Identification of proteins associated with the yeast mitochondrial RNA polymerase by tandem affinity purification.

Abstract

The abundance of mitochondrial (mt) transcripts varies under different conditions, and is thought to depend upon rates of transcription initiation, transcription termination/attenuation and RNA processing/degradation. The requirement to maintain the balance between RNA synthesis and processing may involve coordination between these processes; however, little is known about factors that regulate the activity of mtRNA polymerase (mtRNAP). Recent attempts to identify mtRNAP-protein interactions in yeast by means of a generalized tandem affinity purification (TAP) protocol were not successful, most likely because they involved a C-terminal mtRNAP-TAP fusion (which is incompatible with mtRNAP function) and because of the use of whole-cell solubilization protocols that did not preserve the integrity of mt protein complexes. Based upon the structure of T7 RNAP (to which mtRNAPs show high sequence similarity), we identified positions in yeast mtRNAP that allow insertion of a small affinity tag, confirmed the mature N-terminus, constructed a functional N-terminal TAP-mtRNAP fusion, pulled down associated proteins, and identified them by LC-MS-MS. Among the proteins found in the pull-down were a DEAD-box protein (Mss116p) and an RNA-binding protein (Pet127p). Previous genetic experiments suggested a role for these proteins in linking transcription and RNA degradation, in that a defect in the mt degradadosome could be suppressed by overexpression of either of these proteins or, independently, by mutations in either mtRNAP or its initiation factor Mtf1p. Further, we found that Mss116p inhibits transcription by mtRNAP in vitro in a steady-state reaction. Our results support the hypothesis that Mss116p and Pet127p are involved in modulation of mtRNAP activity.

Full text

Page 1

Yeast
Yeast 2009; 26: 423–440.
Published online 17 June 2009 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/yea.1672
Research Article
Identification of proteins associated with the yeast
mitochondrial RNA polymerase by tandem affinity
purification
Dmitriy A. Markov1,* , Maria Savkina1,2, Michael Anikin1, Mark Del Campo3, Karen Ecker4,
Alan M. Lambowitz3, Jon P. De Gnore5and William T. McAllister1
1Departments of Cell Biology, University of Medicine and Dentistry of New Jersey, School of Osteopathic Medicine, Stratford, NJ, USA
2Graduate School of Biomedical Science, University of Medicine and Dentistry of New Jersey, Stratford, NJ, USA
3Institute for Cellular and Molecular Biology, Department of Chemistry and Biochemistry, and Section of Molecular Genetics and Microbiology,
School of Biological Sciences, University of Texas at Austin, TX, USA
4Summer Undergraduate Research Experience Program, University of Medicine and Dentistry of New Jersey, Stratford, NJ, USA
5Tufts University Core Facility, Department of Physiology, Tufts Medical School, Boston, MA, USA
*Correspondence to:
Dmitriy A. Markov, 2 Medical
Center Drive, Room B220,
Stratford, NJ 08084, USA.
E-mail: markovdm@umdnj.edu
Received: 24 December 2008
Accepted: 22 April 2009
Abstract
The abundance of mitochondrial (mt) transcripts varies under different conditions,
and is thought to depend upon rates of transcription initiation, transcription termina-
tion/attenuation and RNA processing/degradation. The requirement to maintain the
balance between RNA synthesis and processing may involve coordination between
these processes; however, little is known about factors that regulate the activity
of mtRNA polymerase (mtRNAP). Recent attempts to identify mtRNAP–protein
interactions in yeast by means of a generalized tandem affinity purification (TAP)
protocol were not successful, most likely because they involved a C-terminal mtR-
NAP–TAP fusion (which is incompatible with mtRNAP function) and because of
the use of whole-cell solubilization protocols that did not preserve the integrity of
mt protein complexes. Based upon the structure of T7 RNAP (to which mtRNAPs
show high sequence similarity), we identified positions in yeast mtRNAP that allow
insertion of a small affinity tag, confirmed the mature N-terminus, constructed a
functional N-terminal TAP–mtRNAP fusion, pulled down associated proteins, and
identified them by LC–MS–MS. Among the proteins found in the pull-down were
a DEAD-box protein (Mss116p) and an RNA-binding protein (Pet127p). Previous
genetic experiments suggested a role for these proteins in linking transcription and
RNA degradation, in that a defect in the mt degradadosome could be suppressed by
overexpression of either of these proteins or, independently, by mutations in either
mtRNAP or its initiation factor Mtf1p. Further, we found that Mss116p inhibits
transcription by mtRNAP in vitro in a steady-state reaction. Our results support the
hypothesis that Mss116p and Pet127p are involved in modulation of mtRNAP activity.
Copyright  2009 John Wiley & Sons, Ltd.
Keywords:
transcription; tandem affinity purification; TAP-tag; Mss116p; DEAD-box protein
mitochondria; Saccharomyces cerevisiae; Rpo41p; RNA polymerase;
Introduction
Mitochondria are self-replicating organelles that
possess their own genome and are thought to be
derived from an ancient bacterial endosymbiont
(Grey et al., 1999). They are responsible for pro-
duction of most energy in eukaryotic cells through
the process of oxidative phosphorylation (Saraste,
Copyright  2009 John Wiley & Sons, Ltd.

