Secretion across the Bacterial Outer Membrane
Chapter
63
CÉCILE WANDERSMAN
Unlike eukaryotic cells, bacteria have few distinct cellular compartments. Gram-positive bacteria have only one cytoplasmic compartment delimited by a unique membrane. Gram-negative bacteria are more differentiated, with two membranes which delimit two compartments: the periplasm and the cytoplasm. Bacterial proteins synthesized in the cytoplasm are submitted to cellular sorting which distinguishes among them. Some remain in the cytoplasm, whereas others are targeted out of the cytoplasm. This sorting is carried out by a universal mechanism, the signal peptide-dependent general export pathway (GEP). This pathway consisting of the Sec proteins, mostly essential genes, is now well characterized in Escherichia coli. It is also called the sec-dependent pathway (124) and is described in another chapter of this book.
Bacteria also interact with their environment through proteins released into the extracellular medium. These factors are usually not essential for bacterial growth under laboratory culture conditions, but they often enhance bacterial colonization and dissemination during infection via various mechanisms. Bacteria cannot take up macromolecules by endocytosis, so for the most part, they have to degrade them into transportable fragments to get them through their outer membranes. This is achieved by extracellular degradative enzymes. Such hydrolytic secreted enzymes also facilitate bacterial spread by degrading the extracellular host matrix. Most of the other secreted proteins are invasins, which are involved in host cell penetration, and various kinds of toxins, which kill the host cells, or bacteriocins, which destroy bacterial competing strains.
Until recently, the mechanisms involved in protein release in the extracellular medium were poorly understood. Protein secretion without concomitant cell lysis was thought to occur mainly in gram-positive bacteria and in very few gram-negative species. This conviction had arisen from the focusing of bacterial research on E. coli K-12, which does not secrete any protein beyond the outer membrane (112). New genetic tools, allowing the study of a large diversity of bacteria, developed during the last decade have revealed that this idea was wrong and have demonstrated the importance of extracellular proteins in various bacterial physiological processes.
In gram-positive bacteria, extracellular proteins are secreted by the signal peptide-dependent GEP. Translocation through the unique cytoplasmic membrane and cleavage of the signal peptide lead in most cases to the release of the mature polypeptide to the surrounding medium. This review focuses on secretion in gram-negative bacteria; specific aspects of protein secretion in gram-positive bacteria not described here can be found in recent articles (72, 130).
In gram-negative bacteria, some extracellular proteases are directed to the outer membrane and then are autocatalytically excised from it (71). This secretion mechanism will be described in the first part of the present review. However, aside from such special cases, the GEP is usually not sufficient to direct proteins beyond the outer membrane. This was first shown by the expression in E. coli of genes of different bacterial origins coding for extracellular proteins. The foreign proteins were synthesized in E. coli but not secreted to the extracellular medium (20, 64, 91, 119). This indicates that the secreted proteins do not contain sufficient information to allow their secretion and that, very likely, the recipient strain lacks some secretion functions. In several cases the foreign exoproteins had a signal peptide and, when expressed in E. coli, were located in the periplasmic space, suggesting that they were translocated through the inner membrane by the sec-dependent pathway and were stuck there in the absence of secretion functions. Supporting this hypothesis, the foreign proteins were shown to accumulate in the cytoplasm of E. coli sec mutants at a nonpermissive temperature, indicating that their periplasmic localization was indeed dependent on the sec genes (57). The isolation, from the various natural secreting species, of viable secretion-defective mutants which accumulated the polypeptides in the periplasm led to the notion that translocation through the outer membrane was carried out by a nonessential, specific mechanism (3, 40).
Finally, the cloning of the genes required for secretion and the reconstitution of various secretion pathways in E. coli allowed identification of the specific secretion activities (27, 56, 118). These secretion systems are therefore characterized by a two-step process since the secreted polypeptides cross the two membranes by distinct mechanisms, the inner membrane by the GEP and the outer membrane by a specific mechanism. The constitution of these transporters is not uniform. Some are very simple, consisting of a single helper protein located in the outer membrane, such as the Serratia marcescens hemolysin transporter described below (18). Others are more sophisticated, consisting of 13 to 14 helper proteins which span the whole cell envelope. Surprisingly, these transporters are the most common among gram-negative bacteria and constitute a particular secretion pathway called the general secretory pathway (GSP), described in the third part of this review (112). On the other hand, it now appears that a two-step secretion pathway is not the only one in gram-negative bacteria. At least two major pathways have evolved, allowing proteins lacking cleavable signal peptides to traverse the cell envelope by secretion mechanisms independent of the GEP. One of them is well defined and will be described in the fourth part of this review (34, 136), whereas the other is less well characterized and will be described in the last part of this review (135). These signal peptide-independent pathways are not alternative translocation mechanisms since they do not target proteins to the cell envelope. They bypass the GEP and are for extracellular secretion only. This review will be divided into two sections, the sec-dependent and -independent pathways. The first three secretion mechanisms are covered in the first section, and the last two are covered in the second section.
Several extracellular proteases, such as the Neisseria and Haemophilus immunoglobulin A (IgA) proteases (71, 107) and the Serratia serine protease (95), use this pathway, now well characterized for the IgA protease of Neisseria gonorrhoeae, which splits the hinge region of human IgA. This protein is made as a precursor of 169 kDa which has five distinct functional domains: the N-terminal signal peptide, the protease domain with a molecular mass of 106 kDa, the γ domain of 30 amino acids, the 12-kDa α domain, and the C-terminal 45-kDa β domain. After cleavage of the signal peptide, the precursor, transiently located in the periplasm, is incorporated in the outer membrane by its C-terminal β domain. Structure predictions indicate that this domain has a β-barrel secondary structure characteristic of outer membrane proteins (22, 142). Moreover, membrane localization has shown that, indeed, the β polypeptide is associated with the outer membrane fraction. The 31-kDa C-terminal part of this β domain is membrane embedded, whereas its N-terminal part, called the linking region, is exposed together with the mature domain at the cell surface. After incorporation of the β domain in the outer membrane, the enzyme is folded in its active configuration and is cut out of the β domain, which remains in the outer membrane, by autoproteolytic cleavage. The extracellular precursor is further processed by two cleavages which remove the α and γ domains. The internal cleavage sites display a proline-rich consensus sequence similar to the sequence cleaved in IgA1. Hybrid proteins having the precursor of the Vibrio cholerae toxin B subunit (CtxB) fused to the β domain of the IgA protease were constructed (70). Whereas the toxin B subunit expressed in E. coli is normally periplasmic, the fusion proteins were translocated across the outer membrane and exposed at the cell surface, accessible to proteases and antibodies. Nevertheless, the hybrid was not excised from the outer membrane. The efficiency of hybrid protein translocation across the outer membrane increased when the formation of disulfide bonds in the passenger polypeptide was prevented by the addition of sulfhydryl reagents. Similarly, the presence of a dsbA mutation (8), which inactivates the periplasmic oxidoreductase and therefore limits cysteine bond formation, increased the hybrid export through the outer membrane (69). The IgA protease has only two cysteine residues, located in the protease domain and only 11 amino acids apart. It is likely that these cysteine residues do not form intrachain disulfide bonds or produce a loop which is compatible with secretion, since the presence of a dsbA mutation does not increase protease secretion. Nevertheless, the results obtained with the CtxB fusion suggest that stable tertiary structure has an inhibitory effect on this secretion mechanism. The fate of the β domain is unknown: does it remain in the outer membrane or is it more or less rapidly eliminated? There is no evidence that the β domain is reused, as all molecules are made as complete precursors.
It is also unknown whether the α and γ domains, cut out by autoproteolytic processing, have any biological function in the extracellular medium.
Finally, it is not clear whether this secretion system is restricted to proteases which are able to excise from the outer membrane by virtue of their own enzymatic activities. Cleavage in trans by other proteases is indeed observed with CtxB-Iga protease fusions and with S. marcescens serine protease mutants lacking proteolytic activity. The cleavage can be carried out by OmpT, a trypsin-like outer membrane protein of E. coli (51). A secretion mechanism involving release from the outer membrane by proteolytic cleavage could be responsible for the secretion of the Pseudomonas solanacearum endoglucanase (62) and the Helicobacter pylori VacA protein (126). The Shigella surface protein IcsA (VirG) is also released into the supernatant in a cleaved form in strains expressing OmpT (99).
S. marcescens pore-forming hemolysin, ShlA, is a large protein of 1,578 residues. It has an N-terminal signal peptide and is translocated through the inner membrane by the GEP. The crossing of the outer membrane is achieved in the presence of a helper protein, ShlB (18, 125). The expression of ShlA together with ShlB is sufficient to promote hemolysin secretion in E. coli. In the absence of ShlB, inactive ShlA accumulates in the periplasm, showing that ShlB is involved in hemolysin secretion and activation. ShlB has been localized in the outer membrane, and computer predictions suggest the presence of β-sheeted transmembrane strands forming a β-barrel secondary structure typical of outer membrane proteins. This protein might also have a large periplasmic domain. Inactive ShlA protein, extracted from the periplasmic space of mutant strains producing only ShlA, can be activated in vitro. Activation is achieved by incubation with a lysate of cells expressing only ShlB, suggesting that ShlB is able to modify ShlA. The nature and the target of the modification on ShlA are not yet known. The following experiment suggests that modification is in the N-terminal part of ShlA since activation occurs also by incubation of inactive ShlA with nonhemolytic amino-terminal fragments of ShlA secreted in the presence of ShlB. The N-terminal ShlA competent fragments must comprise more than 149 residues and were obtained by in vitro trypsin cleavage of a secreted 255-residue-long ShlA N-terminal fragment. It is proposed that the inactive ShlA monomer interacts with the truncated fragments modified by ShlB and that these heterodimers display hemolytic activity. The ShlB modification has to occur within the first 149 N-terminal residues of ShlA. Moreover, the ShlA signal recognized by ShlB and required for hemolysin secretion is also located in the amino-terminal region of the mature protein since the smallest amino-terminal fragment still secreted has 238 residues.
