These authors contributed equally to the work.
yciM is an essential gene required for regulation of lipopolysaccharide synthesis in Escherichia coli
Article first published online: 24 NOV 2013
© 2013 John Wiley & Sons Ltd
Volume 91, Issue 1, pages 145–157, January 2014
How to Cite
Mahalakshmi, S., Sunayana, M. R., SaiSree, L. and Reddy, M. (2014), yciM is an essential gene required for regulation of lipopolysaccharide synthesis in Escherichia coli. Molecular Microbiology, 91: 145–157. doi: 10.1111/mmi.12452
- Issue published online: 27 DEC 2013
- Article first published online: 24 NOV 2013
- Accepted manuscript online: 6 NOV 2013 05:21AM EST
- Manuscript Accepted: 1 NOV 2013
- Council of Scientific and Industrial Research (CSIR)
- Department of Biotechnology, Government of India
- Top of page
- Experimental procedures
- Supporting Information
The outer membrane of Gram-negative bacteria is an asymmetric lipid bilayer consisting of an essential glycolipid lipopolysaccharide (LPS) in its outer leaflet and phospholipids in the inner leaflet. Here, we show that yciM, a gene encoding a tetratricopeptide repeat protein of unknown function, modulates LPS levels by negatively regulating the biosynthesis of lipid A, an essential constituent of LPS. Inactivation of yciM resulted in high LPS levels and cell death in Escherichia coli; recessive mutations in lpxA, lpxC or lpxD that lower the synthesis of lipid A, or a gain of function mutation in fabZ that increases the formation of membrane phospholipids, alleviated the yciM mutant phenotypes. A modest increase in YciM led to significant reduction of LPS and increased sensitivity to hydrophobic antibiotics. YciM was shown to regulate LPS by altering LpxC, an enzyme that catalyses the first committed step of lipid A biosynthesis. Regulation of LpxC by YciM was contingent on the presence of FtsH, an essential membrane-anchored protease known to degrade LpxC, suggesting that FtsH and YciM act in concert to regulate synthesis of lipid A. In summary, this study demonstrates an essential role for YciM in regulation of LPS biosynthesis in E. coli.
- Top of page
- Experimental procedures
- Supporting Information
The cell envelope of Gram-negative bacteria is made up of three distinct layers: the outer membrane (OM), the peptidoglycan wall (PG), and the inner membrane (IM). The OM serves as an effective permeability barrier to prevent the entry of hydrophobic as well as large hydrophilic molecules and is responsible for the intrinsic resistance of Escherichia coli to antibiotics, detergents and dyes. Structurally, OM is an asymmetric lipid bilayer with the outer leaflet made up of a unique and essential glycolipid called lipopolysaccharide (LPS) and the inner leaflet composed of phospholipids. LPS is the major immunogenic surface molecule consisting of three covalently linked moieties: lipid A, a proximally located hydrophobic anchor that serves as an endotoxin; a distal O-antigen chain; and a core oligosaccharide that connects these two domains (Nikaido, 1996; Raetz, 1996; Bos et al., 2007).
LPS is essential for growth of most Gram-negative bacteria. Any perturbations in synthesis, assembly or function of LPS affect normal barrier properties of OM leading to altered permeability to various hydrophobic antibiotics, detergents, and dyes (Vaara, 1992; Nikaido, 2003). The components of LPS are synthesized on the cytoplasmic surface of IM as discrete units and are transported into the periplasmic side of IM before being assembled to form a functional LPS molecule (Schnaitman and Klena, 1993; Raetz, 1996; Bos et al., 2007; Raetz et al., 2007). The first step in the synthesis of lipid A is fatty acylation of a sugar-nucleotide, UDP-N-acetylglucosamine by LpxA (Fig. 1). This monoacylated molecule is then irreversibly deacetylated by LpxC for further acylation by LpxD, the third enzyme in the biosynthetic machinery of lipid A. In both the acylation steps catalysed by LpxA and LpxD, the fatty acyl donor is R-3-hydroxymyristoyl-ACP generated by the activity of fabZ encoding an ACP dehydrase. R-3-hydroxymyristoyl-ACP also serves as an acyl donor for the synthesis of membrane phospholipids, and hence is situated at a crucial metabolic branch point in the biosynthesis of membrane components (Raetz and Dowhan, 1990; Mohan et al., 1994).
Subsequent steps catalysed by LpxH, LpxB, and LpxK, respectively, result in the synthesis of an intermediate, lipid IVA. Two molecules of an eight carbon sugar, keto-deoxy-D-manno-octulosonic acid (Kdo) are then added to lipid IVA by WaaA (or KdtA), a glycosyl transferase forming Kdo2-IVA, the minimal essential lipid A required for growth of E. coli. Kdo2-IVA is further acylated sequentially by WaaM (LpxL) and WaaN (LpxM) to finally yield Kdo2-lipid A (Fig. 1). Subsequently, the core oligosaccharide is added to Kdo2-lipid A molecule and the nascent core-lipid A is flipped across IM to the periplasmic surface of IM (Trent, 2004; Raetz et al., 2007). Here, the Kdo2-lipid A is ligated to the O-antigen polymer and transported to the outer leaflet of OM by Lpt system (Bos et al., 2007; Ruiz et al., 2009). However, most laboratory strains of E. coli such as K12 and B are deficient in the synthesis of O-antigen and contain LPS molecules that lack O-antigen (Nikaido, 1996).
Lipid A synthesis is regulated at the deacetylation step catalysed by lpxC gene product, UDP-3-O-(R-3-hydroxymyristoyl)-N-acetylglucosamine deacetylase as this is the first committed step of lipid A biosynthetic pathway (Young et al., 1995; Sorensen et al., 1996). LpxC is modulated at the level of proteolysis by an essential membrane-bound AAA-type metalloprotease, FtsH in E. coli (Ogura et al., 1999; Ito and Akiyama, 2005; Führer et al., 2006). Absence of FtsH stabilizes LpxC leading to enhanced levels of LPS and cell death (Ogura et al., 1999; Führer et al., 2006). In addition to regulating LpxC, FtsH is also involved in the turnover of glycosyl transferase (WaaA) that ligates Kdo residues to lipid IVA molecule ensuring a balanced synthesis of mature LPS (Katz and Ron, 2008). Furthermore, LpxC enzyme levels and activity increase 5- to 10-fold if the early steps in lipid A biosynthesis are blocked either by mutations in lpxA, lpxD or by treatment of cells with sublethal concentrations of an inhibitor of LpxC; these findings point to the existence of an as yet unidentified pathway that involves feedback regulation of LpxC synthesis and activity (Onishi et al., 1996; Sorensen et al., 1996; Raetz et al., 2007).
