Escherichia coli one- and two-hybrid systems for the analysis and identification of protein-protein interactions.

Genetic methods based on fusion proteins allow the power of a genetic approach to be applied to the self-assembly of proteins or protein fragments, regardless of whether or not the normal function of the fused assembly domains is either known or amenable to selection or screening. The widespread adoption of variations of the yeast two-hybrid system originally described by S. Fields and O. Song (1989, Nature 340, 245-246) demonstrates the usefulness of these kinds of assays. This review describes some of the many systems used to select or screen for protein-protein interactions based on the regulation of reporter constructs by hybrid proteins expressed in bacteria, including recent implementations of generalizable two-hybrid systems for Escherichia coli.

Gene activation by the AraC protein can be inhibited by DNA looping between AraC and a LexA repressor that interacts with AraC: possible applications as a two-hybrid system.

The Escherichia coli activator and repressor proteins AraC and LexA bind DNA as homodimers. Here we show that their heterodimerization through fused cognate dimerization domains results in repression of AraC-dependent gene activation by LexA. Repression also requires a LexA operator half-site located several helical turns downstream of the AraC operator. This requirement for a specific spatial organization of the operators suggests the formation of a DNA loop between operator-bound Ara/LexA heterodimers, and we propose that heterodimerization with the AraC hybrid provides co-operativity for operator binding and repression by the LexA hybrid. Consistent with a mechanism that involves DNA looping, repression increases when the E. coli DNA looping and transcriptional effector protein IHF binds between the AraC and LexA operators. Thus, we have combined the functions of three distinct transcriptional effector proteins to achieve a new mode of gene regulation by DNA looping, in which the activator protein is an essential part of the repressor complex. The flexibility of the DNA loop may facilitate this novel combinatorial arrangement of those proteins on the DNA. The requirement for protein interactions between the AraC and LexA hybrids for gene regulation suggests that this regulatory circuit may prove useful as an E. coli-based two-hybrid system.

Information essential for cell-cycle-dependent secretion of the 591-residue Caulobacter hook protein is confined to a 21-amino-acid sequence near the N-terminus.

Recent findings suggest that axial flagellar proteins and virulence proteins of Gram-negative bacteria are exported from the cytoplasm via conserved translocation systems. To identify residues essential for secretion of flagellar axial proteins we examined the 591-residue Caulobacter crescentus flagellar hook protein. Western blot assays of the culture media of strains producing mutant hook proteins show that only residues 38-58 are essential for its secretion to the cell surface. We discuss the observation that this unprocessed 21-residue sequence is not conserved in other axial proteins and does not correspond to the SGL-, ANNLAN- and heptad repeat motifs that are located just upstream of the essential secretion information in the hook protein and are conserved near the N-termini of other axial proteins. These motifs, for which an essential role in export or assembly has been proposed, are required for motility. However, we also demonstrate that hook protein can only be secreted when the flagellar basal body is present in the cell envelope. The cell-cycle regulation of hook protein secretion confirms the specificity of the assay used in these studies and suggests that the basal body itself may serve as a secretion channel for the hook protein.

The role of the lipoprotein sorting signal (aspartate +2) in pullulanase secretion.

The analyses of hybrid proteins and of deletion and insertion mutations reveal that the only amino acid at the amino-proximal end of the cell surface lipoprotein pullulanase that is specifically required for its extracellular secretion is an aspartate at position +2, immediately after the fatty acylated amino-terminal cysteine. To see whether the requirement for this amino acid is related to its proposed role as a cytoplasmic membrane lipoprotein sorting signal, we used sucrose gradient floatation analysis to determine the subcellular location of pullulanase variants (with or without the aspartate residue) that accumulated in cells lacking the pullulanase-specific secretion genes. A non-secretable pullulanase variant with a serine at position +2 cofractionated mainly with the major peak of outer membrane porin. In contrast, most (55%) of a pullulanase variant with an aspartate at position +2 cofractionated with slightly lighter fractions that contained small proportions of both outer membrane porin and the cytoplasmic membrane marker NADH oxidase. Only 5% of this pullulanase variant cofractionated with the major NADH oxidase peak, while the rest (c. 40%) remained at the bottom of the gradient in fractions totally devoid of porin and NADH oxidase. When analysed by sedimentation through sucrose gradients, however, a large proportion of this variant was recovered from fractions near the top of the gradient that also contained the major NADH oxidase peak. When this peak fraction was applied to a floatation gradient the pullulanase activity remained at the bottom while the NADH oxidase floated to the top.(ABSTRACT TRUNCATED AT 250 WORDS)