Other proteins seem to use a similar secretion strategy. Proteus mirabilis hemolysin displays sequence homology with ShlA and is activated and secreted by an ShlB analog protein. Heterologous in vitro activation occurs, indicating that the two hemolysins have very similar properties (104). The Bordetella pertussis FhaB protein may also require a protein homologous to ShlB for its secretion (144).
Extracellular proteins secreted by this complex pathway have signal peptides and cross the inner membrane via the GEP. The crossing of the outer membrane requires a secretion machinery composed of 12 to 14 specific proteins. Such secretion apparatuses have been identified over the past 10 years in many gram-negative bacteria, e.g., Klebsiella oxytoca (111), Erwinia carotovora (118), Erwinia chrysanthemi (55), Aeromonas hydrophila (148), Pseudomonas aeruginosa (39), V. cholerae (106), and Xanthomonas campestris (29) Except for K. oxytoca, which secretes only one enzyme, pullulanase, these bacteria secrete several structurally diverse exoenzymes. Nevertheless, each bacterium possesses only one secretion machinery common for all polypeptides. Most of the secretion genes have been identified by isolation of secretion-defective mutants, followed by cloning and sequencing of the genes which complement the secretion defects. In two cases the secretion systems have also been reconstituted in E. coli: the K. oxytoca pullulanase (27) and the E. chrysanthemi pectinase and cellulase secretion apparatuses (56). Reconstitution of secretion in E. coli, which does not have a functional GSP, indicates that the specific components that are not native to E. coli have been identified. In such a way, 14 pul genes were shown to be required to secrete pullulanase and 13 out genes were required for pectinases and cellulases. The secretion genes are clustered and organized in one or two operons. In the case of the pullulanase system, which is dedicated to pullulanase secretion, the structural gene for the exoenzyme is linked to those encoding the secretion proteins. However, this genetic organization is not found in the other bacteria, which secrete several proteins. Among systems which have not yet been reconstituted in E. coli, the best characterized is that of P. aeruginosa with 12 secretion genes, the xcp genes, clustered in two genetic loci and required for the secretion of elastase, exotoxin A alkaline phosphatase, and phospholipase C (2, 6, 39). The absence of secretion in E. coli could result either from the lack of one or more specific secretion proteins or from an incorrect association of these proteins in the E. coli envelope. DNA sequence analysis of secretion genes from diverse systems has revealed that they have homologous proteins, sharing sequence identity ranging from 30 to 60% (Fig. 1) (23). Despite these homologies, usually, there is no cross functioning among the systems: the secretion machinery in one bacterial species is unable to secrete an exopolypeptide from a closely related species, even if it is highly homologous. In most cases, exchange of individual secretion proteins in the various apparatuses results in loss of function. In contrast, nonhomologous proteins from the same bacterium are secreted by the corresponding secretion apparatus. To date, the nature of the common element discerned on these proteins by the secretion machinery is completely unknown. These secretion systems raise several questions. What are the individual functions of the secretion proteins? Which or how many of them recognize the periplasmic intermediate? What is the targeting signal on the secreted polypeptides?
Individual Functions of the Secretion Proteins.
A clue to the function of one of these proteins, XcpA of P. aeruginosa, came with the demonstration that XcpA is identical to the P. aeruginosa PilD protein required for pilin biogenesis (102). One protein is therefore involved in two apparently different processes: pilus synthesis and protein secretion. Pili of P. aeruginosa are located at the cell surface and are composed of subunits of a single polypeptide encoded by the pilA gene. This 15-kDa pilin monomer is very similar to the pilus subunits of other pathogens such as N. gonorrhoeae, Moraxella bovis, or Bacteroides nodosus (133). These type IV pilin polypeptides have a conserved 25-residue N-terminal sequence characterized by the presence of six or seven hydrophilic and positively charged amino acids followed by a stretch of hydrophobic residues. However, despite the similarities with classical signal sequences, only the first six or seven amino-terminal residues are cleaved off by the product of the pilD/xcpA gene, the pilin peptidase. The N-terminal Phe residue exposed on the mature pilin is methylated by the PilD protein, which dispays both peptidase and methylase activities (103). Four Xcp secretion proteins, XcpT, -U, -V, and -W, have N-terminal sequences similar to that found in type IV pilin precursors. They are cleaved by XcpA in vivo (6, 103), and the uncleaved XcpT precursor produced by the xcpA mutant was shown to accumulate in the inner membrane, whereas the mature form produced by the wild-type strain was found to preferentially cofractionate with the outer membrane (6). Four pilin-like proteins are also found in the Pul and Out secretion systems of K. oxytoca and Erwinia spp., which do not have type IV pili (112, 118). These precursors are processed by a specific peptidase, PulO or OutO, encoded by a gene linked to the other pul or out genes (114). It was shown that PulO processes one of the pilin-like proteins, PulG, at the consensus prepilin cleavage site and that mature PulG protein has a methylated amino-terminal phenylalanine residue (113). However, in contrast to the results obtained with XcpT, uncleaved and mature PulG exhibited identical subcellular distributions (115). PulO was shown to correctly cleave and methylate N. gonorrhoeae type IV prepilin (33). Reciprocally, PulG could also be correctly processed in E. coli by the N. gonorrhoeae type IV prepilin peptidase (32), showing that pilus assembly and maturation of pilin-like secretion proteins are carried out by functionally interchangeable enzymes. Each pilus consists of 1,000 pilin repeating subunits assembled in a structure which protrudes from the cells. This assembly requires the maturation of the pilin precursor. By analogy with pilus formation, an attractive hypothesis would be that the pilin-like secretion proteins also assemble in a pilus-like structure. However, neither PulG nor XcpT was found in multiprotein complexes (115). Cross-linking experiments have shown that both proteins can be found as dimers, but the possible dimerization was also observed in the absence of the peptidase PulO or XcpA (5, 115) and was therefore independent of pilin precursor maturation, placing in doubt the biological significance of these experiments.
Besides prepilin peptidase, several proteins are required for pilin assembly. Two such proteins are encoded by genes pilB and pilC linked to the pilA/xcpA locus. PilB presents an ATP-binding site and probably provides energy to the assembly process (134). Whereas the same peptidase is shared for pilus assembly and secretion, PilB and PilC homologs are encoded by genes belonging to the xcp cluster. These genes, xcpR and xcpS, could therefore be dedicated to secretion. Owing to their sequence homologies with PilB and PilC, it was proposed that XcpR and XcpS are required for the assembly of the four pilin-like Xcp proteins. XcpR has no hydrophobic domain and could be a cytoplasmic protein. After cell fractionation, it is found associated with the inner membrane, suggesting that it interacts with some inner membrane components (5). XcpR displays an ATP-binding site consensus sequence, and partially purified XcpR was shown to bind ATP (5). Preliminary results indicate that processing of XcpT requires ATPase activity. However, in an xcpT mutant, processing is still efficient, indicating that the process might occur in an XcpR-independent manner or that an alternative pathway exists. Indeed, it was noticed that in E. coli the XcpT processing is SecA and SecY dependent. This suggests that translocation could be carried by the Sec pathway. Since XcpR and XcpS are necessary for exoprotein secretion, neither the Sec pathway nor the PilB/PilC system probably can substitute for the XcpR/XcpS pathway to achieve a functional assembly of the pilin-like proteins (A. Filloux, personal communication). Proteins with potential ATP-binding sites are also found in the other systems (110).
Secretion through the outer membrane is expected to require helper proteins localized in the outer membrane. In the P. aeruginosa Xcp and E. carotovora Out systems, only one protein is localized in the outer membrane, whereas two outer membrane proteins are found in the K. oxytoca Pul and E. chrysanthemi Out systems, PulS and PulD. PulD (112), which is common to all of these secretion systems, is also closely related to the outer membrane pIV protein required for assembly and secretion of filamentous phages, which is extracted from the outer membrane as a homomultimer composed of 10 to 12 monomers (121). While these proteins are likely to be involved in outer membrane crossing events, the mechanisms allowing the passage of large, folded, and in several cases multiprotein complexes are not yet understood.
The other Xcp and Pul secretion proteins are located in the inner membrane. They do not share significant sequence homologies with other known transport proteins.
Nature of the Secretion Signal on the Mature Exoproteins.
Each of the various GSPs is, in fact, highly specific and displays species specificities. It must therefore recognize specific targeting sequences on the mature exoproteins when the latter reach the periplasm after cleavage of the N-terminal signal peptides. Simple sequence comparison between the exoproteins using the same GSP did not reveal any consensus sequence which might be a secretion signal candidate. Mutagenesis of one E. chrysanthemi extracellular cellulase has shown that several domains carry information relevant for secretion (116). A similar approach has identified key residues required for secretion of aerolysin (147). A study of hybrid proteins carrying the N-terminal part of pullulanase fused to β-lactamase has revealed that the information related to secretion is located in the N-terminal two-thirds of mature pullulanase (73). However, with another GSP, PehA-Bla hybrid proteins carrying up to 87% of the endopolygalacturonase (Peh) were not secreted (122).
Taken together, these results suggest that the secretion signal may involve a tertiary structure which brings together key residues forming one or several secretion motifs recognized by the secretion proteins. This implies that the exoproteins are folded in the periplasm and recognized in that conformation by the transporter. In fact, pullulanase and cellulase EGZ has been shown to be translocated only if folding has occurred, since secretion is prevented in the absence of a disulfide bond in a dsbA strain lacking disulfide oxidoreductase (17, 111). Cholera toxin, as well as the structurally similar E. coli heat-labile enterotoxin, consists of one A subunit and five identical B subunits. Both toxins are secreted by V. cholerae by a partially characterized GSP. It was shown that subunit assembly takes place in the periplasm and that the oligomerized toxin is translocated through the outer membrane (123).