In this study, we demonstrate the existence of an additional player, YciM, a conserved tetratricopeptide repeat (TPR) protein of unknown function in the regulation of lipid A biosynthesis. Absence of yciM increases LpxC enzyme levels resulting in a higher amount of LPS ultimately leading to cell death. Conversely, a modest increase of yciM significantly reduces the amount of LpxC leading to lowered LPS levels and increased OM permeability to various hydrophobic antibiotics, indicating that YciM negatively regulates lipid A synthesis. In addition, we show that the regulation of LPS by YciM is mediated by FtsH, an essential membrane-bound AAA-protease known to degrade LpxC suggesting that YciM may modulate the proteolytic activity of FtsH towards LpxC. In summary, these results demonstrate a novel regulation of lipid A biosynthesis which is a potential target for development of antibacterial therapeutic agents.
- Top of page
- Experimental procedures
- Supporting Information
yciM is essential for growth and viability of E. coli K12
During the course of our studies, we observed that strain JW1272 from the Keio mutant collection (Baba et al., 2006), carrying a deletion of yciM gene (BW25113 ΔyciM::Kan; Table 1), grew very poorly on nutrient agar (NA) at 42°C, although it was able to grow very well on LB or minimal A plates. However, when we attempted to transfer the yciM::Kan deletion from strain JW1272 by P1-phage mediated transduction into either MG1655 (wild-type K12 strain) or BW25113 (parental strain of the Keio collection), no transductants were obtained on LB or minimal A plates. On the other hand, the yciM::Kan deletion could be introduced into a Kan-sensitive yciM+ derivative of JW1272 (constructed by crossing out the region encompassing the yciM::Kan deletion of JW1272 with linked trpB::Tet marker; Table 1); these observations suggested that the yciM gene is essential for the growth of E. coli, and that the original JW1272 strain carries suppressor mutation (s) elsewhere on the chromosome enabling its growth. We further confirmed the requirement of yciM for the growth of E. coli by successful introduction of yciM::Kan deletion into MG1655 strain carrying a plasmid-borne copy of yciM (pMN101) but not into MG1655 carrying the plasmid vector alone.
|Strain||Relevant characteristics||Source or reference|
|BW25113||lacIq rrnB3 ΔlacZ4787 Δ(araBAD)567 Δ(rhaBAD)568 hsdR514||Baba et al. (2006)|
|CAG18455||MG1655 trpB83::Tn10||Lab collection|
|D22||lpxC101 proA23 lac-28 tsx-81 trp-30 his-51 rpsL173 tufA1 ampCp-1||CGSCa; Normark et al. (1969)|
|JW1272||BW25113 ΔyciM::Kan||Baba et al. (2006)|
|MC4100||F− araD139 Δ(argF-lac)U169 rpsL150 relA1 flb5301 deoC1 thiA1 ptsF25 rbsR||Lab collection|
|MG1655||rph1 ilvG rfb-50 (wild-type)||Lab collection|
|RL436||Hfr (PO3) lpxD36(ts) relA1 spoT1 metB1||CGSCa; Lathe et al. (1980)|
|SM101||F− thr-1 araC14 lpxA2(ts) tsx-78 Δ(galK-attLAM)99 hisG4(Oc) rfbC1 rpsL136 xylA5 mtl-1 thiE1||CGSCa; Galloway and Raetz (1990)|
|MR701||JW1272 yciM+ KanS trpB83::Tn10||This study|
|MR703||MG1655 ΔyciM::Kan/pMN103||This study|
|MR704||MG1655 ΔyciS::Kan/pMN103||This study|
|MR705||MG1655 ΔyciSM::Cm/pMN103||This study|
|MR706||MG1655 lpxC1272 leuB::Tn10||This study|
|MR707||MR706 ΔyciSM::Cm||This study|
|MR708||MG1655 lpxD14 (skp::Tn10dTet)||This study|
|MR709||MR708 ΔyciSM::Cm||This study|
|MR710||MG1655 lpxD36 zae-502::Tn10||This study|
|MR711||MR710 ΔyciSM::Cm||This study|
|MR713||MR712 ΔyciM::Kan||This study|
|MR718||MC4100 ΔyciS::lacZ-Kan/pMN102||This study|
|MR719||MC4100 ΔyciM::lacZ-Kan/pMN102||This study|
|MR720||MR705 sfhC21 zad-22::Tn10||This study|
|MR721||MG1655 ftsH1 zgj-3198::Tn10Kan||This study|
|MR722||MR705 ftsH1 zgj-3198::Tn10Kan||This study|
|pAM34||pMB1-based, IPTG-dependent replicon, AmpR, SpcR||Gil and Bouche (1991)|
|pCL1920||pSC101-based, SpcR||Lerner and Inouye (1990)|
|pDSW210||ColE1, AmpR, P206 (trc) promoter||Weiss et al. (1999)|
|pMA2||pSC101 (Ts), AmpR||Reddy (2007); Japanese cloning vector collection|
|pKGE137||KanR, FRT, lacZY+, Ori-R6K||Ellermeier et al. (2002)|
|ASKA-fabZ||CmR, pCA24N-fabZ||Kitagawa et al. (2006)|
|pMN106||pMU2385-plpxC (plpxC::lacZ)||This study|
To further examine these observations, we constructed strain MR703 (MG1655 ΔyciM::Kan carrying a conditional IPTG-dependent replication vector, pAM34 with a cloned copy of yciM, pMN103). The plasmid pAM34 is unique in that the replication of the plasmid itself is dependent on the presence of IPTG (Gil and Bouche, 1991); hence strain MR703 is yciM+ on IPTG-supplementation and yciM− in absence of IPTG. As shown in Fig. 2A, this strain was extremely sick on LB and did not grow at all on minimal A plates at restrictive conditions (without IPTG) whereas it grew very well at the permissive conditions (with IPTG). Further, the terminal phenotype in cells of strain MR703 grown under restrictive conditions was significant cell lysis as evidenced by decrease in optical density and colony forming units (Fig. 2B). The lysing cells exhibited morphological aberrations and cell filamentation as visualized by DIC and live-dead cell staining (Fig. 2C).