Page 2

424D. A. Markov et al.
1999). Reactive oxygen species (ROS) generated as
a by-product of oxidative phosphorylation damage
both cellular and organelle constituents (Cadenas
and Davies, 2000; Huang and Manton, 2004), and
it has been suggested that ROS-induced damage
to mtDNA is a major reason for the loss of mt
function during age-related pathologies (Hiona and
Leeuwenburgh, 2008; Wei et al., 1998). However,
it has recently been found that mtRNA transcripts
may be even more sensitive to oxidative damage
than DNA (Li et al., 2006), and it has been sug-
gested that balancing synthesis and degradation of
mt transcripts is required to preserve the quality of
mtRNAs and organelle integrity (Rogowska et al.,
2006).
Our understanding of protein factors that mod-
ulate mitochondrial transcription or couple RNA
synthesis with RNA processing and degradation
is limited, and is derived primarily from genetic
and biochemical studies in the budding yeast Sac-
charomyces cerevisiae (Calvo et al., 2006; Shadel,
2004). In yeast (as well as in higher eukaryotes)
mtRNA synthesis is driven by a small, single-
subunit DNA-dependent RNA polymerase that is
homologous to bacteriophage T7 RNA polymerase
(Masters et al., 1987). Only a limited number of
genes have been identified as suppressors of muta-
tions in the nuclear gene that encodes yeast mtR-
NAP (RPO41) and, with the exception of the initia-
tion factor Mtf1p (Mangus et al., 1994; Matsunaga
and Jaehning, 2004), none of their products have
been demonstrated to possess a definite function
in vitro. These include the inner membrane pro-
tein Sls1p, which has been suggested to anchor
the transcription complex to the inner mt mem-
brane through direct interaction with the mtRNAP
N-terminal domain (Bryan et al., 2002; Rodeheffer
and Shadel, 2003), and the matrix protein Nam1p
(Mtf2p), which serves as a linker between mtR-
NAP and a pool of other protein factors, includ-
ing Pet309p, Cbp1p and a large 900 kDa protein
complex with still unknown composition (Shadel,
2004).
An earlier approach towards identifying fac-
tors associated with mtRNAP involved passage of
human mt extracts over an affinity column having
immobilized mtRNAP (Wang et al., 2007). While
this identified a ribosomal protein (L12) that stim-
ulates mt transcription in vitro and suggested a
mechanism that may coordinate mt transcription
and translation, it did not uncover any known
or putative transcription factors. Separate attempts
to analyse the mt proteome by two-dimensional
gel electrophoresis also did not provide additional
information on the organization or composition
of the mt transcription apparatus (Balaban, 2006;
Reinders and Sickmann, 2007).
The above considerations called for an alter-
native means to identify proteins associated with
mtRNAP. The use of a tandem affinity purifica-
tion (TAP) protocol to isolate highly pure pro-
tein complexes, while minimizing procedures that
disrupt the association of their components, has
become increasingly popular for such applications
(Collins and Choudhary, 2008; Puig et al., 2001)
and has resulted in the characterization of a num-
ber of protein complexes, including the discovery
of 12 previously unidentified proteins associated
with yeast mt ribosomes (Saveanu et al., 2001).
The potential of the TAP method as a proteomic
tool led to the creation of a genome-wide yeast
library (to date the only eukaryotic library in which
all putative ORFs were tagged), allowing inves-
tigators to perform a comprehensive analysis of
the yeast ‘interactome’ in an attempt to reveal
the association of each TAP-tagged protein with
its functional counterpart(s) (Collins et al., 2007a,
2007b; Gavin et al., 2002; Krogan et al., 2006).
Unfortunately, this approach did not provide infor-
mation about the mtRNAP transcription apparatus.
This lack of success was likely due to a num-
ber of factors. Perhaps most importantly, the yeast
TAP library utilized C-terminal TAP tag fusions.
It is known that alterations at the C-terminus dis-
rupt the activity of T7-like RNAPs, and thus these
strains would have lost functional mitochondria.
TAP pull-downs of transcription factors known
to be associated with Rpo41p by genetic or bio-
chemical approaches (e.g. Mtf1p) also failed to
confirm their association with the core mtRNAP
(Table 1), possibly due to the use of whole cell
extracts, in which the integrity of the mitochon-
drial nucleoid was not maintained (Gavin et al.,
2002), and the use of high-salt cell solubiliza-
tion buffers that interfere with the stability of
DNA–protein complexes and the activity of T7-
like RNAPs (Winkley et al., 1985; Maria Savkina,
unpublished).
By modifying the TAP pull-down method to take
into account these considerations, we successfully
identified a number of proteins that were previously
known or, more importantly, previously unknown,
Copyright  2009 John Wiley & Sons, Ltd.
Yeast 2009; 26: 423–440.
DOI: 10.1002/yea

