Appearance of elongation factor Tu in the outer membrane of sucrose-dependent spectinomycin-resistant mutants of Escherichia coli

When sucrose-dependent spectinomycin-resistant (Sucd-Spcr) mutants of Escherichia coli were grown in the absence of sucrose, a new protein appeared in the membrane fraction insoluble in Triton X-100. The protein had a hydrophobic nature. However, unlike other outer membrane proteins the new protein was extracted with sodium dodecyl sarcosinate. The new protein was found to be identical with elongation factor Tu (EF-Tu), as judged from the electrophoretic mobility in three different gel systems, coprecipitation with the antiserum against EF-Tu, the profiles of peptide fragments produced with three different proteases and analyses of N-terminal and C-terminal amino acids. This membrane EF-Tu accounted for 5-10% of total cell EF-Tu. When spheroplasts were pretreated with trypsin, EF-Tu in the outer membrane disappeared. Incubation of cytosol EF-Tu with the outer membrane did not result in the binding of EF-Tu to the membrane. These results indicate that the appearance of EF-Tu in the outer membrane is not due to artificial binding during membrane preparation. It is suggested that the ribosomal alteration resulted in dislocation of the cytosol protein into the outer membrane.     

The surface-associated elongation factor Tu is concealed for antibody binding on viable pneumococci and meningococci

Proteome analyses revealed that elongation factor-Tu (EF-Tu) is associated with cytoplasmic membranes of Gram-positive bacteria and outer membranes of Gram-negative bacteria. It is still debatable whether EF-Tu is located on the external side or the internal side of the membranes. Here, we have generated two new monoclonal antibodies (mAbs) and polyclonal rabbit antibodies against pneumococcal EF-Tu. These antibodies were used to investigate the amount of surface-exposed EF-Tu on viable bacteria using a flow cytometric analysis. The control antibodies recognizing the pneumococcal surface protein A and phosphorylcholine showed a significant binding to viable pneumococci. In contrast, anti-EF-Tu antibodies did not recognize pneumococcal EF-Tu. However, heat killing of pneumococci lacking capsular polysaccharides resulted in specific antibody binding to EF-Tu and, moreover, increased the exposure of recognized phosphorylcholine epitopes. Similarly, our EF-Tu-specific antibodies did not recognize EF-Tu of viable Neisseria meningitidis. However, pretreatment of meningococci with ethanol resulted in specific antibody binding to EF-Tu on outer membranes. Importantly, these treatments did not destroy the membrane integrity as analysed with control mAbs directed against cytoplasmic proteins. In conclusion, our flow cytrometric assays emphasize the importance of using viable bacteria and not heat-killed or ethanol-treated bacteria for surface-localization experiments of proteins, because these treatments modulate the cytoplasmic and outer membranes of bacteria and the binding results may not reflect the situation under physiological conditions.     

Electron spin resonance studies of lipid fluidity changes in membranes of an uncoupler-resistant mutant of Escherichia coli

The fluidity of the lipids in membrane preparations from a mutant of Escherichia coli resistant to the uncoupler CCCP, grown at different temperatures with and without CCCP, was examined by electron spin resonance using the spin probe 5-doxyl stearic acid. The fluidity of the membrane lipids at the growth temperature, as estimated using electron spin resonance, was less in cells grown at lower temperatures. Precise homeoviscous adaptation was not observed. Growth in the presence of CCCP resulted in a decrease in membrane lipid fluidity, particularly in the inner (cytoplasmic) membrane. There was no change in the proportion of phosphatidylethanolamine, phosphatidylglycerol and cardiolipin in the cell envelope. However, there was an increase in the proportion of unsaturated fatty acids in membranes from cells grown with uncoupler. This was reflected in the increased fluidity of the lipids extracted from these membranes. This result is contrary to that expected from measurements of the fluidity of the lipid in these membranes. The decreased fluidity of the lipid in these membranes may be a consequence of the observed increase in the ratio of protein to phospholipid.     

