Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

FREE

This Article FREE Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Related articles in JB Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Google Scholar Articles by Handford, J. I. Articles by Palmer, T. PubMed PubMed Citation Articles by Handford, J. I. Articles by Palmer, T.

 Previous Article  |  Next Article 

Journal of Bacteriology, August 2009, p. 4732-4749, Vol. 191, No. 15
0021-9193/09/$08.00+0     doi:10.1128/JB.00136-09
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Conserved Network of Proteins Essential for Bacterial Viability ^,

Jennifer I. Handford,^1, Bérengère Ize,^1,2,, Grant Buchanan,³ Gareth P. Butland,^4,5 Jack Greenblatt,⁵ Andrew Emili,⁵ and Tracy Palmer^3*

Department of Molecular Microbiology, John Innes Centre, Colney Lane, Norwich NR4 7UH, United Kingdom,¹ School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, United Kingdom,² Division of Molecular and Environmental Microbiology, College of Life Sciences, University of Dundee, Dundee DD1 5EH, United Kingdom,³ Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, California, 94720,⁴ Banting and Best Department of Medical Research, Terrence Donnelly Center for Cellular and Biomolecular Research, University of Toronto, Toronto, Ontario, Canada M5S 3E1⁵

Received 31 January 2009/ Accepted 7 April 2009

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The yjeE, yeaZ, and ygjD genes are highly conserved in the genomes of eubacteria, and ygjD orthologs are also found throughout the Archaea and eukaryotes. In this study, we have constructed conditional expression strains for each of these genes in the model organism Escherichia coli K12. We show that each gene is essential for the viability of E. coli under laboratory growth conditions. Growth of the conditional strains under nonpermissive conditions results in dramatic changes in cell ultrastructure. Deliberate repression of the expression of yeaZ results in cells with highly condensed nucleoids, while repression of yjeE and ygjD expression results in at least a proportion of very enlarged cells with an unusual peripheral distribution of DNA. Each of the three conditional expression strains can be complemented by multicopy clones harboring the rstA gene, which encodes a two-component-system response regulator, strongly suggesting that these proteins are involved in the same essential cellular pathway. The results of bacterial two-hybrid experiments show that YeaZ can interact with both YjeE and YgjD but that YgjD is the preferred interaction partner. The results of in vitro experiments indicate that YeaZ mediates the proteolysis of YgjD, suggesting that YeaZ and YjeE act as regulators to control the activity of this protein. Our results are consistent with these proteins forming a link between DNA metabolism and cell division.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
One of the major outcomes of mass genome sequencing was the discovery of an array of hitherto undescribed conserved genes encoding proteins of unknown function. In eubacteria, several different approaches, including targeted gene disruption (37, 61, 71), genetic footprinting (19), and lethal insertion trapping (36), have been used to identify sets of genes essential for survival in the laboratory in rich growth media. Although the nature and exact number of essential genes is likely to vary between bacterial species, it is becoming clear that there is a core set of genes that are essential for the viability of probably most bacteria (21), some of which are currently poorly characterized. In this work, we have examined the functions of the products of three conserved essential genes, yjeE, yeaZ, and ygjD, in Escherichia coli.

YjeE is a small (approximately 17 kDa), highly conserved protein unique to eubacteria. The elucidation of a high-resolution X-ray crystal structure of YjeE of Haemophilus influenzae, coupled with biochemical analysis of the E. coli protein, has shown that it is a typical P-loop ATPase that binds and hydrolyzes ATP, leading to speculation that it may act as a hydrolysis-driven molecular switch (3, 60). Studies have shown that the yjeE gene is essential for the viability not only of E. coli (3, 5, 17) but probably also Salmonella enterica (36), Fransicella novicida (18), and Streptococcus pneumoniae (71), although it is dispensable for the survival of Staphylococcus aureus (71), Anabaena (68), and Bacillus subtilis (reference 27, correcting the earlier study in reference 37). The cellular function of YjeE is not known, although loss of yjeE resulted in aberrant heterocyst development in Anabaena (68). Based on the observation that yjeE genes are absent from the genomes of some Mycoplasma and Ureaplasma species, which lack cell walls, together with the fact that yjeE is found in an operon with a gene encoding a cell wall amidase in some gram-negative bacteria, it has been proposed that YjeE plays a role in cell wall biogenesis (60). Recently rstA, the gene encoding the response regulator of the RstAB two-component signal transduction system, was shown to act as a multicopy suppressor of the depletion phenotype in a yjeE conditional expression strain (referred to herein as, for example, "depleted of yjeE" or "yjeE depletion strain") (10).

YeaZ and YgjD (also known as Gcp) are sequence-related proteins, with the E. coli pair sharing 29% identity within their first 100 amino acids. yeaZ shows a distribution similar to that of yjeE, being present uniquely in eubacteria, where it is found in all eubacterial genomes sequenced to date, with the exceptions of the highly reduced genomes of the obligate endosymbionts Carsonella ruddii and Sulcia muelleri (42, 48). The yeaZ gene has been shown or inferred to be essential for the viability of E. coli (6), S. pneumoniae, S. aureus (71), B. subtilis (27, 37), F. novicida (18), Pseudomonas aeruginosa (39), and Mycoplasma genitalium (22). The YeaZ proteins form a family assigned as COG1214 (http://www.ncbi.nlm.nih.gov/Structure/cdd/cddsrv.cgi?uid=COG1214) which is annotated as "Inactive homolog of metal-dependent proteases, putative molecular chaperone." A recent crystal structure of S. enterica serovar Typhimurium YeaZ shows that the protein is part of the HSP70 actin-like-fold (HALF) protein family and suggests the likelihood that upon domain rearrangement, the protein would bind a nucleotide (51). Purification of an affinity-tagged version of YeaZ produced at native levels identified YgjD as a copurifying partner (9), suggesting that these two homologous proteins form a complex in vivo.

At least one copy of the ygjD gene (COG0533) is found in the genome of every sequenced organism to date, with the exception of the highly reduced genomes of C. ruddii and S. muelleri (42, 48). It has been shown or inferred to be essential for the viability of various eubacterial species, such as E. coli (5), S. pneumoniae, S. aureus (61, 71), B. subtilis (27, 37), F. novicida (18), P. aeruginosa (39), and M. genitalium (22) and also for the eukaryote Saccharomyces cerevisiae (20). Surprisingly, and in contrast to its widely essential nature, ygjD has reportedly been disrupted in the cyanobacterium Synechocystis sp. strain PCC6803, where the mutation was associated with a salt-sensitive phenotype (31, 76).

In enteric bacteria, ygjD is often used synonymously with gcp, reflecting the fact that the YgjD ortholog from Mannheimia (formerly Pasteurella) haemolytica has been characterized as an O-sialoglycoprotein endopeptidase. M. haemolytica Gcp has been isolated from culture supernatants and shown to cleave glycoproteins, such as glycophorin A, in erythrocyte membranes (1, 2). The protease activity was shown to require metal ions and to be specific for O-sialoglycosylated proteins (2). The human ortholog of Gcp, termed OSGEP, has also been implicated in proteolysis, bringing about the cleavage of the misfolded nuclear receptor corepressor protein (50). In S. cerevisiae, the ortholog of YgjD, termed Kae1 (for kinase-associated endopeptidase 1), associates with a protein complex involved in transcription, comprising additionally a transcription factor, a kinase, and two further proteins (35). It has also been found as a component of a separate complex, termed KEOPS, which is involved in regulating the length of telomeres (15). Recently, the structure of the Pyrococcus abyssi ortholog of YgjD was solved to 3 Å. The structure revealed that the protein bound ATP, Fe(III), and possibly, additional divalent metal ions (26), and the results of biochemical experiments suggested that it was an atypical DNA binding protein.

In this study, we have explored the roles of the YjeE, YeaZ, and YgjD proteins in E. coli. We have confirmed previous reports that these are essential gene products and shown that their depletion is associated with startling changes in cell morphology and ultrastructure. Using a combination of genetic and biochemical approaches, we demonstrate that the three proteins form an interaction network and that YeaZ is a protease that apparently specifically degrades YgjD.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains. The strains and plasmids used during this study are listed in Table 1. E. coli strain MC4100-A, an arabinose-resistant isolate of MC4100, was the parental strain used for construction of all of the depletion strains in this study. All depletion strains contain a single copy of the gene of interest under the control of the arabinose promoter and optimized ribosome binding site derived from pBAD24 (24) and integrated at the lambda phage attachment site and a deletion of the gene at the native site.

View this table: [in this window] [in a new window]

TABLE 1. Strains and plasmids used in this study

All strains were constructed in a similar fashion. For strain JH17, which conditionally expresses yjeE, the yjeE gene was first cloned into pBAD24, following amplification with primers YjeEApoI and YjeESmaI (see Table S1 in the supplemental material for primer sequences), digestion with ApoI and SmaI, and cloning into EcoRI-SmaI-digested pBAD24. The araC-P_BAD-yjeE region from this construct was amplified by PCR using primers AraCEcoRI and YjeErevBglII, digested with EcoRI and BglII, and cloned into pRS552 that had been previously digested with EcoRI and BamHI. This construct was used to transfer the entire arabinose-regulated yjeE allele to the lambda attachment site (attB) on the chromosome of E. coli host strain MC1061 (56). The attB::(araC⁺ P_BAD yjeE⁺) allele, now 100% linked to a kanamycin resistance marker, was then moved by P1 transduction to the chromosome of MC4100-A. For deletion of the native copy of yjeE, a deletion allele of yjeE was assembled initially in pBluescript. The construct harbored the first 14 codons of yjeE (to preserve the coding sequence for the overlapping yjeF gene), the last 2 codons of yjeE, and the respective 474 bp upstream and 520 bp downstream and was constructed as follows. The upstream part of the deletion was amplified by PCR using primers Yedel1 and Yedel2, digested with XbaI and EcoRI, and cloned into similarly digested pBluescript. This construct was subsequently digested with EcoRI and KpnI, and the downstream part of the deletion, amplified with primers Yedel3 and Yedel4, was similarly digested and cloned in. The deletion allele of yjeE was excised from pBluescript by digestion with KpnI and XbaI and cloned into pMAK705 that had been previously digested with the same enzymes. The native yjeE gene was subsequently deleted from the chromosome of MC4100-A attB::(araC⁺ P_BAD yjeE⁺) according to the method of Hamilton et al. (25), with the modification that the strain was cultured in medium containing 0.2% arabinose and 0.5% glycerol throughout the procedure.

For conditional expression of yeaZ, a similar procedure was followed. Thus, yeaZ was cloned into pBAD24 by amplification with primers yeaZNheI and yeaZSmaI, digested with NheI and SmaI, and cloned into similarly digested pBAD24. The araC-P_BAD-yeaZ region was subsequently cloned, following amplification using primers AraCBglII and RSyeaZBglII and digestion with BglII, into pRS552 that had been previously digested with BamHI. The araC-P_BAD-yeaZ allele was transferred to the chromosome of MC1061 and, subsequently, MC4100-A as described above. A plasmid-encoded deletion allele of yeaZ was assembled by cloning the first two codons of yeaZ and the upstream 500 bp, following amplification with primers YeaZupsXbaI and YeaZupsBamHI, into pBluescript as an XbaI-BamHI fragment. The downstream part of the deletion covered the last two codons of yeaZ and 500 bp of downstream DNA, which was amplified with primers YeaZdownsBamHI and YeaZdownsKpnI and cloned into the above plasmid as a BamHI-KpnI fragment. The yeaZ deletion allele was subsequently moved as an XbaI-KpnI fragment from pBluescript to pMAK705. The native yeaZ gene was subsequently deleted from the chromosome of MC4100-A attB::(araC⁺ P_BAD yeaZ⁺) following the same procedure as for yjeE deletion.

