Loss of Elongation Factor P Disrupts
Bacterial Outer Membrane Integrity
S. Betty Zou, Steven J. Hersch, Hervé Roy, J. Brad Wiggers,
Andrea S. Leung, Stephen Buranyi, Jinglin Lucy Xie, Kiley
Dare, Michael Ibba and William Wiley Navarre
J. Bacteriol. 2012, 194(2):413. DOI: 10.1128/JB.05864-11.
Published Ahead of Print 11 November 2011.
These include:
SUPPLEMENTAL MATERIAL
REFERENCES
CONTENT ALERTS
http://jb.asm.org/content/suppl/2011/12/19/194.2.413.DC1.html
This article cites 61 articles, 31 of which can be accessed free
at: http://jb.asm.org/content/194/2/413#ref-list-1
Receive: RSS Feeds, eTOCs, free email alerts (when new
articles cite this article), more»
Information about commercial reprint orders: http://jb.asm.org/site/misc/reprints.xhtml
To subscribe to to another ASM Journal go to: http://journals.asm.org/site/subscriptions/
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
Updated information and services can be found at:
http://jb.asm.org/content/194/2/413
Loss of Elongation Factor P Disrupts Bacterial Outer
Membrane Integrity
S. Betty Zou,a Steven J. Hersch,a Hervé Roy,b* J. Brad Wiggers,a Andrea S. Leung,a Stephen Buranyi,a Jinglin Lucy Xie,a Kiley Dare,b
Michael Ibba,b and William Wiley Navarrea
Department of Molecular Genetics, University of Toronto, Toronto, Ontario, Canada,a and Department of Microbiology, The Ohio State University, Columbus, Ohio, USAb
E
longation factor P (EF-P) is a translation factor that can facilitate the formation of the first peptide bond in in vitro translation assays (3, 27). It is a 20-kDa protein composed of three
beta-barrels and is similar in size and shape to a tRNA (31). A
crystal structure of Thermus thermophilus EF-P in complex with
the 70S ribosome indicates that EF-P occupies a unique position
between the peptidyl (P) and exit (E) sites to stimulate the formation of the first peptide bond (10). This supports earlier biochemical data suggesting that EF-P can enhance the rate of formation of
the first peptide bond, especially when the amino acid following
fMet has a small side chain (3, 20, 26). However, EF-P is not
absolutely necessary to reconstitute protein synthesis in vitro and
thus far, its biological function remains enigmatic.
EF-P shares significant homology with the eukaryotic translation factor eIF5A and its archaeal counterpart aIF5A, which both
contain the first two beta-barrel domains of EF-P (21). Previous
data support a role for eIF5A in translation elongation beyond
first peptide bond synthesis, although this has recently been challenged (28, 34, 49). Both aIF5A and eIF5A are posttranslationally
modified at a conserved lysyl residue with a spermidine-derived
hydroxylated 4-aminobutyl moiety to generate the unique amino
acid hypusine. Based on the structure of the EF-P/ribosome complex, the long basic side chain of the eIF5A hypusine residue is
predicted to insert into the peptidyl transferase center, where it
could interact with the amino acid linked to the P-site tRNA (10).
Knockout studies with yeast have established that the hypusine
modification is absolutely required for eIF5A activity (12, 45, 46).
Our recent work determined that two enzymes, PoxA and
YjeK, coordinately modify EF-P in a manner analogous to the
modification of aIF5A and eIF5A with hypusine (42, 48). Notably,
the modification of EF-P occurs at a lysyl residue (lysine 34) that
corresponds to the same position as the lysyl residue that is converted to hypusine in aIF5A and eIF5A. Two other laboratories
have independently corroborated aspects of these findings (6, 60).
YjeK belongs to the family of 2,3-lysine aminomutases, ironsulfur cluster-containing enzymes that catalyze the interconver-
0021-9193/12/$12.00
Journal of Bacteriology
p. 413– 425
sion of L-lysine and 3,6-diaminohexanoic acid (also known as
␤-lysine) (8). While PoxA bears close homology to the catalytic
domain of the class II lysyl-tRNA synthetase (LysRS) family of
enzymes that catalyze the addition of lysine to its cognate tRNALys,
a number of studies have failed to show that PoxA can aminoacylate a tRNA (1, 36, 37). PoxA instead catalyzes the ligation of
(R)-␤-lysine to the side chain of the conserved lysyl residue in
EF-P to yield an unusual lysyl-␤-lysine moiety (42, 48, 60). Salmonella poxA and yjeK mutants display nearly identical phenotypes, including increased resistance to S-nitrosoglutathione
(GSNO) and hypersusceptibility to a large number of unrelated
antimicrobial compounds (7, 42, 55). These mutant strains also
display markedly reduced virulence in a mouse model of infection
(36, 42). The mechanisms underlying the pleiotropic phenotypes
of poxA mutants have remained unresolved for over 2 decades.
EF-P was previously reported to be essential for viability in
Escherichia coli (4, 22, 35), a hypothesis that was consistent with
the observation that all bacteria, including those with a highly
reduced genome, encode at least one efp homolog (6, 24, 25, 40).
However, a systematic high-throughput attempt to disrupt every
nonessential gene in E. coli found that at least one strain of E. coli
could tolerate a disruption of efp and remain viable (5, 59). Furthermore, while a transposon insertion in the Agrobacterium homolog of efp, chvH, did not affect bacterial viability, the chvH
mutant displayed reduced virulence and expression of the critical
Received 22 July 2011 Accepted 28 October 2011
Published ahead of print 11 November 2011
Address correspondence to William Wiley Navarre, [email protected].
* Present address: Burnett School of Biomedical Sciences, University of Central
Florida, Orlando, Florida, USA.
Supplemental material for this article may be found at http://jb.asm.org/.
Copyright © 2012, American Society for Microbiology. All Rights Reserved.
doi:10.1128/JB.05864-11
jb.asm.org
413
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
Elongation factor P (EF-P) is posttranslationally modified at a conserved lysyl residue by the coordinated action of two enzymes,
PoxA and YjeK. We have previously established the importance of this modification in Salmonella stress resistance. Here we report that, like poxA and yjeK mutants, Salmonella strains lacking EF-P display increased susceptibility to hypoosmotic conditions, antibiotics, and detergents and enhanced resistance to the compound S-nitrosoglutathione. The susceptibility phenotypes
are largely explained by the enhanced membrane permeability of the efp mutant, which exhibits increased uptake of the hydrophobic dye 1-N-phenylnaphthylamine (NPN). Analysis of the membrane proteomes of wild-type and efp mutant Salmonella
strains reveals few changes, including the prominent overexpression of a single porin, KdgM, in the efp mutant outer membrane.
Removal of KdgM in the efp mutant background ameliorates the detergent, antibiotic, and osmosensitivity phenotypes and restores wild-type permeability to NPN. Our data support a role for EF-P in the translational regulation of a limited number of
proteins that, when perturbed, renders the cell susceptible to stress by the adventitious overexpression of an outer membrane
porin.
Zou et al.
virulence factor VirE2, as well as increased sensitivity to detergents
(47). Disruption of the efp gene in Acinetobacter also does not
affect bacterial viability, although fitness is adversely affected (17).
Given that poxA and yjeK are not essential for the viability of either
E. coli or Salmonella, these findings indicate that efp is a nonessential gene and may instead be required for the synthesis of a subset
of proteins necessary for stress tolerance. In the present study, the
previous observations regarding PoxA and YjeK were extended
with the study of EF-P.
MATERIALS AND METHODS
414
jb.asm.org
Journal of Bacteriology
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
Bacterial strains and plasmids. Wild-type Salmonella enterica serovar
Typhimurium 14028s was obtained from the laboratory of Samuel Miller
at the University of Washington. For all knockout mutants, the null alleles
were moved into a fresh 14028s strain background by transduction using
phage P22 HT105/1 int-201 prior to downstream analyses (50). The origins and properties of the strains used in this study are outlined in Table 1.
Null mutations in efp were constructed by the protocol of Datsenko
and Wanner using the red-gam recombinase as described previously (16).
Briefly, the efp gene of S. Typhimurium strain 14028s was replaced with a
kanamycin resistance gene cassette amplified from plasmid pKD4 using
primers WNp601 and WNp604. Plasmid pCP20, encoding FLP recombinase, was introduced into the efp mutant strain to flip out the FRT-flanked
kanamycin resistance cassette and generate nonpolar mutant strain
BZ070. A similar approach was used in the construction of the ompW and
kdgM (STM1131) null strains, except that primers SGBp13 and SGBp14
and primers BZp058 and BZp059 were used to amplify the kanamycin
resistance cassette from pKD4, respectively.
Plasmid pBZ029 (pWN525-efp) was constructed by PCR amplification of the efp open reading frame from S. Typhimurium 14028s genomic
DNA with primers BZp005 and BZp004. The resulting PCR product was
digested with BamHI and cloned into the corresponding BamHI site on
the pWN525 plasmid. PCR amplification of pBZ029 with primers BZp001
and BZp002 was used to mutagenize the lysine 34 residue to alanine,
generating pBZ031. Following BamHI digestion of pBZ031, the efpK34A
fragment was excised and cloned into the BamHI site of pTH19Kr to
generate pBZ051 (pEF-PK34A). PCR amplification of pBZ051 with primers BZp030 and BZp031 was used to mutagenize the alanine 34 residue to
lysine, thereby generating pBZ087 (pEF-P) containing the wild-type efp
sequence.
