typeii secretion and pathogenesis

7/27/2019 TypeII Secretion and Pathogenesis

1/13

INFECTION AND IMMUNITY,0019-9567/01/$04.000 DOI: 10.1128/IAI.69.6.35233535.2001

June 2001, p. 35233535 Vol. 69, No. 6

Copyright 2001, American Society for Microbiology. All Rights Reserved.

MINIREVIEW

Type II Secretion and PathogenesisMARIA SANDKVIST*

Jerome H. Holland Laboratory, Department of Biochemistry, American Red Cross, Rockville, Maryland 20855

Extracellular secretion of proteins is regarded as a majorvirulence mechanism in bacterial infection. Proteins destinedfor the extracellular environment of gram-negative bacteriahave to cross two membranes during their journey across thebacterial cell envelope. This involves translocation across thecytoplasmic membrane and the outer membrane, which areseparated by the periplasmic compartment and the peptidogly-can layer. Several highly specialized pathways have evolved for

this purpose. The type II secretion pathway is one such systemthat is encoded by at least 12 genes and specifically supportsthe transport of a group of seemingly unrelated proteins acrossthe outer membrane. In order for these proteins to enter thetype II secretion pathway, they have to first translocate acrossthe cytoplasmic membrane via the Sec system and then foldinto a translocation competent conformation in the periplasm.Proteins secreted by the type II pathway include proteases,cellulases, pectinases, phospholipases, lipases, and toxins. Ingeneral, these proteins are associated with destruction of var-ious tissues, which contributes to cell damage and disease.Expression of the genes for these proteins and, in some cases,the secretion genes themselves are under quorum-sensing con-

trol or are strictly regulated by the environment at the site ofcolonization. This type of regulation results in controlled se-cretion of virulence factors that will only occur when the bac-teria have reached their correct location and obtained a critialmass. Species identified to date as harboring type II secretiongenes are members of proteobacteria, and most of them ap-pear to be extracellular pathogens. Genome sequencing ofbacterial pathogens continues to identify additional speciesthat contain DNA highly homologous to the type II secretiongenes, suggesting that this pathway is widely distributed withinthe proteobacterial family. This review discusses the identifi-cation of additional type II secretion genes and the expansionof the type II secretion gene family. The regulation of type IIsecretion and its role in disease are also addressed. Finally, the

finding that a component of the type II secretion pathway inVibrio cholerae can also support bacteriophage extrusion isdiscussed.

CLASSIFICATION OF SECRETION PATHWAYS

While the outer membrane provides a natural barrier andacts together with multidrug pumps to protect the gram-neg-

ative cell from harmful agents such as detergents, disinfectants,dyes, certain antibiotics, and toxins, it also poses a challengewith respect to the uptake of nutrients and the excretion ofby-products (105). For this purpose, the outer membrane hasbeen equipped with different pathways to selectively transportmolecules both into and out of the cell (104, 106). Although ithas been accepted that the outer membrane barrier allows forpassage of small molecules, until very recently it was thought to

prevent the extracellular release of larger macromoleculessuch as proteins. Results from studies accumulated within thelast 10 to 15 years have challenged this view and clearly dem-onstrated that the outer membrane is not inert with regard totransport of such molecules. In fact, these studies have sug-gested that every gram-negative organism may have the capac-ity to selectively secrete proteins across the two membranes tothe extracellular environment. What is perhaps surprising isthat, although extracellular secretion is a relatively rare event,several highly specialized pathways have evolved for this pur-pose (see below). At the moment it is not understood whycertain proteins are secreted via one pathway and not another.It is possible that the choice of secretion pathway for individual

proteins is determined in part by the function they perform atthe extracellular site where they are delivered, as was recentlysuggested by Lory (87).

Six pathways for extracellular secretion have been identifiedto date. These include the signal sequence independent path-way (type I), the main terminal branch of the general secretionpathway (type II), the contact-dependent pathway (type III),the type IV pathway, the Bordetella pertussis filamentous hem-agglutinin secretion pathway (two-partner secretion [TPS]pathway) and the autotransporter pathway (for reviews, seereferences 7, 16, 23, 26, 51, 63, 64, 131, and 133). Some of thesepathways transport proteins across the bacterial cell envelopein one single step. For example, translocation via the type I or

III pathway occurs across the inner and outer membrane si-multaneously, bypassing the periplasmic compartment. Al-though the number of accessory gene products required forsuccessful secretion varies among these pathways from 3 fortype I to at least 20 for type III, both of these pathways supportthe secretion of unfolded proteins and neither utilizes the secsystem. In contrast, the type II, TPS, and autotransporter path-ways involve two steps in which proteins containing an N-terminal signal peptide are first translocated across the cyto-plasmic membrane via the sec machinery. Then, followingremoval of their signal peptides and release into the periplasm,the mature proteins cross the outer membrane in a separatestep. As the name implies, secretion through the autotrans-porter pathway does not require any accessory factors. The

* Corresponding author. Mailing address: Jerome H. Holland Lab-oratory, Department of Biochemistry, American Red Cross, 15601Crabbs Branch Way, Rockville, MD 20855. Phone: (301) 738-0604.Fax: (301) 738-0794. E-mail: [email protected].

3523

atIRISonJanuary31,2008

iai.asm.org

Downloadedfrom
http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/http://iai.asm.org/


2/13

respective C-terminal domains of these secreted proteins arethought to insert into the outer membrane, each forming apore that allows transport of the N-terminal portion to the cellsurface. The mature portion of the protein is then releasedfrom this pore structure by proteolytic cleavage (51, 120). Incontrast, outer membrane translocation through the type IIpathway requires the function of at least 12 gene products andoccurs by a mechanism that is still obscure. This pathway canshuttle a large number of seemingly unrelated proteins in theirfolded states to the extracellular environment (see below).Translocation through the outer membrane via the TPS path-way may only depend on the pore-forming protein FhaC (63).Finally, the type IV pathway is primarily involved in the mo-bilization of DNA either between bacteria or from bacteria to

plant cells; however, recently the type IV system has beenfound to also transport proteins, such as B. pertussis toxin andHelicobacter pylori CagA antigen (16, 112). While the DNA isthought to be transferred as a DNA-protein complex in onesingle step across the cell envelope, pertussis toxin appears tobe transported to the periplasmic compartment in a sec-depen-dent manner and then translocated across the outer membranein a separate step (21).

TYPE II SECRETION PATHWAY

The type II pathway was first discovered in Klebsiella oxytoca,where it was found to be required for secretion of the starch-hydrolyzing lipoprotein, pullulanase (32). Since then, this path-

FIG. 1. Alignment of genes encoding type II secretion pathways. Individual species with the names of their secretion genes and a schematicrepresentation of the genes A to O and S shown as boxes with arrowheads to indicate their orientation. The designations of the xcp genes of

P. aeruginosa and P. putida differ from those of other species. Spaces are introduced to indicate that genes are not adjacent. Lines connecting genesindicate that they are linked but interspersed by opposing promoters or other genes. The names of the O genes are vcpD (pilD) in V. cholerae, tapDin A. hydrophila, xcpA (pilD) in P. aeruginosa, and pilD in L. pneumophila. The gene alignment was done using previously published sequences (1,4, 6, 15, 29, 30, 3234, 40, 44, 49, 59, 61, 66, 67, 82, 89, 129, 137, 140, 144). Sequences of open reading frames from S. putrefaciens, C. crescentus, Y. pestis,and G. sulfurreducens were obtained from The Institute for Genomic Research and Sanger Centre via the NCBI microbial genome database.

3524 MINIREVIEW INFECT. IMMUN.


iai.asm.org

Downloadedfrom


3/13

way has been discovered in a number of bacterial species,including several extracellular pathogens. These include hu-

man pathogens such as Vibrio cholerae and Pseudomonasaeruginosa, fish pathogens such as Aeromonas hydrophila, andplant pathogens, including Erwinia chrysanthemi, Erwinia ca-rotovora, and Xantomonas campestris (Fig. 1). In these speciesseveral of the type II secretion genes have been inactivated andshown to be essential for the translocation of multiple proteins(Table 1). These include several different plant cell wall-de-grading enzymes of the Erwinia species and X. campestris andexotoxin A, elastase, phospholipase C, alkaline phosphatase,and lipases LipA and LipC ofP. aeruginosa (86, 95, 159). In V.cholerae this pathway supports the secretion of cholera toxin(CT), hemagglutinin-protease (HAP), and chitinase (25, 28,137). In addition, while extracellular neuraminidase and lipaseactivities can be detected with wild-type V. cholerae grown on

appropriate indicator media, these activities are abolished instrains carrying mutations in the eps type II secretion genes

(137; M. Bagdasarian, unpublished data).Although the first type II secretion genes were identified forK. oxytoca, A. hydrophila, and P. aeruginosa and sequenced 10years ago (31, 32, 36, 67), the first documented secretion mu-tant was identified 15 years before this (57). In that study, achemically induced V. cholerae mutant strain that did not se-crete CT was isolated. V. cholerae has been, and continues tobe, the subject of intense research, as it remains a significanthealth threat in several parts of the world, where it causes asevere and life-threatening diarrheal disease. Its primary viru-lence factor, CT, is in large part responsible for the symptomsof cholera and is probably the most intensely studied protein ofall those known to be secreted via the type II secretion path-way. CT is a multimeric protein complex and is nearly identical

TABLE 1. Properties of bacteria expressing a type II secretion pathwaya

Species or plasmid (description) Environmental niche(s) Disease(s) or effect(s) Ty pe II secreted protein(s)

V. cholerae (humanpathogen)

Brackish and estuarinewater, associated withcrustaceans, human smallintestine

Cholera CT, HAP, chitinase, neuraminidase,lipase

A. hydrophila(fishpathogen, opportunistic

human pathogen)Aquatic, soil Haemorrhagic septicemia infish, occasional

gastroenteritis, septicemiaand meningitis in humans

Aerolysin, amylase, phospholipase,proteases (both metallo and serine),DNAse

K. oxytoca (opportunisticpathogen of hospitalizedpatients)

Aquatic, soil Nosocomial infections Pullulanase

E. chrysanthemi andcorotovora (plantpathogen)

Aquatic, soil, plant surfaces Soft rot in plant tissues Plant cell wall-degrading enzymes suchas cellulases, pectate lyases, pectinmethyesterase, polygalacturonase

P. aeruginosa (opportunistichuman pathogen)

Aquatic, soil, plant surfaces Chronic infection in cysticfibrosis patients, burn

victims, and cancerpatients

Exotoxin A, elastase, LasA, alkalinephosphatase, lipases LipA and LipC,phospholipase C

X. campestris (plantpathogen)

Aquatic, soil, plant surfaces Black rot in crucifers Polygalacturonate lyase, -amylase,protease and endoglucconase

B. pseudomallei

(opportunistic humanpathogen)

Aquatic, soil Melioidosis Protease, lipase, phospholipase C

B. cepacia (opportunistichuman pathogen)

Aquatic, soil Pulmonary infection in cysticfibrosis patients

Protease, lipase

L. pneumophila (humanpathogen)

Freshwater, intracellularparasite of protozoa andhuman alveolarmacrophages

Legionnaires disease Acid phosphatase, lipase, phospholipaseA, RNase, protease (Msp), lipase

E. coli K-12 (M1655)(nonpathogenic)

Part of the normal flora inhuman large intestine

Nosocomial infections Chitinase

Plasmid pO157 (from E.coli O157:H7, a humanpathogen)

Diarrhea, hemorrhagiccolitis, hemolytic uremicsyndrome

?