Lethal mutations in the structural gene of an outer membrane protein (OmpA) of Escherichia coli K12

Summary The gene ompA encodes a major outer membrane protein of Escherichia coli. Localized mutagenesis of the part of the gene corresponding to the 21-residue signal sequence and the first 45 residues of the protein resulted in alterations which caused cell lysis when expressed. DNA sequence analyses revealed that in one mutant type the last CO₂H-terminal residue of the signal sequence, alanine, was replaced by valine. The proteolytic removal of the signal peptide was much delayed and most of the unprocessed precursor protein was fractioned with the outer membrane. However, this precursor was completely soluble in sodium lauryl sarcosinate which does not solubilize the OmpA protein or fragments thereof present in the outer membrane. Synthesis of the mutant protein did not inhibit processing of the OmpA or OmpF proteins. In the other mutant type, multiple mutational alterations had occurred leading to four amino acid substitutions in the signal sequence and two affecting the first two residues of the mature protein. A reduced rate of processing could not be clearly demonstrated. Membrane fractionation suggested that small amounts of this precursor were associated with the plasma membrane but synthesis of this mutant protein also did not inhibit processing of the wild-type OmpA or OmpF proteins. Several lines of evidence left no doubt that the mature, mutant protein is stably incorporated into the outer membrane. It is suggested that the presence, in the outer membrane, of the mutant precursor protein in the former case, or of the mutant protein in the latter case perturbs the membrane architecture enough to cause cell death.     

Alterations in the Outer Membrane of the Cell Envelope of Heptose-Deficient Mutants of Escherichia coli

The composition of the cell envelope of a heptose-deficient lipopolysaccharide mutant of Escherichia coli, GR467, was studied after fractionation into its outer and cytoplasmic membrane components by means of sucrose density gradient centrifugation. The outer membrane of GR467 had a lower density than that of its parent strain, CR34. Analysis of the fractionated membranes of GR467 indicated that the phospholipid-to-protein ratio had increased 2.4-fold in the outer membrane. The ratio in the mutant cytoplasmic membrane was also increased, although to a lesser extent. By employing a third parameter, the lipid A content of the outer membrane, it was found that the observed phospholipid-to-protein change in the outer membrane was due predominantly to a decrease in the relative amount of protein. This decrease in protein was particularly significant, since it was concomitant with a 68% decrease in the lipid A recovered in the outer membrane of GR467 relative to the lipid A recovered in the outer membrane of CR34. Similar findings were observed in a second heptose-deficient mutant of E. coli, RC-59. The apparent protein deficiency in GR467 was further studied by subjecting solubilized envelope proteins to sodium dodecyl sulfate-polyacrylamide gel electrophoresis. It was found that major envelope proteins which were localized in the outer membrane were greatly diminished in GR467. Two revertants of GR467 with the wild-type amounts of heptose had wild-type relative levels of protein in their outer membranes. A partial heptose revertant had a relative level of protein in its outer membrane between those of the mutant and wild type.     

Effect of uncoupler on assembly pathway for pigment-binding protein of bacterial photosynthetic membranes.

The uncoupler carbonylcyanide m-chlorophenylhydrazone (CCCP) was used to investigate membrane protein assembly in the phototrophic bacterium Rhodobacter capsulatus. As found for Escherichia coli (T. Date, G. Zwizinsky, S. Ludmerer, and W. Wickner, Proc. Natl. Acad. Sci. 77:827-831, 1980) and mitochondrial proteins (N. Nelson and G. Schatz, Proc. Natl. Acad. Sci. USA 76:4365-4369, 1979), assembly across the bacterial photosynthetic membranes was sensitive to CCCP. At uncoupler concentrations which were sufficient to block the export of the periplasmic cytochrome c2 and an outer membrane protein, the integration of pigment-binding protein into the photosynthetic apparatus was abolished. The unassembled protein was detected on the inner surface of the intracytoplasmic membrane. After inactivation of CCCP, accumulated protein continued insertion into the membrane. The data suggest that after binding to the cytoplasmic face of the membrane, translocation of protein into a transmembrane orientation takes place, which is a prerequisite for the formation of a functional pigment-protein complex.     