The isothermal cloning method of Gibson et al. was used to clone
kdgM into high-copy-number vector pBAD18 (23). Primers BZp068
and BZp069 were used to amplify the open reading frame of kdgM. The
resulting PCR product was cloned into the SmaI site of pBAD18 under
the control of an arabinose-inducible promoter. The resulting plasmid
(pBZ185 or pKdgM) was used for downstream complementation analyses.
MOPS minimal medium growth curve. Overnight cultures of each
strain in morpholinepropanesulfonic acid (MOPS) minimal medium
containing either 0.2% glucose or 0.2% glycerol were diluted to an optical
density at 600 nm (OD600) of 0.5. The diluted overnight cultures were
subcultured 1:100 into 50 ml of fresh MOPS minimal medium with the
same carbon source. Plating of the diluted cultures for CFU counting
confirmed that the initial inocula of the strains were comparable. Cells
were grown at 37°C with shaking for 12 h. Samples were taken every hour
and assessed for growth using a spectrophotometer to determine the
OD600.
Immunoblotting. Cell pellets were sonicated in a cell lysis buffer containing 9.32 M urea, 2.67 M thiourea, 0.04 M Tris, and 86.78 mM 3-[(3cholamidopropyl)-dimethylammonio]-1-propanesulfonate (CHAPS; pH
8.5) to extract total protein. Ten micrograms of total cell lysate was added
to an equal volume of 2⫻ SDS sample buffer, boiled at 95°C for 10 min,
and separated by SDS-PAGE on a 12% gel. The gel was transferred onto a
nitrocellulose membrane using the Bio-Rad semidry electrophoretic
transfer cell at 15 V for 1 h. Blocking was carried out overnight at 4°C in
TBST (1⫻ Tris-buffered saline, 0.05% Tween 20) with 5% milk. The blot
was probed with a rabbit primary antibody against EF-P (ProSci, Poway,
CA; 1:1,000 in TBST with 5% milk) for 1 h, followed by incubation with a
goat anti-rabbit secondary antibody conjugated with horseradish peroxidase (diluted 1:10,000 in TBST with 5% milk). As a loading control, DnaK
levels were probed using a mouse primary antibody (1:1,000 in TBST with
5% milk) followed by a goat anti-mouse secondary antibody conjugated
with horseradish peroxidase (1:10,000 in TBST with 5% milk). EF-P levels
were detected with an enhanced chemiluminescence kit (Thermo Scientific, Rockford, IL).
Detergent sensitivity assay. The MICs of various antibiotics were determined using a liquid culture assay. Overnight cultures of wild-type and
mutant Salmonella strains were diluted 1:1,000 in LB medium containing
2-fold serial dilutions of each detergent ranging from 100 ␮g/ml to 0.8
␮g/ml for hexadecyltrimethylammonium bromide (CTAB), dodecyltrimethylammonium bromide (DTAB), and sodium dodecyl sulfate (SDS)
and from 500 mM to 0.8 mM for lauryl sulfobetaine (LSB; Sigma, St.
Louis, MO). Growth was assessed visually after a 16-h incubation at 37°C
with shaking. For CTAB, DTAB, and LSB, the MIC was defined as the
lowest concentration at which growth was completely inhibited. For SDS,
the MIC50 was defined as the concentration at which the OD of the culture
was half of that of the no-detergent control.
GSNO disk diffusion assay. Susceptibility to GSNO was determined
by a disk diffusion method (18). Briefly, 15 ␮l of 500 mM GSNO was
added to a 0.25-in. paper disk placed over a lawn of 106 bacteria on M9
minimal agar with 0.2% glucose (25 ml of agar in a 10-cm petri dish). The
zone of growth inhibition after 16 to 24 h of incubation at 37°C was used
as a measure of susceptibility.
Antibiotic susceptibility assay. The MICs of gentamicin for wild-type
and mutant Salmonella strains were determined using a liquid culture
assay (40). Overnight cultures of Salmonella strains were diluted 100-fold
into LB medium containing 2-fold serial dilutions of gentamicin ranging
from 25 ␮g/ml to 0.8 ␮g/ml. Visible growth was measured by turbidity at
each concentration following 16 h of incubation at 37°C.
Soft agar assay. Salmonella strains were grown for 3 h to mid-log
phase. All cultures were synced to an OD600 of 0.1 and plated for CFU
counting. Comparable numbers of CFU of all of the strains tested were
recovered. Five microliters of the synced culture was spotted onto a 25-ml
soft agar plate (LB containing 0.35% agar). Plates were incubated right
side up at 37°C for 10 h. During the incubation, the diameter of the turbid
zone was measured every 2 h. For each trial, the value for each time point
was the average of three technical replicates.
Hypoosmolarity susceptibility assay. A single colony was streaked
onto 25-ml antibiotic medium 2 agar plates (AB2; Difco, Detroit, MI),
AB2 supplemented with 0.3 M sucrose, or AB2 supplemented with 0.3 M
NaCl. Plates were incubated for 16 h at 37°C, followed by visual assessment of colony size.
Inner and outer membrane extraction. Membrane proteins were extracted from Salmonella using a modification of a previously established
protocol (29). Briefly, a 250-ml culture of each strain was grown to an
OD600 of 1.5 and pelleted by centrifugation. The pellet was resuspended in
20 ml of 20% (wt/wt) sucrose in 0.01 M HEPES (pH 7.8). This and all
subsequent steps were carried out at 4°C. Lysozyme was added to a final
concentration of 0.1 mg/ml. Twenty milliliters of 0.01 M HEPES (pH
7.8)–1 mM EDTA was added with gentle stirring over a 2-min period. The
spheroplasts were collected by centrifugation at 11,000 ⫻ g for 15 min.
The spheroplast pellet was resuspended in 20 ml of 0.01 M HEPES (pH
7.8)– 0.05 mM EDTA (HE buffer) and passed twice through a French
press at 1,000 lb/in2. Membrane proteins were extracted by first layering
the lysed spheroplasts over a two-step sucrose gradient composed of 2 ml
of 55% (wt/wt) sucrose and 0.5 ml of 5% (wt/wt) sucrose in HE buffer.
Gradients were centrifuged in a Beckman SW41 rotor at 35,000 rpm for 3
h. The membrane band at the 55% sucrose interface was collected, diluted
1:1 in HE buffer, and passed through a 27-gauge needle three times. Separation of inner and outer membranes was achieved by layering the di-
Outer Membrane Misregulation in efp Mutants
TABLE 1 Strains, plasmids, and oligonucleotides used in this study
Genotype or relevant characteristics
Reference
Strains
14028s
WN353
WN354
WN356
WN409
WN412
BZ066
BZ070
BZ072
BZ087
BZ093
BZ108
BZ109
BZ152
BZ169
BZ171
BZ186
BZ171
BZ193
BZ195
BZ200
BZ201
Wild-type S. enterica serovar Typhimurium
14028s ⌬poxA::Cm
14028s ⌬yjeK::Cm
14028s ⌬poxA::Km ⌬yjeK::Cm
14028s ⌬poxA::Cm/pWN403
14028s ⌬yjeK::Cm/pWN404
14028s ⌬efp::Km
14028s ⌬efp
14028s ⌬efp/pBZ051
14028s ⌬efp/pBZ087
14028s/pBAD18
14028s ⌬efp/pBZ087/pWN403
14028s ⌬efp/pBZ087/pWN404
14028s ⌬efp ⌬ompW::Km
14028s ⌬kdgM::Km
14028s ⌬efp ⌬kdgM::Km
14028s/pBZ185
14028s ⌬efp ⌬kdgM::Km/pBZ185
14028s ⌬efp/pBAD18
14028s ⌬efp ⌬kdgM::Km/pBAD18
14028s ⌬efp::Km ⌬poxA::Cm
14028s ⌬efp ⌬ompW::Km/pBAD18
ATCC
42
42
42
42
42
This work
This work
This work
This work
This work
This work
This work
This work
This work
This work
This work
This work
This work
This work
This work
Expression vector that places cloned genes under the control of the arabinose-inducible PBAD promoter
Also known as pWN403; has a 1,112-nt fragment encoding the poxA open reading frame cloned into the KpnI and SphI sites
of pBAD18
Also known as pWN404; has a 1,238-nt fragment containing the yjeK open reading frame cloned into the EcoRI and XbaI sites
of pBAD18
pir-deficient plasmid harboring the kanamycin resistance cassette used for gene deletion
Temperature-sensitive plasmid containing the FLP recombinase gene
