Purification, Reconstitution, and Characterization of theCpxRAP Envelope Stress System of Escherichia coli*□S
Received for publication, June 16, 2006, and in revised form, October 16, 2006 Published, JBC Papers in Press, January 26, 2007, DOI 10.1074/jbc.M605785200
Rebecca Fleischer‡, Ralf Heermann§, Kirsten Jung§, and Sabine Hunke‡1
From the ‡Institut fur Biologie, Abteilung Physiologie der Mikroorganismen, Humboldt Universitat zu Berlin, D-10115 Berlin,Germany and the §Department Biologie I, Bereich Mikrobiologie, Ludwig-Maximilians-Universitat, Maria-Ward-Strasse 1a,D-80638 Munchen, Germany
InEscherichia coli theCpx sensor regulator system senses dif-ferent kinds of envelope stress and responds by triggering theexpression of periplasmic folding factors and proteases. It con-sists of the membrane-anchored sensor kinase CpxA, theresponse regulator CpxR, and the periplasmic protein CpxP.TheCpxpathway is induced in vivoby a variety of signals includ-ing pH variation, osmotic stress, and misfolded envelope pro-teins and is inhibited by overproduced CpxP. Because it is notclear how the Cpx pathway is able to recognize and correspondto somany different signals we overproduced, solubilized, puri-fied, and incorporated the complete membrane-integral CpxAprotein into proteoliposomes to analyze its biochemical proper-ties inmore detail. Autokinase and phosphotransfer activities ofthe reconstituted CpxA-His6 protein were stimulated by KCl.NaCl also stimulated the activities but to a lesser extent. Otherosmotic active solutes as glycine betaine, sucrose, and prolinehad no effect. The system was further characterized by testingfor susceptibility to sensor kinase inhibitors. Among these,Closantel inhibited the activities of solubilized but not of thereconstituted CpxA-His6 protein. We further analyzed theeffect of CpxP on CpxA activities. Purified tagless CpxP proteinreduced the phosphorylation status of CpxA to 50% but had noeffect on CpxA phosphotransfer or phosphatase activities. Asthe in vitro system excludes the involvement of other factors ourfinding is the first biochemical evidence for direct protein-pro-tein interaction between the sensor kinase CpxA and theperiplasmic protein CpxP resulting in a down-regulation of theautokinase activity of CpxA.
The bacterial cell wall is involved in a multitude of diversestructural, physiological, and adaptive processes includingtransport, elaboration of virulence factors, and cell division.These processes require specific sets of proteins whose correctfolding and assembly is controlled by periplasmic folding cata-lysts and proteases. In Escherichia coli and related species,
expression of some of the corresponding genes is regulated bythe Cpx sensor regulator system (reviewed in Ref. 1).The Cpx pathway consists of the sensor kinase CpxA, the
response regulator CpxR, and the periplasmicCpxP protein (1).The cpxA gene was originally reported as a gene regulating Fdonor activity in bacterial conjugation (2). It encodes the52-kDa histidine kinase CpxA, which is an integral membraneprotein of the cytoplasmic membrane that contains bothperiplasmic and cytoplasmic domains (3). The 26-kDa responseregulator CpxR is predicted to encode an OmpR-like cytosolictranscriptional activator (4).Signals activating theCpxpathway include elevated pH (5, 6),
altered membrane composition (7), overproduction of outermembrane lipoproteins such as NlpE (8), accumulation of mis-folded variants of the maltose-binding protein (MBP)2 (9),accumulation of pilus subunits (10), indole (11), and increasingosmolarity (12). The molecular mechanism of signal transduc-tion used by the Cpx sensor regulator system is not clear. CpxAfunctions as an autokinase, a CpxR kinase, and a CpxR-P phos-phatase in vivo (13). CpxA deletion mutants are uninducible,demonstrating that CpxA is necessary for signaling (14).Indeed, cpx* gain of function mutations located in the centralregion of the periplasmic domain of CpxA are insensitive tonormally activating signals and probably define a sensorydomain (15).Activation of the Cpx pathway results in increased produc-
tion of proteins involved in protein folding and degradation inthe periplasm, such as the heat shock protease DegP, the pep-tidyl prolyl cis/trans isomerases PpiA and PpiD, and the disul-fide oxidoreductase DsbA (1). In addition, it was shown thatCpxR�P represses motility and chemotaxis genes (16).Interestingly, overproduction of the small (19-kDa) periplas-
mic CpxP protein leads to decreased expression of Cpx-regu-lated genes and prevents Cpx activation by inducing signals (1).This effect depends on an intact CpxA sensory domain (15).Because an MBP-CpxP fusion was able to inhibit Cpx signaltransduction during spheroplast formation, a strong Cpx-in-ducing signal, it was suggested that the interaction betweenCpxA and CpxPmight be direct (17). On the basis of the obser-vation that the cpx* mutants were constitutively activated (15),
* This work was supported by grants from the Deutsche Forschungsgemein-schaft (Hu1011/1-1; to S. H. and R. F.) and (JU270/3-3; to K. J.). The costs ofpublication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked “advertisement” inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.
□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Fig. S1.
1 To whom correspondence should be addressed: Physiologie der Mikroorga-nismen, Chausseestr. 117, D-10115 Berlin, Germany. Tel.: 0049-0-30-2093-8122; Fax: 0049-0-30-2093-8126; E-mail: [email protected].
it was proposed that the presumed CpxA sensory domain nor-mally functions to maintain the kinase in a down-regulatedstate mediated by direct interaction with CpxP. When CpxP istitrated out bymisfolded envelope proteins, theCpx response isactivated (1, 18). The finding that the Cpx response can befurther activated in a cpxP deletion strain indicates that CpxP isnot required for signal transduction (19), suggesting that CpxPmight be responsible for fine-tuning the response. A recentstudy showed that single amino acid substitutions in a pre-dicted �-helix in the N-terminal domain of the CpxP proteinaffect its inhibitory function indicating that the N-terminaldomain of CpxP is critical for interaction with the sensor CpxAand might be the site of inhibitory activity (20).Most of the analysis on Cpx pathway signal transduction
was done in vivo. Here we describe an efficient method tosolubilize, purify, and reconstitute the complete Cpx sensorregulator system into proteoliposomes to study the bio-chemical properties of the signaling cascade in more detail.This in vitro system was further characterized by testing theeffect of known inhibitors of sensor regulator systems of Gram-positive bacteria (21). Finally, we analyzed the inhibitory effectof the periplasmic CpxP protein on the system. Our resultsprovide first biochemical evidence for a direct protein-proteininteraction between CpxA and CpxP resulting in inhibition ofthe CpxA autokinase activity.
