylation destabilizes transcripts (30). To analyze the mature 3’- end with high
resolution we used Northern blot analysis of RNA, which was cut by ribonuclease T1. To
protect some guanylate residues from cleavage we hybridized synthetic DNA
oligonucleotides to the 3’- end of mature COB mRNA and cut the mRNA by G-specific
ribonuclease T1 (Fig 9A). RNA was resolved on sequencing gels and analyzed by
Northern blot using as a probe the same oligonucleotide as for protection. On the
autoradiograms (Fig 9B) we were able to see two bands specific for complete digestion of
the mature 3’- end and the precursor. In the ∆SUV3 and ∆DSS1 strains there was
accumulation of precursors, but there were no changes at the mature 3’- end of COB
mRNA. Therefore, the instability of COB mRNA is not caused by changes at the 3’- end of
mRNA and there is no polyadenylation of this mRNA.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
18
Discussion.
The mitochondrial degradosome contains only two large subunits (suv3p and dss1p)
and co-purifies with mitochondrial ribosomes.
The major enzymatic complex responsible for RNA turnover in mitochondria: the
mitochondrial degradosome was identified more than 10 years ago but there were still
many unanswered questions about its composition and function.
In the present study we have shown that the purified degradosome contains only
two large subunits: Dss1p migrating at 105 kDa and Suv3p migrating at 75 kDa. It has
been proposed that the 90 kDa protein previously seen in degradosome preparations (4,31)
corresponds to the SUV3-encoded protein but direct proof was lacking (12). In contrast to
this, the data presented in this paper show that in degradosome preparations from
mitochondrial extracts obtained in conditions minimizing degradation of Suv3p and Dss1p,
the 90 kDa protein is absent. The procedure described, yielding the pure, active
degradosome has been repeated 5 times and only 2 large (75 and 105 kDa) proteins were
present. Migration of Suv3p as a 75 kDa protein is not surprising since the calculated
molecular weight of this protein after cleavage of the leader is 81 kDa. We think that the
most probable explanation the discrepancy between our present results and previously
published data is that the third protein band of 90kDa was a product of Dss1p degradation.
Indeed, immunoblot analysis revealed that in total cell extracts Dss1p is unstable and
produces a band migrating between Dss1p and Suv3p on SDS-PAGE. Finally, our present
data are in a good agreement with the molecular mass of the native, enzymatically active
complex, estimated as 160 kDa (31).
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
19
Mass spectrometry identification of several of the additional proteins present in
degradosome preparations has shown that they are of ribosomal origin. We identified
proteins from both small and large ribosomal subunits, suggesting that the degradosome is
associated with intact ribosomes. The ribosomal proteins do not seem to be a
contamination since mock purification has shown no proteins. In addition, in contrast to
cytoplasmic ribosomal proteins, mitochondrial ribosomal proteins were not found as
purification artifacts in high-throughput yeast protein complexes purification by the TAP
method (32). Ribosome purification and subsequent immunological analysis confirmed
that degradosome complexes are associated with mitochondrial ribosomes. The ribosome
purification method used in this study pellets only very large complexes and the
mitochondrial degradosome of native molecular weight estimated as 160 kDa could not be
pelleted without association with another large complex (31). Association of the
degradosome with ribosomes is probably not dependent on active translation and/or
mediated by mRNA bridging because ribosome purification was performed in buffers
containing high concentration of EDTA which causes dissociation of ribosome subunits. In
our experiments all degradosome particles were associated with ribosomes but in vivo the
degradosome concentration may be lower than that of ribosomes because there are no
published data on the presence of 75 and 110 kDa proteins in mitochondrial ribosome
preparations analyzed so far (33,34). Our data do not allow us to identify which domains of
the degradosome and of the mitochondrial ribosome are responsible for the binding. In E.
coli the degradosome complex was found to be membrane-associated, but the
physiological significance of this fact is not known (35). It is worth noting that in the null
mutants for the SUV3 or DSS1 genes a strong inhibition of mitochondrial translation was
detected (13). Further research is needed to understand the physiological role and
mechanism of this interaction.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
20
The mitochondrial degradosome has RNA helicase and exoribonuclease activity,
which are both essential for mitochondrial biogenesis.
