Use of an Lpp-OmpA fusion vehicle for bacterial surface display.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to C F Earhart.
Explore the source record for details and available documents.
The cytosolic proteins EntE, EntF, and EntB/G, which are Escherichia coli enzymes necessary for the final stage of enterobactin synthesis, are released by osmotic shock. Here, consistent with the idea that cytoplasmic proteins found in shockates have an affinity for membranes, a small fraction of each was found in membrane preparations. Two procedures demonstrated that the enzymes were enriched in a minor membrane fraction of buoyant density intermediate between that of cytoplasmic and outer membranes, providing indirect support for the notion that these proteins have a role in enterobactin excretion as well as synthesis.
The Lpp'OmpA(46-159) hybrid protein can serve as an efficient targeting vehicle for localizing a variety of procaryotic and eucaryotic soluble proteins onto the E. coli surface, thus providing a system for several possible biotechnology applications. Here we show that fusion between Lpp'OmpA(46-159) and bacterial alkaline phosphatase (PhoA), a normally periplasmic dimeric enzyme, are also targeted to the outer membrane. However, protease accessibility experiments and immunoelectron microscopy revealed that, unlike other periplasmic proteins, the PhoA domain of these fusions is not exposed on the cell surface in cells having an intact outer membrane. Conditions that affect the formation of disulfide bonds and the folding of the PhoA domain in the periplasm not only did not facilitate targeting to the cell surface but led to lethality when the fusion was expressed from a high-copy-number plasmid. Furthermore, E. coli expressing the Lpp'OmpA(46-159)-PhoA fusion exhibited strain- and temperature-dependent alterations in outer-membrane permeability. Our results are consistent with previous studies with other vehicles indicating that PhoA is not displayed on the surface when fused to cell-surface expression vectors. Presumably, the enzyme rapidly assumes a tightly folded dimeric conformation that cannot be transported across the outer membrane. The large size and quaternary structure of PhoA may define a limitation of the Lpp'OmpA(46-159) fusion system for the display of periplasmic proteins on the cell surface. Alkaline phosphatase is a unique protein among a group of five periplasmic proteins (beta-lactamase, alkaline phosphatase, Cex cellulase Cex cellulose-binding domain, and a single-chain Fv antibody fragment), which have been tested as passengers for the Lpp'OmpA(46-159) expression system to date, since it was the only protein not displayed on the surface.
Bacterial cell-surface exposure of foreign peptides and soluble proteins has been achieved recently by employing a fusion protein methodology. An Lpp'-OmpA(46-159)-Bla fusion protein has been shown previously to display the normally periplasmic enzyme beta-lactamase (Bla) on the cell surface of the Gram-negative bacterium Escherichia coli. Here, we have investigated the role of the OmpA domain of the tripartite fusion protein in the surface display of the passenger domain (Bla) and have characterized the effects of the fusion proteins on the integrity and permeability of the outer membrane. We show that in addition to OmpA(46-159), a second OmpA segment, consisting of amino acids 46-66, can also mediate the display of Bla on the cell surface. Other OmpA domains of various lengths (amino acids 46-84, 46-109, 46-128, 46-141 and 46-145) either anchored the Bla domain on the periplasmic face of the outer membrane or caused a major disruption of the outer membrane, allowing the penetration of antibodies into the cell. Detergent and antibiotic sensitivity and periplasmic leakage assays showed that changes in the permeability of the outer membrane are an unavoidable consequence of displaying a large periplasmic protein on the surface of E. coli. This is the first systematic report on the effects that cell surface engineering may have on the integrity and permeability properties of bacterial outer membranes.
Most high-affinity systems for iron uptake in Gram-negative bacteria are thought to employ periplasmic-binding-protein-dependent transport. In Escherichia coli, FepB is a periplasmic protein required for uptake of iron complexed to its endogenously-synthesized siderophore enterobactin (Ent). Direct evidence that ferrienterobactin (FeEnt) binds to FepB is lacking because high background binding by FeEnt prevents use of the usual binding protein assays. Here the membrane localization vehicle LppOmpA [Francisco, J.A., Earhart, C.F. & Georgiou, G. (1992). Proc Natl Acad Sci USA 89, 2713-2717] was employed to place FepB in the E. coli outer membrane. Plasmid pTX700 was constructed and shown to encode, under lac operator control, the 'tribrid' protein LppOmpAFepB; the carboxy-terminal FepB portion lacks at most two amino acids of mature FepB. After short induction periods, most of the tribrid was in the outer membrane. A number of LppOmpAFepB species could be detected; some were degradation products and some may be related to the multiplicity of FepB forms previously observed in minicells and maxicells. Outer membrane harbouring the tribrid and lacking FepA, the normal outer membrane receptor for FeEnt, bound approximately four times more FeEnt than outer membrane from uninduced cells, from cells lacking pTX700 and from cells expressing only an LppOmpA 'dibrid'. Similarly, whole UT5600(fepA)/pTX700 cells induced for tribrid synthesis bound FeEnt and this binding was not affected by energy poisons. The results demonstrated that FepB can bind FeEnt, thereby definitely placing FeEnt transport in the periplasmic permease category of transport systems, and that the LppOmpA localization vehicle can be used with periplasmic binding proteins.