The in vitro insertion of monoamine oxidase B into mitochondrial outer membranes

Bovine monoamine oxidase (MAO) B has been synthesized in vitro using a reticulocyte lysate translation system directed by bovine liver poly(A)+ RNA. The newly synthesized enzyme apparently lacks a cleavable N-terminal extension, but MAO B is readily incorporated into mitochondria or isolated mitochondrial outer membranes prepared from rat liver. ATP is not required for the binding of the newly synthesized enzyme to the outer membranes, but is necessary for the insertion of MAO B into these membrane vesicles. The ATP is not required to generate a mitochondrial membrane potential as assembly occurs under conditions that preclude either the formation or the maintenance of the potential. MAO B will bind to but not become incorporated into outer membrane vesicles which have been treated with trypsin, suggesting that the insertion of MAO B also depends on protein factors present on the outer membranes.     

Base 2661 in Escherichia coli 23S rRNA influences the binding of elongation factor Tu during protein synthesis in vivo.

The binding of the EF-Tu.GTP.aminoacyl-tRNA ternary complex (EF, elongation factor) to the ribosome is known to be strengthened by a 2661G-to-C mutation in 23S ribosomal RNA, whereas the binding to normal ribosomes is weakened if the factor is in an appropriate mutant form (Aa). In this report we describe the mutual effects by the 2661C alteration in 23S rRNA and EF-Tu(Aa) on bacterial viability and translation efficiency in strains with normal or mutationally altered ribosomes. The rrnB(2661C) allele on a multicopy plasmid was introduced by transformation into Escherichia coli K-12 strains, harbouring either the wild-type or the mutant gene (tufA) for EF-Tu as well as normal or mutant ribosomal protein S12 (rpsL). Together with wild-type EF-Tu, the 2661C mutant ribosomes decreased the translation elongation rate in a rpsL+ strain or a non-restrictive rpsL224 strain. This reduction was not seen in strains which harbored EF-Tu(Aa) instead of EF-Tu(As) (As, wild-type form). Nonsense codon suppression by tyrT(Su3) suppressor tRNA was reduced by 2661C in a rpsL224 strain in the presence of EF-Tu(As) but not in the presence of EF-Tu(Aa). The lethal effect obtained by the combination of 2661C and a restrictive ribosomal protein S12 mutation (rpsL282) disappeared if EF-Tu(As) was replaced by EF-Tu(Aa) in the strain. In such a viable strain, 2661C had no effect on either the translation elongation rate or nonsense codon suppression. Our data suggest that the G base at position 2661 in 23S rRNA is important for binding of EF-Tu during protein synthesis in vivo. The interaction between this base and EF-Tu is strongly influenced by the structure of ribosomal protein S12.     

Conformation of the C-terminal domain of the pro-apoptotic protein Bax and mutants and its interaction with membranes

The C-terminal domain of the pro-apoptotic protein Bax is a hydrophobic stretch which, it has been predicted, anchors this protein to the outer mitochondrial membrane when apoptosis is induced in the cell. A 21mer peptide imitating this domain has been synthesized together with two mutants, one with a S184 substituted by K and the other with the S184 deleted. When their structures were studied by infrared spectroscopy, it was seen that the three peptides formed aggregates both in solution and within lipid membranes, and that the peptide changed its secondary structure as a consequence of these two mutations. It was also observed that the wild-type peptide and the two mutants became membrane-integral molecules and changed their conformation when they were incorporated into model membranes with the same composition as the outer mitochondrial membrane. With the peptides incorporated in the membranes the location of W188 was studied by fluorescence quenching using the water soluble quencher acrylamide and different doxyl-PC located in the membrane, this residue being found at different membrane depths in each of the three peptides. The fact that the three peptides were able to perturb the motion of the fluorescent probe diphenylhexatriene confirmed their insertion in the membrane. However, whereas the wild type and the DeltaS184 mutant peptides were very efficient in releasing encapsulated carboxyfluorescein from liposomes, the mutant S184K was less efficient. Taken together, these results showed that the mutation tested changed the conformation of the C-terminal domain of Bax and the positions that they adopted when inserted in membranes, confirming the importance of S184 determining the conformation of this domain. At the same time, these results confirmed that the C-terminal domain of Bax participates in disrupting the barrier properties of biomembranes.     