Suv3p and Dss1p proteins exist in vivo exclusively as a protein complex, which has
both RNA helicase and 3’-5’ exoribonuclease activities. Exoribonuclease activity is
absolutely dependent on RNA helicase, but in vitro the degradosome without RNase
activity is still able to unwind RNA. Nevertheless RNAse activity of the degradosome is
essential for mitochondrial biogenesis.
To the best of our knowledge the interplay between these two helicase and
exoribonuclease activities is unique. The bacterial degradosome, which also has RNA
helicase and exoribonuclease subunits, is able to degrade RNA without ribonucleotide
triphosphates. RNA helicase activity of this complex is only essential for degradation of
RNA containing double stranded regions (36). Dss1p contains an exoribonuclease domain
but does not possess any RNA binding domain present in bacterial exoribonucleases so it is
possible that Suv3p is responsible for interactions with RNA and delivers unwound RNA
to Dss1p.
Is there no polyadenylation of RNA in yeast mitochondria?
RNA polyadenylation exists in nearly all known genetic systems. In the cytoplasm
of eukaryotes polyadenylation stabilizes mRNA. In prokaryotes (37) and plant organelles
(30) (38) polyadenylation destabilizes transcripts and steady state levels of mRNAs
containing poly(A) are very low. In human mitochondria polyadenylation is abundant and
rather stabilizes than destabilizes mRNAs (39). In yeast mitochondria polyadenylation is
not abundant and mRNAs are processed at the 3’- end exactly 2 nucleotides after a
conserved dodecamer sequence (24) but it was possible that as in prokaryotes
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
21
polyadenylation directs RNA to degradation and that only a very small fraction of RNA is
polyadenylated in vivo. In bacteria RNaseII which is homologous to Dss1p can stabilize
mRNA by poly(A) trimming (40). In strains devoid of Dss1p and Suv3p mature 15S rRNA
and COB mRNA are unstable so it was possible that this instability was mediated by
polyadenylation of these RNAs and other mitochondrial RNases were responsible for
degradation. In this case polyadenylated COB mRNA should accumulate in degradosome
deficient strains. Our experiments have shown that there are no differences at the 3’- end of
mature COB mRNA between wild type and ∆SUV and ∆DSS1 strains. This is a suggestion
that in yeast mitochondria there is no polyadenylation of RNA or at least that the COB
mRNA is not polyadenylated.
The mitochondrial degradosome is a part of the mitochondrial RNA surveillance
system.
It has been suggested previously that the main function of the mitochondrial
degradosome is intron turnover and splicing factor recycling (3,12). Our results do not
support this hypothesis as the introduction of intronless mitochondrial genomes to ∆SUV
and ∆DSS1 strains does not restore respiratory competence. Therefore the degradosome
must have other functions essential for mitochondrial biogenesis and the effects observed
on intron metabolism of SUV3 mutants are rather the consequence of general perturbations
in RNA turnover.
Lack of degradosome activity causes accumulation of unprocessed mRNA and
rRNA but the important fact is that at the same time mature RNAs of proper size are
abundant. Therefore the processing of intronless RNA maturation is not abolished in the
∆SUV or ∆DSS1 strains. The degradosome is most probably not involved in RNA
processing directly because processing is endonucleolytic and the degradosome does not
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
22
possess such activity at least in vitro. The most plausible explanation is that the
accumulation of precursors is not caused by impairment of RNA processing but rather by
the lack of RNA degradation. We suggest that the degradosome is a central part of the
mitochondrial RNA surveillance system, which degrades aberrant and unprocessed RNAs.
This hypothesis is supported by the fact that mutations in degradosome components
stabilize and restore expression of unstable mutated mRNAs. Mutation in SUV3 stabilizes
VAR1 mRNA containing a deletion of 206 nucleotides in the 3’- end region (41).
Mutations in DSS1 suppress deletions in the 5’-UTR region of the COB gene, which is
necessary for mRNA stability (42). Association of the degradosome particles with
ribosomes also suggests the function of the degradosome in RNA surveillance since in
mammals many cytoplasmic RNA surveillance factors can be found in ribosomal fractions
(43).