pir-deficient cloning vector containing a chloramphenicol resistance selectable marker
Has a 1,119-bp fragment encoding the efp open reading frame cloned into the BamHI site of pWN525
Has a 1,119-bp fragment encoding the efp open reading frame containing the K34A mutation cloned into the BamHI site of
pWN525
Low-copy-number cloning vector containing a kanamycin resistance selectable marker
Also known as pBZ051; has a 1,119-bp fragment encoding the efp open reading frame containing the K34A mutation cloned
into the BamHI site of pTH19Kr
Also known as pBZ087; has a 1,119-bp fragment encoding the efp open reading frame cloned into the BamHI site of pTH19Kr
Also known as pBZ185; has an 838-bp fragment encoding the ksgM open reading frame cloned into the SmaI site of pBAD18
30
42
Plasmids
pBAD18
pPoxA
pYjeK
pKD4
pCP20
pWN525
pBZ029
pBZ031
pTH19Kr
pEFPK34A
pEFP
pKdgM
Oligonucleotides
BZp001
BZp002
BZp004
BZp005
BZp030
BZp031
WNp601
WNp604
BZp058
BZp059
BZp068
BZp069
SGBp13
SGBp14
WNp555 (16S)
WNp556 (16S)
WNp580 (poxA)
WNp581 (poxA)
WNp255 (yjeK)
WNp256 (yjeK)
WNp582 (efp)
WNp583 (efp)
WNp233 (gyrB)
WNp234 (gyrB)
ALp01 (kdgM)
ALp02 (kdgM)
42
16
13
41
This work
This work
33
This work
This work
This work
5= GTGAATTCGTGAAACCGGGTGCGGGCCAGGCGTTTG 3=
5= CAAACGCCTGGCCCGCACCCGGTTTCACGAATTCAC 3=
5= ATATGGATCCGCGTTTCATCATTGAGGCGCGAAA 3=
5= TCTAGGATCCTGCCTTGATTCTCTGCATACGGGA 3=
5= GTGAATTCGTGAAACCGGGTAAAGGCCAGGCGTTTG 3=
5= CAAACGCCTGGCCTTTACCCGGTTTCACGAATTCAC 3=
5= GCGCCATTTTGTGGCTTAGCTACCAGTTAACAATTTCAGAGTGTAGGCTGGAGCTGCTTC 3=
5= ATGGCGACTTACTATAGCAACGATTTTCGTTCCGGTCTTAGTGTAGGCTGGAGCTGCTTC 3=
5= GGAGTATTTTATGAAAATCAACAAATATCTTCTGGGTATGGTGTAGGCTGGAGCTGCTTC 3=
5= TTAACCTTTCTCAAAAGTGGTATTGTAATCCAACGCGATACATATGAATATCCTCCTTAG 3=
5= TACCCGTTTTTTTGGGCTAGCGAATTCGAGCTCGGTACCCTTTTTTAAATACAAAAATAA 3=
5= CCAAGCTTGCATGCCTGCAGGTCGACTCTAGAGGATCCCCACTCAAGACAAAGTCTAACG 3=
5= ATATTCATCACGCTTTTATAACCATAACGATGGAGCGGGTGTGTAGGCTGGAGCTGCTTC 3=
5= CCCGCCTGGCGGGGTCTTTTTTCGTCACAACATCACGCGACATATGAATATCCTCCTTAG 3=
5= CGGGGAGGAAGGTGTTGTG 3=
5= GAGCCCGGGGATTTCACATC 3=
5= TGCTGCAACTGTTGTTCACGATGG 3=
5= GTGGAAACCATTCGCCAGCTCAAT 3=
5= CGTCACTTCCCGTATGCAGAGAAT 3=
5= GACGTGCTTAATGGCTTCCAGTTG 3=
5= TACAACGACGGTGAGTTCTGGCAT 3=
5= AAAGTTCGGCGGAGTGACAGAGAT 3=
5= GATGGGTTTTCCAGCAGGTATTC 3=
5= AGGTCTGATTGCGGTGGTTTC 3=
5= ACAAGCGGCAACCCTTGATTATCG 3=
5= CCGATTTAACGCTGGCATCGACAT 3=
January 2012 Volume 194 Number 2
jb.asm.org 415
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
Strain, plasmid, or
oligonucleotide
Zou et al.
416
jb.asm.org
Reverse transcription was performed using the iScript cDNA synthesis kit
(Bio-Rad) using random hexamer primers. The cDNA generated was used
for quantitative real-time PCR analysis using iQ SYBR green Mix (BioRad) according to the manufacturer’s protocol. Transcripts of efp, poxA,
yjeK, kdgM (STM1131), the 16S rRNA gene, and gyrB were analyzed by
using the primers listed in Table 1 according to the gene name. The transcript of the 16S rRNA gene and gyrB was used as an internal standard for
normalization for the salt assay and the kdgM expression assay, respectively.
Two-dimensional gel electrophoresis. For each strain, cultures were
grown to early stationary phase (OD600 of 1.5) in LB medium at 37°C.
Cells were pelleted by centrifugation, and total protein was extracted by
sonication. Proteins were precipitated with the Ettan 2-D Cleanup Kit
(GE Healthcare, Piscataway, NJ) according to the manufacturer’s instructions. For isoelectric focusing, 200 ␮g of each protein sample was mixed
with an equal volume of 2⫻ sample buffer (7 M urea, 2 M thiourea, 0.065
M CHAPS [Sigma, St. Louis, MO], 0.13 M dithiothreitol [DTT], and
0.02% IPG buffer, pH 3 to 11 NL (GE Healthcare, Mississauga, Ontario,
Canada]), and incubated on ice for 10 min. Samples were brought to a
final volume of 0.45 ml with rehydration solution (7 M urea, 2 M thiourea,
0.065 M CHAPS, 0.018 M DTT, 0.02% IPG buffer pH 3-11 NL [GE
Healthcare], 0.001% bromophenol blue) and used to passively rehydrate
Immobiline Drystrips (pH 3-11 NL, 24 cm; GE Healthcare) for 14 h at
room temperature. Rehydrated strips were equilibrated and focused under the following conditions: 300 V for 2 h, 1,000 V for 2 h, a gradient of
5,000 V/h to 20,000 V·h, and a gradient of 8,000 V/h to 60,000 V·h.
Second-dimension SDS-PAGE was carried out on 8 to 16% Trisglycine gels (Jule Biotechnologies, Mildford, CT). Gels were run at 80
V, 10 mA/gel, and 1 W/gel for 1 h and then at 500 V, 38 mA/gel, and 13
W/gel until the dye front was approximately 1 cm from the bottom of
the gel.
For colloidal Coomassie staining, gels were fixed overnight in a solution containing 50% ethanol, 3% H3PO4, and 47% water. Following fixation, gels were washed three times with Milli-Q water and allowed to
equilibrate in Neuhoff’s solution [16% (NH4)2SO4, 25% methanol, 5%
H3PO4, 54% water] for 1 h. Coomassie brilliant blue G250 (EMD,
Gibbstown, NJ) was added to each staining tray to a final concentration of
1 g/liter of Neuhoff’s solution. Gels were stained for 3 to 4 days and stored
in 5% acetic acid at 4°C.
RESULTS
Cells lacking EF-P have phenotypes similar to those lacking
PoxA or YjeK. To assess whether efp is essential for viability in
Salmonella, lambda red recombination was used to generate a
complete deletion of the efp gene in S. enterica serovar Typhimurium strain 14028s. The E. coli efp knockout strain was obtained
from the Keio collection and verified as being correctly constructed, a finding that has been corroborated by an independent
study (60). In both species, mutants lacking EF-P exhibited
slightly reduced growth rates compared to those of their wild-type
counterparts when grown in a standard formulation of LB medium (data not shown). In MOPS minimal medium containing
either 0.2% glucose or glycerol, the growth rate of the Salmonella
efp mutant was approximately half that of wild-type Salmonella
(see Fig. S1 in the supplemental material).
If the posttranslational modification catalyzed by PoxA and
YjeK is essential for EF-P activity, the phenotypes of the yjeK,
poxA, and efp mutants would be expected to be similar (Fig. 1A).
Salmonella poxA and yjeK mutants are hypersensitive to a variety
of growth inhibitors, including antibiotics, dyes, detergents, and
sulfometuron methyl, an inhibitor of acetolactate synthase (42,
55). Based on previous findings that poxA and yjeK mutants were
more susceptible to the aminoglycoside gentamicin, the suscepti-
Journal of Bacteriology
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
luted membrane solution onto a seven-step gradient (wt/wt sucrose concentrations are given in parentheses) of 0.4 ml (60%), 0.9 ml (55%), 2.2 ml
(50%), 2.2 ml (45%) 2.2 ml (40%), 1.3 ml (35%), and 0.4 ml (30%). All
solutions were made in HE buffer. Gradients were centrifuged in a Beckman SW41 rotor at 36,000 rpm for 20 h. Inner membrane and outer
membrane protein bands were collected from the 40% and 55% interfaces, respectively. Membrane fractions were assayed for NADH oxidase
activity as previously described (44).
KdgM overexpression. Strains containing either pKdgM or an empty
pBAD18 plasmid were grown to mid-log phase (OD600 of 0.6) and induced for 1 h with 0.2% L-arabinose. Following induction, LSB was added
to a final concentration of 20 mM. Samples were taken prior to and 1 h
after the addition of LSB and plated for CFU counting. LSB sensitivity was
determined as the percentage of colonies recovered following LSB exposure relative to the CFU count before LSB treatment.