EXPERIMENTAL PROCEDURES
Materials—[�-32P]ATP, peroxidase-conjugated anti-mouseand anti-rabbit IgG were purchased from GE Healthcare.Ethodin and Ofloxacin were obtained from Sigma. Vanadatewas from Strem Chemicals. Closantel and TCS were gifts fromThomas Schmulling (Free University, Berlin). Ni2�-NTA resinand His5 antibody were from Qiagen and Protino-Ni fromMacherey&Nagel. Detergentswere obtained fromGlyconBio-chemicals. Purified E. coli lipids were purchased from AvantiPolar lipids. Bio-Beads SM-2 were from Bio-Rad. Trypsin andtrypsin inhibitor were obtained from Roth. Synthetic oligonu-cleotide primers were from Invitrogen, vectors pIVEX2.3 andpIVEX2.4 from Roche Applied Science and vector pET15bfrom Novagen. All other reagents were reagent grade andobtained from commercial source.Bacterial Strains and Plasmids—E. coli JM109 [recA1 endA1
gyrA96 thi hsdR17 supE44 �� relA1 �(lac-proAB)/F’ traD36proA�B� lacIq lacZ�M15] (Amersham Biosciences) was usedas a carrier for the described plasmids. E. coli K-12 strainMG1655 [F� ��] (22) was used as parent strain for the PCR-based amplification of the cpxA, cpxR, and cpxP genes. E. colistrain BL21(DE3)�pLysS� (Novagen) was used for expressionof cpxA, cpxR, and cpxP under the control of the T7 promoter.Construction of cpxA-His6, His6-cxpR, and His6-cpxP—Con-
struction of pI3cpxAwas achieved by amplifying the cpxA cod-ing region using primers CpxA5 (5�-ATCATATGATAGGCAGCT TAA CCG CG-3�) and CpxA3 (5�-ATC CCG GGA CTCCGCTTATACAGCGGCAACC-3�). The resulting fragmentwas cloned into the NdeI and SmaI sites of pIVEX2.3.The cpxR coding region was amplified using primers CpxR5
ATC AG-3�) and cloned into the NcoI and BamHI sites ofpIVEX2.4, resulting in pI4cpxR.Construction of pRF6 was achieved by amplifying the cpxP
coding region using primers 5XhoI_EcoCpxP (5�-GAG ACTCGA GGC TGA AGT CGG TTC AGG CGA TAA C-3�) and3BamHI_EcoCpxP (5�-GAG TGG ATC CCT ACT GGG AACGTG AGT TGC TAT C-3�). The product was digested withXhoI and BamHI and cloned into the corresponding sites ofpET15b. The resulting constructs were confirmed bysequencing.Expression of Proteins and Preparation of Cytosolic Fraction
and Membrane Vesicles—E. coli strain BL21(DE3) �pLysS�was transformed either with pI3cpxA, pI4cpxR, or pRF6,respectively, and grown at 30 °C with aeration in LB mediumsupplemented with 100 �g/ml ampicillin and 30 �g/ml chlor-amphenicol. Gene expression was induced with 0.5 mM isopro-pyl-1-thio-�-D-galactopyranoside for 3–4 h. Cells were har-vested and resuspended in buffer Z (50 mM Tris/HCl pH 7.5,0.15 M NaCl, 20% glycerol (v/v), 0.1 mM phenylmethylsulfonylfluoride). Cells were fractionated by sonication on ice and ultra-centrifugation into membrane fraction (pellet) and cytosolicfraction (supernatant). Themembrane fraction resuspended inbuffer Z, and the cytosolic fraction were frozen in liquid nitro-gen and stored at �80 °C until use.Purification of His6-CpxR—His6-CpxR fusion protein was
purified by means of Ni-affinity chromatography in batch.Binding of the protein (�500mg of cytosolic proteins/2.5 ml ofNi-NTA resin) was done in buffer R (50 mM Tris/HCl, pH 7.5,0.15 M NaCl, 10% glycerol (v/v), 0.1 mM phenylmethylsulfonylfluoride). Bound His6-CpxR was eluted with an imidazole con-centration of 150 mM in buffer R. His6-CpxR-containing frac-tions were mixed and further purified with a Protino Ni 2000prepacked column kit according to themanufacturer’s instruc-tions (Macherey & Nagel). Purified His6-CpxR protein waspassed through a PD10 column (GE Healthcare) to removeimidazole and stored at �80 °C until use.Purification of CpxP—His6-CpxP fusion protein was purified
by means of Ni-affinity chromatography in batch. Binding ofthe protein (600 mg of cytosolic protein/3 ml Ni-NTA) wasdone in the presence of 5 mM imidazole in buffer P (50 mMTris/HCl, pH7.5, 0.5MNaCl, 10%glycerol (v/v), protease inhib-itor mixture (Roche Applied Science)). After washing withbuffer P containing 20 mM imidazole His6-CpxP was eluted byincreasing the imidazole concentration to 150 mM. His6-CpxP-containing fractions were passed through a PD10 column toremove imidazole and further purified by an SP-Sephadex col-umn (Amersham Biosciences). Bound His6-CpxP was washedextensively with buffer P2 (50 mMMOPS/K� pH 6) containingincreasing concentrations of NaCl (50 mM-200 mM), elutedwith buffer P2 containing 0.5 M NaCl and buffered into bufferP3 (50mMTris/HCl, pH 7.5, 0.15 MNaCl, 10% glycerol (v/v)) bypassing through a PD10 column. The His6-fusion from the Nterminus of CpxPwas cleaved by thrombin using the THROM-BIN CleanCleaveKit essentially as described (Sigma).Purification of CpxA-His6—For detergent selection, the fol-
lowing detergents were tested: 1% Triton X-100, 1% lauryldi-methylamine oxide (LDAO); 1% decylmaltoside (DM); 1%dodecylmaltoside (DDM), and 1% octyl-glucoside (OG). After
Cpx Pathway Reconstitution
8584 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 12 • MARCH 23, 2007
incubation on ice for 60 min, mixtures were centrifuged, andaliquots of the supernatant containing solubilized proteinswere analyzed by immunoblotting. The yield of CpxA-His6 sol-ubilization (%) is used as measure for the distribution of theamounts of CpxA-His6 between the soluble and the membranefraction after treatment of membrane vesicles with detergent.1 ml of 10� DM was added to 9 ml of membrane vesicles
(�10 mg/ml) prepared from BL21(DE3) �pLys/pI3cpxA�.Samples were incubated on ice for 1 h and centrifuged at200,000 � g at 4 °C for 30 min. The supernatant fraction con-taining solubilized CpxA-His was purified by affinity chroma-tography. Ni2�-NTA resin equilibrated with buffer A (50 mMTris/HCl, pH 7.5, 0.15 M NaCl, 10% glycerol (v/v), 0.1% DM(w/v), 2 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluo-ride). Binding of CpxA-His6 was carried out by incubating thesolubilized fraction and Ni2�-NTA resin at 4 °C for 1 h. Theprotein-resin complex was packed into a column, and unboundproteins were removed by stepwise washing with buffer A con-taining 10 and 50 mM imidazole. CpxA-His6 was eluted byincreasing the imidazole concentration to 150mM. CpxA-His6-containing fractions were passed through a PD10 column toremove imidazole and stored in 50 mM Tris/HCl, pH 7.5, 10%glycerol (v/v), 0.1% DM (w/v) at �80 °C.Preparation of Proteoliposomes—Purified CpxA-His6 was