Page 3

Identification of yeast mtRNA polymerase-associated proteins by tandem affinity purification 425
Table 1. Results of global proteome search for proteins associated with the mt transcription apparatus
Proteins1
Function2
Associated proteins
by affinity
capture-MS3
Associated proteins
by genetic and
biochemical methods4
Rpo41p
Mt RNAP similar to phage T7
RNAP; requires a specificity factor
Mtf1p for promoter recognition
No data
Mtf1p, Mtf2p
(Nam1p), Sls1p,
Pet309p∗
Mtf1p
Mt RNAP specificity factor
Erg6p, Pzf1p, Ecm27p, Caf4p,
Sub2p, Tor1p, Snf5p, Tfb1p
Rpo41p
Mtf2p (Nam 1p)
Matrix protein coupling transcription,
RNA processing, and translation
Sls1p, Adh3p,
Rvs161p, Skt5p, Urh1p
Rpo41p, Pet309p
Sls1p
Inner membrane protein that
coordinates the gene expression;
may facilitate delivery of mRNA to
translation machinery
Mtf2p (Nam1p) Mtf2p (Nam1p),
Kar2p, Rpo41p
Pet309p∗
Inner membrane protein, specific
translational activator and stabilizer
for the COX1 mRNA
Cbp1p Mtf2p (Nam1p)
1Proteins that are known to interact with the transcription apparatus and2their reported functions (Shadel, 2004).3Associated proteins
identified by combination of affinity capture and mass spectrometry (MS) in a global screen of a library of S. cerevisiae strains with TAP-tagged
ORFs [Collins et al., 2007a, 2007b; Gavin et al., 2002; Krogan et al., 2006; data are available in the Biological General Repository for Interaction
Datasets (BioGRID) database (www. thebiogrid.com)]. Proteins that have not been shown to have mt function are underlined.
4Protein interactions established by genetic and biochemical methods.
∗Pet309p does not interact with Rpo41p directly; instead, it is associated with the core transcription enzyme through Mtf2p (Nam1p).
to be associated with the yeast mt transcription
apparatus. Our results support previous hypotheses
involving the need to balance rates of RNA syn-
thesis and degradation, and suggest mechanisms by
which mtRNA synthesis and degradation may be
coordinated.
Materials and methods
Yeast and bacterial strains, cloning,
and mutagenesis
All reagents were purchased from Sigma-Aldrich
unless otherwise specified. The heterozygotic
diploid Saccharomyces cerevisiae
BY4743 (MATa/MATα his3?1 leu2?0 met15?0
ura3?0) was purchased from Open Biosystems
Inc. Yeast cells were cultured on YEP media
with 1.5% agar at 30◦C unless otherwise speci-
fied. E. coli XL10-Gold competent cells (Strata-
gene) were used for all subcloning and site-
directed mutagenesis procedures. Yeast DNA was
isolated from strain BY4743 as described (Sam-
brook and Russell, 2001); a 5.8 kb DNA frag-
ment encoding the RPO41 gene (Masters et al.,
yeast strain
1987) was amplified from yeast chromosomal
DNA using an Expand Long Template PCR Sys-
tem kit (Roche Applied Science), subcloned into
pT7Blue vector, using a Perfectly Blunt cloning
kit (EMD Bioscience/Novagen), and recloned into
pRS316 and pRS315 yeast shuffling vectors (cen-
tromeric plasmids that exist as a single copy
in the cell and thus ensure the lack of overex-
pression; Sikorski and Hieter, 1989) to produce
pRS316–Rpo41p and pRS315–Rpo41p, respec-
tively. Competent cells were prepared and trans-
fected with the shuffling constructs as described
in BD Yeastmaker Yeast Transformation System 2
User Manual (BD Biosciences). Yeast sporulation,
dissection, and mating to confirm the haploid phe-
notype of the BY4743 derivative (MATa his3?1
leu2?0 met15?0 ura3?0)::pRS316-Rpo41p were
carried out by standard methods (Guthrie and Fink,
1991). Yeast containing shuffling constructs based
on pRS316 and pRS315 vectors were selected on
synthetic complete medium with Dextrose (SD)
lacking either uracil or leucine, respectively, and
prepared as described (Guthrie and Fink, 1991).
To select against the pRS316–Rpo41p vector,
5-FOA medium was prepared by adding 5-FOA
Copyright  2009 John Wiley & Sons, Ltd.
Yeast 2009; 26: 423–440.
DOI: 10.1002/yea