The construction of a ygjD depletion strain was achieved essentially in the same way. The ygjD gene was cloned into pBAD24 by amplification with primers BAD24ygjDEcoRI and BAD24ygjDXbaI, digested with EcoRI and XbaI, and cloned into similarly digested pBAD24. The araC-P_BAD-ygjD region was then cloned, following amplification using primers AraCBglII and QEygjDstopBglII and digestion with BglII, into pRS552 that had been previously digested with BamHI. The araC-P_BAD-ygjD allele was transferred to the chromosome of MC1061 and, subsequently, MC4100-A as described above. For deletion of the native ygjD gene, a method based on the redirect system was employed (14). Briefly, 500 bp of DNA upstream of ygjD with the first two codons of the gene were amplified with YgjDapra1 and YgjDapra2, digested with BglII and EcoRI, and cloned into pBluescript that had been previously digested with BamHI and EcoRI. Into this construct was cloned the apramycin resistance cassette, flanked by FLP recombination target sites, as an EcoRI-ApaI fragment (amplified with primers ApraEcoRI and ApraApaI using plasmid pIJ773 [14] as template). Finally, into this construct the downstream part of the ygjD deletion, covering the last two codons of the gene and 500 bp of downstream DNA, was cloned as an ApaI-KpnI fragment (amplified using primers YgjDapra3 and YgjDapra4). Using primers YgjDapra1 and YgjDapra4, the whole insert was amplified by PCR and transformed by electroporation into strain MC4100-A attB::(araC⁺ P_BAD ygjD⁺) harboring plasmid pIJ790, following the method of Datsenko and Wanner (14). After selection for apramycin-resistant colonies at 37°C and confirmation by PCR analysis that replacement of native ygjD by the marked deletion had occurred, the apramycin cassette was subsequently "flipped out" to leave a nonpolar scar sequence as described previously (14).

Plasmid construction. The plasmids constructed in this work are shown in Table 1, and their construction is described in Methods in the supplemental material. All clones obtained from PCR products were sequenced to ensure that no undesired mismatches had been introduced during the amplification procedure.

Growth conditions. Strains were generally cultured aerobically at 37°C in Luria-Bertani (LB) medium unless stated otherwise. For the conditionally lethal strains JH17, JH19, and JH22, LB medium was supplemented with 0.2% arabinose or 0.2% glucose for growth on solid medium and 0.2% arabinose or 0.4% glucose-6-phosphate plus 0.2% fucose for growth in liquid medium. For scoring bacterial two-hybrid interactions on solid medium, MacConkey indicator medium (55) supplemented with a final concentration of 1% maltose was used. For the sequential peptide affinity (SPA) purification, strains were cultured in 1 liter of terrific broth (TB) medium plus K salts (55) at 32°C until late log phase (73). Antibiotics were used at the following final concentrations: apramycin, 50 µg/ml; ampicillin, 125 µg/ml; kanamycin, 25 µg/ml; and chloramphenicol, 20 µg/ml.

For screening of libraries for interacting clones in the bacterial two-hybrid system, strain BTH101 was initially transformed with the target gene present in pT25. These cells were then transformed with aliquots of each of six chromosomal libraries cloned into pUT18, plated onto MacConkey medium supplemented with maltose, and cultured at 30°C for 24 h. Red colonies, signifying possible interacting clones, were restreaked onto fresh MacConkey-maltose medium. For those colonies that remained red upon restreaking, plasmid DNA was isolated and transformed into BTH101 harboring the original target gene on pT25, pT25 only, the torD gene in pT25, or the sufI gene in pT25. These latter three served as controls to ensure that the interaction observed in the original library screen was specific to the target gene and was not observed with empty vector or with unrelated gene fusions.

For screening of libraries for multicopy suppressors of the deletion phenotypes of the three mutants JH17, JH19, and JH22, each mutant was initially transformed with aliquots of each of seven chromosomal libraries cloned into pUT18 or pGAD10, plated onto LB medium supplemented with glucose (0.2%), and cultured at 37°C for up to 72 h. Plasmid DNA from the growing clones was isolated and was transformed into JH17, JH19, and JH22 before the DNA sequence was obtained.

Growth of strains for protein overproduction was achieved as follows. For overproduction of YjeE-His₆ or YeaZ, strain M15(pREP4) was transformed with plasmid pQEYjeE or pQEYeaZ, respectively, and a 2-liter culture of each strain in LB medium was incubated aerobically at 37°C to an optical density (OD) at 600 nm of 0.6. YjeE-His₆ or YeaZ overproduction was subsequently induced by the addition of 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG). After a further 3 h of growth at 37°C, the cells were harvested, washed, and frozen as cell pellets for further processing. For overproduction of YgjD-His₆, we found that inducing protein production at a growth temperature of 37°C resulted in most of the protein being present in inclusion bodies; therefore, the procedure was modified as follows. Strain M15(pREP4) was transformed with plasmid pQEYgjD, and a 2-liter culture of the strain was incubated, aerobically, at 37°C to an OD at 600 nm of 0.6. The culture was subsequently cooled to 21°C, YgjD-His₆ production was induced following the addition of 1 mM IPTG, and the cells were cultured at 21°C for a further 15 h. Cells were harvested and washed as described above. Since the M. haemolytica ortholog of YgjD is a predicted zinc-binding protein (2) and YeaZ is a homolog of YgjD, 0.5 mM ZnCl₂ was also routinely added to cultures at the same time as IPTG was added to induce overproduction of these two proteins.

Transmission electron microscopy analysis. A 30-µl aliquot of each of strains JH17, JH19, and JH22 was inoculated from glycerol stocks into 5 ml of buffered LB supplemented with kanamycin, 0.4% glucose-6-phosphate, and 0.2% fucose and grown overnight at 37°C. These cultures showed poor but visible growth and were used directly for transmission electron microscopy analysis. The parental strain, MC4100-A, was grown overnight in LB and was then subcultured at 1:100 dilution into fresh medium and grown for 4 h until it reached the same OD as the overnight cultures of the conditional strains. Transmission electron microscopy analysis of E. coli strains was carried out essentially as described by Stanley et al. (58), with the minor exception that the grids were made of copper-palladium rather than copper. Images were captured using a Hamamatsu C8484-05G digital charge-coupled-device camera controlled via computer with AMT Image Capture Engine software, version 5.42.493.

Enzyme assays. ATPase assays were carried out according to the method of Norby and Jensen (52), but were scaled down to a 200-µl reaction mixture volume for use in microtiter plates. For each reaction, 200 µg of purified YjeE-His₆ was used and ATP was added to a final concentration of 0.125, 0.25, 0.5, 1, 2, or 4 mM. Five replicates were carried out at each ATP concentration, and K_m and k_cat values were determined by using the Enzyme Kinetics Module plug-in in SigmaPlot.

For the glycoprotease assay, purified glycophorin A (5 µg; obtained from Sigma) was incubated with purified Gcp from M. haemolytica (4.8 µg; Cedarlane Labs, Hornby, Ontario, Canada), purified YeaZ (2 µg), purified YgjD-His₆ (2 µg), or a mixture of the latter two in a 50 mM HEPES buffer, pH 7.4, for 90 min at 37°C. Samples were then separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (7.5% SDS gel), and glycophorin A degradation was followed by Western blot analysis with an antibody directed against GlpA (1/1,000 dilution; Sigma).

β-Galactosidase assays were performed according to the method of Miller (43).

Protein methods. Purification of overproduced proteins is described in Methods in the supplemental material. Strains in which E. coli chromosomal genes were modified by the addition of an SPA tag were previously constructed as part of a large survey of protein complexes in E. coli (9). Affinity purification of native levels of SPA-tagged proteins and any potential binding partners was performed exactly as previously described (9, 73). For the nucleotide addition experiments, cell extracts harboring YjeE-SPA at the native level were prepared as described above and then incubated with ADP, AMP-PNP, or GMP-PNP (each at 1 mM final concentration) for 3 h on ice prior to purification.

The periplasmic fraction of cells was prepared by using the cold osmotic shock protocol (59), and membranes were isolated by ultracentrifugation of cell extracts that were prepared, following French pressing, as described above. The protein concentration was determined by the method of Lowry et al. (40). SDS-PAGE analysis, with 15% acrylamide gels unless stated otherwise, was carried out according to the method of Laemmli (38), and Western blotting by the method of Towbin et al. (62). For the detection of glycoproteins, SDS-polyacrylamide gels were loaded with 10 µg of cell extracts and, following protein separation, were analyzed by glycoprotein staining with a ProQ-Emerald 300 glycoprotein gel stain kit (Invitrogen). The CandyCane markers (Invitrogen), containing a mixture of glycosylated and nonglycosylated proteins, were used as molecular mass standards. For two-dimensional (2-D) gel analysis, 350 µg of protein extract was loaded onto precast immobilized pH gradient strips (18 cm, pH 3 to 10, nonlinear; Amersham) for the first-dimensional separation as described by Widdick et al. (64). Focused strips were separated in the second dimension by SDS-PAGE with Amersham Biosciences DALT 12.5% gels. For protease digestion experiments, cell extracts (350 µg) were mixed with purified YeaZ (10 µg) or YgjD-His₆ (10 µg) or both proteins for 1 h at 37°C prior to 2-D gel separation as described above. Rabbit polyclonal antibodies were raised to YjeE-His₆ (1:5,000 dilution) with purified protein supplied to Sigma-Genosys and to YeaZ (1:10,000 dilution) with protein supplied to Custom Hybridoma. Anti-FLAG antisera were obtained from Sigma. Overproduced purified proteins were identified by tryptic mass fingerprint analysis carried out by the John Innes Centre proteomics facility. Affinity-purified eluates containing protein complexes were analyzed by SDS-PAGE, individual bands were excised, and proteins were identified by peptide mass fingerprinting using a Bruker Reflex IV matrix-assisted laser desorption ionization-time of flight mass spectrometer. Whole eluates were also subjected to tryptic digestion in solution and directly analyzed by ion trap liquid chromatography-tandem mass spectrometry on a Thermo-Finnigan-LTQ mass spectrometer as previously described (73).

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Genomic organization of yjeE, yeaZ, and ygjD. Genomic context can often reveal clues to the functions of genes, since genes coding for proteins with related cellular functions often cluster together, allowing coregulation. The yjeE gene in E. coli is the second gene of a five-gene operon that also includes genes for DNA mismatch repair (mutL), tRNA modification (miaA), and peptidoglycan remodeling (amiB) (Fig. 1A). A similar genomic context is found in most proteobacteria of the gamma subclass, whereas the pairing of yjeE and yjeF (predicted to encode a sugar kinase) is found in Geobacter species, which are members of the deltaproteobacteria, and in the Acidobacteria. In the betaproteobacterium Neisseria meningitidis, the yjeE ortholog is sandwiched between two genes also involved in cell wall-related processes (Fig. 1A).

View larger version (27K): [in this window] [in a new window]

FIG. 1. Genomic context of yjeE, yeaZ, and ygjD in selected eubacteria. (A) Genomic context of yjeE in E. coli and its ortholog NMB0457 in N. meningitidis. (B) Genomic context of the yjeE orthologs ydiB in B. subtilis and rv3422c in M. tuberculosis. (C) Genomic context of yeaZ in E. coli and its ortholog NMB1871 in N. meningitidis. (D) Genomic context of ygjD in E. coli and its ortholog gcp in N. meningitidis. In each case the yjeE ortholog is shaded with black vertical stripes, yeaZ in grey horizontal stripes, and ygjD with black diamonds. The rimI ortholog is shaded with grey spots. Open reading frame sizes are shown approximately to scale. For further details see the text.

As shown in Fig. 1B, for gram-positive bacteria, the genetic arrangement of the yjeE ortholog (termed ydiB in B. subtilis) is somewhat different. The gene almost always directly precedes the genes that are the yeaZ and ygjD orthologs (termed ydiC and ydiE in B. subtilis). Frequently the rimI ortholog, ydiD, is also found within the same transcription unit. In Mycobacterium tuberculosis and other actinomycetes, the yjeE ortholog is preceded by alr, which encodes alanine racemase that catalyzes the synthesis of d-alanine, an essential precursor for peptidoglycan biosynthesis. In B. subtilis, thiL, encoding thiamine-monophosphate kinase, is found 18 bp upstream of ydiB, but this pairing is generally not conserved, even in firmicute genomes.