The structural organization of the yciM locus indicated that it is the second gene in a bicistronic operon, yciSM, located at 28.85 min on the E. coli chromosome. To examine the role of the upstream yciS gene, we constructed MG1655 ΔyciS::Kan deletion strain (MR704) and a minor effect on cell viability was noticed from the size of the colonies growing on plates at restrictive conditions (Fig. 2A). However, this effect disappeared when the kanamycin determinant from yciS::Kan deletion was flipped out indicating that this defect is likely due to polarity of Kan insertion on the expression of downstream yciM gene. A strain with a complete deletion of yciSM locus carrying a cloned copy of only yciM gene (MG1655/ΔyciSM::Cm/pMN103, designated MR705) also behaved identical to that of the single yciM deletion strain, MR703, showing that absence of yciM alone is responsible for the loss of viability in these mutants (Fig. 2A and B). In summary, the results above demonstrate that yciM is fundamental for growth of E. coli under standard laboratory conditions and that strain JW1272 is viable owing to the presence of compensatory or suppressor mutation (s) unlinked to yciM locus.
Mutations that lower lipid A synthesis suppress the essentiality of yciM
To understand the basis of yciM essentiality, we identified the mutation responsible for the viability of JW1272 with the aid of a transposon-tagging approach as described in Experimental procedures. The suppressor was found to be a recessive mutation in lpxC that is due to T-to-A conversion in codon 186 resulting in a change in the amino acid sequence from isoleucine to asparagine (lpxC-I186N, hereafter referred to as lpxC1272). Fig. 3A shows the extent of suppression conferred by lpxC1272 allele in strain MR705. A known allele of lpxC (lpxC101 or envA1; Normark et al., 1969; Beall and Lutkenhaus, 1987) also suppressed the growth defects of both yciM and yciSM deletion mutants to a moderate extent (data not shown).
The lpxC1272 allele alone (reconstructed in the wild-type strain background MG1655, designated MR706; Table 1) conferred sensitivity to growth on NA at 42°C, indicating that the initial phenotype observed in strain JW1272 is essentially due to lpxC mutation. It is interesting to note that in the chemical genomics screen of Nichols et al. (2011), JW1272 exhibited a strong correlation with asmB mutations that are allelic to lpxC (Kloser et al., 1996), suggesting that most likely the lpxC suppressor mutation would have arisen during the construction of yciM deletion mutant. In an earlier study, strain JW1272 was identified as a mutant deficient in biofilm formation and was also shown to have synthetic lethal phenotypes with surA and yfgA (rodZ) which code for a periplasmic peptidyl-prolyl isomerase and a cytoskeletal membrane protein respectively (Niba et al., 2007); however, the significance of these observations remains to be examined.
We also obtained additional suppressors in strain MR703 after random transposon insertional mutagenesis as described in Experimental procedures. Genetic and molecular characterization of an insertion mutation that conferred very significant suppression to yciM deletion mutant showed an insertion after codon 79 in a gene called skp located at 4 min. However, subsequent studies showed that the resultant suppression phenotype of yciM deletion is a consequence of a polar effect of the insertion on the expression of downstream lpxD gene, as a plasmid encoding lpxD alone was able to abrogate the suppression conferred by this mutation whereas a plasmid carrying skp was unable to do so (for convenience sake, this allele is hereafter referred to as lpxD14). In addition, a Δskp::Kan insertion-deletion from Keio collection (Baba et al., 2006) also suppressed the yciM growth defects, presumably due to the polar effect of the Kan insertion on lpxD. The growth defect of yciM mutant was also abolished by introduction of lpxD36, a known temperature-sensitive allele of lpxD (Lathe et al., 1980) as shown in Fig. 3A. A temperature-sensitive mutation in lpxA (lpxA2; Galloway and Raetz, 1990) also weakly suppressed the absence of yciM at 30°C, although at 37°C, the lpxA2 mutant itself was not viable. On the other hand, deletion mutations in waaM, waaN or lpxT involved in the maturation of lipid A did not rescue the growth phenotypes of yciM deletion mutant (data not shown).
Absence of yciM results in increased LPS
As mutations in lpxA, lpxC or lpxD that are known to lower lipid A biosynthesis (Raetz, 1996) suppressed the yciM growth defects, we speculated that yciM may have a role in lipid A (or LPS) metabolism or regulation. Hence, we measured the levels of LPS in strains MR703 (ΔyciM::Kan/pMN103) and MR705 (ΔyciSM::Cm/pMN103) grown at restrictive conditions (without IPTG), by estimating the amount of Kdo and also by quantifying LPS using SDS-PAGE. Data from Fig. 4B and Table 2 indicate that the amount of LPS is around twofold higher in yciM or yciSM mutant cells compared with that of the parental strain implicating a role for YciM in regulation of LPS.
|Relevant genotype||μg of Kdo per mg of proteina|
|Wild-type (MG1655)||9.0 ± 0.5|
|ΔyciSMb||18.0 ± 2.0|
|ΔyciMb||20.0 ± 1.8|
|lpxC1272||7.3 ± 0.5|
|lpxC1272 ΔyciSM||7.5 ± 0.6|
|lpxD14||7.9 ± 0.4|
|lpxD14 ΔyciSM||10.5 ± 0.3|
|cMG1655/ptrc::yciM||8.3 ± 0.2|
|cMG1655/ptrc::yciM (100 μM IPTG)||6.4 ± 0.2|
|dΔyciM/pfabZ||18.0 ± 0.9|
|dΔyciM/pfabZ (20 μM IPTG)||6.2 ± 0.5|
To examine the basis of suppression of YciM− phenotype by lpxC or lpxD mutations, LPS levels in these single and double mutants were also measured. Mutations in lpxC (lpxC1272) or lpxD (lpxD14) lowered the LPS levels both in the wild-type as well as in the yciSM deletion strain, supporting the notion that the suppression of yciSM growth defects is most likely due to decreased LPS (or lipid A) synthesis (Fig. 4C and Table 2). However, although both lpxC1272 and lpxD14 single mutants showed approximately 50% decrease in LPS (Fig. 4C, lanes 2 and 4), the LPS level of lpxC1272 yciSM double mutant was comparable to that of single lpxC1272 mutant (Fig. 4C, lanes 4 and 5) whereas in lpxD14 yciSM double mutant the LPS was elevated approximately twofold (similar to that of the level seen in the wild-type) compared with the single lpxD14 mutant (Fig. 4C, lanes 2 and 3). In addition, lpxC101 and lpxD36 alleles also behaved similarly in that the LPS levels of lpxC101 yciSM strain remained like that of lpxC101 single mutant whereas in lpxD36 yciSM strain, the LPS levels were higher compared with lpxD36 alone (Fig. S1).