It is not known what discriminates between normal stable mRNA and aberrant
mRNA which should be promptly degraded. Cis stability elements of mitochondrial
mRNAs are located at 5’-UTRs to which stabilizing proteins and translation activators bind
(26,44). RNA turnover has a 3’→5’ direction (45) so 3’- and 5’-ends of mRNA must
interact physically or functionally. At the 3’- end of all mature mitochondrial mRNAs there
is a conserved dodecamer sequence to which the dodecamer binding protein binds (46,47).
Interaction between proteins binding to both ends of RNA could stabilize mRNA, similarly
to cytoplasmic mRNAs. When RNA molecules are not processed, the 3’- end is free and
RNA is recruited for degradation. Unfortunately the gene coding for the dodecamer
binding protein has not been identified and more research is needed to identify putative
interactions of the degradosome with RNA regions or other RNA binding proteins.
Another protein, which may also be involved in mitochondrial RNA surveillance, is
the PET127 gene product (48), which was found to suppress SUV3 and DSS1 disruptions
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
23
when overexpressed (49). This fact prompted us to speculate that the 3’- and 5’-end of
yeast mitochondrial mRNA interact (49). Just like the SUV3 and DSS1 genes, mutations in
the PET127 gene stabilize mRNAs lacking cis stability elements located in their 5’-UTRs
(42,48). In addition Pet127p is involved in 5’-end processing of mitochondrial RNA. The
same mutation in the COB 5’-UTR region is suppressed by mutations in DSS1 and PET127
genes suggesting that both proteins may function in the same pathway. The ability of
PET127 overexpresion to suppress the effects of DSS1 disruption suggests that it could
also activate an alternative degradosome-independent RNA turnover pathway. Further
research is needed to understand the interplay between Pet127p and the degradosome.
It would be interesting to find out if similar RNA surveillance mechanisms exist in
mitochondria of other organisms. SUV3-encoded protein is highly conserved through
evolution and there are orthologs of Suv3p in all eukaryotes, but except for fungi there are
no orthologs of Dss1p. We have analyzed the human ortholog of Suv3p and found that it is
localized in mitochondria and possesses both RNA and DNA helicase activities (50,51);
Minczuk et al in preparation). Research is in progress to identify the function and the
putative partners of human SUV3 RNA helicase.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
24
Acknowledgments
:H WKDQN (ZD %DUWQLN DQG 3LRWU 6áRQLPVNL IRU FULWLFDO UHDGLQJ RI WKH PDQXVFULSW DQG $QQD
Dziembowska for editing the manuscript, Bertrand Seraphin for plasmids containing the
TAP-tag, Andrzej Dziembowski was the recipient of an EMBO short-term fellowship for
work carried out in the Section for Molecular Biology, University of Amsterdam. In the
years 2001 and 2002 Andrzej Dziembowski received FNP (Foundation for Polish Science)
fellowship for young scientists. This work was supported by the State Committee for
Scientific Research, Grants No. 6P04 00319 and 6P04 01818, by the Polish-French Center
for Biotechnology of Plants, and by the Faculty of Biology grant BW.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