Page 4

426D. A. Markov et al.
to SD medium at a concentration of 1 mg/ml
(Zymo Research). RPO41 insertion mutants encod-
ing eight His residues downstream of Asp33,
Glu390, Asn773, Gly858, M928, Ala1001, and
the first 173 amino acids (aa) of a TAP tag
from pMK33–NTAP vector (Veraksa et al., 2005)
downstream of Asp33 at Rpo41p ORF were con-
structed using a mega-primer extension method
(Sarkar and Sommer, 1990) and standard molec-
ular cloning manipulations on the background of
the plasmid pRS315–Rpo41p described above. All
plasmids encoding mutant and wild-type RPO41
genes were purified using QiaQuick Miniprep kits
(Qiagen) and submitted for DNA sequencing to
Genewiz Inc.
Isolation of yeast mitochondria
Isolation of a crude fraction of yeast mitochon-
dria was performed as described (Pon and Schon,
2001), with slight modifications as noted below.
Cells were grown in a modified rich medium
previously used in studies of organelle trans-
port in isolated mitochondria (Daum et al., 1982;
Suissa and Schatz, 1982) and contained the fol-
lowing ingredients: 1% peptone, 1% yeast extract,
2% galactose, 0.1% KH2PO4, 0.12% (NH4)2SO4,
0.05% CaCl2, 0.06% MgCl2, 0.0003% FeCl3, and
10 mg/l tetracycline. After vigorous agitation in
4 l baffled Erlenmeyer flasks with 10 drops of
Antifoam Y-30, cells from mid-log 1.5 l cul-
ture (OD600nm= 5.0) were collected by centrifu-
gation, and spheroplasts were obtained by rocking
cells at 30◦C for 45 min with crude lyophilized
lyticase from Arthrobacter luteus (∼600 units/g
cells). Mitochondria were recovered from sphero-
plasts by osmotic lysis, with the addition of Pro-
tease Inhibitor Cocktail for use with fungal and
yeast extracts (Sigma-Aldrich) and several steps of
differential centrifugation, as described (Pon and
Schon, 2001), and were further purified by centrifu-
gation through a sucrose step gradient as described
(Meisinger et al., 2000). After two washes with
SEM buffer (Meisinger et al., 2000), organelles
were harvested at 12000 × g for 10 min in 1.5 ml
Eppendorf tubes. Pellets (∼200 µl/tube) were used
immediately or flash-frozen in an acetone/dry ice
mixture and stored at −70◦C for subsequent
analysis.
Isolation of mt nucleoids and in vitro
transcription
Mt nucleoids were isolated as originally described
(Miyakawa et al., 1987), with additional modifi-
cations (Meeusen et al., 1999; Miyakawa et al.,
2003). Purified nucleoids were recovered as a
DAPI-positive interphase between 15% and 40%
layers of a sucrose step gradient and resuspended
in nucleoid buffer (20 mM Hepes–KOH, pH 7.6,
0.5 M sucrose, 20 mM KCl, 2 mM EDTA, 0.8 mM
spermidine, 0.4 mM phenylmethanesulphonyl flu-
oride (PMSF), 7 mM β-mercaptoethanol). One µl
nucleoid suspension was diluted into 10 µl tran-
scription buffer (20 mM Hepes–KOH, pH 7.4,
25 mM sucrose, 20–200 mM KCl, 0.8 mM sper-
midine, 7 mM β-mercaptoethanol, 10 mM MgCl2,
300 µM CTP, GTP, ATP; 50 µM UTP; 1 mCi/ml
α-32P-UTP) without addition of any external DNA
template. Transcription reactions were carried out
for 20 min at 30◦C, with or without addition of
10 mg/ml α-amanitin as indicated, and stopped
by the addition of an equal volume of transcrip-
tion stop buffer containing 90% formamide, 50 mM
EDTA and 1 mg/ml bromophenol blue. After heat-
ing at 95◦C for 2 min, radioactively labelled RNA
products were resolved by electrophoresis in poly-
acrylamide denaturing gels (Sambrook et al., 1989)
containing 12% acrylamide:bis-acrylamide mix-
ture (19:1), 1× TBE buffer and 7 M urea, using a
Model S2 Sequencing Gel Electrophoresys Appa-
ratus (GibcoBRL) for 1.5 h at 90 W. Transcrip-
tion rate was quantified as amount of radioactiv-
ity incorporated into high molecular weight RNA
products, using a PhosphorImager Typhoon 9200
system (GE Healthcare) and ImageQuant software.
Tandem affinity purification of protein
from isolated yeast mitochondria
Tandem affinity purification was conducted based
on principles originally described (Rigaut et al.,
1999), except for solubilization procedure and
buffers, which were adopted from methods used
for isolation of intact mitochondria nucleoprotein
complexes (Meeusen et al., 1999; Miyakawa et al.,
1987, 2003), as noted above. All steps were per-
formed at 4◦C except for recovery of mitochondria
(as noted).
200 µl frozen mt pellet were defrosted at 30◦C,
washed once with 5 ml pre-warmed (30◦C) respi-
ration buffer (RB) containing 10 mM Hepes–KOH,
Copyright  2009 John Wiley & Sons, Ltd.
Yeast 2009; 26: 423–440.
DOI: 10.1002/yea