Mutations that lower lipid A synthesis alter the barrier properties of OM leading to hyperpermeability (and hence, increased sensitivity) to hydrophobic antibiotics (Vaara, 1992; Nikaido, 2003). As expected, strains carrying either lpxC1272 (MR706), lpxD14 (MR708) or lpxD36 (MR710) alleles were hypersensitive to antibiotics such as vancomycin and nalidixic acid (Fig. 4A). Interestingly, absence of yciSM reversed the antibiotic hypersensitivity phenotypes of both lpxD14 and lpxD36 mutants but not of lpxC1272 (Fig. 4A). This observation correlates well with the results above wherein the LPS levels of lpxD14 yciSM strain but not of lpxC1272 yciSM are restored back to the wild-type level (Fig. 4C; Table 2) and these results are discussed below.
Increased FabZ also suppresses the absence of yciM
It is known from earlier studies that balanced synthesis of OM components is crucial for growth. The toxic effects of high LPS are alleviated by increasing the formation of membrane phospholipids by multiple copies of FabZ that increases the pool of R-3-hydroxymyristoyl-ACP, a common precursor of both lipid A and phospholipid synthesis (Ogura et al., 1999). Conversely, recessive mutations in fabZ suppress the effects of lpxA or lpxC by balancing the ratio of LPS to phospholipid (Mohan et al., 1994; Kloser et al., 1998; Zeng et al., 2013). Therefore, we examined the effect of multiple copies of fabZ in yciM mutant background and observed that increased expression of plasmid-borne Plac-fabZ (achieved by addition of 20 μM IPTG to the growth medium) suppressed the yciM growth defects extremely well (Fig. 3B) and also lowered the level of LPS (Table 2). In addition, a gain of function allele of fabZ (sfhC21, which is known to suppress ftsH mutations by increasing the formation of phospholipids; Ogura et al., 1999) restored growth equally well to the yciSM deletion mutant (Fig. 3A).
yciM overexpression leads to a decrease in LPS levels which is reversed upon LpxC overexpression
As the experiments above suggested involvement of yciM in modulation of LPS levels, we examined the effect of additional copies of yciM on cell growth and permeability. For this purpose, yciM was cloned in pDSW210 (Weiss et al., 1999), a medium-copy-number, ColE1-based vector under the control of an attenuated trc promoter (Ptrc::yciM; designated pMN104) and introduced into MG1655. Increasing the expression of yciM (by addition of 50 μM IPTG) resulted in severe hypersensitivity to various antibiotics such as vancomycin, rifampin and nalidixic acid indicating a defect in OM barrier functions (Fig. 5A). Likewise, there was a significant reduction in LPS levels when yciM is present in multiple copies (Fig. 5B and Table 2). All these phenotypes could be reversed by a moderate increase in the expression of lpxC which is known to increase the flux into lipid A biosynthesis (Fig. 5A).
YciM regulates LpxC, the rate-limiting enzyme of lipid A biosynthetic pathway
Since LpxC is the rate-limiting enzyme in lipid A biosynthetic pathway and that absence of yciM was not able to elevate the LPS level in lpxC mutants in our studies (Fig. 4A and C), we reasoned that YciM may regulate lipid A synthesis at the step of LpxC. Accordingly, we examined the level of LpxC protein in various mutants by western analysis using anti-LpxC antisera. Fig. 6A shows approximately fivefold upregulation of LpxC enzyme levels in absence of yciM or yciSM, clearly suggesting that LPS overproduction in these mutants is due to increased amount of LpxC. Likewise, there was approximately twofold decrease in LpxC levels in presence of more copies of yciM (Fig. 6B, lanes 1 and 2).
To examine whether these alterations are due to transcriptional regulation of lpxC, a β-galactosidase reporter gene was fused to the promoter of lpxC on a low-copy-number plasmid (as described in Experimental procedures) and introduced into appropriate mutant strains. But, there was no significant difference in the expression of lpxC-lacZ fusion during the absence or overexpression of yciM (Table S1). An lpxC-GFP promoter fusion on a low-copy-number plasmid (Zaslaver et al., 2006) also behaved similarly in that there was no alteration in the expression of GFP both in absence of yciM and in presence of additional copies of yciM (data not shown) suggesting a post-transcriptional regulation of LpxC by YciM.
YciM-mediated regulation of LpxC is dependent on functional FtsH protease
As mentioned above, LpxC is regulated by FtsH, a membrane-anchored essential metalloprotease at the level of proteolysis, and as a consequence, ftsH mutants accumulate LpxC leading to lethal overproduction of LPS and cell death (Ogura et al., 1999; Führer et al., 2006). As absence of either FtsH or YciM results in similar growth defects and that both the mutants are suppressed by factors that lower lipid A synthesis or increase the membrane phospholipid formation (Fig. 3; Ogura et al., 1999), we speculated that FtsH and YciM may participate in a common pathway to regulate the levels of LpxC. To examine whether YciM regulates LpxC via FtsH, LpxC levels were measured in an ftsH mutant strain (carrying a temperature-sensitive allele, ftsH1) with additional copies of yciM (MR721/pMN104). Figure 6B shows higher levels of LpxC in ftsH1 mutant as expected; however, overexpression of yciM did not decrease the LpxC levels in ftsH1 strain background (Fig. 6B, lanes 3 and 4) unlike that of in the wild-type strain (Fig. 6B, lanes 1 and 2). Moreover, LpxC levels remained unaltered in an ftsH1 mutant strain carrying a deletion of yciSM compared with either of the single mutants (Fig. 6C). Both these experiments showed that regulation of LpxC by YciM is mediated through functional FtsH protease.