P. putida (promotes plantgrowth)

Aquatic, soil Rarely associated withclinical infection

?

X. fastidiosa (plant

pathogen)

Xylem of plants Blockage of xylem, leaf

scorching, slows plantgrowth

?

S. putrifaciens (onlyoccasionally associated

with human disease)

Aquatic, soil, oil, fishproducts

Rare bacteremia, soft tissueinfection, otitis media

?

C. crescentus Aquatic, soil ? ?Y. pestis (human pathogen) Rodents, rat fleas Bubonic and pneumonic

plague?

G. sulfurreducens Aquatic sediments ?

a ?, unknown.

VOL. 69, 2001 MINIREVIEW 3525


iai.asm.org

Downloadedfrom


4/13

to the Escherichia coli heat-labile enterotoxin (LT) (146). Thecrystal structures of LT and CT confirm that both toxins arecomposed of a pentameric ring of identical B subunits of 11.6kDa each and one A subunit of 28 kDa (145, 163). Both toxinsbind via their B subunit pentamer to the GM

1ganglioside on

intestinal epithelial cells (98). After entering the epithelial

cells, the A subunit ADP-ribosylates the stimulatory G proteinof the adenylate-cyclase complex. The resulting elevation ofcyclic AMP leads to a rapid and profuse efflux of water andions from the cells, with diarrhea as a consequence (146).

The folding, assembly, and secretion pathways of CT and thealmost identical LT have been analyzed in great detail, and CTserves as a prototype for other proteins secreted via the type IIpathway. The individual toxin subunits are first produced asprecursor proteins with typical N-terminal signal peptides. Af-ter translocation through the cytoplasmic membrane via thesec pathway, the signal peptides are removed and the maturesubunits are released into the periplasm, where they assemblenoncovalently into an AB

5holotoxin complex in a process that

is assisted by the disulfide isomerase DsbA (TcpG) (54, 55,

116, 162). Only when assembled does the toxin complex tra-verse the outer membrane in a second step that requires VcpDand 12 additional gene products that collectively comprise theextracellular protein secretion apparatus (Eps) (Fig. 1) (41, 54,92, 113, 137, 138). The B subunit pentamer carries the infor-mation for outer membrane translocation, and the A subunit issecreted only by virtue of its association with the B

5complex

(56). Based on these early studies and the finding that theperiplasmic toxin complex is very similar, if not identical, to thesecreted form of the toxin, it was proposed that proteins aresecreted in their fully or nearly fully folded forms via the typeII pathway. The suggestion that folding is a prerequisite forouter membrane translocation has been confirmed with a num-

ber of other proteins. However, recent studies suggest that, atleast in some cases, there still may be some structural differ-ences between periplasmic secretion competent proteins andtheir extracellular forms (18). The best examples of this are theelastase of P. aeruginosa and HAP of V. cholerae, which areboth secreted as proforms (13, 47, 96), and it is only when theseproteins have reached the extracellular environment that theirpropeptides are removed.

SEQUENCE AND MUTATIONAL ANALYSIS OF THE

TYPE II SECRETION GENES

Depending on the species, between 12 and 15 genes have

now been identified as essential for type II secretion, and thehomologous genes and gene products have been designated,for most species, by the letters A through O and S (125) (Fig.1). The level of homology between the genes of different spe-cies varies from 25 to 40% identity for the C, L, and M genesto 60 to 80% identity for the E, F, and G genes. The geneticorganization of the type II secretion gene cluster is relativelywell conserved (Fig. 1). Most of the genes appear to be orga-nized in a single operon, and several of the genes are overlap-ping. If variations are detected, they are usually found at the 5and 3 ends of the gene clusters (Fig. 1). Minor differencesbetween the type II secretion gene clusters of relatively closelyrelated species have also been observed. For instance, the Cand D genes of P. aeruginosa (designated xcpP and xcpQ,

respectively) are transcribed in the opposite direction from theE to N genes (xcpR to xcpZ), as well as being transcribed froma different promoter (Fig. 1). This is in contrast to the geneticorganization of most other species, including Pseudomonasputida (30), where the C to N genes are transcribed in alpha-betical order, most likely from the same promoter (Fig. 1).Comparison of the secretion genes of two closely related Er-winia species also shows that one of the genes, outN, is missingin E. chrysanthemi, while it is present and required for secre-tion in E. carotovora (83). The O gene, which in some speciesis the last gene of the type II secretion operon, is homologousto the prepilin peptidase gene required for processing of typeIV prepilin subunits. In species where the O gene is not linkedto the rest of the type II secretion genes, it is found associatedwith a subset of genes required for type IV pilus biogenesis. Inthese species the prepilin peptidase exhibits a dual function,i.e., it is required for processing and methylation of the type IVprepilin subunits as well as the prepilin-like components en-coded by the type II secretion genes G, H, I, J, and K (12, 111,149). In V. cholerae, for instance, VcpD (PilD) is required for

CT secretion via the Eps apparatus and has been shown toprocess the precursor form of EpsI as well as the prepilin formsof the type IV pilin subunits PilA and mannose-sensitive hem-agglutinin (MSHA) A (MshA) (41, 92). DNA compositionanalysis of the secretion genes of V. cholerae suggests that theeps genes are relatively old and that the vcpD gene has beenacquired recently by horizontal gene transfer (50). It is possiblethat the eps cluster originally contained an epsO gene at thedistal end but that when the vcpD gene was acquired, the epsOgene became redundant and was subsequently lost.

The secretion pathways in K. oxytoca and the Erwinia specieshave been reconstituted in laboratory strains of E. coli (83,123). In this background, Possot and colleagues have shown

that all pul genes except for pulB, pulH, and pulN, are requiredfor secretion (123). A similar result was obtained with the E.chrysanthemi out secretion genes expressed in E. coli (83). Inthis case, all out genes, with the exception ofoutH, were foundto be required for secretion of Erwinia pectate lyase. In con-trast, an E. chrysanthemi outH mutant was defective for extra-cellular secretion (83). The discrepancy in requirement for theH gene in Out-dependent secretion is not fully understood;however, it could be due to a copy-number effect (chromosom-al versus plasmid expression) or differences between the spe-cies. It is worth mentioning that even though laboratory strainsof E. coli are not thought to transport proteins to the extra-cellular environment under normal growth conditions, they do

contain what appears to be a complete set of type II secretiongenes, some of which may have compensated for the missinggene(s) in those studies (Fig. 1). However, the level of expres-sion of the E. coli gsp secretion genes is extremely low (124).The lack for requirement of pulN in pullulanase secretion isconsistent with the finding that several of the type II secretiongene clusters do not contain an N gene (Fig. 1). The require-ment for the A gene has also been observed only for A. hy-drophila, and the requirement for the B gene has been ob-served only for Erwinia species and A. hydrophila (60, 66, 83).Likewise, the need for an S protein in type II secretion hasbeen demonstrated only for Erwinia and K. oxytoca (45, 143).

It is interesting to note that in P. aeruginosa a second D genewith overlapping function was identified by Martinez and col-



iai.asm.org

Downloadedfrom


5/13

leagues (94), who noticed that in a strain carrying a mutationin the type II secretion gene D (xcpQ), a low level of extracel-lular secretion could still be detected. This secretion was con-tributed by another D homologue, xqhA (94). Subsequently,when this gene was inactivated in the xcpQ mutant back-ground, extracellular secretion was completely abolished. Fur-

thermore, when the complete genome sequence ofP. aerugi-

nosa was released, it was noted that the xcp genes have severalhomologues, including a complete second set of secretiongenes (148; A. Filloux, personal communication).

The function and interactions of the individual componentsof the type II secretion apparatus have been reviewed recentlyand will not be discussed in great detail here (37, 131, 133).However, a summary of the current knowledge will be pre-sented. Protein D is a member of the secretin family thatincludes proteins required for type IV pilus biogenesis, fila-mentous phage extrusion, and type III secretion (42). Theseproteins are present in the outer membrane, where they arethought to form gated secretion pores (8, 72, 8385, 90, 107,108, 132). The secretins exhibit ion-conducting properties and

an oligomeric structure that can be visualized by electron mi-croscopy (8, 14, 85, 90, 107, 108). Proper outer membraneinsertion of protein D is assisted by protein S in K. oxytoca andE. chrysanthemi (27, 45, 143), but the requirement for the Sprotein in other species has not been confirmed. The presenceof sequences homologous to the S gene in the E. coli O157:H7pO157 plasmid and the Yersinia pestis chromosome may sug-gest that the S gene is restricted to Enterobacteriaceae (Fig. 1).Although specifically required for outer membrane transloca-tion, the remaining components of the type II secretion appa-ratus appear to be associated with the cytoplasmic membrane,where several of them have been found to interact (5, 11, 58,70, 88, 123, 126128, 134136, 141, 151). However, recent

studies suggest that there may also be interactions betweensome of the cytoplasmic membrane components and the se-cretin (10, 24, 75, 121). In addition, outer membrane translo-cation requires the proton motive force of the cytoplasmicmembrane (78, 122). Taken together, these findings suggestthat the components of the secretion apparatus form a multi-protein complex that spans both membranes.