Identification of a new pore in the mitochondrial outer membrane of a porin-deficient yeast mutant

Reconstitution experiments were performed on lipid bilayer membranes in the presence of detergent-solubilized mitochondrial outer membranes of a porin-free yeast mutant and of its parent strain. The addition of the detergent-solubilized material resulted in a strong increase in the membrane conductance which was not observed if only the detergent was added to the aqueous phase. Surprisingly, the membrane conductance induced by the detergent extracts of the mutant membrane was only a factor of 20 less than that caused by the outer membrane of the parent strain under otherwise identical conditions. Single-channel recordings of lipid bilayer membranes in the presence of mitochondrial outer membranes of the yeast mutant suggested the presence of a transient pore. The reconstituted pores had a single-channel conductance of 0.21 nS in 0.1 M KCl and the characteristics of general diffusion pores with an estimated effective diameter of 1.2 nm. The pores present in the mitochondrial outer membranes of the yeast mutant shared some similarities with the pores formed by mitochondrial and bacterial porins although their effective diameter is much smaller than those of the 'normal' mitochondrial porins which have a single-channel conductance of about 0.4 nS in 0.1 M KCl, corresponding to an effective diameter of 1.7 nm. Zero-current membrane-potential measurements suggested that the second mitochondrial porin is slightly cation-selective. Its possible role in the metabolism of mitochondria is discussed.     

The protonophore resistance of Bacillus megaterium is correlated with elevated ratios of saturated to unsaturated fatty acids in membrane phospholipids

Growth of the protonophore-resistant strain of Bacillus megaterium, strain C8, in the presence of oleic acid markedly reduced its resistance to low concentrations of carbonylcyanide m-chlorophenylhydrazone (CCCP). Growth of the CCCP-sensitive wild-type strain in the presence of stearic acid increased the resistance of that strain to growth inhibition by protonophore. Studies of the membrane lipids indicated that in the absence of additions to the medium, membranes from C8 contained greatly reduced levels of monounsaturated fatty acids relative to the wild type; wild-type levels were restored by growth of C8 in the presence of oleic acid, concomitant with the loss of resistance. Conversely, growth of the wild type on stearic acid increased the ratio of saturated/unsaturated fatty acids in the membrane, concomitant with a modest increase in the resistance of the wild-type strain to CCCP. The exogenous oleic acid was preferentially incorporated into phosphatidylethanolamine, diphosphatidylglycerol, and 1,2-diacylglycerol, whereas stearic acid was incorporated preferentially into phosphatidylglycerol, and into the small component of free fatty acids. Depending upon the growth conditions, changes in membrane lipid-to-membrane protein ratio and in the ratios of polar lipid components were observed, but none of those changes correlated as did the changes in saturated fatty-acid-to-unsaturated fatty-acid ratio with protonophore resistance. This latter correlation was further suggested by experiments in which the protonophore resistance of wild type B. megaterium was shown to increase with increasing growth temperature without any temperature-dependent loss of protonophore efficacy. The experiments here support the hypothesis developed from work with Bacillus subtilis that changes in the fatty acid composition of the membrane phospholipids affect energy coupling, and make it clear that simple increases or decreases in the hydrolytic activity of ATPase in the uncoupler-resistant mutants of bacilli are not correlated with resistance in some direct way.     