To test whether YciM is a general modulator of the proteolytic activity of FtsH, we examined the levels of RpoH (σ32), another known target of FtsH (Herman et al., 1995; Tomoyasu et al., 1995) in various strain backgrounds (wild-type, ΔyciM, ΔyciSM and ftsH1) by western analysis using anti-RpoH antisera (Fig. S2). RpoH was not stabilized in absence of YciM indicating that YciM may not have an effect on all the targets of FtsH protease. However, as expected, RpoH levels were high in the ftsH1 mutant background (Fig. S2).
LpxC increase in mutants defective in early steps of lipid A biosynthesis is dependent on YciM
We also observed that single mutations in lpxA, lpxC or lpxD that lower the lipid A synthesis increased the LpxC enzyme levels approximately two- to sixfold (Fig. 6A; Fig. S3) confirming the previous findings of Sorensen et al. (1996). However, the LpxC level in each of these mutants carrying a deletion of yciSM (i.e. lpxC1272 yciSM; lpxC101 yciSM; lpxD14 yciSM and lpxD36 yciSM) was not additive, but similar to that of the single yciSM deletion mutant raising an interesting possibility that the feedback regulation of LpxC in mutants defective in early steps of lipid A biosynthesis is dependent on the presence of YciM (Fig. 6A; Fig. S3).
In addition, the elevated LpxC levels observed in the lpxD yciSM mutants (Fig. 6A, lanes 5 and 6; Fig. S3) allowed us to explain our earlier findings in which the lpxD yciSM double mutants showed increased amounts of LPS leading to restoration of antibiotic hypersensitivity phenotypes of lpxD single mutants (Fig. 4A and C). In other words, higher amounts of functional LpxC in absence of yciM results in elevated LPS in lpxD14 and lpxD36 mutants. However, in lpxC mutants, the accumulation of mutant (non-functional) LpxC enzyme may not contribute to an increase in LPS synthesis and thus do not alter the antibiotic-hypersensitivity phenotypes of lpxC mutants (Fig. 4A and C).
yciS and yciM show transcriptional upregulation by increased RpoH
To examine the physiological regulation of yciSM operon, transcriptional lacZ fusions of yciS and yciM were constructed on the chromosome as described in Experimental procedures. As these reporter gene fusion strains are deleted for yciS or yciM structural genes, they are constructed in strains carrying a plasmid-borne copy of yciM, pMN102 (Table 1). The β-galactosidase values of these lacZ fusions indicated that yciSM operon has a moderate-to-strong promoter (Table S2). However, the lacZ expression levels in these strains were not altered either by decrease or increase of yciM or by presence of lpxC or lpxD mutant alleles, indicating that yciSM is not feedback regulated by lipid A at the transcriptional level. In a global gene expression study, it was earlier reported that the expression of yciS and yciM are upregulated by increased copies of RpoH (Zhao et al., 2005; Nonaka et al., 2006). We experimentally validated these observations by showing that the lacZ expression of both yciS and yciM fusion strains is enhanced approximately threefold by increased expression of RpoH but not of RpoE (Table S2). However, the significance of this observation remains to be understood as expression of yciS/M was unaltered by variations in temperature, osmolarity or pH (data not shown), although in expression profiling experiments, yciM mRNA was reported to be induced at high temperature (47°C) and also at acidic pH conditions in presence of cadmium (Worden et al., 2009; Murata et al., 2011).
Mutations that affect outer membrane assembly or organization also suppress the yciM essentiality
We obtained and characterized several transposon insertion mutations that were weak suppressors of the growth defect in the yciM deletion mutant MR703. The insertions were identified to be in genes encoding, (1) a lipopolysaccharide kinase (rfaP), (2) UDP-glucose pyrophosphorylase (galU), (3) a major OM lipoprotein (lpp) and (4) a protein of unknown function (ybcN). The corresponding deletion mutations from the Keio collection also showed similar weak suppression phenotypes (Fig. S4). Furthermore, deletion mutations in genes encoding other outer membrane proteins such as ompA, tolA, pal also weakly suppressed the yciM growth defects. Though the basis of this suppression is not clear, it is possible that these mutations decrease the lethality associated with high LPS levels by affecting the formation and/or assembly of functional LPS molecules or by altering the effective concentration of LPS in the OM. Alternatively, some of these mutations may alter the ratio of LPS to phospholipid and thus partially contribute to viability. As described earlier for the lpxD alleles (Fig. 4A), the antibiotic sensitivity phenotypes of some of the above mutants were suppressed by absence of yciM indicating that increased LPS in these double mutants is able to reverse the OM permeability defects (data not shown).
- Top of page
- Experimental procedures
- Supporting Information
In this study, we demonstrate that YciM, encoded by a gene of a hitherto unknown function, modulates cellular LPS levels by regulating LpxC, the rate-limiting enzyme of lipid A biosynthesis. Previous studies have shown that expression of LpxC is regulated by FtsH-mediated proteolysis as also by conditions that inhibit early steps of lipid A synthesis (Sorensen et al., 1996; Ogura et al., 1999). Our results show that YciM and FtsH participate in a common pathway to regulate the LpxC levels implicating a role for YciM in modulation of proteolytic activity of FtsH towards LpxC.
Deficiency of YciM, FtsH or overexpression of LpxC result in lethal overproduction of LPS
Inactivation of yciM leads to high LPS levels and cell death (Figs 2 and 4B). As mutations that lower the synthesis of lipid A suppress the lethality of yciM by decreasing the levels of LPS (Fig. 4C; Table 2), the most likely cause of cell death in yciM deletion mutants is high LPS. It is known that in the absence of FtsH protease, LpxC enzyme is stabilized leading to high LPS levels eventually causing cell death (Ogura et al., 1999). Likewise, overexpression of LpxC is toxic to cells as it leads to overproduction of LPS (Sullivan and Donachie, 1984; Beall and Lutkenhaus, 1987; Young et al., 1995; Führer et al., 2006). However, it is not clear why excess LPS leads to lethality. During FtsH inactivation, abnormal membranous structures form in periplasm, which are thought to be deleterious to the cell (Ogura et al., 1999). Alternatively, excess LPS production may lead to depletion of fatty acyl donor molecules needed for the synthesis of membrane phospholipids.