incorporated into liposomes essentially as described (23).Briefly, E. coli phospholipids (Avanti) were dried under astream of nitrogen, and slowly redissolved in 50 mM Tris/HCl,pH 7.5, 10% glycerol (v/v), 0.47% Triton X-100 (v/v) over aperiod of 2 h. To this mixture purified CpxA-His6 was added(ratio of lipid to protein, 100:1 w/w). Themixture was stirred atroom temperature for 10min. Bio-Beads were added in a bead/detergent ratio of 10:1 (w/w), and themixturewas gently stirredat 4 °C overnight. After 16 h, fresh Bio-Beads were added addi-tionally, and the mixture was stirred for another 2 h. The pro-teoliposome solution was pipetted off, and proteoliposomeswere collected by centrifugation for 30min at 200,000 � g. Thepellet was resuspended in 50mMTris/HCl, pH 7.5, 10% glycerol(v/v), and 2mM dithiothreitol. The efficiency of CpxA incorpo-ration into liposomes was calculated as the ratio between theamount of protein in proteoliposomes and in the supernatantafter ultracentrifugation. Proteoliposomes were either usedinstantly or stored in liquid nitrogen. CpxP-containing proteo-liposomes were prepared as described above, but before Bio-Bead treatment, purified CpxP protein was added at a 5–10-fold molar excess to the CpxA protein-lipid mixture.Proteolysis with Trypsin—Proteoliposomes were incubated
with trypsin (15 ng/�l) at a protease/protein ratio of 1:100 atroom temperature. At different times, the reactionwas stoppedby trypsin inhibitor (2 �g/�l), and the proteins were subjectedto Western blot analysis. MalK protein of Salmonella typhi-murium was incorporated into proteoliposomes to controltheir impermeability for trypsin.Phosphorylation and Dephosphorylation Assays—To test
autophosphorylation, solubilized CpxA, CpxA in membranevesicles or CpxA in proteoliposomes was incubated with 40 �M[�-32P]ATP (1.2 Ci/mmol) in phosphorylation buffer (50 mMTris/HCl, pH7.5, 10%glycerol (v/v), 2mMdithiothreitol, 50mMKCl, 5 mM MgCl2) at room temperature. When indicated vari-
ous solutes were additionally added. At different time points,aliquots were removed and mixed with 5� SDS sample buffer.To analyze phosphotransfer, purified His6-CpxR was added tothis mixture after 10 min, additional samples were taken, andthe reaction was stopped by the addition of 5� SDS samplebuffer.To test dephosphorylation, purified His6-CpxR was phos-
phorylated as described abovewithCpxA-His6-containing pro-teoliposomes. After incubation for 25min, the phosphorylationmixture was centrifuged for 30min (13,000� g, 4 °C), and ATPwas removed by the addition of 20 mM MgCl2, 2 mM glucose,and 5.4 units of hexokinase. Dephosphorylation of His6-CpxR�P was initiated by the addition of fresh CpxA-His6 inproteoliposomes. At the indicated times, aliquots were taken,and the reaction was stopped as described above.To test the whole signaling cascade in vitro, CpxA-His6-con-
taining proteoliposomes (0.5�M) and purifiedCpxR (2 or 4�M)were incubated in phosphorylation buffer at room temperature.Phosphorylationwas initiated by the addition of 100�M [�-32P]ATP (0.48 Ci/mmol). Samples were taken at times indicatedand mixed with 5� SDS sample buffer. To analyze CpxA-His6phosphatase activity 1 mM ADP was added after 30 min, addi-tional aliquots were removed at times indicated and stopped asdescribed above.All samples were immediately subjected to SDS-PAGE (24).
Gels were dried and phosphorylated proteins were detected bya PhosphorImager system (Molecular Imager Fx, and SoftwareQuantity One, Bio-Rad) using [�-32P]ATP as a standard.Inhibitor Studies—Inhibitors (except vanadate) were dis-
solved inMe2SO. Vanadate solutions were prepared in water atpH 10 as described (25). The phosphorylation assays were per-formed as described above in the presence of the indicatedinhibitors. To reflect the presence ofMe2SO as a solvent, it wasalso included in control reactions in an equivalent final concen-tration (10% (v/v)).Immunological Analysis—For Western blot analysis protein
samples were subjected to SDS-PAGE. Proteins were electro-blotted, and immunoblots were probed with antiserum topenta-His fusion (mouse) to visualize full-length CpxA-His6 orwith antiserum to MalK (rabbit). Immunodetection was per-formed by using the ECL-kit (GE Healthcare) with a peroxi-dase-conjugated anti-mouse or anti-rabbit IgG.Analytical Methods—Protein content was determined using
the BCA protein assay from Pierce according to the manufac-turer’s instructions.To determine relative protein content of proteoliposomes,
proteins were separated by SDS-PAGE using 12.5% acrylamidegels and stained with Coomassie Blue. Different amounts of thecorresponding purified proteins were used as a standard on thesame gel. Gels were scanned with the Molecular Imager Fx(Bio-Rad), and protein bands were quantified with QuantityOne (Bio-Rad).
RESULTS
Purification of Cpx Components—To allow easy purificationof CpxA a hexa-His tag was attached at its C terminus. Theactivity of the CpxA-His6 protein was comparable to the nativeprotein: both autokinase and kinase activities in membrane
Cpx Pathway Reconstitution
MARCH 23, 2007 • VOLUME 282 • NUMBER 12 JOURNAL OF BIOLOGICAL CHEMISTRY 8585
fractions were only slightly lower than in authentic CpxA (datanot shown). Thus, CpxA-His6 was used for further in vitroanalysis.As a prerequisite for CpxA-His6 purification by affinity chro-
matography the protein had to be efficiently solubilized.Because the yield of solubilizedCpxA-His6 was found to be 80%when decylmaltoside was used as detergent (69% n-dodecyl-maltoside, 38% Triton X-100, 25% octyl-glucoside), this deter-gent was used for all purification steps. Interestingly, the differ-ent detergents did not alter autokinase activity of the solubleCpxA-His6 significantly (data not shown). The highest degreeof purified CpxA-His6 was achieved when the protein wasbound to the Ni2�-NT-agarose in the presence of 50 mM imid-azole and 0.15 M NaCl. In a typical experiment about 1 mg ofCpxA-His6 was obtained from a 1-liter culture. As demon-strated by Coomassie Blue staining (Fig. 1A), the purity ofCpxA-His6 was nearly 90%.
For CpxR purification a hexa-His tag was attached at its Nterminus. The activity of the protein was comparable to nativeCpxR as it was shown previously (11). The His6-CpxR proteinwas purified by the use of two affinity chromatography steps.This procedure yielded 95% pure His6-CpxR protein (Fig. 1B,compare lanes 2 and 3).