Several of the type II secretion genes, including the D, E, F,and O genes, are homologous to genes required for assemblyand export of type IV pili. These are surface appendages thatplay a major role in colonization and virulence of a number ofbacterial pathogens, including P. aeruginosa, Neisseria gonnor-hoeae, and V. cholerae (20, 46, 71, 74, 110). Moreover, proteins

G, H, I, J, and K are similar to the type IV prepilin subunitswith respect to their size and sequence similarity within theN-terminal domains. Because of these similarities, it was sug-gested that they might form a pilus-like structure (109), and arecent paper by Sauvonnet and colleagues demonstrated thatthe components of the type II secretion apparatus can form apilus-like structure when their genes are overexpressed (139).However, when the genes were expressed from a chromosomallocation or from a low-copy-number plasmid, no pilus struc-ture was observed, even though secretion of pullulanase couldbe detected. In support of this finding, a pilus structure en-coded by the xcp type II secretion genes of P. aeruginosa hasbeen identified (D. Nunn, personal communication). Theseresults suggest that a subset of the type II secretion genes has

the capability to form a pilus-like structure; however, it is not

known whether the pilus structure per se directly supportssecretion. Different models for pilus-mediated secretion havebeen proposed. The cytoplasmic membrane-anchored type IIsecretion pilus may act as a piston, and through extension andretraction it may push the secreted proteins through the gatedpore (37, 133, 143). Support for this model comes from recentstudies showing that the type IV pilin subunits assemble intofilaments prior to outer membrane translocation and that thetype IV pili can forcefully retract (99, 157). In this model,protein E, which has been found to exhibit autophosphoryla-tion activity and is present on the cytoplasmic side of thecytoplasmic membrane, may regulate this process via interac-tion with proteins L, M, and C and with the secretin in theouter membrane (5, 121, 123, 127, 133, 134, 136). Alterna-tively, the E protein may supply energy for the outer mem-brane translocation process by promoting polymerization ofthe pilin-like proteins G, H, I, J, and K (133). Because the typeII secretion pilus formed long filaments that bundled into high-er-ordered structures when the pul genes were overexpressed,an alternative role of the pilus in secretion would be to channelsecreted proteins past surface structures such as capsules orpast aggregated bacteria present in biofilm (Fig. 2) (133, 139).

IDENTIFICATION OF ADDITIONAL SPECIES WITH

TYPE II SECRETION GENES

Several of the species that encode the type II secretion

pathway can be isolated from multiple ecological niches (Table1). V. cholerae can not only colonize the human gastrointestinaltract but also live in the aquatic environment either as a free-swimming organism, as an organism associated with plankton,or as a member of biofilm attached to chitinous or abioticsurfaces. The opportunistic human pathogen P. aeruginosa,which frequently causes infections in cystic fibrosis patients,cancer patients, and burn victims, can survive and grow in soilas well as the aquatic milieu and can infect both animals andplants. Another species that can be isolated from soil, water,and plants is Burkholderia cepacia (Table 1). This species isalso associated with infections in cystic fibrosis patients, whereit can cause necrotizing pneumonia. B. cepacia has recentlybeen found to contain a type II secretion pathway that is

FIG. 2. Regulation of secretion. Three steps in the colonization ofa general surface, such as the surface of plant leaves, the mucosalepithelium, or a solid surface in an aquatic milieu. Secretion of viru-lence factors occurs only when the bacteria have reached a criticalconcentration or retrieved the appropriate signal from the surface.



iai.asm.org

Downloadedfrom


6/13

required for protease and lipase secretion, and sequences ho-mologous to epsF-epsK have been deposited in the GenBankdatabase (accession number AF127982). Sequences homolo-gous to the epsC-epsN genes have also been identified in therelated Burkholderia pseudomallei, which is responsible for me-lioidosis, a life-threatening, systemic bacterial infection. In this

species the type II secretion genes are required for secretion ofprotease, lipase, and phospholipase C (33). It is interesting tonote that the order of the secretion genes ofB. cepacia and B.pseudomallei differs from that of the secretion genes of otherspecies. Instead of being upstream of the D gene, the C geneis inserted between the F and G genes (Fig. 1). Furthermore,the orientation of this gene is opposite that of the rest of thegenes in the cluster.

Recently, the first intracellular pathogen to express a type IIsecretion pathway was identified. Two different laboratorieshave identified type II secretion genes in Legionella pneumo-phila (4, 44, 81), the causative agent of a severe form of pneu-monia called Legionnaires disease (154). As many as eightdifferent proteins may depend on these genes for extracellular

secretion. Some of these proteins are listed in Table 1. L.pneumophila is a facultative intracellular pathogen; it can befound outside the human host in the freshwater environmenteither as a free-living organism in biofilms or as an intracellularpathogen of protozoa (154). The type II secretion genes iden-tified to date in L. pneumophila include the lspF-lspKgenes andthe prepilin peptidase gene pilD (Fig. 1) (4, 44, 81). Sequenceanalysis of DNA 1 kb downstream of lspK suggests that L.pneumophila may lack the L, M, and N genes, unless they arepresent elsewhere in the genome (44). Early release of se-quence information from the L. pneumophila sequencingproject at the Columbia Genome Center provides evidencethat sequences homologous to the epsC, epsD, and epsE genes

are present in this species (N. P. Cianciotto, personal commu-nication). By taking advantage of this information, mutationsin the putative lspD and lspE genes have been introduced andfound to prevent extracellular secretion of protease (N. P.Cianciotto, personal communication).

As mentioned above, type II secretion genes have also beenidentified in E. coli K-12 (9, 147); however, they do not appearto be expressed under normal laboratory growth conditions(40). In a study aimed at identifying conditions that increasethe level of E. coli gsp transcription, it was found that expres-sion from the gspC promoter was increased two- to threefold ina global regulatory mutant lacking the nucleoid-structuringprotein H-NS (124). Although the expression was improved in

this mutant background, it was still extremely low and nosecreted protein could be detected.Francetic and colleagues suggested that ChiA is a chitinase,

which may be secreted via the type II secretion pathway in E.coli K-12 (12). The chiA gene is located downstream of thetype II secretion genes and is separated from them by only twoopen reading frames (ORFs). The expression of the chiA geneis, like the secretion genes, regulated by H-NS (38). In addi-tion, ChiA exhibits a low-level homology to the V. choleraeendochitinase, which is secreted via the Eps pathway (25, 137).However, even in the hns mutant background ChiA accumu-lated in the periplasmic compartment, and it was only when theE. coli gsp genes were overexpressed from a plasmid in the hnsmutant that chitinase secretion could be observed (39).

In contrast to E. coli K-12, the food-borne pathogen E. coli0157:H7 does not carry type II secretion genes on the chro-mosome (117; V. Burland, personal communication). From ananalysis of the two E. coli genomes using a genome browsingtool (www.genome.wisc.edu), it is evident that the region con-taining the type II secretion genes as well as the chiA gene inE. coli K-12 is replaced by novel DNA in E. coli O157. Instead,the large 92-kb plasmid pO157, which is found in most entero-hemorrhagic E. coli (EHEC) O157:H7 strains, carries a set oftype II secretion genes, etpC-etpM, etpO, and etpS (Fig. 1) (15,140). This is the first described case in which the type IIsecretion genes are present on a plasmid. It is not known ifthese genes encode a functional secretion apparatus or whatthe substrate for this putative apparatus is. It is possible thatthe etp genes may be required for secretion of a protein en-coded by one of the more than 20 novel ORFs on the plasmidor a chromosome-encoded protein. When the distribution ofetp genes among other E. coli strains was determined by hy-bridization and PCR, etp-like sequences were found in allEHEC strains of serogroup O157 but were less likely to be

identified in other serogroups (140). No positive signal wasobtained with enterotoxigenic, enteropathogenic, enteroinva-sive, or enteroaggregative E. coli. On the other hand, whenprobes of the E. coli K-12 gspGH genes were used in hybrid-ization studies, homologous sequences were restricted to en-teroinvasive and enterotoxigenic E. coli (147). No positive sig-nal was obtained for the Salmonella or Shigella species tested.

Blattner and colleagues at the University of Wisconsin are inthe process of sequencing the genome of another pathogenicE. coli strain, the highly virulent uropathogenic E. coli strainCFT073, which is responsible for acute pyelonephritis. In con-trast to E. coli O157:H7, this pathogen has kept the type IIsecretion genes on the chromosome, and these genes are es-

sentially identical to the corresponding E. coli K-12 genes(www.genome.wisc.edu).

The recently released genome sequence of the economicallyimportant plant pathogen Xylella fastidiosa, which affects pro-duction of citrus fruits and grapes, revealed the presence of atype II secretion gene cluster homologous to that of the relatedX. campestris (144). Genes encoding putative substrates for thissecretion pathway were also present. These include cellulaseand polygalacturonase, which are known virulence factors se-creted via the type II pathway in Erwinia and X. campestris(Table 1).

Analysis of DNA sequences deposited in the National Cen-ter for Biotechnology Information (NCBI) microbial genome

database of completely and incompletely sequenced genomesreveals that additional species harbor DNA homologous to thetype II secretion gene cluster. The genomes of Shewanellaputrefaciens, Caulobacter crescentus, and Y. pestis, sequenced bythe Institute for Genomic Research and Sanger Centre, con-tain ORFs that display homology to the eps genes of V. chol-erae. The genetic organization of the secretion genes in S.putrefaciens is identical to that ofV. cholerae and A. hydrophila(Fig. 1). In fact, when comparing the type II secretion genesfrom different species identified to date, the eps genes exhibitthe overall highest level of homology to the secretion genes inA. hydrophila and S. putrefaciens. S. putrefaciens is a pathogenrarely found in humans; only a few cases of bacteremia havebeen reported. It is an inhabitant of marine ecosystems and is



iai.asm.org

Downloadedfrom


7/13

often associated with spoiling fish (43). C. crescentus is also anenvironmental species that is not associated with human dis-ease but has been studied extensively for its asymmetric celldivision and cell differentiation (65, 142). Of all species havingthe O gene as the last gene of the secretion gene cluster, the Ogene in C. crescentus is transcribed in the opposite orientation

compared with the rest of the secretion genes (Fig. 1). No ORFhomologous to epsN gene was detected in this species whenusing the standard setting for the BLAST search of the NCBImicrobial genome database. However, there is a stretch ofapproximately 870 nucleotides (nt) between the putative Mand O genes that would be large enough to accommodate an Ngene, which is approximately 750 nt in length (137). Y. pestis,the causative agent of the black death, or plague, which killednearly half of the population of Europe in the 14th century alsocontains ORFs homologous to the epsC-epsL and vcpD genes(Fig. 1). Immediately downstream of these ORFs, there is alsoan S gene. Using the standard setting for the BLAST analysis,no ORFs homologous to the epsMand epsNgenes were iden-tified. However, there is a stretch of DNA of approximately

600 nt between the L and O genes in Y. pestis. This spacingwould allow for an M gene but most likely not for an N gene.The epsM gene of V. cholerae is 497 nt (137). Consistent withthis is the finding that all type II pathways identified to datehave an M gene, while only about half of the species have anN gene (Fig. 1). Y. pestis has been extensively studied for itsability to secrete virulence factors via the contact dependenttype III secretion pathway; however, the presence of anotherextracellular secretion pathway in this species has not beenpreviously noted. What the type II pathways in Y. pestis, S.putrifaciens, or C. crescentus might be transporting is notknown. One can only speculate that they may transport ametalloprotease, lipase, phospholipase, or chitinase, as do the

type II secretion pathways of other species (Table 1). Consis-tent with this is the identification of ORFs in Y. pestis and S.putrefaciens whose translated products exhibit homology to V.cholerae endochitinase, while sequences present in the S. pu-trefaciens genome are homologous to V. cholerae HAP and thelipase LipA. Finally, C. crescentus contains DNA homologousto V. cholerae lipA.