Physiological role of yciM
Regulation of lipid A synthesis is crucial as very high or very low amounts of lipid A (or LPS) is detrimental to the growth of a cell. As mentioned above, lipid A is regulated at the step of deacetylation as it is the first committed step in its biosynthesis. Our results show that the deacetylase enzyme, LpxC is regulated by YciM and this regulation is dependent on functional FtsH protease (Fig. 6A and B). However, another well-studied target of FtsH protease, RpoH, is not stabilized in absence of YciM indicating that YciM is a specific modulating factor for FtsH-dependent proteolysis of LpxC.
FtsH protease is well conserved in all eubacteria, mitochondria and chloroplasts (Ito and Akiyama, 2005; Langklotz et al., 2012). On the other hand, yciM gene is completely absent in Gram-positive organisms and is conserved only in a few members of β- and γ-proteobacteria of Gram-negative class. However, it is found to be very well conserved in the enterobacteriaceae family. In this context, it is noteworthy to mention that in Pseudomonas aeruginosa (that belongs to γ-proteobacteria) and also in organisms belonging to α-proteobacteria such as Agrobacterium tumefaciens and Caulobacter crescentus in which degradation of LpxC is independent of FtsH protease (Langklotz et al., 2011), yciM gene is also absent, lending support to the presumed function of YciM.
In addition to LpxC, FtsH protease degrades a variety of cellular targets including phage λ cII, SecY, and RpoH (σ32). It is known that degradation of these targets by FtsH is facilitated by specific adapter proteins; for, e.g. degradation of λ cII is modulated by HflKC complex whereas degradation of RpoH requires the presence of DnaK/J chaperone system (Ito and Akiyama, 2005; Langklotz et al., 2012). These facts also point to the possibility of YciM being a specific adapter protein for modulation of LpxC proteolysis.
Our results also indicate that the feedback regulation of LpxC enzyme observed in mutants that are defective in early steps of lipid A biosynthesis is possibly mediated by YciM (Fig. 6A). Although, it is not clear how YciM facilitates this regulation, one attractive possibility is that YciM communicates the feedback signal to FtsH protease to specifically alter LpxC to maintain optimal levels of lipid A. In support of this, recently, Schakermann et al. (2013) have shown that the FtsH-mediated stabilization of LpxC is linked to cellular growth rate leading them to speculate the existence of additional factor (s) that modulate the LpxC proteolysis.
Structurally, YciM contains six tetratricopeptide (34-amino acid) repeats and a C-terminal Zinc-finger domain. In general, TPR domain proteins mediate protein–protein interactions and facilitate the assembly of multiprotein complexes and are involved in a variety of biological processes including cell cycle regulation, transcriptional control, mitochondrial and peroxisomal protein transport, and bacterial virulence functions (Blatch and Lassle, 1999; D'Andrea and Regan, 2003). Interestingly, an instance of a TPR domain of a eukaryotic protein phosphatase being activated by polyunsaturated fatty acids and anionic phospholipids is earlier known (Sinclair et al., 1999).
To sum up, this study demonstrates an essential role for YciM in lipid A regulation in E. coli. Our results suggest that YciM could be an adapter protein of FtsH protease to facilitate regulated proteolysis of LpxC to maintain optimal cellular lipid A levels. It would be worthwhile to test this model and also examine the mechanistic aspects of LpxC regulation by YciM.
- Top of page
- Experimental procedures
- Supporting Information
Growth media and conditions
The defined and nutrient media were, respectively, minimal A medium (supplemented with 0.2% d-glucose as the C-source) or Luria–Bertani medium (Miller, 1992). Nutrient agar contained 0.5% bacto-peptone and 0.3% beef extract (Miller, 1992). Unless otherwise indicated, growth temperature was 37°C. The following antibiotics were used at the indicated concentrations (μg ml−1): ampicillin (Amp), 50; kanamycin (Kan), 50; chloramphenicol (Cm), 20; spectinomycin (Spc), 50; and tetracycline (Tet), 15.
Bacterial strains, phages and plasmids
The bacterial strains and plasmids used in this study are listed in Table 1. Unless otherwise indicated, all strains are derivatives of E. coli K12 MG1655. The authenticity of the deletion alleles of the Keio mutant collection (Baba et al., 2006) used in this study was confirmed by linkage analysis, PCR, sequencing and phenotype if known. Complete deletion of yciSM locus was made on the chromosome by recombineering as described earlier (Yu et al., 2000). Marker-less strains were made by flipping out the antibiotic resistance marker (KanR) using pCP20 plasmid that encodes a Flp recombinase as described earlier (Datsenko and Wanner, 2000). Phage P1kc was from our laboratory stock.
Molecular and genetic techniques
Standard protocols were followed for experiments involving recombinant DNA and plasmid manipulations (Sambrook and Russel, 2001). Transpositions, P1-phage mediated transductions and β-galactosidase assays were performed using standard methods as described (Miller, 1992). Transposon tagging and identification of the suppressor mutation in JW1272 were done using conventional conjugational and transductional mapping techniques (Miller, 1992) and finally by sequencing and complementation studies.
Identification of transposon insertion mutations that suppressed the yciM mutant phenotypes
Strain MR703 (MG1655 ΔyciM::Kan/pMN103) was subjected to random insertion mutagenesis with transposon Tn10dTet following infection with phage λ1098, as described previously (Miller, 1992). The TetR transposon-mutagenized library was screened for colonies that grew well on plates without IPTG (restrictive condition in which the plasmid pMN103 is unable to replicate leading to the manifestation of YciM− phenotype). The precise location of the insertions that were responsible for suppression of YciM− phenotype was identified by cloning the TetR insertion along with flanking sequences onto the plasmid pCL1920, followed by sequencing the gene junctions using the outwardly directed Tet primers.
Growth and OM permeability assays
The viability of each strain was measured by applying 5–10 μl aliquots of various dilutions (10−2, 10−4, 10−5, 10−6 and 10−7) of overnight cultures onto indicated plates and incubating them at 37°C for 20 to 36 h. Outer membrane barrier function was examined using hypersensitivity to various antibiotics such as vancomycin, nalidixic acid or rifampin at the indicated concentrations. Strains of interest were grown, serially diluted and appropriate dilutions were placed on indicated plates and incubated generally for 16–24 h.