In E. coli, yeaZ is found in the vicinity of yoaA, encoding a probable ATP-dependent helicase, and genes for a probable outer-membrane protein (yeaY), fatty acyl coenzyme A synthase (fadD), and RNase D (rnd) (Fig. 1C). The context of yeaZ with yoaA is also found in the betaproteobacterial organism Methylobacillus flagellatus. In N. meningitidis and many other betaproteobacteria, yeaZ is immediately upstream of rimI (Fig. 1C), in an arrangement mirroring that observed for many gram-positive organisms. RimI is an acetyltransferase that catalyzes the N-acetylation of the N-terminal alanine residue (following cleavage of the initiator methionine) of the ribosomal protein S18 (28). Remarkably, in the three sequenced Bordetella genomes, the yeaZ and rimI genes are apparently fused to give a single reading frame. A similar fused gene is also seen in the Mycobacterium leprae genome sequence.

The ygjD gene in E. coli (Fig. 1D) is found as a monocistronic unit, flanked on either side by a gene for a probable anion permease (ygjE) and the rpsU gene, encoding ribosomal protein S21. The divergent arrangement of ygjD with rpsU is conserved among the gamma- and many of the betaproteobacteria, as is the linkage with dnaG, encoding DNA primase, and rpoD, encoding the sigma factor ⁷⁰. In N. meningitidis, the ortholog of ygjD is found at the end of a four-gene cluster (Fig. 1D), with the three preceding genes being related to cytochrome c biogenesis and function.

YjeE, YeaZ, and YgjD are essential in E. coli. Previous studies from other groups have shown or inferred that yjeE, yeaZ, and ygjD are essential for E. coli viability (3, 5, 17, 19). Although a strain allowing conditional expression of yjeE has previously been described (3), there are currently no strains allowing for conditional expression of yeaZ or ygjD. Therefore, we constructed individual yjeE, yeaZ, and ygjD conditional-expression strains which we designated JH17, JH19, and JH22, respectively. Each strain was constructed in a similar fashion in background strain MC4100-A, an arabinose-resistant derivative of our standard laboratory strain. In each case, a copy of the gene of interest was placed under the control of the arabinose-inducible, glucose-repressible araBAD promoter at the phage lambda attachment site, after which the native copy was deleted in the presence of arabinose, ensuring that the P_BAD-controlled copy would be induced.

Plating each of the three conditional strains onto LB medium supplemented with either 0.2% arabinose or 0.2% glucose confirmed that each of the three genes encoded an essential protein, since no colonies could be observed on plates supplemented with glucose, even after prolonged periods of incubation (shown in Fig. 6A, "Vector only" panels). The growth characteristics of each strain cultured under gene depletion conditions are shown in Fig. 2. As the parental strain is auxotrophic for arabinose, suppression of gene expression relies upon the arabinose being diluted out rather than metabolized, and therefore, during the first round of growth in liquid culture, the conditional strains appear to grow normally in the presence of glucose-6-phosphate plus fucose. However, upon subculture into glucose-6-phosphate plus fucose-containing medium followed by a second round of incubation, each of the conditional strains ceased growing, confirming that yjeE, yeaZ, and ygjD are indeed essential genes in E. coli.

View larger version (79K): [in this window] [in a new window]

FIG. 6. Genetic complementation of the yjeE, yeaZ, and ygjD depletion strains. (A) Strain MC4100-A (wild type [WT]) and the yjeE (JH17), yeaZ (JH19), and ygjD (JH22) depletion strains were transformed with pQE60 (vector only), pQEYjeEhis, pQEYeaZ, pQEYgjDhis, clone 22.7.27, clone 126, or clone 116. Half of each transformed strain was plated onto LB medium supplemented with 0.2% arabinose (Ara), and the remaining half onto LB medium supplemented with 0.2% glucose (Glc). The plated strains were allowed to grow overnight (16 h) or, where indicated, for 24 h, 48 h, or 72 h at 37°C. A representative part of each of the plates is shown. (B) The shaded areas indicate E. coli genomic DNA that is present on each of clones 22.7.27, 116, and 126. Clone 22.7.27 was isolated from an E. coli K12 MATCHMAKER genomic library cloned into the vector pGAD10. Clone 126 was constructed by cloning a portion of clone 22.7.27 into pGAD10 and covers rstA and the first 867 bp of rstB (for details, see Table S1 in the supplemental material). Clone 116 was constructed by cloning rstA into pBluescript under the control of the lac and tatA promoters.

View larger version (25K): [in this window] [in a new window]

FIG. 2. The yjeE, yeaZ, and ygjD gene products are essential in E. coli K12. (A) Strain MC4100-A (wild type) (filled circles) and the yjeE depletion strain, JH17, subcultured in duplicate (filled squares and filled triangles), were cultured overnight in LB medium containing 0.02% arabinose (Ara). At time zero, each strain was subcultured at a 1:100 dilution into fresh LB medium containing 0.02% arabinose. After aerobic growth for 2 h at 37°C, the cells were harvested, washed twice in LB medium, and finally, resuspended into either LB containing 0.2% arabinose (for MC4100-A and one of the JH17 cultures [filled squares]) or buffered LB containing 0.4% glucose-6-phosphate and 0.2% fucose (G6P) (for the second JH17 culture [filled triangles]). The cells were cultured aerobically for a further 6 h at 37°C, when they were subcultured (again at 1:100 dilution) into fresh LB containing 0.2% arabinose (for MC4100-A and the first JH17 culture) or buffered LB containing 0.4% glucose-6-phosphate and 0.2% fucose (for the second JH17 culture). Cells were then cultured at 37°C for a further 8 h. (B and C) Growth curves, carried out exactly as described for panel A, were repeated with the yeaZ (B) and ygjD (C) depletion strains, JH19 and JH22, respectively. These growth curves were all carried out concurrently, with the same wild-type control, but the results have been separated into three panels for clarity. (D) A 100-µl aliquot of cells was removed from each culture at the 2-h, 8-h, and 16-h time points during the growth curve shown in panel A, and total cell proteins were separated by SDS-PAGE (15% acrylamide) prior to analysis by immunoblotting with a rabbit polyclonal antiserum raised against YjeE. (E) A 100-µl aliquot of cells was withdrawn from each culture at the time points indicated during the growth curve shown in panel B, and total cell proteins were separated by SDS-PAGE (15% acrylamide) prior to analysis by immunoblotting with a rabbit polyclonal antiserum raised against YeaZ. Note that it was not possible to analyze the depletion of YgjD production by Western blotting since we were unable to raise a suitable antibody against this protein. WT, wild type.

As shown in Fig. 2A to C, depletion of YjeE (Fig. 2A) and YeaZ (Fig. 2B) resulted in an inability to detect the cognate protein after Western blotting, consistent with their absence upon prolonged depletion. Interestingly, the Western blot results also demonstrate that the production of each of these proteins in the conditional strains grown in the presence of arabinose is noticeably higher than the native level. This indicates that the levels of transcription and translation of yjeE and yeaZ driven by the araBAD promoter and araB ribosome binding site are higher than their native levels of expression in wild-type cells. This increased level of expression did not have any obvious impact on the physiology of the cells (see Fig. S1 in the supplemental material).

Depletion of YjeE, YeaZ, and YgjD is associated with unusual cellular morphologies. The functions of YjeE, YeaZ, and YgjD are currently unknown. To obtain clues about the cellular roles of these proteins, we carried out transmission electron microscopy analysis of each of the three depletion strains, JH17, JH19, and JH22, grown under conditions where the cell was depleted of each of these essential proteins.

As shown in Fig. 3B, repression of the expression of yjeE resulted in the formation of massive cells which were very elongated and also typically 1.5 to 1.75 times wider than wild-type cells (shown to the same scale as an inset in Fig. 3B), with the occasional cell up to 2.75 times wider. Most of the cells also showed evidence of branching or curving, which is consistent with a defect in cell wall metabolism (seen, for example, in strains lacking penicillin binding protein 5 [PBP5] [49]). Similar to the yjeE depletion strain, strains lacking expression of PBP5 also show an increase in cell diameter. However, unlike cells defective in PBP5 expression, depletion of yjeE also produced cells routinely displaying an unusual distribution of the DNA, which appears to be located around the periphery of the cell rather than throughout the cytoplasm. Interestingly, it has been reported that E. coli cells limited for thymine form very enlarged and branched cells (72) that show a strikingly similar peripheral distribution of the nucleoids (65). However, we were unable to suppress the lethal depletion of yjeE in the presence of added thymine (at 5, 50, or 500 µg/ml) on either liquid or solid medium (not shown). Nonetheless, these observations may suggest that depletion of yjeE leads to a defect in the rate of DNA synthesis.

View larger version (90K): [in this window] [in a new window]

FIG. 3. Cells depleted of YjeE, YeaZ, or YgjD display unusual morphologies. Transmission electron micrographs of strains MC4100-A (wild type [WT]) after growth in LB containing 0.2% arabinose (A) and JH17, JH19, or JH22 (depleted for yjeE, yeaZ, or ygjD, respectively) after growth in buffered LB containing 0.4% glucose-6-phosphate (B to D). MC4100-A is shown in inset, where appropriate, at the same scale as the main picture.

Transmission electron micrographs of E. coli cells with suppressed production of YeaZ are shown in Fig. 3C. In general, depletion of yeaZ also resulted in cells that were enlarged, although not nearly as dramatically as those seen when yjeE was depleted. Cells were also occasionally misshapen, with rare cells showing signs of branching. However, the most startling feature of cells depleted for yeaZ was the appearance of the nucleoid. Almost all of the cells we examined displayed nucleoids that were highly condensed. Indeed, the appearance of yeaZ-depleted cells resembles wild-type E. coli that has been treated with the translational inhibitor chloramphenicol (46, 75). Other treatments that inhibit translation, for example, amino acid starvation, also reportedly cause nucleoid condensation (75). E. coli nucleoid condensation has also been observed when histone-like DNA binding proteins, for example, the chlamidial Hc1 protein (7) or H-NS (57), are overproduced. Thus, the depletion phenotype of yeaZ repression is consistent with cells experiencing a block in protein synthesis.

Transmission electron micrographs of E. coli cells with suppressed production of YgjD are shown in Fig. 3D. It is apparent that depletion of this essential protein is also associated with unusual cellular morphologies. In this case, however, the cells had a more varied appearance. Approximately 60% of cells (from a total of 29 captured on micrographs) looked very much like the strain depleted for yjeE, i.e., cells were enlarged and showed the same unusual distribution of DNA around the cell periphery (e.g., Fig. 3D, right panel). This is suggestive evidence that there might be some interrelationship between the two proteins. As for the conditional yjeE strain, we were unable to suppress the lethal depletion of ygjD in the presence of added thymine. A subpopulation (40%) of ygjD-depleted cells, on the other hand, was rather smaller (although still enlarged relative to the wild type) and their DNA showed a wild-type localization. It is not clear why we observe two different populations, although it may possibly arise due to incomplete depletion of ygjD in a subpopulation of the culture. However, it should be noted that regardless of which of the two populations of cells were examined, in all cases the membranes of the cells showed an unusual ruffled appearance. It is not clear what the basis of this ruffling phenotype might be, although it clearly reflects an aberration connected with the cell envelope.

As shown in Fig. 2, growth of the yjeE and yeaZ depletion strains in the presence of arabinose resulted in an overproduction of YjeE or YeaZ relative to the levels in the parental strain. Electron micrographs of each of the three conditional strains grown in the presence of arabinose are shown in Fig. S1 in the supplemental material. In general, the appearance of each strain was similar in size and morphology to the wild type. There was some grainy material, most probably mesosomes, associated with each of the three conditional strains that was not so prominent in the wild-type strain. However, it is generally accepted that mesosomes are artifacts produced by the chemical fixation techniques (16), so the significance of this observation is not clear.

YjeE interacts with YeaZ in the bacterial two-hybrid assay. Since YjeE has been proposed to act as a molecular switch, it presumably must interact with other cellular proteins under certain conditions. A large-scale study investigating native levels of protein complexes in E. coli suggested that there was a reciprocal interaction between YjeE and YjeF (9), which is encoded immediately upstream of yjeE in E. coli and many other bacterial species. We chose to examine putative interaction partners of YjeE using a genetic approach based on the bacterial two-hybrid system (32). This assay is based around the reconstitution of adenylate cyclase activity from two noninteracting fragments in an E. coli cya strain.

When YjeE was produced as a fusion with the T18 and the T25 fragments of Bordetella pertussis CyaA, a strong interaction was observed, as evidenced by the production of red colonies on MacConkey-maltose medium (not shown) and detection of significant levels of β-galactosidase activity, shown in Fig. 4A and B. This indicates that the YjeE protein is able to form a dimer. Interestingly, oligomerization of purified E. coli YjeE and the B. subtilis YjeE ortholog, YdiB, has recently been reported (34).