The outer membrane of Gram-negative bacteria presents an effective barrier that restricts the release of proteins from the cell. Virtually all extracellular proteins of Gram-negative bacteria are exported by specialized systems requiring the action of several gene products. We have constructed a tripartite fusion consisting of (i) the signal sequence and first nine N-terminal amino acids of the mature major Escherichia coli lipoprotein, (ii) amino acids 46-159 of the outer membrane protein OmpA, and (iii) the complete mature beta-lactamase (EC 3.5.2.6) sequence. This protein had an enzymatically active beta-lactamase and was found predominantly in the outer membrane. Immunofluorescence microscopy, the accessibility of the fusion protein to externally added proteases, and the rates of hydrolysis of nitrocefin and penicillin G by whole cells demonstrated that a substantial fraction (20-30%) of the beta-lactamase domain of the fusion protein was exposed on the external surface of E. coli. In cells grown at 24 degrees C the localization of beta-lactamase on the cell surface was almost quantitative (greater than 80% of the enzymatically active protein was exposed to the extracellular fluid) as determined by nitrocefin and penicillin G hydrolysis and trypsin accessibility. These results demonstrated that a soluble protein, beta-lactamase, can be transported through--and become anchored on--the outer membrane by fusion to the proper targeting and localization signals.
In Escherichia coli, iron assimilation by means of the siderophore enterobactin requires two hydrophobic cytoplasmic membrane proteins, FepD and FepG, which are essential components of a binding-protein-dependent transport system. Such components are typically difficult to detect. Here we report observation of the fepD and fepG gene products in polyacrylamide gels; they appeared as diffuse bands at positions consistent with smaller sizes than those predicted by sequence analysis. Translational coupling was suggested by the lack of a detectable product from the fepG message in the absence of translation of the upstream fepD message. The orientation of FepD/FepG in the membrane was predicted based on their similarities in sequence and hydrophobicity to FhuB.
Transposon mutagenesis and plasmid complementation studies have identified two genes, fepD and fepG, which are essential for ferrienterobactin transport in Escherichia coli. These genes mapped in the enterobactin gene cluster and genetic evidence indicated that they are transcribed as part of an operon (fepD, fepG, fepC). The nucleotide sequence of fepD was determine; it could encode a hydrophobic 33.8 kDa protein with sequence homologies to other iron and vitamin B12 transport proteins. Also identified, between fepD and fepB, was an open reading frame (ORF43) with no detectable function; its 43 kDa protein product (P43) was seen on polyacrylamide gels. The fepD-C operon and ORF43 were divergently transcribed from a 110bp region containing a binding site for the repressor protein Fur.
The last steps in the biosynthesis of the Escherichia coli siderophore enterobactin (Ent) are carried out by Ent synthetase, a multienzyme complex believed to be composed of the entD, -E, -F, and -G products (EntD to -G). However, sequencing data showed that there is no separate entG gene and, unlike EntD to -F, no distinct EntG polypeptide has been identified. In this study, genetic, biochemical, and immunological approaches were used to study the anomalies associated with EntG activity. Two plasmids, pJS43 and pJS100, were isolated that had mutations resulting in truncated EntB proteins; both had the phenotype EntB+ EntG-. PJS43 had a Tn5 inserted 198 bp from the entB termination codon, and pJS100 had the last 25 codons of entB deleted. Plasmids isolated with Tn5 insertions in the 5' half of entB had the phenotype EntB- EntG+. These latter Tn5 mutations were EntB- EntG- when moved to the bacterial chromosome. Polyclonal antiserum was prepared and shown to react only with intact EntB in Western immunoblots. Addition of anti-EntB antiserum to Ent synthetase assays resulted in complete inhibition of enzyme activity, whereas preimmune serum had no effect. Lastly, AN462, the type strain for entG which was derived by Mu insertion and which has the phenotype EntB-G-A-, was characterized. Southern blot data showed a Mu insertion, presumably with polar effects, in the vicinity of the 5' end of entB. In summary, EntG activity was found to be encoded by the entB 3' terminus. The evidence, while not rigorously eliminating the possibility that a separate EntG polypeptide exists, strongly supports the idea that EntB is a bifunctional protein.