For the purification of CpxP we constructed an expressionplasmid in which the 5�-sequence of cpxP corresponding to thesignal peptide was replaced by six His codons. C-terminalfusions with the full-length protein in different vector systemscould not be stably expressed in the periplasm even in a degPstrain (data not shown). Therefore, the protein was overpro-duced and purified from the cytosol. After purification by Ni-chelate chromatography the N-terminal histidine residueswere cut off using a thrombin cleavage site resulting in a taglessCpxP protein (Fig. 1C). Purified CpxP preparations containedan additional protein with an approximate size of 70 kDa. MSspectrometry identified this protein as the chaperone DnaK(data not shown). Because DnaK was not found in proteolipo-somes after reconstitution (Fig. 7A, lane 3) the preparation wasnot purified further.Incorporation of CpxA-His6 into Proteoliposomes—Purified
CpxA-His6 protein was incorporated into E. coli phospholipidsusing the detergent-mediated method as described (26). Theefficiency of CpxA-His6 incorporation into liposomes was�40% (data not shown).To analyze the orientation of CpxA-His6 in proteoliposomes
the susceptibility of its C-terminal domain to trypsin wastested. In intact cells, the C terminus is on the cytoplasmic sideof themembrane. Because 82% of the CpxA-His6 in the proteo-liposomes lost a C-terminal peptide as a result of trypsin treat-ment (as shown byWestern blotting using an anti-His antibodyand calculated from the average of two independent experi-ments), the majority of the CpxA-His6 protein in proteolipo-somes is in the inside-out orientation (supplemental Fig. S1). Asa control for the integrity of the proteolipsomes, soluble MalKATPase was incorporated along with CpxA-His6. No proteo-lytic cleavage of MalK was detected after 1.5 min, while thefreely accessible MalK protein was fully degraded by trypsinafter 1 min (27).
Characterization of CpxA-His6 Activities in Proteoliposomes—CpxA-His6 proteoliposomes were autophosphorylated in thepresence of [�-32P]ATP at room temperature. The autophos-phorylation increased almost linearly for 10 min (Fig. 2A).Autokinase activity of CpxA-His6 proteoliposomes was 1.5times higher compared with that of CpxA-His6 in detergent(data not shown).Next, we tested transfer of the phosphoryl group fromCpxA-
His6 proteoliposomes to purified His6-CpxR. In this experi-ment CpxA-His6 in proteoliposomes was phosphorylated for
FIGURE 1. Purification of CpxA, CpxR, and CpxP. Samples from differentstages of the purification procedure were separated on 12.5% SDS-PAGE:A, solubilization and purification of CpxA: Lanes: 1, everted membrane vesi-cles; 2, decylmaltoside-solubilized protein; 3, membrane vesicles after solubi-lization; 4, purified CpxA-His6 after elution from Ni-NTA agarose with imidaz-ole (3.5 �g of protein). B, purification of CpxR: Lanes: 1, cytoplasmic fraction; 2,His6-CpxR after elution from Ni-NTA-agarose with imidazole; 3, purified His6-CpxR after elution from Protino-Ni-agarose with imidazole (4 �g of protein).C, purification of CpxP: Lanes: 1, cytoplasmic fraction; 2, His6-CpxP after elu-tion from Ni-NTA-agarose with imidazole; 3 purified CpxP after thrombincleavage (4 �g of protein). The proteins were stained with Coomassie Blue.The molecular mass markers are shown on the left.
Cpx Pathway Reconstitution
8586 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 12 • MARCH 23, 2007
10 min, and subsequently His6-CpxR was added. The mixturewas incubated, and samples were taken at the times indicated.Transfer of the phosphoryl group to His6-CpxR was observedwithin 0.1 min (Fig. 2B).Next we studied the phosphatase activity of CpxA-His6 in
proteoliposomes. First, we tested the intrinsic stability of phos-pho-CpxR. Purified His6-CpxR protein was phosphorylated byCpxA-His6-containing proteoliposomes as described above.Then, His6-CpxR�32P was separated from the reaction mix-ture and incubated in the presence and absence of ADP. Sam-ples were taken at various time points. Phosphorylated His6-CpxR was stable for a period of 60 min independently of thepresence of ADP (Fig. 2C).To test now the dephosphorylation of His6-CpxR�32P by
CpxA, CpxA-His6-containing proteoliposomes were added,and dephosphorylation was monitored over a period of 60min.The half-life of His6-CpxR�32P was 55 min. In the presence ofADP the half-life was shortened to 20 min (Fig. 2C). In sampleslacking ADP the level of His6-CpxR�32P increased over thefirst 5 min. This effect was constantly observed but, as it lies inthe range of the standard deviation, it was not studied further.Apparently, ADP seems to trigger the phosphatase activity ofCpxA. A similar effect was reported for EnvZ and its cognateresponse regulator OmpR (23). Taken together, CpxA-His6-proteoliposomes and His6-CpxR catalyzed all known enzy-matic activities.Influence of Various Solutes and pH Variations on the Auto-
phosphorylation Activity of CpxA-His6 in Proteoliposomes—Invivo data suggested that the Cpx system might be involved insensing environmental osmolarity (12). To prove this hypothe-sis we tested the autokinase activity of CpxA-His6 in proteoli-posomes in the presence of various solutes. It was found thatthis activity of CpxA-His6 was significantly stimulated in thepresence of KCl and RbCl (Fig. 3A). NH4Cl also had a stimula-tory effect but to a lower extent. In contrast, other osmoticactive solutes as NaCl, sucrose, and trehalose did not influenceCpxA autokinase activity. Furthermore, the compounds gly-cine betaine, proline, and glutamate that are accumulated incells exposed to an osmotic upshift had no stimulatory effect.We also analyzed the concentration-dependent effect of KCl
on CpxA-His6 autophosphorylation activity using a concentra-tion range between 0.5 and 500mMKCl. The stimulatory effectof KCl was found to be linear until a concentration of 50mM, athigher concentrations the stimulation followed a saturationcurve (data not shown).In vivo data showed that the Cpx system is activated by mild
alkaline pH (6). Therefore, we analyzed the effect of pH varia-tions on the autophosphorylation activity of CpxA-His6 pro-teoliposomes. We observed a stimulation of CpxA autokinase
FIGURE 2. Enzymatic activities of purified CpxA-His6 in proteoliposomes.A, time-dependent autophosphorylation activitiy of CpxA-His in proteolipo-somes. CpxA (1 �M) was incubated with [�-32P]ATP in buffer containing 50 mM
KCl. B, phosphotransfer activity of CpxA in proteoliposomes. CpxA was phospho-rylated under standard conditions, at time point 0 min, 3 �M purified CpxR wasadded to initiate phosphotransfer activity. C, phosphatase activity of CpxA in
proteoliposomes. CpxR was phosphorylated by CpxA-His6-proteoliposomeswith [�-32P]ATP (described under “Experimental Procedures”). His6-CpxR�Pwas incubated in the presence or absence of ADP with MgCl2 and dephos-phorylation was initiated by adding CpxA-containing proteoliposomes (0.5�M). The reaction was started at time 0 with 2 �M [32P-CpxR]. In all assays,samples were taken at the times indicated, separated by SDS-PAGE and ana-lyzed on a PhosphorImager (upper part). The amounts of [32P]phospho-CpxAand [32P]phospho-CpxR were determined with a PhosphorImager (lower part)using [�-32P] ATP as standard. Shown are averages S.E. from at least threedifferent experiments.