25
References
1. Carpousis, A. J., Vanzo, N. F., and Raynal, L. C. (1999) Trends Genet 15(1), 24-8
2. Butler, J. S. (2002) Trends Cell Biol 12(2), 90-6
3. Margossian, S. P., and Butow, R. A. (1996) Trends Biochem Sci 21(10), 392-6
4. Dziembowski, A., and Stepien, P. P. (2001) Methods Enzymol 342, 367-78
5. Gagliardi, D., Perrin, R., Marechal-Drouard, L., Grienenberger, J. M., and Leaver, C.
J. (2001) J Biol Chem 13(8), 1803-18
6. Hanekamp, T., and Thorsness, P. E. (1999) Curr Genet 34(6), 438-48
7. Vincent, R. D., Hofmann, T. J., and Zassenhaus, H. P. (1988) Nucleic Acids Res 16(8),
3297-312
8. Zassenhaus, H. P., Hofmann, T. J., Uthayashanker, R., Vincent, R. D., and Zona, M.
(1988) Nucleic Acids Res 16(8), 3283-96
9. Stepien, P. P., Margossian, S. P., Landsman, D., and Butow, R. A. (1992) Proc Natl
Acad Sci U S A 89(15), 6813-7
10. Dmochowska, A., Golik, P., and Stepien, P. P. (1995) Curr Genet 28(2), 108-12
11. Min, J. J., and Zassenhaus, H. P. (1991) SAAS Bull Biochem Biotechnol 4, 1-5
12. Margossian, S. P., Li, H., Zassenhaus, H. P., and Butow, R. A. (1996) Cell 84(2), 199-
209
13. Dziembowski, A., Malewicz, M., Minczuk, M., Golik, P., Dmochowska, A., and
Stepien, P. P. (1998) Mol Gen Genet 260(1), 108-14
14. Stepien, P. P., Kokot, L., Leski, T., and Bartnik, E. (1995) Curr Genet 27(3), 234-8
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
26
15. Golik, P., Szczepanek, T., Bartnik, E., Stepien, P. P., and Lazowska, J. (1995) Curr
Genet 28(3), 217-24
16. Conrad-Webb, H., Perlman, P. S., Zhu, H., and Butow, R. A. (1990) Nucleic Acids Res
18(6), 1369-76
17. Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M., and Seraphin, B. (1999)
Nat Biotechnol 17(10), 1030-2
18. Daum, G., Sperka-Gottlieb, C. D., Hermetter, A., and Paltauf, F. (1988) Biochim
Biophys Acta 946(2), 227-34
19. Chacinska, A., Boguta, M., Krzewska, J., and Rospert, S. (2000) Mol Cell Biol 20(19),
7220-9
20. Shevchenko, A., Wilm, M., Vorm, O., and Mann, M. (1996) Anal Chem 68(5), 850-8
21. Gallinari, P., Brennan, D., Nardi, C., Brunetti, M., Tomei, L., Steinkuhler, C., and De
Francesco, R. (1998) J Virol 72(8), 6758-69
22. Berk, A. J. (1989) Methods Enzymol 180, 334-47
23. Dziembowski, A., and Stepien, P. P. (2001) Anal Biochem 294(1), 87-9
24. Hofmann, T. J., Min, J., and Zassenhaus, H. P. (1993) Yeast 9(12), 1319-30
25. Mian, I. S. (1997) Nucleic Acids Res 25(16), 3187-95
26. Grivell, L. A. (1995) Crit Rev Biochem Mol Biol 30(2), 121-64
27. Reichert, A. S., and Morl, M. (2000) Nucleic Acids Res 28(10), 2043-8
28. Hendler, F. J., Padmanaban, G., Patzer, J., Ryan, R., and Rabinowitz, M. (1975)
Nature 258(5533), 357-9
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
27
29. Groot, G. S., Flavell, R. A., Van Ommen, G. J., and Grivell, L. A. (1974) Nature
252(5479), 167-9
30. Gagliardi, D., and Leaver, C. J. (1999) Eur J Biochem 18(13), 3757-66
31. Min, J., Heuertz, R. M., and Zassenhaus, H. P. (1993) J Biol Chem 268(10), 7350-7
32. Gavin, A. C., Bosche, M., Krause, R., Grandi, P., Marzioch, M., Bauer, A., Schultz, J.,
Rick, J. M., Michon, A. M., Cruciat, C. M., Remor, M., Hofert, C., Schelder, M.,
Brajenovic, M., Ruffner, H., Merino, A., Klein, K., Hudak, M., Dickson, D., Rudi, T.,
Gnau, V., Bauch, A., Bastuck, S., Huhse, B., Leutwein, C., Heurtier, M. A., Copley,
R. R., Edelmann, A., Querfurth, E., Rybin, V., Drewes, G., Raida, M., Bouwmeester,
T., Bork, P., Seraphin, B., Kuster, B., Neubauer, G., and Superti-Furga, G. (2002 Jan
10) Nature 415(6868), 141-7.