Folding of Pseudomonas glumae lipase is achieved in the periplasm by the LipB product, a periplasmic protein anchored in the inner membrane by its N-terminal hydrophobic segment. In the absence of LipB, secretion is prevented and there is periplasmic accumulation of lipase. Nevertheless, missense mutant lipases lacking lipolytic activity are secreted in normal amounts. This suggests that proper folding, but not lipase activity, is linked to secretion (42).
Protein folding was also shown to be a prerequisite for P. aeruginosa elastase secretion. Elastase is synthesized as a preproprotein. After processing of the signal peptide, the periplasmic zymogen is matured by autoproteolytic removal of the propeptide, which remains noncovalently bound to the mature protein until the mature part is secreted through the outer membrane (68). When expressed without propeptide, elastase was inactive in E. coli. The propeptide, provided in trans, restored proteolytic activity, showing that elastase, like several other proteases, utilizes a propeptide as an intramolecular chaperone but can also recognize and use it as an intermolecular chaperone (129, 151). In P. aeruginosa, no elastase was found in the extracellular medium in the absence of the propeptide. Moreover, expression of the propeptide in trans restored elastase secretion. This shows that the propeptide is also required for secretion, indicating that, here again, proper folding is essential for secretion through the outer membrane by the GSP (P. Braun, personal communication).
At present, several such GSP systems have been identified and most of the secretion genes have been sequenced. Much work is needed to understand how the GSPs are assembled in the cell envelope, what the precise functions of these 10 to 12 inner membrane proteins are in a process occurring in the outer membrane, and finally what the molecular recognition steps are between the substrates and the apparatus.
Proteins using the ABC transporters cross both membranes in a single step without a periplasmic intermediate. These secretion pathways are characterized in each case by the presence of a membrane ATPase as a component of the secretion apparatus. These ATPases display a hydrophobic domain with six transmembrane segments and a cytoplasmic domain bearing a nucleotide-binding site (50). Resulting, presumably, from genetic duplication, several of these ATPases have homologous halves, each bearing a hydrophobic and a cytoplasmic domain. Whereas the hydrophobic domains are poorly conserved, the homologies in the cytoplasmic domains cover a 200-amino-acid stretch extending well beyond the consensus ATP-binding sequences. It has been proposed that the cytoplasmic domain constitutes a conserved cassette linked to variable membrane domains. For this reason, this group of ATPases has been named the ABC family (ATP-binding cassette) and, by extension, ABC transporters, those which constitute an ABC protein (58). ABC proteins are involved in membrane transport of various substrates in a large number of organisms. Eukaryotic proteins include the multidrug resistance (MDR) transporter, which can pump hydrophobic drugs out of the cell; the cystic fibrosis gene product (CFTR), which forms the chloride channel (120); the 70-kDa mammalian peroxisomal protein (1, 97); the yeast STE6 protein required for secretion of a mating factor across the plasmic membrane (76); the yeast MDR proteins involved in multidrug resistance (7, 14); the yeast vacuolar protein involved in heavy metal resistance (105); an essential yeast mitochondrial ABC transporter (78); and finally, the Tap1 and Tap2 major histocompatibility complex-encoded peptide transporter of the mammalian endoplasmic reticulum (65) (Fig. 2). In gram-positive bacteria, ABC proteins are implicated in drug efflux, lantibiotic secretion such as nisin and subtilin, and Enterococcus faecalis cytolysin secretion (34, 47, 90, 101).
In gram-negative bacteria, ABC proteins are involved in the uptake of various substrates, including maltose, histidine, oligopeptides, and iron siderophore complexes. To accomplish uptake, the ABC proteins interact with specific binding proteins (58). Finally, they are also required for protein secretion (34, 136). ABC proteins are usually highly specific for one substrate transported in one direction: uptake or export. Nevertheless, there are notable exceptions to this specificity rule. First, the mammalian MDR protein, which is overproduced in tumoral multidrug-resistant cells, pumps structurally unrelated drugs out of the cells (50). Moreover, the mouse MDR protein expressed in a Saccharomyces cerevisiae mutant strain lacking the STE6 ABC protein can secrete the a mating factor, which does not have homology with drugs (117). The Tap1 and Tap2 ABC proteins recognize peptides which do not share extensive sequence homologies, since they are derived from intracellular protein breakdown, although short hydrophobic peptides are preferentially transported (100).
We will focus on the gram-negative bacterial ABC protein exporters, but several recent reviews have been written on ABC proteins (50, 59, 77).
General Characteristics of the ABC Protein Exporters.
The first protein exporter to be identified in bacteria was the α-hemolysin transporter present in some uropathogenic E. coli strains (63, 143). At present, more than 10 ABC protein exporters have been identified and they secrete a great diversity of proteins (Fig. 3). Nevertheless, the proteins using these pathways can be classified in families which share sequence and functional homologies (Fig. 3). The best-characterized families are the pore-forming toxins and the metalloproteases. Secretion of E. coli hemolysin (37, 49) and E. chrysanthemi (138) and P. aeruginosa (54) metalloproteases has been reconstituted in E. coli. The functional transport apparatus always requires three specific envelope proteins: one ABC protein and two auxiliary proteins, one in the inner membrane and one in the outer membrane. In each species the ABC transporters are specifically devoted to the secretion of only one protein or a group of isoenzymes. Moreover, in a few cases, several ABC transporters secreting unrelated exoproteins are found in a single bacterial strain.
This specificity also appears in the genetic organization of the transporters, which are usually coded for by genes linked to the structural gene of the extracellular polypeptide. For example, the four highly homologous proteases of E. chrysanthemi (45, 138) are encoded by contiguous genes clustered with the three genes encoding their common ABC transporter: PrtD, the ABC protein; PrtE, the inner membrane auxiliary protein; and PrtF, the outer membrane protein (26, 79). The alkaline protease genetic determinant of P. aeruginosa exhibits similar organization, with all genes for synthesis and secretion in a cluster (30, 54). Similarly, the B. pertussis adenylcyclase structural gene, cyaA, is linked to the three genes coding for the secretion apparatus, cyaB, cyaD, and cyaE (48). The E. coli α-hemolysin structural gene hlyA is adjacent to the hlyB and hlyD genes, which code for the ABC protein and the inner membrane auxiliary protein, respectively (36, 49). In this special case, the gene coding for the outer membrane component, tolC, is unlinked to the hly locus and is also used by another E. coli ABC transporter responsible for the secretion of colicin V (35, 137). TolC is a multifunctional protein also involved in colicin E1 permeation and chromosome segregation and could form a channel in the outer membrane selective for peptides (11, 61, 96, 139). Finally, the TolC protein can combine with two S. marcescens inner membrane proteins to form an active hybrid ABC transporter in E. coli required for the secretion of the S. marcescens hemoprotein HasA (83). HasA is an extracellular protein able to bind heme and acquire heme from hemoglobin. It is essential for the ability of S. marcescens to use heme or hemoglobin as iron sources (82). This protein has a C-terminal secretion signal but does not show sequence similarities with other proteins. The ABC protein, HasD, and the auxiliary inner membrane protein, HasE, are encoded by two adjacent genes located just downstream from the HasA structural gene in the same iron-regulated operon. In E. coli mutant strains lacking TolC, the reconstituted HasA secretion system can also combine with PrtF, the outer membrane component of the E. chrysanthemi protease transporter, to form an active system (83). Within each family, the three components of the ABC transporters are very homologous and the secretion systems are basically interchangeable (35, 53, 60, 87). Hence, these systems are highly selective but able to recognize distinct proteins. This implies the presence of common secretion signals on the exoproteins belonging to the same family.
Nature of the Secretion Signal on the Exoproteins.
The proteins secreted by this pathway do not have a cleavable N-terminal signal peptide (36). The presence of a C-terminal secretion signal located in the last 60 amino acids was first identified on α-hemolysin by using deletions and gene fusions (86). Similarly, C-terminal signals were found on other homologous toxins, such as the Proteus vulgaris and Morganella morganii hemolysins (74) and the Pasteurella hemolytica leukotoxin (149). Nevertheless, in this family the overall homologies are too high to deduce from sequence comparison that a conserved stretch of amino acids constitutes a secretion signal. Adenylcyclase of B. pertussis is also highly homologous to the hemolysin of E. coli in its C-terminal hemolysin domain. Its C-terminal secretion signal is recognized by the E. coli hemolysin transporter (128). In this case again, sequence comparison did not allow the definition of a consensus secretion signal. Site-directed mutagenesis has failed to produce single missense mutations in hlyA which significantly reduce the secretion level. However, the combination of several mutations strongly impairs secretion. This suggests that the C-terminal signal adopts a tertiary structure in which scattered key residues are brought together (66, 67, 132).
Similarly, the highly homologous metalloproteases secreted by E. chrysanthemi, S. marcescens, and P. aeruginosa were shown to have C-terminal secretion signals (25, 44, 45, 80; F. Duong, Thèse de Doctorat de l’Université Aix-Marseilles II, 1994). Using one E. chrysanthemi protease, PrtG, it was shown that the smallest C-terminal sequence allowing efficient secretion contains the last 29 residues of PrtG and that low but significant secretion can be promoted by the last 15 residues of PrtG. Moreover, the extreme C-terminal motif, consisting of a negatively charged residue followed by several hydrophobic residues, must be exposed and is conserved among the homologous proteases (46). Sequence comparison has revealed that in addition to proteases, several other proteins (Rhizobium leguminosarum NodO and S. marcescens and Pseudomonas fluorescens lipases) which can be secreted by the E. chrysanthemi or P. aeruginosa ABC transporters display a very similar C-terminal motif (31, 46). The importance of this motif is accentuated by the lack of overall sequence homology between these proteins and the protease family. Nuclear magnetic resonance study of the purified protease G C-terminal fragment has shown that it forms a stable α helix located just upstream of the last seven or eight residues (146). In conclusion, the protease and toxin secretion signals are different and specific. Heterologous complementation between transporters of proteins belonging to these two families is very low, and the biological significance of this complementation remains unclear since its dependence on the C-terminal signal has not been determined (25, 35). Nevertheless, each signal can promote secretion of foreign proteins by its specific transporter.