View larger version (27K): [in this window] [in a new window]

FIG. 4. YjeE interacts with YeaZ in the bacterial two-hybrid assay. Bacterial two-hybrid interaction analysis was carried out using yjeE and yjeS, yjeF, amiB, a truncated version of amiB lacking the DNA encoding the signal sequence (ssamiB), mutL, miaA, or murI (A) or alr, thiL, yeaZ, rimI, or ygjD (B). Interactions were tested reciprocally (i.e., both with yjeE cloned in pT18 and the partner gene in pT25 and vice versa), with the exceptions of amiB, which was only tested as a C-terminal protein fusion in pT25, and ssamiB, yjeS, and thiL, which we were unable to clone into pT18. Each block of four bars also has two controls where the gene of interest cloned in either pT18 or pT25 was tested reciprocally against the empty plasmid vector lacking yjeE. Strain BTH101 was cotransformed with the pT18 and pT25 plasmid derivatives, a single colony was picked from each transformation and cultured in LB medium supplemented with appropriate antibiotics, and β-galactosidase assays were carried out using the method of Miller (43). Error bars represent standard errors of the means; n = 3.

We next tested whether YjeE could interact with proteins encoded in the genetic neighborhood of yjeE in gram-negative bacteria (Fig. 4A) and gram-positive bacteria (Fig. 4B). No significant interaction was observed between YjeE and any of the proteins YjeS, AmiB (produced with and without its N-terminal signal peptide), MutL, MiaA, MurI, and surprisingly, YjeF. Thus, in each case, strain BTH101 harboring plasmids coproducing YjeE and each of these proteins as CyaA fusions gave white colonies on MacConkey agar plates (data not shown) and a low level of β-galactosidase activity (Fig. 4A and B). However, a strong interaction was observed between YjeE and YeaZ (Fig. 4B), the ortholog of which is encoded directly downstream of the yjeE ortholog in many gram-positive organisms. The interaction was observed when yjeE was present on the high-copy number vector (pT18) and yeaZ on the lower-copy-number vector (pT25) and also when the genes were present in the reciprocal vectors, where the level of β-galactosidase activity was notably higher. No significant interaction between YjeE and ThiL, Alr, RimI, or YgjD could be detected by this method (Fig. 4B). Thus, we conclude from the results of these experiments that there is an interaction between the essential proteins YjeE and YeaZ.

YgjD and YeaZ show reciprocal interaction using the bacterial two-hybrid system. YeaZ and YgjD at native levels have already been shown to form a complex that is stable during purification (9). To examine the interaction of these two proteins and to probe for additional interacting partner proteins, we constructed vectors that produced each protein as a fusion with either the T18 or T25 fragment of B. pertussis adenylate cyclase.

The interaction of YgjD with proteins encoded in its genetic neighborhood in gram-positive bacteria is shown in Fig. 5A. It is apparent that YgjD can form a strong homodimer since coexpression of pT18ygjD and pT25ygjD in the cya strain BTH101 gave bright red colonies on MacConkey-maltose medium (data not shown) and β-galactosidase activity levels of close to 1,200 Miller units (Fig. 5A). In addition to the significant self-interaction, YgjD also showed clear interaction with YeaZ, confirming the results of the copurification experiments of Butland et al. (9) which showed that these two proteins form a complex. The levels of β-galactosidase activity (taken as a measure of the strength of interaction) indicate that the strongest interaction was observed when YgjD was produced as a fusion to the N terminus of T18 (from the higher-copy-number pT18 vector) rather than as a fusion to the C terminus of T25 (from the lower-copy-number pT25 vector). Again, no significant level of interaction was observed when YgjD was tested against YjeE, confirming the observations shown in Fig. 4B. In addition, there was no significant interaction of YgjD with RimI or ThiL when tested using this system.

View larger version (21K): [in this window] [in a new window]

FIG. 5. YeaZ interacts with YgjD, YjeE, and RimI in the bacterial two-hybrid assay. Bacterial two-hybrid interaction analysis was carried out using ygjD and each of thiL, yjeE, yeaZ, and rimI (A) and yeaZ and each of thiL, yjeE, rimI, and ygjD (B). Interactions were tested reciprocally, and each block of four bars also has two controls where the gene of interest cloned in either pT18 or pT25 was tested reciprocally against the empty plasmid vector lacking either ygjD or yeaZ. Strain BTH101 was cotransformed with the pT18 and pT25 plasmid derivatives, a single colony was picked from each transformation and cultured in LB medium supplemented with appropriate antibiotics, and β-galactosidase assays were carried out using the method of Miller (43). Error bars represent standard errors of the means; n = 3. (C) BTH101 was cotransformed with pT18 and pT25, pT18yjeE and pT25yeaZ, or pT18yjeEygjD and pT25yeaZ. β-Galactosidase assays were carried out using the method of Miller (43). Error bars represent standard errors of the means; n = 3.

The interaction of YeaZ with proteins encoded in its genetic neighborhood in gram-positive bacteria is shown in Fig. 5B. Coexpression of pT18yeaZ and pT25yeaZ in strain BTH101 gave strongly red colonies on MacConkey-maltose indicator medium and a high level of β-galactosidase activity, suggesting that, like YjeE and YgjD, this essential protein also forms homo-oligomers. As shown in Fig. 5B, the interaction of YeaZ with YjeE and YgjD was also confirmed. When YeaZ was produced as a fusion protein with T18, there was no apparent evidence of interaction with either T25-ThiL or T25-RimI. When RimI was produced as a T18 fusion protein and YeaZ as a T25 fusion, the level of β-galactosidase activity that was measured was slightly higher, suggesting that there may be some weak interaction between the two proteins that is dependent on copy number or on positional effects of the protein fusion. We were unable to clone thiL into pT18 and so were unable to test interactions with YeaZ reciprocally. However, taken together, the data presented here indicate that YeaZ can homo-oligomerize and, in addition, is able to form complexes with YgjD and YjeE.

In addition to the targeted two-hybrid analyses described above, we also screened six random genomic libraries cloned into the two-hybrid vector pUT18, using pT25yeaZ, pT25ygjD, or pT25yjeE as bait. After screening out false-positive interactors, as described in Materials and Methods, a single positive clone that gave a consistent and specific interaction with pT25ygjD (β-galactosidase activity measurements of 1,020 ± 33 Miller units [mean ± standard deviation]) and one which gave a consistent and specific interaction with pT25yeaZ (β-galactosidase activity measurements of 2,682 ± 111 Miller units) were identified. In both cases, the clone was identical and harbored a fragment of yeaZ encoding amino acids 1 to 213 fused in frame to the N terminus of T18. When this clone was tested against pT25yjeE, a significant level of interaction was also seen (β-galactosidase activity measurements of 861 ± 33 Miller units). It is not clear why we did not isolate this truncated yeaZ clone when we carried out our screening using pT25yjeE as bait, but it may be that we did not obtain sufficient coverage during our screening experiments with this bait plasmid.

YeaZ may partner switch between YjeE and YgjD. Using the bacterial two-hybrid system described above, we have demonstrated that the essential protein YeaZ can apparently interact with the other two essential proteins YjeE and YgjD. The fact that these proteins may interact and that they are all essential for E. coli viability suggests that they may be involved in the same pathway. However, we have been unable to demonstrate any interaction between YjeE and YgjD, which therefore places YeaZ as the central component in this protein network.

To explore this further, we next investigated whether in the presence of overproduced YgjD, the interaction between YeaZ and YjeE that was seen using the bacterial two-hybrid assay was affected. To perform these experiments, we modified our pT18yjeE vector by cloning into this the gene for native ygjD. The ygjD gene was cloned directly after yjeET18 gene fusion and was thus under the control of the same promoter. As shown in Fig. 5C, β-galactosidase activity measurements for BTH101 harboring pT18yjeEygjD and pT25yeaZ were very low, almost indistinguishable from the level in the negative control. The simplest explanation for this drop in β-galactosidase activity when ygjD is also co-overexpressed is that the interaction of YeaZ with YjeE and YgjD is mutually exclusive and that under the growth conditions tested, the preferred partner for YeaZ is YgjD. This might also explain why in the native-level affinity purification experiments carried out by Butland et al. (9) under standard growth conditions, the only copurifying partner protein identified for affinity-tagged YeaZ was YgjD.

The YjeE, YeaZ, and YgjD depletion strains can all be partially complemented by plasmids harboring rstA. Since our experimental evidence was strongly suggestive of an interrelationship between YjeE, YeaZ, and YgjD, we went on to test whether the normally lethal depletion of any of these proteins could be complemented by overproduction of the putative partner proteins. The outcomes of these complementation analyses are shown in Fig. 6A. It can clearly be seen that the lack of growth of the yjeE depletion strain, JH17, in the presence of glucose could be overcome in the presence of multicopy yjeE but not yeaZ or ygjD. Likewise, in the presence of glucose, the yeaZ depletion strain, JH19, could only grow if it harbored multicopy yeaZ and not yjeE or ygjD. Finally, depletion of chromosomal ygjD expression in strain JH22 could only be complemented by plasmid-borne ygjD and not yeaZ or yjeE.

It has been reported previously by Campbell et al. (10) that rstA is a multicopy suppressor of the lethal phenotype of yjeE depletion. To identify possible multicopy suppressors of the depletion phenotypes of our conditional yjeE, yeaZ, and ygjD strains, we carried out transformations of each of these strains with libraries carrying fragments of E. coli chromosomal DNA and plated the transformants onto medium containing glucose.

After transforming strain JH17 with approximately 0.8 million clones, we isolated 28 clones that supported growth of the strain after repression of the expression of chromosomal yjeE. Of those clones (listed in Table S2 in the supplemental material), 23 harbored the yjeE gene, including 1 which contained only the coding sequence for the first 85 amino acids (from a total length of 152) of the protein. Of the five clones that we isolated as multicopy suppressors of the yjeE depletion phenotype that did not harbor yjeE, one of these, clone 17.2.3, carried the entire purE gene and the first one-third of purK. Both of these genes are involved in the de novo pathway for purine biosynthesis. A further clone, 17.2.27, harbored most of the yieG gene. This gene, which is currently uncharacterized, is predicted with high confidence to encode an inner-membrane protein of the xanthine-uracil permease superfamily. The isolation of these clones again suggests a link between DNA synthesis and YjeE. Of the remaining clones, 17.2.4 encodes fragments of the uncharacterized proteins YfaT and YfaS, the latter of which is a predicted alpha-2-macroglobulin domain protein; clone 17.1.18 harbors most of the ygfF gene, which is predicted to encode a short-chain dehydrogenase; and clone 17.7.20 carries fragments of two genes, ymdA and ymdB, that form part of a cryptic curli-encoding operon.

After transforming strain JH19 with approximately 1.15 million clones, we isolated 35 plasmids that supported growth of the strain after suppression of the expression of chromosomal yeaZ. However, all of these complementing clones (listed in Table S3 in the supplemental material) harbored at least a portion of the yeaZ gene. The minimal complementing clone encoded only the first 178 amino acids of the 231-amino-acid YeaZ, indicating that even a significantly C-terminally truncated form of YeaZ retains some functionality.

After transforming strain JH22 with more than 2 million clones, we isolated only 19 plasmids that supported growth of the strain after suppression of the expression of chromosomal ygjD. Of those clones (listed in Table S4 in the supplemental material), 17 harbored all or part of ygjD, with the minimally complementing clone encoding just the first 256 amino acids (from a total of 337). Two of the complementing plasmids did not contain any part of ygjD. Clone 22.2.33 carries the entire ybiV gene, which encodes a probable sugar phosphatase (54), along with very small fragments of ybiW and ybiU. Most interestingly, however, clone 22.7.27 carries a cluster of four genes, one of which is rstA (Fig. 2B), which has been previously identified as a multicopy suppressor of the lethal phenotype of a strain depleted for yjeE. To test whether clone 22.7.27 could also complement our yjeE and yeaZ depletion strains, we transformed this construct into strains JH17 and JH19, respectively, and plated the new strains onto glucose-containing medium. As shown in Fig. 6A, this plasmid could clearly also act as a multicopy suppressor of the depletion phenotype for strains lacking YjeE and YeaZ, as well as YgjD.