The anaerobic Geobacter sulfurreducens (17, 22), isolatedfrom aquatic sediments, is also likely to contain a type IIsecretion apparatus. Genes homologous to the epsC-epsIgenesare present in one sequence contig, while sequences homolo-gous to the epsA and epsJ genes are present on others. Thegenomes of many other species contain sequences homologous

to the D, E, and F genes of the type II secretion gene cluster.However, since these sequences are also homologous to genesrequired for type IV pilus biogenesis, it is not possible at themoment to determine whether these sequences encode com-ponents of the type II secretion apparatus.

REGULATION OF SECRETION

Quorum sensing is the mechanism whereby a number ofbacterial species control gene expression in response to celldensity. One form of quorum sensing involves the productionand detection of acylated homoserine lactones. Activation (orrepression) of genes occurs when these diffusable autoinducersaccumulate and reach a critical concentration as the cell den-

sity increases. This means of cell-to-cell communication tocoordinate gene expression occurs apparently both within andbetween different species and is thought to contribute to suc-cessful colonization and infection. The first species identifiedthat depends on quorum sensing for regulated gene expressionwas V. fischeri, which requires quorum sensing for biolumines-cence (35, 103). Subsequently, the production of many extra-cellular virulence factors has been found to be regulated in agrowth-phase-dependent manner as a result of quorum sensingor autoinduction. In P. aeruginosa, quorum sensing involvestwo different systems, lasRIand rhlRI(114, 115). These systemsoverlap and in some instances regulate the same genes. Inaddition, rhlRI-regulated genes require the lasRI system formaximal expression (118). The genes for several of the sub-strates for the type II secretion pathway Xcp, including exo-toxin A and elastase, are regulated by these systems, as are thexcp secretion genes themselves (19). This dual level of regula-tion should not only prevent premature secretion of proteinsbefore the cells have reached a critical mass but also conserveenergy. If there is no substrate to secrete, there is no reason to

produce and assemble the secretion apparatus. LasRI homo-logues have also been identified for B. cepacia (80). In a B.cepacia cepR mutant, no extracellular protease can be de-tected, but it is not yet known if CepR regulates proteaseproduction and/or biogenesis of the type II secretion apparatus(80). Another recent report suggests that the expression oflipase and phospholipase C is also under quorum-sensing con-trol in B. cepacia (156). In addition to intraspecies regulation,there is also apparent cross talk and interspecies signalingbetween P. aeruginosa and B. cepacia (97). Cell-free mediumtaken from stationary-phase cultures ofP. aeruginosa increasesthe production of lipase and protease when added to culturesof B. cepacia (97). A quorum-sensing system has also been

identified for E. carotovora, another species with a type IIsecretion pathway. The virulence of this species is dependenton production and secretion of a number of plant cell wall-degrading enzymes (Table 1). When a strain with a mutation inthe E. carotovora epxI gene, which is homologous to the V.fischeri luxl gene, was analyzed, the production of these exoen-zymes and the ability of the mutant strain to propagate andcause disease were reduced (69, 119). Finally, expression of theErwinia out secretion genes themselves is greatly increased inearly stationary phase, suggesting that these genes may also beunder quorum-sensing control (82).

In contrast to expression of the type II secretion genes ofPseudomonas and Erwinia, expression of the exe and eps secre-

tion genes ofA. hydrophila and V. cholerae may be constitutive(150; M. Sandkvist and V. J. DiRita, unpublished studies ofV.cholerae). Addition of purified A. hydrophila autoinducer N-(butanoyl)-L-homoserine lactone to cultures of A. hydrophilaexpressing an exeL::lacZ fusion had no effect on lacZ expres-sion, and reverse transcriptase PCR analysis of exeD mRNAproduction in wild-type A. hydrophila showed that the exeDgene is constitutively expressed (150). While AhyRI does notregulate the exe secretion genes in A. hydrophila, the expressionof the secreted serine protease AspA is growth phase regulatedby this quorum sensing system (150). In V. cholerae, growth-phase-dependent expression of the type II secretion genes wasalso not observed since the expression of an epsE::lacZ fusionremained unchanged during exponential and stationary-phase



iai.asm.org

Downloadedfrom


8/13

growth (M. Sandkvist and V. J. DiRita, unpublished data). Inaddition, when wild-type V. cholerae was cultured and sampleswere collected during growth to analyze for EpsE protein pro-duction by immunoblotting, the same level of EpsE protein percell mass was produced during all growth phases tested (M.Sandkvist and V. J. DiRita, unpublished data). It is possible

that constitutive expression of the secretion apparatus may benecessary for species that secrete substrates in several differentenvironments. For instance, the Eps apparatus must be activenot only in the human intestine, where it secretes CT, but alsoin the aquatic environment, where it is thought to secretechitinase (Table 1) (25). Although the expression of the epsgenes appears to be constitutive, it is still possible that certaingrowth conditions may affect the activity of the secretion ap-paratus. Interestingly, maltose has been found to inhibit secre-tion of CT in vitro (73). When maltose was present, only 22%of total amount of CT was secreted following 6 h of growth.The rest of CT accumulated inside the cells. In contrast, with-out maltose, 95% of CT was extracellular. Following overnightgrowth (17 h), the effect of maltose was even greater, with only

5% of the total amount of toxin detected in the growth media.The mechanism of this inhibition is not understood but itappears to be at least partially growth phase dependent, be-cause no inhibition of CT secretion was observed following 3 hof growth. Interestingly, addition of maltose to wild-type V.cholerae cultures also had an adverse effect on secretion ofHAP and biogenesis of the type IV pilus MSHA (73). HAP,like CT, is secreted via the Eps pathway (137). MSHA, on theother hand, requires another set of genes for its assembly andouter membrane translocation (46, 91, 93). However, secretionof CT, HAP, and MSHA requires the function of one commoncomponent (41, 92). The prepilin peptidase VcpD (PilD) pro-cesses the MSHA prepilin subunit as well as the pilin-like

proteins EpsG, -H, -I, -J, and -K, which are required for CTand HAP secretion (92). Thus, one explanation could be thatmaltose interferes with the expression of the vcpD gene or theactivity of the VcpD protein, thereby inhibiting secretion.

While the eps genes may be constitutively expressed andextracellular secretion is likely to occur in environments asdiverse as the human intestine and the aquatic milieu, expres-sion of the substrates for this pathway is tightly controlled. Theregulation of expression of the CT genes ctxAB is complex anddiffers during in vitro growth and infection (76). In vitro, ex-pression of ctxAB is greatly increased at 30C in Luria broth(pH 6.5) with low osmolarity and requires the transcriptionalactivators ToxR, ToxT, and TcpP (48, 53, 101, 102). Interest-

ingly, during colonization of the infant mouse intestine, TcpPis not required for toxin expression. Instead, ctxAB expressionrequires the toxin-coregulated pilus. In an elegant study, Leeand colleagues showed that the ctxAB and tcpA genes aresequentially expressed during infection and that expression ofctxAB requires prior tcpA expression (76). This led the authorsto suggest that colonization is a prerequisite for toxin produc-tion. With the imposition of such a regulatory control, toxinproduction and secretion will only occur when V. cholerae hasarrived at the intestinal epithelium (76).

When V. cholerae is cultured in vitro, HAP production isupregulated in stationary phase (68). This regulation occurs atthe transcriptional level by HapR, a 203-amino-acid proteinwith 71% identity to LuxR ofVibrio harveyi. HapR may directly

activate hap expression in V. cholerae or may indirectly regu-late HAP production via the stationary-phase sigma factorRpoS. Yildiz and Schoolnik recently showed that inactivationof the rpoS gene in V. cholerae led to reduced levels of HAPproduction (160). Finally, just as HAP production is increasedin late growth phase, the production of extracellular neuramin-idase is increased more than 10-fold when V. cholerae culturesreach stationary phase, and addition of sialic acid to the cultureincreases the production further (153).

Thus, the expression of most proteins secreted via the type IIpathway is growth phase dependent or strictly regulated byenvironmental signals. In addition, the type II secretion genesthemselves are under quorum-sensing control in some species.Whether the regulatory control occurs at the transcriptionallevel of a particular virulence factor or at the functional levelof its secretion apparatus, extracellular secretion of the viru-lence factor will generally occur only when the bacteria havereached a critical mass and location. This is schematically de-picted in Fig. 2, where in the case of V. cholerae, the coloni-zation surface may represent the intestinal epithelium, where

CT is secreted or the exoskeleton of crustaceans, where chiti-nase is released.