Secretion of the cell surface lipoprotein pullulanase in Escherichia coli. Cooperation or competition between the specific secretion pathway and the lipoprotein sorting pathway.

The fatty acid-acylated enzyme pullulanase is normally found in either of two locations in Escherichia coli, depending on whether or not the producing strains also express the genes specifically required for the second step in pullulanase secretion. When they are expressed, the enzyme is localized to the cell surface, while in their absence, it is directed to an unidentified location in the cell envelope which, upon lysis, forms vesicles whose density is intermediate between those of outer and cytoplasmic membrane vesicles. In order to test the role of the putative lipoprotein sorting signal, Asp2, in pullulanase sorting and secretion, the structural gene (pulA) was subjected to site-directed mutagenesis. Replacement of the Asp2 residue by Asn, Glu, or Ser caused the enzyme to fractionate with outer membrane-derived vesicles rather than with intermediate density vesicles from E. coli cells devoid of pullulanase secretion genes. A pronounced secretion defect was observed in a two-step secretion assay in which the first (sec gene-dependent) and second (pul gene-dependent) secretion steps were uncoupled. We propose that the Asp residue increases the efficiency of pullulanase secretion by allowing the enzyme to be initially sorted to a region of the cell envelope wherein most of the pullulase-specific secretion factors are located.

Outer membrane translocation of the extracellular enzyme pullulanase in Escherichia coli K12 does not require a fatty acylated N-terminal cysteine.

Site-directed mutagenesis was used to construct three mutant derivatives of the extracellular, cell surface lipoprotein pullulanase (PulA) in which the normally fatty acylated cysteine of the signal peptide-bearing precursor was replaced by other amino acids. When produced in Escherichia coli expressing all genes required for pullulanase secretion, approximately 90% of the PulA derivatives persisted as cell-associated precursors, indicating inefficient signal peptide processing. Processed (intermediate-sized) forms of the two derivatives that were studied in detail were found to result from proteolytic cleavage at different sites within the signal peptide. Both were further processed to smaller polypeptides by cleavage at an undetermined site that is presumably close to their C termini. The intermediate-sized pullulanase derived from prepullulanase in which Cys+1 had been replaced by Leu and Gly-1 by Glu (PulA:C1L/G-1E) appeared rapidly, was apparently entirely extracellular, and accounted for approximately 10% of synthesized PulA. Prolonged incubation did not result in further conversion of the precursor to the intermediate form, and the precursor remained anchored to the cytoplasmic membrane. The smaller processed form was also found extracellularly. The active form of the extracellular enzyme was monomeric, which is again in contrast to the fatty acylated, wild-type enzyme. Taken together, these results indicate that replacement of Cys+1 of prePulA eliminates processing by lipoprotein signal peptidase and does not permit processing by leader peptidase, but allows inefficient, aberrant processing by an unknown peptidase and immediate secretion of the resulting polypeptide, which retains most of its signal peptide. Processing and secretion only occur when the pullulanase secretion functions are expressed.

Two distinct steps in pullulanase secretion by Escherichia coli K12.