One possibility for the ability of clone 22.7.27 to complement the loss of any of the proteins YjeE, YeaZ, and YgjD is that it may interfere with the regulation of the P_araBAD promoter such that there is insufficient repression of the expression of the essential gene. The studies of Campbell et al. (10) already ruled out this possibility by confirming that multicopy rstA did not result in any overexpression of a luciferase reporter gene cloned under the control of the P_BAD promoter. However, we also confirmed this by showing that clone 22.7.27 could not support the growth of an otherwise isogenic conditionally lethal lepB strain in the presence of glucose (data not shown). None of the other clones that we identified as multicopy suppressors for JH17, JH19, or JH22 could cross-complement any of the other strains, indicating that clone 22.7.27 was unique in this respect (data not shown). Since we have isolated a single clone that is able to suppress the lethality of yjeE, yeaZ, or ygjD depletion, this is very strong evidence that these three genes are involved in a related pathway.

The arrangement of genes on the multicopy suppressor clone 22.7.27 is shown in Fig. 6B. Since Campbell et al. (10) showed that rstA could specifically complement the normally lethal depletion of yjeE, it seemed likely that the presence of rstA was responsible for the multicopy suppression seen with this plasmid. To test this, we first removed the ydgC and partial folM reading frames to give clone 126. As shown in Fig. 6A, this clone could also act as a multicopy suppressor for JH17, JH19, and JH22 grown in the presence of glucose, although the complementation of JH22 was not as strong as that seen with the full-length clone. When we subcloned rstA alone under the sole control of the constitutive tatA promoter in the medium-copy-number plasmid pT7.5, it could not complement any of the three conditional expression strains (not shown). However, when we subcloned the P_tatA-rstA fragment into the high-copy-number vector pBluescript so that it was also under the control of the lac promoter (clone 116), the clone was able to complement all three of the conditional expression strains grown in the presence of glucose. Thus, the expression level of rstA is critical for the complementation of JH17, JH19, and JH22, in agreement with the findings of Campbell et al. (10), who demonstrated that high levels of expression of rstA were critical for its ability to complement their conditional yjeE expression strain.

Purified YjeE, YeaZ, and YgjD form homodimers. As a prerequisite to the ultimate characterization of complexes formed between YjeE, YeaZ, and YgjD, we overproduced and purified each of the three essential proteins. YjeE, which was supplied with a C-terminal His tag, was purified initially by immobilized metal affinity chromatography, followed by size exclusion chromatography. As shown in Fig. 7A, left panel, YjeE-His₆ (confirmed by peptide mass fingerprinting) eluted from the size exclusion column in two distinct peaks, corresponding to molecular masses of 39 kDa and 22 kDa, respectively, which are close to the estimated masses for dimeric (36 kDa) and monomeric (18 kDa) forms of YjeE-His₆. These data confirm the results of the bacterial two-hybrid analysis (Fig. 4A), which gave a strong indication that YjeE could self-interact. We also confirmed that the as-purified YjeE-His₆ protein had ATPase activity with an estimated K_m of 169 µM and a k_cat of 0.0026 s^–1 (data not shown), which compares favorably with the values estimated by Teplyakov et al. (60) for the H. influenzae YjeE protein, Allali-Hassani et al. (3) for E. coli YjeE, and Karst et al. (34) for B. subtilis YdiB. In this latter publication, the authors also demonstrated that ATPase activity was higher with monomeric than with oligomeric YdiB (34).

View larger version (38K): [in this window] [in a new window]

FIG. 7. In vitro characterization of YjeE, YeaZ, and YgjD. (A) Size exclusion chromatographic analysis of YjeE-His6 (left), YeaZ (center), and YgjD-His6 (right). The insets in each panel show SDS-PAGE analysis of the source of each of the peaks indicated on the chromatogram. Vo indicates the position of the void volume of the column. (B) SDS-PAGE analysis of E. coli strain M15(pREP4)(pQEYgjDhis) prior to (U) and 15 h after (I) induction of YgjD-His6 production by addition of IPTG and of peak fraction containing YgjD-His6 following elution from a nickel-charged immobilized metal affinity column (imac). The bands indicated with arrows were subjected to tryptic mass peptide fingerprint analysis. (C) Silver-stained SDS-PAGE analysis of YgjD-SPA (lane 1) and YeaZ-SPA (lane 2) affinity-purified eluates. The proteins indicated as present in specific bands were identified by matrix-assisted laser desorption ionization-time of flight peptide mass fingerprint analysis. (D) Left, SDS-PAGE analysis of as-purified YgjD-His6 (8 µg of protein) and YeaZ (10 µg of protein); right, purified YgjD-His6 (8 µg of protein) was incubated alone or in the presence of 10 µg of purified YeaZ in a final reaction mixture volume of 20 µl for 180 min at 37°C, after which the samples were quenched by the addition of SDS loading buffer and immediately analyzed by SDS-PAGE. (E) Purified YjeE-His6 (10 µg of protein) and purified YeaZ (20 µg of protein) were incubated separately or together for 180 min at 37°C in a final reaction mixture volume of 20 µl, after which the samples were quenched and analyzed as described for panel D. For the SDS-PAGE analyses in all panels, the molecular mass markers are indicated to the left, with sizes in kDa.

For purification of YeaZ, we overproduced the native protein as described in Methods in the supplemental material and isolated YeaZ following Q-Sepharose ion exchange chromatography and size exclusion chromatography. The protein was confirmed to be native YeaZ by peptide mass fingerprint analysis. As shown in Fig. 7A, center panel, YeaZ was found in two distinct peaks after size exclusion chromatography, a major peak corresponding to a molecular mass of 62 kDa and a more minor peak at a mass of 27 kDa, close to the estimated masses of dimeric (50 kDa) and monomeric (25 kDa) forms of YeaZ. This observation supports the contention resulting from bacterial two-hybrid analysis that YeaZ has a strong propensity to homodimerize.

YgjD was overproduced as a C-terminally histidine-tagged fusion protein, as described in Methods in the supplemental material. Interestingly, as shown in Fig. 7B in the lane labeled imac, following purification of YgjD-His₆ by nickel affinity chromatography, we observed that the protein had been subject to degradation, despite the inclusion of a protease inhibitor cocktail (Roche Complete EDTA-free) throughout the cell lysis and purification steps. Analysis of the bands numbered 1 to 4 in Fig. 7B indicated that each of them was derived from YgjD/YgjD-His₆. Band number 1 corresponded to full-length YgjD-His₆, while band 2 was an N-terminal fragment of the protein, most probably formed from cleavage of the protein between F₁₉₅ and V₁₉₆. The molecular mass of band 3 is consistent with it also being an N-terminal fragment of the protein, generated following cleavage between amino acids K₁₇₁ and L₁₇₂. Although band 4 was also clearly derived from YgjD, it was not possible to determine a cleavage site for this form of the protein since tryptic fragments from both the N- and C-terminal regions of the protein were detected. It may be that band 4 contains a mixture of N- and C-terminally processed forms of YgjD-His₆.

We were able to separate out the full-length form of YgjD-His₆ from the proteolysis products following size exclusion chromatography (Fig. 7A, right panel). The protein eluted predominantly as a single peak after gel filtration, with an estimated molecular mass of 66 kDa, close to the expected mass of a YgjD-His₆ dimer (74 kDa). The small shoulder, labeled 1 in the right panel in Fig. 7A, which corresponds to an estimated mass of 111 kDa, also contained low levels of a putative trimer form of YgjD-His₆. These observations again support the conclusions from bacterial two-hybrid analysis that YgjD can oligomerize. None of the three essential proteins that we purified showed any evidence of any copurifying partner proteins, which may be because they are so massively overproduced that any interacting factors are out-titrated.

Preliminary analysis of the interaction between YeaZ and YgjD indicates that YeaZ mediates proteolysis of YgjD. The results presented above clearly show that YgjD-His₆ is very sensitive to proteolytic degradation in crude cell extracts, even in the presence of added protease inhibitors. Interestingly, when YgjD was purified at native levels as a SPA-tag fusion protein, it was also seen to be subject to proteolysis (Fig. 7C). As reported previously, the native-level purification of tagged YgjD clearly also resulted in coisolation of the nontagged partner protein YeaZ. When the reciprocal experiment was performed, i.e., when native-level SPA-tagged YeaZ was isolated, nontagged YgjD was copurified, and again it was observed that YgjD was subject to proteolytic degradation, even though a protease inhibitor cocktail was also included throughout the purification steps. This observation rules out an effect of C-terminal tagging on the susceptibility of YgjD to proteolysis and indicates that the native protein is intrinsically sensitive to proteolytic cleavage.

Since proteolysed YgjD was found at native expression levels complexed to YeaZ and both of these proteins are predicted to be proteases, this raised the possibility that the proteolytic degradation of YgjD seen here was mediated by YeaZ. To investigate this further, we dialyzed YgjD-His₆ that had been purified by immobilized metal chelate chromatography to remove the excess imidazole and incubated it for 3 h at 37°C in the presence or absence of purified YeaZ. It can clearly be seen in Fig. 7D (right panel) that when YeaZ was added to the incubation mixture, YgjD-His₆ was fully degraded, whereas incubation of YgjD-His₆ alone was without effect. Therefore, we conclude that the proteolysis of YgjD-His₆ is mediated by YeaZ. When we additionally included purified YjeE in the incubation mixture, it did not prevent the degradation of YgjD-His₆ mediated by YeaZ (data not shown).

While our experiments do not distinguish the possibility that YgjD-His₆ is undergoing autoproteolysis that is activated by YeaZ, we suggest that it is more likely that the proteolysis of YgjD is catalyzed by YeaZ because (i) YeaZ overproduced in the absence of zinc supplementation of the growth medium was inactive in bringing about the proteolysis of YgjD-His₆ and (ii) purification of a C-terminally His-tagged variant of YeaZ by immobilized metal affinity chromatography also yielded tagged YeaZ that was inactive for YgjD-His₆ proteolysis, but if the tagged variant was purified by anion exchange and gel filtration chromatography, it was functional for catalyzing YgjD-His₆ degradation (data not shown). However, we suggest that if YeaZ is a protease, it is a highly specific one, since it was not able to mediate significant degradation of YjeE-His₆ (Fig. 7E) and when it was incubated with an E. coli crude cytoplasmic extract and analyzed by 2-D gel electrophoresis, no protein spots disappeared, but if purified YgjD-His₆ was added to the same mixture of YeaZ with crude cytoplasmic proteins, it was still specifically degraded (not shown). Thus, our results point strongly toward the contention that the essential cellular protein, YeaZ, is a protease and that YgjD is its major, if not only, substrate.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In this paper, we report for the first time that the three essential eubacterial proteins YjeE, YeaZ, and YgjD form an interaction network. Numerous lines of evidence support this contention, including the fact that YeaZ interacts with YjeE and YgjD in bacterial two-hybrid experiments, YeaZ forms a complex with YgjD at native levels that is stable to purification, and strains with depleted production of each of these essential proteins can be rescued by a multicopy plasmid harboring the rstA gene. Our observations help to explain why these three proteins are encoded in the genomes of almost all eubacteria and why they are all, in general, critical for growth, since they all interact in the same essential, unknown cellular pathway.

Of these three proteins, YjeE and YeaZ are specific to eubacteria, whereas YgjD is ubiquitous, also having an essential function in eukaryotes. Interestingly, in archaea and eukaryotes, it appears that the function of YgjD, which has been termed Kae1 (for kinase-associated endopeptidase 1) is regulated by a protein kinase, Bud32. This is supported by the fact that fusions of Kae1 with a Bud32 ortholog are encoded in the genomes of some archaea and that Kae1 and Bud32 have been found in the same protein complex as Kae1 in yeast (15, 35, 41). Indeed, crystallographic evidence has shown very tight complex formation between Kae1 and Bud32 in the KEOPS complex and in the fusion protein from Methanocaldococcus jannaschii, where it was also revealed that Bud32 maintains the ATP-binding site of Kae1 in an inactive configuration (26a, 26b). Thus, it seems that one of the functions of Bud32 is to regulate the activity of Kae1. Since there is no eubacterial ortholog of Bud32, we propose that the regulation of YgjD activity in these organisms is achieved through the action of YeaZ and YjeE. Possible models for the regulatory network between YjeE, YeaZ, and YgjD are shown in Fig. 8.