SECRETION AND DISEASE

The type II pathway appears to be typically associated withorganisms that colonize surfaces, and in most cases these or-ganisms do not invade cells. Several of these species are alsoknown to form biofilms, and mutations in the eps genes inter-fere with polysaccharide production and biofilm formation inV. cholerae (2). Even so, it is unlikely that the Eps pathway isdirectly involved in the transport of polysaccharides to the cellsurface, because the biosynthesis of polysaccharides appears to

be controlled by another pathway in V. cholerae that is homol-ogous to the polysaccharide biosynthesis systems of other spe-cies (161). Thus, the defect in biofilm formation in the epsmutants could be due to pleiotropic effects or a weakenedouter membrane (137). Alternatively, it is possible that anas-yet-unidentified polysaccharide-processing enzyme requiredfor assembly of polysaccharide on the cell surface could use theEps system for outer membrane translocation.

The role of the type II pathway in pathogenesis has not beeninvestigated in great detail. However, in species where it se-cretes known virulence factors, it is assumed to be essential forpathogenesis. In the case of animal and human pathogens, thetype II secreted proteins might be required for the establish-

ment of an infection at the mucosal surfaces of the respiratoryand gastrointestinal tract. Secreted proteases, lipases, phos-pholipases, and toxins may aid in the colonization or tissuedestruction by defeating host defenses and/or providing nutri-ents. In V. cholerae, CT and toxin coregulated pilus are themajor virulence factors. Both factors have been analyzed ingreat detail and tested both in animal and human models. Thesymptoms, including severe diarrhea, vomiting, and headaches,observed with wild-type V. cholerae were considerably attenu-ated when a toxin-negative mutant was given to human volun-teers (52). The number of volunteers with diarrhea and thenumber of diarrheal stools were reduced. In addition, themean diarrheal stool volume was decreased from 5 liters forthe wild-type strain to 0.2 liters for the toxin mutant. More-



iai.asm.org

Downloadedfrom


9/13

over, ingestion of purified CT caused large volumes of diarrheain a dose-dependent manner in volunteers (79).

We have hypothesized that the type II secretion pathwayEps is essential for V. cholerae virulence, since it secretes CT.Unfortunately, the eps secretion mutants of V. cholerae havenot been tested for their ability to colonize and cause disease

in vivo. These mutants grow slowly and are unable to competewith wild-type V. cholerae when cultured together in vitro(Z. Y. Dossani and M. Sandkvist, unpublished data); thus, anyresults from eps mutant colonization studies in the infantmouse model would be difficult to interpret. However, the roleof VcpD, which is required for proper assembly and functionof the Eps apparatus, has been tested in virulence and coloni-zation studies. These results indicated that inactivation of thevcpD gene led to a 100-fold increase in the 50% lethal dose(LD

50) for the cholera infant mouse model and to an approx-

imately 100-fold reduction in colonization efficiency (92). Sincethe vcpD mutant is pleiotropic, the reduced virulence of thismutant cannot be contributed to lack of CT secretion only.

Type II secretion mutants of other species have been ana-

lyzed for their ability to cause disease. Mutagenesis of the outsecretion genes of the plant pathogen E. chrysanthemi pro-duces mutants that are unable to secrete enzymes that digestplant cell walls and are severely attenuated (3, 152). On theother hand, when P. aeruginosa xcp mutants were tested in amouse burn infection model, they were found to be as virulentas the parental strain (158). The reason for this result is notclear, although it may suggest that low-level release of viru-lence factors such as exotoxin A and elastase from the xcpmutants is sufficient to cause disease. Alternatively, the level oftoxin and protease gene expression may be increased to com-pensate for the reduced secretion level in the xcp mutants, orthe second type II secretion gene cluster, recently identified in

the P. aeruginosa genome, may partly complement the muta-tions. Similarly, secretion mutants ofB. pseudomallei displayedonly modest defects in ability to cause disease in the Syrianhamster model (33). The LD

50for these mutants was increased

3- to 13-fold. In the case ofL. pneumophila, the type II secre-tion pathway was found to be required for multiplication inamoebae (44, 81). However, inactivation of the lspG and lspHsecretion genes did not have an effect on macrophage cytotox-icity, since like the wild-type strain, the lspGHmutant was ableto kill HL-60-derived macrophages (44). On the other hand, apilD knockout mutant demonstrated severe defects in the abil-ity to grow within and kill macrophage-like U937 cells (81). Inaddition, the LD

50for the pilD mutant was increased 100-fold

compared to that of the wild-type strain administered intratra-cheally to Guinea pigs (81). These results suggest that the typeII secretion pathway in L. pneumophila is not required forgrowth in macrophages, and that the attenuated phenotypeobserved with the pilD mutant may be due to lack of biogenesisof a pilus, yet to be identified, since PilD may affect both typeII secretion and pilus biogenesis.

EXPLOITATION OF A SECRETION PATHWAY

COMMON TO ALL VIBRIOS FOR TOXIN SECRETION

AND HORIZONTAL GENE TRANSFER

Since the type II pathway is widely distributed it was ofinterest to determine whether all species have the capability to

secrete CT. Although species such as P. aeruginosa, K. oxytoca,and E. chrysanthemi contain type II secretion genes homolo-gous to the V. cholerae eps genes, they were unable to secretethe toxin B subunit when a plasmid encoding the B subunitgene was introduced (100). Yet several environmental vibriosother than V. cholerae were found to secrete plasmid-encoded

B subunits to the extracellular environment (100). In addition,when the B subunit was expressed in the marine Vibrio sp. 60,which is related to Vibrio anguillarum, it was also secreted (77).However, in the extracellular secretion mutant MVT1192,which is unable to secrete protease (62), the B subunit re-mained in the periplasm (77). When the MVT1192 mutant wasanalyzed further, we found that the secretion defect could becomplemented with the V. cholerae epsE gene (M. Sandkvistand A. Ichige, unpublished data). Likewise, overproduction ofan EpsE protein with a mutation in the ATP-binding site in thewild-type Vibrio sp. 60 inhibited protease secretion (Sandkvistand Ichige, unpublished data). This dominant-negative effectof the mutant EpsE protein was also observed for wild-type V.

cholerae and is most likely a result of competition betweenwild-type and mutant EpsE proteins for interaction with EpsLand the rest of the secretion apparatus (134). The findings thatthe B subunit could be secreted from most Vibrio species andthat the V. cholerae epsE gene was able to restore the secretionphenotype in a marine Vibrio mutant, suggests that the Epspathway is common to the Vibrionaceae family. Most of thespecies in this family are not thought of as human pathogensand are not known to produce enterotoxins. It is thereforelikely that the secretion apparatus in Vibrio is used primarilyfor secretion of proteins common to all vibrios, possibly pro-teases and chitinases. Vibrio species are common inhabitants ofthe aquatic environment, and many, if not all, produce andsecrete chitinases as well as proteases homologous to V. chol-erae HAP. It is believed that the Eps apparatus did not origi-nally evolve for the purpose of secreting CT but rather that CThas evolved in such a way as to be recognized and efficientlysecreted by this machinery.

The structural genes for CT are encoded by the filamentousbacteriophage CTX, which is related to the E. coli phage f1(155). The CTX genome is 7000 bp in length and chromo-somally integrated, but it can also replicate as a plasmid (155).Like the coliphage f1, CTX is actively secreted without lysingthe host cell. Secretion of CTX requires EpsD, the compo-nent of the Eps type II secretion apparatus that is likely to formthe outer membrane translocation pore (28). This is consistentwith the notion that CTX does not encode its own D pore forsecretion and is in contrast to the homologous E. coli f1 phage,which expresses its own outer membrane pore, the EpsD ho-mologue pIV (84, 130). No other Eps protein is thought to berequired for the extracellular release of CTX, suggesting thatthe secretory pathways for CT and CTX have converged atthe site of the EpsD pore (28). This finding is intriguing andsuggests that CTX has taken advantage of a preexisting se-cretion pore that is widely distributed within the Vibrionaceaefamily. As such, the EpsD protein has been converted to avirulence-export machinery. In this way, EpsD contributes tothe pathogenesis of both the V. cholerae bacterium and thephage by actively secreting the toxin and by releasing the phageto allow for horizontal gene transfer.



iai.asm.org

Downloadedfrom


10/13

CONCLUDING REMARKS

The type II secretion pathway is widely distributed amongplant, animal, and human pathogens of the proteobacteria.Representatives are present in all subdivisions of the pro-teobacteria but are mostly clustered in the gamma subdivisionand include species in the Aeromonadaceae, Alteromona-

daceae, Legionellaceae, Enterobacteriaceae, Pseudomonaceae,Vibrionaceae, and Xanthomonadales families. It is possible thatthis pathway is confined to and is an intrinsic property of theproteobacteria. Alternatively, other bacteria may contain thispathway; however, if they are further removed from and lessrelated to the proteobacteria, their putative secretion genesmay not be identified by simple homology searches since thelevel of homology may be too low. For that reason, identifica-tion of this pathway in bacteria belonging to other groups mayhave to be obtained through functional studies. Although se-quencing the secretion genes is important and provides valu-able information, it cannot fully explain how proteins cross theouter membrane. Thus, determining the organization of the

type II secretion apparatus and identifying the putative secre-tion signal on the secreted proteins is the focus of intenseresearch at the moment. This information should lead to anincreased understanding of the mechanism of secretionthrough the type II secretion pathway and may generate toolsuseful for therapeutic interventions. The wide distribution ofthe type II secretion pathway and its similarity to the type IVpilus biogenesis system suggests that the identification of spe-cific drugs or peptides that can interfere with both pathwaysmay simultaneously block secretion of virulence factors andcolonization and could thereby provide a reasonable alterna-tive to antibiotic use.

ACKNOWLEDGMENTS

I thank V. Burland, N. Cianciotto, A. Filloux, N.-T. Hu, D. Nunn,and A. P. Pugsley for providing information prior to publication. Manythanks to M. Bagdasarian, V. J. DiRita, K. C. Ingham, D. A. Lawrence,and M. Scott for critically reading the manuscript. Preliminary se-quence data was obtained from The Institute for Genomic Research

website (http://www.tigr.org) and the Sanger Centre website (http://www.sanger.ac.uk/Projects).

REFERENCES

1. Akrim, M., M. Bally, G. Ball, J. Tommassen, H. Teerink, A. Filloux, and A.Lazdunski. 1993. Xcp-mediated protein secretion in Pseudomonas aerugi-nosa: identification of two additional genes and evidence for regulation ofxcp gene expression. Mol. Microbiol. 10:431443.

2. Ali, A., J. A. Johnson, A. A. Franco, D. J. Metzger, T. D. Connell, J. G.Morris, Jr., and S. Sozhamannan. 2000. Mutations in the extracellularprotein secretion pathway genes (eps) interfere with rugose polysaccharideproduction in and motility ofVibrio cholerae. Infect. Immun. 68:19671974.