33. Graack, H. R., and Wittmann-Liebold, B. (1998) Biochem J 329(Pt 3)), 433-48
34. Saveanu, C., Fromont-Racine, M., Harington, A., Ricard, F., Namane, A., and
Jacquier, A. (2001) J Biol Chem 276(19), 15861-7
35. Liou, G. G., Jane, W. N., Cohen, S. N., Lin, N. S., and Lin-Chao, S. (2001) Proc Natl
Acad Sci U S A 98(1), 63-8
36. Coburn, G. A., Miao, X., Briant, D. J., and Mackie, G. A. (1999) Genes Dev 13(19),
2594-603
37. Coburn, G. A., and Mackie, G. A. (1999) Prog Nucleic Acid Res Mol Biol 62, 55-108
38. Lupold, D. S., Caoile, A. G., and Stern, D. B. (1999) Plant Cell 11(8), 1565-78
39. Ojala, D., Montoya, J., and Attardi, G. (1981) Nature 290(5806), 470-4
40. Marujo, P. E., Hajnsdorf, E., Le Derout, J., Andrade, R., Arraiano, C. M., and Regnier,
P. (2000) Rna 6(8), 1185-93
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
28
41. Butow, R. A., Zhu, H., Perlman, P., and Conrad-Webb, H. (1989) Genome 31(2), 757-
60
42. Chen, W., Islas-Osuna, M. A., and Dieckmann, C. L. (1999) Genetics 151(4), 1315-25
43. Mangus, D. A., and Jacobson, A. (1999) Methods 17(1), 28-37
44. Fox, T. D. (1996) Experientia 52(12), 1130-5
45. Min, J., and Zassenhaus, H. P. (1993) J Bacteriol 175(19), 6245-53
46. Li, H., and Zassenhaus, H. P. (2000) Curr Genet 37(6), 356-63
47. Li, H., and Zassenhaus, H. P. (1999) Biochem Biophys Res Commun 261(3), 740-5
48. Wiesenberger, G., and Fox, T. D. (1997) Mol Cell Biol 17(5), 2816-24
49. Wegierski, T., Dmochowska, A., Jablonowska, A., Dziembowski, A., Bartnik, E., and
Stepien, P. P. (1998) Acta Biochim Pol 45(4), 935-40
50. Dmochowska, A., Kalita, K., Krawczyk, M., Golik, P., Mroczek, K., Lazowska, J.,
Stepien, P. P., and Bartnik, E. (1999) Acta Biochim Pol 46(1), 155-62
51. Dmochowska, A., Stankiewicz, P., Golik, P., Stepien, P. P., Bocian, E., Hansmann, I.,
and Bartnik, E. (1998) Cytogenet Cell Genet 83(1-2), 84-5
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
29
Figure legends
Figure 1 Dss1p is partially degraded in total native yeast extracts but is more stable in native
mitochondrial extracts. Stability of Dss1p protein containing a TAP-tag was analyzed by
western blotting using anti–TAP antibodies. As a control we used the same procedure for a
yeast strain containing TAP-tagged Suv3 protein which was more stable in both types of
extracts.
Figure 2 The degradosome is composed of Suv3 and Dss1 proteins. The degradosome was
purified from Tap-tagged yeast strains DSSTAP and SUVTAP and from the wild type strain
W303 as a control. After purification all eluted fractions were pooled and divided into equal
portions. The first portion was dialyzed against the appropriate buffer and assayed for
exoribonuclease activity. The other was TCA-precipitated and subjected to SDS-PAGE. A)
NTP-exoribonuclease activity assay. Protein samples were incubated with internally ([32P]-
labelled) UTP labeled RNA with and without NTPs and subjected to PEI-Cellulose TLC
under conditions in which the substrate (RNA) remains at the bottom of the chromatogram
and reaction products (UMP) migrates with the front. B) SDS-PAGE analysis of the purified
degradosome using silver staining. Proteins identified by mass spectrometry are indicated by
arrows. Molecular weight markers are indicated on the left.
Figure 3 The degradosome is associated with mitochondrial ribosomes. 100 µg of
mitochondrial extract proteins from the DSSTAP strain was separated into ribosomal and
soluble fractions by centrifugation through a sucrose cushion. All fractions (total extract,
soluble and ribosomal) were analyzed for the presence of ribosomal protein Nam9p and the
degradosome component Dss1p by western blotting using anti-TAP and anti-Nam9
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
30
antibodies. A)SDS-PAGE B) Western blot. Dss1 protein was present exclusively in the
ribosomal fraction.