Localization of a 64-kDa phosphoprotein in the lumen between the outer and inner envelopes of pea chloroplasts

The identification and localization of a marker protein for the intermembrane space between the outer and inner chloroplast envelopes is described. This 64-kDa protein is very rapidly labeled by [gamma-32P]ATP at very low (30 nM) ATP concentrations and the phosphoryl group exhibits a high turnover rate. It was possible to establish the presence of the 64-kDa protein in this plastid compartment by using different chloroplast envelope separation and isolation techniques. In addition comparison of labeling kinetics by intact and hypotonically lysed pea chloroplasts support the localization of the 64-kDa protein in the intermembrane space. The 64-kDa protein was present and could be labeled in mixed envelope membranes isolated from hypotonically lysed plastids. Mixed envelope membranes incorporated high amounts of 32P from [gamma-32P]ATP into the 64-kDa protein, whereas separated outer and inner envelope membranes did not show significant phosphorylation of this protein. Water/Triton X-114 phase partitioning demonstrated that the 64-kDa protein is a hydrophilic polypeptide. These findings suggest that the 64-kDa protein is a soluble protein trapped in the space between the inner and outer envelope membranes. After sonication of mixed envelope membranes, the 64-kDa protein was no longer present in the membrane fraction, but could be found in the supernatant after a 110,000 x g centrifugation.     

Purification of a hexokinase-binding protein from the outer mitochondrial membrane.

Brain hexokinase (ATP:D-hexose-6-phosphotransferase, EC 2.7.1.1) binds selectively to the outer membrane of rat liver mitochondria but not to inner mitochondrial or microsomal membranes nor to the plasma membrane of human erythrocytes. A protein having subunit molecular weight of 31,000, determined by sodium dodecyl sulfate-gel electrophoresis, has been highly purified from the outer mitochondrial membrane by repetitive solubilization with octyl-beta-D-glucopyranoside followed by reconstitution into membranous vesicles when the detergent is removed by dialysis. When incorporated into lipid vesicles, the protein confers the ability to bind brain hexokinase in a Glc-6-P-sensitive manner as is seen with the intact outer mitochondrial membrane. Hexokinase binding ability and the 31,000 subunit molecular weight protein co-sediment during sucrose density gradient centrifugation. Both hexokinase binding ability and the 31,000 subunit molecular weight protein are resistant to protease treatment of the intact outer mitochondrial membrane while other membrane proteins are extensively degraded. It is concluded that this protein, designated the hexokinase-binding protein (HBP), is an integral membrane protein responsible for the selective binding of hexokinase by the outer mitochondrial membrane.     

Localization and assembly into the Escherichia coli envelope of a protein required for entry of colicin A.

Mutations in tolQ, previously designated fii, render cells tolerant to high concentrations of colicin A. In addition, a short deletion in the amino-terminal region of colicin A (amino acid residues 16 to 29) prevents its lethal action, although this protein can still bind the receptor and forms channels in planar lipid bilayers in vitro. These defects in translocation across the outer membrane in the tolQ cells or the colicin A mutant cannot be bypassed by osmotic shock. The TolQ protein, which is constitutively expressed at a low level, was studied in recombinant plasmid constructs allowing the expression of various TolQ fusion proteins under the control of the inducible caa promoter. The TolQ protein was thus "tagged" with an epitope from the colicin A protein for which a monoclonal antibody is available. A fusion protein containing the entire TolQ protein plus the 30 N-terminal residues of colicin A was shown to complement the tolQ mutation. Pulse-chase labeling followed by gradient fractionation indicated that the bulk of the overproduced fusion protein was rapidly incorporated into the inner membrane, with small amounts localized to regions corresponding to the attachment sites between inner and outer membranes and to the outer membrane itself. However, most of the protein was rapidly degraded, leaving only that localized to the attachment sites and the outer membrane remaining at very late times of chase.     