Two distinct steps in the secretion of the extracellular, cell-surface-anchored lipoprotein pullulanase by Escherichia coli were uncoupled by allowing export of the enzyme to the cytoplasmic membrane via the signal peptide/sec-gene-dependent general export pathway, and then inducing the pulC-O operon of genes required for translocation to the cell surface. The secretion intermediate cofractionated mainly with intermediate-density vesicles when cells were gently lysed and the resulting vesicles were separated by isopycnic sucrose density centrifugation. Cytoplasmic forms of pullulanase (which are not exported because they lack a functional signal peptide) are more sensitive to heat inactivation, denaturation by sodium dodecyl sulphate and carboxymethylation than the intermediate and cell-surface forms. The latter are distinguished only by the fact that the secretion intermediate is less susceptible to proteinase K and trypsin, and is partially inaccessible to substrate or in an inactive conformation in sphaeroplasts. These and other results indicate that the secretion intermediate can acquire considerable higher-ordered structure, including disulphide bridges, before it is transported to the cell surface; this seems to rule out the possibility that it is threaded through this membrane as a locally unfolded polypeptide.

The general protein-export pathway is directly required for extracellular pullulanase secretion in Escherichia coli K12.

Pullulanase is an extracellular, cell surface-anchored lipoprotein produced by Gram-negative bacteria belonging to the genus Klebsiella. Its correct localization in recombinant Escherichia coli requires the products of 14 genes that are linked to the enzyme structural gene in the Klebsiella chromosome. In addition, we show here that six sec genes (secA, secB, secD, secE, secF and secY) are all required for processing of the prepullulanase signal peptide to occur. This implies that pullulanase crosses the cytoplasmic membrane via the general export pathway of which the sec gene products are essential components. Removal or drastic alteration of the prepullulanase signal peptide cause the enzyme to remain cytoplasmic. We propose that pullulanase secretion occurs in two steps, the first of which is common to all signal peptide-bearing precursors of exported and secreted proteins, whereas the second is specifically involved in translocating pullulanase to the cell surface.

The normally periplasmic enzyme beta-lactamase is specifically and efficiently translocated through the Escherichia coli outer membrane when it is fused to the cell-surface enzyme pullulanase.

Hybrid proteins were constructed in which C-terminal regions of the bacterial cell surface and extracellular protein pullulanase were replaced by the mature forms of the normally periplasmic Escherichia coli proteins beta-lactamase or alkaline phosphatase. In E. coli strains expressing all pullulanase secretion genes, pullulanase-beta-lactamase hybrid protein molecules containing an N-terminal 834-amino-acid pullulanase segment were efficiently and completely transported to the cell surface. This hybrid protein remained temporarily anchored to the cell surface, presumably via fatty acids attached to the N-terminal cysteine of the pullulanase segment, and was subsequently specifically released into the medium in a manner indistinguishable from that of pullulanase itself. These results suggest that the C-terminal extremity of pullulanase lacks signal(s) required for export to the cell surface. When beta-lactamase was replaced by alkaline phosphatase, the resulting hybrid also became exposed at the cell surface, but exposition was less efficient and specific release into the medium was not observed. We conclude that proteins that do not normally cross the outer membrane can be induced to do so when fused to a permissive site near the C-terminus of pullulanase.

Analysis of the subcellular location of pullulanase produced by Escherichia coli carrying the pulA gene from Klebsiella pneumoniae strain UNF5023.

Three different techniques, protease accessibility, cell fractionation and in situ immunocytochemistry, were used to study the location of the lipoprotein pullulanase produced by Escherichia coli K12 carrying the cloned pullulanase structural gene (pulA) from Klebsiella pneumoniae, with or without the K. pneumoniae genes required to transport pullulanase to the cell surface (secretion-competent and secretion-incompetent, respectively). Pullulanase produced by secretion-competent strains could be slowly but quantitatively released into the medium by growing the cells in medium containing pronase. The released pullulanase lacked the N-terminal fatty-acylated cysteine residue (and probably also a short N-terminal segment of the pullulanase polypeptide), confirming that the N-terminus is the sole membrane anchor in the protein. Pullulanase produced by secretion-incompetent strains was not affected by proteases, confirming that it is not exposed on the cell surface. Pullulanase cofractionated with both outer and inner membrane vesicles upon isopycnic sucrose gradient centrifugation, irrespective of the secretion competence of the strain. Examination by electronmicroscopy of vesicles labelled with antipullulanase serum and protein A-gold confirmed that pullulanase was associated with both types of vesicles. When thin-sectioned cells were examined by the same technique, pullulanase was found to be located mainly on the cell surface of the secretion-competent cells and mainly in the proximity of the inner membrane in the secretion-incompetent cells. Thus, while the results from three independent techniques (substrate accessibility, protease accessibility and in situ immunocytochemistry) show that pullulanase is transported to the cell surface of secretion-competent cells, this could not be confirmed by cell-fractionation techniques. Possible explanations for this discrepancy are discussed.