Versatility of the C-Terminal Secretion Signal.
Several studies using hybrid proteins carrying the HlyA carboxy-terminal signal fused to different passenger polypeptides have shown that the C-terminal secretion signal is sufficient to promote the secretion of chimeric proteins through the specific hemolysin transporter. This targeting sequence was able to secrete a large diversity of polypeptides, including proteins that exhibit enzymatic activity such as β-lactamase (15) or streptokinase (I. Kern, personal communication) and proteins that are normally cytoplasmic such as dihydrofolate reductase and chloramphenicol acetylase (15, 98). Similarly, the E. chrysanthemi C-terminal protease secretion signal can promote the specific secretion of fused passenger polypeptides (25, 84). However, the 4-amino-acid C-terminal motif is not sufficient by itself to promote secretion. The study of fusion protein secretion has revealed the role of a domain located just upstream of the C-terminal signal on most of the exoproteins. Toxins, proteases, and lipases secreted by ABC pathways all have such a domain consisting of a glycine-rich sequence (GGXGXD) that is repeated 4 to 36 times, depending on the protein. This region has been implicated in calcium binding (16), cytolytic activity, and secretion. (38). Three-dimensional structures of P. aeruginosa alkaline protease and S. marcescens metalloprotease have shown that the repeats form a β parallel roll structure which binds calcium ions (9, 10). Comparison of the secretion level of various heterologous polypeptides either fused directly to the signal or separated from it by the glycine-rich domain has shown that these repeats play a critical role in the secretion of some polypeptide passengers (84). An attractive hypothesis would be that the glycine-rich repeats act as internal chaperones, allowing better signal separation from the remainder of the protein.
In contrast to the export via the sec pathway of recombinant proteins fused to a signal peptide, heterologous protein secretion promoted by the C-terminal secretion signal is efficient and is not restricted to extracytoplasmic proteins (15). Moreover, this secretion system is presently the only one which, in E. coli, exports proteins to the extracellular medium, making recombinant protein purification easier.
ABC Protein Function.
Since the ABC protein has a nucleotide-binding site, it is likely that this protein provides energy for the translocation. The cytoplasmic half of HlyB and the whole PrtD ABC proteins were overproduced and purified. They exhibit ATPase activity in vitro (24, 75). Biochemical evidence suggested a direct interaction between the secreted polypeptide and the ABC protein. It has been shown that a fusion protein containing β-galactosidase fused to the hemolysin C-terminal signal was strongly associated with the inner membrane in the presence of the ABC protein HlyB (43). The ATPase activity of purified PrtD protein was specifically inhibited by the addition of a C-terminal protease signal (24). A genetic study, also suggesting a direct role of the ABC protein in substrate recognition, revealed the existence of mutant HlyB proteins that had acquired the ability to secrete a truncated HlyA polypeptide (150). Nevertheless, the biological significance of this experiment remains unclear since the mutant HlyA protein used lacked the last 29 amino acids in the C-terminal secretion signal.
The determination of the ABC protein function in vivo was facilitated by the identification of the S. marcescens HasA ABC transporter, which is highly homologous to the E. chrysanthemi protease transporter but dedicated to the secretion of HasA. Whereas proteases can use their own genuine transporter and the HasA transporter as well, HasA is secreted only by its specific transporter (81, 83). Functional analysis of protease and HasA secretion through hybrid transporters obtained by combining components from each system demonstrated that the ABC protein is responsible for the substrate specificity (13a).
Functions of the Two Helper Proteins.
The inner membrane helper protein belongs to a novel family of transport accessory proteins, found mostly in gram-negative bacteria, which function in conjunction with membrane transporters such as ABC proteins or drug antiporters (28). Their structural genes are adjacent to the genes coding for the energy-providing proteins (ABC or antiporter proteins) (85). They are involved in the export of peptides, proteins, drugs, metal cations, and oligosaccharides. The membrane topology of these proteins suggests that they may be involved in localized fusion between the two membranes. Indeed, sequence structure predictions indicate that they have an N-terminal hydrophobic segment probably anchored in the inner membrane, a large hydrophilic segment presumably located in the periplasmic space, and a β-sheeted structural C-terminal domain which could interact with the outer membrane. This protein family was recently termed the membrane fusion protein family (MFP) (28). In the case of the hemolysin transporter, it was shown that the inner membrane auxiliary protein HlyD has a large periplasmic domain and is found associated with the inner and outer membranes, after membrane separation, while a truncated HlyD protein lacking the last 10 residues fractionates exclusively with the inner membrane (127). This suggests a role of the C terminus in the interaction with the outer membrane.
An outer membrane helper protein is always required in the ABC protein transporters. Similarly, some drug efflux systems also have a specific outer membrane component coded for by a gene clustered with the other two (108).
The absence of periplasmic intermediates in mutants lacking the outer membrane component (S. Létoffé, personal communication) suggests that this protein does not function independently of the other two and does not form a nonspecific pore. Actually, there is no evidence that the outer membrane component could recognize the substrates (81, 83). It could interact with the inner membrane helper protein (13a). It appears, therefore, that interaction between the three components of the transporter could be required for the substrate translocation, in one step, across both membranes.
On the contrary, in the mammalian MDR system, it was shown that the purified MDR protein binds drugs and ATP analogs and exhibits ATPase activity in vitro. The purified protein, reconstituted in liposomes, is able to translocate drugs in the presence of ATP by itself (50). Moreover, mouse MDR expression in E. coli confers resistance to tetraphenylphosphonium and tetraphenylarsenium, two MDR substrates to which E. coli is sensitive. Resistance resulted from an increased efflux of these drugs, indicating that the MDR protein is functional in E. coli and directly involved in export (13). However, these experiments do not demonstrate that the MDR protein alone promotes drug efflux across both membranes in E. coli. It is likely that it interacts with as yet unidentified E. coli helper proteins, one inner membrane MFP and one outer membrane protein.
At present, it appears that protein secretion by ABC transporters is widespread among prokaryotes and eukaryotes. However, several major questions remain concerning the definition of the steps requiring ATP hydrolysis and/or membrane potential and the form of the protein substrate competent for the transporter. It is possible that unfolding followed by refolding is required and promoted by as yet unidentified proteins common to the various bacteria or provided by the transporter itself after recognition of the signal. In contrast, studies with the mammalian MDR proteins are more advanced since drug efflux has been reconstituted in liposomes containing purified MDR proteins. Nevertheless, in this case, the physiologically important cellular substrates for MDR transporters are still unknown and need to be identified. Similarly, the function of the yeast mitochondrial ABC protein remains unclear (78), even though it is the first ABC protein known to be required for normal growth.
Several extracellular proteins of plant and animal pathogens are secreted through a new, recently discovered pathway first identified in the virulent Yersinia species Y. pestis, Y. enterocolitica, and Y. pseudotuberculosis. All three species possess a common 70-kb virulence plasmid which promotes the synthesis and secretion of 11 anti-host proteins called Yops (for a review, see references 21 and 41). The term Yops originally signified Yersinia outer membrane proteins (109), but it is now well established that these proteins are secreted to culture supernatants (21). Loss of the virulent plasmid or disruption of individual yop genes makes the bacteria avirulent. However, functions have been assigned to only a few of the proteins. YopH and YopE are cytotoxins involved in interference with phagocytosis. YopH has a tyrosine phosphatase activity, and YopE disrupts the actin microfilament structure. The YopM protein interacts with thrombin and inhibits platelet aggregation. The YopH and YopE proteins are not cytotoxic when added to culture cells. Upon infection, bacteria adhere to the cell surface and transfer the toxins across the host cell plasma membrane. Translocation of YopE and YopH into host cells requires their secretion and also the secretion of YopB and YopD, which could act as translocators for YopE and YopH (131). The presence of adhesin at the cell surface is also required, but not the internalization of the bacteria themselves.
All of the Yop proteins are secreted by the same secretory apparatus also coded for by the virulence plasmid. The proteins do not have N-terminal signal peptides (94). Secretion is independent of the general export pathway. Study of protein fusions showed that hybrid proteins, with the amino-terminal part of the Yop proteins fused to alkaline phosphatase lacking its signal peptide, are efficiently secreted to the extracellular medium. These experiments localized the secretion signals within the 50 to 100 N-terminal residues of several Yop proteins (92). Sequence comparison of these domains failed to reveal any similarity, suggesting that the recognition of the common secretory apparatus is not via direct interaction of the secretion domain with the apparatus. Very recently, three small cytoplasmic proteins, SycD, SycE, and SycH, were identified. They are coded for by genes adjacent to yopD, yopE, and yopH, respectively. Each is specifically involved in the secretion of the corresponding Yop protein and in vitro can bind its N-terminal domain carrying the secretion signal (140, 141). It was proposed that the Syc proteins constitute individual chaperones which guide the Yop proteins to the secretion machinery. However, this just pushes the same problem back one step: how are the various Syc proteins able to recognize a common transporter? Sequence analysis does not reveal a common motif on these Syc proteins.