View larger version (14K): [in this window] [in a new window]

FIG. 8. Possible models for the interaction network formed between YjeE, YeaZ, and YgjD. Both models assume that under nonstress conditions, the default situation is that YeaZ is complexed with YgjD rather than YjeE. (A) The complex of YeaZ and YgjD results in the proteolysis and inactivation of YgjD. When cells are stressed (signified by the arrow marked "Input signal"), YjeE complexes with YeaZ, either liberating YgjD or allowing newly synthesized YgjD to accumulate and perform its essential cellular function. (B) The complex of YeaZ and YgjD results in the proteolysis and activation of YgjD, with YgjD carrying out its essential housekeeping functions. An input signal results in YeaZ forming a complex with YjeE, leaving YgjD uncomplexed and inactive. In both situations, nucleotide exchange or hydrolysis by YjeE is proposed to reset the switch, returning YjeE to the noncomplexed state.

Prior experiments have shown that YjeE is an ATP-binding protein with low intrinsic ATPase activity, and it has been speculated that the protein acts as a molecular switch (60). Our findings are in keeping with this proposal since the interaction between YjeE and YeaZ that we see using the bacterial two-hybrid assay is disrupted in the presence of multicopy ygjD. We therefore propose that one of the main functions of YjeE is to control or regulate the interaction between YeaZ and YgjD. Since YjeE can interact with adenine nucleotides and catalyze the slow hydrolysis of ATP, one way that complex formation between YjeE and YeaZ may be modulated is by nucleotide binding or hydrolysis. To investigate this further, we incubated cell extracts of a strain producing, at the native level, a SPA-tagged variant of YjeE with ADP or nonhydrolysable analogs of ATP (AMP-PNP) or GTP (GMP-PNP) and then isolated YjeE using the protein tag. However, none of these nucleotides was sufficient to promote the formation of a complex between YjeE-SPA and YeaZ that was stable to purification (data not shown). It is possible that the interaction between YjeE and YeaZ is disrupted by the conditions used during protein isolation, or alternatively, the interaction may be transient or require the input of further proteins or other cofactors, such as metabolites, that are not present at sufficient levels during the growth of these strains in rich medium.

Clearly, however, under the same growth conditions, YeaZ and YgjD do form a strong complex. This is backed up by the observations of Butland et al. (9) that at native levels of expression, YeaZ and YgjD copurify when E. coli strains are cultured aerobically in rich growth medium. It was also observed on MacConkey indicator medium and during liquid growth in LB that in bacterial two-hybrid experiments, the preferred partner for YeaZ is YgjD. Thus, according to the models shown in Fig. 8, the "default" situation is that YgjD is complexed with YeaZ. One of the most striking observations is that within this complex, YgjD is subject to proteolysis. The results of experiments where we mixed together the purified YeaZ with YgjD strongly suggest that this proteolysis is catalyzed by YeaZ. This complex formation and proteolysis of YgjD by YeaZ could serve either of two functions. It could serve to inactivate the function of YgjD, as depicted in Fig. 8B, such that it is unable to carry out its essential cellular function(s). Alternatively, it may serve to activate YgjD (Fig. 8A) and it may be that a truncated form of YgjD is the active one. A prediction of this second model is that the production of a truncated variant of YgjD equivalent to that generated by proteolysis of the protein may be able to suppress the lethal phenotype of yeaZ and/or ygjD depletion. However, the expression of constructs encoding either the first 171 or 195 amino acids of YgjD was unable to support growth of the yeaZ or ygjD depletion strain in the presence of glucose (not shown), although it should be noted that we isolated a functional construct containing amino acids 1 to 256 of ygjD in our library screen for suppressors of ygjD depletion. Nonetheless, taken together, our results are consistent with the proteolysis of YgjD resulting in its inactivation.

What are the cellular roles of YjeE-YeaZ-YgjD? The exact functions of the three key cellular proteins we have studied here have so far been enigmatic. However, it is clear that the loss of any of these three proteins results in startling changes to the morphology of E. coli. Depletion of yjeE and ygjD similarly gave rise to (at least a subset of) very enlarged cells that had defective morphologies, consistent with a link to cell envelope processes. It should be noted that conditional mutation of gcp in Staphylococcus aureus has also been linked to effects on cell wall turnover (74). In each case, the enlarged cells seen during yjeE and ygjD depletion showed a highly unusual pattern of DNA distribution that has only been seen previously when E. coli strains are starved for thymine. However, an effect on the rate of DNA synthesis cannot be the sole reason for the loss of viability of the yjeE and ygjD conditional expression strains, as we were unable to rescue their growth by supplementation of the growth medium with any of thymine (at 5, 50, and 500 µg/ml), uracil (at 10, 100, and 1000 µg/ml), or adenine (at 10 and 100 µg/ml).

The YgjD ortholog of M. haemolytica is the only protein of the three we have studied here that has been previously ascribed a function. The protein, which was termed Gcp, was isolated from culture supernatants of M. haemolytica A serotypes and shown to possess glycoprotease activity that was specific for O-sialoglycoproteins, such as glycophorin A (1, 2, 63). It is not clear how the secretion of Gcp is mediated, since the secreted protein was not N-terminally processed and lacked any recognizable signal sequence; however, when gcp was expressed in E. coli, the gene product was found in the periplasmic fraction (1). Gcp is 77% identical (88% similar) to E. coli YgjD, making it extremely likely that the two proteins share a common function. Interestingly, fractionation of E. coli cells producing SPA-tagged YgjD at native levels followed by Western blotting against the FLAG epitope revealed that the E. coli protein has a strictly cytoplasmic location (as did SPA-tagged YjeE and YeaZ; data not shown). Moreover, we were unable to demonstrate any sialoglycoprotease activity against glycophorin A of either purified YgjD-His₆ or YeaZ (data not shown), and, in addition, we were also unable to detect, following staining with Pro-Q Emerald, the presence of glycoproteins in cell extracts of wild-type E. coli strains; in any of our yjeE, yeaZ, or ygjD depletion strains; or in strains overproducing YjeE, YeaZ, or YgjD (data not shown).

Despite the very high sequence identity between E. coli YgjD and M. haemolytica Gcp, we were also unable to suppress the lethal depletion of ygjD by the expression of gcp (data not shown), suggesting that despite their high sequence conservation, the two proteins have diverged in function, possibly resulting in an inability to interact with specific cellular targets or partner proteins. Nonetheless, given the homology between YeaZ and YgjD, the fact that Gcp cleaves peptide bonds, and our observations which show that YeaZ is a specific protease that degrades YgjD, we speculate that YgjD is also a protease. However, the exact cellular target(s) of YgjD are currently not known. Overproduction of YgjD or YgjD plus YeaZ in cells is not lethal, and there is no apparent wholesale degradation of cellular proteins (e.g., see Fig. 7B), nor does supplementation of crude cell extracts with purified YgjD-His₆ followed by analysis by 2-D gel electrophoresis result in any apparent degradation of the proteins (not shown). Therefore, we speculate that the proteins may need to be modified in some way to mark them as targets for YgjD. Clearly, glycosylation may be one such way to modify the proteins, but other modifications also exist, some of which, such as acetylation (69) or nonenzymatic glycation (4, 45), have been observed in eubacteria. Interestingly, glycated proteins arise when intermediates in sugar metabolism, such as methylglyoxal (67) or 6-phosphogluconolactone (4), accumulate, resulting in protein inactivation and toxic side effects such as protein precipitation (47), suggesting that a cellular mechanism should exist for the targeted degradation of such products.

Finally, we note that multicopy rstA is able to suppress the otherwise lethal phenotype of depletion of yjeE, yeaZ, or ygjD. RstA is the response regulator of a two-component system where RstB is the cognate histidine kinase (66). The rstAB genes are under the transcriptional control of a second two-component system, PhoPQ, which is involved in magnesium homeostasis (44). The RstAB regulon has not currently been well defined, although a genomic approach has identified two promoters that are bound by RstA, one of which was that for csgD (53). Interestingly, CsgD regulates the curli biosynthetic operon that includes ymdA and ymdB (8), which were also found to be multicopy suppressors for the yjeE depletion strain. However, these genes in multicopy did not suppress the lethality associated with yeaZ or ygjD depletion, and therefore, it is likely that other genes controlled by RstA are also important to rescue growth in the absence of YeaZ and YgjD. Clearly, further definition of the RstAB regulon will be required to understand the basis of the suppression seen here.

To summarize, the work described in this study is the first to link the three essential E. coli proteins YjeE, YeaZ, and YgjD to the same essential cellular process. Our evidence indicates that the major, if not only roles, of YjeE and YeaZ are to regulate the activity of YgjD, most probably via proteolytic degradation. Our studies pave the way for a more complete study of the interaction between these three key proteins and the essential processes which they control.

 
We thank Pascale Ferrigno, Yiliang Ding, Alex Graf, and Michael Rose for their help in constructing some of the plasmids that were used in this study and Govind Chandra for his assistance in genome searching for YjeE, YeaZ, and YgjD orthologs. Robert Davies is thanked for providing us with Mannheimia haemolytica genomic DNA. We thank Jeff Errington, Simon Foster, Mark Buttner, Simon Andrews, Alison Hunt, and Matt Hutchings for helpful discussion and Gary Sawers and Nicola Stanley-Wall for critical reading of the manuscript.

This work is supported by the BBSRC through grant BB/D000386/1, via a Ph.D. studentship to J.H., and through the MRC by the awarding of an MRC Senior Non-Clinical Fellowship to T.P.

 
* Corresponding author. Mailing address: Division of Molecular Microbiology, College of Life Sciences, Dow Street, University of Dundee, Dundee DD1 5EH, United Kingdom. Phone: 44-01382 386464. Fax: 44-01382 388216. E-mail: t.palmer{at}dundee.ac.uk

Published ahead of print on 17 April 2009.

Supplemental material for this article may be found at http://jb.asm.org/.

These authors contributed equally to this work.