The Escherichia coli entE gene encodes a polypeptide necessary in the latter stages of biosynthesis of the siderophore enterobactin. The entE gene and adjacent DNA were sequenced. The predicted EntE polypeptide consists of 536 amino acids and has a Mr of 58,299 and a net charge of -7.33. Genetic evidence combined with this and previous sequencing data indicate that the genes entCEB(G)A are transcribed as unit from a promoter upstream of entC.
The Escherichia coli entD gene encodes a product necessary for the synthesis of the iron-chelating and transport molecule enterobactin (Ent); cells harbouring entD mutations fail to grow in iron-deficient environments. For unknown reasons, it has not been possible to identify the entD product. The nucleotide sequence of the entD region has now been determined. An open reading frame extending in the same direction as the adjacent fepA gene and capable of encoding an approximately 24 kDa polypeptide was found; it contained a high percentage of rare codons and two possible translational start sites. Complementation data suggested that EntD proteins truncated at the carboxy terminus retain some activity. Two REP sequences were present upstream of entD and an IS186 sequence was observed downstream. RNA dot-blot hybridizations demonstrated that entD is transcribed from the strand predicted by the sequencing results. An entD-lacZ recombinant plasmid was constructed and shown to express low amounts of a fusion protein of the anticipated size (approximately 125 kDa). The evidence suggests a number of possible explanations for difficulties in detecting the entD product. Sequence data indicate that if entD has its own promoter, it is weak; the REP sequences suggest that entD mRNA may be destabilized; and translation may be slow because of the frequency of rare codons and a possible unusual start codon (UUG). The data are also consistent with previous evidence that the entD product is unstable.
The Escherichia coli fepB gene encodes a periplasmic protein required for ferrienterobactin transport; four fepB-related polypeptides are resolved by standard sodium dodecyl sulfate-polyacrylamide gel electrophoresis. In vitro DNA-directed protein-synthesizing systems and experiments with the inhibitors dinitrophenol, carbonyl cyanide m-chlorophenylhydrazone, and ethanol demonstrated that the initial fepB translation product is processed. The nucleotide sequence of fepB and neighboring regions was determined. The predicted proFepB has a molecular weight of 34,255, consists of 318 amino acids, and is devoid of cysteine residues. A leader peptide is present, as are three possible leader peptidase cleavage sites after positions 22, 23, and 26. The upstream regulatory region included a Fur box, indicating that fepB is iron regulated, which was verified by RNA dot blot experiments. The regulatory region included a 68-amino-acid open reading frame (ORF) that encompassed a sequence capable of forming a large stem-and-loop structure. Indirect evidence indicated that this ORF must be translated for fepB transcription to occur. Six copies of the nonomer CCCTC(A/T)CCC or its invert were present in the stem-and-loop region. An ORF of unknown significance was found downstream from fepB; its product would have a molecular weight of 18,036 and be rich in proline and alanine. Processing of proFepB remains unclear, but the appearance of the three smaller members of the FepB family required the action of leader peptidase and the presence of the entire fepB gene.
The fepA gene of Escherichia coli encodes the outer-membrane receptor protein for ferrienterobactin. Previous genetic studies indicated that fepA mutations occur frequently and suggested that most of the mutations were deletions. In this work seven spontaneous fepA mutations were analyzed by enzyme assay (enterobactin synthase and enterobactin esterase) and by DNA hybridization studies. In two strains, UT500 and UT700, the mutations were confined to the fepA gene. In the remaining mutants, the mutations were large deletions; in several cases, 27 kb or more of DNA had been lost. The deletions, all of which eliminated approximately the left half of the enterobactin gene cluster, extended from the vicinity of the fepC gene counterclockwise into the chromosome. A minimum of three clockwise endpoints were identified and at least two counterclockwise endpoints were detected. The variation in endpoints among the deletions argues against the involvement of a normal transposon in their formation. Also, unexpected homology was found between enterobactin gene cluster DNA and lacPOZ and pSC101.