1-N-Phenylnaphthylamine (NPN) accumulation assay. The NPN
accumulation assay protocol was adapted from others previously established to measure the intrinsic permeability of bacterial membranes (32,
38, 53). LB broth containing appropriate antibiotics was inoculated from
overnight cultures of the test strains. The cells were grown to early stationary phase (10 h), centrifuged (13,000 rpm for 1 min), and washed twice in
the assay buffer (5 mM HEPES pH 7.2, 137 mM NaCl). The cells were
resuspended in buffer, and the OD600 was adjusted to 1.0. Fractions from
these suspensions were diluted and plated in triplicate to acquire CFU
counts. One-hundred-microliter volumes were also loaded into the wells
of a black 96-well fluorescence microplate (Greiner Bio-One). Controls
containing only buffer were included, and three technical replicates were
conducted. A 20 ␮M solution of NPN (Sigma-Aldrich) was prepared in
the assay buffer just prior to experimentation, and 100-␮l volumes were
added to appropriate wells of the microplate, yielding a final bacterial
OD600 of 0.5 and a final NPN concentration of 10 ␮M. Controls were
included in which buffer was added instead of dye. Immediately after the
addition of the dye, the plate was inserted into a Tecan Infinite M200
microplate reader (a delay of approximately 15 s) and read at excitation
and emission wavelengths of 355 nm and 402 nm, respectively. Readings
were taken every 25 s for 10 min. Analysis of the fluorescence values was
conducted using Microsoft Excel. Background fluorescence (NPN in buffer only) was subtracted from the raw values, and the result was divided by
the number of CFU in the corresponding sample. Finally, the fluorescence
value of wild-type Salmonella at time zero was defined as 100% and all
other values were normalized accordingly.
Ethidium bromide (EtBr) accumulation assay. EtBr uptake was measured very similarly to NPN uptake, with several differences: The buffer
used was 50 mM KH2PO4 (pH 7.0)–137 mM NaCl. As well, cells were
diluted to an OD600 of 0.4 (final OD600 of 0.2 after addition of EtBr). The
concentration of the stock EtBr solution was 12 ␮M (final concentration
of 6 ␮M after addition to cells). EtBr fluorescence was read at excitation
and emission wavelengths of 545 nm and 600 nm, respectively. Finally,
since wild-type fluorescence values were at background levels, the value
for the wild type at time zero was zero for some trials. The values could
therefore not be represented as a percentage of the wild type at time zero.
They were instead presented in arbitrary units (data with the background
subtracted divided by the number of CFU in the sample).
Quantitative PCR of gene expression. For gene expression under various salt conditions, wild-type Salmonella 14028s was grown to an OD600
of 0.3. Cell pellets were then washed and resuspended in fresh, prewarmed
LB medium containing 0, 170, or 300 mM NaCl and grown for an additional 30 min at 37°C. For kdgM gene expression, strains were grown in LB
medium at 37°C to early stationary phase (OD600 of 1.5). To assess kdgM
expression from the complementation plasmid pKdgM, the wild-type
strain containing pKdgM was grown to an OD600 of 0.6 and induced with
0.2% L-arabinose for 1 h. For all experiments, 1.5 ml of each sample was
pelleted in RNA Protect Bacterial Reagent (Qiagen) according to the manufacturer’s instructions and stored at ⫺80°C. Subsequent RNA preparations were performed using the Aurum Total RNA Mini kit (Bio-Rad).
Outer Membrane Misregulation in efp Mutants
bility of efp mutants to the drug was assessed (42). Salmonella efp
mutants were slightly more susceptible to gentamicin than either
the yjeK or the poxA mutant (Fig. 1B). Sensitivity to gentamicin
was only partially rescued by complementation with efp on a lowcopy-number vector and was exacerbated by the expression of
efpK34A in trans.
Salmonella poxA and yjeK mutants also display enhanced resistance to the nitrosating agent GSNO for reasons that are currently
unclear (42). The Salmonella efp mutant was assessed for GSNO
sensitivity by a disk diffusion assay on M9 minimal medium agar
(Fig. 1C). As previously observed, poxA and yjeK mutants were
found to have smaller zones of growth inhibition than the wildtype strain. The efp mutant was more resistant to GSNO than
January 2012 Volume 194 Number 2
either the poxA or the yjeK mutant. A poxA efp double mutant was
not more sensitive to gentamicin or more resistant to GSNO than
a poxA or efp single mutant, indicating that all three genes (poxA,
yjeK, and efp) display epistasis, which further supports the biochemical and genetic data indicating that PoxA, YjeK, and EF-P
function in the same pathway (42, 48, 60).
Given the broad phenotypic overlap with poxA and yjeK, a
broader set of Salmonella efp mutant phenotypes were analyzed
using a Biolog phenotype microarray (see Table S1 in the supplemental material). In this assay, wild-type and mutant strains are
cultivated in microplates with each well containing a different
nutrient or growth inhibitor condition. A relative measure of fitness under each condition is quantified colorimetrically by the
jb.asm.org 417
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
FIG 1 Salmonella poxA, yjeK, and efp mutants display epistasis and are hypersusceptible to gentamicin but resistant to GSNO. (A) Schematic representation of
the PoxA- and YjeK-mediated posttranslational modification of EF-P. (B) MIC of gentamicin for Salmonella wild-type, mutant, and complemented strains as
measured by growth in LB containing 2-fold dilutions of gentamicin ranging from 25 to 0.8 ␮g/ml. Values reported represent the lowest concentrations at which
growth was inhibited in three independent experiments. (C) Results from a disk diffusion assay to measure Salmonella resistance to GSNO. A paper disk was
soaked with 15 ␮l of 500 mM GSNO and placed on a lawn of 106 salmonellae. The diameter of the zone of growth inhibition was measured after 16 to 24 h of
growth. Error bars represent the standard error of the mean of at least three independent experiments.
Zou et al.
derivatives, and complemented strains were plated on AB2 agar or AB2 agar supplemented with either 0.3 M sodium chloride or 0.3 M sucrose as indicated.
Representative plates after 16 h of growth are shown. The AB2 (low-salt) plate was photographed a second time after 20 h to show the emergence of efp mutant
microcolonies (white arrowheads) compared to the colonies formed by the wild-type strain (bottom left panel). (B) Immunoblot analysis of 10 ␮g total cell lysate
from wild-type Salmonella grown under different osmolarity conditions. DnaK was used as a loading control.
turnover of a tetrazolium redox dye that reports total respiration
over the course of the assay.
The phenotype microarray examined approximately 1,900 different conditions and revealed that Salmonella efp mutants display
highly pleiotropic phenotypes that were nearly identical to those
previously observed for poxA and yjeK mutants. Of particular interest was the large number of unrelated stress conditions under
which the efp mutant grew poorly, including high and low pHs
and antibiotics from different pharmacological classes and with
different mechanisms of action. In contrast, efp, poxA, and yjeK
mutants all displayed increased respiration when utilizing certain
nitrogen, sulfur, and phosphate sources. This growth advantage
was especially pronounced when amino acids and dipeptides containing methionine or branched-chain amino acids were present
as nitrogen sources. Like poxA and yjeK mutants, efp mutants were
unable to utilize ␥-glutamyl-glycine as a nitrogen source. These
results highlight the highly pleiotropic and overlapping phenotypes of Salmonella lacking poxA, yjeK, or efp. Furthermore, the
modification of EF-P by PoxA and YjeK appears to be essential for
its activity in vivo.
E. coli and S. Typhimurium strains deficient in poxA, yjeK,
and efp display poor growth under hypoosmolarity conditions.
Salmonella poxA mutants had previously been reported to display
418
jb.asm.org
poorer growth on AB2 agar (6 g peptone, 1.5 g beef extract, and 3
g yeast extract per liter) than on LB (10 g tryptone, 5 g yeast
extract, and 10 g NaCl per liter), where their growth is only slightly
impaired compared to that of the wild type (36). This phenotype
was confirmed for Salmonella poxA mutants, as well as yjeK and
efp mutants, which also displayed poor growth on AB2 agar (Fig.
2A, top panel). The growth defect was the most pronounced for
the efp mutant, which required 20 h of growth at 37°C to produce
visible microcolonies. Given the compositions of the AB2 and LB
media, it seemed likely that the observed growth defect was due to
the absence of sodium chloride in AB2 agar. Consistent with this
hypothesis, growth was restored to nearly wild-type levels on AB2
agar supplemented with 300 mM NaCl (Fig. 2A). Accordingly, all
three mutants also displayed significantly reduced growth on LB
medium prepared without NaCl (LBNS; data not shown). Growth
of mutant Salmonella on AB2 or LBNS could also be restored by
the addition of 300 mM sucrose, indicating that the growth defect
was due to the low osmolarity of the medium and not due to the
lack of either sodium or chloride per se (Fig. 2A, bottom right).
Similar to what was observed for Salmonella, the poxA, yjeK, and
efp E. coli mutants also displayed very poor growth on lowosmolarity medium (data not shown).