Secretion of Yops occurs by a transporter consisting of more than 20 secretion proteins. Three operons are involved in the secretion process: the virA, virB, and virC operons (93). Individual functions of these secretion proteins have not yet been identified, with the exception of one virB protein which could be a cytoplasmic ATPase (145). Another animal pathogen, Shigella flexneri, secretes proteins into the extracellular medium by a similar pathway. IpaB, -C, and -D proteins (for invasion plasmid antigens) are plasmid-encoded invasins. They form a multiprotein complex in the extracellular medium essential for phagocytosis and escape from the phagosome. They do not have signal peptides and are secreted via a specific transporter coded for by plasmid mxi and spa genes, sharing homology with the ysc genes (12). When grown in culture, most of the Ipa proteins are not secreted but stored in the cytoplasm. Formation of cytoplasmic Ipa complexes is prevented by a 17-kDa protein which is not secreted and binds the IpaB and IpaC proteins in the cytoplasm (89). This protein could act as a specific Ipa chaperone, like the Y. enterocolitica Syc proteins. Interaction with cultured epithelial cells or addition of bovine fetal serum provides the signal for Ipa secretion. In the absence of IpaB or IpaD, the remaining Ipa proteins are secreted even in the absence of the secretion inducer. The transporters seem to be assembled in the cells even in the absence of the signal, but they are jammed by IpaB and IpaD proteins, which act as antisecretion factors. When the signal is provided, the antisecretion proteins are unhooked and secretion occurs (88). This modulation of secretion by several exoproteins, themselves substrates of the transporter, is unique and could allow rapid delivery of large amounts of presynthesized invasins, if necessary.
A homologous genetic region was found on the chromosome of Salmonella typhimurium (official designation, Salmonella enterica serovar Typhimurium). It is involved in epithelial cell invasion. Moreover, an S. typhimurium invasion mutant can be rescued by complementation with the S. flexneri secretion genes (52). This suggests that the newly identified genes of S. typhimurium actually code for a secretion apparatus, even though the Salmonella extracellular proteins involved in invasion remain to be identified. To date, only one extracellular S. typhimurium protein has been identified. This protein, InvJ, lacks a typical N-terminal signal sequence (19).
Large gene clusters homologous to the Yersinia ysc and Shigella spa and mxi regions were recently identified in several plant pathogens such as P. solanacearum, X. campestris, Pseudomonas syringae, and Erwinia amylovora (for a review, see reference 135). Genes belonging to two such clusters were shown to be required for the secretion of proteins lacking signal peptides. In one case, it is a hairpin protein required for disease development on compatible plants and for induction of a hypersensitive response in incompatible plants (55). The second exoprotein secreted by this pathway, PopA1, is only required for the hypersensitive reaction (4).
During the last decade, evidence has accumulated that there is no general mechanism analogous to the GEP for translocating proteins across the outer membrane. At present, five secretion mechanisms have been identified. Three of them are two-step processes initiated by translocation across the inner membrane via the GEP. The self-promoted protease extracellular secretion is well elucidated. It involves incorporation in the outer membrane of a helper domain and subsequent proteolytic cleavage to release the extracellular polypeptide from the membrane. Nevertheless, the physiological use of such a simple secretion mechanism is not known: is it a pathway restricted to certain proteases? S. marcescens hemolysin secretion is also quite simple, based on a requirement for only a few accessory proteins. It functions with one helper protein located in the outer membrane also required for hemolysin activation. The nature of the modification introduced by the helper protein is still unknown. This is the only secretion mechanism in which chemical modification leading to activation is a prerequisite for secretion. In this case also, it is not known whether this mechanism is restricted to a few hemolysins. On the contrary, there is no doubt that GSP is the most widespread secretion pathway present in plant and animal pathogens, secreting hydrolytic enzymes as well as toxins. Unfortunately, it is poorly understood at present. The nature of the signal on the secreted proteins, the molecular mechanism of protein sorting, and the individual roles of the 14 helper proteins, most of them located in the cytoplasmic membrane, remain open questions. Among the two Sec-independent secretion pathways presently identified, the ABC pathway appears to be ubiquitous, found in organisms ranging from bacteria to humans. In bacteria it secretes a large diversity of proteins. The secretion steps are beginning to be elucidated. The basic feature of this secretion mechanism is that the exoproteins recognize the inner membrane ABC protein and cross both membranes in one step at localized membrane fusion points created by a specific interaction between the two helper proteins. However, the major element of this model, the existence of adhesion zones between membranes, postulated more than 20 years ago still remains to be demonstrated. The last secretion pathway also occurs in many gram-negative bacterial pathogens. In this very recently discovered pathway, the components of the transporter have not all been identified and localized in the cell. This pathway exhibits some very peculiar properties, such as the existence of specific chaperones and the secretion modulation by blocking the transporters. This secretion modulation, which appears to be different in the various pathogens, could be important in the distinct strategies developed by these bacteria for entry into host cells.
I am grateful to many colleagues for communicating unpublished results and for their help, especially to Andrée Lazdunski, Tony Pugsley, Philippe Delepelaire, and Richard D’Ari.
References
1. Aitchison, J. D., W. M. Nuttley, R. K. Szilard, A. M. Brade, J. R. Glover, and R. A. Rachubinski. 1992. Peroxisome biogenesis in yeast. Mol. Microbiol. 6:3455–3460.
2. Akrim, M., M. Bally, G. Ball, J. Tommassen, A. Filloux, and A. Lazdunski. 1993. Xcp mediated protein secretion in Pseudomonas aeruginosa: identification of two additional genes and evidence for regulation of xcp gene expression. Mol. Microbiol. 10:431–443.
3. Andro, T., J. P. Chambost, A. Kotoujanski, J. Cattaneo, Y. Bertheau, F. Barras, F. Van Gijsegen, and A. Coleno. 1984. Mutants of Erwinia chrysanthemi defective in secretion of pectinase and cellulase. J. Bacteriol. 160:1199–1203.
4. Arlat, M., F. Van Gijsegem, J. C. Huet, J. C. Pernollet, and C. A. Boucher. 1994. PopA1, a protein which induces a hypersensitivity-like response on specific Petunia genotypes, is secreted via the Hrp pathway of Pseudomonas solanacearum. EMBO J. 13:543–553.
5. Bally, M. 1994. La voie générale de sécrétion chez Pseudomonas aeruginosa et d’autres bacteries a Gram négatif. Cah. Imabio 77:65–71.
6. Bally, M., A. Fillioux, M. Akrim, G. Ball, A. Lazdunski, and J. Tommassen. 1992. Protein secretion in Pseudomonas aeruginosa—characterization of 7 xcp genes and processing of secretory apparatus components by prepilin peptidase. Mol. Microbiol. 6:1121–1131.
7. Balzi, E., M. Wang, S. Leterme, L. Van Dyck, and A. Goffeau. 1994. PDR5, a novel yeast multidrug resistance conferring transporter controlled by the transcription regulator PDR1. J. Biol. Chem. 269:2206–2214.
8. Bardwell, J. C., K. MacGovern, and J. Beckwith. 1991. Identification of a protein required for disulfide bond formation in vivo. Cell 67:581–589.
9. Baumann, U. 1994. Crystal structure of the 50kDa metallo protease from Serrratia marcescens. J. Mol. Biol. 242:244–251.
10. Baumann, U., S. Wu, K. M. Flaherty, and D. B. McKay. 1993. Three-dimensional structure of the alkaline protease of Pseudomonas aeruginosa: a two-domain protein with a calcium binding parallel beta roll motif. EMBO J. 72:3357–3364.
11. Benz, R., E. Maier, and I. Genrschev. 1993. TolC of Escherichia coli functions as an outer membrane channel. Zentralbl. Bakteriol. 278:186–196.
12. Bergman, T., K. Erickson, E. Galyov, C. Persson, and H. Wolf-Watz. 1994. The lcrB (yscN/U) gene cluster of Yersinia pseudotuberculosis is involved in Yop secretion and shows high homology to the spa gene clusters of Shigella flexneri and Salmonella typhimurium. J. Bacteriol. 176:2619–2626.
13. Bibi, E., P. Gros, and H. R. Kaback. 1993. Functional expression of mouse mdr1 in Escherichia coli. Proc. Natl. Acad. Sci. USA 90:9209–9213.
13a. Binet, R., and C. Wandersman. 1995. Protein secretion by hybrid bacterial ABC-transporters: specific functions of the membrance ATPase and the membrane fusion protein. EMBO J. 14:2298–2306.
14. Bissinger, P. H., and K. Kuchler. 1994. Molecular cloning and expression of the Saccharomyces cerevisiae STS1 gene product. J. Biol. Chem. 269:4180–4186.
15. Blight, M. A., and B. Holland. 1994. Heterologous protein secretion and the versatile Escherichia coli haemolysin translocator. Trends Biotechnol. 12:450–455.
16. Boehm, D. F., R. A. Welch, and I. S. Snyder. 1990. Domains of Escherichia coli hemolysin (HlyA) involved in binding of calcium and erythrocyte membranes. Infect. Immun. 58:1959–1964.
17. Bortoli-German, I., E. Brun, B. Py, M. Chippaux, and F. Barras. 1994. Periplasmic disulphide bond formation is essential for cellulase secretion by the plant pathogen Erwinia chrysanthemi. Mol. Microbiol. 11:545–553.
18. Braun, V., R. Schönherr, and S. Hobbie. 1993. Enterobacterial hemolysins: activation, secretion and pore formation. Trends Microbiol. 1:211–216.
19. Collazo, C., M. Zierler, and J. E. Galan. 1995. Functional analysis of the Salmonella typhimurium invasion genes invI and invJ and identification of a target of the protein secretion apparatus encoded by the inv locus. Mol. Microbiol. 15:25–38.
20. Collmer, A., C. Schoedel, D. Roeder, J. L. Ried, and J. L. Rissler. 1985. Molecular cloning in Escherichia coli of Erwinia chrysanthemi genes encoding multiple forms of pectate lyase. J. Bacteriol. 161:913–920.
21. Cornelis, G. R., T. Biot, C. Lambert, T. Michiels, B. Mulder, C. Sluiters, M. P. Sory, M. Van Bouchaute, and J. C. Vanooteghen. 1989. The Yersinia Yop regulon. Mol. Microbiol. 3:1455–1459.
22. Cowan, S. W., T. Schirmer, G. Rummel, M. Steiert, R. Ghosh, R. A. Pauptit, J. N. Jansonius, and J. P. Rosenbush. 1992. Crystal structures explain functional properties of two E. coli porins. Nature (London) 358:727–733.