Escherichia coli genes specifically required for transport of iron by the siderophore enterobactin are designated fep. The studies reported here were initiated to identify and localize the fepB product. The plasmid pCP111, which consisted of an 11-kilobase E. coli DNA fragment containing fepB ligated to pACYC184, was constructed. The fepB gene was subcloned; in the process, complementation tests and Tn5 mutagenesis results provided evidence for the existence of a new fep gene, fepC. The order of the transport genes in the ent gene cluster is as follows: fepA fes entF fepC fepB entE. Minicell, maxicell, and in vitro DNA-directed protein synthesizing systems were used to identify the fepB and fepC products. The fepC polypeptide was 30,500 daltons in standard sodium dodecyl sulfate-polyacrylamide gels. The fepB gene was responsible for the appearance of three or four bands (their apparent molecular weights ranged from 31,500 to 36,500) in sodium dodecyl sulfate-polyacrylamide gels, depending on the gel system employed. The largest of these was tentatively designated proFepB, since it apparently had a leader sequence. Localization experiments showed that FepC was a membrane constituent and that mature FepB was present in the periplasm. An additional polypeptide (X) was also encoded by the bacterial DNA of pCP111, but its relationship to iron transport is unknown. The results indicated that ferrienterobactin uptake is mediated by a periplasmic transport system and that genes coding for outer membrane (fepA), periplasmic (fepB), and cytoplasmic membrane (fepC) components have now been identified.
The temperature-dependent expression of OmpF and OmpC, the major channel-forming proteins of the Escherichia coli K-12 outer membrane, was studied. In wild-type cells, decreasing growth temperatures resulted in increased amounts of OmpF protein and correspondingly decreased quantities of OmpC protein. Bacteria deleted for the 13-min chromosomal region did not exhibit this temperature-dependent fluctuation in porin proteins. Plasmid pML22, which consists of pBR322 containing a 0.5-megadalton E. coli chromosomal DNA insert, complemented the thermoregulatory defect. The regulatory gene was named envY. In minicells, pML22 directed the synthesis of an envelope polypeptide (EnvY) having an apparent molecular weight of 25,000. The EnvY protein was synthesized in minicells in greater amounts at 27 degrees C than at 37 degrees C, and a reducing agent was necessary in the solubilization buffer for its subsequent detection on polyacrylamide gels. The results describe the initial characterization of a regulatory system which, along with proteins of the ompB operon, the cyclic AMP system, and the tolC gene product, is involved in a complex network affecting major porin expression.
Escherichia coli mutants defective in the assimilation of iron from ferrienterochelin were isolated and characterized. One mutant was able to bind ferrienterochelin to its outer membrane but could not transport it into the cell. Complementation tests with lambda hybrid phage were employed to distinguish the defective gene, which we term fepB, from fepA, the structural gene for the outer membrane ferrienterochelin receptor protein. These same physiological and genetic tests were employed to tentatively classify several previously described fep mutants as carrying either fepA or fepB. The data demonstrate the existence of fepB and provide an explanation for previous difficulties in identifying fepB mutants.
We characterized UC-1, a previously undescribed Escherichia coli phage. UC-1 was observed to have an icosahedral head and a long, flexible, noncontractile tail: its genome consisted of linear double-stranded DNA having a molecular weight of 34 X 10(6). The product of the tonA gene served as at least part of the receptor for UC-1. E. coli tonA strains neither plated nor adsorbed UC-1 well, tonA mutants were selected on the basis of UC-1 resistance, and ferrichrome, a siderophore which utilizes TonA as its receptor, blocked infection. Restriction analyses, DNA-DNA hybridization experiments, and guanine-plus-cytosine determinations demonstrated that UC-1 DNA was unrelated to that of other phages (T1, T5, and phi 80) which employ TonA as a receptor. Also, mutants specifically resistant to UC-1 were isolated. UC-1 may be useful as a probe for investigating TonA, which functions as a receptor for more ligands than any other membrane protein. Study of the resistant mutants may improve our understanding of how phage DNA penetrates the cell envelope.
Iron uptake in pseudorevertants of Escherichia coli K-12 strains which lack the ability to synthesize enterochelin, 2,3 dihydroxybenzoate, and the ferrienterochelin receptor protein was characterized. In four independent pseudorevertants, the suppressor mutations which permitted growth in iron-poor environments appeared to be located in omp B, the regulatory locus for the porin proteins. Unlike wild-type cells, the pseudorevertants were unable to utilize ferrienterochelin and could acquire iron from citrate without induction by prior growth in citrate. The energy requirements of the pseudorevertant system appeared to be identical to those of the enterochelin system. Evidence that loss of the porin proteins results in the secretion by the pseudorevertants of a molecule with siderophore activity is presented; this siderophore is able to remove iron from the non-biological iron chelators nitrilotriacetic acid and alpha, alpha'-dipyridyl but not fom the siderophores ferrichrome and enterochelin.