Present address: Laboratoire d'Ingénierie des Systèmes Macromoléculaires, Institut de Microbiologie de la Méditerranée, UPR 9027, 31 Chemin Joseph Aiguier, 13402 Marseille Cedex 20, France.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Abdullah, K. M., R. Y. Lo, and A. Mellors. 1991. Cloning, nucleotide sequence, and expression of the Pasteurella haemolytica A1 glycoprotease gene. J. Bacteriol. 173:5597-5603.[Abstract/Free Full Text]
2
2. Abdullah, K. M., E. A. Udoh, P. E. Shewen, and A. Mellors. 1992. A neutral glycoprotease of Pasteurella haemolytica A1 specifically cleaves O-sialoglycoproteins. Infect. Immun. 60:56-62.[Abstract/Free Full Text]
3
3. Allali-Hassani, A., T. L. Campbell, A. Ho, J. W. Schertzer, and E. D. Brown. 2004. Probing the active site of YjeE: a vital Escherichia coli protein of unknown function. Biochem. J. 384:577-584.[CrossRef][Medline]
4
4. Aon, J. C., R. J. Caimi, A. H. Taylor, Q. Lu, F. Oluboyede, J. Dally, M. D. Kessler, J. J. Kerrigan, T. S. Lewis, L. A. Wysocki, and P. S. Patel. 2008. Suppressing posttranslational gluconoylation of heterologous proteins by metabolic engineering of Escherichia coli. Appl. Environ. Microbiol. 74:950-958.[Abstract/Free Full Text]
5
5. Arigoni, F., F. Talabot, M. Peitsch, M. D. Edgerton, E. Meldrum, E. Allet, R. Fish, T. Jamotte, M. L. Curchod, and H. Loferer. 1998. A genome-based approach for the identification of essential bacterial genes. Nat. Biotechnol. 16:851-856.[CrossRef][Medline]
6
6. Baba, T., T. Ara, M. Hasegawa, Y. Takai, Y. Okumura, M. Baba, K. A. Datsenko, M. Tomita, B. L. Wanner, and H. Mori. 2006. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol. Syst. Biol. 2:2006.0008.[Medline]
7
7. Barry, C. E., III, S. F. Hayes, and T. Hackstadt. 1992. Nucleoid condensation in Escherichia coli that express a chlamydial histone homolog. Science 256:377-379.[Abstract/Free Full Text]
8
8. Brombacher, E., A. Baratto, C. Dorel, and P. Landini. 2006. Gene expression regulation by the curli activator CsgD protein: modulation of cellulose biosynthesis and control of negative determinants for microbial adhesion. J. Bacteriol. 188:2027-2037.[Abstract/Free Full Text]
9
9. Butland, G., J. M. Peregrin-Alvarez, J. Li, W. Yang, X. Yang, V. Canadien, A. Starostine, D. Richards, B. Beattie, N. Krogan, M. Davey, J. Parkinson, J. Greenblatt, and A. Emili. 2005. Interaction network containing conserved and essential protein complexes in Escherichia coli. Nature 433:531-537.[CrossRef][Medline]
10
10. Campbell, T. L., C. S. Ederer, A. Allali-Hassani, and E. D. Brown. 2007. Isolation of the rstA gene as a multicopy suppressor of YjeE, an essential ATPase of unknown function in Escherichia coli. J. Bacteriol. 189:3318-3321.[Abstract/Free Full Text]
11
11. Casadaban, M. J. 1976. Transposition and fusion of the lac genes to selected promoters in Escherichia coli using bacteriophage lambda and Mu. J. Mol. Biol. 104:541-555.[CrossRef][Medline]
12
12. Casadaban, M. J., and S. N. Cohen. 1980. Analysis of gene control signals by DNA fusion and cloning in Escherichia coli. J. Mol. Biol. 138:179-207.[CrossRef][Medline]
13
13. Cherepanov, P. P., and W. Wackernagel. 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.[CrossRef][Medline]
14
14. Datsenko, K. A., and B. L. Wanner. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97:6640-6645.[Abstract/Free Full Text]
15
15. Downey, M., R. Houlsworth, L. Maringele, A. Rollie, M. Brehme, S. Galicia, S. Guillard, M. Partington, M. K. Zubko, N. J. Krogan, A. Emili, J. F. Greenblatt, L. Harrington, D. Lydall, and D. Durocher. 2006. A genome-wide screen identifies the evolutionarily conserved KEOPS complex as a telomere regulator. Cell 124:1155-1168.[CrossRef][Medline]
16
16. Dubochet, J., A. W. McDowall, B. Menge, E. N. Schmid, and K. G. Lickfeld. 1983. Electron microscopy of frozen-hydrated bacteria. J. Bacteriol. 155:381-390.[Abstract/Free Full Text]
17
17. Freiberg, C., B. Wieland, F. Spaltmann, K. Ehlert, H. Brotz, and H. Labischinski. 2001. Identification of novel essential Escherichia coli genes conserved among pathogenic bacteria. J. Mol. Microbiol. Biotechnol. 3:483-489.[CrossRef][Medline]
18
18. Gallagher, L. A., E. Ramage, M. A. Jacobs, R. Kaul, M. Brittnacher, and C. Manoil. 2007. A comprehensive transposon mutant library of Francisella novicida, a bioweapon surrogate. Proc. Natl. Acad. Sci. USA 104:1009-1014.[Abstract/Free Full Text]
19
19. Gerdes, S. Y., M. D. Scholle, J. W. Campbell, G. Balazsi, E. Ravasz, M. D. Daugherty, A. L. Somera, N. C. Kyrpides, I. Anderson, M. S. Gelfand, A. Bhattacharya, V. Kapatral, M. D'Souza, M. V. Baev, Y. Grechkin, F. Mseeh, M. Y. Fonstein, R. Overbeek, A. L. Barabasi, Z. N. Oltvai, and A. L. Osterman. 2003. Experimental determination and system level analysis of essential genes in Escherichia coli MG1655. J. Bacteriol. 185:5673-5684.[Abstract/Free Full Text]
20
20. Giaever, G., A. M. Chu, L. Ni, C. Connelly, L. Riles, S. Veronneau, S. Dow, A. Lucau-Danila, K. Anderson, B. Andre, A. P. Arkin, A. Astromoff, M. El-Bakkoury, R. Bangham, R. Benito, S. Brachat, S. Campanaro, M. Curtiss, K. Davis, A. Deutschbauer, K. D. Entian, P. Flaherty, F. Foury, D. J. Garfinkel, M. Gerstein, D. Gotte, U. Guldener, J. H. Hegemann, S. Hempel, Z. Herman, D. F. Jaramillo, D. E. Kelly, S. L. Kelly, P. Kotter, D. LaBonte, D. C. Lamb, N. Lan, H. Liang, H. Liao, L. Liu, C. Luo, M. Lussier, R. Mao, P. Menard, S. L. Ooi, J. L. Revuelta, C. J. Roberts, M. Rose, P. Ross-Macdonald, B. Scherens, G. Schimmack, B. Shafer, D. D. Shoemaker, S. Sookhai-Mahadeo, R. K. Storms, J. N. Strathern, G. Valle, M. Voet, G. Volckaert, C. Y. Wang, T. R. Ward, J. Wilhelmy, E. A. Winzeler, Y. Yang, G. Yen, E. Youngman, K. Yu, H. Bussey, J. D. Boeke, M. Snyder, P. Philippsen, R. W. Davis, and M. Johnston. 2002. Functional profiling of the Saccharomyces cerevisiae genome. Nature 418:387-391.[CrossRef][Medline]
21
21. Gil, R., F. J. Silva, J. Pereto, and A. Moya. 2004. Determination of the core of a minimal bacterial gene set. Microbiol. Mol. Biol. Rev. 68:518-537.[Abstract/Free Full Text]
22
22. Glass, J. I., N. Assad-Garcia, N. Alperovich, S. Yooseph, M. R. Lewis, M. Maruf, C. A. Hutchison III, H. O. Smith, and J. C. Venter. 2006. Essential genes of a minimal bacterium. Proc. Natl. Acad. Sci. USA 103:425-430.[Abstract/Free Full Text]
23
23. Gust, B., G. L. Challis, K. Fowler, T. Kieser, and K. F. Chater. 2003. PCR-targeted Streptomyces gene replacement identifies a protein domain needed for biosynthesis of the sesquiterpene soil odor geosmin. Proc. Natl. Acad. Sci. USA 100:1541-1546.[Abstract/Free Full Text]
24
24. Guzman, L. M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose P_BAD promoter. J. Bacteriol. 177:4121-4130.[Abstract/Free Full Text]
25
25. Hamilton, C. M., M. Aldea, B. K. Washburn, P. Babitzke, and S. R. Kushner. 1989. New method for generating deletions and gene replacements in Escherichia coli. J. Bacteriol. 171:4617-4622.[Abstract/Free Full Text]
26
26. Hecker, A., N. Leulliot, D. Gadelle, M. Graille, A. Justome, P. Dorlet, C. Brochier, S. Quevillon-Cheruel, E. Le Cam, H. van Tilbeurgh, and P. Forterre. 2007. An archaeal orthologue of the universal protein Kae1 is an iron metalloprotein which exhibits atypical DNA-binding properties and apurinic-endonuclease activity in vitro. Nucleic Acids Res. 35:6042-6051.[Abstract/Free Full Text]
26
27. Hecker, A., R. Lopreiato, M. Graille, B. Collinet, P. Forterre, D. Libri, and H. van Tilbeurgh. 2008. Structure of the archaeal Kae1/Bud32 fusion protein MJ1130: a model for the eukaryotic EKC/KEOPS subcomplex. EMBO J. 27:2340-2351.[CrossRef][Medline]
26
28. Hecker, A., M. Graille, E. Madec, D. Gadelle, E. Le Cam, H. van Tilbergh, and P. Forterre. 2009. The universal Kae1 protein and the associated Bud32 kinase (PRPK), a mysterious protein couple probably essential for genome maintenance in Archaea and Eukarya. Biochem Soc Trans. 37:29-35.[CrossRef][Medline]
27
29. Hunt, A., J. P. Rawlins, H. B. Thomaides, and J. Errington. 2006. Functional analysis of 11 putative essential genes in Bacillus subtilis. Microbiology 152:2895-2907.[Abstract/Free Full Text]
28
30. Isono, K., and S. Isono. 1980. Ribosomal protein modification in Escherichia coli. II. Studies of a mutant lacking the N-terminal acetylation of protein S18. Mol. Gen. Genet. 177:645-651.[Medline]
29
31. Ize, B., F. Gerard, M. Zhang, A. Chanal, R. Voulhoux, T. Palmer, A. Filloux, and L. F. Wu. 2002. In vivo dissection of the Tat translocation pathway in Escherichia coli. J. Mol. Biol. 317:327-335.[CrossRef][Medline]
30
32. Jack, R. L., G. Buchanan, A. Dubini, K. Hatzixanthis, T. Palmer, and F. Sargent. 2004. Coordinating assembly and export of complex bacterial proteins. EMBO J. 23:3962-3972.[CrossRef][Medline]
31
33. Karandashova, I., I. Elanskaya, K. Marin, J. Vinnemeier, and M. Hagemann. 2002. Identification of genes essential for growth at high salt concentrations using salt-sensitive mutants of the cyanobacterium Synechocystis sp. strain PCC 6803. Curr. Microbiol. 44:184-188.[CrossRef][Medline]
32
34. Karimova, G., J. Pidoux, A. Ullmann, and D. Ladant. 1998. A bacterial two-hybrid system based on a reconstituted signal transduction pathway. Proc. Natl. Acad. Sci. USA 95:5752-5756.[Abstract/Free Full Text]
33
35. Karimova, G., A. Ullmann, and D. Ladant. 2001. Protein-protein interaction between Bacillus stearothermophilus tyrosyl-tRNA synthetase subdomains revealed by a bacterial two-hybrid system. J. Mol. Microbiol. Biotechnol. 3:73-82.[Medline]
34
36. Karst, J. C., A. E. Foucher, T. L. Campbell, A. M. Di Guilmi, D. Stroebel, C. S. Mangat, E. D. Brown, and J. M. Jault. 2009. The ATPase activity of an "essential" Bacillus subtilis enzyme, YdiB, is required for its cellular function and is modulated by oligomerization. Microbiology 155:944-956.[Abstract/Free Full Text]
35
37. Kisseleva-Romanova, E., R. Lopreiato, A. Baudin-Baillieu, J. C. Rousselle, L. Ilan, K. Hofmann, A. Namane, C. Mann, and D. Libri. 2006. Yeast homolog of a cancer-testis antigen defines a new transcription complex. EMBO J. 25:3576-3585.[CrossRef][Medline]
36
38. Knuth, K., H. Niesalla, C. J. Hueck, and T. M. Fuchs. 2004. Large-scale identification of essential Salmonella genes by trapping lethal insertions. Mol. Microbiol. 51:1729-1744.[CrossRef][Medline]
37