TLR4-dependent activation of inflammatory cytokine response in macrophages by Francisella elongation factor Tu

The bacterial determinants of pulmonary Francisella induced inflammatory responses and their interaction with host components are not clearly defined. In this study, proteomic and immunoblot analyses showed presence of a cytoplasmic protein elongation factor Tu (EF-Tu) in the membrane fractions of virulent Francisella novicida, LVS and SchuS4, but not in an attenuated F. novicida mutant. EF-Tu was immunodominant in mice vaccinated and protected from virulent F. novicida. Moreover, recombinant EF-Tu induced macrophages to produce inflammatory cytokines in a TLR4 dependent manner. This study shows immune stimulatory properties of a cytoplasmic protein EF-Tu expressed on the membrane of virulent Francisella strains.     

Specific alterations of the EF-Tu polypeptide chain considered in the light of its three-dimensional structure

Specific alterations of the elongation factor Tu (EF-Tu) polypeptide chain have been identified in a number of mutant species of this elongation factor. In two species, Ala-375, located on domain II, was found by amino acid analysis to be replaced by Thr and Val, respectively. These replacements substantially lower the affinity of EF-Tu.GDP for the antibiotic kirromycin. Since kirromycin can be cross-linked to Lys-357, also located on domain II but structurally very far from Ala-375, these data suggest that the replacements alter the relative position of domains I and II. The Ala-375 replacements also lower the dissociation rates of the binary complexes EF-Tu.GTP and the binding constants for EF-Tu.GTP and Phe-tRNA. It is conceivable that these effects are also mediated by movements of domains I and II relative to each other. Replacement of Gly-222 by Asp has been found in another mutant by DNA sequence analysis of the cloned tufB gene, coding for this mutant EF-Tu. Gly-222 is part of a structural domain, characteristic for a variety of nucleotide binding enzymes. Its replacement by Asp does not abolish the ability of EF-Tu to sustain protein synthesis. It increases the dissociation rate of EF-Tu.GTP by approximately 30%. In the presence of kirromycin this mutant species of EF-Tu.GDP does not bind to the ribosome, in contrast to its wild-type counterpart. A possible explanation is now open for experimental verification.     

Study of the transverse diffusion of spin-labeled phospholipids in biological membranes. II. Inner mitochondrial membrane of rat liver: use of phosphatidylcholine exchange protein.

Spin-labeled phosphatidylcholine was incorporated into the membrane of isolated "inner membrane+matrix" particles of rat liver mitochondria by incubation with sonicated spin-labeled phosphatidylcholine vesicles at 22 degrees C. When the spin label was on the acyl chain the incorporation of phosphatidylcholine into the membrane was stimulated by the presence of the phosphatidylcholine exchange protein extracted from rat or beef liver. On the other hand no stimulation was observed when the nitroxide was on the polar head-group. When spin-labeled phosphatidycholine was incorporated into the mitochondrial membrane in the absence of phosphatidylcholine exchange protein, ascorbate treatment at 0 degrees C reduced the EPR signal of the spin-labeled membranes by approximately 50%, indicating that fusion incorporates molecules equally on both sides of the membrane. On the other hand when spin-labeled phosphatidylcholine was incorporated in the presence of the exchange protein most of the EPR signal could be destroyed by the ascorbate treatment at 0 degrees C, indicating that the spin-labeled phosphatidylcholine had been selectively incorporated in the outer layer of the membrane. Finally when the label is on the polar head-group the inner content of mitochondria reduces the label facing the matrix, thus creating again an anisotropy of the labeling. The anisotropic distribution of spin-labeled phosphatidylcholine in the mitochondrial membrane was found to be stable at 25 degrees C for more than 2 h. It is therefore concluded that the rate of outside-inside and inside-outside transitions are extremely slow (half-life greater than 24 h).     

The assembly pathway of outer membrane protein PhoE of Escherichia coli.