Growth on AB2 or LBNS agar was also completely restored
Journal of Bacteriology
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
FIG 2 Salmonella strains harboring mutations in efp, poxA, or yjeK display poor growth on low-osmolarity medium. (A) Wild-type Salmonella 14028s, mutant
Outer Membrane Misregulation in efp Mutants
when the poxA and yjeK mutants were complemented with their
corresponding open reading frames in trans on the high-copy,
arabinose-inducible vector pBAD18. Notably, full complementation was apparent even in the absence of arabinose, suggesting that
low levels of PoxA and YjeK expression were sufficient to restore
wild-type growth. When efp was cloned into a high-copy-number
plasmid, the construct not only failed to restore growth on AB2
agar, but it also exacerbated the growth defect (data not shown).
Partial rescue was achieved by cloning the efp open reading frame
into the low-copy-number vector pTH19Kr (pSC101 replicon,
⬃5 copies per cell). A mutant plasmid, pEF-P(K34A), was also
generated using pTH19Kr wherein the conserved lysine residue in
EF-P was mutated to an alanine, such that the resulting protein
could no longer be modified by PoxA and YjeK. Complementation analysis revealed that the pEF-P(K34A) plasmid could not
rescue the growth defect of the Salmonella efp mutant and instead
exacerbated the poor-growth phenotype of these mutants on AB2
agar (Fig. 2A, top middle). Immunoblot analysis revealed that
EF-P levels in the complemented strains were significantly higher
than EF-P levels in the wild-type strain (see Fig. S2A in the supplemental material).
The observation that overexpression of either EF-P or EFP(K34A) exacerbated the phenotypes of an efp null mutant suggests that when overexpressed, there is an accumulation of unmodified EF-P that may act in a dominant negative fashion. The
accumulation of unmodified EF-P could conceivably be caused by
limiting amounts of YjeK. The yjeK and efp genes are divergently
transcribed with a 40-bp intergenic region between the two open
reading frames. Although the yjeK and efp promoters have not
been mapped, it is plausible that the promoter of yjeK may overlap
the efp open reading frame and vice versa such that knockout of
one gene may have affected the expression of the other. In the yjeK
null mutant used in this study, 200 nucleotides (nt) of the 5= end of
yjeK was left intact to prevent disruption of efp expression. Robust
expression of EF-P was detected by immunoblot analysis in yjeK
mutants, confirming that efp expression was unaffected by the
inactivation of yjeK (data not shown). However, it is possible that
the more extensive deletion of efp reduced the expression of yjeK
January 2012 Volume 194 Number 2
through secondary effects on the yjeK promoter. To address this
possibility, efp mutant strains complemented with pEF-P or pEFP(K34A) were supplemented with the compatible pYjeK plasmid
or pPoxA, as a control. While pPoxA did not confer additional
restoration of the wild-type phenotype, pYjeK produced nearly
full complementation of the ⌬efp pEF-P strain in both the hypoosmolarity and gentamicin sensitivity assays (see Fig. S2B and C in
the supplemental material). The enhanced complementation by
pYjeK was not evident in the strain harboring the pEF-P(K34A)
plasmid, demonstrating epistasis between efp and yjeK. These
findings indicate that the effects of YjeK occur via its role in the
modification of EF-P and not through a secondary EF-Pindependent mechanism. Where appropriate, the complementation of the efp mutant was carried out with both pYjeK and pEF-P
in subsequent assays.
Given that EF-P is required for optimal growth under lowosmolarity conditions, it is possible that EF-P participates in an
osmoadaptive response and may be differentially expressed in response to various salt concentrations. Levels of EF-P in wild-type
Salmonella were assessed by immunoblot analysis after growth in
0, 170, or 300 mM sodium chloride (Fig. 2B). No significant differences in the level of endogenous EF-P could be observed under
the growth conditions tested. poxA and yjeK transcript levels were
also largely unaffected by changes in osmolarity (see Fig. S3 in the
supplemental material). These findings indicate that poxA, yjeK,
and efp are unlikely to constitute an adaptive system that responds
to changes in osmolarity.
Salmonella poxA, yjeK, and efp mutants have a migration defect. Bearson et al. (7) reported that Salmonella poxA mutants
displayed a migration defect, which may be linked to their observation that expression of the flagellar subunit FliC was also decreased in the poxA mutant. Analysis of the Salmonella efp, poxA,
and yjeK mutants confirmed that these strains also displayed significant migration defects compared to the wild-type strain but
that the severity of the phenotype varied between strains (Fig. 3).
The yjeK mutant displayed the mildest phenotype, while the efp
mutant had the most severe phenotype. Complementation of the
jb.asm.org 419
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
FIG 3 Salmonella poxA, yjeK, or efp mutants exhibit migration defects. poxA, yjeK, and efp mutants have a lower rate of migration over time. Approximately 4 ⫻
108 bacteria were spotted onto a soft agar plate (LB with 0.35% agar). Error bars represent the standard error of the mean for four independent experiments.
Zou et al.
⌬efp mutant with pEF-P and pYjeK restored migration to wildtype levels.
Salmonella efp mutants are sensitive to detergent. poxA mutants are markedly more sensitive to LSB than wild-type Salmonella (7, 42). LSB is a zwitterionic detergent with a dodecyl hydrocarbon chain identical to that of SDS. To explore the underlying
basis of the detergent sensitivity phenotype, the abilities of the efp,
poxA, and yjeK mutants to withstand a panel of related detergents
were tested. Among the detergents tested, only LSB is zwitterionic.
CTAB and DTAB are cationic, whereas SDS is anionic. While
wild-type salmonellae were able to grow at and tolerate up to 500
mM LSB, the growth of the Salmonella efp mutant was inhibited at
3.125 mM (Fig. 4). Further exploration of this phenotype revealed
that while efp mutants display general defects in detergent sensitivity, a dramatic difference in susceptibility was observed only
with LSB and not with any of the nonzwitterionic detergents
tested. SDS and DTAB are both smaller than LSB and have a critical micellar concentration (CMC) that is slightly higher than that
of LSB, while CTAB is similar in total mass to LSB and has a
significantly lower CMC. The efp mutant strains were 2-fold more
susceptible to CTAB and DTAB (Fig. 4). The mutant strains were
also more susceptible to SDS, although increasing concentrations
led to a gradual effect on growth, unlike the stark differences observed over narrow concentration ranges for the other detergents.
These findings indicate that the zwitterionic nature of LSB, rather
than its CMC or molecular weight, is a key determinant of its
potent antimicrobial activity against the poxA, yjeK, and efp mutant strains.
Salmonella efp mutants are selectively more permeable to
NPN. The inability of poxA, yjeK, and efp mutants to grow on
low-osmolarity medium, their migration defects, and their hypersensitivity to a broad range of compounds, including detergents,
420
jb.asm.org
suggest that these mutants may have an altered cell envelope. A
defect in overall membrane permeability would provide a parsimonious explanation for the observation that these mutants display increased susceptibility to various pharmacologically unrelated classes of antibiotics.
The permeability of the Salmonella efp mutant strain was assessed by measuring the uptake of the fluorescent probes EtBr and
NPN, two dyes which fluoresce upon entering the cell but for
different reasons (38, 39, 53). EtBr will not fluoresce unless it
permeates both the inner and outer membranes and intercalates
into nucleic acid. NPN is a nonpolar dye that fluoresces strongly in
hydrophobic environments, such as the lipid bilayer of the inner
membrane or the inner leaflet of the outer membrane. The outer
leaflet of the outer membrane is composed of lipopolysaccharide
(LPS), which forms a tightly packed semicrystalline structure that
effectively excludes both hydrophilic and hydrophobic compounds, including NPN (38, 43). Uptake of NPN was significantly
higher in the efp mutant than in the wild-type or complemented
strain (Fig. 5A). The rapid kinetics of NPN uptake by efp mutants
is highlighted by the observation that equilibrium was almost
achieved within 10 to 15 s of dye addition. In contrast, no significant difference in permeability to EtBr was observed between
wild-type Salmonella and efp mutants (Fig. 5B). Isogenic Salmonella strains lacking the AcrAB multidrug efflux pump served as a
positive control for dye uptake. While this strain was expected to
exhibit the same rate of dye influx as wild-type Salmonella, a decrease in the cell’s efflux capability would cause the dye to gradually accumulate. As expected, the Salmonella acrAB mutant
showed significantly higher total uptake of both NPN and EtBr,
although this mutant achieved equilibrium at a substantially lower
rate than the efp mutant. The kinetics of the efp mutant strain
suggest that although the strain contains a functional acrAB efflux
Journal of Bacteriology
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
FIG 4 Salmonella poxA, yjeK, and efp mutants display various degrees of sensitivity to different detergents. MICs of the Salmonella yjeK, poxA, and efp mutant
strains were assessed as the ability to grow in the presence of the indicated detergents (corresponding structures shown). MICs of LSB (500 to 0.8 mM), CTAB,
and DTAB (100 to 0.8 ␮g/ml) are reported as the lowest concentrations of the detergents at which growth was inhibited. Increasing concentrations of SDS (100
to 0.8 ␮g/ml) had a gradual effect on Salmonella growth, and hence, the MIC is reported as the concentration at which growth was inhibited by 50%. MIC
experiments were replicated three times with identical results.