Figure 4 Mitochondrial degradosome has RNA helicase activity which precedes RNA
hydrolysis by this complex. A) RNA helicase substrate in which the shorter RNA strand was
radioactively labeled at the 5’-end. B) Analysis of RNA helicase activity by the strand
displacement method. 100 fmoles of substrate were incubated in the presence of UTP with
increasing concentrations of TAP purified degradosome from the DSSTAP strain. Reaction
products were resolved in native acrylamide gels in a 10 min time course. The amount of
degradosome was estimated by SDS-PAGE and was about 0.1, 0.3 and 0.5 ng of Suv3p
protein, respectively. An aliquot of the duplex substrate was heat-denatured at 90°C for 5 min
and served as a marker for the position of single-stranded RNA (indicated as 90°C).
Figure 5 Change of tyrosine 814 to serine in Dss1p causes the loss of degradosome
exoribonuclease activity but has no influence on RNA helicase activity. The degradosome
purified from mitochondria from a W303/DSSSER W303/DSSTAP diploid yeast strain was
assayed for exoribonuclease and RNA helicase activities as described in Figs 2 and 4,
respectively. For both experiments the same amounts of purified degradosome were used.
Figure 6 Inactivation of degradosome components causes accumulation of precursors not
processed at the 3’- end but does not inhibit the processing completely.
A) Analysis of the 3’- ends of 21S rRNA, 15S rRNA and COB and COX1 mRNAs from
BWG (wt), ∆SUV3 and ∆DSS1 strains by S1-nuclease mapping. Reaction products for 15S
rRNA and COB and COX1 were resolved in 6% acrylamide 7M urea gels, and for 21S rRNA
in denaturing agarose formaldehyde gels. The position of mature RNA is indicated on each
gel.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
31
B) Analysis of 3’- end processing by Northern blot hybridization using oligonucleotide probes
specific for the mature RNA and 3’- precursor. The positions of the mature RNA and RNA
size markers are indicated on each gel.
Figure 7 Inactivation of degradosome components causes accumulation of precursors not
processed at the 5’-end. Analysis of 5’-end processing of RNA in ∆SUV3 and ∆DSS1 strains
by primer extension using oligonucleotides labeled with γ[32P]-labeled ATP. We analyzed all
specifically processed RNAs: 15S rRNA and COB, VAR1, ATP6/8 mRNAs. As a control we
used the unprocessed RNAs: 21S rRNA, COX1 mRNA and COX3 as an example of an
mRNA with its 5’-end released by tRNA excision:. An analysis was also performed for
tRNAVal. On each gel positions of the mature RNA and precursor are indicated. Numbers in
brackets indicate the difference in length of precursor and mature RNA.
Figure 8 The mitochondrial degradosome is not involved in tRNA maturation. Northern blot
analysis of mitochondrial tRNAs in wt and SUV3 and DSS1-disrupted strains. RNA was
resolved in 6% acrylamide/7M urea gels. Val and Thr tRNAs are transcribed from
complementary strands and their transcripts partially overlap. Single stranded DNA fragments
derived from the same PCR product were used as probes. Ala and Ile tRNAs are derived from
the same transcript, so a single stranded DNA fragment complementary to both tRNAs
including the transcribed intergenic region was used as a probe for hybridization. Single-
stranded DNA size markers are shown on the left.
Figure 9 There are no changes at the 3’- end or polyadenylation of mature COB mRNA in
∆SUV3 or ∆DSS1 strains. Analysis of the 3’- end of COB mRNA in ∆SUV3 and ∆DSS1
strains by ribonuclease T1 digestion and Northern blot hybridization. Two bands (143 and
223 nucleotides) which represent complete digestion products of mature mRNA and a
precursor not processed at the 3’- end are visible on autoradiograms. A) Mapping strategy. B)
Northern blot.
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from
Dmochowska, Michel Siep, Hans van der Spek, Les Grivell and Piotr P. StepienAndrzej Dziembowski, Jan Piwowarski, Rafal Hoser, Michal Minczuk, Aleksandrahelicase and RNase activities and the role in mitochondrial RNA metabolism
The yeast mitochondrial degradosome: its composition, interplay between RNA
published online November 7, 2002J. Biol. Chem.
10.1074/jbc.M208287200Access the most updated version of this article at doi:
Alerts:
When a correction for this article is posted•
When this article is cited•
to choose from all of JBC's e-mail alertsClick here
by guest on January 7, 2020http://w
ww
.jbc.org/D
ownloaded from