Cpx Pathway Reconstitution
MARCH 23, 2007 • VOLUME 282 • NUMBER 12 JOURNAL OF BIOLOGICAL CHEMISTRY 8587
activity bymild alkaline pH but not by neutral ormild acetic pH(Fig. 3B).Inhibition of the Autokinase and Phosphotransfer Activities of
CpxA-His6 in Proteoliposomes by Inhibitors of Two-componentSystems—We further characterized themolecular properties ofCpxA inmore detail by studying the effect of chemicals that areknown to inhibit senor kinases. The potential inhibitors used inthis study are the salicylanilides Closantel (28) and Tetrachlo-rosalicylanilid (TCS) (29), the small hydrophobic intercalator-like molecule Ethodin (21), the quinolone Ofloxacin (29), andthe phosphate analogue vanadate.Because all compounds except vanadate were dissolved in
Me2SO, this solvent was used as an additional control. Interest-ingly, Me2SO alone caused a reduction of the autophosphoryl-ation activity of reconstituted CpxA-His6 by about 30% in com-parison to the control without any additive (Fig. 4).For CpxA-His6 in proteoliposomes we observed that the
inhibitory effect of Closantel did not differ significantly fromthe inhibitory effect of the solvent (Fig. 4), Thus, Closantel waswithout effect. In contrast, theClosantel analogueTCS stronglyaffected the autophosphorylation activity of CpxA-His6 (Fig. 4).The IC50 value was determined to be 150.8 40.9 �M. ForEthodin we observed reduced autophosphorylation activity of
CpxA-His6 (Fig. 4)with a IC50 value of 219.2 42.4�M.Ofloxa-cin slightly inhibited the autophosphorylation activity ofCpxA-His6 (Fig. 4). The proportion of phosphorylated CpxA proteinwas 87% of the Me2SO control reaction.In addition to the established sensor kinase inhibitors we
choose vanadate, a classical inhibitor of P-type ATPases andreconstituted ABC transporters. Vanadate is supposed toinhibit the formation of a phosphorylated intermediate in thecase of P-type ATPases (31) and to block the release of ADP inthe case of ABC transporters (32, 33). Vanadate caused a slightreduction of the autophosphorylation activity of CpxA-His6 to75% of the control without any additive (Fig. 4).We also analyzed the effect of these potential inhibitors on
the phosphotransfer activity of CpxA-His6 in proteoliposomes.No remarkable differences were observed for most of the com-pounds in comparison to the effect on the autophosphorylationactivity. However, CpxA phosphotransfer activity was slightlyincreased by vanadate to 116% of the control (data not shown).Closantel Inhibits the Autophosphorylation Activity of the
Detergent-solubilized CpxA-His6 but Not of CpxA-His6 inProteoliposomes—TCS and Ethodin were found to be inhibi-tors for autophosphorylation and phosphotransfer activitiesof reconstituted CpxA-His6. Interestingly and to our sur-prise, Closantel did not inhibit these two activities whenCpxA-His6 was reconstituted in proteoliposomes. To studythis phenomenon further, we compared the effect of Closan-tel on the autophosphorylation activity of detergent-solubi-lized with that of reconstituted CpxA-His6. In contrast to theeffect on the reconstituted CpxA protein, the activity of solubi-lized protein was reduced to 6% of the control by 0.2 mM Clos-antel (Fig. 5). To exclude that the hydrophobic componentClosantel was not able to inhibit reconstituted CpxA-His6because of its dilution in the lipid phase, we tested a higherinhibitor concentration. After addition of 1 mM Closantel, no
FIGURE 3. Influence of different solutes and pH variations on the auto-phosphorylation activity of CpxA. Autokinase activity of CpxA-His6 in pro-teoliposomes (1 �M) was probed in buffers containing various solutes (A).Shown are the results obtained with various solutes tested at a low (0.5 mM) ora high (0.5 M) concentration. B, CpxA-His6 containing proteoliposomes (1 �M)were prepared in the presence of indicated pH, collected and resuspended inphosphorylation buffer containing 50 mM KCl, with either a constant pH of 7.5(white bars) or with indicated pH (black bars). For both experiments sampleswere taken after 10 min. The averages S.E. from three different experimentsare shown.
FIGURE 4. Inhibition of the autokinase activity of CpxA-His6-proteolipo-somes by histidine kinase inhibitors. Purified CpxA was incorporated intoliposomes and the effect of different inhibitors (200 �M) on autokinase activ-ity was tested. The phosphorylation assay was performed as described under“Experimental Procedures” with CpxA containing proteoliposomes (0.5 �M)in buffer containing 50 mM KCl. The relative amount of [32P]phospho-CpxAafter incubation the CpxA-containing proteoliposomes (1 �M) with differentinhibitors as indicated. Shown are averages S.E. from two different experi-ments with two duplicates each. For CpxA without any inhibitor a phospho-rylation level of 1 nmol of [32P]CpxA/mg was obtained after 30 min.
Cpx Pathway Reconstitution
8588 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 12 • MARCH 23, 2007
activity was observed for the solubilized CpxA-His6, but theactivity of the protein in proteoliposomes was not inhibited(Fig. 5).Effect of These Inhibitors on the Autokinase Activity of
EnvZ-His6—The inhibitor studies with CpxA-His6 indicatedclear differences between the detergent-solubilized and theprotein reconstituted in proteoliposomes. To examine whetherthe difference in susceptibility of the two forms is a uniquetrait of CpxA or if it is a general property of sensor kinases weanalyzed the effect of Closantel and of TCS on the EnvZ
sensor kinase of E. coli. Purification and reconstitution ofEnvZ was carried out as described before (23). As for theCpxA protein we observed inhibition by the solvent Me2SOalone. The amount of phosphorylated EnvZ protein wasreduced for the solubilized protein to 50% and for the recon-stituted protein to 68% (Fig. 6). In support of our finding forCpxA only the solubilized EnvZ was inhibited by Closantelbut not the reconstituted protein. In contrast, TCS was notable to inhibit the activity of EnvZ-His6 under the testedconditions (Fig. 6).Taken together, the results of our inhibitor studies indicate
that the soluble and reconstituted forms of membraneanchored sensor kinases differ significantly in certain biochem-ical properties. This clearly underlines the importance of ana-
FIGURE 5. Effect of Closantel on solubilized CpxA-His6 protein and CpxA-His6 in proteoliposomes. Different concentrations of Closantel were testedon the autophosphorylation activity of CpxA-His6 solubilized in detergent(soluble) or CpxA-His6-proteoliposomes (PLS) as described under “Experi-mental Procedures”. After incubation with Me2SO maximal values of 0.001nmol of [32P-CpxA]/min � mg soluble and [32P]CpxA incorporated into pro-teoliposomes were obtained. Shown are averages S.E. from three differentexperiments.