Outer Membrane Misregulation in efp Mutants
(⌬efp), and complemented (⌬efp/pEF-P/pYjeK) strains. The Salmonella ⌬acrAB strain was included as a positive control. Values are averages of four biological
replicates. P values were calculated using a two-tailed t test (assuming unequal variances) comparing test strains to the wild type at 10 min (ⴱⴱ, P ⬍ 0.01; ⴱⴱⴱ, P ⬍
0.001).
system, alterations in its membrane permeability greatly enhance
the rate of dye influx such that it exceeds the capacity of the efflux
system.
Outer membrane protein expression is altered in efp mutants. Permeability defects, hypersensitivity to antibiotics, and
poor growth under conditions of low osmolarity are reminiscent
of Salmonella mutants deficient in LPS or osmoregulated periplasmic glucan (OPG) biosynthesis. Like poxA and yjeK mutants, Salmonella deep rough and mdoGH mutants display decreased virulence and increased susceptibility to multiple compounds and
grow poorly in low-salt medium (2, 9, 19, 43, 51, 56).
Analysis of lipid A composition by thin-layer chromatography
(TLC) revealed no apparent differences between the mutant and
wild-type strains (see Fig. S4A in the supplemental material). The
degrees of O-antigen modification in wild-type Salmonella and efp
mutants were also examined by SDS-PAGE and silver staining,
and no obvious differences were found (see Fig. S4B in the supplemental material). To assess OPG content, OPG from the wildtype and mutant strains was solvent extracted and analyzed by
January 2012 Volume 194 Number 2
TLC and gel filtration. No significant changes in the composition
or total amount of OPG were observed, even when cells were
grown under hypoosmotic conditions (data not shown). These
data indicate that the outer membrane lipid composition and
OPG content of our mutants are largely similar to those of wildtype Salmonella, even under conditions of low osmolarity.
The lack of a clearly observable difference in the LPS and OPG
composition of the mutants suggested that the observed phenotypes might be due to changes in membrane protein expression.
Beta-barrel proteins known as porins can constitute approximately 50% of the outer membrane by mass and serve as the primary channel through which many small compounds, including
antibiotics, enter the cell. Inner and outer membranes were isolated from wild-type Salmonella, an efp mutant, and an efp mutant
strain complemented with pEF-P and pYjeK (Fig. 6). Most proteins in the inner and outer membranes appeared to be expressed
at similar levels in all strains. However, two proteins with masses
of approximately 26 kDa and 15 kDa were present at considerably
higher levels in the efp mutant. Mass spectrometry and N-terminal
jb.asm.org 421
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
FIG 5 Salmonella efp mutants are permeable to NPN but not EtBr. Fluorescent dye uptake for NPN (A) or EtBr (B) by the Salmonella wild-type, efp mutant
Zou et al.
Edman sequencing identified the 26-kDa protein as KdgM (also
called SkrP2, encoded by STM1131), a poorly characterized member of the monomeric oligogalacturonate-specific porin superfamily (57). The 15-kDa outer membrane protein was identified
as the outer membrane lipoprotein, SlyB.
The soluble proteome of efp mutants is similar to that of yjeK
and poxA mutants. The limited changes observed in the membrane proteome of efp mutants are consistent with previous observations that fewer than 90 proteins were differentially expressed between the soluble proteomes of wild-type Salmonella
and the poxA and yjeK mutants (42). Two-dimensional gel analysis was used to compare the soluble proteomes of the Salmonella
wild-type and efp mutant strains (see Fig. S5 in the supplemental
material). Similar to what had previously been observed for the
poxA and yjeK mutants, the efp mutant proteome varied only
slightly from that of the wild type. While the expression of most
proteins was unaffected, a number of spots were more highly expressed in the efp mutant strain than in the wild-type strain, including a cluster of spots previously identified as being upregulated in poxA and yjeK mutants. These findings were corroborated
by a recent study which reported changes of equally small magnitudes in the soluble proteome of an E. coli efp mutant (60). This
finding is also consistent with data obtained for the efp (chvH)
mutant of Agrobacterium tumefaciens, where few differences were
observed in the protein profiles of the mutant and wild-type
strains using one-dimensional SDS-PAGE, except for a single, unidentified 32-kDa protein that was expressed at much higher levels
in the mutant (47).
422
jb.asm.org
Journal of Bacteriology
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
FIG 6 Inner and outer membrane proteomes of Salmonella efp mutants
differ from those of wild-type Salmonella. Inner and outer membrane
preparations of wild-type (wt) Salmonella, the efp mutant (⌬efp), and the
efp mutant complemented with plasmids pEF-P and pYjeK (⌬efp pE pY)
were separated by 4 to 20% SDS-PAGE prior to staining with colloidal
Coomassie. Outer membrane proteins were identified by mass spectrometry as indicated. The identity of KdgM (STM1131) was further verified by
Edman sequencing. Identities of bands are indicated and are as follows: 1,
outer membrane protein A (OmpA, STM1070); 2, nucleoside channel
(Tsx, STM0413); 3, oligogalacturonate-specific porin protein (KdgM,
STM1131); 4, outer membrane protein W (OmpW, STM1732); 5,
peptidoglycan-associated outer membrane lipoprotein (Pal, STM0749); 6,
outer membrane lipoprotein (SlyB, STM1445).
Disruption of kdgM in efp mutants restores growth on lowsalt medium, increases antibiotic resistance, and decreases
membrane permeability. To address the possibility that the increased permeability observed in efp mutants was due to the adventitious overexpression of KdgM, the kdgM (STM1131) gene
was replaced with a kanamycin drug resistance cassette. The resulting mutation was introduced via phage transduction into fresh
wild-type Salmonella strain 14028s and the isogenic efp mutant.
Analysis of outer membrane proteins from the efp kdgM double
mutant confirmed that the 26-kDa overexpressed protein observed in the efp mutant was, in fact, KdgM (Fig. 7A).
Deletion of KdgM in an efp mutant background rescued many
of the phenotypes, including those associated with the cell surface.
Specifically, growth in MOPS medium was restored to normal
levels in the efp kdgM double mutant (see Fig. S1 in the supplemental material). Also, the low-osmolarity growth defect and the
susceptibility to LSB and gentamicin were completely reverted to
wild-type levels in the absence of KdgM (Fig. 7B to D). Deletion of
kdgM also restored the growth of yjeK and poxA mutants on lowsalt agar (Fig. 7B). Furthermore, the increased permeability of efp
mutants to the hydrophobic dye NPN was completely rescued by
the loss of KdgM (Fig. 8). As a control for these studies, the efp
kdgM mutant was compared to an isogenic efp ompW mutant.
OmpW is an outer membrane protein similar in size to KdgM.
Although the OmpW protein is present in the outer membrane at
levels similar to those of KdgM, removal of OmpW did not rescue
any of the efp mutant phenotypes (data not shown). These data
suggest that the observed phenotypes are likely attributable to specific properties of the KdgM porin and are not due to general
membrane destabilization caused by other factors, such as membrane overcrowding. To verify that the phenotypic reversion observed was due to deletion of kdgM and not due to a secondary
mutation, the kdgM mutation was independently rederived (via
transduction) four times in the Salmonella efp mutant background. Strains containing the ompW mutation were generated in
parallel. Each independently derived double mutant was tested for
growth on AB2 agar, and in all cases, the removal of KdgM restored growth whereas loss of OmpW had no effect (data not
shown).
Complementation of the kdgM deletion was carried out by
cloning the kdgM open reading frame into the high-copy-number
vector pBAD18 to generate pKdgM. Induction of pKdgM in an efp
kdgM double mutant strain restored the LSB hypersensitivity phenotype. Furthermore, wild-type Salmonella overexpressing KdgM
was as sensitive to LSB as the efp mutant empty-vector control (see
Fig. S6A in the supplemental material). Transcript analyses confirmed that the kdgM transcript levels were 25-fold higher in the
wild-type strain harboring pKdgM following induction than in
wild-type Salmonella. These findings indicate that the detergent
sensitivity caused by KdgM overexpression is not specific to an efp
null background but rather due to an intrinsic property of the
KdgM porin.
To determine whether the upregulation of KdgM occurred at
the transcriptional or translational level, the effects of the yjeK,
poxA, and efp mutations on the expression of kdgM were assessed
using quantitative real-time PCR (see Fig. S6B in the supplemental
material). Steady-state kdgM transcript levels were 16-fold higher
in the efp mutant strain than in wild-type Salmonella. Under the
same conditions, kdgM transcript levels were 4-fold higher in a
poxA mutant strain whereas kdgM transcript levels in a yjeK mu-
Outer Membrane Misregulation in efp Mutants
tant were similar to those of wild-type Salmonella. Of note, the
degree of KdgM overexpression in the poxA, yjeK, and efp mutants
correlates with the relative severity of their phenotypes in many
different assays. Given the role of EF-P as a translation factor,
these findings indicate that the effects of EF-P on KdgM expression are likely indirect.
DISCUSSION
FIG 8 Deletion of kdgM restores wild-type levels of permeability to NPN.
Fluorescent dye uptake for wild-type Salmonella, the efp mutant (⌬efp), the
complemented strain (⌬efp/pEFP/pYjeK), the kdgM mutant (⌬kdgM), and the
efp kdgM double mutant (⌬efp ⌬kdgM). A ⌬acrAB strain was included as a
positive control. Values are averages of four biological replicates. P values were
calculated using a two-tailed t test (assuming unequal variances) comparing
test strains to the ⌬efp mutant at 10 min (ⴱⴱ, P ⬍ 0.01).