Microscopy and viability measurements
To monitor growth, indicated cells were grown overnight in permissive conditions and next day they were washed and diluted 1:500 into fresh medium and grown at both permissive and restrictive conditions. At indicated time points, absorbance at 600 nm and colony forming units were measured and additional 0.5 ml culture was drawn for microscopy. Cell viability was also measured after staining cells with LIVE/DEAD Baclight bacterial viability kit (Molecular Probes, Invitrogen). Cells were immobilized on a thin agarose pad and visualized on a Zeiss apotome microscope by DIC (Nomarski optics) and fluorescence microscopy using GFP and DsRed filters.
Determination of Kdo (keto-deoxy-d-manno-8-octanoic acid)
Kdo was measured essentially as described earlier (Karkhanis et al., 1978). Strains of interest were grown in LB broth, washed, resuspended in 10 mM HEPES buffer (pH 7.4) and cell lysates were prepared by sonication. Kdo was measured following acid hydrolysis of these lysates as described earlier (Ogura et al., 1999) and the values are expressed per mg of protein. Protein concentrations were determined using Quickstart-Bradford reagent from Bio-Rad with bovine serum albumin (BSA) as the standard.
Quantification of LPS
LPS levels were measured using the cell lysates prepared as described above. The cell lysates (normalized to the protein concentrations) were mixed with equal volume of tricine sample buffer (100 mM Tris-HCl, pH 6.8, 24% w/v glycerol, 8% w/v SDS, 5% v/v 2-β-mercaptoethanol, 0.02% w/v bromophenol blue), and boiled for 10 min. To 50 μl of boiled sample, 10 μl of proteinase K solution (2.5 mg ml−1 in sample buffer) was added and incubated further at 60°C for 60 min followed by centrifugation at 16 000 g for 30 min. The supernatants were loaded on 18% tricine-SDS polyacrylamide gels and LPS was visualized by silver staining (Austin et al., 1990). LPS was quantified using Image-J densitometric software.
Strains of interest were grown in LB broth, washed, resuspended in 10 mM HEPES buffer (pH 7.4) and cell lysates were prepared by sonication. Total protein was estimated and equal amount of protein was electrophoresed by SDS-PAGE. Western analysis was done as described earlier (Sambrook and Russel, 2001). LpxC antiserum (a kind gift from F Narberhaus) and RpoH antibodies (ab26890; purchased from AbCam) were used at 1:20000 and 1:2000 dilution respectively. Appropriate secondary anti-rabbit-HRP was used at a dilution of 1:10000. Blots were developed using ECL chemiluminiscent detection reagents (Roche) and quantified using Image-J software.
The details of plasmid constructions, oligonucleotides used and additional experimental procedures are described in the Supplemental information. It also includes additional references, four figures and two tables.
- Top of page
- Experimental procedures
- Supporting Information
We thank NBRP: E. coli for the Keio mutant collection and ASKA plasmids; Coli Genetic Stock Centre for lpx mutants; Carol Gross for ftsH1 and sfhC21 mutants; Thomas Silhavy and Dante Ricci for support and valuable discussions; and J Gowrishankar for advice on the manuscript. We would like to thank Jan Tommassen for helpful suggestions and Franz Narberhaus for generous sharing of LpxC antisera. This work was supported in part by funds from Council of Scientific and Industrial Research (CSIR), and Department of Biotechnology, Government of India.
Conflict of interest
The authors declare that they have no conflict of interest.
- Top of page
- Experimental procedures
- Supporting Information
- 1990) Genetic analysis of lipopolysaccharide core biosynthesis by Escherichia coli K-12: insertion mutagenesis of the rfa locus. J Bacteriol 172: 5312–5325. , , , , and (
- 2006) Construction of Escherichia coli K-12 in-frame, single-gene knock-out mutants: the Keio collection. Mol Syst Biol 2: 2006.0008. , , , , , , et al. (
- 1987) Sequence analysis, transcriptional organization and insertional mutagenesis of the envA gene of Escherichia coli. J Bacteriol 169: 5408–5415. , and (
- 1999) The tetratricopeptide repeat: a structural motif mediating protein-protein interactions. Bioessays 21: 932–939. , and (
- 2007) Biogenesis of the Gram-negative bacterial outer membrane. Annu Rev Microbiol 61: 191–214. , , and (
- 2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci USA 97: 6640–6645. , and (
- 2003) TPR proteins: the versatile helix. Trends Biochem Sci 28: 655–662. , and (
- 2006) The C-terminal of LpxC is required for degradation by the FtsH protease. Mol Microbiol 59: 1025–1036. , , and (
- 1990) A mutant of Escherichia coli defective in the first step of endotoxin biosynthesis. J Biol Chem 265: 6394–6402. , and (
- 1991) ColE1-type vectors with fully repressible replication. Gene 105: 17–22. , and -P. (
- 1995) Degradation of σ32, the heat shock regulator in Escherichia coli is governed by HflB. Proc Natl Acad Sci USA 92: 3516–3520. , , , and (
- 2005) Cellular functions, mechanism of action, and regulation of FtsH protease. Annu Rev Microbiol 59: 211–231. , and (
- 1978) A new and improved microassay to determine 2-keto-3-deoxyoctonate in lipopolysaccharide of Gram-negative bacteria. Anal Biochem 85: 595–601. , , , and (
- 2008) Dual role of FtsH in regulating lipopolysaccharide biosynthesis in Escherichia coli. J Bacteriol 190: 7117–7122. , and (
- 2006) Complete set of ORF clones of Escherichia coli ASKA library (a complete set of E. coli K-12 ORF archive): unique resources for biological research. DNA Res 12: 291–299. , , , , , , et al. (
- 1998) Modulations in lipid A and phospholipid biosynthesis pathways influence outer membrane protein assembly in Escherichia coli K-12. Mol Microbiol 27: 1003–1008. , , , and (