FIGURE 6. Inhibition of the autophosphorylation activity of solubilizedEnvZ-His6 and EnvZ-His6 in proteoliposomes. The effect of two concentra-tions of Closantel and TCS was tested on the autophosphorylation activity ofpurified EnvZ solubilized in detergent (soluble) or incorporated into proteo-liposomes (PLS) as described in Ref. 23. Shown is a representative of twoindependent experiments.
FIGURE 7. Inhibition of autophosphorylation activity of CpxA-His6 byCpxP. Proteoliposomes containing purified CpxA-His6 and CpxP proteinswere prepared and tested for autophosphorylation as described under“Experimental Procedures.” A, CpxA-His6 containing proteoliposomes loadedwith CpxP or buffer were separated on a 12.5% SDS-gel, and stained by Coo-massie Blue. Protein was quantified in the scanned gels with Quantity Onesoftware. B, autophosphorylation activity of CpxA-His in proteoliposomesloaded with purified CpxP (1 �M), His6-CpxP (5 �M), or buffer. CpxA (1 �M) wasincubated with [�-32P]ATP (40 �M) in buffer containing 50 mM KCl. Sampleswere separated by SDS-PAGE and analyzed with a PhosphoImager (upperpart). The amounts of [32P]CpxA were determined on a PhosphorImager(lower part) using [�-32P]ATP as a standard. In the absence of CpxP, an auto-phosphorylation rate of 0.04 0.007 nmol/min � mg was calculated. Shownare averages S.E. from three different experiments.
Cpx Pathway Reconstitution
MARCH 23, 2007 • VOLUME 282 • NUMBER 12 JOURNAL OF BIOLOGICAL CHEMISTRY 8589
lyzing the reconstituted and not the soluble version of a sensorkinase.Inhibition of CpxA-His6 Autophosphorylation Activity by
CpxP—In vivo data showed that overproduction of theperiplasmic CpxP protein leads to inhibition of the Cpx signal-ing cascade (17). Our defined in vitro system provides an exper-imental platform to test if this is because of direct protein-protein interaction between the CpxP protein and the CpxAsensor kinase because of the involvement of other factors can beexcluded. In addition, our in vitro system allows us to analyzewhich catalytic function of the CpxA protein is affected byCpxP. CpxP might either inhibit the autophosphorylationactivity or induce the phosphatase activity of CpxA (17).To analyze the inhibitory effect of CpxP on CpxA-His6, we
mixed purified CpxP protein with solubilized CpxA-His6.AlthoughCpxPwas provided in a 5-fold excess, no inhibition ofautophosphorylation activity of the solubilized CpxA-His6 pro-tein by CpxP was observed (data not shown). To analyze nowthe effect of CpxP on reconstituted CpxA-His6, purified CpxPwas incorporated into proteoliposomes to allow access of thisprotein to the periplasmic loop of CpxA. Although we variedthe ratio between CpxP and CpxA-His6 during the reconstitu-tion process up to 10:1, the CpxP:CpxA ratio found in the pro-teoliposomes was almost 1:1 (Fig. 7A). Under these conditionsCpxP inhibited the autophosphorylation activity of CpxA-His6by 51% (Fig. 7B). Interestingly, the His6-CpxP variant wasunable to inhibit CpxA autophosphorylation activity (Fig. 7B).Inhibition of the in Vitro CpxRA Signal Transduction Cas-
cade by CpxP—Next, we examined the effect of CpxP on theCpxR phosphorylation status. Because our experimental setupfor monitoring CpxA kinase activity (Fig. 2B) would be influ-enced by the inhibitory effect of CpxP on CpxA autophospho-rylation and because in whole cells a stepwise addition of thecomponents of a signaling pathway is not possible, we estab-lished an in vitro signal transduction cascade for the CpxARPsystem. PurifiedCpxA-His6 in proteoliposomes andHis6-CpxRwere mixed at a molar ratio of 1:8, and the reaction was startedby the addition of [�-32P]ATP. Samples were taken after 15 and30 min. Then, ADP was added to shift CpxA activities todephosphorylation, and samples were taken after an additional15 and 30min (Fig. 8A,white symbols). As shown before, CpxAhas a autokinase and phosphotransfer activities (Fig. 8A, whitesymbols up to 30 min) and a dephosphorylation activity towardCpxR (Fig. 8A, white squares after 30 min).This in vitro signaling cascade system was now used to ana-
lyze the effect of CpxP on the CpxRA system. Proteoliposomescontaining CpxA-His6 and purified CpxP were mixed withHis6-CpxR at a molar ratio 1:1:8, and the experiment was car-ried out as described above. As shown before, CpxP caused adecrease in CpxA autophosphorylation (Fig. 8A, black cir-cles). Interestingly, CpxP did not influence the phosphoryl-
ation of CpxR (Fig. 8A, black squares) indicating that CpxPdid not induce the phosphatase activity of CpxA as specu-lated by others (17).As we could not observe an inhibitory effect of CpxP on the
CpxR phosphorylation status we speculated if thismight be dueto an excess of CpxR protein in the reaction mix. Thus wereduced the molar ratio between the CpxP:CpxA:CpxR pro-teins to 1:1:4. This time, not only the phosphorylation status ofthe CpxA protein (Fig. 8B, black circles) was reduced in CpxP-containing proteoliposomes but also that of phosphorylatedCpxR protein (Fig. 8B, black squares).Taken together, our data indicate that CpxP inhibits CpxA
autokinase activity, but does not induce phosphatase activity. Inaddition, the balance between the sensor kinase CpxA and theresponse regulator CpxR seems to be critical for an inhibitoryeffect mediated by the periplasmic CpxP protein on the com-plete signaling cascade.