Molecular characterization of pulA and its product, pullulanase, a secreted enzyme of Klebsiella pneumoniae UNF5023.

The determined nucleotide sequence of the Klebsiella pneumoniae UNF5023 gene pulA comprises a single open reading frame coding for a 1090-residue precursor of the secreted protein pullulanase. The predicted sequence of this protein is highly homologous to that of pullulanase of Klebsiella aerogenes strain W70. However, the UNF5023 pullulanase lacks a collagen-like sequence present at the N-terminus of the mature W70 enzyme and differs further from the W70 pullulanase around residue 300 and at the C-terminus. Pullulanases with or without the collagen-like sequence could not be separated by gel electrophoresis under denaturing or non-denaturing conditions, and were unaffected by collagenase. A large central domain which is highly conserved in both UNF5023 and W70 polypeptides contains eight short sequences that are also found in amylases and iso-amylases. Linker mutations in the region of the UNF5023 pulA gene coding for this domain abolished catalytic activity without affecting transport of the polypeptide across the outer membrane. Hybrid proteins comprising at least the amino-terminal 656 residues of prepullulanase fused to alkaline phosphatase were partially localized to the cell surface, as judged by their accessibility to anti-pullulanase serum in immuno-fluorescence tests. On the basis of these results, we tentatively propose that secretion signals required for recognition and translocation across the outer membrane via the pullulanase-specific extension of the secretion pathway are located near the N-terminus of the pullulanase polypeptide.

Klebsiella pneumoniae strain K21: evidence for the rapid secretion of an unacylated form of pullulanase.

Klebsiella pneumoniae strain PAP996 was previously shown to secrete fatty acylated, aggregated (micellar) pullulanase only after the end of exponential growth. Here we show that the closely related strain K21 secretes large amounts of unacylated, non-aggregated (monomeric) pullulanase during exponential growth. Only a small amount (less than 10%) of the secreted pullulanase was initially retained by the exponentially growing cells to be subsequently secreted in a fatty acylated, aggregated form. Despite the absence of fatty acids in secreted monomeric pullulanase, the effects of the antibiotic globomycin on pullulanase maturation indicated that all of the enzyme synthesized by strain K21 is processed by lipoprotein signal peptidase.

A new regulatory locus of the maltose regulon in Klebsiella pneumoniae strain K21 identified by the study of pullulanase secretion mutants.

This study has shown that Klebsiella pneumoniae strain K21 differs from the previously characterized and closely related K. pneumoniae strain PAP996 in that expression of the pullulanase gene (pulA) and other genes of the maltose regulon is partially independent of exogenous inducer (maltose/maltotriose). Mutants of strain K21 which are defective in pullulanase synthesis and/or secretion were isolated following Tn10 mutagenesis. Three phenotypic classes of mutants were identified. Class I mutants were defective in the surface localization and secretion of pullulanase. Class II mutants did not secrete detectable levels of pullulanase but were able to export pullulanase to the cell surface. Class II mutants also expressed pullulanase and other maltose-regulated genes at markedly lower levels than those found in the parent strain under non-inducing conditions. The single class III mutant was intermediate between K21 and class I mutants; most of the cell-associated pullulanase was localized at the cell surface whilst a significant amount was secreted into the medium. Mapping indicated that all but three of the Tn10 insertions were adjacent to, and at either side of, pulA. One class II mutant carried a Tn10 insertion in or close to malT whereas in the remaining class II mutants the insertions were located at least 4 kb upstream of pulA in a region which may define a new regulatory locus of the maltose operon.