23. De Groot, A., A. Filloux, and J. Tommassen. 1991. Conservation of xcp genes, involved in the two-step protein secretion process, in different Pseudomonas species and other gram-negative bacteria. Mol. Gen. Genet. 229:278–284.
24. Delepelaire, P. 1994. PrtD, the integral membrane ATP-binding cassette component of the Erwinia chrysanthemi metalloprotease secretion system, exhibits a secretion signal-regulated ATPase activity. J. Biol. Chem. 269:27952–27957.
25. Delepelaire, P., and C. Wandersman. 1990. Protein secretion in Gram-negative bacteria. The extracellular metalloprotease B from Erwinia chrysanthemi contains a C terminal secretion signal analogous to that of Escherichia coli α-haemolysin. J. Biol. Chem. 265:9083–9089.
26. Delepelaire, P., and C. Wandersman. 1991. Characterization, localisation and transmembrane organization of the three proteins PrtD, PrtE and PrtF necessary for protease secretion by the Gram-negative bacterium Erwinia chrysanthemi. Mol. Microbiol. 5:2427–2434.
27. D’Enfert, C., A. Ryter, and A. P. Pugsley. 1987. Cloning and expression in Escherichia coli of the Klebsiella pneumoniae genes for production surface localization and secretion of the lipoprotein pullulanase. EMBO J. 6:3531–3538.
28. Dinh, T., I. T. Paulsen, and M. H. J. Saier, Jr. 1994. A family of extracytoplasmic proteins that allow transport of large molecules across the outer membranes of gram-negative bacteria. J. Bacteriol. 176:3825–3831.
29. Dums, F., J. M. Dow, and M. J. Daniels. 1991. Structural characterisation of protein secretion genes of the bacterial phytopathogen Xanthomonas campestris pathovar campestris: relatedness to secretion systems of other gram-negative bacteria. Mol. Gen. Genet. 229:357–364.
30. Duong, F., A. Lazdunski, B. Cami, and M. Murgier. 1992. Nucleotide sequence of a cluster of genes involved in synthesis and secretion of alkaline protease in Pseudomonas aeruginosa: relationships to other secretion pathways. Gene 121:47–54.
31. Duong, F., C. Soscia, A. Lazdunski, and M. Murgier. 1994. The Pseudomonas fluorescens lipase has a C-terminal secretion signal and is secreted by a three-component bacterial ABC-exporter system. Mol. Microbiol. 11:1117–1126.
32. Dupuy, B., and A. P. Pugsley. 1994. Type IV prepilin peptidase gene of Neisseria gonorrhoeae MS11: presence of a related gene in other piliated and nonpiliated Neisseria strains. J. Bacteriol. 176:1323–1331.
33. Dupuy, B., M. K. Taha, O. Possot, C. Marchal, and A. P. Pugsley. 1992. PulO, a component of the pullulanase secretion pathway of Klebsiella oxytoca, correctly and efficiently processes gonococcal type IV prepilin in Escherichia coli. Mol. Microbiol. 6:1887–1894.
34. Fath, M. J., and R. Kolter. 1993. ABC transporters: bacterial exporters. Microbiol. Rev. 57:995–1017.
35. Fath, M. J., R. C. Skvirsky, and R. Kolter. 1991. Functional complementation between bacterial MDR-like export systems: colicin V, alpha-hemolysin, and Erwinia proteases. J. Bacteriol. 173:7549–7556.
36. Felmlee, T., S. Pellett, E.-Y. Lee, and R. A. Welch. 1985. Escherichia coli hemolysin is released extracellularly without cleavage of a signal peptide. J. Bacteriol. 163:88–93.
37. Felmlee, T., S. Pellett, and R. A. Welch. 1985. Nucleotide sequence of an Escherichia coli chromosomal hemolysin. J. Bacteriol. 163:94–105.
38. Felmlee, T., and R. A. Welch. 1988. Alterations of amino acid repeats in the Escherichia coli hemolysin affect cytolytic activity and secretion. Proc. Natl. Acad. Sci. USA 85:5269–5273.
39. Filloux, A., M. Bally, G. Ball, M. Akrim, T. Tommassen, and A. Lazdunski. 1990. Protein secretion in Gram-negative bacteria: transport across the outer membrane involves common mechanism in different bacteria. EMBO J. 9:4323–4329.
40. Filloux, A., M. Murgier, B. Wretlind, and A. Lazdunski. 1987. Characterization of two Pseudomonas aeruginosa mutants with defective secretion of extracellular proteins and comparison with other mutants. FEMS Microbiol. Lett. 40:159–163.
41. Forsberg, A., R. Rosqvist, and H. Wolf-Watz. 1994. Regulation and polarized transfer of the Yersinia outer proteins (Yops) involved in anti-phagocytosis. Trends Microbiol. 2:14–20.
42. Frenken, L. G. J., A. De Groot, J. Tommassen, and C. T. Verrips. 1993. Role of the Lip B gene product in the folding of the secreted lipase of Pseudomonas glumae. Mol. Microbiol. 9:591–599.
43. Gentschev, I., and W. Goebel. 1992. Topological and functional studies on HlyB of Escherichia coli. Mol. Gen. Genet. 232:40–48.
44. Ghigo, J. M., and C. Wandersman. 1992. Cloning, nucleotide sequence and characterization of the gene encoding the Erwinia chrysanthemi B374 metalloprotease: a third metalloprotease secreted via a C-terminal secretion signal. Mol. Gen. Genet. 236:135–144.
45. Ghigo, J. M., and C. Wandersman. 1992. A 4th metalloprotease gene in Erwinia chrysanthemi. Res. Microbiol. 143:857–867.
46. Ghigo, J. M., and C. Wandersman. 1994. A carboxyl-terminal four-amino acid motif is required for secretion of the metalloprotease PrtG through the Erwinia chrysanthemi protease secretion pathway. J. Biol. Chem. 269:8979–8985.
47. Gilmore, M. S., R. O. Segarra, M. C. Booth, C. Bogie, L. R. Hall, and D. B. Clewell. 1994. Genetic structure of the Enterococcus faecalis plasmid pAD1-encoded cytolytic toxin system and its relationship to lantibiotic determinants. J. Bacteriol. 176:7335–7344.
48. Glaser, P., H. Sakamoto, J. Bellalou, A. Ullmann, and A. Danchin. 1988. Secretion of cyclolysin, the calmodulin-sensitive adenylate cyclase-haemolysin bifunctional protein of Bordetella pertussis. EMBO J. 7:3997–4004.
49. Goebel, W., and J. Hedgpeth. 1982. Cloning and functional charaterization of the plasmid-encoded hemolysin determinant of Escherichia coli. J. Bacteriol. 151:1290–1298.
50. Gottesman, M. M., and I. Pastan. 1993. Biochemistry of multidrug resistance mediated by the multidrug transporter. Annu. Rev. Biochem. 62:385–427.
51. Grodberg, J., and J. J. Dunn. 1988. ompT encodes the Escherichia coli outer membrane protease that cleaves T7 RNA polymerase during purification. J. Bacteriol. 170:1245–1253.
52. Groisman, E., and H. Ochman. 1993. Cognate gene clusters govern invasion of host epithelial cells by Salmonella typhimurium and Shigella flexneri. EMBO J. 12:3779–3787.
53. Guzzo, J., F. Duong, C. Wandersman, M. Murgier, and A. Lazdunski. 1991. The secretion genes of Pseudomonas aeruginosa alkaline protease are functionally related to those of Erwinia chrysanthemi proteases and Escherichia coli α-haemolysin. Mol. Microbol. 5:447–453.
54. Guzzo, J., M. Murgier, A. Filloux, and A. Lazdunski. 1990. Cloning of the Pseudomonas aeruginosa alkaline protease gene and secretion of the protease into the medium by Escherichia coli. J. Bacteriol. 172:942–948.
55. He, S., H. C. Huang, and A. Collmer. 1993. Pseudomonas syringae pv syringae hairpinpss: a protein that elicits the hypersensitive response in plants. Cell 73:1255–1266.
56. He, S. Y., M. Lindeberg, A. Chatterjee, and A. Collmer. 1991. Cloned Erwinia chrysanthemi out genes enable Escherichia coli to selectively secrete a diverse family of heterologous proteins to its milieu. Proc. Natl. Acad. Sci. USA 88:1079–1083.
57. He, S. Y., C. Schoedel, A. K. Chatterjee, and A. Collmer. 1991. Extracellular secretion of pectate lyase by the Erwinia chrysanthemi Out pathway is dependent upon Sec-mediated export across the inner membrane. J. Bacteriol. 173:4310–4317.
58. Higgins, C. F. 1992. ABC transporters—from microorganisms to man. Annu. Rev. Cell Biol. 8:67–113.
59. Higgins, C. F., and M. Gottesman. 1992. Is the multidrug transporter a flippase? Trends Biochem. Sci. 17:18–21.
60. Highlander, S. K., M. J. Engler, and G. M. Weinstock. 1990. Secretion and expression of the Pasteurella haemolytica leukotoxin. J. Bacteriol. 172:2343–2350.
61. Hiraga, S., H. Niki, T. Ogura, C. Ichinose, H. Mori, B. Ezaki, and A. Jaffé. 1989. Chromosome partitioning in Escherichia coli: novel mutants producing anucleate cells. J. Bacteriol. 171:1496–1505.
62. Huang, J. Z., and M. A. Schell. 1992. Role of the two-component leader sequence and mature amino acid sequences in extracellular export of endoglucanase EGL from Pseudomonas solanacearum. J. Bacteriol. 174:1314–1323.
63. Hull, S. I., R. A. Hull, B. Minshew, and S. Falkow. 1982. Genetics of hemolysin of Escherichia coli. J. Bacteriol. 151:1006–1011.
64. Keen, N. T., D. Dahlbeck, B. Staskawicz, and W. Belser. 1985. Molecular cloning of pectate lyase genes from Erwinia chrysanthemi and their expression in Escherichia coli. J. Bacteriol. 159:825–831.