DISCUSSION
The aim of the present study was to establish an in vitrosystem to investigate the biochemical properties of the CpxRAenvelope stress system and its interaction with the periplasmicCpxP protein. Here, we describe a protocol to overproduce,purify, and reconstitute the full-length membrane-integralCpxA protein of E. coli into proteoliposomes as a C-terminalhexa-His fusion protein. The use of a C-terminal His tag fusionis well established for the purification and reconstitution ofmembrane-anchored histidine kinases such as the E. coliKdpD(35) and EnvZ (23) proteins. All known enzymatic activitieswere detectable for CpxA-His6-proteoliposomes under stand-ard phosphorylation conditions, although rates differed some-what compared with the purified cytoplasmic histidine kinasedomain of the CpxA protein (36). In case of the full-lengthCpxA protein, transfer of the phosphoryl group to CpxR pro-tein was slower and was not complete after 20 min. In contrast,a very rapid phosphotransfer to CpxR was shown for the cyto-plasmic histidine kinase domain of CpxA (36). Thus, our dataare in agreement with a suggestion made for the reconstitutedKdpD protein that interactions between the domains of sensorkinases may fine-tune their enzymatic activities (35).Our studies on the effect of solutes and pH conditions sup-
port in vivo data of Cpx pathway activation by mild alkaline pH(6). In addition, our data revealed the importance of KCl forCpxA autophosphorylation and phosphortransfer activities.Other solutes such as RbCl, NH4Cl, and NaCl also mediatedactivation of CpxA activities although to a lower extent com-pared with KCl. Because sucrose and trehalose did not showany effect on CpxA activities the data also imply that stimula-tion is not caused by an increase of osmolality per se as sug-gested by recent in vivo data (12). Furthermore, the compoundsproline or glycine betaine, which are might to be accumulated
FIGURE 8. Effect of CpxP on the Cpx signaling cascade in vitro. Phosphorylation assay for the CpxA/CpxR signal transduction cascade was performed asdescribed under “Experimental Procedures.” A, CpxA-His6-containing proteoliposomes (0.5 �M) and His6-CpxR (4 �M) were mixed in reaction buffer containing50 mM KCl. The reaction was started by the addition of 100 �M [�-32P]ATP, and samples were taken at the indicated times. After 30 min, ADP (1 mM) was added,and further samples were taken. The experiment was repeated with CpxP (0.5 �M) loaded in proteoliposomes. B, effect of CpxP on the whole Cpx signalingcascade containing a reduced amount of CpxR. 2 �M His6-CpxR protein was added to CpxA-His6 or CpxA-His6/CpxP-containing proteoliposomes. For both setsof experiments, samples were separated by SDS-PAGE and analyzed on a PhosphorImager. The amounts of [32P]CpxA and [32P]CpxR were quantified on aPhosphorImager using [�-32P]ATP as a standard. Shown are averages S.E. from three different experiments.
Cpx Pathway Reconstitution
MARCH 23, 2007 • VOLUME 282 • NUMBER 12 JOURNAL OF BIOLOGICAL CHEMISTRY 8591
in osmotically stressed cells did not exhibit a stimulatory effecton CpxA autokinase activity.Both the solubilized and reconstituted CpxA-His6 were fur-
ther characterized for susceptibility to known sensor kinaseinhibitors. To the best of our knowledge, this is the first timethat inhibitors of two-component systems have been tested onamembrane-anchored sensor kinase that was purified as a full-length protein and reconstituted into proteoliposomes. Untilnow, soluble histidine kinases have received the most attentionas potential targets of new antimicrobial agents. Examples arethe B. subtilis KinA kinase that is essential for sporulation (37),the E. coli NRII (GlnL) kinase involved in nitrogen regulation(21) and the AlgR2 kinase essential for alginate production ofmucoid strains of Pseudomonas aeruginosa (38). Although theeffect of sensor kinase inhibitors on some membrane-integralhistidine kinases was also studied, these proteins were neveranalyzed after purification and reconstitution. Either artificiallytruncated, soluble forms, as in the case of Thermotoga mari-tima HpkA kinase (29), or membrane preparations, as in thecase of the VanS kinase of Enterococcus faecium (39) were used.
We investigated TCS and Ethodin as inhibitors for autoki-nase and phosphotransfer activities of both the soluble and thereconstituted CpxA-His6 protein. Ofloxacin and vanadate didnot inhibit any activities of the CpxA protein.Interestingly, Closantel inhibited only the activities of solu-
bilized CpxA-His6 protein but not of CpxA-His6 in proteolipo-somes. This observation was confirmed by the use of the recon-stituted EnvZ protein of E. coli. EnvZ, like CpxA, is an innermembrane protein with a short cytosolic N terminus, a 114-amino acid periplasmic domain flanked by two membrane-spanning domains and a cytoplasmic C-terminal domain (40).As observed for CpxA, only the soluble EnvZ was inhibited byClosantel.Taken together, our results clearly indicate that purified
membrane-integral histidine kinases mimic their natural activ-ities only after reconstitution into proteoliposomes. Our obser-vations for the histidine kinase inhibitor agent Closantel under-score the importance of studying the full-length membraneprotein.The availability of purified components allowed the recon-
struction of the whole CpxRAP signal transduction cascade invitro to analyze in more detail the in vivo observation that theperiplasmic protein CpxP inhibits the Cpx signaling cascade(1). In particular we addressed the question, whether a directinteraction between CpxP andCpxA is involved. ACpxP:CpxAratio of 1:1 was sufficient to inhibit the autophosphorylationactivity of the sensor kinase CpxA up to 51%. In contrast, aCpxP variant that was fused to an N-terminal His tag reducedthe autophosphorylation activity of CpxA only by 20%. Thisfinding supports recent in vivo data showing that the N-termi-nal part of the CpxP protein is critical for its inhibitory activityon the Cpx signaling cascade (20). Interestingly, an inhibitoryeffect on the phosphorylation status of the response regulatorCpxR was not observed indicating that neither the phospho-transfer nor the phosphatase activities of CpxA are influencedby CpxP. Therefore, we suggest that the CpxA-mediated phos-photransfer reaction is much faster than autophosphorylation.In other words, phosphorylated CpxA immediately transfers
the phosphoryl group to CpxR. Concomitantly, we found thatthe balance between the sensor protein CpxA and its cognateregulator CpxR is critical for the signaling cascade. CpxPdecreased significantly the amount of phosphorylated CpxRwhen themolar ratio ofCpxP:CpxA:CpxRwas adjusted to 1:1:4.In summary, we demonstrated that CpxA catalyzes several
reactions: autophosphorylation, the transfer of the phospho-ryl group to CpxR and the dephosphorylation of CpxR�P. Inaddition, by establishing an in vitro signal transduction cas-cade for the whole CpxRAP systemwe were able to show thatthe periplasmic CpxP protein inhibits only the autokinaseactivity of CpxA and does not stimulate its phosphataseactivity as speculated before (16). As the in vitro systemexcludes the involvement of other factors, this is the firstbiochemical indication for direct protein-protein interac-tion between these two proteins. Thus, we suggest that CpxPintervenes at the initial step of signal transduction, keepingthe pathway in a resting state.
Acknowledgments—We thank Erwin Schneider (Humboldt Univer-ity, Berlin) for fruitful discussions,MalK protein,MalK antibody, andfor further manifold support and Jean-Michel Betton (Institut Pas-teur, Paris) for initial help and discussions. We thank Heidi Land-messer for excellent technical assistance. We acknowledge ThomasSchmulling (Free University, Berlin) for Closantel and TCS. We aregrateful to Ed Schwartz (Humboldt University, Berlin) for criticalreading of the manuscript.