39. Kobayashi, K., S. D. Ehrlich, A. Albertini, G. Amati, K. K. Andersen, M. Arnaud, K. Asai, S. Ashikaga, S. Aymerich, P. Bessieres, F. Boland, S. C. Brignell, S. Bron, K. Bunai, J. Chapuis, L. C. Christiansen, A. Danchin, M. Debarbouille, E. Dervyn, E. Deuerling, K. Devine, S. K. Devine, O. Dreesen, J. Errington, S. Fillinger, S. J. Foster, Y. Fujita, A. Galizzi, R. Gardan, C. Eschevins, T. Fukushima, K. Haga, C. R. Harwood, M. Hecker, D. Hosoya, M. F. Hullo, H. Kakeshita, D. Karamata, Y. Kasahara, F. Kawamura, K. Koga, P. Koski, R. Kuwana, D. Imamura, M. Ishimaru, S. Ishikawa, I. Ishio, D. Le Coq, A. Masson, C. Mauel, R. Meima, R. P. Mellado, A. Moir, S. Moriya, E. Nagakawa, H. Nanamiya, S. Nakai, P. Nygaard, M. Ogura, T. Ohanan, M. O'Reilly, M. O'Rourke, Z. Pragai, H. M. Pooley, G. Rapoport, J. P. Rawlins, L. A. Rivas, C. Rivolta, A. Sadaie, Y. Sadaie, M. Sarvas, T. Sato, H. H. Saxild, E. Scanlan, W. Schumann, J. F. Seegers, J. Sekiguchi, A. Sekowska, S. J. Seror, M. Simon, P. Stragier, R. Studer, H. Takamatsu, T. Tanaka, M. Takeuchi, H. B. Thomaides, V. Vagner, J. M. van Dijl, K. Watabe, A. Wipat, H. Yamamoto, M. Yamamoto, Y. Yamamoto, K. Yamane, K. Yata, K. Yoshida, H. Yoshikawa, U. Zuber, and N. Ogasawara. 2003. Essential Bacillus subtilis genes. Proc. Natl. Acad. Sci. USA 100:4678-4683.[Abstract/Free Full Text]
38
40. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.[CrossRef][Medline]
39
41. Liberati, N. T., J. M. Urbach, S. Miyata, D. G. Lee, E. Drenkard, G. Wu, J. Villanueva, T. Wei, and F. M. Ausubel. 2006. An ordered, nonredundant library of Pseudomonas aeruginosa strain PA14 transposon insertion mutants. Proc. Natl. Acad. Sci. USA 103:2833-2838.[Abstract/Free Full Text]
40
42. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275.[Free Full Text]
41
43. Mao, D. Y., D. Neculai, M. Downey, S. Orlicky, Y. Z. Haffani, D. F. Ceccarelli, J. S. Ho, R. K. Szilard, W. Zhang, C. S. Ho, L. Wan, C. Fares, S. Rumpel, I. Kurinov, C. H. Arrowsmith, D. Durocher, and F. Sicheri. 2008. Atomic structure of the KEOPS complex: an ancient protein kinase-containing molecular machine. Mol. Cell 32:259-275.[CrossRef][Medline]
42
44. McCutcheon, J. P., and N. A. Moran. 2007. Parallel genomic evolution and metabolic interdependence in an ancient symbiosis. Proc. Natl. Acad. Sci. USA 104:19392-19397.[Abstract/Free Full Text]
43
45. Miller, J. H. 1992. A short course in bacterial genomics: a laboratory manual and handbook for Escherichia coli and related bacteria. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
44
46. Minagawa, S., H. Ogasawara, A. Kato, K. Yamamoto, Y. Eguchi, T. Oshima, H. Mori, A. Ishihama, and R. Utsumi. 2003. Identification and molecular characterization of the Mg²⁺ stimulon of Escherichia coli. J. Bacteriol. 185:3696-3702.[Abstract/Free Full Text]
45
47. Mironova, R., T. Niwa, H. Hayashi, R. Dimitrova, and I. Ivanov. 2001. Evidence for non-enzymatic glycosylation in Escherichia coli. Mol. Microbiol. 39:1061-1068.[CrossRef][Medline]
46
48. Morgan, C., H. S. Rosenkranz, H. S. Carr, and H. M. Rose. 1967. Electron microscopy of chloramphenicol-treated Escherichia coli. J. Bacteriol. 93:1987-2002.[Abstract/Free Full Text]
47
49. Munch, G., K. Berbaum, C. Urban, and R. Schinzel. 2005. Proteins of Thermus thermophilus are resistant to glycation-induced protein precipitation: an evolutionary adaptation to life at extreme temperatures? Ann. N. Y. Acad. Sci. 1043:865-875.[CrossRef][Medline]
48
50. Nakabachi, A., A. Yamashita, H. Toh, H. Ishikawa, H. E. Dunbar, N. A. Moran, and M. Hattori. 2006. The 160-kilobase genome of the bacterial endosymbiont Carsonella. Science 314:267.[Abstract/Free Full Text]
49
51. Nelson, D. E., and K. D. Young. 2000. Penicillin binding protein 5 affects cell diameter, contour, and morphology of Escherichia coli. J. Bacteriol. 182:1714-1721.[Abstract/Free Full Text]
50
52. Ng, A. P., J. Howe Fong, D. Sijin Nin, J. L. Hirpara, N. Asou, C. S. Chen, S. Pervaiz, and M. Khan. 2006. Cleavage of misfolded nuclear receptor corepressor confers resistance to unfolded protein response-induced apoptosis. Cancer Res. 66:9903-9912.[Abstract/Free Full Text]
51
53. Nichols, C. E., C. Johnson, M. Lockyer, I. G. Charles, H. K. Lamb, A. R. Hawkins, and D. K. Stammers. 2006. Structural characterization of Salmonella typhimurium YeaZ, an M22 O-sialoglycoprotein endopeptidase homolog. Proteins 64:111-123.[CrossRef][Medline]
52
54. Norby, J. G., and J. Jensen. 1988. Measurement of binding of ATP and ADP to Na⁺,K⁺-ATPase. Methods Enzymol. 156:191-201.[Medline]
53
55. Ogasawara, H., A. Hasegawa, E. Kanda, T. Miki, K. Yamamoto, and A. Ishihama. 2007. Genomic SELEX search for target promoters under the control of the PhoQP-RstBA signal relay cascade. J. Bacteriol. 189:4791-4799.[Abstract/Free Full Text]
54
56. Roberts, A., S.-Y. Lee, E. McCullagh, R. E. Silversmith, and D. E. Wemmer. 2005. YbiV from Escherichia coli K12 is a HAD phosphatase. Proteins 58:790-801.[CrossRef][Medline]
55
57. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
56
58. Simons, R. W., F. Houman, and N. Kleckner. 1987. Improved single and multicopy lac-based cloning vectors for protein and operon fusions. Gene 53:85-96.[CrossRef][Medline]
57
59. Spurio, R., M. Durrenberger, M. Falconi, A. La Teana, C. L. Pon, and C. O. Gualerzi. 1992. Lethal overproduction of the Escherichia coli nucleoid protein H-NS: ultramicroscopic and molecular autopsy. Mol. Gen. Genet. 231:201-211.[CrossRef][Medline]
58
60. Stanley, N. R., K. Findlay, B. C. Berks, and T. Palmer. 2001. Escherichia coli strains blocked in Tat-dependent protein export exhibit pleiotropic defects in the cell envelope. J. Bacteriol. 183:139-144.[Abstract/Free Full Text]
59
61. Stanley, N. R., T. Palmer, and B. C. Berks. 2000. The twin arginine consensus motif of Tat signal peptides is involved in Sec-independent protein targeting in Escherichia coli. J. Biol. Chem. 275:11591-11596.[Abstract/Free Full Text]
60
62. Teplyakov, A., G. Obmolova, M. Tordova, N. Thanki, N. Bonander, E. Eisenstein, A. J. Howard, and G. L. Gilliland. 2002. Crystal structure of the YjeE protein from Haemophilus influenzae: a putative ATPase involved in cell wall synthesis. Proteins 48:220-226.[CrossRef][Medline]
61
63. Thanassi, J. A., S. L. Hartman-Neumann, T. J. Dougherty, B. A. Dougherty, and M. J. Pucci. 2002. Identification of 113 conserved essential genes using a high-throughput gene disruption system in Streptococcus pneumoniae. Nucleic Acids Res. 30:3152-3162.[Abstract/Free Full Text]
62
64. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354.[Abstract/Free Full Text]
63
65. Watt, M. A., A. Mellors, and R. Y. Lo. 1997. Comparison of the recombinant and authentic forms of the Pasteurella haemolytica A1 glycoprotease. FEMS Microbiol. Lett. 147:37-43.[CrossRef][Medline]
64
66. Widdick, D. A., K. Dilks, G. Chandra, A. Bottrill, M. Naldrett, M. Pohlschröder, and T. Palmer. 2006. The twin-arginine translocation pathway is a major route of protein export in Streptomyces coelicolor. Proc. Natl. Acad. Sci. USA 103:17927-17932.[Abstract/Free Full Text]
65
67. Woldringh, C. L., A. Zaritsky, and N. B. Grover. 1994. Nucleoid partitioning and the division plane in Escherichia coli. J. Bacteriol. 176:6030-6038.[Abstract/Free Full Text]
66
68. Yamamoto, K., K. Hirao, T. Oshima, H. Aiba, R. Utsumi, and A. Ishihama. 2005. Functional characterization in vitro of all two-component signal transduction systems from Escherichia coli. J. Biol. Chem. 280:1448-1456.[Abstract/Free Full Text]
67
69. Yim, M. B., H. S. Yim, C. Lee, S. O. Kang, and P. B. Chock. 2001. Protein glycation: creation of catalytic sites for free radical generation. Ann. N. Y. Acad. Sci. 928:48-53.[Medline]
68
70. Yoon, H. S., M. H. Lee, J. Xiong, and J. W. Golden. 2003. Anabaena sp. strain PCC 7120 hetY gene influences heterocyst development. J. Bacteriol. 185:6995-7000.[Abstract/Free Full Text]
69
71. Yu, B. J., J. A. Kim, J. H. Moon, S. E. Ryu, and J. G. Pan. 2008. The diversity of lysine-acetylated proteins in Escherichia coli. J. Microbiol. Biotechnol. 18:1529-1536.[Medline]
70
72. Yu, D., H. M. Ellis, E. C. Lee, N. A. Jenkins, N. G. Copeland, and D. L. Court. 2000. An efficient recombination system for chromosome engineering in Escherichia coli. Proc. Natl. Acad. Sci. USA 97:5978-5983.[Abstract/Free Full Text]
71
73. Zalacain, M., S. Biswas, K. A. Ingraham, J. Ambrad, A. Bryant, A. F. Chalker, S. Iordanescu, J. Fan, F. Fan, R. D. Lunsford, K. O'Dwyer, L. M. Palmer, C. So, D. Sylvester, C. Volker, P. Warren, D. McDevitt, J. R. Brown, D. J. Holmes, and M. K. Burnham. 2003. A global approach to identify novel broad-spectrum antibacterial targets among proteins of unknown function. J. Mol. Microbiol. Biotechnol. 6:109-126.[CrossRef][Medline]
72
74. Zaritsky, A., and C. L. Woldringh. 1978. Chromosome replication rate and cell shape in Escherichia coli: lack of coupling. J. Bacteriol. 135:581-587.[Abstract/Free Full Text]
73
75. Zeghouf, M., J. Li, G. Butland, A. Borkowska, V. Canadien, D. Richards, B. Beattie, A. Emili, and J. F. Greenblatt. 2004. Sequential peptide affinity (SPA) system for the identification of mammalian and bacterial protein complexes. J. Proteome Res. 3:463-468.[CrossRef][Medline]
74
76. Zheng, L., C. Yu, K. Bayles, I. Lasa, and Y. Ji. 2007. Conditional mutation of an essential putative glycoprotease eliminates autolysis in Staphylococcus aureus. J. Bacteriol. 189:2734-2742.[Abstract/Free Full Text]
75
77. Zusman, D. R., A. Carbonell, and J. Y. Haga. 1973. Nucleoid condensation and cell division in Escherichia coli MX74T2 ts52 after inhibition of protein synthesis. J. Bacteriol. 115:1167-1178.[Abstract/Free Full Text]
76
78. Zuther, E., H. Schubert, and M. Hagemann. 1998. Mutation of a gene encoding a putative glycoprotease leads to reduced salt tolerance, altered pigmentation, and cyanophycin accumulation in the cyanobacterium Synechocystis sp. strain PCC 6803. J. Bacteriol. 180:1715-1722.[Abstract/Free Full Text]

---

Journal of Bacteriology, August 2009, p. 4732-4749, Vol. 191, No. 15
0021-9193/09/$08.00+0     doi:10.1128/JB.00136-09
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Related articles in JB:

Grasping at Shadows: Revealing the Elusive Nature of Essential Genes

Tarek Msadek
JB 2009 191: 4701-4704. [Full Text]  

This article has been cited by other articles:

• Choi, E., Groisman, E. A., Shin, D. (2009). Activated by Different Signals, the PhoP/PhoQ Two-Component System Differentially Regulates Metal Uptake. J. Bacteriol. 191: 7174-7181 [Abstract] [Full Text]  
• Msadek, T. (2009). Grasping at Shadows: Revealing the Elusive Nature of Essential Genes. J. Bacteriol. 191: 4701-4704 [Full Text]  

This Article FREE Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Related articles in JB Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Google Scholar Articles by Handford, J. I. Articles by Palmer, T. PubMed PubMed Citation Articles by Handford, J. I. Articles by Palmer, T.