The assembly of the wild-type and several mutant forms of the trimeric outer membrane porin PhoE of Escherichia coli was investigated in vitro and in vivo. In in vivo pulse-chase experiments, approximately half of the wild-type PhoE molecules assembled within the 30-s pulse in the native conformation in the cell envelope. The other half of the molecules followed slower kinetics, and three intermediates in this multistep assembly process were detected: a soluble trypsin-sensitive monomer, a trypsin-sensitive monomeric form that was loosely associated with the cell envelope and a metastable trimer, which was integrated into the membranes and converted to the stable trimeric configuration within minutes. The metastable trimers disassembled during sample preparation for standard SDS/PAGE into folded monomers. In vitro, the isolated PhoE protein could efficiently be folded in the presence of N,N-dimethyldodecylamine-N-oxide (LDAO). A mutant PhoE protein, DeltaF330, which lacks the C-terminal phenylalanine residue, mainly followed the slower kinetic pathway observed in vivo, resulting in increased amounts of the various assembly intermediates. It appears that the DeltaF330 mutant protein is intrinsically able to fold, because it was able to fold in vitro with LDAO with similar efficiencies as the wild-type protein. Therefore, we propose that the conserved C-terminal Phe is (part of) a sorting signal, directing the protein efficiently to the outer membrane. Furthermore, we analysed a mutant protein with a hydrophilic residue introduced at the hydrophobic side of one of the membrane-spanning amphipathic beta strands. The assembly of this mutant protein was not affected in vivo or in vitro in the presence of LDAO. However, it was not able to form folded monomers in a previously established in vitro folding system, which requires the presence of lipopolysaccharides and Triton. Hence, a folded monomer might not be a true assembly intermediate of PhoE in vivo.     

Carboxy-terminal phenylalanine is essential for the correct assembly of a bacterial outer membrane protein

Bacterial outer membrane proteins are supposed to span the membrane repeatedly, mostly in the form of amphipathic beta-sheets. The last ten C-terminal amino acid residues of PhoE protein are supposed to form such a membrane-spanning segment. Deletion of this segment completely prevents incorporation into the outer membrane. Comparison of the last ten amino acid residues of other outer membrane proteins from different Gram-negative bacteria revealed the presence of a potential amphipathic beta-sheet with hydrophobic residues at positions 1 (Phe), 3 (preferentially Tyr), 5, 7 and 9 from the C terminus, in the vast majority of these proteins. Since such sequences were not detected at the C termini of periplasmic proteins, it appears to be possible to discriminate between the majority of outer membrane proteins and periplasmic proteins on the basis of sequence data. The highly conserved phenylalanine at the C termini of outer membrane proteins suggests an important function for this amino acid in assembly into the outer membrane. Site-directed mutagenesis was applied to study the role of the C-terminal Phe in PhoE protein assembly. All mutant proteins were correctly incorporated into the outer membrane to some extent, but the efficiency of the process was severely affected. It appears that both the hydrophobicity and the aromatic nature of Phe are of importance.     

Verification of the protein in the outer membrane of Bdellovibrio bacteriovorus as the OmpF protein of its Escherichia coli prey.

Two research groups showed that several Bdellovibrio strains incorporated into their outer membranes intact OmpF porin proteins derived from their Escherichia coli prey. These results could not be reproduced by another group using Bdellovibrio bacteriovorus 109J. They showed that a major protein appearing in the Bdellovibrio Triton X-100-insoluble outer membrane was coded for by the bdellovibrios. We reconciled these results by examining the strain used by this group and by reviving a freeze-dried culture of strain 109J which had been stored for almost 9 years. B. bacteriovorus 109J failed to acquire substantial amounts of the OmpF protein from E. coli ML35, and a protein coded for by the bdellovibrios was expressed in its place. However, B. bacteriovorus 109J incorporated the OmpF protein from rough K-12 strains of E. coli, and the revived 9-year-old culture of B. bacteriovorus 109J incorporated more of the OmpF protein from the smooth E. coli ML35 than did its contemporary counterpart. The protein isolated from the outer membrane of the bdellovibrios was identified as the OmpF protein of E. coli by its protease peptide profile on sodium dodecyl sulfate-polyacrylamide gel electrophoresis and by Western blot analysis. This confirmed that bdellovibrios relocalize outer membrane proteins from their prey, but relocalization may be an unstable trait which can be influenced by the prey.