- 1996) asmB, a suppressor locus for assembly-defective outer-membrane proteins of E. coli, is allelic to envA (lpxC). J Bacteriol 178: 5138–5143. , , and (
- 2011) Control of lipopolysaccharide biosynthesis by FtsH-mediated proteolysis of LpxC is conserved in enterobacteria but not in all Gram-negative bacteria. J Bacteriol 193: 1090–1097. , , and (
- 2012) Structure and function of the bacterial AAA protease FtsH. Biochim Biophys Acta 1823: 40–48. , , and (
- 1980) Prokaryotic histone-like protein interacting with RNA polymerase. Proc Natl Acad Sci USA 77: 3548–3552. , , , and (
- 1990) Low copy number plasmids for regulated low-level expression of cloned genes in Escherichia coli with blue/white insert screening capability. Nucleic Acids Res 18: 4631. , and (
- 1992) A Short Course in Bacterial Genetics: A Laboratory Manual and Handbook for Escherichia coli and Related Bacteria. Cold Spring Harbor: Cold Spring Harbor Laboratory. (
- 1994) An Escherichia coli gene (FabZ) encoding (3R)-hydroxymyristoyl acyl carrier protein dehydrase. J Biol Chem 269: 32896–32903. , , , , and (
- 2011) Molecular strategy for survival at a critical high temperature in Escherichia coli. PLoS ONE 6: e20063. , , , , , , et al. (
- 2007) A genome-wide approach to identify the genes involved in biofilm formation in Escherichia coli. DNA Res 14: 237–246. , , , , and (
- 2011) Phenotypic landscape of a cell. Cell 144: 143–156. , , , , , , et al. (
- 1996) Outer membrane. In Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd edn. Neidhardt, F.C. , Curtiss, R., III , Ingraham, J.L. , Lin, E.C.C. , Low, K.B. , Magasanik, B. , et al. (eds). Washington, DC: ASM press, pp. 29–47. (
- 2003) Molecular basis of bacterial outer membrane permeability revisited. Microbiol Mol Biol Rev 67: 593–656. (
- 2006) Regulon and promoter analysis of the Escherichia coli heat-shock factor, sigma32, reveals a multifaceted cellular response to heat stress. Genes Dev 20: 1776–1789. , , , , and (
- 1969) Mutant of Escherichia coli with anomalous cell division and ability to decrease episomally and chromosomally mediated resistance to ampicillin and several other antibiotics. J Bacteriol 97: 1334–1342. , , and (
- 1999) Balanced biosynthesis of major membrane components through regulated degradation of the committed enzyme of lipid A biosynthesis by the AAA protease FtsH (HflB) in Escherichia coli. Mol Microbiol 31: 833–844. , , , , , , et al. (
- 1996) Antibacterial agents that inhibit lipid A biosynthesis. Science 274: 980–982. , , , , , , et al. (
- 1996) Bacterial lipopolysaccharides: a remarkable family of bioactive macroamphiphiles. In Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd edn. Neidhardt, F.C. , Curtiss, R. III , Ingraham, J.L. , Lin, E.C.C. , Low, K.B. , Magasanik, B. , et al. (eds). Washington, DC: ASM press, pp. 1035–1063. (
- 1990) Biosynthesis and function of phospholipids in Escherichia coli. J Biol Chem 265: 1235–1238. , and (
- 2007) Lipid A modification systems in Gram-negative bacteria. Annu Rev Biochem 76: 295–329. , , , and (
- 2007) Role of FtsEX in cell division of Escherichia coli: viability of ftsEX mutants is dependent on functional SufI or high osmotic strength. J Bacteriol 189: 98–108. (
- 2009) Transport of lipopolysaccharide across the cell envelope: the long road of discovery. Nat Rev Microbiol 7: 677–683. , , and (
- 2001) Molecular Cloning: A Laboratory Manual, 3rd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. , and (
- 2013) FtsH-mediated coordination of lipopolysaccharide biosynthesis in Escherichia coli correlates with the growth rate and the alarmone (p)ppGpp. J Bacteriol 195: 1912–1919. , , and (
- 1993) Genetics of lipopolysaccharide biosynthesis in enteric bacteria. Microbiol Rev 57: 655–682. , and (
- 1999) The tetratricopeptide repeat domain and a C-terminal region control the activity of Ser/Thr protein phosphatase 5. J Biol Chem 274: 23666–23672. , , , , , and (
- 1996) Regulation of UDP-3-O-[R-3-hydroxymyristoyl]-N-acetylglucosamine deacetylase in Escherichia coli. J Biol Chem 271: 25898–25905. , , , , , and (
- 1984) Transcriptional organization within Escherichia coli cell division gene cluster: direction of transcription of cell separation gene envA. J Bacteriol 160: 724–732. , and (
- 1995) Escherichia coli FtsH is a membrane-bound ATP-dependent protease which degrades the heat-shock transcription factor σ32. EMBO J 14: 2551–2560. , , , , , , et al. (
- 2004) Biosynthesis, transport, and modification of lipid A. Biochem Cell Biol 82: 71–86. (
- 1992) Agents that increase the permeability of the outer membrane. Microbiol Rev 56: 395–411. (
- 1999) Localization of FtsI (PBP3) to the septal ring requires its membrane anchor, the Z Ring, FtsA, FtsQ, and FtsL. J Bacteriol 181: 508–520. , , , , and (
- 2009) Environmental pH affects transcriptional responses to cadmium toxicity in Escherichia coli K-12 (MG1655). FEMS Microbiol Lett 293: 58–64. , , , and (
- 1995) The envA permeability/cell division gene of Escherichia coli encodes the second enzyme of lipid A biosynthesis. J Biol Chem 270: 30384–30391. , , , , , , et al. (
- 2000) An efficient recombination system for chromosome engineering in Escherichia coli. Proc Natl Acad Sci USA 97: 5978–5983. , , , , , and (
- 2006) A comprehensive library of fluorescent transcriptional reporters for Escherichia coli. Nat Methods 3: 623–628. , , , , , , et al. (
- 2013) Mutants resistant to LpxC inhibitors by rebalancing cellular homeostasis. J Biol Chem 288: 5475–5486. , , , , , and (
- 2005) The global transcriptional response of Escherichia coli to induced sigma 32 protein involves sigma 32 regulon activation followed by inactivation and degradation of sigma 32 in vivo. J Biol Chem 280: 17758–17768. , , and (
- Top of page
- Experimental procedures
- Supporting Information
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.