3. Andro, T., J.-P. Chambost, A. Kotoujansky, J. Cattaneo, Y. Bertheau, F.Barras, F. Van Gijsegem, and A. Coleno. 1984. Mutants of Erwinia chry-santhemi defective in secretion of pectinase and cellulase. J. Bacteriol.160:11991203.

4. Aragon, V., S. Kurtz, A. Flieger, B. Neumeister, and N. P. Cianciotto. 2000.Secreted enzymatic activities of wild-type and pilD-deficient Legionellapneumophila Infect. Immun. 68:18551863.

5. Ball, G., V. Chapon-Herve, S. Bleves, G. Michel, and M. Bally. 1999.Assembly of XcpR in the cytoplasmic membrane is required for extracel-lular protein secretion in Pseudomonas aeruginosa. J. Bacteriol. 181:382388.

6. Bally, M., A. Filloux, M. Akrim, G. Ball, A. Lazdunski, and J. Tommassen.1992. Protein secretion in Pseudomonas aeruginosa: characterization ofseven xcp genes and processing of secretory apparatus components byprepilin peptidase Mol. Microbiol. 6:11211131.

7. Binet, R., S. Letoffe, J. M. Ghigo, P. Delepelaire, and C. Wandersman.1997. Protein secretion by Gram-negative bacterial ABC exportersa re-view. Gene 192:711.

8. Bitter, W., M. Koster, M. Latijnhouwers, H. De Cock, and J. Tommassen.1998. Formation of oligomeric rings by XcpQ and PilQ, which are involvedin protein transport across the outer membrane of Pseudomonas aerugi-nosa. Mol. Microbiol. 27:209219.

9. Blattner, F. R., G. Plunkett III, C. A. Bloch, N. T. Perna, V. Burland, M.Riley, J. Collado-Vides, J. D. Glasner, C. K. Rode, G. F. Mayhew, J. Gregor,N. W. Davis, H. A. Kirkpatrick, M. A. Goeden, D. J. Rose, B. Mau, and Y.

Shao. 1997. The complete genome sequence of Escherichia coli K-12.Science 277:14531474.

10. Bleves, S., M. Gerard-Vincent, A. Lazdunski, and A. Filloux. 1999. Struc-ture-function analysis of XcpP, a component involved in general secretorypathway-dependent protein secretion in Pseudomonas aeruginosa. J. Bac-teriol. 181:40124019.

11. Bleves, S., A. Lazdunski, and A. Filloux. 1996. Membrane topology of threeXcp proteins involved in exoprotein transport by Pseudomonas aeruginosa.J. Bacteriol. 178:42974300.

12. Bleves, S., R. Voulhoux, G. Michel, A. Lazdunski, J. Tommassen, and A.Filloux. 1998. The secretion apparatus of Pseudomonas aeruginosa: iden-tification of a fifth pseudopilin, XcpX (GspK family). Mol. Microbiol.27:3140.

13. Braun, P., J. Tommassen, and A. Filloux. 1996. Role of the propeptide infolding and secretion of elastase of Pseudomonas aeruginosa. Mol. Micro-biol. 19:297306.

14. Brok, R., P. Van Gelder, M. Winterhalter, U. Ziese, A. J. Koster, H. DeCock, M. Koster, J. Tommassen, and W. Bitter. 1999. The C-terminaldomain of the Pseudomonas secretin XcpQ forms oligomeric rings withpore activity. J. Mol. Biol. 294:11691179.

15. Burland, V., Y. Shao, N. T. Perna, G. Plunkett, H. J. Sofia, and F. R.Blattner. 1998. The complete DNA sequence and analysis of the largevirulence plasmid of Escherichia coli O157:H7. Nucleic Acids Res. 26:41964204.

16. Burns, D. L. 1999. Biochemistry of type IV secretion. Curr. Opin. Micro-biol. 2:2529.

17. Caccavo, F., Jr., D. J. Lonergan, D. R. Lovely, M. Davis, J. F. Stolz, andM. J. McInerney. 1994. Geobacter sulfurreducens sp. nov., a hydrogen- andacetate-oxidizing dissimilatory metal-reducing microorganism. Appl. Envi-ron. Microbiol. 60:37523759.

18. Chapon, V., H. D. Simpson, X. Morelli, E. Brun, and F. Barras. 2000.Alteration of a single tryptophan residue of the cellulose-binding domainblocks secretion of the erwinia chrysanthemi cel5 cellulase (ex-EGZ) viathe type II system. J. Mol. Biol. 303:117123.

19. Chapon-Herve, V., M. Akrim, A. Latifi, P. Williams, A. Lazdunski, and M.Bally. 1997. Regulation of the xcp secretion pathway by multiple quorum-sensing modulons in Pseudomonas aeruginosa. Mol. Microbiol. 24:11691178.

20. Chiang, S. L., R. K. Taylor, M. Koomey, and J. J. Mekalanos. 1995. Singleamino acid substitutions in the N-terminus of Vibrio cholerae TcpA affectcolonization, autoagglutination, and serum resistance. Mol. Microbiol. 17:11331142.

21. Christie, P. J. 1997. Agrobacterium tumefaciens T-complex transport appa-ratus: a paradigm for a new family of multifunctional transporters in eu-bacteria. J. Bacteriol. 179:30853094.

22. Coates, J. D., E. J. P. Phillips, D. J. Lonergan, H. Jenter, and D. R. Lovely.1996. Isolation of Geobacter species from diverse sedimentary environ-ments. Appl. Environ. Microbiol. 62:15311536.

23. Collazo, C. M. and J. E. Galan. 1997. The invasion-associated type-IIIprotein secretion system in Salmonellaa review. Gene 192:5159.

24. Condemine, G., and V. E. Shevchik. 2000. Overproduction of the secretinOutD suppresses the secretion defect of an Erwinia chrysanthemi outBmutant. Microbiology 146:639647.

25. Connell, T. D., D. J. Metzger, J. Lynch, and J. P. Folster. 1998. Endochiti-nase is transported to the extracellular milieu by the eps-encoded generalsecretory pathway of Vibrio cholerae. J. Bacteriol. 180:55915600.

26. Cornelis, G. R., and F. Van Gijsegem. 2000. Assembly and function of typeiii secretory systems. Annu. Rev. Microbiol. 54:735774.

27. Daefler, S., I. Guilvout, K. R. Hardie, A. P. Pugsley, and M. Russel. 1997.The C-terminal domain of the secretin PulD contains the binding site for itscognate chaperone, PulS, and confers PulS dependence on pIVf1 function.Mol. Microbiol. 24:465475.

28. Davis, B. M., E. H. Lawson, M. Sandkvist, A. Ali, S. Sozhamannan, andM. K. Waldor. 2000. Convergence of the secretory pathways for choleratoxin and the filamentous phage, CTXphi. Science 288:333335.

29. de Groot, A., G. Gerritse, J. Tommassen, A. Lazdunski, and A. Filloux.1999. Molecular organization of the xcp gene cluster in Pseudomonasputida: absence of an xcpX (gspK) homologue. Gene 226:3540.

30. de Groot, A., J. J. Krijger, A. Filloux, and J. Tommassen. 1996. Character-ization of type II protein secretion (xcp) genes in the plant growth-stimu-lating Pseudomonas putida, strain WCS358. Mol Gen. Genet. 250:491504.

31. dEnfert, C., I. Reyss, C. Wandersman, and A. P. Pugsley. 1989. Proteinsecretion by gram-negative bacteria. Characterization of two membraneproteins required for pullulanase secretion by Escherichia coli K-12. J. Biol.Chem. 264:1746217468.



iai.asm.org

Downloadedfrom


11/13

32. dEnfert, C., A. Ryter, and A. P. Pugsley. 1987. Cloning and expression inEscherichia coli of the Klebsiella pneumoniae genes for production, surfacelocalization and secretion of the lipoprotein pullulanase. EMBO J. 6:35313538.

33. DeShazer, D., P. J. Brett, M. N. Burtnick, and D. E. Woods. 1999. Molec-ular characterization of genetic loci required for secretion of exoproducts inBurkholderia pseudomallei. J. Bacteriol. 181:46614664.

34. Dums, F., J. M. Dow, and M. J. Daniels. 1991. Structural characterizationof protein secretion genes of the bacterial phytopathogen Xanthomonas

campestris pathovar campestris: relatedness to secretion systems of othergram-negative bacteria. Mol. Gen. Genet. 229:357364.

35. Engebrecht, J., K. Nealson, and M. Silverman. 1983. Bacterial biolumines-cence: isolation and genetic analysis of functions from Vibrio fischeri. Cell32:773781.

36. Filloux, A., M. Bally, G. Ball, M. Akrim, J. Tommassen, and A. Lazdunski.1990. Protein secretion in gram-negative bacteria: transport across theouter membrane involves common mechanisms in different bacteria.EMBO J. 9:43234329.

37. Filloux, A., G. Michel, and M. Bally. 1998. GSP-dependent protein secre-tion in gram-negative bacteria: the Xcp system of Pseudomonas aeruginosa.FEMS Microbiol. Rev. 22:177198.

38. Francetic, O., C. Badaut, S. Rimsky, and A. P. Pugsley. 2000. The ChiA(YheB) protein of Escherichia coli K-12 is an endochitinase whose gene isnegatively controlled by the nucleoid-structuring protein H-NS. Mol. Mi-crobiol. 35:15061517.

39. Francetic, O., D. Belin, C. Badaut, and A. P. Pugsley. 2000. Expression ofthe endogenous type II secretion pathway in Escherichia coli leads to

chitinase secretion. EMBO J. 19:66976703.40. Francetic, O., and A. P. Pugsley. 1996. The cryptic general secretory path-

way (gsp) operon of Escherichia coli K-12 encodes functional proteins. J.Bacteriol. 178:35443549.

41. Fullner, K. J., and J. J. Mekalanos. 1999. Genetic characterization of a newtype IV-A pilus gene cluster found in both classical and E1 Tor biotypes ofVibrio cholerae. Infect. Immun. 67:13931404.

42. Genin, S., and C. A. Boucher. 1994. A superfamily of proteins involved indifferent secretion pathways in gram-negative bacteria: modular structureand specificity of the N-terminal domain. Mol. Gen. Genet. 243:112118.