65. Kelly, A., S. H. Powis, L. A. Kerr, I. Mockridge, T. Elliott, J. Bastin, B. Uchanska-Ziegler, A. Ziegler, J. Trowsdale, and A. Towsend. 1992. Assembly and function of the two ABC transporter proteins encoded in the human major histocompatibility complex. Nature (London) 355:641–642.
66. Kenny, B., C. Chervaux, and I. B. Holland. 1994. Evidence that residues –15 to –46 of the haemolysin secretion signal are involved in early steps in secretion leading to recognition of the translocator. Mol. Microbiol. 11:99–109.
67. Kenny, B., S. Taylor, and I. B. Holland. 1992. Identification of individual amino acids required for secretion within the haemolysin (HlyA) C-terminal targeting region. Mol. Microbiol. 6:1477–1489.
68. Kessler, E., and M. Safrin. 1988. Synthesis, processing, and transport of Pseudomonas aeruginosa elastase. J. Bacteriol. 170:5241–5247.
69. Klauser, T., J. Kramer, K. Otzelberger, J. Pohlner, and T. F. Meyer. 1993. Characterization of the Neisseria Iga beta core, the essential unit for outer membrane targeting and extracellular protein secretion. J. Mol. Biol. 234:579–593.
70. Klauser, T., J. Pohlner, and T. F. Meyer. 1992. Selective extracellular release of cholera toxin-B subunit by Escherichia coli—dissection of Neisseria Iga beta-mediated outer membrane transport. EMBO J. 11:2327–2333.
71. Klauser, T., J. Pohlner, and T. F. Meyer. 1993. The secretion pathway of IgA protease-type proteins in Gram-negative bacteria. Bioessays 15:799–805.
72. Kontinen, V. P., P. Saris, and M. Sarvas. 1991. A gene prsA of Bacillus subtilis involved in a novel late stage of protein export. Mol. Microbiol. 5:1273–1283.
73. Kornacker, M. G., and A. P. Pugsley. 1990. The normally periplasmic enzyme β lactamase is specifically and efficiently translocated through the Escherichia coli outer membrane when it is fused to the cell surface enzyme pullulanase. Mol. Microbiol. 4:1101–1109.
74. Koronakis, V., M. Cross, B. Senior, E. Koronakis, and C. Hughes. 1987. The secreted hemolysins of Proteus mirabilis, Proteus vulgaris, and Morganella morganii are genetically related to each other and to the alpha-hemolysin of Escherichia coli. J. Bacteriol. 169:1509–1515.
75. Koronakis, V., C. Hughes, and E. Koronakis. 1993. ATPase activity and ATP/ADP-induced conformational change in the soluble domain of the bacterial protein translocator HlyB. Mol. Microbiol. 8:1163–1175.
76. Kuchler, K. 1989. Saccharomyces cerevisiae STE6 gene product: a novel pathway for protein export in eukaryotic cells. EMBO J. 8:3973–3984.
77. Kuchler, K. 1993. Unusual routes of protein secretion: the easy way out. Trends Cell Biol. 3:421–426.
78. Leighton, J., and G. Schatz. 1995. An ABC transporter in the mitochondrial inner membrane is required for normal growth of yeast. EMBO J. 14:188–195.
79. Létoffé, S., P. Delepelaire, and C. Wandersman. 1990. Protease secretion by Erwinia chrysanthemi: the specific secretion functions are analogous to those of E. coli α-haemolysin. EMBO J. 9:1375–1382.
80. Létoffé, S., P. Delepelaire, and C. Wandersman. 1991. Cloning and expression in Escherichia coli of the Serratia marcescens metalloprotease gene: secretion of the protease from E. coli in the presence of the Erwinia chrysanthemi protease secretion functions. J. Bacteriol. 173:2160–2166.
81. Létoffé, S., J. M. Ghigo, and C. Wandersman. 1993. Identification of two components of the Serratia marcescens metalloprotease transporter: protease SM secretion in Escherichia coli is TolC dependent. J. Bacteriol. 175:7321–7328.
82. Létoffé, S., J. M. Ghigo, and C. Wandersman. 1994. Iron acquisition from heme and hemoglobin by Serratia marcescens extracellular protein. Proc. Natl. Acad. Sci. USA 91:9876–9880.
83. Létoffé, S., J. M. Ghigo, and C. Wandersman. 1994. Secretion of the Serratia marcescens HasA protein by an ABC transporter. J. Bacteriol. 176:5372–5377.
84. Létoffé, S., and C. Wandersman. 1992. Secretion of CyaA-PrtB and HlyA-PrtB fusion proteins in Escherichia coli: involvement of the glycine-rich repeat domain of Erwinia chrysanthemi protease B. J. Bacteriol. 174:4920–4927.
85. Lewis, K. 1994. Multidrug resistance pumps in bacteria: variations on a theme. Trends Biochem. Sci. 19:119–123.
86. Mackman, N., K. Baker, L. Gray, and I. B. Holland. 1987. Release of a chimeric protein into the medium from Escherichia coli using the C-terminal secretion signal of haemolysin. EMBO J. 6:2835–2841.
87. Masure, H. R., D. C. Au, M. K. Gross, M. G. Donovan, and D. R. Storm. 1990. Secretion of the Bordetella pertussis adenylate cyclase from Escherichia coli containing the hemolysin operon. Biochemistry 29:140–145.
88. Ménard, R., P. Sansonetti, and C. Parsot. 1994. The secretion of the Shigella flexneri Ipa invasins is activated by epithelial cells and controlled by IpaB and IpaD. EMBO J. 13:5293–5302.
89. Ménard, R., P. Sansonetti, C. Parsot, and T. Vasselon. 1994. Extracellular association and cytoplasmic partitioning of the IpaB and IpaC invasins of S. flexneri. Cell 79:515–525.
90. Mergeay, M. 1991. Towards an understanding of the genetics of bacterial metal resistance. Trends Biotechnol. 9:17–24.
91. Michaelis, S., C. Chapon, C. D’Enfert, A. P. Pugsley, and M. Schwartz. 1985. Characterization and expression of the structural gene for pullulanase, a maltose-inducible secreted protein of Klebsiella pneumoniae. J. Bacteriol. 164:633–638.
92. Michiels, T., and G. R. Cornelis. 1991. Secretion of hybrid proteins by the Yersinia Yop export system. J. Bacteriol. 173:1677–1685.
93. Michiels, T., J.-C. Vanooteghem, C. L. L. de Rouvroit, B. China, A. Gustin, P. Boudry, and G. R. Cornelis. 1991. Analysis of virC, an operon involved in the secretion of Yop proteins by Yersinia enterocolitica. J. Bacteriol. 173:4994–5009.
94. Michiels, T., P. Wattiau, R. Brasseur, J.-M. Ruysschaert, and G. Cornelis. 1990. Secretion of Yop proteins by yersiniae. Infect. Immun. 58:2840–2849.
95. Miyazaki, H., N. Yanagida, S. Horinouchi, and T. Beppu. 1989. Characterization of the Serratia marcescens serine protease and COOH-terminal processing of the precursor during its excretion through the outer membrane of Escherichia coli. J. Bacteriol. 171:6566–6572.
96. Morona, R., P. A. Manning, and P. Reeves. 1983. Identification and characterization of the TolC protein, an outer membrane protein from Escherichia coli. J. Bacteriol. 153:693–699.
97. Mosser, J., A. M. Douar, C. O. Sarde, P. Kioskis, R. Feil, H. Moser, A. M. Pouska, J. L. Mandel, and P. Allbourg. 1993. Putative X-linked adrenoleukodystrophy gene shares unexpected homology with ABC transporters. Nature (London) 361:726–727.
98. Nakano, H., Y. Kawakami, and H. Nishimura. 1992. Secretion of genetically-engineered dihydrofolate reductase from Escherichia coli using an E. coli alpha-hemolysin membrane translocation system. Appl. Microbiol. Biotechnol. 37:765–771.
99. Nakata, N., T. Tobe, I. Fukuda, T. Susuki, K. Komatsu, M. Yoshikava, and C. Sasakava. 1993. The absence of a surface protease, OmpT, determines the intercellular spreading of Shigella: relationship between the ompT and kcpA loci. Mol. Microbiol. 9:459–968.
100. Neefjes, J. J., F. Momburg, and G. J. Hämmerling. 1993. Selective and ATP-dependent translocation of peptides by the MHC-encoded transporter. Science 261:769–771.
101. Nies, D. H. 1992. Resistance to cadmium, cobalt, zinc and nickel in microbes. Plasmids 27:17–28.
102. Nunn, D., and S. Lory. 1992. Components of the protein excretion apparatus of Pseudomonas aeruginosa are processed by the prepilin peptidase. Proc. Natl. Acad. Sci. USA 89:47–51.
103. Nunn, D. N., and S. Lory. 1993. Cleavage, methylation and localization of the Pseudomonas aeruginosa export proteins XcpT, -U, -V, and -W. J. Bacteriol. 175:4375–4382.
104. Ondraczek, R., S. Hobbie, and V. Braun. 1992. In vitro activation of the Serratia marcescens hemolysin through modification and complementation. J. Bacteriol. 174:5086–5094.
105. Ortiz, D. F., L. Kreppel, D. Speiser, G. Sheel, G. MacDonald, and D. W. Ow. 1992. Heavy metal tolerance in the fission yeast requires an ATP binding cassette-type vacuolar membrane transporter. EMBO J. 11:3491–3499.
106. Overbye, J. O., M. Sandkvist, and M. Bagdasarian. 1993. Genes required for extracellular secretion of enterotoxin are clustered in Vibrio cholerae. Gene 132:101–106.
107. Pohlner, J., R. Halter, K. Beyreuther, and T. F. Meyer. 1987. Gene structure and extracellular secretion of Neisseria gonorrhoeae IgA protease. Nature (London) 325:458–462.