January 2012 Volume 194 Number 2
Since its identification in 1975, the role of EF-P in protein synthesis has remained enigmatic. While EF-P is not required to reconstitute protein synthesis in vitro, some studies reported a mild
stimulation of peptide synthesis upon the addition of EF-P (27,
52). The current findings indicate that EF-P is not essential for
viability in Salmonella or E. coli and that loss of EF-P affects the
production of relatively few proteins, many of which are more
highly expressed in the absence of EF-P. The hypersusceptibility of
poxA, yjeK, and efp mutants to a wide variety of antimicrobials,
their poor growth under low-osmolarity conditions, reduced migration, and attenuated virulence are indicative of a perturbation
in the cell envelope. The rescue of the cell surface-associated phe-
jb.asm.org 423
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
FIG 7 Loss of KdgM rescues the osmosensitivity, drug resistance, and detergent sensitivity phenotypes of Salmonella efp mutants. (A) Inner and outer membrane
preparations of wild-type (wt) Salmonella strain 14028s, the efp mutant, and the efp kdgM double mutant. The arrow indicates KdgM.(B) Effects of kdgM
knockout on the growth of poxA, yjeK, and efp mutant Salmonella strains on AB2 agar. The ⌬efp strain produces poorly visible microcolonies on the agar (Fig.
2). MICs of gentamicin (25 to 0.8 ␮g/ml) (C) and LSB (100 to 0.8 mM) (D) for the various mutant Salmonella strains as determined in Fig. 1 and 4, respectively.
MIC experiments were replicated three times with identical results.
Zou et al.
424
jb.asm.org
gets its cognate transcripts and the role of this factor in bacterial
physiology.
ACKNOWLEDGMENTS
We thank Mohammed Adil Khan and Russell Bishop for lipid A analysis.
We also thank Hiroshi Nikaido for a helpful discussion regarding the
assessment of outer membrane permeability and Trevor Moraes for help
with the isolation of bacterial membranes.
M.I. is supported by a grant from the National Institutes of Health
(GM065183). W.W.N. received support from the Canadian Institutes of
Health Research (MOP-86683) and the Natural Sciences and Engineering
Research Council of Canada (NSERC; RGPIN 386286-10). S.B.Z. is the
recipient of a Vanier Canada Graduate Scholarship from NSERC.
REFERENCES
1. Ambrogelly A, O’Donoghue P, Soll D, Moses S. 2010. A bacterial
ortholog of class II lysyl-tRNA synthetase activates lysine. FEBS Lett. 584:
3055–3060.
2. Ames BN, Lee FD, Durston WE. 1973. An improved bacterial test system
for the detection and classification of mutagens and carcinogens. Proc.
Natl. Acad. Sci. U. S. A. 70:782–786.
3. Aoki H, Adams SL, Turner MA, Ganoza MC. 1997. Molecular characterization of the prokaryotic efp gene product involved in a peptidyltransferase reaction. Biochimie 79:7–11.
4. Aoki H, Dekany K, Adams SL, Ganoza MC. 1997. The gene encoding the
elongation factor P protein is essential for viability and is required for
protein synthesis. J. Biol. Chem. 272:32254 –32259.
5. Baba T, et al. 2006. Construction of Escherichia coli K-12 in-frame, singlegene knockout mutants: the Keio collection. Mol. Syst. Biol. 2:2006.0008.
6. Bailly M, de Crécy-Lagard V. 2010. Predicting the pathway involved in
post-translational modification of elongation factor P in a subset of bacterial species. Biol. Direct. 5:3.
7. Bearson SM, Bearson BL, Brunelle BW, Sharma VK, Lee IS. 2011. A
mutation in the poxA gene of Salmonella enterica serovar Typhimurium
alters protein production, elevates susceptibility to environmental challenges, and decreases swine colonization. Foodborne Pathog. Dis.
8:725–732.
8. Behshad E, et al. 2006. Enantiomeric free radicals and enzymatic control
of stereochemistry in a radical mechanism: the case of lysine 2,3aminomutases. Biochemistry 45:12639 –12646.
9. Bhagwat AA, et al. 2009. Osmoregulated periplasmic glucans of Salmonella enterica serovar Typhimurium are required for optimal virulence in
mice. Microbiology 155:229 –237.
10. Blaha G, Stanley RE, Steitz TA. 2009. Formation of the first peptide
bond: the structure of EF-P bound to the 70S ribosome. Science 325:
966 –970.
11. Chatterjee I, Gross SR, Kinzy TG, Chen KY. 2006. Rapid depletion of
mutant eukaryotic initiation factor 5A at restrictive temperature reveals
connections to actin cytoskeleton and cell cycle progression. Mol. Genet.
Genomics 275:264 –276.
12. Chattopadhyay MK, Park MH, Tabor H. 2008. Hypusine modification
for growth is the major function of spermidine in Saccharomyces cerevisiae
polyamine auxotrophs grown in limiting spermidine. Proc. Natl. Acad.
Sci. U. S. A. 105:6554 – 6559.
13. Cherepanov PP, Wackernagel W. 1995. Gene disruption in Escherichia
coli: TcR and KmR cassettes with the option of Flp-catalyzed excision of
the antibiotic-resistance determinant. Gene 158:9 –14.
14. Clement PM, et al. 2003. Identification and characterization of eukaryotic initiation factor 5A-2. Eur. J. Biochem. 270:4254 – 4263.
15. Clement PM, Johansson HE, Wolff EC, Park MH. 2006. Differential
expression of eIF5A-1 and eIF5A-2 in human cancer cells. FEBS J. 273:
1102–1114.
16. Datsenko KA, Wanner BL. 2000. One-step inactivation of chromosomal
genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci.
U. S. A. 97:6640 – 6645.
17. de Crécy E, et al. 2007. Development of a novel continuous culture device
for experimental evolution of bacterial populations. Appl. Microbiol. Biotechnol. 77:489 – 496.
18. De Groote MA, et al. 1995. Genetic and redox determinants of nitric
oxide cytotoxicity in a Salmonella typhimurium model. Proc. Natl. Acad.
Sci. U. S. A. 92:6399 – 6403.
Journal of Bacteriology
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
notypes of efp mutants by loss of the KdgM porin indicates that
many of the phenotypes ascribed to poxA, yjeK, and efp mutants
are the result of misregulation of this porin and not due to a defect
in protein synthesis per se.
The basis for a previous report (4) that efp is an essential gene
and is required for protein synthesis in E. coli remains unclear. It is
possible that a difference in strain background or growth condition, such as high temperature or low-osmolarity medium, led to
the earlier conclusion that this factor is essential. In addition to
Salmonella, EF-P has been successfully knocked out in E. coli,
Agrobacterium, and Acinetobacter (5, 17, 47, 59). In the case of the
Acinetobacter efp mutant, suppressor mutations conveying enhanced growth were obtained after a few generations (17). Taken
together, these studies suggest that EF-P may be nonessential in
many bacterial species.
It has been consistently observed that disruption of yjeK leads
to milder phenotypes than disruption of poxA does (42). These
findings are extended by showing that the most severe phenotypes
are obtained by deletion of efp or, in the most extreme case,
through the overexpression of EF-P(K34A) in an efp mutant background. The relative phenotypic severity of the yjeK, poxA, and efp
mutant strains correlates with their respective degree of overexpression of the outer membrane porin KdgM. Recently published
data by Roy et al. place YjeK upstream of PoxA in the modification
pathway. YjeK first converts lysine to ␤-lysine, which is then ligated onto EF-P by PoxA in an ATP-dependent reaction to generate a novel lysyl-␤-lysine amino acid modification (48). Several
studies have also demonstrated that PoxA is able to use lysine as a
substrate for the modification of EF-P in vitro, albeit at a greatly
reduced rate (1, 42, 48, 60). One plausible model for the relative
severity of phenotypes is that in the absence of YjeK, PoxA attaches
a lysine instead of a ␤-lysine to EF-P to generate a partially active
protein. However, in the absence of PoxA, no such modification
can occur, resulting in EF-P with an unmodified lysyl side chain
that has little to no activity. Mutation of the lysyl side chain to an
alanine could conceivably reduce the activity of EF-P even further,
thereby creating a dominant negative effect and further exacerbating the phenotypes of the efp mutant.
The role of EF-P in bacterial physiology remains unclear. The
novel finding that EF-P plays a role in membrane integrity in
Salmonella is interesting in light of two reports which link eIF5A to
cell wall integrity in yeast and may also be true for the distantly
related genus Agrobacterium (11, 54). Several eukaryotic species
encode multiple isoforms of eIF5A. It has been reported that eIF5a
can selectively bind certain RNA molecules, supporting the notion
that these factors may also modulate the translation of specific
transcripts instead of acting as a general translation factor (14, 15,
58). aIF5A and eIF5A are generally considered to be essential for
cellular growth, and some reports indicate that they play a role in
peptide chain elongation beyond the formation of the first peptide
bond, although these points remain controversial (28, 34, 49).