Page 5

Identification of yeast mtRNA polymerase-associated proteins by tandem affinity purification 427
pH 7.4, 25 mM sucrose, 75 mM sorbitol, 100 mM
KCl, 10 mM K2HPO4, 10 mM MgCl2, 2 mM ADP,
1 mM EDTA, 5 mM pyruvate, 5 mM malate, 1 mM
glutamate, 1 mg/ml BSA, 50 µM dNTPs, 300 µM
NTPs (adapted from Enriquez et al., 1994, 1996;
Micol et al., 1997) and then resuspended in pre-
warmed RB at 2 mg/ml of mitochondrial pro-
tein (1 mg is ∼50 µl of mt pellet). After gen-
tle agitation in 15 ml culture tubes for 30 min at
30◦C to allow organelles to respirate and restore
their metabolic activity, mitochondria were trans-
ferred to Eppendorf tubes, harvested by centrifu-
gation (12000 × g for 5 min) and immediately
resuspended in 3 ml ice-cold solubilization buffer
(20 mM Hepes–KOH, pH 7.4, 250 mM sucrose,
200 mM KCl, 1 mM EDTA, 1 mM spermidine,
7 mM β-mercaptoethanol, 0.2 mM PMSF). 75 µl
20% NP-40 were added slowly, drop-by-drop, to
the mt suspension to a final concentration of 0.5%.
After 15 min of incubation on ice, each lysate was
split into two 1.5 ml tubes and non-soluble compo-
nents were removed by centrifugation at 20000 ×
g for 10 min. After centrifugation was repeated,
the cleared supernatant was transferred into 1.5 ml
non-stick microtubes (Phenix Research Products
Inc.), which contained 50 µl pre-washed Protein A
Sepharose 4B FF beads (Sigma-Aldrich). After 1 h
of incubation at slow rotation to prevent bead sed-
imentation, the supernatant was separated from the
beads on a touch-spin table microfuge (1000 × g
maximum), using Micro Bio-Spin columns (Bio-
Rad), and transferred to a fresh non-stick tube with
25 µl pre-washed IgG Sepharose 6 FF beads (GE-
Healthcare/Amersham Bioscience). After 3 h rota-
tion the supernatant was discarded and the beads
were washed five times with 1.5 ml solubilization
buffer containing 0.1% NP-40 and lacking EDTA
(‘wash buffer’). Beads from two 1.5 ml tubes were
combined in 0.6 ml non-stick microtubes (Phenix
Research Products Inc.), resuspended in 500 µl
wash buffer containing 10 µl AcTEV protease (100
U; Invitrogen) and rotation was continued for 6 h.
The eluate was separated from the Sepharose beads
on the Micro Bio-Spin columns and transferred
to a fresh 0.5 ml tube containing 25 µl Calmod-
ulin–Sepharose 4B beads (GE Healthcare) that had
been pre-washed with wash buffer containing 2 mM
CaCl2. After 3 h rotation, the beads were settled,