43. Gram, L., and H. H. Huss. 1996. Microbiological spoilage of fish and fishproducts. Int. J. Food Microbiol. 33:121137.

44. Hales, L. M., and H. A. Shuman. 1999. Legionella pneumophila contains atype II general secretion pathway required for growth in amoebae as well asfor secretion of the Msp protease. Infect. Immun. 67:36623666.

45. Hardie, K. R., S. Lory, and A. P. Pugsley. 1996. Insertion of an outermembrane protein in Escherichia coli requires a chaperone-like protein.EMBO J. 15:978988.

46. Hase, C. C., M. E. Bauer, and R. A. Finkelstein. 1994. Genetic character-ization of mannose-sensitive hemagglutinin (MSHA)-negative mutants ofVibrio cholerae derived by Tn5 mutagenesis. Gene 150:1725.

47. Hase, C. C., and R. A. Finkelstein. 1991. Cloning and nucleotide sequenceof the Vibrio cholerae hemagglutinin/protease (HA/protease) gene and con-struction of an HA/protease-negative strain. J. Bacteriol. 173:33113317.

48. Hase, C. C., and J. J. Mekalanos. 1998. TcpP protein is a positive regulatorof virulence gene expression in Vibrio cholerae. Proc. Natl. Acad. Sci. USA95:730734.

49. He, S. Y., M. Lindeberg, A. K. Chatterjee, and A. Collmer. 1991. ClonedErwinia chrysanthemi out genes enable Escherichia coli to selectively se-crete a diverse family of heterologous proteins to its milieu. Proc. Natl.Acad. Sci. USA88:10791083.

50. Heidelberg, J. F., J. A. Eisen, W. C. Nelson, R. A. Clayton, M. L. Gwinn,R. J. Dodson, D. H. Haft, E. K. Hickey, J. D. Peterson, L. Umayam, S. R.Gill, K. E. Nelson, T. D. Read, H. Tettelin, D. Richardson, M. D. Ermo-

laeva, J. Vamathevan, S. Bass, H. Qin, I. Dragoi, P. Sellers, L. McDonald,

T. Utterback, R. D. Fleishmann, W. C. Nierman, and O. White. 2000. DNAsequence of both chromosomes of the cholera pathogen Vibrio cholerae.Nature 406:477483.

51. Henderson, I. R., F. Navarro-Garcia, and J. P. Nataro. 1998. The greatescape: structure and function of the autotransporter proteins. TrendsMicrobiol. 6:370378.

52. Herrington, D. A., R. H. Hall, G. Losonsky, J. J. Mekalanos, R. K. Taylor,and M. M. Levine. 1988. Toxin, toxin-coregulated pili, and the toxR regulonare essential for Vibrio cholerae pathogenesis in humans. J. Exp. Med.168:14871492.

53. Higgins, D. E., and V. J. DiRita. 1994. Transcriptional control of toxT, aregulatory gene in the ToxR regulon of Vibrio cholerae. Mol. Microbiol.14:1729.

54. Hirst, T. R., and J. Holmgren. 1987. Conformation of protein secretedacross bacterial outer membranes: a study of enterotoxin translocation fromVibrio cholerae. Proc. Natl. Acad. Sci. USA 84:74187422.

55. Hirst, T. R., and J. Holmgren. 1987. Transient entry of enterotoxin subunitsinto the periplasm occurs during their secretion from Vibrio cholerae. J.Bacteriol. 169:10371045.

56. Hirst, T. R., J. Sanchez, J. B. Kaper, S. J. Hardy, and J. Holmgren. 1984.

Mechanism of toxin secretion by Vibrio cholerae investigated in strainsharboring plasmids that encode heat-labile enterotoxins of Escherichia coli.Proc. Natl. Acad. Sci. USA 81:77527756.

57. Holmes, R. K., M. L. Vasil, and R. A. Finkelstein. 1975. Studies on toxi-nogenesis in Vibrio cholerae. III. Characterization of nontoxinogenic mu-tants in vitro and in experimental animals. J. Clin. Investig. 55:551560.

58. Housby, J. N., J. D. Thomas, S. D. Wharam, P. J. Reeves, and G. P.Salmond. 1998. Conditional mutations in OutE and OutL block exoenzymesecretion across the Erwinia carotovora outer membrane. FEMS Microbiol.

Lett. 165:91102.59. Howard, S. P., J. Critch, and A. Bedi. 1993. Isolation and analysis of eight

exe genes and their involvement in extracellular protein secretion and outermembrane assembly in Aeromonas hydrophila. J. Bacteriol. 175:66956703.

60. Howard, S. P., H. G. Meiklejohn, D. Shivak, and R. Jahagirdar. 1996. ATonB-like protein and a novel membrane protein containing an ATP-binding cassette function together in exotoxin secretion. Mol. Microbiol.22:595604.

61. Hu, N. T., P. F. Lee, and C. Chen. 1995. The type IV pre-pilin leaderpeptidase of Xanthomonas campestris pv. campestris is functional withoutconserved cysteine residues. Mol. Microbiol. 18:769777.

62. Ichige, A., K. Oishi, and S. Mizushima. 1988. Isolation and characterizationof mutants of a marine Vibrio strain that are defective in production ofextracellular proteins. J. Bacteriol. 170:35373542.

63. Jacob-Dubuisson, F., C. El Hamel, N. Saint, S. Guedin, E. Willery, G.Molle, and C. Locht. 1999. Channel formation by FhaC, the outer mem-brane protein involved in the secretion of the Bordetella pertussis filamen-tous hemagglutinin. J. Biol. Chem. 274:3773137735.

64. Jacob-Dubuisson, F., B. Kehoe, E. Willery, N. Reveneau, C. Locht, andD. A. Relman. 2000. Molecular characterization ofBordetella bronchisepticafilamentous haemagglutinin and its secretion machinery. Microbiology 146:12111221.

65. Jacobs, C., and L. Shapiro. 1998. Microbiol asymmetric cell division: local-ization of cell fate determinants. Curr. Opin. Genet. Dev. 8:386391.

66. Jahagirdar, R., and S. P. Howard. 1994. Isolation and characterization of asecond exe operon required for extracellular protein secretion in Aeromo-nas hydrophila. J. Bacteriol. 176:68196826.

67. Jiang, B., and S. P. Howard. 1992. The Aeromonas hydrophila exeE gene,required both for protein secretion and normal outer membrane biogenesis,is a member of a general secretion pathway. Mol. Microbiol. 6:13511361.

68. Jobling, M. G., and R. K. Holmes. 1997. Characterization of hapR, apositive regulator of the Vibrio cholerae HA/protease gene hap, and itsidentification as a functional homologue of the Vibrio harveyi luxR gene.Mol. Microbiol. 26:10231034.

69. Jones, S., B. Yu, N. J. Bainton, M. Birdsall, B. W. Bycroft, S. R. Chhabra,A. J. Cox, P. Golby, P. J. Reeves, and S. Stephens. 1993. The lux autoin-

ducer regulates the production of exoenzyme virulence determinants inErwinia carotovora and Pseudomonas aeruginosa. EMBO J. 12:24772482.70. Kagami, Y., M. Ratliff, M. Surber, A. Martinez, and D. N. Nunn. 1998. Type

II protein secretion by Pseudomonas aeruginosa: genetic suppression of aconditional mutation in the pilin-like component XcpT by the cytoplasmiccomponent XcpR. Mol. Microbiol. 27:221233.

71. Kaufman, M. R., C. E. Shaw, I. D. Jones, and R. K. Taylor. 1993. Biogenesisand regulation of the Vibrio cholerae toxin-coregulated pilus: analogies toother virulence factor secretory systems. Gene 126:4349.

72. Kazmierczak, B. I., D. L. Mielke, M. Russel, and P. Model. 1994. pIV, afilamentous phage protein that mediates phage export across the bacterialcell envelope, forms a multimer. J. Mol. Biol. 238:187198.

73. Lang, H., G. Jonson, J. Holmgren, and E. T. Palva. 1994. The maltoseregulon of Vibrio cholerae affects production and secretion of virulencefactors. Infect. Immun. 62:47814788.

74. Lauer, P., N. H. Albertson, and M. Koomey. 1993. Conservation of genesencoding components of a type IV pilus assembly/two-step protein exportpathway in Neisseria gonorrhoeae. Mol. Microbiol. 8:357368.

75. Lee, H. M., K. C. Wang, Y. L. Liu, H. Y. Yew, L. Y. Chen, W. M. Leu, D. C.Chen, and N. T. Hu. 2000. Association of the cytoplasmic membrane pro-tein XpsN with the outer membrane protein XpsD in the type II proteinsecretion apparatus ofXanthomonas campestris pv. campestris. J. Bacteriol.182:15491557.

76. Lee, S. H., D. L. Hava, M. K. Waldor, and A. Camilli. 1999. Regulation andtemporal expression patterns of Vibrio cholerae virulence genes duringinfection. Cell 99:625634.

77. Leece, R., and T. R. Hirst. 1992. Expression of the B subunit of Escherichiacoli heat-labile enterotoxin in a marine Vibrio and in a mutant that ispleiotropically defective in the secretion of extracellular proteins. J. Gen.Microbiol. 138:719724.

78. Letellier, L., S. P. Howard, and J. T. Buckley. 1997. Studies on the ener-getics of proaerolysin secretion across the outer membrane of Acromonasspecies. Evidence for a requirement for both the protonmotive force andATP. J. Biol. Chem. 272:1110911113.

79. Levine, M. M., J. B. Kaper, R. E. Black, and M. L. Clements. 1983. Newknowledge on pathogenesis of bacterial enteric infections as applied tovaccine development. Microbiol. Rev. 47:510550.



iai.asm.org

Downloadedfrom


12/13

80. Lewenza, S., B. Conway, E. P. Greenberg, and P. A. Sokol. 1999. Quorumsensing in Burkholderia cepacia: identification of the LuxRI homologs Ce-pRI. J. Bacteriol. 181:748756.

81. Liles, M. R., P. H. Edelstein, and N. P. Cianciotto. 1999. The prepilinpeptidase is required for protein secretion by and the virulence of theintracellular pathogen Legionella pneumophila. Mol. Microbiol. 31:959970.

82. Lindeberg, M., and A. Collmer. 1992. Analysis of eight out genes in a clusterrequired for pectic enzyme secretion by Erwinia chrysanthemi: sequence

comparison with secretion genes from other gram-negative bacteria. J.Bacteriol. 174:73857397.