While the mechanism by which EF-P, aIF5A, and eIF5A affect
peptide bond synthesis at the peptidyl transferase center is likely
similar, there are also considerable differences among the three
proteins (61). Importantly, both aIF5A and eIF5A lack the third
beta-barrel domain present in EF-P. The second and third domains of EF-P are similar to an S1 RNA-binding domain, which,
according to its location in the EF-P/ribosome complex, may interact with the 5= untranslated region of the translating mRNA
(10). Future investigation will focus on how EF-P specifically tar-
Outer Membrane Misregulation in efp Mutants
January 2012 Volume 194 Number 2
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
derived by comparison of complete bacterial genomes. Proc. Natl. Acad.
Sci. U. S. A. 93:10268 –10273.
Navarre WW, et al. 2006. Selective silencing of foreign DNA with low GC
content by the H-NS protein in Salmonella. Science 313:236 –238.
Navarre WW, et al. 2010. PoxA, YjeK, and elongation factor P coordinately modulate virulence and drug resistance in Salmonella enterica. Mol.
Cell 39:209 –221.
Nikaido H. 2003. Molecular basis of bacterial outer membrane permeability revisited. Microbiol. Mol. Biol. Rev. 67:593– 656.
Osborn MJ, Gander JE, Parisi E, Carson J. 1972. Mechanism of assembly
of the outer membrane of Salmonella typhimurium. Isolation and characterization of cytoplasmic and outer membrane. J. Biol. Chem. 247:
3962–3972.
Park MH. 2006. The post-translational synthesis of a polyamine-derived
amino acid, hypusine, in the eukaryotic translation initiation factor 5A
(eIF5A). J. Biochem. 139:161–169.
Park MH, Nishimura K, Zanelli CF, Valentini SR. 2010. Functional
significance of eIF5A and its hypusine modification in eukaryotes. Amino
Acids 38:491–500.
Peng WT, Banta LM, Charles TC, Nester EW. 2001. The chvH locus of
Agrobacterium encodes a homologue of an elongation factor involved in
protein synthesis. J. Bacteriol. 183:36 – 45.
Roy H, et al. 2011. The tRNA synthetase paralog PoxA modifies elongation factor P with (R)-␤-lysine. Nat. Chem. Biol. 7:667– 669.
Saini P, Eyler DE, Green R, Dever TE. 2009. Hypusine-containing
protein eIF5A promotes translation elongation. Nature 459:118 –121.
Schmieger H. 1971. A method for detection of phage mutants with altered
transducing ability. Mol. Gen. Genet. 110:378 –381.
Schnaitman CA, Klena JD. 1993. Genetics of lipopolysaccharide biosynthesis in enteric bacteria. Microbiol. Rev. 57:655– 682.
Swaney S, et al. 2006. Characterization of a high-throughput screening
assay for inhibitors of elongation factor P and ribosomal peptidyl transferase activity. J. Biomol. Screen. 11:736 –742.
Vaara M, Plachy WZ, Nikaido H. 1990. Partitioning of hydrophobic
probes into lipopolysaccharide bilayers. Biochim. Biophys. Acta 1024:
152–158.
Valentini SR, Casolari JM, Oliveira CC, Silver PA, McBride AE. 2002.
Genetic interactions of yeast eukaryotic translation initiation factor 5A
(eIF5A) reveal connections to poly(A)-binding protein and protein kinase
C signaling. Genetics 160:393– 405.
Van Dyk TK, Smulski DR, Chang YY. 1987. Pleiotropic effects of poxA
regulatory mutations of Escherichia coli and Salmonella typhimurium, mutations conferring sulfometuron methyl and alpha-ketobutyrate hypersensitivity. J. Bacteriol. 169:4540 – 4546.
Vuorio R, Vaara M. 1992. The lipid A biosynthesis mutation lpxA2 of
Escherichia coli results in drastic antibiotic supersusceptibility. Antimicrob. Agents Chemother. 36:826 – 829.
Wirth C, et al. 2009. NanC crystal structure, a model for outer-membrane
channels of the acidic sugar-specific KdgM porin family. J. Mol. Biol.
394:718 –731.
Xu A, Jao DL, Chen KY. 2004. Identification of mRNA that binds to
eukaryotic initiation factor 5A by affinity co-purification and differential
display. Biochem. J. 384:585–590.
Yamamoto N, et al. 2009. Update on the Keio collection of Escherichia coli
single-gene deletion mutants. Mol. Syst. Biol. 5:335.
Yanagisawa T, Sumida T, Ishii R, Takemoto C, Yokoyama S. 2010. A
paralog of lysyl-tRNA synthetase aminoacylates a conserved lysine residue
in translation elongation factor P. Nat. Struct. Mol. Biol. 17:1136 –1143.
Zou SB, Roy H, Ibba M, Navarre WW. 2011. Elongation factor P
mediates a novel post-transcriptional regulatory pathway critical for bacterial virulence. Virulence 2:147–151.
jb.asm.org 425
Downloaded from http://jb.asm.org/ on April 19, 2012 by University of Western Ontario
19. Fiedler W, Rotering H. 1988. Properties of Escherichia coli mutants lacking membrane-derived oligosaccharides. J. Biol. Chem. 263:
14684 –14689.
20. Ganoza MC, Aoki H. 2000. Peptide bond synthesis: function of the efp
gene product. Biol. Chem. 381:553–559.
21. Ganoza MC, Kiel MC, Aoki H. 2002. Evolutionary conservation of
reactions in translation. Microbiol. Mol. Biol. Rev. 66:460 – 485.
22. Gerdes SY, et al. 2003. Experimental determination and system level
analysis of essential genes in Escherichia coli MG1655. J. Bacteriol. 185:
5673–5684.
23. Gibson DG, et al. 2009. Enzymatic assembly of DNA molecules up to
several hundred kilobases. Nat. Methods 6:343–345.
24. Gil R, Silva FJ, Pereto J, Moya A. 2004. Determination of the core of a
minimal bacterial gene set. Microbiol. Mol. Biol. Rev. 68:518 –537.
25. Glass JI, et al. 2006. Essential genes of a minimal bacterium. Proc. Natl.
Acad. Sci. U. S. A. 103:425– 430.
26. Glick BR, Chladek S, Ganoza MC. 1979. Peptide bond formation stimulated by protein synthesis factor EF-P depends on the aminoacyl moiety
of the acceptor. Eur. J. Biochem. 97:23–28.
27. Glick BR, Ganoza MC. 1975. Identification of a soluble protein that
stimulates peptide bond synthesis. Proc. Natl. Acad. Sci. U. S. A. 72:
4257– 4260.
28. Gregio AP, Cano VP, Avaca JS, Valentini SR, Zanelli CF. 2009. eIF5A
has a function in the elongation step of translation in yeast. Biochem.
Biophys. Res. Commun. 380:785–790.
29. Guy-Caffey JK, Rapoza MP, Jolley KA, Webster RE. 1992. Membrane
localization and topology of a viral assembly protein. J. Bacteriol. 174:
2460 –2465.
30. Guzman LM, Belin D, Carson MJ, Beckwith J. 1995. Tight regulation,
modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121– 4130.
31. Hanawa-Suetsugu K, et al. 2004. Crystal structure of elongation factor P
from Thermus thermophilus HB8. Proc. Natl. Acad. Sci. U. S. A. 101:
9595–9600.
32. Hancock RE. 1984. Alterations in outer membrane permeability. Annu.
Rev. Microbiol. 38:237–264.
33. Hashimoto-Gotoh T, et al. 2000. A set of temperature sensitivereplication/-segregation and temperature resistant plasmid vectors with
different copy numbers and in an isogenic background (chloramphenicol,
kanamycin, lacZ, repA, par, polA). Gene 241:185–191.
34. Henderson A, Hershey JW. 2011. Eukaryotic translation initiation factor
(eIF) 5A stimulates protein synthesis in Saccharomyces cerevisiae. Proc.
Natl. Acad. Sci. U. S. A. 108:6415– 6419.
35. Joyce AR, et al. 2006. Experimental and computational assessment of
conditionally essential genes in Escherichia coli. J. Bacteriol. 188:
8259 – 8271.
36. Kaniga K, Compton MS, Curtiss R, III, Sundaram P. 1998. Molecular
and functional characterization of Salmonella enterica serovar Typhimurium poxA gene: effect on attenuation of virulence and protection. Infect.
Immun. 66:5599 –5606.
37. Kong L, Fromant M, Blanquet S, Plateau P. 1991. Evidence for a new
Escherichia coli protein resembling a lysyl-tRNA synthetase. Gene 108:
163–164.
38. Loh B, Grant C, Hancock RE. 1984. Use of the fluorescent probe 1-Nphenylnaphthylamine to study the interactions of aminoglycoside antibiotics with the outer membrane of Pseudomonas aeruginosa. Antimicrob.
Agents Chemother. 26:546 –551.
39. Murata T, Tseng W, Guina T, Miller SI, Nikaido H. 2007. PhoPQmediated regulation produces a more robust permeability barrier in the
outer membrane of Salmonella enterica serovar Typhimurium. J. Bacteriol. 189:7213–7222.
40. Mushegian AR, Koonin EV. 1996. A minimal gene set for cellular life