REFERENCES1. Raivio, T. L., and Silhavy, T. J. (2001) Annu. Rev. Microbiol. 55, 591–6242. McEwan, J., and Silverman, P. (1980) Proc. Natl. Acad. Sci. U. S. A. 77,
513–5173. Weber, R. F., and Silverman, P. J. (1988) J. Mol. Biol. 203, 467–4764. Dong, J., Iuchi, S., Kwan, H. S., Lu, Z., and Lin, E. C. (1993) Gene (Amst.)
136, 227–2305. Nakayama, S. I., and Watanabe, H. (1995) J. Bacteriol. 180, 3522–35286. Danese, P. N., and Silhavy, T. J. (1998) J. Bacteriol. 180, 831–8397. Mileykovskaya, E., and Dowhan, W. (1997) J. Bacteriol. 179, 1029–10348. Snyder, W. B., Davis, L. J., Danese, P. N., Cosma, C. L., and Silhavy, T. J.
(1995) J. Bacteriol. 177, 4216–42239. Hunke, S., and Betton, J.-M. (2003)Mol. Microbiol. 50, 1579–158910. Jones, C. H., Danese, P. N., Pikner, J. S., Silhavy, T. J., and Hultgren, S. J.
(1997) EMBO J. 21, 6394–640611. Hirakawa, H., Inazumi, Y., Masaki, T., Hirata, T., and Yamaguchi, A.
(2005)Mol. Microbiol. 55, 1113–112612. Jubelin, G., Vianney, A., Beloin, C., Ghigo, J. M., Lazzaroni, J. C., Lejeune,
P., and Dorel, C. (2005) J. Bacteriol. 187, 2038–204913. Raivio, T. L., and Silhavy, T. J. (1999) Curr. Opin. Microbiol. 2, 159–16514. Danes, P. N., Snyder, W. B., Cosma, C. L., Davis, J. B., and Silhavy, T. J.
(1995) Genes Dev. 9, 387–39815. Raivio, T. L., and Silhavy, T. J. (1997) J. Bacteriol. 179, 7724–773316. De Wulf, P., Kwon, O., and Lin, E. C. (1999) J. Bacteriol. 181, 6772–677817. Raivio, T. L., Laird, M.W., Joly, J. C., and Silhavy, T. J. (2000)Mol. Microbiol.
37, 1186–119718. Isaac, D. D., Pinkner, J. S., Hultgren, S. C., and Silhavy, T. J. (2005) Proc.
Natl. Acad. Sci. U. S. A. 102, 17775–1777919. Raivio, T. L., Popkin, D. L., and Silhavy, T. J. (1999) J. Bacteriol. 181,
5263–527220. Buelow, D. R., and Raivio, T. L. (2005) J. Bacteriol. 187, 6622–663021. Stephenson, K., Yamaguchi, Y., and Hoch, J. A. (2000) J. Biol. Chem. 275,
38900–3890422. Blattner, F. R., Plunkett, G. III, Bloch, C. A., Perna, N. T., Burland, V., Riley,
M., Collado-Vides, J., Glasner, J. D., Rode, C. K., Mayhew, G. F., Gregor, J.,
Cpx Pathway Reconstitution
8592 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 12 • MARCH 23, 2007
Davis, N. W., Kirkpatrick, H. A., Goeden, M. A., Rose, D. J., Mau, B., andShao, Y. (1997) Science 277, 1453–1474
23. Jung, K., Hamann, K., and Revermann, A. (2001) J. Biol. Chem. 276,40896–40902
24. Laemmli, U. K. (1970) Nature 227, 680–68525. Goodno, C. C. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 2620–262426. Rigaud, J.-L., Pitard, B., and Levy, D. (1995) Biochim. Biophys. Acta 1231,
223–24627. Schneider, E., Wilken, S., and Schmid, R. (1994) J. Biol. Chem. 269,
20456–2046128. Hlasta, D. J., Demers, J. P., Foleno, B. D., Fraga-Spano, S. A., Guan, J.,
Hilliard, J. J., Macielag, M. J., Ohemeng, K. A., Sheppard, C. M., Sui, Z.,Webb,G. C.,Weidner-Wells,M.A.,Werblood,H., and Barrett, J. F. (1998)Bioorg. Med. Chem. Lett. 8, 1923–1928
29. Foster, J. E., Sheng, Q., McClain, J. R., Bures, M., Nicas, T. I., Henry, K.,Winkler, M. E., and Gilmour, R. (2004)Microbiology 150, 885–896
30. Tanaka, T., Saha, S. K., Tomomori, C., Ishima, R., Liu, D., Tong, K. I., Park,H., Dutta, R., Qin, L., Swindells, M. B., Yamazaki, T., Ono, A. M., Kainosho,M., Inouye, M., and Ikura, M. (1998)Nature 396, 88–92
31. Hilliard, J. J., Krause, H. M., Bernstein, J. I., Fernandez, J. A., Nguyen, V.,
Ohemeng, K. A., and Barrett, J. F. (1995) Adv. Exp. Med. Biol. 390, 59–6932. Macara, I. G. (1980) Trends Biochem. Sci. 5, 92–9433. Hunke, S., Drose, S., and Schneider, E. (1995) Biochem. Biophys. Res. Com-
mun. 216, 589–59434. Sharma, S., and Davidson, A. L. (2000) J. Bacteriol. 182, 6570–657635. Jung, K., Tjaden, B., and Altendorf, K. (1997) J. Biol. Chem. 272,
10847–1085236. Yamamoto, K., Hirao, K., Oshima, T., Aiba, H., Utsumi, R., and Ishihama,
A. (2005) J. Biol. Chem. 280, 1448–145637. Barrett, J. F., Goldschmidt, R. M., Lawrence, L. E., Foleno, B., Chen, R.,
Demers, J. P., Johnson, S., Kanojia, R., Fernandez, J., Bernstein, J., Li-cata, L., Donetz, A., Huang, S., Hlasta, D. J., Macielag, M. J., Ohemeng,K., Frechette, R., Frosco, M. B., Klaubert, D. H., Whiteley, J. M., Wang,L., and Hoch, J. A. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 5317–5322
38. Roychoudhury, S., Zielinski, N. A., Ninfa, A. J., Allen, N. E., Jungheim,L. N., Nicas, T. I., and Cakarbarty, A. M. (1993) Proc. Natl. Acad. Sci.U. S. A. 90, 965–969
39. Ulijasz, A. T., and Weisblum, B. (1999) J. Bacteriol. 181, 627–63140. Forst, S., Comeau, D., Norioka, S., and Inouye, M. (1987) J. Biol. Chem.
262, 16433–16438
Cpx Pathway Reconstitution
MARCH 23, 2007 • VOLUME 282 • NUMBER 12 JOURNAL OF BIOLOGICAL CHEMISTRY 8593