washed five times with 0.5 ml wash buffer with
2 mM CaCl2, and twice with CaCl2-free buffer.
Protein was eluted from the beads with three con-
secutive washes, each using 50 µl wash buffer with
2 mM EGTA, on ice for 15 min. The washes were
combined, polypeptides were precipitated with ace-
tone/TCA at −20◦C for 3 h and separated by elec-
trophoresis in 4–12% Bis–Tris NuPAGE Novex
protein gels in an MES buffer system (Invitrogen).
The gels were stained with Coomassie Brilliant
Blue R-250 (Bio-Rad), and each lane containing
visible protein bands was sliced into 18 equal gel
samples, which were submitted for tryptic digestion
and LC–MS–MS analysis, as described below.
LC–MS–MS analysis
Excised bands were subjected to in-gel reduc-
tion, alkylation and enzymatic digestion (Roche
Applied Science, Indianapolis, IN, USA) in a
HEPA-filtered hood to reduce the keratin back-
ground. LC–MS–MS analysis was performed on
the in-gel digest extracts, using an Agilent 1100
binary pump (Santa Clara, CA, USA) directly cou-
pled to a mass spectrometer. 2–8 µl of sample
was injected onto the column using a LC Pack-
ings FAMOS autosampler (Sunnyvale, CA, USA).
Nanobore electrospray columns were constructed
from a 360 µm o.d., 75 µm i.d. fused silica cap-
illary with the column tip tapered to a 10 µm
opening (New Objective, Woburn, MA, USA). The
columns were packed with 200 ˚ A 5 µm C18 beads
(Michrom BioResources, Auburn, CA, USA), a
reverse-phase packing material, to a length of
10 cm. The pre-column was used to achieve a flow
rate of 320 nl/min. The mobile phase used for gra-
dient elution consisted of (a) 0.3% acetic acid and
99.7% water, and (b) 0.3% acetic acid and 99.7%
acetonitrile. Tandem mass spectra (LC–MS–MS)
were acquired on a Thermo Finnigan LTQ ion
trap mass spectrometer (Thermo Corp., San Jose,
CA, USA). Needle voltage was set to 3 kV, isola-
tion width was 3 Da, relative collision energy was
30% and dynamic exclusion was used to exclude
recurring ions. Ion signals above a predetermined
threshold automatically triggered the instrument to
switch from MS to MS–MS mode for generating
fragmentation spectra. The MS–MS spectra were
searched against the NCBI non-redundant protein
sequence database, using the SEQUEST computer
algorithm (Sadygov et al., 2004), to produce a list
of proteins identified in each sample. Allowance
Copyright  2009 John Wiley & Sons, Ltd.
Yeast 2009; 26: 423–440.
DOI: 10.1002/yea