83. Lindeberg, M., G. P. Salmond, and A. Collmer. 1996. Complementation ofdeletion mutations in a cloned functional cluster of Erwinia chrysanthemiout genes with Erwinia carotovora out homologues reveals OutC and OutDas candidate gatekeepers of species-specific secretion of proteins via thetype II pathway. Mol. Microbiol. 20:175190.

84. Linderoth, N. A., P. Model, and M. Russel. 1996. Essential role of a sodiumdodecyl sulfate-resistant protein IV multimer in assembly-export of fila-mentous phage. J. Bacteriol. 178:19621970.

85. Linderoth, N. A., M. N. Simon, and M. Russel. 1997. The filamentous phagepIV multimer visualized by scanning transmission electron microscopy.Science 278:16351638.

86. Lindgren, V., and B. Wretlind. 1987. Characterization of a Pseudomonasaeruginosa transposon insertion mutant with defective release of exoen-zymes. J. Gen. Microbiol. 133:675681.

87. Lory, S. 1998. Secretion of proteins and assembly of bacterial surfaceorganelles: shared pathways of extracellular protein targeting. Curr. Opin.

Microbiol. 1:2735.88. Lu, H. M., S. T. Motley, and S. Lory. 1997. Interactions of the components

of the general secretion pathway: role of Pseudomonas aeruginosa type IVpilin subunits in complex formation and extracellular protein secretion.Mol. Microbiol. 25:247259.

89. Makino, K., K. Ishii, T. Yasunaga, M. Hattori, K. Yokoyama, C. H. Yut-sudo, Y. Kubota, Y. Yamaichi, T. Iida, K. Yamamoto, T. Honda, C. G. Han,E. Ohtsubo, M. Kasamatsu, T. Hayashi, S. Kuhara, and H. Shinagawa.

1998. Complete nucleotide sequences of 93-kb and 3.3-kb plasmids of anenterohemorrhagic Escherichia coli O157:H7 derived from Sakai outbreak.DNA Res. 5:19.

90. Marciano, D. K., M. Russel, and S. M. Simon. 1999. An aqueous channelfor filamentous phage export. Science 284:15161519.

91. Marsh, J. W., D. Sun, and R. K. Taylor. 1996. Physical linkage of the Vibriocholerae mannose-sensitive hemagglutinin secretory and structural subunitgene loci: identification of the mshG coding sequence. Infect. Immun.64:460465.

92. Marsh, J. W., and R. K. Taylor. 1998. Identification of the Vibrio cholerae

type 4 prepilin peptidase required for cholera toxin secretion and pilusformation. Mol. Microbiol. 29:14811492.

93. Marsh, J. W., and R. K. Taylor. 1999. Genetic and transcriptional analysesof the Vibrio cholerae mannose-sensitive hemagglutinin type 4 pilus genelocus. J. Bacteriol. 181:11101117.

94. Martinez, A., P. Ostrovsky, and D. N. Nunn. 1998. Identification of anadditional member of the secretin superfamily of proteins in Pseudomonasaeruginosa that is able to function in type II protein secretion. Mol. Mi-crobiol. 28:12351246.

95. Martinez, A., P. Ostrovsky, and D. N. Nunn. 1999. LipC, a second lipase ofPseudomonas aeruginosa, is LipB and Xcp dependent and is transcription-ally regulated by pilus biogenesis components. Mol. Microbiol. 34:317326.

96. Mclver, K. S., E. Kessler, J. C. Olson, and D. E. Ohman. 1995. The elastasepropeptide functions as an intramolecular chaperone required for elastaseactivity and secretion in Pseudomonas aeruginosa. Mol. Microbiol. 18:877889.

97. McKenney, D., K. E. Brown, and D. G. Allison. 1995. Influence of Pseudo-monas aerugino sa exoproducts on virulence factor production in Burkhold-

eria cepacia: evidence of interspecies communication. J. Bacteriol. 177:69896992.

98. Merritt, E. A., T. K. Sixma, K. H. Kalk, B. A. van Zanten, and W. G. Hol.1994. Galactose-binding site in Escherichia coli heat-labile enterotoxin(LT) and cholera toxin (CT). Mol. Microbiol. 13:745753.

99. Merz, A. J., M. So, and M. P. Sheetz. 2000. Pilus retraction powers bacterialtwitching motility. Nature 407:98102.

100. Michel, L. O., M. Sandkvist, and M. Bagdasarian. 1995. Specificity of theprotein secretory apparatus: secretion of the heat-labile enterotoxin B sub-unit pentamers by different species of gram-bacteria. Gene 152:4145.

101. Miller, V. L., and J. J. Mekalanos. 1984. Synthesis of cholera toxin ispositively regulated at the transcriptional level by toxR. Proc. Natl. Acad.Sci. USA 81:34713475.

102. Miller, V. L., and J. J. Mekalanos. 1988. A novel suicide vector and its usein construction of insertion mutations: osmoregulation of outer membraneproteins and virulence determinants in Vibrio cholerae requires toxR. J.Bacteriol. 170:25752583.

103. Nealson, K. H., T. Platt, and J. W. Hastings. 1970. Cellular control of the

synthesis and activity of the bacterial luminescent system. J. Bacteriol.104:313322.

104. Nikaido, H. 1994. Porins and specific diffusion channels in bacterial outermembranes. J. Biol. Chem. 269:39053908.

105. Nikaido, H. 1996. Multidrug efflux pumps of gram-negative bacteria. J.Bacteriol. 178:58535859.

106. Nikaido, H., and M. H. Saier, Jr. 1992. Transport proteins in bacteria:common themes in their design. Science 258:936942.

107. Nouwen, N., N. Ranson, H. Saibil, B. Wolpensinger, A. Engel, A. Ghazi, andA. P. Pugsley. 1999. Secretin PulD: association with pilot PulS, structure,and ion-conducting channel formation. Proc. Natl. Acad. Sci. USA 96:81738177.

108. Nouwen, N., H. Stahlberg, A. P. Pugsley, and A. Engel. 2000. Domainstructure of secretin PulD revealed by limited proteolysis and electronmicroscopy. EMBO J. 19:22292236.

109. Nunn, D. 1999. Bacterial type II protein export and pilus biogenesis: morethan just homologies? Trends Cell Biol. 9:402408.

110. Nunn, D., S. Bergman, and S. Lory. 1990. Products of three accessorygenes, pilB, pilC, and pilD, are required for biogenesis of Pseudomonasaeruginosa pili. J. Bacteriol. 172:29112919.

111. Nunn, D. N., and S. Lory. 1993. Cleavage, methylation, and localization ofthe Pseudomonas aeruginosa export proteins XcpT, -U, -V, and -W. J.Bacteriol. 175:43754382.

112. Odenbreit, S., J. Puls, B. Sedlmaier, E. Gerland, W. Fischer, and R. Haas.2000. Translocation of Helicobacter pylori CagA into gastric epithelial cellsby type IV secretion. Science 287:14971500.

113. Overbye, L. J., M. Sandkvist, and M. Bagdasarian. 1993. Genes requiredfor extracellular secretion of enterotoxin are clustered in Vibrio cholerae.Gene 132:101106.

114. Passador, L., J. M. Cook, M. J. Gambello, L. Rust, and B. H. Iglewski. 1993.Expression of Pseudomonas aeruginosa virulence genes requires cell-to-cellcommunication. Science 260:11271130.

115. Pearson, J. P., L. Passador, B. H. Iglewski, and E. P. Greenberg. 1995. Asecond N-acylhomoserine lactone signal produced by Pseudomonas aerugi-nosa. Proc. Natl. Acad. Sci. USA 92:14901494.

116. Peek, J. A., and R. K. Taylor. 1992. Characterization of a periplasmicthiol:disulfide interchange protein required for the functional maturation ofsecreted virulence factors of Vibrio cholerae. Proc. Natl. Acad. Sci. USA89:62106214.

117. Perna, N. T., G. Plunkett, V. Burland, B. Mau, J. D. Glasner, D. J. Rose,G. F. Mayhew, P. S. Evans, J. Gregor, H. A. Kirkpatrick, G. Posfai, J.

Hackett, S. Klink, A. Boutin, Y. Shao, L. Miller, E. Grotbeck, N. W. Davis,

A. Lim, E. T. Dimalanta, K. D. Potamousis, J. Apodaca, T. S. Ananthara-man, J. Lin, G. Yen, D. C. Schwartz, R. A. Welch, and F. R. Blattner. 2001.Genome sequence of enterohaemorrhagic Escherichia coli 0157:H7. Na-

ture 409:529533.118. Pesci, E. C., J. P. Pearson, P. C. Seed, and B. H. Iglewski. 1997. Regulation

of las and rhl quorum sensing in Pseudomonas aeruginosa. J. Bacteriol.179:31273132.

119. Pirhonen, M., D. Flego, R. Heikinheimo, and E. T. Palva. 1993. A smalldiffusible signal molecule is responsible for the global control of virulenceand exoenzyme production in the plant pathogen Erwinia carotovora.EMBO J. 12:24672476.

120. Pohlner, J., R. Halter, K. Beyreuther, and T. F. Meyer. 1987. Gene struc-ture and extracellular secretion of Neisseria gonorrhoeae IgA protease.Nature 325:458462.

121. Possot, O. M., M. Gerard-Vincent, and A. P. Pugsley. 1999. Membraneassociation and multimerization of secreton component pulC. J. Bacteriol.181:40044011.

122. Possot, O. M., L. Letellier, and A. P. Pugsley. 1997. Energy requirement forpullulanase secretion by the main terminal branch of the general secretorypathway. Mol. Microbiol. 24:457464.

123. Possot, O. M., G. Vignon, N. Bomchil, F. Ebel, and A. P. Pugsley. 2000.

Multiple interactions between pullulanase secreton components involved instabilization and cytoplasmic membrane association of PulE. J. Bacteriol.182:21422152.

124. Pugsley, A. P., and O. Francetic. 1998. Protein secretion in Escherichia coliK-12: dead or alive? Cell. Mol. Life Sci. 54:347352.

125. Pugsley, A. P., O. Francetic, O. M. Possot, N. Sauvonnet, and K. R. Hardie.1997. Recent progress and future directions in studies of

typeii secretion and pathogenesis

Documents