108. Poole, K., K. Krebes, C. McNally, and S. Neshat. 1993. Multiple antibiotic resistance in Pseudomonas aeruginosa: evidence for involvement of an efflux operon. J. Bacteriol. 175:7363–7372.
109. Portnoy, D. A., S. L. Moseley, and S. Falkow. 1981. Characterization of plasmids and plasmid-associated determinants of Yersinia enterocolitica pathogenesis. Infect. Immun. 31:775–782.
110. Possot, O., and A. P. Pugsley. 1994. Molecular characterization of PulE, a protein required for pullulanase secretion. Mol. Microbiol. 12:287–299.
111. Pugsley, A. P. 1992. Translocation of a folded protein across the outer membrane in Escherichia coli. Proc. Natl. Acad. Sci. USA 89:12058–12060.
112. Pugsley, A. P. 1993. The complete general secretory pathway in gram-negative bacteria. Microbiol. Rev. 57:50–108.
113. Pugsley, A. P. 1993. Processing and methylation of PulG, a pilin-like component of the general secretory pathway of Klebsiella oxytoca. Mol. Microbiol. 9:295–308.
114. Pugsley, A. P., and B. Dupuy. 1992. An enzyme with type IV prepilin peptidase activity is required to process a component of the general extracellular protein secretion pathway of Klebsiella oxytoca. Mol. Microbiol. 6:751–760.
115. Pugsley, A. P., and O. Possot. 1993. The general secretory pathway of Klebsiella oxytoca: no evidence for relocalization or assembly of pilin like PulG protein into a multiprotein complex. Mol. Microbiol. 10:665–674.
116. Py, B., M. Chippaux, and F. Barras. 1993. Mutagenesis of cellulase Egz for studying the general protein secretory pathway in Erwinia chrysanthemi. Mol. Microbiol. 7:785–793.
117. Raymond, M., P. Gros, M. Whiteway, and D. Y. Thomas. 1992. Functional complementation of yeast ste6 by a mammalian multidrug resistance mdr gene. Science 256:232–234.
118. Reeves, P. J., D. Whitcombe, M. Gibson, G. Allison, N. Bunce, R. Barallon, P. Douglas, V. Mulholland, S. Stevens, D. Walker, and G. P. C. Salmond. 1993. Molecular cloning and characterization of 13 out genes from Erwinia carotovora subspecies carotovora: genes encoding members of a general secretory pathway (GSP) widespread in Gram negative bacteria. Mol. Microbiol. 8:443–456.
119. Reverchon, S., N. Hugouvieux-Cotte-Pattat, and J. Robert-Baudouy. 1985. Cloning of genes encoding pectolytic enzymes from a genomic library of phytopathogenic bacterium, Erwinia chrysanthemi. Gene 35:124–130.
120. Riordan, J. R., J. M. Romens, B. S. Kerem, N. Alon, R. Rozmahel, Z. Grzelczak, J. Zielenski, S. Lok, N. Plavsic, J. L. Chou, M. L. Drumm, M. C. Iannuzzi, F. S. Collins, and L. C. Tsui. 1989. Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science 245:1066–1073.
121. Russel, M. 1994. Mutants at conserved positions in gene IV, a gene required for assembly and secretion of filamentous phages. Mol. Microbiol. 14:357–369.
122. Saarilahti, H. T., M. Pirhonen, M. B. Karlsson, D. Flego, and E. T. Palva. 1992. Expression of peh-bla gene fusions in Erwinia carotovora, subsp carotovora and isolation of regulatory mutants affecting polygalacturonase production. Mol. Gen. Genet. 234:81–88.
123. Sandkvist, M., and M. Bagdasarian. 1993. Suppression of temperature sensitive assembly mutants of heat labile enterotoxin B subunits. Mol. Microbiol. 10:635–645.
124. Schatz, P. J., and J. Beckwith. 1990. Genetic analysis of protein export in Escherichia coli. Annu. Rev. Genet. 24:215–248.
125. Schiebel, E., H. Schwarz, and V. Braun. 1989. Subcellular location and unique secretion of the hemolysin of Serratia marcescens. J. Cell Biol. 264:16311–16320.
126. Schmitt, W., and R. Haas. 1994. Genetic analysis of the Helicobacter pylori vacuolating cytoxin: structural similarities with the IgA protease type of exported protein. Mol. Microbiol. 12:307–319.
127. Schülein, R., I. Gentschev, J. J. Mollenkops, and W. Goebel. 1992. A topological model for the haemolysin translocator protein HlyD. Mol. Gen. Genet. 234:155–163.
128. Sebo, P., and D. Ladant. 1993. Repeat sequences in the Bordetella pertussis adenylate cyclase toxin can be recognized as alternative carboxy-proximal secretion signals by the Escherichia coli α-hemolysin translocator. Mol. Microbiol. 9:999–1009.
129. Silen, J., and D. Agard. 1989. The α lytic protease pro-region does not require a physical linkage to activate the protease domain in vivo. Nature (London) 341:462–464.
130. Simonen, M., and I. Palva. 1993. Protein secretion in Bacillus species. Microbiol. Rev. 57:109–137.
131. Sory, M. P., and G. R. Cornelis. 1994. Translocation of a hybrid YopE-adenylate cyclase from Yersinia enterolytica into Hela cells. Mol. Microbiol. 14:583–594.
132. Stanley, P., V. Koronakis, and C. Hugues. 1991. Mutational analysis supports a role for multiple structural features in the C-terminal secretion signal of Escherichia coli haemolysin. Mol. Microbiol. 5:2391–2403.
133. Strom, M. S., D. Nunn, and S. Lory. 1991. Multiple roles of the pilus biogenesis protein PilD: involvement of PilD in excretion of enzymes from Pseudomonas aeruginosa. J. Bacteriol. 173:1175–1180
134. Turner, L. R., J. C. Lara, D. N. Nunn, and S. Lory. 1993.Mutations in the consensus ATP-binding sites of XcpR and PilB eliminate extracellular protein secretion and pilus biogenesis in Pseudomonas aeruginosa. J. Bacteriol. 175:4962–4969.
135. Van Gijsegen, F., S. Genin, and C. Boucher. 1993. Conservation of secretion pathways for pathogenicity determinants of plant and animal bacteria. Trends Microbiol. 1:175–180.
136. Wandersman, C. 1992. Secretion across the bacterial outer membrane. Trends Genet. 8:317–321.
137. Wandersman, C., and P. Delepelaire. 1990. TolC, an Escherichia coli outer membrane protein required for hemolysin secretion. Proc. Natl. Acad. Sci. USA 87:4776–4780.
138. Wandersman, C., P. Delepelaire, S. Létoffé, and M. Schwartz. 1987. Characterization of Erwinia chrysanthemi extracellular proteases: cloning and expression of the protease genes in Escherichia coli. J. Bacteriol. 169:5046–5053.
139. Wandersman, C., and S. Létoffé. 1993. Involvement of lipopolysaccharide in the secretion of Escherichia coli α-haemolysin and Erwinia chrysanthemi proteases. Mol. Microbiol. 7:141–150.
140. Wattiau, P., B. Bernier, P. Deslée, T. Michiels, and G. R. Cornelis. 1994. Individual chaperones required for Yop secretion by Yersinia. Proc. Natl. Acad. Sci. USA 91:10493–10497.
141. Wattiau, P., and G. R. Cornelis. 1993. SycE, a chaperone-like protein of Yersinia enterocolytica involved in the secretion of YopE. Mol. Microbiol. 8:123–131.
142. Weiss, M. S., U. Abele, J. Weckesser, W. Welte, E. Schiltz, and G. E. Schultz. 1991. Molecular architecture and electrostatic properties of a bacterial porin. Science 254:1627–1630.
143. Welch, R. A., E. Patchen-Dellinger, B. Minshew, and S. Falkow. 1981. Haemolysin contributes to virulence of extraintestinal E. coli infection. Nature (London) 294:665–667.
144. Willems, R. J., C. Geujen, H. C. J. Van der Heide, G. Renauld, P. Bertin, W. M. R. Van der Akker, C. Locht, and F. R. Mooi. 1994. Mutational analysis of the Bordetella pertussis fim/fha gene cluster: identification of a gene with sequence similarities to haemolysin accessory genes involved in export of FHA. Mol. Microbiol. 11:337–347.
145. Woestyn, S., A. Allaoui, P. Wattiau, and G. R. Cornelis. 1994. YscN, the putative energizer of the Yersinia Yop secretion machinery. J. Bacteriol. 176:1561–1569.
146. Wolff, N., J. M. Ghigo, P. Delepelaire, C. Wandersman, and M. Delepierre. 1994. C-terminal secretion signal of an Erwinia chrysanthemi protease secreted by a signal peptide-independent pathway: proton NMR and CD conformational studies in membrane-mimetic environments. Biochemistry 33:6792–6801.
147. Wong, K. R., and J. T. Buckley. 1991. Site directed mutagenesis of a single tryptophan near the middle of the channel forming aerolysin inhibits its transfer across the outer membrane of Aeromonas salmonicida. J. Biol. Chem. 266:14451–14456.
148. Wong, K. R., M. Green, and J. T. Buckley. 1989. Extracellular excretion of cloned aerolysin and phospholipase by Aeromonas salmonicida. J. Bacteriol. 171:2523–2527.
149. Zhang, F., D. I. Greig, and V. Ling. 1993. Functional replacement of the hemolysin A transport signal by a different primary sequence. Proc. Natl. Acad. Sci. USA 90:4211–4215.
150. Zhang, F., J. A. Sheps, and V. Ling. 1993. Complementation of transport deficient mutants of Escherichia coli α-hemolysin by second-site mutations in the transporter hemolysin B. J. Biol. Chem. 268:19889–19895.
151. Zhu, X., Y. Ohta, F. Jordan, and M. Inouye. 1989. Pro-sequence of subtilisin can guide the refolding of denatured subtilisin in an intermolecular process. Nature (London) 339:483–484.
Chapter63
